Minerva Access is the Institutional Repository of The University of Melbourne
Author/s:
Bassirat, Maryam
Title:
Mechanisms underlying changes in microvascular blood flow in a diabetic rat model:
relevance to tissue repair
Date:
2002-08
Citation:
Bassirat, M. (2002). Mechanisms underlying changes in microvascular blood flow in a
diabetic rat model: relevance to tissue repair, PhD thesis, Medicine, The Royal Melbourne
Hospital.
Publication Status:
Published
Persistent Link:
http://hdl.handle.net/11343/38988
File Description:
Mechanisms underlying changes in microvascular blood flow in a diabetic rat model:
relevance to tissue repair
Terms and Conditions:
Terms and Conditions: Copyright in works deposited in Minerva Access is retained by the
copyright owner. The work may not be altered without permission from the copyright owner.
Readers may only download, print and save electronic copies of whole works for their own
personal non-commercial use. Any use that exceeds these limits requires permission from
the copyright owner. Attribution is essential when quoting or paraphrasing from these works.
MECHANISMS UNDERLYING CHANGES IN
MICROVASCULAR BLOOD FLOW IN A DIABETIC RAT MODEL:
RELEVANCE TO TISSUE REPAIR
Maryam Bassirat
Submitted in total fulfilment of the requirements of the degree of Doctor of
Philosophy
August 2002
Department of Medicine
Royal Melbourne Hospital
The University of Melbourne
1
DECLARATION
This is to certify that:
(i)
The thesis comprises only my original work, completed while I was a
postgraduate student in the faculty of Medicine at the University of Melbourne.
This thesis contains no material that has been accepted for the award of any other
degree or diploma at any other university or institution.
(ii)
To the best of my knowledge and belief, the thesis contains no material previously
written or published by another person except where due reference is made in the
text
(iii)
This thesis is less than 10,000 words in length, exclusive of tables, figures,
references and appendices.
Signed:
Maryam Bassirat
2
ACKNOWLEDGMENTS
This work was performed at the National Ageing Research Institute (NARI), University
of Melbourne. I would like to extent my unreserved gratitude to Associate Professor
Zeinab Khalil for her supervision over the past 6 years. Her knowledge, enthusiasm,
integrity, and generosity have been an inspiration. She has been a wonderful supervisor
and it was with her invaluable guidance and enduring patience that I was able to complete
this work. Her training over the past 9 years has allowed me to broaden my knowledge
and scientific skills and I feel extremely privileged to have had the opportunity to work
and study under her supervision.
I am also grateful to Professor Robert Helme for giving me the opportunity to work and
study at the National Ageing Research Institute.
I would like to sincerely thank Associate Professor Marjorie Dunlop and her team at the
Department of Medicine (Royal Melbourne Hospital) for assisting me in establishing the
STZ-diabetic rat model in the early stages of this study.
I extend a special thank you to Dr. Michael Farrell for his review of my draft thesis and
his valuable comments. I am extremely grateful to Dr. Judi Dunai, of NARI, for her
guidance in statistical analysis of the data. Thank you also to the staff of the Statistical
Consulting Centre at The University of Melbourne for providing directives that
contributed to solving the complex issues that emerged from this research.
A big thank you is extended to Mr. Michael Gorman for his general technical assistance,
especially in using the CO2 laser to induce thermal wounds. I am also grateful to Dr. Paul
Andrews for his ongoing support with all computer-related issues.
To all NARI staff and colleagues, thank you so much for all your support,
encouragement, and friendship.
Finally, I would like to sincerely thank my husband Gerard for his assistance in proof
reading my thesis and giving me his total dedication and support. He always believed that
this thesis would eventuate and his never-ending strength, patience, and love made it all
so
much
3
easier.
SUMMARY
Diabetes mellitus is a chronic syndrome affecting carbohydrate, protein, and fat
metabolism. It is characterized primarily by relative or absolute insufficiency of insulin
secretion (type I diabetes or IDDM) or concomitant insensitivity / resistance to the
metabolic action of insulin on target tissues (Type II diabetes or NIDDM), both resulting
in hyperglycaemia. Diabetes mellitus is known to induce microvascular changes and
alterations to neuronal functions.
The neurovascular system comprising of unmyelinated primary afferent sensory neurones
and the microvasculature innervated by these nerves play a major role in modulating
inflammatory and tissue repair processes. Sensory nerve terminals respond to injury via
the release of sensory neuropeptides which mediate inflammation and tissue repair. These
processes are known to be altered in diabetes.
This thesis is concerned with the role of diabetes in modulating microvascular blood flow
directly and indirectly via modulating sensory nerve activity and the effect of these
changes on repair processes in skin of 4 weeks streptozotocin (STZ)-induced diabetic
rats. The following hypotheses were examined:
1. That factors implicated in long-term diabetic vascular damage play a role in altering
skin microvascular function in early diabetes.
2. That preventing the deleterious effects of these factors could improve skin
microvascular blood flow and skin repair processes in early diabetes.
Specifically, the thesis examined the involvement of a number of factors including
hyperglycaemia, endothelins (ETs), free radicals, and advanced glycated end-products
(AGEs) in modulating skin microvascular blood flow in early diabetes. It was shown that
preventing the deleterious effects of these factors, using specific inhibitors / antagonists,
could improve microvascular blood flow and consequently modulate tissue repair
processes in early diabetes.
4
Chapter 1 reviews issues related to diabetes mellitus, neurogenic inflammation, and
wound repair. The function of sensory neuropeptides and their interaction with other
agents (e.g. free radicals, ETs, AGEs) during inflammation is discussed in detail. The
literature review reveals that investigations of early changes in microvascular blood flow
in the skin is lacking and that this area needs further attention as it impacts on possible
future early intervention approaches to limit long-term diabetic vascular complications.
Chapter 2 describes the general methodology including induction of diabetes with STZ
(in Sprague-Dawley rats) as well as methods used to induce an inflammatory response
and to measure two specific parameters of this response, namely plasma extravasation
and vasodilatation (using laser Doppler flowmetry). Procedures to electrically stimulate
C-fibres and to induce a full thickness skin wound are also described.
In chapter 3, the effect of diabetes on sensory nerve function, as determined by the ability
of these nerves to mount an inflammatory response, was examined. One of the aims was
to examine changes in inflammatory responses with time. Sensory-independent effects of
diabetes on the microvasculature were examined in sensory-denervated (capsaicintreated) rats. In addition, the hyperglycaemic effect of diabetes, independent of possible
neurotoxic effects of STZ, was examined in alloxan-induced diabetic rats. The first set of
results demonstrated that there is an overall reduction in the neurogenic inflammatory
responses (vasodilatation and plasma extravasation) over 10 weeks of diabetes and this
reduction reached a maximum 4 weeks after induction of diabetes. Using two different
laser Doppler flowmeter machines, with different laser light penetration capabilities, it
was shown that capillary blood flow at both superficial and deeper compartments of the
skin is reduced after 4 weeks of diabetes. In addition, it was also demonstrated that postterminal changes in microvascular function are more apparent at this early stage of
diabetes. Overall, the results supported the notion that there are disturbances in
endothelial cells and smooth muscle cell function in skin microvasculature in early
diabetes.
5
In chapter 4, the role of parasympathetic nerves and local tissue factors in modulation of
microvascular blood flow in early diabetes were examined. Results showed a significant
reduction in microvascular responses to acetylcholine (ACh), muscarine chloride,
nicotine, and bradykinin (BK) in early diabetes. Reduction in the above responses could
be attributed to a number of factors that were further explored in chapters 6, 7, and 8.
In chapter 5, guanethidine-treated diabetic rats were used to examine the role of the
sympathetic nerves in the modulation of microvascular blood flow in early diabetes.
Results demonstrated that the contribution of sympathetic efferents in regulating
microvascular blood flow is increased in early diabetes.
Chapters 6, 7, and 8 examined possible role of ETs, free radicals, and AGEs, in
modulating microvascular blood flow in diabetes. This was investigated using specific
inhibitors and receptor antagonists. The evidence in chapter 6 showed that ETs are
involved in altering microvascular blood flow and endothelial cell permeability in 4
weeks diabetic rats. It was also shown that local pharmacological manipulation at the
level of the skin microcirculation, using ET receptor antagonists (e.g. BQ-123 & BQ788), could normalize local diabetic vascular responses with subtle differences in the
relative contribution of these two receptor subtypes.
Results in chapter 7 demonstrated that free radicals play a role in modulating diabetic
microvascular blood flow at this early stage. In addition, evidence was provided to
suggest that the inhibitory effects of ETs and free radicals on microvascular blood flow
were additive. Results from chapters 6 and 7 collectively showed that ETs play a greater
role than free radicals in altering endothelial cell permeability. Using different free
radical scavengers, there were subtle differences in the relative contribution of different
free radicals in altering microvascular blood flow. The results again showed that local
pharmacological manipulation at the level of peripheral skin microvasculature could
normalize diabetic vascular responses.
6
In chapter 8, evidence was provided that short-term exposure of control microvasculature
to early (e.g. amadori) and late (AGEs) glycation products could modulate microvascular
function in these rats. Furthermore, amadori products exerted a positive enhancing effect
on microvascular function in early diabetes. It was proposed that in diabetic rats, amadori
products could provide a respite for the microvasculature and hence, improve the
inflammatory responses. When anti-RAGE IgG was used, results showed that AGE
receptors are involved in inhibiting the inflammatory responses in diabetic rats.
Furthermore, evidence was provided to suggest that short-term aminoguanidine (AG)
treatment could improve the inflammatory responses in diabetic rats and that the duration
of treatment could play an important role with reference to the extent of preventing
neurovascular damage.
Chapter 9 describes the role of insulin treatment in the modulation of skin blood flow in
early diabetes. Results in this chapter showed that early onset insulin treatment could
have an important role in preventing early skin microvascular alterations by diabetes.
This raised the possibility that adequate control of hyperglycaemia at the early stages of
diabetes may reduce the long-term impact on microvascular and tissue repair processes.
The experiment in chapter 10 was devised to utilize the information gained from previous
chapters in order to examine the impact of preventing the deleterious effects of a number
of factors namely ETs, free radicals, AGEs and hyperglycaemia on modulating tissue
repair of full thickness wounds in early diabetes. The results showed that all treatments
used have contributed, to varying degrees, to different stages of wound repair and to the
overall improvement in repair processes in diabetic rats. Furthermore, Pharmacological
manipulations at the level of microvasculature (e.g. local injection of BQ-123 and
superoxide dismutase (SOD)) and systemic treatments (e.g. insulin, AG, Tirilazad
mesylate) were shown to improve repair processes in early diabetes.
Chapter 11 summarizes the overall findings of this thesis.
The thesis concludes that a number of factors that are known to cause long-term vascular
complications, also contribute to early changes in skin microvascular blood flow in
7
diabetes. The treatment protocols used in this thesis (chapters 6, 7, 8, and 9) showed that
the action of these factors is reversible at the early stages of diabetes. These same
treatments were also effective in improving the repair processes in these rats. This work
raised the exciting possibility that early intervention could prevent microvascular
alterations and argues that further research is needed to determine the impact of early
intervention on decelerating late diabetic complications.
8
GLOSSARY OF ABBREVIATIONS
Acetylcholine
Adenine diphosphate
Adenine triphosphate
Advanced glycated end-products
American diabetes association
Aminoguanidine hydrochloride
Analysis of variance
Angiotensin-converting enzyme
Antibody against RAGE
Arachidonic acid
L-arginine
Arteriovenous anastomoses
Basal blood flow
Basic fibroblast growth factor
Bisindolylmaleimide
Bovine serum albumin
Bradykinin
Calcitonin gene-related peptide
Central nervous system
Colecystokinin
Colony stimulating factor
Cyclic adenosine monophosphate
Cyclic Guanosine monophosphate
Cyclo-oxygenase
Diabetes control and complications trial
Diabetes mellitus
Diacylglycerol
Deoxyribonucleic acid
Dorsal root ganglia
Electrical stimulation
Endothelial NOS
Endothelin
Endothelin-1
Endothelium-derived hyperpolarizing factor
Endothelium-derived relaxing factor
Endothelium-derived constrictor factor
Epidermal growth factor
Fasting plasma glucose
Fibroblast growth factor
Glomerular filtration rate
Glucose transporter
Glycosaminoglycan
Guanosine monophosphate
Guanosine triphosphate
Hertz
Human leucocyte antigen
Hydrogen peroxide
5-Hydroxytryptamine
Inducible NOS
ACh
ADP
ATP
AGEs
ADA
AG
ANOVA
ACE
Anti-RAGE IgG
AA
L-Arg
AVA
BBF
bFGF
BIM
BSA
BK
CGRP
CNS
CCK
CSF
cAMP
cGMP
COX
DCCT
DM
DAG
DNA
DRG
ES
eNOS
ET
ET-1
EDHF
EDRF
EDCF
EGF
FPG
FGF
GFR
GLUT
GAG
GMP
GTP
Hz
HLA
H2O2
5-HT
iNOS
9
International association for the study of pain
Inositol triphosphate
Insulin-dependent diabetes mellitus
Insulin-like growth factor
Interleukin
Intradermal
Intraperitoneal
Intravenous
Kilo pascal
Laser Doppler flowmeter
Low density lipoprotein
Malonyldialdehyde
MAP kinase
Messenger ribonucleic acid
Moor laser Doppler flowmeter
N-acetylcysteine
Nerve growth factor
Neurogenic inflammation
Neurokinin A
Neurokinin-1
Neuronal NOS
Nicotinamide adenine dinucleotide
Nicotinamide adenine dinucleotide hydrogen
Nicotinamide adenine dinucleotide phosphate
Nicotinamide adenine dinucleotide hydrogen phosphate
Nitric oxide
Nitric oxide synthase
NG-nitro-L-arginine methyl ester
Non insulin-dependent diabetes mellitus
Noradrenaline
Neuropeptide Y
Oral glucose tolerance test
Oxidised LDL
Perimed laser Doppler flow meter
Peripheral vascular disease
N-Phenacyl thiazolium bromide
Phosphoinositol
Phospholipase A2
Phospholipase C
Phospholipase D
Plasma extravasation
Platelet-derived angiogenesis factor
Platelet-derived growth factor
Polymorphonuclear Leucocyte
Potassium chloride
Prostacyclin
Protein kinase C
Reactive oxygen species
Receptor for advanced glycated end-products
Sodium nitroprusside
Soluble RAGE
10
IASP
IP3
IDDM
IGF
IL
i.d.
i.p.
i.v.
kPa
LDF
LDL
MDA
MAPK
mRNA
MLDF
NAC
NGF
NI
NKA
NK-1
nNOS
NAD
NADH
NADP
NADPH
NO
NOS
L-NAME
NIDDM
NA
NPY
OGGT
oxLDL
PLDF
PVD
PTB
PI
PLA2
PLC
PLD
PE
PDAF
PDGF
PMN
KCl
PGI2
PKC
ROS
RAGE
SNP
sRAGE
Standard error of means
Subcutaneous
Substance P
Substance P-like immunoreactivity
Superoxide dismutase
Streptozotocin
Thiobarbituric acid
Transforming growth factor
Tumour necrosis factor
Vascular endothelial growth factor
Vasoactive intestinal contractor endothelin
Vasoactive intestinal peptide
Vasodilatation
Volts
World health organization
SEM
s.c.
SP
SPLI
SOD
STZ
TBAR
TGF
TNF
VEGF
VIC
VIP
VD
V
WHO
11
PUBLICATIONS ARISED FROM THIS WORK AND ABSTRACTS
PRESENTED TO SCIENTIFIC SOCIETIES
Bassirat M and Khalil Z. Endothelin and Free radicals modulate microvascular
responses in streptozotocin-induced diabetic rat model. Microvascular Research 59:8898 (2000).
Bassirat M and Khalil Z. Short- and long-term modulation of inflammation in
streptozotocin-induced diabetic rats by glycosylated products. Diabetes (Submitted)
(2001)
Bassirat M and Khalil Z. Factors modulating microvascular blood flow in
streptozotocin-induced diabetic rats: Relevance to tissue repair (In preparation) (2001)
Bassirat M and Khalil Z. The role of endothelin and superoxide anions in modulating
inflammation in streptozotocin-induced diabetic rat model.
Proceeding of the Australian Neuroscience Society (ANS) 9:144 (Canberra, Jan 1998)
Bassirat M and Khalil Z. Endothelin and reactive oxygen-species modulate
inflammation in streptozotocin-induced diabetic rats.
The Australasian Society for Immunology (ASI): Society for Leukocytes and Inflammation
Research (SLIR)
(Melbourne, Dec 1998)
Bassirat M and Khalil Z. Modulation of microvascular responses in streptozotocininduced diabetic rat by Glycosylated products.
The Science of Ageing 1:45 (Melbourne, Nov 1999)
Bassirat M and Khalil Z. Short- and long-term modulation of inflammation in
streptozotocin-induced diabetic rats by glycosylated products.
The Australasian Society of Clinical and Experimental Pharmacologists and
Toxicologists (ASCEPT)
(Sydney, Dec 1999)
Bassirat M and Khalil Z. Modulation of microvascular responses in streptozotocininduced diabetic rat by aminoguanidine and insulin.
Proceeding of the Australian Neuroscience Society (ANS) (Melbourne, Jan 2000)
12
TABLE OF CONTENTS
DECLARATION .............................................................................................................................................................. ii
ACKNOWLEDGMENTS ............................................................................................................................................... iii
SUMMARY..................................................................................................................................................................... iv
GLOSSARY OF ABBREVIATIONS ........................................................................................................................... viii
PUBLICATIONS ARISED FROM THIS WORK AND ABSTRACTS PRESENTED TO SCIENTIFIC SOCIETIES xi
LIST OF TABLES……………………………………………………………………………………………………….xv
LIST OF
FIGURES………………………………………………………………………………………………………xvi
CHAPTER 1:LITERATURE REVIEW .................................................................................................... 1
1.1. DIABETES MELLITUS: HISTORICAL PROSPECTIVE ............................................................................ 2
1.2. CHARACTERIZATION AND DIAGNOSIS .............................................................................................. 2
1.3. AETIOLOGY ..................................................................................................................................... 3
1.3.1.
Type I diabetes mellitus .......................................................................................................... 3
1.3.2.
Type II diabetes mellitus......................................................................................................... 4
1.4. LONG-TERM COMPLICATIONS .......................................................................................................... 5
1.4.1.
Macrovascular disease ........................................................................................................... 5
1.4.2.
Microvascular disease ............................................................................................................ 7
1.4.2.1.
1.4.2.2.
1.4.2.3.
Retinopathy ........................................................................................................................................ 9
Nephropathy..................................................................................................................................... 10
Neuropathy....................................................................................................................................... 11
1.5. MECHANISMS OF DIABETIC COMPLICATIONS ................................................................................. 15
1.5.1.
Polyol pathway ..................................................................................................................... 16
1.5.2.
Advanced Glycated End-products (AGEs)............................................................................ 18
1.5.3.
Oxidative stress and antioxidants ......................................................................................... 29
1.5.4.
Alterations in smooth muscle reactivity in diabetes.............................................................. 35
1.5.4.1.
1.5.4.2.
Vasoconstrictor responses ................................................................................................................ 35
Vasodilator responses....................................................................................................................... 39
1.6. ANIMAL MODELS OF DIABETES ...................................................................................................... 45
1.6.1.
History .................................................................................................................................. 45
1.6.2.
Chemical induction of diabetes ............................................................................................ 46
1.6.2.1. Alloxan............................................................................................................................................. 46
1.6.2.2. Streptozotocin .................................................................................................................................. 48
1.6.2.2.1. Role of nicotinamide and vitamin E ........................................................................................... 49
1.7. NEUROGENIC INFLAMMATION ....................................................................................................... 50
1.7.1.
History .................................................................................................................................. 50
1.7.2.
Sensory nerves ...................................................................................................................... 51
1.7.3.
Contribution of SP and CGRP to NI..................................................................................... 52
1.7.4.
Electrical stimulation of the sciatic nerve ............................................................................ 55
1.7.5.
Measuring NI........................................................................................................................ 57
1.8. MODULATORS OF SENSORY NERVE FUNCTION ............................................................................... 59
1.8.1.
Nitric oxide ........................................................................................................................... 59
1.8.2.
Capsaicin.............................................................................................................................. 62
1.8.3.
Sympathetic nerve................................................................................................................. 64
1.8.4.
Endothelin............................................................................................................................. 69
1.8.4.1.
1.8.5.
1.8.5.1.
1.8.5.2.
1.8.5.3.
1.8.5.4.
Endothelin and diabetes ................................................................................................................... 73
Other modulators.................................................................................................................. 78
Acetylcholine ................................................................................................................................... 79
Bradykinin........................................................................................................................................ 82
Insulin .............................................................................................................................................. 85
Protein Kinase C (PKC) .................................................................................................................. 89
1.9. WOUND HEALING .......................................................................................................................... 91
1.9.1.
Skin ....................................................................................................................................... 91
1.9.1.1.
1.9.1.2.
1.9.1.3.
Structure / extracellular matrix......................................................................................................... 91
Vascular supply / Endothelium ........................................................................................................ 94
Innervation ....................................................................................................................................... 96
13
1.9.2.
1.9.2.1.
1.9.2.2.
1.9.3.
1.9.3.1.
1.9.3.2.
Mechanism of tissue repair................................................................................................... 97
Role of sensory nerves ................................................................................................................... 100
Role of growth factors.................................................................................................................... 101
Effect of diabetes ................................................................................................................ 102
Structural changes .......................................................................................................................... 102
Physiological changes .................................................................................................................... 103
1.10. AIMS AND HYPOTHESES ............................................................................................................... 106
CHAPTER 2:MATERIALS AND GENERAL METHODS ................................................................ 108
2.1. MATERIALS ................................................................................................................................. 109
2.2. ANIMALS ..................................................................................................................................... 109
2.3. TYPES OF FOOD ........................................................................................................................... 110
2.3.1.
Mouse Breeder.................................................................................................................... 110
2.3.2.
GR2 ..................................................................................................................................... 110
2.4. INDUCTION OF DIABETES ............................................................................................................. 111
2.4.1. Choosing a diabetogenic agent ................................................................................................ 111
2.4.2. Control rats .............................................................................................................................. 112
2.4.3. Alloxan treatment ..................................................................................................................... 112
2.4.4. Streptozotocin ........................................................................................................................... 112
2.4.5. STZ vs Alloxan.......................................................................................................................... 113
2.4.6. Modifications made to the original procedure ......................................................................... 113
2.4.6.1.
Establishing a time frame............................................................................................................... 115
2.5. INSULIN TREATMENT PROTOCOL ................................................................................................. 116
2.6. BLISTER MODEL OF NEUROGENIC INFLAMMATION....................................................................... 116
2.7. ELECTRICAL STIMULATION OF SCIATIC NERVE ............................................................................ 118
2.8. NEONATAL CAPSAICIN PRE-TREATMENT ..................................................................................... 119
2.9. BLOOD FLOW MEASUREMENTS USING LASER DOPPLER FLOWMETER (LDF)................................ 120
2.9.1.
Perimed LDF (PLDF)......................................................................................................... 120
2.9.1.1.
Digital Planimeter .......................................................................................................................... 121
2.9.2.
Moor LDF (MLDF) ........................................................................................................... 121
2.10. SYMPATHETIC NERVE BLOCKADE ................................................................................................ 122
2.11. WOUND HEALING ........................................................................................................................ 122
2.11.1. Induction of thermal burn................................................................................................... 122
2.11.2. Drug treatments.................................................................................................................. 123
2.11.3. Measurement of healing ...................................................................................................... 123
2.12. PROTEIN ASSAY TECHNIQUE ......................................................................................................... 124
2.13. STATISTICAL METHODS OF ANALYSES .......................................................................................... 124
2.14. LIMITATIONS OF THIS STUDY ........................................................................................................ 125
CHAPTER 3: DIABETES AND SENSORY NERVE FUNCTION..................................................... 127
3.1. NEUROGENIC INFLAMMATION: PART 1 (POST-TERMINAL EVENTS).............................................. 128
3.1.1.
Introduction ........................................................................................................................ 128
3.1.2.
Method................................................................................................................................ 131
3.1.2.1.
Expression of data and statistical analyses ..................................................................................... 133
3.1.3.
Results................................................................................................................................. 133
3.1.4.
Discussion........................................................................................................................... 147
3.2. NEUROGENIC INFLAMMATION: PART 2 (PRE-TERMINAL AND POST-TERMINAL EVENTS) ............. 153
3.2.1.
Introduction ........................................................................................................................ 153
3.2.2.
Method................................................................................................................................ 154
3.2.2.1.
3.2.3.
3.2.4.
Expression of data and statistical analyses ..................................................................................... 155
Results................................................................................................................................. 155
Discussion........................................................................................................................... 158
CHAPTER 4: DIABETES AND MODULATION OF MICROVASCULAR BLOOD FLOW: ROLE
OF LOCAL TISSUE FACTORS AND PARASYMPATHETIC NEUROTRANSMITTERS.......... 161
4.1.
4.2.
INTRODUCTION ............................................................................................................................ 162
METHOD ...................................................................................................................................... 165
14
4.2.1.
Expression of data and statistical analyses ........................................................................ 166
4.3. RESULTS ...................................................................................................................................... 166
4.4. DISCUSSION ................................................................................................................................. 174
CHAPTER 5: DIABETES AND MODULATION OF MICROVASCULAR BLOOD FLOW: ROLE
OF SYMPATHETIC NERVES…………………………………………. .............................................. 179
5.1. INTRODUCTION ............................................................................................................................ 180
5.2. METHOD ...................................................................................................................................... 181
5.2.1.
Expression of data and statistical analyses ........................................................................ 182
5.3. RESULTS ...................................................................................................................................... 182
5.4. DISCUSSION ................................................................................................................................. 186
CHAPTER 6: DIABETES AND MODULATION OF MICROVASCULAR BLOOD FLOW: ROLE
OF ENDOTHELIAL-DERIVED CONSTRICTOR FACTORS AND PROTEIN KINASE C......... 190
6.1. INTRODUCTION ............................................................................................................................ 191
6.2. METHOD ...................................................................................................................................... 193
6.2.1.
Expression of data and statistical analyses ........................................................................ 194
6.3. RESULTS ...................................................................................................................................... 195
6.4. DISCUSSION ................................................................................................................................. 202
CHAPTER 7: DIABETES AND MODULATION OF MICROVASCULAR BLOOD FLOW:. ROLE
OF REACTIVE OXYGEN SPECIES (ROS)………………………………….. ................................... 209
7.1. INTRODUCTION ............................................................................................................................ 210
7.2. METHOD ...................................................................................................................................... 211
7.2.1.
Expression of data and statistical analyses ........................................................................ 212
7.3. RESULTS ...................................................................................................................................... 213
7.4. DISCUSSION ................................................................................................................................. 220
CHAPTER 8: DIABETES AND MODULATION OF MICROVASCULAR BLOOD FLOW: ROLE
OF ADVANCED GLYCATED END-PRODUCTS (AGES) (SHORT- AND LONG-TERM
EFECTS)……………………………………………………………........................................................ 227
8.1. INTRODUCTION ............................................................................................................................ 228
8.2. METHOD ...................................................................................................................................... 229
8.2.1.
Expression of data and statistical analyses ........................................................................ 230
8.3. RESULTS ...................................................................................................................................... 231
8.4. DISCUSSION ................................................................................................................................. 238
CHAPTER 9: HYPERGLYCAEMIA AND CHANGES IN MICROVASCULAR BLOOD FLOW:
ROLE OF INSULIN…………………………………………………………………… ......................... 243
9.1. INTRODUCTION ............................................................................................................................ 244
9.2. METHOD ...................................................................................................................................... 245
9.2.1.
Expression of data and statistical analyses ........................................................................ 246
9.3. RESULTS ...................................................................................................................................... 246
9.4. DISCUSSION ................................................................................................................................. 252
CHAPTER 10: WOUND HEALING IN DIABETIC RATS ................................................................ 257
10.1. INTRODUCTION ............................................................................................................................ 258
10.2. METHOD ...................................................................................................................................... 260
10.2.1. Expression of data and statistical analyses ........................................................................ 261
10.3. RESULTS ...................................................................................................................................... 262
10.4. DISCUSSION ................................................................................................................................. 271
CHAPTER 11: GENERAL DISCUSSION ............................................................................................ 284
11.1. GENERAL OVERVIEW AND METHODOLOGY ................................................................................. 285
11.2. NEURONAL AND LOCAL TISSUE FACTORS MODULATING MICROVASCULAR BLOOD FLOW IN EARLY
DIABETES ................................................................................................................................................ 287
15
11.3. ENDOTHELIAL-MEDIATED FACTORS MODULATING MICROVASCULAR BLOOD FLOW IN EARLY
DIABETES……………………………………………………………………….. ................................... 293
11.4. FACTORS INVOLVED IN TISSUE REPAIR PROCESS .......................................................................... 302
11.5. CONCLUDING REMARKS AND FUTURE DIRECTIONS ...................................................................... 306
CHAPTER 12: BIBLIOGRAPHY.......................................................................................................... 308
APPENDICES .......................................................................................................................................... 354
16
LIST OF TABLES
Table 1-1: Epidemiologically identified detriments of type I diabetes (Adopted from Pickup, J.C. &
Williams, G. (1997)) __________________________________________________________ 4
Table 1-2: Clinical features associated with diabetic autonomic neuropathy (Adopted from Davidson
, M.B. (1998) & Ellenberg, M. /Rifkin, H. (1983)) __________________________________ 13
Table 1-3: Some of the consequences of AGE formation and deposition (Adopted from Bierhaus, A. et al
(1998) & Vlassara, H. et al (1994))______________________________________________ 20
Table 1-4: Effects of AG (Adopted from Bierhaus, A. et al (1998) & Friedman, E.A. (1999)) __________ 27
Table 1-5: A number of ET-1’s biological effects (Adopted from Ruffolo, R.R. Jr.(1995)). ____________ 72
Table 2-1: Major differences between the content of Mouse breeder and GR2. ____________________ 111
Table 3-1: Average weight of control and diabetic rats at week zero and week ten. ________________ 138
Table 3-2: Average weight of STZ- and alloxan-treated rats at week zero and week four. ___________ 144
Table 4-1: Vascular responses to ACh (100μM) in control and diabetic rats (4 weeks) measured by the
PLDF and the MLDF, presented as percent above basal ± SEM (n = 6). ________________ 169
Table 8-1: Vascular responses to SP (1μM) and SNP (100μM) in control and diabetic rats before and
after amadori (1mg/ml), AGEs (10mg/ml), and anti-RAGE IgG (100μg/ml) (n = 4-10). Data
are presented as mean ± SEM and their 95% CI.__________________________________ 232
Table 8-2: Vascular responses to SNP (100μM), SP (1μM), CGRP (1μM), and ES (20V, 5Hz, 2msec,
for 1 minute) in diabetic rats treated with AG (25mg/kg, i.p) (2 or 4 weeks) (n = 4-10). Data
are presented as mean ± SEM and their 95% CI. __________________________________ 235
Table A-1: The basal blood flow (BBF) and vascular responses to SNP (100μM), SP, CGRP(both 1μM)
and ES (20V, 15Hz, 2 mSec, for 1 min) in control rats over time (n = 6-9). Data are
presented as mean ± SEM and their 95% confidence intervals (95%CI). ______________ _ 355
Table B-1: PE in diabetic rats over time (n = 6-9). Data are presented as mean ± SEM and their 95%
confidence intervals (95% CI). ________________________________________________ 355
Table B-2: PE in control and diabetic rats after {SP (1μM), SP (1μM) + L-NAME (100μM)}, control
and diabetic rats after {BK (10μM), BK (10μM) + L-NAME (100μM}, control rats after
Nicotine (100μM) or Muscarine chloride (100μM), and diabetic rats after Nicotine
(100μM) orMuscarine chloride (100μM) (n = 6-9). Data are presented as mean ± SEM
and their 95% confidence intervals (95% CI).
C=Control D= Diabetic______________ 356
Table B-3: PE in control and diabetic rats before and after guanethidine treatment (50mg/kg, i.p)
(n = 4-10). Data are presented as mean ± SEM and their 95% confidence intervals
(95% CI). N = Guanethidine non-treated T = Guanethidine-treated __________________ 356
Table B-4: PE in control, diabetic (Diab), control and diabetic rats after perfusing BQ-123, BQ-788
(both at 10μM), and diabetic rats after perfusing BIM (1μM) (n = 4-10). Data are presented
as mean ± SEM and their 95% confidence intervals (95% CI). ______________________ 357
Table B-5: PE in control, diabetic (Diab), diabetic rats after perfucing SOD (100U/ml), NAC (100μM),
BQ-123 (10μM), combined [BQ-123 (10μM) + SOD (100U/ml)], and diabetic rats treated
with Tirilazad mesylate (5mg/kg/i.p)(n = 4-10). Data are presented as mean ± SEM and their
95%
confidence intervals (95% CI). ________________________________________________ 357
Table B-6: PE in control, diabetic (Diab), diabetic rats after perfusing amadori (1mg/ml), AGE (10mg/ml),
anti-RAGE (100μg/ml), and diabetic rats treated with AG (25mg/kg, i.p) (2 or 4 weeks).
Data are presented as mean ± SEM and their 95% confidence intervals (95% CI)._______ 358
Table B-7: PE in control, diabetic (Diab) (2 &4 weeks), and diabetic rats (4 weeks) treated with
insulin (2-6IU/day) for specific periods of time (late and early interventions)(n = 4-10).
Data are presented as mean ± SEM and their 95% confidence intervals (95% CI).________ 358
17
LIST OF FIGURES
Fig 1-1: Glucose metabolism and the polyol pathway. (Adopted from Pickup, J.C. & Williams, G (1997)).
___________________________________________________________________________________ 16
Fig 1-2: Scheme of AGEs formation in the “Maillard” reaction (Adopted from Bierhaus, A. et al (1998)).
___________________________________________________________________________________ 18
Fig 1-3: The chemical structure of aminoguanidine (Adopted from Vlassara, H. et al (1994)) _________ 26
Fig 1-4: Formation of oxygen derived radicals (Adopted from Gryglewski, R.J. et al (1986), Munzel,
T. et al (1997), and Kehrer, J.P. (1993)) ___________________________________________ 29
Fig 1-5: Structure of alloxan (Adopted from Lenzen, S. & Munday, R. (1991)). ____________________ 47
Fig 1-6: Structure of Streptozotocin (adopted from Weiss, R.B. (1982)). _________________________ 48
Fig 1-7: Schematic representation of post-synaptic receptors for sympathetic neurotransmitter NA and
their mechanism of action. Please note that in order to keep the diagram simple NPY and ATP
are not included (Adopted from Bennett, T. & Gardiner, S.M. (1996)).____________________ 64
Fig 1-8: Schematic representation illustrating locations and selective ligands for endothlin receptor
subtypes (Adopted from Ruffolo, R.R.. Jr (1995)). ____________________________________ 71
Fig 1-9: Intracellular signal transduction pathways activated by ET-1 in vascular smooth muscle cells
(Adopted from Reid, J.J. & La, M. (1995)). _________________________________________ 73
Fig 1-10: Effects of high glucose on factors related to ET-1 production and action (Adopted from Hopfner,
R.L. et al (1999)). ____________________________________________________________ 75
Fig 1-11: Receptor activation of vascular endothelium by ACh or BK (Adopted from Moncada, S. &
Higgs, A. (1993)). ____________________________________________________________ 80
Fig 1-12: Bradykinin synthesis, vascular effects and degradation (Adopted from Dray, A. & Perkins, M.
(1993)). ____________________________________________________________________ 83
Fig 1-13: Possible mechanisms involved in insulin secretion by glucose (Adapted from Pickup, J.C. &
Williams, G. (1997)). __________________________________________________________ 85
Fig 1-14: Effects of insulin on factors related to ET-1 production and action (Adopted from Hopfner, R.L.
et al (1999)). ________________________________________________________________ 88
Fig 2-1: A schematic representation of an experiment set up for electrical stimulation of the sciatic nerve._
__________________________________________________________________________________ 118
Fig 3-1: BBF in rats before and after induction of diabetes (changes over time), presented as mean ± SEM
(n = 6-9). ___________________________________________________________________ 134
Fig 3-2: Vascular responses to SNP (100 μM), SP, and CGRP (both at 1μM), presented as percent above
basal ± SEM (n = 6-9). _______________________________________________________ 135
Fig 3-3: Vascular responses to SNP (100 μM), SP, and CGRP (both at1μM), presented as mean ± SEM
(n = 6-9). _________________________________________________________________ 136
Fig 3-4: Changes in the vascular response to SNP (100 μM) in diabetic rats over time, presented as mean
± SEM (n = 6-9) ____________________________________________________________ 137
Fig 3-5: Changes in the vascular response to SP (1μM) in diabetic rats over time, presented as mean ±
SEM (n = 6-9). ______________________________________________________________ 137
Fig 3-6: Changes in the vascular response to CGRP (1μM) in diabetic rats over time presented as mean ±
SEM (n = 6-9). ______________________________________________________________ 138
Fig 3-7: Weight in non-diabetic rats over time, presented as mean ± SEM (n = 6-9) _______________ 139
Fig 3-8: Weight in diabetic rats over time, presented as mean ± SEM (n = 6-9). __________________ 139
Fig 3-9: PE in diabetic rats over time, presented as percent above basal (n = 6-9). ________________ 140
Fig 3-10: Vascular responses to SP (1μM) in control and diabetic rats (2 and 4 weeks) at 10 minute
intervals measured by the PLDF, presented as percent above basal ± SEM (n = 6-9).______ 141
Fig 3-11: Vascular responses to SP (1μM) in control and diabetic rats (2 and 4 weeks) at 10 minute
intervals measured by the MLDF, presented as percent above basal ± SEM (n = 6-9). _____ 141
Fig 3-12: BBFs in control, diabetic, capsaicin pre-treated, and capsaicin pre-treated diabetic rats,
presented as percent of control ± SEM (n = 6). ____________________________________ 142
Fig 3-13: Vascular responses to SNP (100 μM) in control, diabetic, capsaicin pre-treated, and capsaicin
pre-treated diabetic rats, presented as percent of control ± SEM (n = 6).________________ 143
18
Fig 3-14: Vascular responses to SP (1μM) in control, diabetic, capsaicin pre-treated, and capsaicin pretreated diabetic rats, presented as percent of control ± SEM (n = 6). ___________________ 144
Fig 3-15: BBFs in control, STZ-treated, and alloxan-treated diabetic rats, presented as mean ± SEM
(n = 6). ___________________________________________________________________ 145
Fig 3-16: Vascular responses to SNP (100 μM) in control, STZ-treated, and alloxan-treated diabetic rats,
presented as mean ± SEM (n = 6). _____________________________________________ 145
Fig 3-17: Vascular responses to SP (1μM) in control, STZ-treated, and alloxan-treated diabetic rats,
presented as mean ± SEM (n = 6). _____________________________________________ 146
Fig 3-18: Vascular responses to CGRP (1μM) in control, STZ-treated, and alloxan-treated diabetic rats,
presented as mean ± SEM (n = 6). _____________________________________________ 146
Fig 3-19: Changes in the vascular response to SNP (100 μM) in diabetic rats over time, presented as mean
± SEM (n = 6-9). ___________________________________________________________ 156
Fig 3-20: Changes in the vascular response to ES (20V, 5Hz, 2msec for 1 minute) in diabetic rats over
time, presented as mean ± SEM (n = 6-9). _______________________________________ 156
Fig 3-21: Vascular responses to ES (20V, 5Hz, 2msec for 1 minute) in control, diabetic, capsaicin pretreated, and capsaicin pre-treated diabetic rats, presented as percent of control ± SEM (n = 6).
__________________________________________________________________________________ 157
Fig 3-22: Comparing the response to ES (20V, 5Hz, 2msec for 1 minute) in control, STZ-treated, and
alloxan-treated diabetic rats, presented as mean ± SEM (n = 6). ______________________ 158
Fig 4-1: Vascular responses to SNP (100μM) in control and diabetic rats, presented as mean ± SEM
(n = 6-7). __________________________________________________________________ 167
Fig 4-2: Vascular responses to ACh, Nicotine, and Muscarine chloride (all 100μM) in control rats,
presented as mean ± SEM (n = 6-7). _____________________________________________ 167
Fig 4-3: Vascular responses to ACh, Nicotine, and Muscarine chloride (all 100μM) in diabetic (Diab)
rats,presented as mean ± SEM (n = 6-7).__________________________________________ 168
Fig 4-4: Response profile to ACh (100μM, 30 min) in control rat skin microvasculature, as recorded
by the PLDF. ________________________________________________________________ 168
Fig 4-5: Response profile to SNP (100μM, 10 min) and ACh (100μM, 30 min) in control rat skin
microvasculature, as recorded by the MLDF. ______________________________________ 169
Fig 4-6: Vascular responses to ACh (100μM) in control and diabetic (Diab) rats (4 weeks) at 10 minutes
intervals measured by the PLDF, presented as percent above basal ± SEM (n = 6). ________ 170
Fig 4-7: Vascular responses to ACh (100μM) in control and diabetic (Diab) rats (4 weeks) at 10 minutes
intervals measured by the MLDF, presented as percent above basal ± SEM (n = 6). ________ 170
Fig 4-8: Vascular responses to SNP (100μM) in control and diabetic rats, presented as mean ± SEM
(n = 6-9). __________________________________________________________________ 171
Fig 4-9: Vascular responses to BK (10μM) in control and diabetic rats, before and after L-NAME (100μM)
perfusion, presented as percent of control ± SEM (n = 6-9). ___________________________ 172
Fig 4-10: Vascular responses to SP (1μM) in control and diabetic rats, before and after L-NAME (100μM)
perfusion, presented as percent of control ± SEM (n = 6-9). __________________________ 172
Fig 4-11: PE in control {SP (1μM), BK (10μM)}, diabetic (Diab) {SP (1μM), BK (10μM)}, control after
Nicotine (100μM) or Muscarine chloride (100μM), and diabetic rats after Nicotine (100μM) or
Muscarine chloride (100μM), presented as percent above basal (n = 6-9). _______________ 173
Fig 5-1: BBFs in control and diabetic rats before and after guanethidine treatment (50mg/kg, i.p),
presented as mean ± SEM (n = 4-10). ____________________________________________ 183
Fig 5-2: Vascular responses to SNP (100μM) in control and diabetic rats before and after guanethidine
treatment (50mg/kg, i.p), presented as mean ± SEM (n = 4-10). ________________________ 183
Fig 5-3: Vascular responses to SP (1μM) in control and diabetic rats before and after guanethidine
treatment (50mg/kg, i.p), presented as mean ± SEM (n = 4-10). _______________________ 184
Fig 5-4: Vascular responses to ES (20V, 5Hz, 2msec, for 1 minute) in control and diabetic rats before and
after guanethidine treatment (50mg/kg, i.p), presented as mean ± SEM (n = 4-8). __________ 185
Fig 5-5: PE in control (C) and diabetic (D) rats before and after guanethidine treatment (50mg/kg, i.p),
19
presented as percent above basal (n = 4-10) ______________________________________ 186
Fig 6-1: BBF in control rats before and after BQ-123 and BQ-788 (both at 10μM), presented as mean ±
SEM (n = 4-9). ______________________________________________________________ 195
Fig 6-2: BBF in diabetic (Diab) rats before and after BQ-123 and BQ-788 (both at 10μM), presented as
mean ± SEM (n = 4-10)._______________________________________________________ 195
Fig 6-3: The vascular response to SNP (100μM) in control rats before and after BQ-123 and BQ-788
(both at 10μM), presented as mean ± SEM (n = 8-12).________________________________ 196
Fig 6-4: The vascular responses to SP (1μM) in control rats before and after BQ-123 and BQ-788 (both
at10μM), presented as mean ± SEM (n = 4-9). _____________________________________ 196
Fig 6-5: The vascular response to SNP (100μM) in diabetic (Diab) rats before and after BQ-123 and
BQ-788 (both at 10μM), presented as mean ± SEM (n = 8-10). _________________________ 197
Fig 6-6: The vascular response to SP (1μM) in diabetic (Diab) rats before and after BQ-123 and BQ-788
(both at 10μM), presented as mean ± SEM (n = 4-10).________________________________ 198
Fig 6-7: The vascular response to ES (20V, 5Hz, 2msec, for I minute) in control rats before and after
BQ-123 and BQ-788 (both at 10μM), presented as mean ± SEM (n = 8-12)._______________ 199
Fig 6-8: The vascular response to ES (20V, 5Hz, 2msec, for I minute) in diabetic (Diab) rats before and
after BQ-123 and BQ-788 (both at 10μM), presented as mean ± SEM (n = 8-10). __________ 199
Fig 6-9: BBFs in control, control after BIM, diabetic (Diab), and diabetic rats after BIM (1μM),
presented as mean ± SEM (n = 5). _______________________________________________ 200
Fig 6-10: Vascular responses to SNP (100μM) in control, control after BIM, diabetic (Diab), and diabetic
rat after BIM (1μM), presented as mean ± SEM (n = 5). _____________________________ 201
Fig 6-11: Vascular responses to SP (1μM) in control, control after BIM, diabetic (Diab), and diabetic
rats after BIM (1μM), presented as mean ± SEM (n = 5). ____________________________ 201
Fig 6-12: PE in control, diabetic, and diabetic rats before and after BQ-123, BQ-788 (both at 10μM),
and BIM (1μM), presented as percent above basal (n = 4-10). ________________________ 202
Fig 7-1: BBF in control rat, and control rats after SOD (100U/ml), NAC (100μM), and combined
[BQ-123 (10μM) + SOD (100U/ml)], presented as mean ± SEM (n = 4-10). ______________ 213
Fig 7-2: BBFs in diabetic (Diab) rats, diabetic rats after SOD (100U/ml), NAC (100μM), combined
[BQ-123 (10μM) + SOD (100U/ml)], and diabetic rats treated with Tirilazad mesylate
(5mg/kg/i.p),
presented as mean ± SEM (n = 4-10). _____________________________________________ 214
Fig 7-3: The vascular response to SNP (100μM) in control rats and control rats after SOD (100U/ml),
NAC (100μM), and combined [BQ-123 (10μM) + SOD (100U/ml)], presented as mean ± SEM
(n = 4-10). _________________________________________________________________ 215
Fig 7-4: The vascular response to SNP (100μM) in diabetic (Diab) rats , diabetic rats after SOD
(100U/ml),
NAC (100μM), combined [BQ-123 (10μM) + SOD (100U/ml)], and diabetic rats treated with
Tirilazad mesylate (5mg/kg/i.p), presented as mean ± SEM (n = 4-10). ___________________ 215
Fig 7-5: The vascular response to SP (1μM) in control rats and control rats after SOD (100U/ml), NAC
(100μM), and combined BQ-123 (10μM) + SOD (100U/ml), presented as mean ± SEM
(n = 4-10). _________________________________________________________________ 216
Fig 7-6: The vascular response to SP (1μM) in diabetic (Diab) rats, diabetic rats after SOD (100U/ml),
NAC (100μM), combined [BQ-123 (10μM) + SOD (100U/ml)], and diabetic rats treated with
Tirilazad mesylate (5mg/kg/i.p), presented as mean ± SEM (n = 4-10). __________________ 217
Fig 7-7: The vascular response to CGRP (1μM) in control, diabetic (Diab), and diabetic rats treated with
Tirilazad mesylate (5mg/kg/i.p), presented as mean ± SEM (n = 4-10). ___________________ 218
Fig 7-8: The vascular response to ES (20V, 5 Hz, 2 msec for 1 minute) in control rats and control rats after
SOD (100U/ml), NAC (100μM), and combined [BQ-123 (10μM) + SOD (100U/ml)], presented
as mean ± SEM (n = 5-10). ____________________________________________________ 219
Fig 7-9: The vascular response to ES (20V, 5 Hz, 2 msec for 1 minute) in diabetic (Diab) rats, diabetic rats
after SOD (100U/ml), NAC (100μM), combined [BQ-123 (10μM) + SOD (100U/ml), and diabetic
20
rats treated with Tirilazad mesylate (5mg/kg/i.p), presented as mean ± SEM (n = 5-10)._____ 219
Fig 7-10: PE in control rats, diabetic rats, diabetic rats after SOD (100U/ml), NAC (100μM), combined
[BQ-123 (10μM) + SOD (100U/ml)], and diabetic rats pre-treated with Tirilazad mesylate
(5mg/kg/i.p), presented as percent above basal (n = 4-10). ___________________________ 220
Fig 8-1: BBFs in control rats and control rats after amadori (1mg/ml), AGEs (10mg/ml), and anti-RAGE
IgG (100μg/ml), presented as mean ± SEM (n = 4-8).________________________________ 231
Fig 8-2: BBFs in diabetic (Diab) rats, diabetic rats after amadori (1mg/ml), AGEs (10mg/ml), and antiRAGE IgG (100μg/ml), and diabetic rats treated with AG (25mg/kg, i.p) (2 or 4 weeks), presented
as mean ± SEM (n = 4-10). ____________________________________________________ 232
Fig 8-3: The vascular response to SNP (100μM) in control rats and control rats after amadori (1mg/ml),
AGEs (10mg/ml), and anti-RAGE IgG (100μg/ml), presented as mean ± SEM (n = 4-8). _____ 233
Fig 8-4: The vascular response to SP (1μM) in control rats and control rats after amadori (1mg/ml), AGEs
(10mg/ml), and anti-RAGE IgG (100μg/ml), presented as mean ± SEM (n = 4-8).__________ 233
Fig 8-5: The vascular response to SNP (100μM) in diabetic (Diab) rats, diabetic rats after amadori
(1mg/ml), AGEs (10mg/ml), anti-RAGE IgG (100μg/ml), and diabetic rats treated with AG
(25mg/kg, i.p) (2 or 4 weeks), presented as mean ± SEM (n = 4-10). ____________________ 234
Fig 8-6: The vascular response to SP (1μM) in diabetic (Diab) rats, diabetic rats after amadori (1mg/ml),
AGEs (10mg/ml), anti-RAGE IgG (100μg/ml), and diabetic rats treated with AG (25mg/kg, i.p)
(2 or 4 weeks), presented as mean ± SEM (n = 4-10). ________________________________ 234
Fig 8-7: The vascular response to CGRP (1μM) in control, diabetic (Diab), and diabetic rats treated with
AG (25mg/kg, i.p) (2 or4 weeks), presented as mean ± SEM (n = 4-10). __________________ 236
Fig 8-8: The vascular response to ES (20V, 5Hz, 2msec for 1 minute) in control, diabetic (Diab), and
diabetic rats treated with AG (25mg/kg, i.p) (2 or 4 weeks), presented as mean ± SEM
(n = 4-8). ___________________________________________________________________ 236
Fig 8-9: PE in control rats, diabetic rats, diabetic rats after amadori (1mg/ml), AGEs (10mg/ml, antiRAGE IgG (100μg/ml), and diabetic rats treated with AG (25mg/kg, i.p) (2 or 4 weeks), presented
as percent above basal (n = 4-10). _______________________________________________ 237
Fig 9-1: BBFs in control rats, diabetic (Diab) (2 & 4 weeks) rats, and diabetic rats (4 weeks) treated
with insulin (2-6IU/100g/day) (late and early interventions), presented as mean ± SEM
(n = 4-10). __________________________________________________________________ 247
Fig 9-2: Weigh in control rats, diabetic (Diab) (2 & 4 weeks) rats, and diabetic rats (4 weeks)
treated with insulin (2-6IU/100g/day) (late and early interventions), presented as mean ± SEM
(n = 4-7). __________________________________________________________________ 248
Fig 9-3: Vascular responses to SNP (100μM) in control rats, diabetic (Diab) (2 & 4 weeks) rats, and
diabetic rats (4 weeks) treated with insulin (2-6IU/100g/day) (late and early interventions)
(n = 4-10). _________________________________________________________________ 248
Fig 9-4: Vascular responses to SP (1μM) in control rats, diabetic (Diab) (2 & 4 weeks) rats, and diabetic
rats (4 weeks) treated with insulin (2-6IU/100g/day) (late and early interventions), presented as
mean ± SEM (n = 4-10). _______________________________________________________ 249
Fig 9-5: Vascular responses to CGRP (1μM) in control rats, diabetic (Diab) (2 & 4 weeks) rats, and
diabetic rats (4 weeks) treated with insulin (2-6IU/100g/day) (late and early interventions),
presented as mean ± SEM (n = 4-6). _____________________________________________ 250
Fig 9-6: Vascular responses to ES (20V, 5 Hz, 2 msec, for 1 minute) in control rats, diabetic (Diab)
(2 & 4 weeks) rats, and diabetic rats (4 weeks) treated with insulin (2-6IU/100g/day)(late and
early interventions), presented as mean ± SEM (n = 4-8). _____________________________ 251
Fig 9-7: PE in control rats, diabetic (Diab)(2 &4 weeks) rats, and diabetic rats (4 weeks) treated with
insulin (2-6IU/100g/day) (late and early interventions), presented as percent above basal
(n = 4-10). __________________________________________________________________ 252
Fig 10-1: Wound healing profile in control, diabetic (Diab) non-treated, and diabetic rats treated with
insulin, AG, Tirilazad mesylate, and combined (BQ-123 + SOD), presented as mean ± SEM
(n = 5- 6). _________________________________________________________________ 262
21
Fig 10-2: Wound healing profile in control and diabetic (Diab) rats (both non-treated), presented as mean
± SEM (n = 6). _____________________________________________________________ 264
Fig 10-3: Wound healing profile in control, diabetic (Diab), and insulin (2-6IU/100g)-treated diabetic rats,
presented as mean ± SEM (n = 5). ______________________________________________ 265
Fig 10-4: Wound healing profile in control, diabetic (Diab), and AG (25mg/kg)-treated diabetic rats,
presented as mean ± SEM (n = 6). ______________________________________________ 266
Fig 10-5: Wound healing profile in control, diabetic (Diab), and Tirilazad mesylate (5 mg/kg)-treated
diabetic rats, presented as mean ± SEM (n = 6). __________________________________ 267
Fig 10-6: Wound healing profile in control, diabetic (Diab), and combined (BQ-123 + SOD) (100μM +
100U)- treated diabetic rats, presented as mean ± SEM (n = 6)._______________________ 268
Fig 10-7: Wound healing profile in diabetic (Diab) rats treated with insulin, AG, Tirilazad mesylate, and
combined(BQ-123 + SOD) (same concentrations as before), presented as mean ± SEM (n =5-6).
_________________________________________________________________________________ 270S
22
CHAPTER 1
Literature Review
23
1.1.
Diabetes Mellitus: Historical prospective
Diabetes mellitus (DM) is a disease that was recognized in antiquity. The term “diabetes
mellitus” was first used in the 18th century to distinguish the sweet taste of diabetic urine
from other polyuric states in which the urine was tasteless. The first major breakthrough
in the history of this disease occurred in 1921 when Banting and Best discovered insulin.
They showed that insulin reduced the blood glucose level in diabetics and significantly
improved patient’s lives. Since then, major advances in understanding diabetes have
included: the characterization of the insulin molecule (1955) and elucidating its structure
(1969), the measurement of insulin concentration by radioimmunoassay (RIA) (1959),
the isolation of pro-insulin (1967), the identification of the insulin receptor (1971), and
the sequencing of the insulin receptor gene (1985). Milestones in the management of
diabetes have included: the development of long-acting insulin (Isophane) (1936), the
first therapeutic use of phenformin (1957), the introduction of test strips for measuring
blood glucose, and definitive proof from Diabetes Control and Complications Trial
(DCCT) data (1993) (DCCT, 1993) that strict glycaemic control could slow or prevent
the development of diabetic microvascular complications (Pickup and Williams, 1997).
Diabetes associations, aiming to help diabetics and promote research into it, now exist in
many countries.
1.2.
Characterization and diagnosis
DM is a syndrome with metabolic, vascular, and neuropathic components that are all
interrelated. The metabolic syndrome is characterised by alterations in carbohydrate, fat,
and protein metabolism. These are secondary to absent or markedly diminished insulin
secretion and / or ineffective insulin action.
The classic symptoms of this disease are excessive thirst (polydipsia), tiredness, excess
passing of urine (polyuria), presence of glucose in urine (glucosuria), large food intake
(polyphagia), weight loss, and presence of ketone in urine (ketonuria). Early symptoms of
some individuals may also include itchy rashes, blurred vision, and pins and needles in
hands and feet. These individuals will show high glucose levels in their blood stream
(hyperglycaemia) which can be easily measured.
24
Recently it was shown that the oral glucose tolerance test (OGTT) is not reproducible,
inconvenient, costly, and not often used clinically to diagnose diabetes. The World
Health Organization (WHO) and the American Diabetes Association (ADA) have
therefore, adopted the fasting plasma glucose test (FPG) as a preferred method to make
diagnosis and to judge therapy. Individuals with an FPG of 3.5-5.5 mmol/lit (around 60110mg/dl) are considered as healthy but anything greater than 7mmol/lit (126mg/dl) will
be considered as diabetic (Davidson, 1998).
1.3.
Aetiology
Recent understanding of the pathogenesis and aetiology of the various categories of
diabetes have led to a new classification of this disease by the ADA and WHO
(Davidson, 1998). According to this classification there are 2 major types of diabetes:
-
Type I (previously known as juvenile or insulin-dependent DM) (IDDM)
-
Type II (previously known as adult-onset or non-insulin dependent DM) (NIDDM)
a) Non-obese b) Obese
Diabetes may also occur as a result of malnutrition. Gestational diabetes only appears
during pregnancy in women with no history of type I or type II diabetes and disappears
after childbirth.
The aetiology of other specific types of diabetes is listed below:
- Pancreatic disease
- Endocrine disease
- Genetic defects of β-cell function
- Abnormalities of insulin or its receptor
- Drug or chemical induction
- Infections and uncommon immune defects
- Genetic syndromes
1.3.1. Type I diabetes mellitus
This accounts for only 10-15% of all cases of diabetes. Result from a study of 27
countries showed that the incidence of type I diabetes is on the rise by around 3% per
year (Onkamo et al., 1999). Type I diabetes can be subdivided into a) Autoimmune and
b) Idiopathic form. The autoimmune form is by far the most common and results from a
25
cell-mediated autoimmune destruction of the pancreatic beta cells. The peak incidence
occurs in children and adolescence but individuals up to 30 years of age can develop type
I diabetes. The rate of cell destruction is variable but is more rapid in children than in
adults. The autoimmune destruction of beta cells has a genetic predisposition, which can
be identified by human leucocyte antigen (HLA). Epidemiological studies provide strong
evidence that the emergence of type I diabetes is largely determined by the effect of
adverse environmental factors (Pickup and Williams, 1997) (Table 1-1).
A small proportion of patients (e.g. African Americans) with type I diabetes will fall into
the idiopathic form (Umpierrez et al., 1995). This form of diabetes is strongly inherited,
caused by decreased insulin secretion and lacks immunological evidence for
autoimmunity.
Possible causative
Genetic factors
Possible precipitating
Seasonality / climatological factors
Virus (Coxsackie?)
Life-style?
Psychosociological stress?
HLA-linked
(HLA-DQ factors)
Non-HLA-linked
(Insulin gene?)
Non genetic factors
Virus (Rubella?)
Nutrients (cow’s
Nitrosamines?)
milk
protein?,
Table 1-1: Epidemiologically identified detriments of type I diabetes (Adopted from Pickup, J.C. &
Williams, G. (1997))
1.3.2. Type II diabetes mellitus
Individuals with this type of diabetes have insulin resistance in combination with a
relative deficiency of insulin secretion. Its aetiology is not known and frequently these
patients are undiagnosed for many years because the elevated glucose concentrations are
not high enough to elicit the classic symptoms of uncontrolled diabetes. Since the gap
between the onset of hyperglycaemia and the diagnosis is large (9-12 years), these
patients are at risk of developing microvascular complications and neuropathy during this
period.
Around 80% to 90% of patients with this form of diabetes are obese, which itself adds to
insulin resistance and, therefore, weight loss is a vital part of their treatment. Age, life-
26
style, race, family history, and genetic susceptibility are also risk factors. The insulin
secretion is defective in these patients and the presence of insulin resistance poses an
additional problem. Although with weight loss the insulin resistance will reduce in these
patients, the genetic susceptibility component will remain. Ketoacidosis rarely occurs in
these patients since there is some effective insulin action but it can occur with stress and
illness (Pickup and Williams, 1997).
1.4.
Long-term complications
Long-term complications of DM occur in both forms of the condition in humans and have
an important role in the increased morbidity and mortality suffered by these individuals.
Macrovascular complications include coronary heart disease, atherosclerosis, and
peripheral
vascular
disease.
Microvascular
complications
include
retinopathy,
nephropathy, and neuropathy (Pickup and Williams, 1997). Diabetic neuropathy is the
most common complication of this disease and its manifestations can be divided into two
broad categories, somatic (peripheral) and visceral. Disease of the large and small vessels
results in myocardial infarction, stroke, and gangrene of the lower extremities.
1.4.1. Macrovascular disease
Diabetes is one of the six independent risk factors for developing macrovascular disease
in the general population. The presence of diabetes, in addition to any or all the other 5
risk factors, approximately doubles the chance of developing macrovascular disease and
the available evidence suggests that strict diabetic control does not prevent or delay these
complications. The other 5 potential factors are smoking, hypertension, hyperlipidaemia,
obesity, and genetic factors (Davidson, 1998).
Macrovascular complications or disease of large vessels is due to accelerated
atherosclerosis and includes ischaemic heart disease, cerebrovascular disease and
peripheral vascular disease (Cooper et al., 1997; McMillan, 1997). Hence, diabetic people
are at a high risk of suffering from myocardial infarction, coronary heart failure, stroke,
and ischaemic gangrene (Stehouwer et al., 1997).
27
Mechanisms by which diabetes causes acceleration of atherosclerosis are not clear.
However, a previous study showed that this acceleration does not involve glucose
metabolism and will remain unaffected even if near euglycaemia is achieved (Davidson,
1998). Results from advances in molecular genetics and oxidation chemistry have shown
that genetic variants of lipoprotein lipase and apolipoprotein are suitable candidates for
mediating diabetic vascular risk (Sememkovich and Heinecke, 1997; Mullarkey et al.,
1990). For example, variants of these proteins can produce hypertriglyceridaemia.
Lipoproteins retained in the arterial wall are also subject to oxidative modifications
(Mullarkey et al., 1990), which could be dependent on glycoxidation, the enzyme
myeloperoxides, or the reactive nitrogen species derived from nitric oxide (NO).
Accelerated vascular disease in diabetes is therefore likely to be the result of complex
interactions between metabolic derangement such as hyperglycaemia, mutations in genes
controlling lipid metabolism, and antioxidant defence mechanisms (Sememkovich and
Heinecke, 1997). It was shown that apart from hypertriglyceridaemia, other factors such
as hyperinsulinaemia, increased platelet aggregability, and increased clotting activity of
thrombogenic factors can potentiate atherosclerosis (McMillan, 1997).
The increased incidence of congestive heart failure and the increased mortality and
morbidity in diabetics following myocardial infarction or coronary artery bypass graft
can be explained by the presence of diabetic cardiomyopathy. This is characterized by an
early diastolic dysfunction and a later systolic one, with intracellular retention of calcium
and sodium ions and loss of potassium (Macgoub and Abdel-Fattah, 1998). This
condition is associated with structural, functional, and biochemical abnormalities of the
cardiac cells (Fein and Sonnenblick, 1994; Malhotra and Sanghi, 1997). Cellular changes,
including defects in calcium transport and fatty acid metabolism, may lead to myocellular
hypertrophy and myocardial fibrosis (Bell, 1995). A previous study also showed
mitochondrial respiratory impairment in streptozotocin (STZ)-diabetic rat’s heart (2-14
weeks), which may contribute to cardiac dysfunction (Tomita et al., 1996). It remains
unclear whether changes in the diabetic heart microvasculature, such as increased
basement membrane thickening and increased capillary microaneurysms, could also be
28
significant in the development of altered myocardial functioning (Hardin, 1996; Li,
1992).
Diabetes is associated with cerebrovascular disease featuring non-fatal or small infarction
especially with multiple occurrences. A previous study showed that high blood pressure
and poor blood glucose control are associated with the higher incidence of cerebral
infarctions in diabetic patients (Kameyama et al., 1994). It is also a risk factor for
ischaemic but not haemorrhagic stroke. Impaired autoregulation of cerebral blood flow,
impaired cerebrovascular reactivity, and damage to large and small extra- and intracranial cerebral vessels have been found in humans and animals with diabetes (Cipolla et
al., 1997; Mankovsky et al., 1996). In one study, it was shown that acute glucose
exposure dilates posterior cerebral arteries with intrinsic tone and impairs cerebrovascular
reactivity to transmural pressure via an endothelium-mediated mechanism that involves
NO and prostaglandins (Cipolla et al., 1997).
Peripheral vascular disease (PVD) in diabetes characteristically affects the arterioles
below the knee, such as the tibial and peroneal arteries and their branches. Such vessels
are also commonly affected by atherosclerosis and may display capillary basement
membrane thickening. Narrowing of the arteriolar lumen often leads to tissue ischaemia
causing ulceration and gangrene of the lower extremities of the diabetic and ultimately
amputation of the affected area (Armstrong and Lavery, 1998; Ellenberg and Rifkin,
1983). There are multiple risk factors involved in the development of diabetic PVD and
no universal agreement as to which is the most important. These are blood pressure,
smoking, age, duration of diabetes, blood glucose, and lipids (e.g. cholesterol,
triglycerides) (Cooper et al., 1997).
1.4.2. Microvascular disease
Important functional changes associated with microvascular complications are increased
organ blood flow, increased vascular permeability, abnormal blood viscosity, and
abnormal platelet and endothelial function (Stehouwer et al., 1997). Increased
microvascular flow and pressure results in injury to the endothelium, which can in turn
cause adaptive microvascular sclerosis. This contributes to a loss of vasodilatory reserve
29
and autoregulatory capacity with increasing disease duration (Took, 1996). The structural
hallmark of diabetic microangiopathy is thickening of the capillary basement membrane
(McMillan, 1997). These changes may lead to occlusive angiopathy and tissue hypoxia
and damage (Dahl-Jorgensen, 1998).
The major microvascular complications of diabetes are retinopathy, nephropathy, and
neuropathy. Mechanisms by which diabetic microangiopathy develop are not clear, but
may include genetic influences. Several biochemical changes are involved in causing
these complications, one important change being increased protein glycation (Wautier
and Guillausseau, 1998) (section 1.5.2). Reactive oxygen species (ROS) or free radicals
are also involved in causing tissue damage in diabetes (Giugliano et al., 1996). ROS
generation is enhanced by non-enzymatic glycation and by increased activity of the
polyol pathway, while the ability to neutralize free radicals is reduced in diabetic tissues
(Cameron et al., 1992; Mak et al., 1996) (section 1.5.3). A role for endothelin (ET) in
diabetic microvascular complications (Deng et al., 1999) has also attracted a great interest
(section 1.8.4).
All these studies and those mentioned in the following chapters have enriched the
knowledge base, to different degrees, of what factors may contribute to diabetic vascular
complications. Delayed skin wound repair is one of the major complications of diabetes,
especially among the older population. Adequate microvascular blood flow is an essential
prerequisite for tissue repair and it is known that sensory nerves play an important role in
modulating microvascular blood flow (section 1.7.2). Previous studies have indicated that
the function of sensory nerves is compromised at the early stage of diabetes, leading to
changes in skin microvascular blood flow (Gyorfi et al., 1996; Rendell et al., 1993).
Previous studies have also shown that factors such as ETs, free radicals, advanced
glycated end-products (AGEs) and hyperglycaemia were involved in long-term vascular
complications of diabetes mellitus (sections 1.8.4, 1.5.3, 1.5.2 & 1.8.5.3).
It is the aim of this PhD study to provide information about the mechanisms underlying
changes in skin microvascular blood flow in early diabetes and its effect on tissue repair
30
processes. First, the involvement of ETs, free radicals, AGEs, and hyperglycaemia in
modulating skin microvascular blood flow in 4 weeks streptozotocin (STZ)-induced
diabetic rats was established. Then, by using specific inhibitors / antagonists, the
possibility of preventing the deleterious effects of these factors and improving
microvascular blood flow and tissue repair processes was examined.
1.4.2.1.
Retinopathy
Diabetic retinopathy and blindness from diabetes became a significant problem after the
discovery of insulin in 1922 due to the increase in the life expectancy of diabetics.
Diabetes causes various changes in the retinal circulation, causing two forms of
retinopathy: non-proliferative and the most severe proliferative retinopathy. Nonproliferative retinopathy is recognised by the formation of microaneurysms in the retinal
capillaries, possibly due to slight or moderate dilatation of venules. A previous study in
diabetic rats (3-4 weeks) showed an increase in the production of vasodilatory
prostaglandins (E and I2) in the rat retinal vasculature (Johnson et al., 1999), which may
contribute to abnormal retinal haemodynamics in early diabetic retinopathy. However,
another study provided direct evidence to show that an increase in diacylglycerol (DAG)
and subsequent protein kinase C (PKC) activation are the primary biochemical sequel
responsible for the development of this abnormality in diabetic (2 weeks) rats (Bursell et
al., 1997).
The presence of small retinal haemorrhages from which erythrocytes and serious fluid
may leak is also common; serum leakage or oedema results in formation of hard exudates
(Ellenberg and Rifkin, 1983; Nathan, 1993). Such disturbances do not usually lead to loss
of vision. However, if serum leaks from defective microvessels near the maculae, visual
acuity may be affected (Nathan, 1993). Proliferative retinopathy follows from the
development of increasingly severe non-proliferative retinopathy. Initially, what are
known as ‘cotton wool spots’ develop, which represent the infarction and ischaemia of
retinal tissue. Retinal vessel ischaemia leads to retinal neovascularization and glial
proliferation; the newly formed vessels are particularly fragile and prone to
haemorrhaging. If haemorrhaging occurs into the vitreous cavity, loss of vision will be
the result (Ellenberg and Rifkin, 1983; Nathan, 1993). Endothelin-1 (ET-1) is a potent
31
vasoconstrictor of the porcine ophthalmic microcirculation and, thus, may play an
important role in the physiological control of ocular perfusion and diabetic retinopathy
(Ruffolo, 1995). It was shown that ETs might alter extracellular matrix protein’s gene
expression in the retina leading to the development of structural changes (Deng et al.,
1999).
1.4.2.2.
Nephropathy
Diabetic nephropathy occurs in approximately one third of both types of diabetics
(O'Bryan and Hostetter, 1997) and can be divided into 5 stages. Stage I occurs at the
onset of the disease and is characterised by 30%-40% increase in the glomerular filtration
rate (GFR) above normal. Other changes are increase in kidney size, increased
glomerular diameter, and tubular size (Marshall, 1998; Nathan, 1993). An increase in
GFR is mainly driven by increases in plasma flow and glomerular capillary pressure,
which may cause a transient increase in albumin excretion. Renal blood flow can be
reduced by ET-1 resulting in the reduction of GFR. It was suggested that the alteration in
ET activity in diabetic kidney might contribute to the increase in GFR (Lam et al., 1995).
Elevation in the capillary pressure could damage glomerular endothelial, epithelial, and
mesangial cells. Many factors have been suggested to be involved in increasing the
capillary pressure such as alterations in the endogenous level of atrial natriuretic peptide,
endothelial-derived
relaxing
factor
(EDRF),
angiotensin
II,
prostaglandins,
thromboxanes, and kinins among others (Anderson and Vora, 1995). PKC activity is
altered in the kidney of 4 weeks diabetic rats (Kang et al., 1999). There is also evidence
to suggest that PKC may regulate renal haemodynamics by increasing or decreasing NO
production (Nishio and Watanabe, 1996) and this may be important in kidney vascular
blood flow.
Ultrastructural thickening of the glomerular basement is not present at the time of onset
of insulin dependence and occurs over time (Ellenberg and Rifkin, 1983). The messenger
ribonucleic acid (mRNA) level of extracellular matrix components such as glomerular
alpha 1 (I), (III), and (IV) collagen, laminin B1 and B2, and growth factors such as
tumour necrosis factor (TNF)-alpha, platelet-derived growth factor (PDGF)-B chain,
transforming growth factor (TGF)-beta, and basic fibroblast growth factor (bFGF) all
32
increase in glomeruli with duration of diabetes (Flyvbjerg et al., 1994). Increase in ETs
may play an important role in causing structural changes in the glomeruli since this upregulation was completely blocked by treatment with endothelin-A (ET-A) receptor
antagonists (Nakamura et al., 1995). Mesangial cell abnormalities including cell
expansion and hypertrophy and increased mesangial cell matrix also occurs with
nephropathy about 2-3 years after the onset of diabetes (Ellenberg and Rifkin, 1983).
Stage II is characterised by normal excretion of albumin regardless of the duration of the
diabetes. Stage III or incipient diabetic nephropathy is characterised by microalbuminuria
at rest, which can be detected by a sensitive method such as radioimmunoassay. Diabetic
patients often, at this stage, have higher blood pressure than non-diabetics do. It was
shown that endothelial dysfunction leads to the onset of microalbuminuria, which was
observed at this stage, and the level of glycaemic control seems to be the strongest factor
influencing the transition between normoalbuminuria to microalbuminuria (Marshall,
1998). Stage IV or overt diabetic nephropathy is characterized by clinical proteinuria and
even higher blood pressure than stage III. The development of proteinuria and decreased
GFR identify end-stage renal disease (Marshall, 1998; Nathan, 1993).
1.4.2.3.
Neuropathy
Diabetic neuropathies are complex, heterogeneous disorders that encompass a wide range
of abnormalities affecting both peripheral and autonomic nervous systems. They may be
diffuse or focal (acute onset neuropathy), proximal or distal, and involve somatic and
autonomic nerves (Vinik, 1999). Diffuse neuropathy includes peripheral and autonomic
neuropathy, whereas Focal neuropathy, which is less common, may affect single or
multiple peripheral nerves, cranial nerves, brachial and lumbasacral plexuses, and nerve
roots (Davidson, 1998; Vinik, 1999). Focal syndromes are classified as 1)
mononeuropathies or 2) entrapment syndromes. Mononeuropathies are usually due to an
inflammatory, vasculitic, or autoimmune condition.
The cause of diabetic neuropathy is poorly understood but many studies have suggested
that neurotrophic proteins such as nerve growth factors (NGFs) are involved (Apfel et al.,
1994; Hellweg and Hartung, 1990; Schmidt et al., 1999). In 4 weeks diabetic rodents, a
reduction in NGF level was observed in the serum and the sciatic nerve (Hellweg and
33
Hartung, 1990). Additionally, decreased NGF receptor expression in dorsal root ganglia
(DRG) of STZ-treated rats (Maeda et al., 1996) supports a role for impaired NGFmediated neurotrophism in the pathogenesis of experimental diabetic neuropathy.
Another animal study suggested that neuropathy may arise as a consequence of the loss
of neurotrophic insulin-like growth factor (IGF) activity due to diabetes (Zhuang et al.,
1996).
Peripheral neuropathy is a generalized sensorimotor polyneuropathy of gradual onset
that is usually progressive. It is the earliest and most widely recognized form of diabetic
neuropathy, which affects the lower limbs. Patients initially experience sensory
manifestations (e.g. burning sensation, paraesthesia, and dysesthesias). Loss of deep
tendon reflexes and poor sensation (vibration, touch) can also be present. Poor sensation
in the feet can lead to foot problems and may contribute to the development of diabetic
ulcers. Involvement of motor fibres can cause muscle weakness and atrophy leading to
feet deformities, which may also contribute to ulcers (Davidson, 1998).
Previous animal studies showed that peripheral impairments develop within weeks after
induction of diabetes (Biessels et al., 1999; Ralevic et al., 1995a; Tong et al., 1996). In
STZ-treated diabetic rats (6-14 weeks) this is characterised by abnormalities in thermal
pain threshold, reductions in DRG levels of substance P (SP) and calcitonin gene-related
peptide (CGRP), and slowing of nerve conduction velocity (Apfel et al., 1994). Early
changes (after 1 week) in the sciatic nerve endoneurial blood flow and function were also
shown (Cameron et al., 1991). However, no detectable structural basis has been found for
these early functional abnormalities i.e. reduction in nerve conduction velocity, pain
threshold, and blood flow (Walker et al., 1999). PKC may play a role in diabetic
neuropathy but the exact nature of this role is not clear (see also section 1.8.5.4).
Autonomic neuropathy is thought to encompass alterations to motor, sensory, and reflex
functions of the cardiovascular, gastrointestinal and genitourinary systems, impaired
sudomotor and vasomotor thermoregulatory mechanisms, disturbance of autonomic
control of endocrine secretion, changes in pupillary reflex function, and disturbances in
respiratory control (Davidson, 1998; Ellenberg and Rifkin, 1983; Vinik, 1999) (Table 1-
34
2). Symptoms may be non-specific, ranging from minor disturbances to extreme
disability.
Location
of
autonomic Clinical features
neuropathy
Cardiovascular system
Resting tachycardia
Postural hypotension:-
Gastrointestinal system
dizziness, vertigo,
nausea, vomiting,
faintness, syncope
Oesophageal atony:-
gastric reflex,
Dysphagia.
Gastric atony:-
nausea, vomiting,
anorexia,
postprantial bloating
Diabetic diarrhoea
Genitourinary system
Colonic atony:-
constipation
Bladder atony:-
loss of bladder
sensation, urinary
stasis, incontinence
Impotence and retrograde ejaculation
Thermoregulatory system
Sudomotor dysfunction:- anhydrosis
Loss of reflex vasodilatation and
vasoconstriction of skin vessels
Others
Pupillary disturbances:- decreased resting
pupillary diameter,
delayed response
to light
Defective respiratory
reflexes:cardiorespiratory arrest
Table 1-2: Clinical features associated with diabetic autonomic neuropathy (Adopted from Davidson
, M.B. (1998) & Ellenberg, M. /Rifkin, H. (1983))
35
Resting tachychardia is usually the initial clinical cardiovascular effect of autonomic
neuropathy. The parasympathetic arm of the autonomic nervous system is often affected
earlier and more profoundly than the sympathetic one. A reduction in the resting heart
rate, and an increase in the heart rate variability were observed in diabetic rats (1-17
weeks) but the mean arterial pressure was unchanged (Schaan et al., 1997). Exercise
intolerance may occur when the sympathetic nervous system is affected, because the
necessary increase in cardiac output and skeletal blood flow is blunted. Standing results
in venous pooling in the legs and splanchnic beds, with a decrease in venous return and
cardiac output. Normally, increased sympathetic outflow to the heart and peripheral
vessels will maintain blood pressure, and such reflex responses are mainly maintained via
the carotid sinus and aortic baroreceptors. Autonomic neuropathy of these reflex
pathways in diabetes is believed to contribute to postural hypotension (Davidson, 1998).
A study in diabetic rats (2 weeks) showed a reduction in the diastolic pressure and an
impairment in the cardiovascular response evoked by baroreflex and chemoreflex
activation (Dall'Ago et al., 1997). It was suggested that alterations in cardiovascular
responses might be secondary to the autonomic dysfunction of cardiovascular control.
Autonomic neuropathy of the heart muscle is also believed to contribute to the increased
incidence of painless myocardial infarction in diabetes, which can lead to sudden death
(Davidson, 1998).
Disturbance in most regions of the gastrointestinal system has been demonstrated to
involve damage to the parasympathetic, sympathetic, and afferent sensory fibres
(Davidson, 1998). Symptoms include oesophageal atony, gastric atony, delayed gastric
emptying, gall bladder atony, and colonic atony leading to constipation. However, the
most commonly recognised manifestation of autonomic neuropathy involving the gut is
diabetic diarrhoea. In rats, diabetic diarrhoea was observed after only 1 week pretreatment with STZ (Musabayane et al., 1995). Although the precise cause is unknown,
proposed mechanisms include pancreatic insufficiency, electrolyte imbalance, altered
gastrointestinal hormonal production, altered bile acid metabolism, and gastric vagal fibre
damage leading to decreased acid secretion and gut stasis, which may promote bacterial
overgrowth and small intestine contamination (Beebe and Walley, 1992).
36
Another major system demonstrating obvious alterations is the genitourinary system.
Neurogenic bladder disturbances may contribute to bladder atony, loss of bladder
sensations, incontinence, and urinary stasis (Ellenberg and Rifkin, 1983). Impotence may
also result in diabetic males following damage to the parasympathetic fibres of erectile
tissue (Faerman et al., 1974). For example, erectile dysfunction was documented in 4
weeks diabetic rats (Ari et al., 1999). This was shown to be initially reversible by insulin
treatment but with time (more than 6 months) it became irreversible. Further problems
that may arise from autonomic neuropathy include sweating disorders and vasomotor
instability. Loss of sweating and inability of the arterioles to respond to temperature
changes may result from damage to the sympathetic nerve pathways to the skin. For
example, in the case of minor tissue injury with secondary infection fluid loss, dryness,
and cracking of the skin can further exacerbate the infection. In diabetics this may lead to
gangrene and amputation. Complete loss of spontaneous variation in arterial blood flow
(e.g. in the feet) also indicates loss of reflex vasomotor responses in these diabetic
patients (Ellenberg and Rifkin, 1983). Hypoglycaemic unawareness, pupillary
abnormalities, and disturbed respiratory control have also been attributed to autonomic
neuropathy (Davidson, 1998). The relationship between changes in sensory nerve
function and diabetes and its relevance to this thesis is discussed later in this chapter
(sections 1.7.2. and 1.7.3).
1.5.
Mechanisms of diabetic complications
Chronic hyperglycaemia may contribute to the pathogenesis of the long-term
complications of DM, on the basis that effective glycaemic control reduces the incidence
of these complications in both animal models and patients with diabetes (Chase et al.,
1989; Reichard et al., 1993; Service et al., 1985). Results from a study by the DCCT in
1993 supported this hypothesis by showing that intensified insulin treatment delayed the
development and progression of diabetic retinopathy, microalbuminuria, and clinical
neuropathy (DCCT, 1993).
It is not clearly understood how hyperglycaemia contributes to tissue damage and
functioning but progress to date suggests that glucose-induced derangement of several
37
biochemical pathways in diabetes may be linked to the generation of diabetic
complications.
1.5.1. Polyol pathway
Aldose reductase is the first and rate-limiting enzyme of the polyol pathway, which
converts monosaccharides (e.g. glucose), to their polyols or sugar alcohols (e.g. sorbitol).
This enzyme is widely distributed throughout the body, including those tissues that are
susceptible to chronic diabetic complications (e.g. retina, lens, cornea, glomeruli,
schwann cells of peripheral nerves, and endothelium) (Greene et al., 1987; Harrison et al.,
1989). Aldose reductase has a low affinity for glucose, which at physiological glucose
concentrations is preferentially channelled into the glycolytic or pentose-phosphate
pathway under the action of hexokinase, which has a much higher affinity for glucose
(Fig 1-1). Under hyperglycaemic conditions in tissues where insulin does not regulate
glucose entry, intracellular glucose levels rise, and glucose is then increasingly diverted
through the polyol pathway to be converted into sorbitol. Sorbitol is poorly diffusable and
metabolized only slowly to fructose and therefore tends to accumulate within these
tissues (Pickup and Williams, 1997).
Low glucose concentrations
ATP
High glucose concentrations
Fig 1-1: Glucose metabolism and the polyol
NADPH
Hexokinase
ADP
Aldose Reductase
NADP+
Sorbitol
NAD+
Glucose- 6-phosphate
Sorbitol Dehydrogenase
Glycolytic Pentose phosphate
pathway
pathway
NADH
Fructose
pathway. Glucose enters the polyol
pathway only under conditions of
hyperglycaemia {Adenosine
diphosphate (ADP), Adenosine
triphosphate(ATP), Nicotinamide
adenine dinucleotide (NAD),
Nicotinamide adenine dinucleotide
hydrogen (NADH), Nicotinamide
adenine dinucleotide phosphate
(NADP), Nicotinamide adenine
dinucleotide hydrogen phosphate
(NADPH)} (Adopted from Pickup,
J.C. & Williams, G. (1997)).
Polyol pathway
Increased flux through the polyol pathway is associated with various other cellular and
metabolic abnormalities. These include depletion of myo-inositol, reduced activity of
Na+/K+-ATPase, an abnormality of kinase C-dependent protein phosphorylation, a
38
reduction in intracellular reduced glutathione concentrations, and an alteration in the
redox state, which may impair the activity of many cellular enzymes (Hotta, 1997). An
example is a study by Obrosova et al in 1999 showing reduction in the free NAD+/NADH
ratio and glutathione concentration in 3 weeks STZ-treated rats (Obrosova et al., 1999).
In the lens, sorbitol accumulation is thought to cause osmotic effects leading to cataract
(Kinoshita and Nishimura, 1988) and in the nerve, the reduced Na+/K+-ATPase activity
may be related to falls in the conduction velocity and impaired axonal transport (Mayer
and Tomlinson, 1983). In the kidneys, an increase in polyol pathway activity in early
diabetes can increase prostaglandin production, which may ultimately lead to changes
associated with diabetic nephropathy (Keogh et al., 1997). An enhanced polyol pathway
may lead to an increase in oxidative stress, which can inactivate NO (Honing et al.,
1998).
Development of aldose reductase inhibitors (Sarges and Oates, 1993) has permitted
investigation of the polyol pathway’s involvement in contributing to diabetes-induced
defects. Many animal studies showed that aldose reductase inhibitors are mainly effective
in preventing diabetic complications from developing rather than reversing the
established tissue damage (Cameron et al., 1997a; Hotta et al., 1996; Tomlinson, 1993).
In humans, clinical benefits of aldose reductase inhibitors are less convincing. For
example, although some positive effects were reported from clinical trials on patients
with diabetic neuropathy and retinopathy, overall results were inconsistent (Christensen
et al., 1985; Clark and Lee, 1995; Gonen et al., 1991).
39
1.5.2. Advanced Glycated End-products (AGEs)
Synthesis/Structure/Function
The glycation process is a natural biochemical event occurring as a consequence of the
glucose-rich milieu that constantly permeates living tissues. When proteins or lipids are
exposed to aldose sugars, they undergo non-enzymatic glycation and oxidation. Initially
this leads to the formation of early glycated products, schiff bases, and amadori products
(Maillard, 1912). Early glycated products are still in equilibrium with plasma glucose, but
when glucose levels fall they can dissociate to the native proteins. Alternatively, if
glycation continues, further molecular rearrangements (glycation, glycoxidation, and
autoxidative glycosylation) occur to generate the irreversible, heterogeneous AGEs
(Friedman, 1999) (Fig 1-2). Glycation is a slow process under normal ambient sugar
concentrations.
Glucose
+
NH2-R
Schiff
base
R
NH2
+
O
O
C
(CHOH)
Fig 1-2: Scheme of AGEs formation in the
Weeks, Months
Days
Hours
AGEs
Amadori
R
R
+NH
C
(CHOH)
CH2OH
R
O
NH
CH2
N
N
O
O
R
C
(CHOH)
CH2OH
CH2OH
“Maillard” reaction: Reducing
sugars condense with amino
groups of macromolecules to
form reversible schiff base
adducts. Intramolecular
rearrangements lead to
chemically stabilized amadori
products. Further rearrangements,
dehydration and condensation
reactions result in the irreversible
formation of AGEs that are
capable to physically cross-link
amino groups. (Adopted from
Bierhaus,A. et al (1998))
The Maillard reaction has been implicated in the development of pathophysiology in agerelated diseases such as DM, atherosclerosis, Alzheimer’s disease, and in dialysis-related
amyloidosis (Nawroth et al., 1999).
AGEs are fluorescent yellow-brown pigments, which are found on red blood cells,
platelets, extracellular matrix, and vessel wall components including endothelial and
smooth muscle cells (Wautier and Guillausseau, 1998). AGE formation changes the
40
physicochemical properties of proteins, lipids, and nucleic acids. Some of AGEs
biological properties are listed below (Bierhaus et al., 1998):
-
Induction of RAGE, one of the known receptors for AGE
Impairment of vasodilatory effects by quenching NO
Increased vasoconstriction by inducing ET-1
Induction of intracellular oxidative stress and activation of transcription factor NF-κB
Lipid peroxidation
Reduction of intracellular antioxidative defence mechanisms (e.g. glutathione,
vitamin C)
Increased macrophage uptake of AGE-low density lipoprotein (LDL) and possible
atheroma formation
Induction of cytokines and growth factors (Interleukin (IL)-1α, TNF, IGF-1A, PDGF)
by monocytes / macrophages folowed by vascular cell proliferation
Induction of smooth muscle cell proliferation
Enhancement of deoxyribonucleic acid (DNA) mutation rate
AGE receptor
RAGE, a cell surface receptor for AGE, is a member of the immunoglobulin superfamily
and exists in two forms, 35kDa and 45kDa (Thornalley, 1998). Hyperglycaemia was
identified as one of the factors that stimulates RAGE expression (Thornalley, 1998) and
this could play an important role in diabetic complications. The 45kDa form of RAGE is
known as soluble RAGE or sRAGE with a proteolytic property which is used, in vivo, as
a pharmacological agent to prevent the vascular effects of AGEs (acts as an antagonist) in
experimental diabetes (Renard et al., 1997). For example, vascular hyperpermeability in
diabetic rats was inhibited in the presence of sRAGE in a dose-dependent manner
(Wautier et al., 1996). Anti-RAGE IgG can also inhibit AGE-RAGE interaction and the
rise of vascular complications in diabetes (Schmidt et al., 1995; Wautier et al., 1994; Yan
et al., 1994).
Normally, AGE-modified proteins are either repaired or replaced and degraded in vivo.
The recognition and degradation of such proteins is mediated by specific AGE receptors,
which are found on monocytes (Schmidt et al., 1996), macrophages (Nawroth et al.,
1999), endothelial cells (Lalla et al., 1998; Wautier et al., 1994; Yan et al., 1994),
vascular smooth muscle cells, pericytes, podocytes, astrocytes, microglia, and neurons
(Thornalley, 1998). Interaction of AGE with RAGE on macrophages stimulates these
41
cells to produce and release cytokines, growth factors, proteolytic enzymes, and increase
expression of extracellular matrix proteins and vascular adhesion molecules, which are all
required for normal tissue remodelling (Nawroth et al., 1999) (Table 1-3).
______________________________________________________________
AGEs in atherosclerosis
- Accumulation in the vascular matrix Æ narrowing & occlusion
- Vascular endothelial dysfunction (e.g. inactivating NO) Æ procoagulant state,
vasoconstriction, hypertension
- Glyoxidation of LDLÆ slow degradation of LDL, lipid peroxidation, oxidative stress
- Trapping of plasma proteins Æ initiation of complement activation, oxidation
- Monocyte chemotaxis/activation Æ cytokine & growth factor release Æ vascular
tissue proliferation
- Increased endothelial cell permeability Æ vascular leakage
AGEs in renal disease
- Matrix expansion, vascular leakage & basement membrane thickening Æ glomerular
hypertrophy & glomerular sclerosis
- Glomerular sclerosis Æ albuminurea
- Delayed clearence of AGE-peptide Æ uremic complications
AGEs in diabetic neuropathy
- Accumulation in vasa nervosum Æ wall thickening & occlusion, ischemia
- Vascular endothelial dysfunction Æ occlusion, ischemia
- Glycation of myelin Æ myelin damage
- Glycation of growth factors (NGF, fibroblast growth factor(FGF)) Æ loss of function
- Accumulation on macrophages Æ macrophage activity Æ myelin & vascular
degeneration
AGEs in diabetic retinopathy
- Increased endothelial cell permeability Æ vascular leakage & retinal damage
- Vessel wall thickening Æ occlusion, ischemia
- Coagulaton Æ occlusion, ischemia
- Induction of autocrine vascular endothelial growth factor (VEGF) synthesis Æ
angiogenesis, neovascularization
_______________________________________________________________________
Table 1-3: Some of the consequences of AGE formation and deposition (Adopted from Bierhaus, A. et al
(1998) & Vlassara, H. et al (1994))
Functional alterations or saturation of the macrophage system would allow anomalous
tissue accumulation of AGEs leading to reduced structural protein turnover, increased
collagen cross-links, and excessive degeneration and / or proliferation of tissue
42
components as observed in ageing and diabetes (Bierhaus et al., 1998; Grossi and Genco,
1998; Sensi et al., 1991) (Table 1-3).
AGE-RAGE interaction can also alter cellular phenotypes generating a pro-inflammatory
environment where induction of vascular perturbation, development of vascular lesions,
and impairment of normal reparative responses are initiated (Lalla et al., 1998)
independent of macrophage activity (Wautier et al., 1994; Yan et al., 1994). Introduction
of AGEs into the extracellular matrix can also interfere with the endothelial function at
different levels. For example, AGEs quench NO, during its passage from endothelial cell
to smooth muscle cells, inhibiting its vasodilatation and antiproliferative actions (Bucala
et al., 1991).
Collagenous proteins are especially prone to non-enzymatic glycation because they
contain several dibasic amino acid residues with free amino groups, have a very slow
turnover rate, and are exposed to ambient levels of glucose. The main functional property
of collagen is to provide a supporting framework to almost all tissues. AGEs inhibit a
normal network formation by type IV collagen and decrease heparan sulphate
proteoglycan binding by vitronectin and laminin in diabetes (Brownlee, 1994). AGEs
fusion with albumin was also shown to interact with the glomerular basement membrane
inducing glomerular sclerosis and albuminuria in rats (Vlassara et al., 1994).
AGEs and diabetes
Under hyperglycaemia and / or conditions where the protein and lipid turnover period is
prolonged, the advanced glycation process is enhanced (Lalla et al., 1998; Nawroth et al.,
1999). For example, AGEs were identified in mesenteric vessels of STZ-treated diabetic
rats within 3 weeks (Rumble et al., 1997) and in their skeletal muscle arteries within 4-6
weeks (Hill and Ege, 1994).
It was postulated that AGEs contribute to the development of vascular diseases associated
with diabetes (Chibber et al., 1997; Makita et al., 1995; Wautier and Guillausseau, 1998).
AGEs formation on the matrix component of the vessel wall can cause structural damage
by decreasing elasticity, increasing thickness, rigidity, and narrowing of the vessel lumen
(Bierhaus et al., 1998; Turk et al., 1999). AGEs increase collagen cross-linking leading to
43
the arterial stiffness that is commonly observed in normal ageing but at an accelerated
rate in diabetes (Odetti et al., 2000; Wolffenbuttel et al., 1998). In diabetic rats, some
AGEs were found in the aortic collagen after 4 and 12 weeks but this was significantly
increased by 20 weeks (Turk et al., 1999).
AGEs change endothelial cell properties, relevant to pathogenesis of vascular disease, by
forming cross links, recognizing RAGE on cellular surfaces (Yan et al., 1994) and
generating ROS by undergoing autoxidation (Chappey et al., 1997; Schmidt et al., 1994).
Interaction of AGEs with RAGE on vascular matrix proteins may generate superoxide
anions (Galle et al., 1998), which can inactivate NO (Bucala et al., 1991). This is
supported by previous studies showing glycated haemoglobin (Hb) induced endothelial
dysfunction in rat (Angulu et al., 1996) and human vessels (Vallejo et al., 2000). AGEs
also induce ET-1 expression and change endothelial function towards vasoconstriction
(Quehenberger et al., 1995). An in vitro study demonstrated that ET-1 could potentiate
the oxidized LDL-induced superoxide anion formation (Galle et al., 2000). Hence, it is
possible that ET-1 could play a role in inhibiting NO in the diabetic condition via this
mechanism. A previous study supported this hypothesis and showed that reduction of
cytochrome C by glycated proteins, a process involving superoxide anions, could induce
lipid peroxidation (Ceriello, 1999). It was suggested that increased formation of ROS
(e.g. superoxide anions and hydroxyl radicals) participate in the vascular endothelial cell
dysfunction (Gryglewski et al., 1986; Suzuki et al., 1991) (section 1.5.3).
It is known that transmembrane signalling of contractile cells is generally inhibited by
high glucose. In the kidney, it was hypothesized that glycated macromolecules with
RAGE might inhibit transmembrane calcium signalling (Mene et al., 1999). Also,
reduced function of the L-type calcium channel that occurs in retinal pericyte culture in
high glucose maybe due to the glycation of a channel protein (McGinty et al., 1999).
Both free glucose (Booth et al., 1997) and protein-glucose adduct (Taniguchi et al., 1996)
could undergo oxidation in the presence of trace amounts of metal ions, to generate free
radicals and reactive carbonyls. Amadori products undergo metal-free oxidation to form
superoxide anions (Mossine et al., 1999). In one study, incubation of cultured smooth
muscle cells with early glycated products (after 2 weeks) and cupric ion (Cu2+) was
44
shown to be capable of damaging these cells (Sakata et al., 1998). It was hypothesized
that the generated hydrogen peroxide, under this condition, was a candidate for this
damage. It is possible that hydrogen peroxide damages smooth muscle cells by collagen
cross-linking (Elgawish et al., 1996).
ROS formed by glucose and protein-glucose adducts may be involved directly in the
formation of AGEs and in AGE-induced pathologic alterations in gene expression
(Brownlee, 2000). This was supported by previous studies, in vitro, showing antioxidants
(e.g. vitamin E, alpha-lipoic acid, dimethyl sulfoxide) can inhibit the hyperglycaemiainduced AGE-formation (Bierhaus et al., 1997; Giardino et al., 1996). AGE-RAGE
interaction also depletes the cellular antioxidant defence mechanism (e.g. glutathione,
vitamin C, superoxide dismutase (SOD)) in diabetes (Bierhaus et al., 1997; Ceriello,
1999; Mak et al., 1996). This was supported by a previous study demonstrating
inactivation of Cu-Zn-SOD by the non-enzymatic glycation (Kashiwagi et al., 1996).
Also it was reported that the level of glycated Cu-Zn-SOD was increased in patients with
diabetes (Taniguchi et al., 1996). It is possible therefore, that the combination of both an
increase in ROS and the reduction in antioxidant efficiency could contribute to tissue
damage in diabetes.
Evidence exists that transition metals may be involved in causing diabetic vascular and
nerve abnormalities (Qian and Eaton, 2000). This author hypothesized that heavily
glycated proteins, known to accumulate in diabetic patients, may gain an increased
affinity for transition metals such as iron and copper. As a result, proteins such as elastin
and collagen within the arterial wall can accumulate bound metal (i.e. copper), causing
catalytic destruction of EDRF and engendering a state of chronic vasoconstriction. This
in turn could impair nerve blood flow and ultimately cause nerve death. Administration
of chelators could prevent or reverse slow peripheral nerve conduction, neuronal blood
flow, as well as impaired endothelium-dependent arterial relaxation, supporting the above
hypothesis (Qian and Eaton, 2000).
Diabetic neuropathy is associated with some early defects of axonal transport in STZtreated animals (Di Giulio et al., 1995; Tomlinson and Mayer, 1984). Axonal transport is
45
dependent on intact microtubules and unsubstituted lysine residues of tubulin are
essential for microtubule polymerization. Reduced polymerization of the “glycated”
tubulin in diabetes (Williams et al., 1982) could in part be responsible for alterations of
the axoplasmic transport observed in diabetic neuropathy. There was more than a 4-fold
increase in the extent of glycation of tubulin in the sciatic nerve of 2 weeks diabetic rats
(Cullum et al., 1991). Increased non-enzymatic glycation of myelin and myelinassociated proteins have also been reported in diabetes, both at the peripheral and central
nervous system levels (Ryle et al., 1997; Sensi et al., 1991). For example, high levels of
the early glycated product (furosine) and low levels of AGEs were found in the sciatic
nerve and cytoskeletal preparations of 6 weeks STZ-treated rats (Ryle et al., 1997). Also,
significantly lower levels of AGEs were found in the spinal cord and spinal nerve of
these rats. It was suggested that enhanced accumulation of AGEs in cytoskeletal as well
as myelin and peripheral nerve proteins might contribute to the axonal degeneration
polyneuropathy in diabetes (Ryle and Donaghy, 1995). Quantitative analysis of the
myelinated nerve fibres (peroneal nerve) showed no significant morphometric changes in
2 (Nukada, 1993) or 4 weeks diabetic rats but a reduction in the myelinated nerve calibre
was observed after 8 weeks of diabetes (Ochodnicka et al., 1995). Over-glycation of
myelin, in the diabetic condition, seems to act as a signal for recognition and degradation
by macrophages, causing segmental demyelination (Ryle and Donaghy, 1995). Moreover,
the presence of AGE on myelin appears to modify the properties of its proteins by
trapping plasma proteins (covalent bonding). Presence of these molecules can start
immunological reactions responsible for demyelination. The central nervous system is
not involved in this immunological phenomenon and is protected by the blood-brain
barrier.
AGE-RAGE interaction in diabetes, involving erythrocytes, increases vascular
permeability (Bonnardel-Phu et al., 2000). It was proposed that this increase could be
mediated by free radicals since antioxidants inhibited hyperpermeability in both in vitro
and in vivo conditions (Wautier et al., 1996). VEGF may also influence the tightness of
the endothelial barrier. This hypothesis was based on a previous study showing AGEs
46
play a role in increasing the TGF-beta expression in experimental diabetes (Rumble et al.,
1997), which in turn could enhance VEGF levels (Pertovaara et al., 1994).
Non-enzymatic glycation affects proteins involved in the clotting cascade and in diabetes
this modification may interfere with vascular wall homeostasis. For example, in diabetes,
increased glycation of platelet membrane proteins contributes to functional platelet
defects and microangiopathy (Yamagishi et al., 1998). It was suggested that the
underlying mechanism might involve cyclic adenosine monophosphate (cAMP). The
same study also showed that AGE-RAGE interaction could inhibit the production of
prostaglandin I2 and up-regulate plasminogen activator inhibitor-1 (PAI-1), by
microvascular endothelial cells, through a reduction in cAMP. Erythrocyte membrane
proteins undergo non-enzymatic glycation over the course of the red blood cell life, and
the amount of amadori products on these proteins increases in the diabetic state.
Previously, it was reported that excessive membrane glycation may play a role in altering
some properties of the erythrocyte membrane proteins which can affect doublet
formation, microviscosity, and increased oxygen affinity (Sensi et al., 1991). It was also
shown that erythrocyte AGEs increase vascular permeability due to extensive contact
with the vessel wall RAGE, leading to vascular complications (Wautier and Wautier,
1999).
LDL, a lipid transport protein that plays an important role in the development of
atherosclerosis, is glycated as shown in both in vitro and in vivo conditions (Thorp and
Baynes, 1996). The increase in LDL glycation causes a reduction in the uptake and
degradation of LDL by tissue fibroblasts and macrophages, contributing to premature
atherosclerosis in diabetics (Vlassara, 1996). Lipid peroxidation is known to induce
cross-linking of collagen with a high rate in the presence of high glucose (Fu et al.,
1994). Advanced glycated collagen was shown to be capable of covalently trapping LDL
(e.g. in the arterial wall) causing it to be modified oxidatively by free radicals (Mullarkey
et al., 1990). It was proposed that LDL oxidation may contribute to the formation of
“foam” cells (Wautier and Guillausseau, 1998).
47
Aminoguanidine (AG)
AG is a small nucleophilic hydrazine compound. It blocks AGE formation by specifically
binding to non-protein-bound glucose-derived intermediates of early glycated products
and preventing their further rearrangements (protein-protein or protein-lipid cross-links)
(Brownlee et al., 1986; Friedman, 1999) (Fig 1-3). Inhibition of AGEs by AG involves
interaction between the reactive hydrazine moiety of AG and the reactive carbonyl
groups of amadori or glycoxidation products (Corbett et al., 1992; Yu and Zuo, 1997).
+
NH2
Fig 1-3: The chemical structure of aminoguanidine
(Adopted from Vlassara, H. et al (1994))
H2N NH
NH2
The glycation process can decrease the susceptibility of molecules to proteolysis. The
reverse is true of protein oxidation, which has previously been shown to increase the
susceptibility of proteins to proteolytic digestion (Skamarauskas et al., 1996). AG is
capable of enhancing the proteolytic digestion of glycated bovine serum albumin (BSA),
a protein shown to be protease resistant (Skamarauskas et al., 1996). High doses of AG,
in vitro, inhibits catalase, generates hydrogen peroxide, and other oxidants similar to
hydroxyl radicals in activity (Ou and Wolff, 1993; Skamarauskas et al., 1996). However,
in vivo and at low doses, AG is regarded as virtually a non-toxic substance. It can act as
an antioxidant, quenching hydroxyl radicals and lipid peroxidation in cells and tissues as
well as preventing oxidant-induced apoptosis (Giardino et al., 1998).
AG can also retard or prevent the development of nephropathy (Soulis-Liparota et al.,
1991), retinopathy (Hammes et al., 1991) and neuropathy (Cameron et al., 1992) in the
STZ-induced diabetic rat model (Table 1-4).
48
______________________________________________________________________
Effects on the development of atherosclerotic lesions
- prevention of collagen- to- collagen cross-linking
- decreased collagen stability
- increased elasticity and decreased fluid filtration in large arteries
- decreased trapping of lipoproteins
- inhibition of glycation and oxidation of LDL
- decrease in LDL-cholesterol, very low density lipoprotein (VLDL)-cholesterol, total
cholesterol and triglycerides
Effects in kidney disease
- reduction of AGE accumulation in the renal glomerulus
- prevention of diabetes-induced thickening of basement membranes
- prevention of diabetic nephropathy in animal models
- reduction of albuminuria in diabetic and hypertensive animals
- decreased trapping of IgG by basement membranes of renal capillaries
Effects on the diabetic peripheral nerve
- preservation of nerve conduction velocity
- normalization of nerve action potential amplitude and peripheral nerve blood flow
Effects on diabetic retinopathy
- prevention of lens protein cross-linking
- prevention of AGE-formation and formation of acellular capillaries in the retina
- reduction of microaneurysms
- reduction of pericyte drop-out
- inhibition of hypertension dependent accelerated diabetic retinopathy
________________________________________________________________________
Table 1-4: Effects of AG (Adopted from Bierhaus, A. et al (1998) & Friedman, E.A. (1999))
A previous study in vivo showed that generation of AGEs at least in rat kidneys was timedependent and closely linked to the development of metabolic and structural changes up
to 32 weeks after induction of diabetes (Soulis-Liparota et al., 1996). This may be an
important factor to consider when using preventative treatment regimes involving AG in
diabetics with the risk of developing complications associated with AGEs (chapter 8).
Furthermore, AG was shown to have no influence on the body weight or hyperglycaemia
in STZ-treated rats during the 32 weeks post-diabetes (Soulis-Liparota et al., 1996). A
different study showed an increase in regional vascular albumin permeation by about 2 to
3 fold in the ocular tissue, sciatic nerve, and aorta of 5 weeks STZ-treated diabetic rats
which was normalized by AG treatment (Tilton et al., 1993). While AG prevents the
49
ongoing AGE formation, it may not be effective in patients with a long history of disease
that already resulted in extensive tissue-AGE accumulation. Also, the need to remove
irreversibly bound AGEs from connective tissues and matrix compounds recently led to
the design of AGE-cleaving agents such as prototypic AGE cross-link “breaker” Nphenacyl thiazolium bromide (PTB). Vasan et al in 1996 performed the first in vivo and
in vitro experiments to show that PTB can reduce AGE cross-links (Vasan et al., 1996)
but further studies on “breakers” are required.
AGEs and skin microvasculature
Direct in vivo evidence regarding the pathogenic influence of AGEs, independent of
diabetes, was presented by Vlassara et al (1992). They showed that short-term (2-4
weeks) administration of in vitro prepared AGE-albumin to normal rats and rabbits
produced vascular defects similar to those associated with experimental diabetes, such as
vascular leakage and mononuclear cell extravasation, as well as unresponsiveness to
vasodilator agents. These defects were significantly inhibited by AG treatment in control
rats. In a study by Soulis-Liparota et al (1996), using diabetic rat kidneys (32 weeks
duration), it was suggested that a relationship exists between duration of AG treatment
and its protective effects but not the commencement of treatment (Soulis-Liparota et al.,
1996).
Many studies have investigated the effects of AGEs and AG on diabetic vascular
responses in different tissues and at different stages of the disease (Bucala et al., 1991;
Quehenberger et al., 1995). However, the role of AGEs in modulating skin microvascular
blood flow in early diabetes has not been previously examined. The aim of chapter 8 was
to assess the short- and long-term effects of early and late products of glycation (e.g.
amadori and AGEs) in modulating microvascular blood flow in 4 weeks STZ-induced
diabetic rats. The effect of AGEs in microvascular changes in diabetic rats was
investigated with regards to the duration of diabetes and the treatment with AG.
50
1.5.3. Oxidative stress and antioxidants
Free radicals are defined as atoms or molecules that contain one or more unpaired
electrons. The term reactive oxygen species (ROS) refers to oxygen radicals (superoxide
anions (O2•) and hydroxyl radicals (OH•)) and a number of related species, such as
hydrogen peroxide (H2O2), that do not themselves contain unpaired electrons but are
often involved in the generation of free radicals (Fig 1-4).
Reduction
O2
NADPH
Oxidase
XO
.O2- SOD
+
NO.
ONOO-
Oxidation
H2O2
Fig 1-4: Formation of oxygen derived radicals.
Catalase
[molecular oxygen (O2), superoxide
anion (•02-), hydrogen peroxide
(H2O2), hydroxyl radical (•OH-),
superoxide dismutase (SOD),
xanthine oxidase (XO), nitric oxide
(NO•), peroxynitrite (ONOO-),
nitrogen dioxide (NO2)] (Adopted
from Gryglewski, R.J. et al (1986),
Munzel, T. et al (1997), and Kehrer,
J.P. (1993))
H2O + O2
Fe (II)
(Haber-Weiss and Fenton reaction)
.OHNO2
ROS are formed continuously as normal by-products of cellular metabolism and
sometimes by accident and they have the potential to produce high levels of tissue
damage (oxidative stress). They damage many cellular components including DNA but
primarily phospholipid A-containing cellular membranes, which are converted to
malonyldialdehyde (MDA) by lipid peroxidation (Stohs, 1995).
Antioxidants are substances capable of neutralizing oxygen free radicals and other similar
agents. The body’s natural protection system against free radicals involves different
enzymes such as SOD, that scavenges superoxide anions, and catalase and glutathione
redox enzymes (especially glutathione peroxidase) that act as a secondary defence system
(Thomas, 1995). Additional antioxidant defence is provided by certain vitamins (C, E,
beta-carotene), minerals (selenium), and other substances with redox capability
(ubiquinone, coenzyme Q10) and is obtained through intake of food since they are not
synthesized by the body (Ernster and Dallner, 1995).
51
Physiological effects of free radicals
ROS generated by endothelial cells (e.g. superoxide anions, hydrogen peroxide, and
hydroxyl radicals) (Bonnardel-Phu and Vicaut, 2000a) can directly modify vascular
endothelium by interfering with the availability of NO, peroxidation of membrane lipids,
and up-regulation of adhesion molecules to platelets and leucocytes. ROS are important
for physiological activity of smooth muscle cells and in the pathogenesis of various
diseases, where the function of smooth muscles is altered. One study, using normal
smooth muscles from different body parts, showed that ROS are involved in changing the
muscle tone, membrane conductance, calcium homeostasis and dependent processes, as
well as in eicosanoid and NO metabolism (Bauer et al., 1999).
Superoxide anions can interact directly with NO to reduce its biological activity
(Gryglewski et al., 1986) causing contraction of the vascular smooth muscle (Kaustic and
Vanhoutte, 1989). Hydrogen peroxide is produced by dismutation of superoxide anions
either spontaneously or catalyzed by SOD (Fig 1-4). SOD can be either cytotoxic
(Rubanyi, 1988) or act as a smooth muscle relaxant (Zembowics et al., 1993), depending
on its concentration, where in the latter case it can accelerate NO release (Rubanyi and
Vanhoutte, 1986).
In the absence of superoxide anions, NO engages in nitrosative chemistry to yield stable
N-nitrosamine derivatives. As the flux of superoxide anions is increased, nitrosative
reactions are suppressed and oxidative chemistry is enhanced (Jourd'heuil et al., 1997).
Thus, depending upon the flux of each radical, either nitrosation or oxidation chemistry
may predominate. The chemical reaction between superoxide anions and NO generates a
highly reactive peroxynitrite (ONOO-) (Jourd'heuil et al., 1997) (Fig 1-4), which in turn
stimulates cyclo-oxygenase catalysis, lipid peroxidation, and increased prostanoid
production, resulting in endothelial dysfunction (Bouloumie et al., 1997). In physiologic
pH, peroxynitrite leads to the formation of hydroxyl radicals (Prasad and Bharadwaj,
1996). Interaction between hydrogen peroxide and superoxide anions also generates
hydroxyl radicals (Haber-Weiss/Fenton reaction) (Fig 1-4). It is believed that these
radicals can elicit contraction of the vascular smooth muscle (Kontos and Wei, 1993),
however, the exact mechanism is not clear. Evidence exist that hydroxyl radicals are
52
capable of both stimulating and inhibiting prostaglandin synthesis and stimulating
guanylate cyclase to trigger smooth muscle relaxation (Prasad and Bharadwaj, 1996;
Rubanyi, 1988)
ROS also play a role in alteration of calcium homeostasis. Under normal conditions ROS,
generated by the hypoxanthine-xanthine oxidase system, can mobilize calcium from
intracellular stores and induce a marked rise in the cytosolic calcium level in different
cells (Dreher and Junod, 1995). They can also decrease (Ca2+)-stimulated ATPase
activity, increase (Mg2+)-ATPase activity and reduce sulphydryl (SH) group content
(Suzuki et al., 1991).
ROS and diabetes
In the diabetic condition, the level of free radicals is altered and the function of the
antioxidant defence system is compromised. Evidence exists that an acute increase in
plasma glucose concentration may enhance the production of free radicals by
autoxidation of glucose (Pieper et al., 1997b) and the advanced glycation process that
attenuates the activity of antioxidant enzymes (Mak et al., 1996). The advanced glycation
process can also activate the polyol pathway, intracellularly, which produces an
imbalance in the NADH / NAD+ ratio and favours the production of free radicals (Ido et
al., 1996).
The activity of total SOD (Cu-Zn SOD & Mn-SOD) was increased in the heart, aorta, and
blood of rats from the 2nd week after diabetes onset and continued up to the 4th week,
after which it tended to decline (Kakkar et al., 1996). In the rat brain, an increase in fatty
acid and MDA level and a decrease in SOD and catalase activities were observed after 2
weeks treatment with STZ (Kumar and Menon, 1993). Catalase activity, on the other
hand, was increased in the heart and aorta of diabetic rats (0-6 weeks) but showed
significance only after 4 weeks (Kakkar et al., 1996). Activities of catalase and SOD
were significantly increased after 4 weeks of diabetes in the rat brain (Ramanathan et al.,
1999). No change in the level of glutathione peroxidase activity was observed in the
heart, blood or brain up to 4 weeks after STZ treatment in rats (Kakkar et al., 1996;
Ramanathan et al., 1999). The gene expression of catalase and SOD (both types) was not
reduced in L4, L5 of DRG and superior cervical ganglion of 3 or 12 months diabetic rats
53
(Kishi et al., 2000). It was suggested that changes in enzyme activity might be related to
duration of diabetes or due to post-translational modifications.
Pathological effects of free radicals
An increase in the production of plasma free radicals (Mullarkey et al., 1990; Ohkuwa et
al., 1995; Traverso et al., 1998) and a general reduction in antioxidant defences (Kamata
and Kobayashi, 1996; Mak et al., 1996) were shown to be related to diabetic
complications such as endothelial dysfunction (Stehouwer et al., 1997).
Exposure to glucose causes an increase in intracellular sorbitol and fructose levels
leading to a reduction in NADPH cell stores. This could inhibit NO synthase and
glutathione reductase activities. Decreased levels of NO can lead to vasoconstriction and
tissue injury while reduced levels of glutathione increases the susceptibility of endothelial
cells to damage by hydrogen peroxide (Paolisso and Giugliano, 1996). In early (3 weeks)
diabetic rat’s synaptosomes, the reduction in glutathione and alpha-tocopherol (Vitamin
E) level was suggested to be the reason for an increase in the level of hydroxyl radical
and hydrogen peroxide (Aragno et al., 2000).
SOD protects NO from breakdown by decreasing the concentration of superoxide anions
(Diederich et al., 1994; Gryglewski et al., 1986). One study suggested that the rapid
destruction of NO by superoxide anions, in STZ-treated rats, might be due to the decrease
in mRNA expression of Mn-SOD and Cu-Zn-SOD (Kamata and Kobayashi, 1996).
Exogenously or endogenously-derived ROS were previously shown to be capable of
enhancing ET-1 production in diabetic glomeruli (1 & 4 weeks) and this was inhibited by
SOD and catalase (Chen et al., 2000). Hyperglycaemia or insulin deficiency was also
suggested to be partially responsible for the increase in ET-1 production in these animals.
A previous study reported that ROS increased oxidation of LDL, formation of AGE, and
activation of platelets and monocytes (Stehouwer et al., 1997). Hydroxyl radicals are
capable of oxidizing lipids, damaging cell membranes, and oxidizing thiol groups
(Munzel et al., 1997) and have been implicated in diabetes-induced endothelial
dysfunction (Pieper et al., 1997b).
54
The presence of MDA, a product of lipid peroxidation, can easily be detected by
thiobarbituric acid (TBAR) (Stohs, 1995). It was proposed that products of lipid
peroxidation might reduce NO bioactivity without affecting endothelial NOS (eNOS)
mass or its catalytic activity (Hayakawa and Raij, 1999). In diabetic rats (0-6 weeks), a
significant increase in TBAR levels in the aorta, blood and liver was observed (Kakkar et
al., 1996). Lipid peroxide levels were also significantly increased in rat pancreas and
heart as early as 2 weeks after induction of diabetes (Kakkar et al., 1998; Tatsuki et al.,
1997).
In oxidative stress, changes in intracellular calcium homeostasis are thought to contribute
to the cell dysfunction. Sasaki et al (1993) supported the view that hydroxyl radicals are
capable of potentiating the voltage-dependent influx of calcium (Sasaki and Okabe,
1993). This author hypothesized that, in smooth muscle cells, hydroxyl radicals might
damage the sarcoplasmic reticulum (SR), which could reduce calcium release. A different
study showed that in STZ-diabetic (4 weeks) rat aorta, superoxide anions might be
responsible, at least in part, for the impaired endothelial integrity, enhanced alphaadrenergic receptor-mediated phosphoinositol (PI) turnover, and the augmented
contractility possibly through modification of calcium channels (Chang et al., 1993).
Lipid peroxidation by-products may also interfere with the calcium homeostasis by
inhibiting microsomal glucose 6-phosphate, plasma membrane Ca2+/ Mg2+-activated
ATPase, and the Na+/K+ activated ATPase pump (Dianzani, 1990; Wells et al., 1997).
Diabetes and exogenous antioxidants
N-acetylcysteine (NAC) has widely been used, in vivo and in vitro experiments, as an
exogenous antioxidant. It was shown that NAC is a powerful scavenger of hydroxyl
radicals and reacts slowly with hydrogen peroxide but no reaction with superoxide anions
was detected (Aruoma et al., 1989). It is also a sulphydryl donor and a precursor for
glutathione synthesis (Pieper and Siebeneich, 1998; Sagara et al., 1996).
In diabetic rats, NAC corrected the decreased GSH and increased lipid peroxide levels in
plasma without any effect on the level of blood glucose and nerve glucose, sorbitol or
cAMP content (Sagara et al., 1996). Long-term treatment with NAC could remove
55
radicals released as by-products during the glycation process and prevent the defective
relaxation response to acetylcholine (ACh) in diabetes (Pieper and Siebeneich, 1998).
NAC was also reported to inhibit the development of functional and structural
abnormalities of the peripheral nerve in STZ-induced diabetic rats (12-15 weeks) (Sagara
et al., 1996).
Tirilazad mesylate (Freedox) is a potent inhibitor of lipid peroxidation (Saniova, 1997)
and was the first lazaroid compound to be used in clinical practice and disease conditions
where lipid peroxidation caused further tissue damage. Tirilazad mesylate cytoprotective
properties include scavenging free radical intermediates, stabilizing cell, and preserving
vitamin E content in the membrane (Saniova, 1997).
A previous study showed that treatment with tirilazad mesylate could normalize the
increase in plasma lipid peroxides in rats after 8 weeks of diabetes without any change in
the level of serum insulin or concentration of blood glucose (Pieper et al., 1997a). In rats,
oral pre-treatment with tirilazad mesylate (for 8 weeks) did not prevent the diabetesinduced impairment in endothelium-dependent relaxation caused by ACh (Pieper et al.,
1997a). Tirilazad mesylate also protected cells by preserving the post-injury calcium
homeostasis (Saniova, 1997).
ROS and skin microvasculature
It was shown that ROS are important in the physiological activity of smooth muscle cells
and the active species produced by endothelial cells can directly modify vascular
endothelium (Bauer et al., 1999; Bonnardel-Phu and Vicaut, 2000a). ROS also play an
important role in the pathogenesis of various diseases such as diabetes (Stehouwer et al.,
1997). For example, the synthesis and release of hydroxyl radicals was suggested to be
responsible for the impairment of nitric oxide synthase (NOS)-dependent dilatation of the
basilar artery in 3-4 weeks diabetic rats (Mayhan and Patel, 1998). The beneficial effects
of antioxidants in the diabetic condition have been previously documented (Kaneto et al.,
1999; Pieper et al., 1997a). For example, long-term treatment (e.g. 8-13 weeks) of
diabetic rats with NAC was shown to inhibit endothelial dysfunction and development of
abnormalities in the peripheral nerve (Pieper and Siebeneich, 1998; Sagara et al., 1996).
56
However, none of the previous studies to date have examined the role of ROS in
modulating skin microvascular blood flow in early diabetes.
The aim of chapter 7 was to investigate the quantitative contribution of superoxide
anions, hydroxyl radicals, and lipid peroxidation products to endothelial (inflammatory)
and vascular smooth muscle responses in 4 weeks STZ-induced diabetic rats. For this
purpose, the changes in diabetic skin blood flow are examined after using three free
radical scavengers / antioxidants (SOD, NAC, and Tirilazad mesylate). A relationship
between ROS and enhancement of ET-1 was previously reported (Chen et al., 2000).
Therefore, the possible interaction between the effects of ET and free radicals was also
assessed, using combined treatment of the ET-1 receptor antagonist, BQ-123, and
superoxide anion scavenger, SOD.
1.5.4. Alterations in smooth muscle reactivity in diabetes
1.5.4.1.
Vasoconstrictor responses
The reactivity to vasoconstrictors such as noradrenaline (NA), angiotensin II, serotonin,
and potassium chloride (KCl), has been widely investigated in numerous vascular tissues
from animals with chemically induced diabetes. However, results from these studies have
been inconsistent, reporting an increase, decrease, or no change in the reactivity of
diabetic tissues (Kawasaki, 1997; Lieu and Ried, 1994; Ralevic et al., 1995a).
Results with respect to the vascular reactivity of aorta to NA in rats varied with the
duration of diabetes. Enhanced responsiveness to NA was reported in aorta from 1 and 4
weeks (Orie and Aloamaka, 1993) as well as 8-12 weeks of diabetes (Kawasaki, 1997;
Wong, 1996). Similar responses were also observed in the following tissues from STZtreated rats: the sciatic nerve (Pa and Tomlinson, 1997), mesenteric vessels (Savage et al.,
1995), and kidneys from 4-6 weeks (Martinez-Nieves and Dunbar, 1999), small
cremaster muscles from 2-16 weeks (Morff, 1990), mesenteric vessels from 10-12 weeks
(White and Carrier, 1990), and the eye (Su et al., 1995) and tail artery after 4 weeks
(Ramanadham et al., 1984).
Diabetes-induced changes in reactivity to other vasoconstrictor agents have also been
examined. For example, studies using the alpha-adrenoreceptor agonists phenylephrine
and methoxamine, demonstrated significantly greater contractile responses in tissues
57
from 3 days and 4 weeks STZ-induced diabetic rats compared to control (Chang et al.,
1993; Wald et al., 1988). Further supporting studies have shown an increase in reactivity
to NA, angiotensin II, serotonin, KCl, and ET-1 in aorta (Abebe et al., 1994; Murat et al.,
1999), atria (Reid et al., 1989), skeletal muscle arterioles (Ungvari et al., 1999), bladder
(Waring and Wendt, 2000), renal arteries (Inazu et al., 1991), rat tail artery / vein
(Ramanadham et al., 1984; Tam et al., 1997), and gastric fundus (Sakai et al., 1994) after
induction of diabetes with STZ (early stages).
From studies in humans, Bodmer and co-workers (1999), demonstrated an exaggerated
sensitivity to NA-induced vasoconstriction in the dorsal hand vein from diabetic patients
with microalbuminuria, a predictor of micro- and macrovascular complications, as
compared to non diabetic patients and normoalbuminuric diabetic patients (Boder et al.,
1999). Furthermore, Christlieb et al (1976) reported an increase in vascular
responsiveness to infused angiotensin II and NA in subjects with retinopathy (Christlieb
et al., 1976). Together these two studies suggest that in a diabetic state a non specific
increase in the vascular smooth muscle function to vasoconstrictor agonists occurs
(Waring and Wendt, 2000).
A range of mechanisms has been proposed to explain vascular hyperresponsiveness to
vasoconstrictor agents in diabetes. One suggested mechanism relates the contractile
response to the enhanced extracellular influx of calcium through either voltage-dependent
or receptor-dependent calcium channels (White and Carrier, 1990). Indeed, the uptake of
radioactively labelled calcium induced by NA or serotonin, was shown to be significantly
greater in diabetic rat renal arteries (Inazu et al., 1991) than in control groups. Numerous
studies have demonstrated that the voltage-dependent calcium channel agonist, BAY K
8644, produces a greater force of contraction in the mesenteric, renal, and isolated
skeletal muscle arterioles from STZ-treated rats than in tissues from corresponding
control rats (Inazu et al., 1991; Ungvari et al., 1999; White and Carrier, 1990). These
authors suggested that diabetes might have increased the activity and / or number of
calcium ion channels. In another report by Waring and Wendt (2000), it was postulated
that there are no major impairments in either intracellular calcium regulation or the
contractile function in the bladder smooth muscle after 8 weeks of STZ-induced diabetes.
58
However, a non-specific enhancement in the contractile force was observed in these rats,
which was associated with an increase in the intracellular calcium level. This change
implied that the apparent sensitivity to intracellular calcium is enhanced in the bladder
smooth muscle from diabetic rats (Waring and Wendt, 2000).
Another mechanism, which may contribute to enhanced contractile responsiveness,
involves the pathway wherein phosphatidylinositol 4’, 5’-bisphosphate is hydrolyzed to
generate inositol triphosphate (IP3) and DAG. IP3 can release intracellular calcium,
whereas DAG activates PKC to initiate extracellular calcium influx and vascular
contraction (Abebe and MacLeod, 1990; Legan, 1989). One study showed a decrease in
PI hydrolysis after 4 weeks of diabetes and suggested that the mobilization and utilization
of calcium ions during contraction occur independently of the receptor-stimulated PI
hydrolysis (Legan, 1989). A change in the PKC-mediated contractile response may also
contribute to the hyperresponsiveness of diabetic vascular preparations. The agent
phorbol dibutyrate, which mimics DAG and its activation of PKC, produced contractions
in the aorta and mesenteric artery from diabetic rats which were greater than those from
control rats (Abebe and MacLeod, 1990; Kawasaki, 1997). Up-regulation of PKC was
also supported by other studies (Sakai et al., 1994; Ungvari et al., 1999), where each
study suggested alternative possible causes. For example, the former suggested an
increase in the activity of the voltage-dependent calcium channel as a possible cause and
the latter study ruled out the possibility of an alteration in calcium channel density.
Ramnadham and co-workers (1984) have demonstrated supersensitivity to alphaadrenoreceptor agonists NA and methoxamine with a simultaneous decrease in
catecolamine content, in a caudal artery from STZ-treated rats of 4 weeks duration. It was
suggested that autonomic neuropathy, leading to the depletion of caudal artery-NA stores,
could induce a supersensitivity phenomenon as a consequence of changes in alphaadrenoreceptor numbers and / or affinity (Ramanadham et al., 1984).
Another theory was that increased vascular smooth muscle reactivity to vasoconstrictor
agents in diabetes may occur as a consequence of a diabetes-induced impairment in the
opposing vasodilatation response mediated by endothelium-derived NO (Chang and
59
Stevens, 1992). Indeed, enhanced sensitivity of aorta from diabetic rats to phenyephrine
was abolished and restored to control levels after removal of the endothelium.
In contrast to these findings, there are a number of reports demonstrating a decrease or no
change in the reactivity of vascular smooth muscle to vasoconstrictor agents. Diminished
contractile responsiveness to NA was reported in the cremaster muscle (Mayer and
Messina, 1996), aorta (Wong and Tzeng, 1992), and caudal artery (Ramanadham et al.,
1984) of rats with 4 weeks of diabetes. Similar reduction in contractile responses were
obtained using vasoconstrictor agents including NA, adrenaline, serotonin, KCl, and ET1, in the kidney (1 week) (Garcia et al., 1999), mesenteric artery (2 weeks) (Makino and
Kamata, 1998), aorta (2, 8, 12 and 14 weeks) (Fulton et al., 1991; Hattori et al., 1996;
Hopfner et al., 1999; Orie and Aloamaka, 1993; Utkan et al., 1998), cremaster muscle
arterioles (4-8 weeks) (Mayer and Messina, 1996), atria ( 8-12 weeks) (Lieu and Ried,
1994), and caudal artery (12-16 weeks) (Tam et al., 1997) of STZ-induced diabetic rats.
Several possible mechanisms could contribute to the diminished contractile ability of
diabetic preparations. These include impaired extracellular influx of calcium, due to
alterations in calcium channel activity or alterations in intracellular calcium binding and
storage, contributing to defective excitation-contraction coupling; reduced ATP levels,
leading to decreased contractile ability; and defective contractile protein sensitivity or
function (Hattori et al., 1996; Pfaffman et al., 1982; Zemel, 1995).
Other studies have demonstrated no change in reactivity to vasoconstrictor substances by
diabetes in rat caudal artery (Vo and Reid, 1995), aorta (Fulton et al., 1991) and portal
vein (MacLeod and McNeill, 1985). Numerous reports have shown that effects of
diabetes on vascular smooth muscle reactivity are dependent on the duration of the
disease. Orie and Aloamaka (1993) showed the responsiveness of isolated aortic rings to
NA and 5-hydroxytryptamine (5-HT) was significantly enhanced after 1 and 4 weeks of
diabetes but not after 12 weeks (Orie and Aloamaka, 1993). In another study in vivo,
direct measures were made of the arteriolar response to NA in an intact microvascular
bed using rats in acute and chronic stages of diabetes (2 to 32 weeks) (Morff, 1990).
Results showed normal responses from large arterioles at all stages of diabetes studied,
60
whereas smaller arterioles were hyperresponsive at the acute stage of the disease but
returned to normal sensitivity as the duration of diabetes progressed to the chronic stage
(32 weeks). The dependency of diabetes-induced changes to the duration of this disease
has similarly been demonstrated by a number of other investigators (Legan, 1989;
MacLeod and McNeill, 1985; Wong, 1996).
It is apparent that the reported effects of diabetes on smooth muscle reactivity to
vasoconstrictor agents are inconsistent. Factors that may contribute to differences
between studies include variation in vascular preparations, the duration of diabetes, dose
of the diabetogenic agent, and the route of its administration. Nevertheless, the majority
of reports indicate that contractile responsiveness to NA and to other vasoconstrictor
agents is enhanced by the diabetic state. This effect in itself may particularly be
dependent on the duration of the disease and the type of tissue under investigation.
1.5.4.2.
Vasodilator responses
The importance of vascular endothelium in generating a number of vasoactive substances
has been well established. A physiological role for endothelium-derived NO has been
proposed in the regulation of vascular smooth muscle tone and in the modulation of
vascular smooth muscle reactivity (Becker et al., 2000; Ralevic et al., 1995b; Rongen et
al., 1994; Singh and Evans, 1997) (see also section 1.9.1). Therefore, any alteration to the
endothelium may contribute to the development of vascular complications including
those that occur as a result of diabetes mellitus. Endothelial cell damage has been
previously documented in animals and patients with diabetes (Chan et al., 2000; Lorenzi
and Cagliero, 1991).
Morphological changes include enhanced endothelial adhesion of leucocytes, platelets,
and fibrin-like material; increased endothelial cell proliferation; increased number of
multinucleated endothelial cells; thickening of basement membrane; increased
endothelial cell surface area; increased fluidity of endothelial cell membrane; and
increased capillary diffusion capacity (Chan et al., 2000).
61
Previous reports are indicative of abnormal endothelium-dependent vasorelaxation in the
diabetic state (Kobayashi and Kamata, 1999a; Pieper, 1999). However, studies assessing
endothelial function in both animal and human models of type I diabetes have produced
conflicting results. In experimental animal models, variation in some factors between
different studies including duration of diabetes, vascular bed under study, and the
presence or absence of diabetic complications may account for conflicting results.
There is substantial evidence that vasodilatation, mediated by endothelium-derived NO,
is impaired in animal models of diabetes (Azadzoi and De Tejada, 1992; Brands and
Fitzgerald, 1998; Palmer et al., 1998; Teraka et al., 1999). The importance of disease
duration has been clearly demonstrated in aortic tissues. For example, endotheliumdependent relaxations produced by ACh and histamine were significantly attenuated in 810 weeks STZ-induced diabetic rats (Kobayashi and Kamata, 1999a; Oyama et al., 1986),
whereas responses to the endothelium-independent relaxant agent, sodium nitroprusside
(SNP), were unaffected (Oyama et al., 1986). However, evidence exists that this defect in
diabetic aorta may occur as early as 2 weeks after induction of diabetes (Zhu et al., 1999)
whereas one study showed no change at 4 weeks (Kobayashi and Kamata, 1999a). These
reports, however, seem to agree with results in a recent study by Pieper (1999) who
demonstrated a relationship between the triphasic response of aorta and the disease
duration. The author showed an increase in endothelium-dependent relaxation to ACh
after 24 hrs, normal response after 1-2 weeks, but impaired after 8 weeks of diabetes,
compared to controls. Again, in this study, the vascular response to another endotheliumindependent relaxing agent, nitroglycerine, was unaltered at all stages (Pieper, 1999).
Further studies demonstrated diabetes-induced attenuation of endothelium responses, in
vitro, in rat hind limb vessels (Brands and Fitzgerald, 1998) and in vivo, in cerebral,
hindquarter (Diederich et al., 1994; Kiff et al., 1991b; Mayhan et al., 1991), and sciatic
nerve arterioles of rats (Teraka et al., 1999). There is a substantial difference in timing
between impairment of the endothelial-dependent relaxation and the onset of diabetes in
different vascular beds. The range is from 7-11 days in rat hind limb (Brands and
Fitzgerald, 1998) and intestinal arterioles (Lash and Bohlen, 1991) to between 3 and 6
weeks in rat cremaster muscles (Alsip et al., 1996) and mesenteric arteries (Diederich et
al., 1994; Kiff et al., 1991b; Palmer et al., 1998). Relaxations to SNP and glyceryl
62
trinitrate were not significantly altered in different diabetic tissues (Brands and
Fitzgerald, 1998; Kiff et al., 1991b).
A similar disparity has been observed in studies using resistance vessels, with
endothelium-dependent relaxation being found to be normal (Gebremedhin et al., 1999),
impaired (Alsip et al., 1996; Taylor et al., 1995) or even enhanced (Heygate et al., 1995).
Again, the disease duration and differences in the vascular bed under study could play an
important role in observed discrepancies.
Abnormalities in diabetic vascular tissues were shown to occur in humans. For example,
in type 1 diabetic patients, using the occlusion plethysmography technique, impairment
(Elliott et al., 1993; O'Driscoll et al., 1997), no change (Huvers et al., 1999), or
enhancement (Makimattila et al., 1997) was observed in endothelium-dependent
relaxation. Most studies using the above technique found that the vascular response to
SNP was not changed. Using the vascular Doppler ultrasound technique also showed
conflicting results, with endothelium-dependent vascular responses being impaired
(Zenere et al., 1995) or unchanged (Enderle et al., 1998). The vascular response to
glyceryl trinitrate was found to be either impaired (Zenere et al., 1995) or unchanged
(Enderle et al., 1998). The difference in these results could be due to variations in the
degree of acute hyperglycaemia and the effect of glycaemic control (Huvers et al., 1999;
O'Driscoll et al., 1997), sex differences (Huvers et al., 1999; Makimattila et al., 1997),
and effects of diabetic complications (Elliott et al., 1993; Makimattila et al., 1997). A role
for a decrease in the stimulated release of endothelium-derived NO in diabetes was not
supported by these studies (both diabetic animals and humans) since no change or
enhancement in endothelium-dependent vascular response was observed.
Mechanisms by which diabetes contributes to endothelial dysfunction are not completely
understood, but hyperglycaemia may initiate it. The first substantive evidence linking
hyperglycaemia and endothelial dysfunction was shown by Tesfamariam and co-workers
(Tesafemariam et al., 1990; Tesfamariam et al., 1991). These authors showed exposure of
arteries to elevated concentrations of glucose (in vitro) caused impairment in
endothelium-dependent relaxation.
63
Hyperglycaemia-induced endothelial dysfunction may result from the decreased synthesis
and / or release of endothelium-derived NO, decreased responsiveness of the smooth
muscle to relaxing factor, release of endothelium-derived constricting factors, and
accelerated destruction of NO. Impaired functioning of other endothelium-derived
vasodilator factors such as prostacyclin and endothelium-derived hyperpolarizing factor
(EDHF) could also be induced by the diabetic state. EDHF is defined as the substance
which produces vascular smooth muscle hyperpolarization and this effect cannot be
explained by NO or by a cyclo-oxygenase product such as prostacyclin.
Decreased production of NO may be responsible for impotence in some men with
diabetes, since isolated strips of corpus cavernosum from these men were less sensitive to
the relaxant effect of ACh and nerve stimulation than from non-diabetic men (Saenz De
Tejada et al., 1989). Pre-treatment of isolated tissues from STZ-treated rats with Larginine (L-Arg), the precursor of NO synthesis, reversed the impairment in relaxation to
ACh in diabetic aortic rings, but had no effect on responses from the control group
(Pieper and Peltier, 1995). This suggests that utilization of L-Arg by NOS to produce NO
may be defective in diabetes. Aldose reductase, an enzyme present in endothelial cells,
requires NADPH for converting glucose to sorbitol in glucose metabolism through the
polyol pathway. NADPH is also an essential cofactor for NOS and synthesis of NO, and
its depletion, as a result of hyperglycaemia, could lead to a reduction in NO synthesis.
Aldose reductase inhibitors have shown early promise in reversing glucose-induced
changes in sorbitol and myo-inositol metabolism and endothelial dysfunction in
experimental diabetic animals (Cameron et al., 1997a; Sarges and Oates, 1993) (see also
section 1.5.1).
Some studies show that the release of NO is impaired in experimentally induced DM. In
one study, the vasodilator effect of insulin in skeletal muscle was shown to be mediated
via an increase in NO release (Steinberg et al., 1994). Therefore, in the presence of
insulin deficiency in type I diabetes there could be a reduction in NO release and this
could contribute to the endothelial dysfunction. Basal NO release can be inhibited by LArg analogue inhibitors of NOS, NG-nitro-L-arginine methyl ester (L-NAME). The
64
ability of this substance to increase the mean arterial blood pressure and to produce
vasoconstriction in the rat hindquarter vasculature was found to be attenuated in diabetic
animals (Kiff et al., 1991b). Also L-NAME was less effective in producing decreases in
skin blood flow in diabetic rats than in control rats (Lawrence and Brain, 1992). These
studies raised the possibility that beside a possible reduction in NO release, vascular
sensitivity to NO may also be diminished in diabetes.
There is substantial evidence that hyperglycaemia-induced endothelial dysfunction is
mediated by free radicals (sections 1.5.3) produced through increased arachidonic acid
(AA) metabolism (Giugliano et al., 1996; Tesfamariam and Cohen, 1992). It has been
shown that in human aortic endothelial cells, prolonged exposure to high concentrations
of glucose increases eNOS gene expression and NO release, which is also associated with
a pronounced concomitant increase in superoxide anion production (Cosentino et al.,
1997). These anions inactivate NO (Gryglewski et al., 1986) and furthermore, they
interact with NO, leading to production of peroxynitrite (Jourd'heuil et al., 1997), which
may contribute to endothelial dysfunction (Bouloumie et al., 1997; Galle et al., 1992)
(section 1.5.3). Superoxide dismutase has been shown to normalise NO-mediated
vasorelaxation impaired by increased glucose concentration (Bassirat and Khalil, 2000;
Bauer et al., 1999). Interaction between superoxide anions and hydrogen peroxide as well
as peroxynitrite itself can also lead to generation of hydroxyl radicals (section 1.5.3, Fig
1-4). A previous study suggested that the synthesis and / or release of hydroxyl radicals
might be responsible for the impairment of NOS-dependent vasodilation in diabetic rats
(Mayhan and Patel, 1998). This was further supported by Pieper et al (1997), where
inhibiting hydroxyl radical formation by DETAPAC markedly enhanced relaxation to
ACh in aortic rings from STZ-induced diabetic rats (Pieper et al., 1997b). These studies
show that hyperglycaemia could induce increased NO production as well as a reduced
NO availability due to inactivation mediated by free radicals. The role of free radicals
including hydroxyl radicals in diabetes is discussed in more detail in section 1.5.3.
65
Activation of PKC, by an increase in hyperglycaemia-induced DAG, has also been
suggested as a mechanism for endothelial dysfunction and vascular complications in
diabetes (Park et al., 1999) (section 1.8.5.4). Indeed PKC activation is responsible for
several vascular alterations in diabetes (Park et al., 1999) including an increase in
vascular prostanoids (Tesfamariam et al., 1991). Isolated rabbit aorta exposed to
increased glucose concentrations showed impaired endothelium-dependent relaxation to
ACh after 10 min treatment with 4-phorbol 12-myristate 13-acetate, a PKC activator
(Tesfamariam et al., 1991). The same study demonstrated that indomethacin increased
relaxation induced by ACh, suggesting a role for vascular prostanoids, and this abnormal
relaxation was restored with sphingosine, a PKC inhibitor (Tesfamariam et al., 1991).
Another proposed mechanism contributing to endothelial dysfunction and diabetic
vascular complications is the formation of AGEs. Accumulation of AGEs over time has
been shown to reduce NO availability. Previous studies in vitro and in vivo, using a rat
model of STZ-induced diabetes, showed that reactive intermediates form in the early
stages of the glycation pathway, which then react with and quench NO rapidly (< 5 sec)
(Bucala et al., 1991) (section 1.5.2). In addition to NO inactivation, AGEs have been
shown to impair the effect of NO on mesangial cell antiproliferation, an early and
characteristic lesion of diabetic vasculopathy (Hogan et al., 1992). Furthermore, in
diabetic rats, AG has been shown to partially restore endothelium-dependent relaxation
(Bucala et al., 1991) and retard the development of diabetic nephropathy (Soulis-Liparota
et al., 1996). It should, however, be noted that AG has multiple actions including direct
effects on NO generation (Corbett et al., 1992). AGEs could also cause endothelial
dysfunction by other means, which are described in section 1.5.2.
Vasodilator agents other than NO, such as prostacyclin and EDHF, are produced by the
vascular endothelium, and may also be involved in defective endothelium-dependent
vasodilation. The decreased vascular production of prostacyclin in both diabetic animals
and humans has been previously reported and alterations in prostacyclin receptor number
and / or turnover were also apparent (Hodgson et al., 1992). ACh is capable of producing
endothelium-dependent hyperpolarization in vascular smooth muscle, independent of NO
66
release (Garland and Mcpherson, 1992). While the identity of EDHF is not very clear, the
potassium ion (K+) is gaining support as a possible candidate (Edwards et al., 1998;
Makino et al., 2000) (see also section 1.8.5.1). There are controversial reports on whether
diabetes can attenuate functions of EDHF. Recent evidence showed that vasodilatation
induced by ACh and K+ are both greatly impaired in mesenteric arteries of diabetic rats
(Fukao et al., 1997; Makino et al., 2000). However, Endo et al (1995) demonstrated that
impaired relaxation to ACh in diabetic rat aorta could not be attributed to a defect in the
function of the hyperpolarizing factor-mediated component (Endo et al., 1995).
An array of findings exist describing diabetes-induced functional changes in the vascular
smooth muscle reactivity to both vasoconstrictor and vasodilator agents, in both animal
models and humans. The majority of reports indicate that vasoconstrictor responsiveness
is increased in diabetes and endothelium-derived NO reactivity is impaired. Such
alterations may therefore contribute to the development of clinical vascular complications
observed in diabetes. It should be noted that vascular responses to endotheliumindependent relaxants (e.g. SNP) were mainly unchanged in different diabetic tissues at
all stages of diabetes. It is well known that inflammation is an important factor for the
tissue repair process. Since none of these experiments were performed on skin with
inflammation, further work is required to obtain the relevant information necessary for a
deeper understanding of mechanisms involved in delaying wound repair in diabetes.
1.6.
Animal models of diabetes
1.6.1. History
Permanent diabetes in animals offers an excellent model for studying the interaction
between hereditary and environmental factors such as drugs, toxic, and infectious agents.
This is also useful in the determination of therapeutic and preventative measures (Shafrir,
1990).
Surgery, viral infections, hormone administration, and chemical agents can all produce
diabetes (Mendez and Ramos, 1994). DM has been documented in monkeys, dogs, rats,
cats, shrews, squirrels, seals, dolphins, hippopotamus, antelopes, and different farm
animals. Some animals may develop spontaneous diabetes, which is more frequent in
67
rodents (rats, mice, hamsters), but it is considered as an inadequate model for studying
natural development of this disease (Mendez and Ramos, 1994). It is well known that the
ethics code of practice in using animals for research purposes should be respected and
obeyed by any qualified person who conducts these experiments. All techniques and
surgical procedures conducted in this PhD study were approved by the Royal Melbourne
Hospital Research Foundation (Animal Ethics Committee) and adhered to IASP
(International Association for the Study of Pain) guidelines.
1.6.2. Chemical induction of diabetes
The use of chemical agents to cause diabetes allows for precise studies to be carried out
on the biochemical, hormonal, and morphological changes that occur during the
induction of the diabetic state and afterwards. Alloxan and streptozotocin are the most
widely used diabetogenic agents because of their selectivity for beta cells of the islets of
Langerhans (Mendez and Ramos, 1994). A diabetic dosage is defined as the amount of an
inducing agent which in 80% of the animals of a given species produces sustained
hyperglycaemia and beta cell necrosis, but does not cause harm to other organs (Goldner
and Gomori, 1944). Diabetogenic dose can vary depending on the type of animal used,
the weight/age, whether they are fasted or fed before the treatment, and the route of its
administration.
1.6.2.1.
Alloxan
Alloxan was first synthesised in 1838 by F. Wohler and J. Leibig but its diabetogenic
property was first reported in 1943, showing islet necrosis in rats (Dunn and McLetchie,
1943). Alloxan is an unstable crystal that is freely soluble in water and causes diabetes in
experimental animals by destroying the insulin secreting beta cells of the pancreas,
leaving alpha-cells unharmed. Morphological evidence from rat islets treated with
alloxan exhibited multiple cellular necrosis, marked degranulation, and extensive
vesiculation of endoplasmic reticulum and golgi complex (Abdel-Rahman et al., 1992a).
Mitochondrial enlargements with disrupted cristae and mitochondrial ruptures were
prominent and this was one of the targets through which alloxan destroyed beta cells.
68
Alloxan’s stability in aqueous solution is mainly dependent on pH and temperature.
Below pH 3 it is fairly stable at room temperature, whereas at pH 7 the solution must be
kept under 4 °C and its life in blood is less than 1 min (Rerup, 1970) (Fig1-5).
o
o
4
HN 3
5
2
6
Lenzen, S. & Munday, R. (1991)).
o
1
NH
Fig 1-5: Structure of alloxan (Adopted from
o
Four factors that may be important in determining the toxicity of alloxan are stability,
hydrophilicity, reactivity toward thiol groups, and generation of free radicals (Lenzen and
Munday, 1991). Hydrophilicity may facilitate its access to those areas of pancreatic beta
cells in which its toxic effects may be expressed. Glucose transporter-2 (GLUT-2) is
expressed in rodent’s pancreatic beta cells mediating glucose uptake. It was shown that
selective uptake by GLUT-2 is not a prerequisite for alloxan diabetogenecity (Munday et
al., 1993). Alloxan reacts with thiols and inhibits thiol enzymes such as glucokinase and
hexokinase (Lenzen and Panten, 1988).
The mechanism behind alloxan’s diabetogenic effect is not very clear. However, it has
been suggested that it damages pancreatic beta cells through formation of superoxide
anions and hydrogen peroxide but not hydroxyl radicals (Jorns et al., 1999; Sentman et
al., 1999). This causes DNA strand-breaks (Yamamoto et al., 1981) and depletes NAD+
stores (Askar and Baquer, 1994; Le Doux et al., 1988). One study suggested that the
generation of free radicals through redox cycling with glutathione was not enough to
cause diabetes by alloxan but it seems that hydrophilicity of this compound is a necessary
condition for its diabetogenecity (Munday et al., 1993). Administration of alloxan at
higher than a diabetogenic dose can cause serious kidney damage (Rerup, 1970).
69
1.6.2.2.
Streptozotocin
STZ is a broad-spectrum antibiotic isolated from streptomyces achromogenes (Herr et al.,
1959). It is a fairly effective carcinogen (Arison and Feudale, 1967) as well as an
antileukaemic agent (Evans et al., 1965). During the pre-clinical toxicology studies, it
became apparent that when the drug is administered intravenously at a sufficient dose, it
produced hyperglycaemia in rats and dogs within a few hours of the treatment (Rakienten
et al., 1963). The structure is composed of a nitrosouria moiety with a methyl group
attached at one end and a glucose molecule at the other (Fig1-6).
CH2OH
O
H
HO
OH
OH
Fig 1-6: Structure of Streptozotocin (adopted
from Weiss, R.B. (1982)).
NH
C
O
N
NO
CH3
STZ is a colourless crystalline water-soluble powder. When it is dissolved in saline or
distilled water at room temperature and neutral pH it decomposes within a few minutes
(visible bubbles). In solution, this compound is stable at pH 4 and a low temperature
(Weiss, 1982). Its half-life in plasma is between 5 to 35 min and concentrates in the liver,
kidney and pancreas but not in the brain.
The kidney rapidly eliminates STZ but renal dysfunction is the major dose-limiting toxic
effect of this drug. Vomiting is the most noxious side effect of STZ reported in treated
patients. Other potential troublesome side effects of this drug are myelosuppression and
hepatotoxicity, which are both uncommon (Weiss, 1982).
STZ produces permanent diabetes in some animals (rat, mouse, dog, hamster, monkey
and guinea pig) by destroying pancreatic beta cells similar to alloxan. Once again, the
diabetogenic dose of STZ is depend on the species, weight / age, if they are fasted or fed,
and the route of its administration. In rats, STZ produces hyperglycaemia with normal
levels of blood ketones, plasma-free fatty acids and heart glycogen (Weiss, 1982). This is
different from alloxan-treated rats where all these parameters are markedly elevated
70
(Mansford and Opie, 1968). In rats, the severity of experimental diabetes and its
persistence depends on the dose of STZ used. Intraperitoneal injection of STZ at a dose
of 45mg/kg is known to cause a transient state of diabetes (Simon and West, 1992).
Spontaneous recovery from diabetes was previously reported in rats injected with STZ in
doses up to 50mg/kg (Iwase et al., 1991).
A single high dose of STZ can cause fragmentation of DNA in beta cells, through the
formation of free alkylating radicals, leading to a reduction in the cellular levels of
nucleotides and related compounds, in particular NAD+, causing rapid necrosis (Thulesen
et al., 1997). This is supported by Hoorens et al (1999) study which showed
nicotinamide, an inhibitor of the NAD+-degrading enzyme (poly (ADP-ribose)
polymerase), could protect islet cells against STZ-damage through protection of the
intracellular NAD+ pool (Hoorens and Pipeleers, 1999).
The diabetogenic action of STZ in animals is not identical to alloxan but is very similar in
many aspects. It was suggested that the nitrosouria moiety is necessary for DNA
alkylation and antitumour efficacy, and the sugar moiety is necessary to facilitate islet
tissue uptake and minimise bone marrow toxicity (Weiss, 1982). GLUT-2 may mediate
the cellular uptake of streptozotocin (Schnedl et al., 1994).
1.6.2.2.1.
Role of nicotinamide and vitamin E
Vitamins E and nicotinamide have been previously shown to play an important role in the
delay or prevention of diabetes onset in a model of spontaneous experimental diabetes or
STZ-treated mouse islet cells (Beales et al., 1994; Kaneto et al., 1999; Strandell et al.,
1989). Vitamin E may protect islet cells by reducing cytotoxocity mediated by cytokines
and their products as well as inhibiting lipid peroxidation of cell membranes preventing
damage by free radicals (Beales et al., 1994; Kaneto et al., 1999). The earliest report
about nicotinamide protection against diabetes onset was as early as 1947 and in alloxantreated rats (Lazarow, 1947), an effect later reproduced with STZ (Dulin et al., 1969).
Nicotinamide, the amide derivative of nicotinic acid, offers protection to pancreatic beta
cells against toxic stimuli. At high doses it is a free radical scavenger, and a potent
inhibitor of poly (ADP-ribose) polymerase, and protects against depletion of intracellular
NAD (Gale, 1996). Nicotinamide pre-treatment in, in vivo and in vitro studies, was
71
capable of protecting islet cells against damage as well as showing other beneficial
effects (Petley et al., 1995; Strandell et al., 1989).
1.7. Neurogenic inflammation
1.7.1. History
Upon tissue injury, inflammation occurs which is a protective response that initiates the
process of tissue repair. Inflammation is characterized by a vascular response involving
alterations in vessel calibre and blood flow; swelling and exudation which results from
increased vascular permeability; the accumulation of inflammatory cells; and changes in
ground substance. Sensory nerves play an important role in mediating “neurogenic
inflammation” (Brain, 1996; Holzer, 1998) and subsequent tissue repair. The notion that
certain sensory nerves are involved in cutaneous inflammation can be traced back to the
classical work of Stricker and Bayliss. Stricker (1876) found that stimulation of DRG (the
site of the nerve bodies of primary afferent nerves) caused them to conduct impulses
centrally, but also showed them to have an efferent function that produce vasodilatation
(Stricker, 1876). Later, Bayliss (1901) demonstrated that the observed cutaneous
vasodilatation was infact an axon reflex, whereby impulses produced in afferent nerve
fibres travel antidromically through collateral fibres (Bayliss, 1901; Maggi, 1995).
The classic signs of inflammation of skin are best described by Lewis (1927) as a “triple
response” (Lewis, 1927). This response consists of vasodilatation or an increase in skin
blood flow through dilated arteries (manifest as redness and warmth); plasma
extravasation (PE) or an increase in the permeability of post-capillary venules to bloodborne proteins, allowing these to escape into the extracellular fluid (seen as swelling);
and hyperaemia (flare) arising from axon reflexes that takes place between collaterals of
single sensory nerve fibres. Injury to the skin and stimulation of small diameter afferent
sensory neurons can send impulses orthodromically to the brain producing the pain
component of the inflammatory process and antidromically to the spinal cord producing
responses similar to inflammatory “triple response”.
72
1.7.2. Sensory nerves
Cell bodies of human skin sensory nerves are found in the DRG of the spinal cord. Axons
from these cells travel centripetally to connections in the spinal cord and peripherally in
main trunks of mixed nerves to reach the cutis, where they serve functions of pain,
temperature and other tactile sensations (Wood and Bladon, 1985). In the skin many
sensory neurons terminate in specialized receptors but many end in unmyelinated freenerve endings. These latter tend to form plexus from which they branch between the cells
to detect touch and pain stimulus.
Pain results from the activation of nociceptors that are divided into 2 types: highthreshold mechanoreceptors (usually arising from Aδ fibres) (Burgess and Pearl, 1967),
and polymodal nociceptors that are associated with 80% of C-fibres. The former responds
to strong pressure, heat, and skin damage (Lynn, 1989). The latter responds to heat,
pressure, and irritant chemicals and are the most common sensory unit in the skin. An
important point is that polymodal nociceptors can be especially sensitized by the presence
of endogenous substances including ACh, BK, histamine, and prostaglandins.
Primary afferent sensory nerves contain several neuropeptides including SP and other
tachykinins (neurokinin A (NKA), neuropeptide K, neuropeptide gamma, etc), CGRP,
vasoactive intestinal peptide (VIP), cholecystokinin, angiotensin II, somatostatin, galanin
and dynorphin. Upon activation, some of these sensory neuropeptides are released at the
peripheral terminal to initiate inflammatory responses (Maggi, 1991).
It has been shown that unmyelinated small diameter nociceptive C-fibres are prerequisite
for neurogenic inflammation induced by chemical or mechanical irritations (e.g.
capsaicin, electrical nerve stimulation (ES)) (Brain, 1996; Holzer, 1991; Szolscanyi et al.,
1988). It was suggested that in diabetes mellitus these fibres are altered (Gyorfi et al.,
1996; Walmsley and Wiles, 1991) and neurogenic inflammatory and anti-inflammatory
responses are impaired (Nemeth et al., 1999b). The aim of the study, described in
chapter 3, is to examine the effect of diabetes on sensory nerve function as determined by
the ability of these nerves to mount an inflammatory response.
73
1.7.3. Contribution of SP and CGRP to Neurogenic Inflammation
Neurogenic inflammatory responses can be induced experimentally in a number of ways.
They include the exogenous application of capsaicin, introducing exogenous sensory
neuropeptides themselves (e.g. SP, CGRP) or indirectly through modulators such as
histamine. However, the most popular technique involves the use of antidromic nerve
stimulation, which is described in the next section (1.7.4). The tachykinin SP, with 11
amino acids, has been located in the small unmyelinated nerve fibres within the sensory
nerves that terminate in the superficial dorsal horn of the spinal cord; and is present in
both the peripheral and central endings of these nerves (Hokfelt et al., 1981). Most of the
SP synthesized in the cell bodies of the DRG is transported to peripheral nerve terminals
where it is released by nerve stimulation (e.g. antidromically) (Khalil et al., 1994). SP can
produce many of the physiological changes associated with inflammation including
vasodilatation and PE (Khalil and Helme, 1990b), the activation of macrophages
(Brunelleschi et al., 1990), and degranulation of mast cells with a local release of
histamine (Johnson and Erdos, 1973; Khalil and Helme, 1989).
When SP is released, it can act upon neurokinin-1 (NK-1) receptors on the endothelium
to activate phosphatidyl-inositol-bisphosphate (PI) pathway. This increases intracellular
calcium and activates NOS causing the release of NO. NO can then diffuse into smooth
muscle resulting in its relaxation and subsequent vasodilation (Khalil and Helme, 1989;
Ralevic et al., 1992). SP-induced increase in blood flow is short acting and accompanied
by tachyphylaxis (Khalil and Helme, 1996; Khalil et al., 1994). SP is also capable of
activating membrane-linked G-proteins on the membrane of mast cells which causes
degranulation and subsequent histamine release (Johnson and Erdos, 1973; Khalil and
Helme, 1989). The latter then acts on histamine-1 (H1) receptors on sensory nerve
terminals causing further release of sensory neuropeptides. SP acts more on mast cells
when it is administered exogenously and at relatively high concentrations (Tausk and
Undem, 1995).
SP (through NK-1 receptors) and NKA are the main messengers for increase in vascular
permeability (Holzer, 1998). The early phase of the neurogenic- and SP-induced
extravasation arises from a direct action of the peptide on the endothelium post-capillary
74
venules, whereas the later phase depends in part on the release of 5-HT, histamine,
prostaglandins, leukotrienes, and other mediators from mast cells (Brain, 1996; Walsh et
al., 1995). The initial phase of the exudative response takes place in the superficial
dermis where peptidergic afferent nerve fibres are located with few mast cells. Although
there are contact sites between afferent fibres and mast cells in the skin, most mast cells
are located in a deeper layer of the dermis and hence are mainly involved in the postacute phase of neurogenic inflammation (Kowalski et al., 1990).
CGRP, a 37 amino acid peptide, is found in C- and Aδ-fibres and commonly co-localized
with SP (Mione et al., 1990). Release of CGRP can occur via ES of sensory nerves (at
specific stimulus parameters) (section 1.7.3), and / or activation of receptors for
inflammatory mediators (such as capsaicin, histamine, and BK). NO and prostaglandins,
which are produced at inflammatory sites also have potential to modulate the release and
actions of CGRP (Brain, 1996). Receptors for CGRP have been identified on vascular
smooth muscle and two receptor subtypes have been verified and designated as CGRP1
and CGRP2, according to their sensitivity to the receptor antagonist CGRP8-37. The latter
subtype is unaffected by actions of the antagonist (Dennis et al., 1990). CGRP1 receptors
are found in skin and cardiovascular tissues, whereas CGRP2 receptors are located in the
vas deferens, aorta, and urinary bladder.
CGRP is a potent vasodilator in many preparations (Holzer, 1992), but the most
important physiological action of CGRP is its vasodilatation effect (long acting) on the
microvascular system (Brain, 1996). The peptide-induced hyperaemia (e.g. in skin),
which is mediated by CGRP1 receptors that are coupled to the adenylate cyclase system,
arises from a direct action on the vascular smooth muscle and does not involve NO or
prostaglandins (Brain and Williams, 1988). CGRP can also induce vasodilation mediated
by NO such as in rabbit skin (Hughes and Brain, 1994). One way to show that CGRP is
involved in neurogenic vasodilatation was by pharmacological antagonism of the
endogenously released peptide. It was reported that dilatation of cutaneous arterioles
evoked by electrical or chemical stimulation of afferent fibres was inhibited by the
CGRP1 receptor antagonist, CGRP8-37. CGRP is unable to induce an increase in PE,
however it is capable of enhancing exudative responses to sensory nerve stimulation,
tachykinin and a variety of inflammatory mediators (Brain, 1996). This facilitation of PE
75
is thought to result from the peptide’s vasodilator activity, although inhibition of SP
degradation by CGRP may also play a role (Holzer, 1992). Initially, SP was promoted as
a major mediator of neurogenic inflammation; this was likely due to historical reasons in
that SP was discovered much earlier than CGRP. However, with the development of
specific receptor antagonists for the two sensory neuropeptides, interactive involvement
of CGRP and SP in vascular responses has been determined. Although some interaction
effects were shown between these two peptides, several lines of evidence indicate that SP
is the major mediator of the PE responses. This is supported by the ability of NK1
receptor antagonists and SP immunoneutralization to inhibit neurogenic PE to a variety of
stimuli (Brain, 1996; Walsh et al., 1995). On the other hand, CGRP has now been shown
to be more prominent in local neurogenic vasodilator responses and is the principal
transmitter involved in neurogenic dilatation of arterioles in all species that have been
tested (Brain, 1996). This is supported by studies using CGRP8-37 and an NK-1 receptor
antagonist (SR140333), both showing inhibition of PE responses following heat injury in
the rat skin (Siney and Brain, 1996). SP is also known to be capable of terminating an
existing vasodilator response to CGRP (Brain and Williams, 1988).
In addition to their action on the vascular system, SP and CGRP can influence the activity
of granulocytes, monocytes and lymphocytes (Holzer, 1998). They stimulate not only
adhesion of leucocytes to the vessel wall and their migration into the inflamed skin tissue
(Brain, 1997) but can also affect the release of mediator substances from white blood
cells. SP induces monocytes to release cytokines and prostanoids, stimulates basophils to
release histamine, and activates neutrophils to produce free radicals (Brain, 1996; Brain,
1997; Holzer, 1992). Many of these mediators are vasoactive and are likely to exert a
modulatory action on vessel diameter and permeability.
One possible reason for reduction in neurogenic inflammatory responses observed in
experimental animals and patients with diabetes (Gyorfi et al., 1996; Nemeth et al.,
1999b; Walmsley and Wiles, 1991) may be related to the reduction in tissue sensory
neuropeptide content (Diemel et al., 1992; Karanth et al., 1990). So far reports are
controversial and it seems that the type of tissue under investigation and duration of
diabetes play a significant role in observed differences. A previous study on skin samples
76
(lip & footpad) of STZ-induced diabetic rats showed no change in the distribution or
relative density of immunoreactive nerve fibres at 2, 4 or 8 weeks (Karanth et al., 1990).
However, in the same study, a marked increase in CGRP-immunoreactive fibres was
observed at 12 weeks in the dermis and epidermis in these rats. In a different study, the
SP-like immunoreactivity (SPLI) was increased in skin at 11 weeks but no change was
observed in atrial myocardium tissue of STZ-treated rats (Willars et al., 1989). Also,
reductions in SPLI and preprotachykinin mRNA level were demonstrated in the
gastrointestinal tract (11 weeks) (Willars et al., 1989) and anterior pituitary of STZtreated rats (6 weeks) (Rittenhouse et al., 1996b). SPLI in arterial wall of 6 weeks
diabetic rats was, however, increased by 2 fold (Markle and Walker, 1996) but strict
insulin treatment partially (significantly) reversed this effect towards normal. Recently,
treatment with insulin was also demonstrated to prevent loss of neurogenic inflammatory
responses (vasodilatation and PE) in diabetic rats (8 weeks) and NGF pre-treatment
managed to return these responses towards normal (Bennett et al., 1998). This author
showed that insulin was more effective in reversing the neurogenic oedema but that both
agents appeared equally effective in restoring the neurogenic vasodilator response. These
results extended previous findings, where insulin showed beneficial effects on neurogenic
inflammatory responses in diabetic rats and NGF was able to restore sensory nervemediated microvascular function in diabetes (Biessels et al., 1996; Diemel et al., 1992).
The improvement in neurogenic responses by insulin may be related to inhibition of
peptide loss. Insulin, by reducing high levels of glucose in the nerve, can prevent
formation of glucose by-products against neuronal oedema and nerve ischaemia and
hence prevent impairment in peptide expression and transport (Diemel et al., 1992). The
role of NGF could be related to improvement in SP production, since SP and CGRP
genes are known to be regulated by NGF (Lindsay et al., 1989). Results in Bennett’s
study (1998) also suggested that neurogenic responses in diabetic rats were not markedly
affected by modulation of SP or CGRP neuropeptide receptor-mediated mechanisms.
1.7.4. Electrical stimulation of the sciatic nerve
Neurogenic inflammatory responses can be induced experimentally by antidromic nerve
stimulation (chapter 2, section 2.7) whereby sensory nerve fibres are electrically
77
stimulated to mimic the axon-reflex phenomenon, leading to the release of sensory
neurotransmitters. The rat sciatic nerve is a frequently used peripheral nerve in
experimental studies. This nerve originates from the spinal lumbar ganglia L3-L6 (Swett
et al., 1991), and its branches innervate muscles and skin of the leg including the planter
surface of the rat foot. Although electrical stimulation of the sciatic nerve relies on an
invasive procedure in terms of nerve preparation, the main advantage of antidromic
stimulation is that the physiological responses observed are induced via the endogenous
release of peptides.
Antidromic vasodilatation and PE can occur across a range of frequencies (1-15Hz) but
this must be accompanied by high intensity voltages that are capable of activating Cfibres (White and Helme, 1985). The parameters used in this study were 20V, 5Hz,
2mSec for the duration of 1 min (Merhi et al., 1998). These parameters were shown to
cause sufficient sensory neurotransmitter release to cause vasodilatation in the rat hind
footpad but does not cause PE due to the short stimulation period (Lembeck and Holzer,
1979; White and Helme, 1985). It is known that stimulation parameters that activate
small unmyelinated sensory nerves will also activate the sympathetic efferents (Kawasaki
et al., 1990). Neurotransmitters of sympathetic origin, including NA and neuropeptide Y
(NPY), can modulate the action of sensory nerves at both pre- and post-junctional sites.
The role of sympathetic nerves in modulating sensory nerve responses is described in
section 1.8.3 (see also chapter 5).
When the sciatic nerve is stimulated antidromically, sensory neurotransmitters such as SP
and CGRP, which are synthesized in the DRG and transported (axonal transport) to the
nerve terminals, are released in the periphery to induce neurogenic inflammatory
responses. A reduction in content and transport of SP and CGRP was reported in sciatic
nerves of diabetic rats at 4 weeks (Di Giulio et al., 1995; Diemel et al., 1992) and 11
weeks (Willars et al., 1989). This coexisted with a decrease in SP and CGRP content of
lumbar DRG (Terenghi et al., 1994) and the footpad skin (Karanth et al., 1990). It was
shown that the metabolism and synthesis of SP and CGRP are dependent upon NGFs
uptake and retrograde transport and this is impaired in the sciatic nerve of diabetic rats
78
(3-4 weeks) (Diemel et al., 1992; Lindsay et al., 1989; Schmidt et al., 1999). Nemeth and
co-workers (1999a) used isolated diabetic rat trachea and showed the release of SP,
CGRP and somatostatin to be substantially diminished in response to electrical field
stimulation after 4 weeks and suggested the cause to be of neuronal origin. Impairment in
nerve axonal transport (both anterograde and retrograde) and motor nerve conduction
velocity (Di Giulio et al., 1995; Lockett and Tomlinson, 1992) have also been reported in
experimental diabetic rats. There is also evidence that abnormal sciatic nerve myoinositol metabolism (Finegold et al., 1983) in experimental diabetes impairs axoplasmic
flow (Mayer and Tomlinson, 1983) and reduces Na+/K+-ATPase activity (Greene and
Lattimer, 1983) by affecting phosphoinositide turnover (Lockett and Tomlinson, 1992).
Insulin was shown to be able to reverse alterations in nerve conduction velocity (sciatic
nerve, ascending and descending pathway of the spinal cord) up to 6 months after
induction of diabetes in STZ-treated rats (Biessels et al., 1999). It was suggested that the
reduction in glucose level might be responsible for this improvement although insulin’s
direct effect on the vasculature was not overlooked. Some aldose reductase inhibitors
were shown to have an effect on nerve conduction velocity and axonal transport in
diabetic rats (Cameron et al., 1997a; Hotta et al., 1996). For example, acetyl-l-carnitine
(ALCAR) prevented diabetes-induced SP loss in the sciatic nerve and spinal cord by
improving altered metabolic pathways and hence preventing the reduction in synthesis
and axonal transport of SP (Di Giulio et al., 1995) (section 1.5.1). All findings in this
section and the previous section indicate that reduction in neurogenic inflammatory
responses are probably due to a combination of reduction in synthesis and transport as
well as the release of sensory peptides in diabetes.
1.7.5. Measuring NI
A variety of different techniques have been used to monitor and quantify changes in
blood flow and have been reviewed by a number of authors (Oberg, 1990; Rowell, 1993).
Examples of these techniques are plethysmography, Doppler ultrasound flowmetery,
thermal and dye-dilution, and radioisotope infusion. Antidromic cutaneous vasodilatation
has also been measured using these various methods.
79
For example, Newby and co-workers (1999) used venous occlusion plethysmography to
measure changes in blood flow of the human forearm after SP infusion, whereas Arora
and collegues (1998) measured vasodilatation responses to iontophoresis of ACh and
SNP in forearm skin using Doppler ultrasound flowmeter. Others have collected venous
outflow from the hind limb (Lembeck and Holzer, 1979). While the above methods
provide quantitative information about blood flow, some are invasive in nature and both
demanding and time consuming to perform. Furthermore, they do not accurately
represent dilatation changes in the cutaneous microvasculature since they measure overall
blood flow in a larger area than skin innervated by the stimulated nerve.
Laser Doppler flowmetry is a non-invasive, simple-to-perform technique that produces
reproducible and reliable measurements of relative blood flow. This technique was
developed in early 1970’s but was first adopted for use in rat skin by Gamse & Saria in
1987, where they showed its suitability as a non-invasive technique for measuring
antidromic vasodilatation. This technique has also been used for early diagnosis of
diabetic microangiopathies (Balabolkin et al., 1994; Rendell et al., 1993). The heliumneon laser light (632.8 nm) was used to measure blood flow based on a Doppler
principle, where the resulting photocurrent of reflected laser light proportional to the
number and mean velocity of moving red blood cells is converted to a continuous
electrical input signal (in volts). The effective depth of laser penetration into the skin has
been previously reported to be approximately 1-2 mm (Halloway and Watkins, 1977;
Rendell et al., 1993) and Rendell et al (1998) have estimated the area of measurement to
be about 2 mm2. The principle and mechanism underlying laser Doppler flowmetry and 2
different types of laser Doppler flowmeters (LDFs) used in this study (Perimed and
Moor) are described in full detail in chapter 2, section 2.9.
PE, the other major feature of inflammation initiated by antidromic nerve stimulation, has
also been extensively studied. This is commonly assessed using Evans blue dye or by the
Bradford technique (Bradford, 1976). In experiments where inflammatory mediators
were perfused over the blister base, PE was collected and the level of protein in the
exudate was analyzed using the Bradford technique (see also chapter 2, section 2.12). In
experiments where the sciatic nerve was stimulated electrically, it was not possible to
80
measure PE due to the short stimulation period (1 min) and only changes in blood flow
was measured.
1.8. Modulators of sensory nerve function
1.8.1. Nitric oxide
NO was first described as an endothelium-derived relaxing factor by Furchgott and
Zawadzki in 1980. Because of their substantially similar actions, it was deduced that NO
and EDRF were the same agent (Ignarro et al., 1987). NO is synthesized as a mediator of
vasodilatation from its precursor L-Arg by NOS, which is present in an inducible form
and a constitutive form. Inducible NOS (iNOS) is cytosolic, requires protein for its
expression and stimulated in immunological reactions by cytokines and bacterial wall
proteins (Moncada and Higgs, 1993). This form of NOS is calcium-independent, and
produces nanomoles of NO over long periods but is dependent on NADPH (Moncada et
al., 1991). Many cell types including macrophages, vascular smooth muscle cells,
Kupffer cells, and hepatocytes are capable of expressing iNOS. Constitutive NOS is both
cytosolic and particulate, is NADPH-dependent, requires calcium / calmodulin, and
releases picomoles of NO for short periods in response to receptor stimulation by
calcium-mobilizing agonists (Moncada and Higgs, 1993; Moncada et al., 1991). This
type of NOS is localized in vascular endothelial cells, central and peripheral nervous
tissue, and platelets. NOS has been further divided into three isoforms: Neuronal NOS
(nNOS) describes the constitutive form localized in neuronal tissues; iNOS identifies the
inducible form; and eNOS represents the constitutive form found in vascular endothelial
cells. Each of the NOS isoforms found in brain, endothelium, macrophages and various
other cells from animals or humans, have been successfully purified and synthesized
(Nankane et al., 1993; Ohshima et al., 1992; Pollock et al., 1991; Xie et al., 1992).
The physiological actions of NO per se relate not only to its identity as a free radical
species (or NO•), but also to its other redox forms: the nitrosonium cation (or NO+), and
the nitroxyl anion (or NO-) (chapter 1, section 1.5.3, Fig 1-4) (Butler et al., 1995). Hence,
under physiological conditions, the term NO does not identify its redox forms but refers
to a variety of NO-containing molecules which may be present.
81
NO plays an important role in cell communication, defence, and injury. The biological
effect of NO depends upon both its concentration at the site of action as well as the
specific location where it is generated. Small quantities of this agent are generated by
constitutive NOS such as that present in the vascular endothelium, while large quantities
of NO are synthesized by iNOS in response to cytokines or bacterial products. As a
ubiquitous biological mediator, NO has been shown to have physiological roles in diverse
settings as well as being implicated in the pathogenesis of some diseases (e.g.
hypertension, dementia). There have been many reviews detailing various roles of NO
(Moncada et al., 1991; Singh and Evans, 1997) but the following discussion will be
limited to the issues relevant to this PhD study.
NO has a short half-life but is a potent endogenous vasodilator. Many of the target cell
responses of NO, such as the regulation of cardiovascular tone and the mediation of
neurotransmission, result from an increase in intracellular levels of cyclic guanosine
monophosphate (cGMP) via the stimulation of soluble guanylate cyclase (Moncada et al.,
1991). A continuous basal release of NO from vascular endothelial cells is suggested to
provide a constant vasodilator tone in the vasculature. This is supported by previous
studies in humans (Vallance et al., 1989) and rats (Rees et al., 1989) using NOS inhibitor.
These findings, as well as other studies in the literature, led to the conclusion that there is
a physiologic, NO-dependent vasodilator tone that is essential for regulating blood flow
and pressure (Moncada and Higgs, 1993).
As an intracellular signalling molecule, NO has important physiological roles in
nociception, neuronal degeneration, and neuroprotection (Lipton, 1999; Meller and
Gebhart, 1993). There are growing numbers of reports to propose neurotransmitter roles
for NO within the peripheral sensory nervous system. An immunological study has
demonstrated the existence of NOS in neurones and fibres of the DRG (Ruda et al., 1994)
suggesting that NO is released from sensory nerves. NOS immunoreactivity is also found
in peripheral nerves in the myenteric plexus of the intestinal tract, in the retina and the
adrenal gland (Bredt and Synder, 1990). Furthermore, NO was shown to be released from
mast cells and endothelium (Moncada et al., 1991). Taken together these observations
indicate that NO is well placed to mediate its effects in the periphery.
82
There is evidence showing that NO interacts with sensory neuropeptides such as CGRP.
A previous study using the NOS inhibitor, L-NAME, in rabbit skin microvasculature
showed that the release of CGRP from sensory nerves was NO-dependent but not its
mechanism of vasodilator action (Hughes and Brain, 1994). In a different study it was
suggested that NO could contribute to the activation and release of neurotransmitters
from sensory nerve fibres (Holzer and Jocic, 1994).
In the rat skin, CGRP participates in neurogenic inflammatory processes but the
cutaneous vasodilator action of exogenous CGRP and the CGRP-mediated vasodilation,
evoked by antidromic stimulation of afferent nerve fibres, do not depend on NO
formation (Holzer et al., 1995). Ralevic et al (1992), on the other hand, showed that
dilatation produced by ACh in the rat skin relies on a NO component (Ralevic et al.,
1992). These studies suggest that perhaps vasodilatation caused by interaction between
NO and CGRP may depend on the vascular tissue under study (Holzer et al., 1995).
Increasing evidence indicates that NO may play a role in inflammation. In one study,
treatment with inhibitors of NOS reduced the degree of acute inflammation in rats (Ialenti
et al., 1992). This active agent (NO) has also been shown to have a role in the
vasodilatation and inflammatory response to BK (Khalil and Helme, 1992), SP and VIP
(Khalil et al., 1994; Ralevic et al., 1995b). NO generation is also involved in the SPinduced extravasation response in rats (Hughes et al., 1990), since actions of NO were
inhibited by L-NAME.
The nature of endothelial dysfunction in diabetes is controversial. Increased NO
production is thought to occur in early diabetes due to NOS activation mediated by
hyperglycaemia (Pieper, 1998). It is also well established that the metabolic
dysregulation of diabetes is associated with inhibition of NO release and action (Pieper,
1998). Considerable evidence implicated differences in the duration of exposure to
diabetes in determining the nature and direction of these changes. Hyperglycaemia could
activate EDRF in the short-term (section 1.8.4.1, Fig 1-10). However, as the duration of
diabetes progresses atherosclerosis changes and other events such as generation of AGEs
and superoxide radicals promote pathological changes in endothelial function leading to
suppressed EDRF action (Bucala et al., 1991; Kaustic and Vanhoutte, 1989) (see also
83
sections 1.6.2 and 1.6.3). There is substantial evidence that vasodilatation mediated by
endothelium-derived NO is impaired in humans and animals with type I diabetes (Took,
1996; Utkan et al., 1998). Accordingly, recent evidence showed that vasodilatation
mediated by NO is accentuated in short-term (1-2 weeks) STZ-induced rats but becomes
paradoxically impaired with increasing duration of diabetes (8 weeks) (Pieper, 1999).
Moreover, Hopfner et al (1999) observed attenuated EDRF-mediated vasodilatation in 14
weeks, but not 2 weeks, STZ-treated diabetic rats (Hopfner et al., 1999). Alterations of
NO in diabetes could also appreciably affect ET-1 activity, which is discussed in section
1.8.4.
Functional alteration of the endothelial L-Arg-NO pathway leads to impaired
vasodilatation, reduced antithrombic properties, and structural changes (Cosentino and
Luscher, 1998). For example, it was suggested that in early diabetes, metabolic defects
may cause a decrease in synthesis of NO in either vascular endothelium or sympathetic
ganglia leading to decreased nerve blood flow (Stevens, 1995).
The relation between diabetes and premature cardiovascular disease is well established
(Bell, 1995). Atherosclerosis and microangiopathy are principle causes of morbidity and
mortality in patients with diabetes mellitus (Dahl-Jorgensen, 1998; McMillan, 1997).
Although the link between diabetes and cardiovascular disease is not understood, loss of
the modulatory role of the endothelium may be implicated in the pathogenesis of diabetic
vascular disease. This may be initiated by hyperglycaemia leading to decreased
production of NO, increased free radical formations and / or alterations in the level of
vasoconstrictors, which all can interfere directly and indirectly with NO activity (Chan et
al., 2000; Cosentino and Luscher, 1998) (see also section 1.5.4.2).
In chapter 4, the role of NO in modulating microvascular blood flow in early diabetes is
examined.
1.8.2. Capsaicin
An important tool used in the elucidation of the role of primary afferent nerves in various
processes is capsaicin (trans 8-methyl-N-vanillyl-6-noneamide), the irritant principal of
Hungarian red peppers. Capsaicin is a selective neurotoxin for certain types of sensory
nerves, especially unmyelinated C-fibres with polymodal nociceptors, involving the
84
activation of the specific receptor (vanilloid) (Holzer, 1991). Activation of the vanilloid
receptor leads to an influx of sodium and calcium ions, which at low doses (in the μg/kg
range) excites afferent neurons with small cell bodies, and unmyelinated (C-fibre) or
thinly myelinated (A-fibre) axons (Szallasi and Blumbeeg, 1996). This leads to release of
neurotransmitters, which results in the development of neurogenic inflammation and
nociception (Szolscany, 1984).
Acute topical application of capsaicin in small doses was shown to be capable of exciting
polymodal nociceptors (C and Aδ fibres) causing flare and PE responses without leaving
any permanent damage to these fibres (Fitzgerald, 1983). Long-term treatment with
capsaicin, depending on the treatment regime, can degenerate sensory nerves leading to a
decrease in their neurotransmitter content which in turn, attenuates neurogenic responses
and / or desensitizes sensory fibres to stimuli (e.g. electrical and chemical) (Holzer, 1991;
Szolscany, 1984; Szolscanyi et al., 1988). Systemic administration of high doses of
capsaicin (in the mg/kg range) can produce ultrastructural (e.g. mitochondrial swelling)
or even toxic effects (cell death) on primary afferents (Holzer, 1991). The extent of this
damage can vary depending on the dosage of capsaicin, the route of administration,
animal species, and age of the animals used. For example, pre-treatment of neonatal rats
with capsaicin at 50mg/kg, i.p. (chapter 2, section 2.8), causes selective degeneration of
up to 90% of unmyelinated primary afferent fibres (Holzer, 1992; Jancso et al., 1977) and
destruction of small DRG cells (Fitzgerald, 1983). Presence of capsaicin-activated ion
channels in DRG was shown before (Oh et al., 1996) and it is believed that their
activation may underlie neurotoxic effects of capsaicin.
The importance of sensory nerves and wound healing became evident from experiments
in animals pre-treated with capsaicin. Capsaicin-treated animals exhibit a number of
trophic changes and abnormal responses to injury, diminished skin integrity, reduced
inflammatory response, and poor wound healing (Gamse et al., 1980; Kjartansson et al.,
1987). Deficiencies in repair processes in capsaicin-treated animals can be reversed by
exogenous administration of sensory peptides. Reduced healing rate in capsaicin-treated
or old rats, following a burn, can be normalized by intradermal (i.d.) injection of SP and
CGRP into the wound (Khalil and Helme, 1996). CGRP can also increase the survival of
85
skin flaps in capsaicin-treated rats with skin injury (Kjartansson et al., 1987) (see also
section 1.9).
Loss of neurogenic inflammation in response to tissue injury may be an important
complication of diabetes mellitus and can influence wound healing. A previous study
found that there is loss of capsaicin-related neurogenic inflammation in the vasa
nervorum of experimental diabetes (4 months) (Zochodne and Ho, 1993). The authors
suggested that a similar deficit following nerve injury could impair the milieu for axonal
regeneration in diabetes. In chapter 3, sensory-independent effects of diabetes on the
microvasculature are examined using capsaicin pre-treated diabetic rats.
1.8.3. Sympathetic nerve
There is now a substantial body of evidence showing that NA, ATP and NPY act as
neurotransmitters and are released from sympathetic nerves in variable proportions
depending on the type of tissue, species used and parameters of stimulation (Burnstock,
1986). NA acts upon alpha (α-) and beta (β-) adrenoreceptors to mediate its biological
response. In most peripheral vascular beds in the skin the contractile response to NA is
mediated via post-junctional alpha 1-adrenoreceptors. In the vascular smooth muscle,
both alpha 1- and alpha 2-subtypes mediate vasoconstriction and beta-2 mediate
vasodilatation (Bennett and Gardiner, 1996) (Fig 1-7).
NA
NA
Fig 1-7: Schematic representation of post
NA
α2
α1
-synaptic receptors for sympathetic
neurotransmitter NA and their
mechanism of action. Please note
that in order to keep the diagram
simple NPY and ATP are not
included (Adopted from Bennett, T.
& Gardiner, S.M. (1996)).
β2
? Ca2+ influx
PI
cAMP
Free cytosolic
Ca2+
Contraction
cAMP
Vascular Smooth
Muscle Cell
Free Ca2+
Relaxation
86
In the rat DRG, three alpha 2-adrenergic receptor subtypes have been identified, namely
alpha 2A, 2B, and 2C (Gold et al., 1997). Existence of alpha 2-adrenoreceptors was
previously reported in the membrane of unmyelinated nerve axons (Fuder and Selbach,
1993; Nicholas et al., 1993). Furthermore, neuropeptide release from pulmonary C-fibre
afferents evoked by only low frequency electrical stimulation was also inhibited by
stimulation of pre-junctional alpha 2-adrenoreceptors (De Luca et al., 1990).
Functional studies have suggested that NA has two distinct actions on sensory nerves
mediated by different adrenergic subtypes. NA can prevent pre-junctional neuropeptide
release by acting on alpha 2B-adrenoreceptor and inhibiting voltage-gated calcium
currents (Bean, 1989; Fuder, 1994; Kawasaki et al., 1990). NA can also directly inhibit
primary afferent activity by acting on alpha 2C-adrenoreceptors (Khasar et al., 1995). It
was also reported that C-fibre responses to nerve stimulation were reduced and delayed
following sympathetic stimulation (Shyu et al., 1989).
In STZ diabetic rats, the spinal release of NA was markedly suppressed (4 weeks) (Bitar
et al., 1999) and circulating NA was reduced (2-5 weeks) (Ohtani et al., 1997). A
diminished tissue NA concentration was also reported in skin (Ahlgren and Levine,
1993), certain parts of the brain such as pons and medulla (Ohtani et al., 1997;
Ramakrishnan and Nmasivayam, 1995), and in the duodenum (Patel et al., 1997) but
significantly enhanced in kidney and liver of 4 weeks diabetic rats (Patel et al., 1997).
Impairment
in
neurotransmitter
release
from
both
bladder
sympathetic
and
parasympathetic efferent nerve endings have also been previously reported in early (2
weeks) experimental diabetes (Tong et al., 1996). The effect of diabetes on the vascular
response to NA is strongly related to the type of tissue used and the duration of diabetes
and this was discussed in section 1.5.4.1).
A reduction in the expression of messenger ribonucleic acid (mRNA) level encoding for
alpha 2-adrenoreceptor subtypes was previously reported in the spinal cord of diabetic
rats (4 weeks) (Bitar et al., 1999). The number of beta-adrenoreceptors, in the diabetic
heart, was also decreased by 29% after 1 week and 50% after 3 weeks (Nishio et al.,
1988; Yu et al., 1994). The decrease in beta-adrenoreceptor population size in 4 weeks
diabetic rat’s ventricular tissue has been related to an increase in circulating NA level
(Ramanadham and Tenner, 1987). The densities of beta 1- and beta 2-adrenoreceptors in
87
the brain of STZ-treated rats were increased in hypothalamus, thalamus, and amygdala
(Bitar and DeSouza, 1990). It was previously suggested that diabetes may impair
reactivity of large peripheral arteries and arterioles to activation of beta-adrenoreceptors
(Mayhan, 1994). The relaxation of vascular smooth muscle induced by stimulation of
beta-adrenoreceptors was also demonstrated to be decreased in humans (Harada et al.,
1999) and diabetic rats (Kamata et al., 1992).
Angiotensin II is a potent vasoconstrictor and plays a vital role in regulating blood
pressure. An in vitro study has shown that high extracellular glucose concentrations (1520mM) could directly down regulate angiotensin II and its function in vascular smooth
muscle cells (Williams et al., 1992). However, no difference in plasma angiotensin II
levels was observed between 4 weeks diabetic rats and controls (Kihara et al., 1999). It
has been hypothesized that activation of pre-junctional angiotensin II-receptors by
angiotensin II may facilitate NA release mediated by beta 2-adrenoreceptors under
physiological conditions (Schlicker et al., 1988; Ziogas and Story, 1991). On the other
hand, reduced angiotensin II formation and enhanced prostaglandins formation might
reduce NA release (Schwieler et al., 1992). An increase in angiotensin converting
enzyme (ACE) activity was shown in type II diabetic patients (Ustundag et al., 2000). An
increase in serum ACE level was also observed in rats, 12 days after STZ injection
(Ustundag et al., 1999). Previous studies monitored diabetes-associated alterations in
heart rate and blood pressure in STZ-treated rats over a 10 week period and showed a
time-dependent reduction in heart rate (Fazan et al., 1999; Hicks et al., 1998). Although
diastolic blood pressure and mean arterial pressure did not differ significantly, when
compared to non-diabetics during this period, the pulse pressure was diminished. Reninangiotensin system (RAS) may also play a role in the NO / ROS balance, as angiotensin
II has been directly implicated in generation of superoxide anions in smooth muscle cells.
Angiotensin II has been shown to regulate a membrane-bound flavin-containing NADH /
NADPH oxidase that produces oxygen radicals (Fukui et al., 1997).
NPY and ATP are also released from sympathetic nerves upon stimulation but in variable
proportions depending on tissue type, species, and parameter of stimulation (Burnstock,
88
1986). NPY is a transmitter often co-localized with NA and co-released from sympathetic
nerve endings (Bennett and Gardiner, 1996). Plasma levels of NPY are correlated with
those of NA and its resting plasma levels are low in most species (Pernow, 1988). It is
preferentially released at high frequency nerve stimulation (10 Hz or more) and has a
long duration of vasoconstrictor action (Kennedy et al., 1997; Pinter et al., 1997). There
are two NPY receptor subtypes, Y1 and Y2, and both are expressed in sensory ganglia
fibres (Zhang et al., 1995). NPY acts as a neuromodulator controlling the release and
action of NA and ATP (Bennett and Gardiner, 1996) and hence capable of regulating skin
microcirculation (Pinter et al., 1997).
Pre-junctionally, this transmitter reduces the release of NA and ATP and postjunctionally it enhances their actions (Bennett and Gardiner, 1996). It was suggested that
NPY initiates an intracellular calcium-sensitive mechanism, which increases alphaadrenoreceptor sensitivity (Fallgren et al., 1993). This results in a significant increase of
sarcoplasmic calcium and stronger contractile responses to NA. Although both NA and
NPY have a vasoconstrictor effect, the action of NPY is not alpha-adrenoreceptormediated (Wahlestedt et al., 1986). It has also been shown that NPY release is enhanced
after alpha-adrenoreceptor blockade or beta-adrenoreceptor stimulation, indicating that
NPY release is also regulated pre-junctionally by NA (Dahlof et al., 1986).
Diabetes is associated with altered both central and peripheral nervous content of NPY.
An increase in NPY level in various brain regions of STZ-treated rats (7 weeks) has been
observed (Dunbar et al., 1992). This caused an increase in mean arterial pressure (MA) in
diabetic rats but did not affect the heart rate. It was demonstrated that vascular
conductances (e.g. regional blood flow) in response to NPY were attenuated in diabetic
vessels (4 weeks) especially the iliac and renal arteries (Hu and Dunbar, 1997).
Furthermore, vascular smooth muscle contractile responses (post-junctional event) to
NPY were markedly reduced in human and diabetic animals (Andersson et al., 1992;
Lind et al., 1995) and this was shown to be endothelium-independent.
ATP is present in both small and large dense-cored vesicles together with NA. It is
synthesized in the mitochondria and incorporated into vesicles by carrier-mediated
transport. ATP is released from sympathetic nerves and activates purinergic receptors (P1
and P2). The main role of ATP appears to be in the initiation of vascular contraction,
89
although its rapid degradation complicates the interpretation of its effect. Evidence has
been presented that ATP acts as an excitatory co-transmitter (at least in some species)
with NA causing vasoconstriction via excitatory P2X purinoceptors located on the
vascular smooth muscle (Bennett and Gardiner, 1996). The contractile effect of ATP is
dependent on an influx of extracellular calcium. ATP can also act as a vasodilator
through P2Y receptors on endothelial cells or when it is broken down to adenosine by
acting on P1 purinoceptors on the vascular smooth muscle (Pinter et al., 1997). ATP
contribution in response to sympathetic nerve stimulation varies considerably in different
vessels. In diabetes, there was no significant change in ATP concentration or its release
between control and diabetic rats (2 and 8 weeks respectively) (Belai et al., 1991;
Carlsson and Arnqvist, 1981).
ETs have also been reported to both stimulate and inhibit neurotransmitter release from
post-ganglionic sympathetic neurons. For example, it was demonstrated that ET-1 (1-30
nmol/l) significantly inhibited but ET-3 (300 nM) potentiated ATP and NA release from
rat isolated tail artery (Mutafova-Yambolieva and Westfall, 1998). Threshold and
subthreshold concentrations of ET-1 can also enhance the contractile response to other
vasoconstrictor agents (e.g. NA, 5-HT, vasopressin, histamine, and ATP). In certain
tissues, this response is post-junctional and partially due to facilitation of calcium through
L-type calcium channels (Reid and La, 1995). There was no change in positive
chronotropic and inotropic effects of ET-1 to sympathetic nerve stimulation and NA in
early diabetes (Lieu and Ried, 1994).
Sympathetic nerve and skin microvasculature
Under short-term physiological conditions there is a dynamic balance between
sympathetic nerve-mediated vasoconstrictor tone and endothelial-mediated vasodilator
tone. This balance may be altered by diabetes leading to neurovascular complications.
One study demonstrated changes in skeletal muscle haemodynamics after only one week
of STZ-induced diabetes (Hill et al., 1985). This was characterized by decreased
perfusion and diminished ability to lower resistance by vasodilation. Alterations in the
level of NA (circulating and tissue content), adrenoreceptors function and their densities
90
in diabetic condition were also previously reported. The overall information presented
raised the possibility that sympathetic nerves may modulate skin microvascular blood
flow in early diabetes, which is examined in chapter 5.
1.8.4. Endothelin
ET is a potent vasoconstrictor polypeptide that was first isolated from porcine aortic
endothelial cells and sequenced in 1988 by Yanagisawa and his colleagues. Four
isoforms of ET have been identified: ET-1, ET-2, ET-3, and vasoactive intestinal
contractor ET (VIC). These isoforms are not species specific and share the total number
of 21 amino acids, ET-2 > ET-1 > VIC >> ET-3 (Nayler, 1990).
ET-1 is produced by many cells other than endothelial cells, including vascular smooth
muscle, kidney, tracheal cells, renal glomeruli, motor and dorsal horn neurones of the
spinal cord, retinal endothelial cells, and even certain tumour cells (Nayler, 1990). In
healthy individuals, plasma levels of ET-1 range from 1.5 to 3.7 pg/ml (Battistini et al.,
1993). This can increase by approximately 2 to 10 fold in a number of cardiovascular
disorders including myocardial infarction, atherosclerosis, chronic heart failure,
cardiogenic shock, diabetes, renal failure, subarachnoid haemorrhage, and hypertension
(Battistini et al., 1993).
ET synthesis and release is triggered by platelets, thrombin, other vasoconstrictors
(angiotensin II, vasopressin, adrenaline), growth factors (platelet-derived growth factor
(PDGF), TGF-beta, EGF), cytokines (IL-1), TNF-alpha, insulin, cellular stress (e.g.
hypoxia), etc with no apparent feed back response (Ruffolo, 1995). The mechanism of
agonist-induced ET-1 production and cellular release possibly involves the receptormediated mobilization of intracellular calcium and activation of PKC (Yanagisawa et al.,
1989). A negative pathway involving the cGMP mechanism also regulates the production
of ET-1. Agents such as NO and nitrosovasodilators reduce the secretion of ET-1
(Ruffolo, 1995). ET-1 has a short half-life (1-7 min) before it is eliminated from the
circulation but its biological actions are maintained for a longer period (about 30 min).
91
ET isopeptides mediate their biological responses (in vivo and in vitro) through binding
to receptors termed, ET-A and ET-B receptors (a third ET-C has also been identified)
(Ruffolo, 1995). There is evidence that ET-A and ET-B receptors exist in the rat skin
(Lawrence et al., 1995). ET-A receptors are characterised by selectivity for ET-1 and ET2 and are found in the highest density in vascular smooth muscle to mediate
vasoconstriction and smooth muscle cell proliferation (Ruffolo, 1995). ET-B receptors
(and subtypes) are “non-selective” and have equivalent affinities for ET-1, -2, and -3.
Pharmacological data indicate the existence of two anatomically separate and distinct ETB receptors; an ET-B receptor on the endothelium that mediates the release of NO and
prostacyclin (Fozard and Part, 1992) to stimulate vasodilatation and an ET-B receptor on
the vascular smooth muscle that directly stimulates smooth muscle contraction. The
endothelial ET-B receptor has been termed ET-B1 and the other receptor that stimulates
vascular smooth muscle contraction has been termed ET-B2 (Gellai et al., 1996). The ETC receptor, to which ET-3 shows a higher affinity than ET-1, has been identified in
bovine endothelial cells (Emori et al., 1990). Species differences exist in ET receptors
both at structural and functional levels (Lodge et al., 1995; Ruffolo, 1995).
Binding of ET to its receptor(s) is slow, pH sensitive, and calcium-independent (Ruffolo,
1995). Various pathological conditions, including hypertension, ischaemia, and STZinduced diabetes can modify the density of ET receptor binding (Nayler, 1990). Various
receptor agonists and antagonists have been developed over time to aid scientists with
understanding ETs mechanisms of action under physiological as well as pathological
conditions (Fig 1-8).
92
ET RECEPTOR SUBTYPES IN THE BLOOD VESSEL
ET-3, ET-1, IRL 1620
SB 209670, BQ-788, Bosentan,
PD 142893, RES 701
ET-B1
ENDOTHELIUM
ET-3, ET-1,
IRL 1620
ET-1
SB 209670
NO
ET-B2
BQ-123, Bosentan,
BMS-182874,
PD 142893
Fig 1-8: Schematic representation
illustrating locations and
selective ligands for
endothlin receptor
subtypes (Adopted from
Ruffolo, R.R.. Jr (1995)).
ET-A
VASCULAR SMOOTH MUSCLE
VASODILATATION
VASOCONSTRICTION
ET-1 has a wide range of biological actions in a variety of tissues, apart from its potent
vasoconstrictor activity. A number of ET-1’s biological effects are shown in Table 1-5.
93
Biological effects of ET-1
Vascular effects
- Protracted constriction of vascular smooth muscle
- Mitogenic action in cell culture
- Proliferative actions in vivo
- Synergy with other vasoconstrictors & growth factors
Cardiac effects
- Positive chronotropic & inotropic effect
- Coronary vasoconstriction
- Arrhythmogenic
- Myocardial hypertrophy
Endothelium
- Release of NO & prostanoids
- Induction of adhesion molecule expression
Non-vascular smooth muscle
- Constriction of intestinal, tracheal, mesangial, bladder,
uterine, and prostatic smooth muscle
Renal function
- Decrease in renal blood flow & GFR
- Increase in sodium excretion
- Increase in plasma renin activity
Respiratory
- Bronchoconstriction
Endocrine
- Stimulation of neuropeptide & pituitary hormone release
- Increase in release of atrial natriuretic peptide
- Stimulation of aldosterone biosynthesis
Nervous tissue
- Modulation of neurotransmitter release
- Potent cerebrovascular spasm
Other effects
- Increase in intraocular pressure
- Stimulate matrix production
- Increase in bone reabsorption
- Stimulate neutrophil superoxide production
Table 1-5: A number of ET-1’s biological effects (Adopted from Ruffolo, R.R. Jr.(1995)).
94
It has been shown that binding of ET-1 to its receptor activates a variety of signal
transduction pathways to produce responses. These signal transduction processes include
adenylate cyclase, guanylate cyclase, Phospholipase (PL) C, A2, and D, PKC, and MAP
kinase (MAPK) (Ruffolo, 1995). As a result, a number of second messengers are
generated such as cAMP, cGMP, IP, intracellular calcium, DAG, and AA. These second
messengers activate other enzymes to phosphorylate or dephosphorylate key cellular
proteins, leading ultimately to biological responses (Fig 1-9).
Ca2+0
ET-1
Ca2+0
Fig 1-9: Intracellular signal transduction
R
Cation channels
or ROCC
Depolarization
G
Pathways activated by ET-1 in vascular
Smooth muscle cells (intracellular
calcium store (Ca2+s), receptor-operated
calcium channels (ROCC), voltageoperated calcium channels (VGCC), free
cytoplasmic calcium (Ca2+i), phosphatidic
acids (PA),prostaglandins E2 (PG E2),
thromboxane A2 (TXA2)) (Adopted from
Reid, J.J. & La, M. (1995)).
L-type
VOCC
PLC PLD PLDA2
PA
Ca2+i
Ca2+i
IP3 + DAG
PKC
Ca2+i
AA
Ca2+i
PGE2
TXA2
+
CONTRACTION
The physiological importance of ET-1 is not very well known. It is possible that, as a
locally acting hormone, ET-1 exerts its physiological effects on the vascular tone by
interacting with other vasoconstrictor agonists within the local vicinity of smooth muscle
cells to produce or enhance vasoconstriction. Non-vascular smooth muscle, such as
bronchial muscle, ileum, uterus, and vas deferens can also respond to ET but they are less
sensitive than vascular smooth muscle and a higher concentration of ET will be required
for their contraction (Ruffolo, 1995). The vasoconstrictor activity of ET is not mediated
by alpha-adrenergic, histaminergic, cholinergic, or serotonergic mechanisms, and neither
vasopressin nor angiotensin seems to be involved (Ruffolo, 1995).
1.8.4.1.
Endothelin and diabetes
ET-1 synthesis and release
In patients with type I diabetes, plasma ET-1 has been reported to be both increased
(Haak et al., 1992; Takahashi et al., 1990) and decreased (Smulders et al., 1994) but in
patients with type II diabetes, this was either unchanged (Bertello et al., 1994; Nugent et
al., 1996) or again increased (Donatelli et al., 1994; Morise et al., 1995). The status of
95
ET-1 plasma concentrations in type I diabetes in animals is also controversial. In STZinduced diabetic rats, most studies agree that ET-1 concentration is increased in
mesenteric artery and aorta after 8 weeks of diabetes (Hopfner et al., 1999; Makino and
Kamata, 1998; Takahashi et al., 1991). However, this did not seem to be the case in
earlier stages of the disease. Some showed reduction in ET-1 concentration in kidney,
aorta, and cultured vascular smooth muscle before 5 weeks (Frank et al., 1993; Hopfner
et al., 1998b; Shin et al., 1995), whereas others demonstrated no change (mesenteric
artery, aorta) at 4 days, 2 or 4 weeks after induction of diabetes (Takahashi et al., 1991;
Wu and Tang, 1998). One study showed an increase in ET-1 after 2 weeks of diabetes in
the above tissues (Wu and Tang, 1998). It appears that the duration of diabetes can
determine the direction of change in plasma ET-1 in STZ-treated animals (Hopfner et al.,
1999). In addition, inter-study variability in the metabolic state, level of control, presence
of established macrovascular complications, and methodology for measuring ET-1 are all
likely to contribute to observed discrepancies. Importantly it was demonstrated that
altered plasma ET-1 in diabetes could be returned to normal level by restoration of
metabolic control (Hopfner et al., 1998b).
Key metabolic variables that are changed in DM such as plasma glucose, insulin, and
lipids are well known to directly regulate the release of ET-1 from endothelial cells and
modulate receptor expression and responses to the peptide (Hopfner and Gopalakrishnan,
1999) (see also Figs 1-10 & section 1.8.5.3, Fig 1-15). Prolonged and excessive exposure
of both endothelial cells and vascular smooth muscle cells to the metabolic dysregulation
of diabetes could also promote morphological changes that indirectly affect the release
and action of ET-1 at various target sites (Hopfner and Gopalakrishnan, 1999).
Hyperglycaemia, in vitro, can increase (Yamauchi et al., 1990), decrease (Hattori et al.,
1991), or have no effect (Baumgartner-Parzer et al., 1996) on the release of ET-1 from
endothelial cells. Species type as well as glucose concentration and incubation period
used by these studies may have accounted for observed discrepancies. High glucose
levels increase EDRF production through activation of NOS (Hopfner and
Gopalakrishnan, 1999) and EDRF inhibition, through endothelial damage and formation
96
of AGEs and superoxide radicals (Fig 1-10). Formation of the above substances could
also regulate ET-1 production and action (Hopfner and Gopalakrishnan, 1999).
Studies examining the effect of high glucose levels on receptors and responses to ET-1
have yielded somewhat similar results and are summarized in Fig 1-10.
Glucose
+
Fig 1-10: Effects of high glucose on factors
+
Endothelial
damage
Free radicals
AGE products
+/-
related to ET-1 production and
action (Adopted from Hopfner,
R.L. et al (1999)).
+/-?
_
EDRF
_
ET-B
EC
ET-1
ET-A
[Ca2+]i
VSMC
Hopfner et al. (1999) and a number of other investigators have gathered evidence to show
that insulin can modulate ET-1 production and actions in both in vitro and in vivo settings
and the summary of their findings is shown in Fig 1-14 (Hopfner and Gopalakrishnan,
1999) (section 1.8.5.3).
Alterations in NO in diabetes could also appreciably affect ET-1 activity. ET is the main
vasoconstrictor / mitogenic mediator released by the vascular endothelium whereas
EDRF modulates vasodilation anti-growth effects. At the gene level, EDRF inhibits ET1. Conversely, ET-1 might inhibit EDRF in pathological conditions. It is well known that
EDRF suppresses ET-1 release and action (Kourembanas et al., 1993) and endothelial
damage is thought to be associated with increased release of ET-1 (Huszka et al., 1997;
Takahashi et al., 1990). It was shown that changes in plasma concentration of ET-1 and
vascular responses to ET-1 in STZ-treated rats are duration-dependent and occur in
parallel with changes in endothelial function (Hopfner et al., 1999).
Raised plasma lipoproteins are present in both type I and type II diabetic patients,
particularly in those with poor metabolic control. Considerable evidence indicates a
97
profound ability of plasma lipoproteins to affect both the release and action of ET-1.
Oxidized LDL increases ET-1 gene expression and raised lipoproteins in vivo accentuate
vascular responses to agonists (including ET-1) before the onset of atherosclerosis
(Ruffolo, 1995). After the onset of atherosclerosis, ET receptors are decreased and ET-1
peptide increased in animal models of the disease (Hopfner and Gopalakrishnan, 1999).
Moreover, endothelial damage resulting from the cascade of atherosclerotic events
resulting from dyslipidaemia would be expected to modulate the release and action of
ET-1. Thus, it is reasonable to suggest that hyperlipidaemia could regulate plasma
concentrations of ET-1 as well as its receptors, and tissue reactivity to the peptide
(Hopfner and Gopalakrishnan, 1999).
ET-receptors density and activity
Up-regulation of ET receptors was reported in the heart (2-6 weeks), retina (4 weeks),
prostate, vas deferens (8 weeks), and urethra (16 weeks) after induction of diabetes (Deng
et al., 1999; Nakamura et al., 1997; Saito et al., 1996; Vesci et al., 1995). However, in
diabetic kidney (1-6 weeks), no change in ET-A receptor density was observed (Awazu et
al., 1991; Vesci et al., 1995) but ET-B receptors were down- regulated (4 weeks) (Awazu
et al., 1991; Masao et al., 1999). Diabetes-induced PKC activation was suggested to be
responsible for up-regulation of ET-1 in the retina (Koya and King, 1998). In the kidney,
an increase in ET-1 could down-regulate ET-B receptors by activating PKC and
intracellular calcium mobilization (Masao et al., 1999). These suggest that in diabetes
there may be a relationship between a high level of ET and PKC activity. Experiments on
patients with type II diabetes showed no reduction in forearm blood flow when ET-1 was
infused locally (Nugent et al., 1996). They presented several possible explanations for
their results including down-regulation of ET-A and ET-B receptors on smooth muscle,
up-regulation of endothelial ET-B receptors, or perhaps an increase in concentration of
big ET, which is a weak agonist for ET receptors.
Vascular response to ET
In diabetic states, few studies have assessed haemodynamic actions of ET-1 in vivo,
therefore no definite conclusions can be drawn from them. However, based on existing
98
evidence, this appears to be dependent on the pre-existing duration of diabetes, the status
of endothelial function, and vascular bed under consideration (Hopfner et al., 1999) (see
also section 1.6.4.2). The in vivo response to ET-1 in different vascular beds itself seems
to depend on the expression and density of vasodilator ET-B and vasoconstrictor ET-A
and ET-B receptors.
Most in vitro studies examining vascular responsiveness to vasoactive stimuli (including
ET-1) in models of diabetes use aorta as the prototype vessel. In early diabetes (2 weeks),
some suggested that significant attenuation of the aortic response to ET-1 (Makino and
Kamata, 1998; Utkan et al., 1998) could be attributed to down-regulation of ET receptors
by either PKC activation or increased ET-1 production (Awazu et al., 1991; Guillon et
al., 1998). Others showed an exaggerated response to ET-1 in the same tissue (Hopfner et
al., 1999; Hopfner et al., 1998b), and suggested that expression and action of calcium
channels are altered or PKC is activated (Abebe and MacLeod, 1990; Reid and La, 1995)
(see also section 1.6.4.1).
Endothelins and skin microvasculature
ET production is regulated at the level of mRNA transcription / translation and is made
upon demand, for example in an emergency and / or in a defensive event such as
inflammation and wound repair (Nakamura et al., 1990; Yanagisawa and Masaki, 1989).
ET is a potent vasoconstrictor. In vascular smooth muscle, ET causes a biphasic increase
in calcium concentration. A transient increase is believed to be related to the mobilization
of intracellular calcium while a sustained increase is due to the influx of extracellular
calcium (Highsmith et al., 1992). The influx of calcium induced by ET-1 may depend on
the opening of L-type calcium channels, but the involvement of non-selective cation
channels was also suggested (Highsmith et al., 1992; Reid and La, 1995). A previous
study suggested that activation of ET receptors (ET-A or ET-B) mediates
vasoconstriction in the rat cutaneous microvasculature but has no effect on the increased
microvascular permeability (Lawrence et al., 1995). Arteries, arterioles, veins, and
venules are all affected by ET, with veins being more sensitive than arteries but venules
and arterioles are equisensitive.
99
Presence of ETs and their binding sites in both peripheral neurons and the central nervous
system suggests a neuromodulatory role for these peptides (Ruffolo, 1995). For example,
it was suggested that the vasoconstrictor action of endogenous ET may contribute to
peripheral nerve dysfunction in experimental diabetes (5-8 weeks) (Stevens and
Tomlinson, 1995; Zochodne et al., 1996). It also appears that ET does not contribute to
SP-induced vasodilation in the rat skin (Khalil and Helme, 1996) and, in fact, inhibits
vasodilatation induced by NO and nitroprusside (Chander et al., 1988). ET can also
inhibit plasma extravasation induced in the rat skin by the i.d. injection of inflammatory
mediators 5-HT, histamine, BK, and platelet activating factor (PAF) (Chander et al.,
1988). The role of ET-1 in modulation of SP and CGRP release from capsaicin-sensitive
neurones has been reported (Dymshitz and Vasko, 1994). However, in this study, it was
not clear whether ET-1 acted directly on these nerves, or released some chemical
mediator from other types of neurons, or from satellite cells present in DRG cultures.
There has been much speculation regarding the pathophysiological significance of ET-1.
For example, convincing evidence exists that ET-1 might be directly linked to the
development of hypertension (Yokokawa et al., 1991). Elevated ET-1 plasma levels have
also been reported in many cardiovascular disorders and diabetes (Battistini et al., 1993;
Haak et al., 1992; Takahashi et al., 1990). However, conclusions drawn about the direct
association of increased ET-1 levels with these disease states needs to be interpreted with
caution.
In chapter 6, the role of ETs in modulating skin microvascular blood flow in early
diabetes is investigated. Specifically, the role of ET receptors (ET-A and ET-B) and their
quantitative contribution to microvascular responses in 4 weeks diabetic rats is examined.
1.8.5. Other modulators
Several endogenous and local tissue factors appear to modulate release and action of
neuropeptides from sensory nerves. However, only those factors that are relevant to this
thesis are discussed. Factors which activate the efferent function of sensory nerves
include serotonin (Khalil and Helme, 1990b), ACh, cholecystokinin (CCK) (Bennett and
Gardiner, 1996; Maggi, 1991), and BK (Khalil and Helme, 1992). Other mediators inhibit
100
the efferent function at the pre-junctional level including galanin and somatostatin
(Maggi, 1991). Furthermore, there are reports of drugs and / or transmitters that could
play a dual effect via different receptors. For example, opioid peptides can inhibit plasma
extravasation and vasodilatation responses (Khalil and Helme, 1990a), but they may also
release SP from the central nervous system (Chahl, 1990). Others in this group include
capsaicin, histamine, γ-aminobutyric acid, and adenosine (Maggi, 1991). The role of the
sympathetic nervous system in modulating sensory nerve function is also important and is
discussed in section 1.8.3.
1.8.5.1.
Acetylcholine
ACh, a neurotransmitter, is synthesized in nerve cells from choline and acetyl coenzyme
A in the presence of choline acetyl transferase. It is stored in synaptic vesicles and
released at synapses of parasympathetic nerves and at neuromuscular junctions. ACh acts
upon muscarinic and nicotinic receptors. The muscarinic receptors are on endothelial
cells, smooth muscle cells, and glands that are coupled via a G-protein to adenylate
cyclase or PLC. Nicotinic receptors are in motor end plates of skeletal muscle,
sympathetic ganglia and many parts of the central and peripheral nervous systems
(Bennett and Gardiner, 1996; Burnstock, 1988). Nicotinic receptors are ion channels that
open up when ACh is present, leading to an influx of sodium ions and hence a
depolarizing potential. This depolarization leads to the release of other transmitters and
peptides in a cascade fashion. Stimulation of these receptors produces sensation of
prickle, itch, burning pain, and flare which spreads via the axon reflex (Grunfeld et al.,
1991).
ACh has a role in the control of local blood flow. It can produce smooth muscle
relaxation and vasodilatation mediated by the release of EDRFs, mainly NO (Furchgott
and Zawadzki, 1980), from the vascular endothelium in response to stimulation of
muscarinic receptors. If ACh is applied to blood vessels with damaged endothelium,
however, it may produce vasoconstriction by acting on the muscarinic receptors on the
underlying smooth muscle (M1 receptors) (Burnstock, 1988).
Prostacyclin or pre-junctional inhibition of adrenergic neurotransmission, in vivo, may
also be involved in the vasodilatation response to ACh. At the level of skin
101
microvasculature, ACh may be released from cholinergic post-ganglionic sympathetic
fibres that innervate sweat glands and from parasympathetic post-ganglionic nerve
endings at the neuromuscular junction. In addition to ACh, the cholinergic nerve also
contains VIP, which produces vasodilation by directly acting on vascular smooth muscle
(Bennett and Gardiner, 1996). Presence of NOS in the neuronal tissue and the coexistence of NO and VIP in parasympathetic post-ganglionic nerve terminals suggest that
NO may be synthesized and released from cholinergic vascular nerve terminals (Kummer
et al., 1992). This is supported by a previous study where inhibition of NO production
attenuated parasympathetic, non-cholinergic vasodilatation (Garthwaite, 1991).
It has been suggested that ACh released from post-ganglionic sympathetic nerve fibres
contributes to the inflammatory response since antidromic vasodilation and neurogenic
plasma extravasation were reduced by atropine, a muscarinic receptor antagonist
(Furchgott and Zawadzki, 1980). A previous study investigating the mechanism
underlying vasodilator action to ACh in the rat skin microvasculature, with intact
endothelium, indicated that this response is comprised of two distinct components. One
part of the response is mediated through NO, probably arising after direct stimulation of
muscarinic receptors (M3 receptors) on endothelial cells, and the other part of the
response is mediated through nicotinic receptors located on sensory nerves, leading to the
release of sensory neurotransmitters (Ralevic et al., 1992).
ACh BK Ca2+
Endothelium
Fig 1-11: Receptor activation of vascular
endothelium by ACh or BK
(Intracellular calcium ([Ca2+]),
Constitutive nitric oxide synthase (cNOS),
L-arginine (L-Arg), Soluble guanylate
cyclase (sGC), Guanosine triphosphate
(GTP))(Adopted from Moncada, S. &
Higgs, A. (1993)).
L-Arg
+
+ [Ca2+] + NOS
NO
+
sGC
Smooth muscle
GTP
Cyclic GMP
Relaxation
102
Superoxide anions depress, and hydroxyl radicals facilitate, endothelium-dependent
relaxations caused by activation of muscarinic receptors (Rubanyi and Vanhoutte, 1986).
NO is increased by ACh, which promotes the reaction between NO and superoxide
anions (Prasad and Bharadwaj, 1996).
In rat arteries, muscarinic agonists such as ACh release EDHF (Edwards et al., 1998).
One in vitro study, using hepatic artery, proposed that the stimulation of endothelial cells
by agonists such as ACh or BK might activate apamin-and charbdotoxin-sensitive K+
channels. Any resulting K+ efflux might then hyperpolarize the adjacent smooth muscle,
suggesting EDHF to be endothelium-derived K+ (Edwards et al., 1998). Relaxation
responses, induced by potassium channel activators, were significantly reduced in STZinduced diabetic rat aorta (Kamata et al., 1989), providing indirect evidence for the
reduced potassium channels activity in these animals (section 1.5.4.2).
An in vitro study showed that basal release of ACh in the brain of STZ-treated rats,
monitored for 7 weeks, started to decline after 1 week of diabetes (Welsh and Wecker,
1991). The concentration of ACh in the heart of 2 and 4 weeks diabetic rats was not
significantly different from control rats (Akiyama et al., 1989). It was shown that diabetes
causes an alteration in muscarinic receptor biosynthesis and function. Increase in the
density of these receptors was observed in vas deferens (about 8 weeks) (Kanda et al.,
1997) and the bladder (2 weeks) (Kanda et al., 1997; Tong et al., 1999) of diabetic rats.
Impaired receptor activity was also reported in isolated gastric antral smooth muscles
(Soulie et al., 1992). It was suggested that early alteration in the muscarinic ACh system
in diabetes is associated with increased production of NO and cGMP (Wald et al., 1998).
Another study hypothesized that muscarinic receptor up-regulation could be related to the
polyol-pathway and activation of aldose reductase (Kanda et al., 1997). Alteration to the
receptor property and function in experimental diabetes was suggested to be due to
changes of GTP-binding proteins coupled to muscarinic receptors (Soulie et al., 1992).
Alteration in the vasodilator response to ACh has been described earlier by other studies
(section 1.5.4.2) and inconsistency was found among results. Duration of diabetes and the
type of vascular bed under study were nominated as possible contributing factors to these
discrepancies (Chan et al., 2000; Pieper, 1999). In chapter 4, the role of parasympathetic
103
nerves in modulation of skin microvascular blood flow in early diabetes is investigated.
In this chapter, changes in microvascular responses to ACh and the function of its
receptors (nicotinic and muscarinic) are examined in 4 weeks diabetic rats.
1.8.5.2.
Bradykinin
Kinins are generated from plasma and tissue precursors at the site of tissue injury and
contribute to many aspects of both acute and chronic inflammation including oedema
formation, vasodilatation, and pain (Cambridge and Brain, 1995; Dray and Perkins, 1993;
Khalil and Helme, 1992). BK, a naturally occurring polypeptide, has diverse actions on
the vascular endothelium, smooth muscle, and cellular function and is a powerful
analgesic agent, which can both sensitize and directly stimulate sensory nerve terminals
(Dray and Perkins, 1993) (Fig 1-12). BK (10μM) stimulates vascular endothelial cells to
produce NO and constricts vascular endothelial cells to produce plasma extravasation
(Khalil and Helme, 1992). At concentrations higher than 2μM, BK caused a significant
histamine release from rat peritoneal mast cells when extracellular calcium was removed
(Zhao et al., 1996). There is evidence to suggest that BK can stimulate sympathetic
nerves to release stored neurotransmitters (Jones et al., 1995) which was not supported in
an in vivo situation (Cambridge and Brain, 1995). This is because the mechanism of a
pro-inflammatory effect on the sympathetic nerve is not well understood. One study
suggested that vascular permeability and plasma extravasation observed with BK may be
mediated at least in part through activation of 12-lipoxygenase in the rat skin (Wang et
al., 1999). BK also contracts a variety of smooth muscle tissues, stimulates prostanoid
production in some tissues, exerts chemotactic and stimulatory effects on immune cells,
and degranulates mast cells (Dray and Perkins, 1993) (Fig 1-12).
BK receptors, B1 and B2, modulate the cellular action of kinins and have been localised
in animal neuronal tissues (Hall, 1997; Raidoo and Bhoola, 1998). Evidence suggests that
B2 receptors are more involved in acute inflammatory events, such as oedema and
inflammatory pain and this was supported by Khalil et al (1992) in a study of the rat skin
(Khalil and Helme, 1992). These authors showed a significant part of the response to BK
is mediated through an excitatory action on peripheral nociceptive C-fibres, involving B2
receptors, where sensory neurotransmitters are released to produce vasodilatation and
104
plasma extravasation via NO release from endothelial cells. B1 receptors, on the other
hand, appear to be involved in chronic inflammatory responses (Farmer et al., 1991; Hall,
1997). Induction of B1 receptors and responses to des-Arg9-BK are increased by
stimulation of macrophages as well as inflammatory mediators themselves such as
cytokines and ILs (Farmer et al., 1991). B1 agonists can increase prostacyclin and EDRF
in cultures of both endothelial and vascular smooth muscle cells (D'Orleans-Juste et al.,
1989).
Injury anoxia
Inactive
metabolites
inflammation low pH
Fig 1-12: Bradykinin synthesis,
Kallikreins
Kininogens
Kininase II
Endopeptidase
Bradykinin
Kininase I
[Des-Arg9]Bradykinin
Endothelium
Epithelium
Immune cells
Mast cells
B2
B2
B2
B2
B2
Prostanoid
production
vascular effects and
degradation (Adopted
from Dray, A. & Perkins,
M. (1993)).
B2
Smooth muscle
Nociceptive neurone
Excitation and sensitization
Sympathetic
neurones
There is a relationship between the metabolism of kinins and the control of blood flow by
the endothelium (Mombouli and Vanhoutte, 1995). For example, angiotensin-converting
enzyme (ACE) catalyzes the degradation and inactivation of BK, SP, and other peptides.
On the other hand, ACE inhibitors potentiate the action of kinins, normalize endothelial
function (Mombouli and Vanhoutte, 1995), and are capable of enhancing prostaglandin
production (Bennett and Gardiner, 1996).
A reduction in kallikrein has been found in diabetic individuals, suggesting that an
impaired kallikrein-kinin system (KKS) contributes to the development of diabetic
complications (e.g. nephropathy). High levels of urinary BK were detected in 12 weeks
STZ-induced diabetic rats (Tschope et al., 1999a). It was suggested that this increase is
related to increased filtration of components of the plasma KKS and / or renal kininogen
synthesis in combination with decreased renal kinin-degrading activity. The same study
also showed an impairment in the main kinin-forming enzymes, implying a reduced
105
capacity to generate BK in these animals (Tschope et al., 1999b), which was not
compensated by elevation in the level of BK receptors.
In a previous study, in vitro, treatment of vascular smooth muscle cell for 24hrs with high
glucose significantly increased the number of BK receptors (Velarde et al., 1997). The
same study suggested that BK could play a role in atherosclerotic vascular disease in
diabetes by increasing DNA synthesis and cell number, inducing MAP kinase activation,
and nuclear translocation in vascular smooth muscle cells. In STZ-treated mouse (10
days), B1 and B2 receptors were over expressed in the stomach fundus but not in the
urinary bladder (Pheng et al., 1997). Involvement of B1 kinin receptors in experimental
diabetic mice have also been reported (Zuccollo et al., 1996).
The vascular response to BK in diabetic rats, once again, seems to depend on the vascular
tissue under study as well as the duration of diabetes. It was previously reported that in
cremaster muscles of diabetic rats (3 weeks), the vascular response to BK was unchanged
(Alsip et al., 1996) but in the limb hindquarter of these animals (4 weeks), the response
was attenuated (Kiff et al., 1991b). The author of the latter study suggested that this
reduction was particularly dependent on NO synthesis that was selectively impaired in
STZ-treated rats. It was reported that the endothelial release of BK-stimulated relaxing
factor, in rat intestinal arterioles, was not compromised after 1-2 or 7-8 weeks of diabetes
(Lash and Bohlen, 1991) (see also section 1.6.4.2).
In normal rat tail artery, BK elicits a concentration-dependent vasoconstriction that
although is endothelium independent, is dependent on extracellular calcium entry and
intracellular calcium release. This contraction was previously reported to be significantly
reduced in diabetic rats treated with STZ (Wang et al., 1998). In the same study, it was
demonstrated that B2 receptors located on vascular smooth muscle mediates this
contraction. Altered B2 receptor-mediated vascular activity might play an important role
in vascular complications associated with diabetes.
The role of BK in modulating skin microvascular blood flow in early diabetes is
examined in chapter 4, an issue that was not previously addressed in other studies.
106
1.8.5.3.
Insulin
The Insulin molecule consists of two polypeptide chains (A & B), linked by two
disulphide bridges and is synthesized in the pancreas by beta cell of islets of Langerhans
(Pickup and Williams, 1997). Its secretion is modulated by numerous metabolic
hormones (e.g. glucagon, opioids, VIP, adrenaline, NA, etc) and neural factors (e.g.
vagal, sympathetic) (Pickup and Williams, 1997). In mammals, the main physiological
detriment of insulin release is the blood glucose concentration. Glucose must be
metabolized within beta cells to produce ATP, which closes specific ATP sensitive
channels in the cell membrane. This prevents potassium ions leaving the cell, causing
depolarization, which in turn opens voltage-gated calcium channels in the membrane
allowing calcium ions to enter the cell. The increase in cytosolic calcium then initiates
granule translocation and exocytosis of insulin.
glucose
G LUT2
ATP-sensitive K +
channel closes
glucose
K+
_
G lucokinase
G lucose 6-P
Ca 2+ channel
opens
Insulin
+
Depolarization
+
+
Exocytosis
Fig 1-13: Possible mechanisms involved
Granule
translocation
in insulin secretion by glucose
(Adapted from Pickup, J.C. &
Williams, G. (1997)).
ATP
beta cell
Insulin exerts its biological effects by binding to cell surface glycoprotein receptors on
the target tissues (Marino-Buslje et al., 1999). This causes a conformational change in the
receptor and activation of tyrosine kinase, essential for insulin signalling. The postreceptor signalling mechanisms are still largely unknown.
Insulin-receptor structure and function are affected by many physiological (exercise,
diet), hormonal (insulin), and pathological states (obesity, diabetes). Insulin resistance
characterizes patients with type I and type II diabetes (Yki-Jarvinen and Koivisto, 1986).
In type I, insulin resistance could be due to a post-receptor defect, diminished receptor
tyrosine kinase activity, and / or altered receptor carbohydrate structure (Pickup and
Williams, 1997). An in vitro study revealed that insulin receptors in skeletal muscle of
107
STZ-induced diabetic rats after 7 days diabetes were reduced and substrate kinase activity
was diminished (Burant et al., 1986). They also observed alterations in the mobility of
diabetic derived beta-subunits (structural change). A study, in vivo, on STZ-treated rats (2
weeks) suggested that tissue specific modulation of insulin receptors occurred at the level
of mRNA (Sechi et al., 1992).
Insulin’s vascular effects are mediated both by the sympathetic nervous system (chapter
10) and the L-Arg NO pathway (Scherrer et al., 1996). Insulin is an endotheliumdependent vasodilator (Baron, 1999; Cardillo et al., 1999; Nava et al., 1997). A study, in
vitro, showed that insulin can regulate the expression of the eNOS gene, in endothelial
cells and microvessels, mediated by the activation of phosphoinositide 3-kinase (PI-3)
(Kuboki et al., 2000). At physiological concentrations, insulin increases both cGMP and
cAMP in human vascular smooth muscle, endothelial cells, and platelets (Aljada and
Dandina, 2000; Trovati et al., 1997; Trovati et al., 1996). Increase in cGMP is via NOmediated stimulation of soluble guanylate cyclase and increase in cAMP may be
mediated by insulin-receptors. A previous study showed topical application of insulin
(200μU/ml) to the hamster cremaster muscle caused a significant increase in arteriolar
diameter (McKay and Hester, 1996), which could be inhibited by adenosine receptor
antagonism and blockage of ATP-sensitive potassium channels. As shown in Fig 1-13,
ATP-sensitive potassium channels are important in maintaining beta cell membrane
potential and may provide a crucial link between metabolic and electrophysiological
effects of glucose.
Evidence exists that insulin is involved in pro-inflammatory condition (Vianna and
Garcia-Leme, 1995) and insulin-sensitive depletion of SP and CGRP was also reported in
sensory nerves of STZ-treated diabetic rats (Diemel et al., 1992). When rats underwent
daily insulin treatment soon after the onset of diabetes, for the duration of 4 weeks, SP
and CGRP deficits in their sciatic nerve were reversed (Diemel et al., 1992), and
alterations in vasoreactivity and endothelial cell function were prevented (Stevens et al.,
1993). In diabetic rats, insulin treatment also restored neurogenic oedema towards normal
(Bennett et al., 1998). Intensive insulin treatment prevented sciatic nerve ischaemia in
rats with short-term diabetes (1-4 weeks) (Stevens et al., 1994) partially through
108
glycaemic control (Biessels et al., 1996). A previous study showed that insulin
deficiency, in type I diabetes, often leads to sensory, motor, and autonomic nerve deficits
(Vinik, 1999). It was suggested that type I diabetes (1 week) is associated with altered
levels of insulin signalling components (Bonini et al., 1995).
Na+/K+-ATPase, an integral component of the sodium pump, is involved in the
maintenance of cellular integrity and function such as contractility, growth, and
differentiation (Greene and Lattimer, 1983). There is evidence that insulin stimulates
Na+/K+-ATPase in skeletal muscle endothelium (Tack et al., 1996). The mechanism by
which insulin stimulates peripheral sensory nerves may also involve the Na+/K+-ATPase
system. This is supported by previous studies suggesting that the reduced Na+/K+-ATPase
activity in diabetic nerve may be responsible for a reduction in conduction velocity and
eventual structural abnormalities (Di Giulio et al., 1995; Vinik, 1999). Activity of both
Na+/K+-ATPase and Ca2+-stimulated ATPase but not Mg2+-ATPase was reported to be
depressed in diabetic rat heart (1-4 weeks) (Golfman et al., 1998; Ver et al., 1997). These
authors supported the view that sarcolemmal membrane abnormalities in Na+/K+ATPase, Na+/Ca2+-exchange and Ca2+-pump activities may lead to the occurrence of
intracellular Ca2+ overload in diabetes. Short-term insulin treatment of cultured vascular
smooth muscle cells, in vitro, caused a significant increase in mRNA encoding the alpha
2-subunit of Na+/K+-ATPase (Tirupattur et al., 1993). On the vascular smooth muscle, it
was suggested that insulin may act directly via stimulation of the Na+/K+-ATPase and
Na+-H+ exchanger, leading to hyperpolarization of the cell membrane and consequent
closure of voltage-gated calcium channels (Cleland et al., 1998). The Na+-H+ antiport of
retinal epithelial cells was down-regulated when these cells were exposed to high glucose
levels (in vitro, 26mM) for 2 weeks (Civan et al., 1994).
Insulin regulation of vasoconstriction and relaxation appears to be secondary to the
regulation of intracellular calcium, as insulin attenuates both voltage- and receptormediated calcium influx and stimulates both the transcription and activity of calciumATPase in vascular smooth muscle cell (Zemel, 1995). Although intracellular calcium
plays a poorly understood role in insulin signalling, increases beyond an optimal range
results in impaired insulin sensitivity, possibly by calcium inhibition of insulin-induced
109
dephosphorylation of insulin-sensitive substrates (Zemel, 1995). In liver cells, influx of
calcium was required for the stimulation of PI-3 kinase activity (Benzeroual et al., 2000).
Glucose control plays an important role in the pro-oxidant / antioxidant balance. It has
been hypothesized that oxidative stress can impair insulin action through: 1) changes in
the physical state of the plasma membrane of target cells, 2) an increase in intracellular
calcium content, and 3) a reduction in NO availability (Paolisso and Giugliano, 1996).
Evidence exists that insulin treatment reversed the increase in lipid peroxidation in
diabetes (1 week) (Rungby et al., 1992). This treatment also retarded and reduced the
intensity of advanced glycation process by regulating glycaemia (Turk et al., 1999).
It appears that insulin is the modulator of ET-1 peptide, ET receptors, and ET-1 mediated
responses in both in vitro and in vivo settings (Ferri et al., 1995; Frank et al., 1993;
Hopfner et al., 1998b) (Fig 1-14).
I n s u lin
Fig 1-14 Effects of insulin on factors
+
-
E T -B E T -A
related to ET-1 production and
action (Adopted from Hopfner,
R.L. et al (1999)).
ECs
+
+
EDRF
E T -1
E T -A
[C a 2 + ] i
VSM C
Physiological insulin concentrations can increase the release of ET-1 from both human
endothelial cells and vascular smooth muscle cells (Ferri et al., 1995; Hopfner et al.,
1998a; Hopfner et al., 1998b). At this concentration it can also up-regulate ETA receptor
expression and ETA-mediated intracellular free calcium level signalling in aortic smooth
muscle cells (Hopfner et al., 1998a). It was also hypothesized that insulin deficiency or
hyperglycaemia may partly be responsible for the enhanced ET-1 production induced by
ROS in 1 and 4 weeks diabetic kidney (Chen et al., 2000). Therefore, the relationship
between insulin, ET-1, NO, ROS, and AGEs may be of importance in pathological
condition such as diabetes.
110
In chapter 9, the role of insulin in the modulation of skin blood flow in early diabetes is
examined. Based on information presented including the findings of previous studies the
possibility was raised that adequate control of hyperglycaemia may prevent changes in
skin microvasculature in early diabetes.
1.8.5.4.
Protein Kinase C (PKC)
PKC, a family of kinases vital to intracellular signalling, consists of at least 11 isoforms.
The conventional PKCs (alpha, beta I, beta II, gamma) are calcium-dependent, binding
sites for DAG, and calcium / phospholipids. New PKCs (delta, epsilon, eta, theta, micro)
are DAG-sensitive but calcium-independent. Multiple isoforms of PKC are expressed in
each cell with several of them capable of mediating a similar range of functions that may
be cell-specific (sections 1.5.4.1 & 1.5.4.2).
One study (in vitro) showed that PKC isoforms binding to DAG are activated by high
glucose (20mM) in vascular smooth muscle cells (Haller et al., 1995). Total PKC activity
was increased, in the above study, after six hours, causing translocation of alpha, beta,
and epsilon isoforms into the nucleus. Also, PKC delta showed strong association with
cytoskeletal structures. Previous studies have demonstrated that diabetes has a selective
effect on the expression and subcellular distribution of isoforms of PKC (-alpha, -beta,
-gamma and -zeta). For example, it was suggested that PKC-beta and delta-isoforms were
activated preferentially in the vasculature of diabetic animals (Inoguchi et al., 1992; Koya
and King, 1998). A recent immunohistochemical study revealed a marked increase in
PKC-alpha immunoreactivity of cardiac capillaries, skeletal muscles and endothelial cells
of larger arteries of 4 weeks diabetic rats (Kang et al., 1999). The author concluded that
the increase in PKC-alpha in the endothelial cell supported a role for this isoform in the
functional endothelial impairments observed in diabetes. There are controversial reports
regarding PKC activity in the sciatic nerve of 4 weeks diabetic rats. One study showed
around 47% reduction in calcium-dependent PKC in these rats (Gabbay et al., 1990),
whereas others showed no significant change from control (Ido et al., 1994; Nakamura et
al., 1999). Studies on the sciatic nerve, spinal cord, and DRG of 6 weeks diabetic rats
111
were more supportive of the notion that diabetes-induced changes in the expression of
PKC beta II contributed to nerve damage (Roberts and McLean, 1997).
The role of PKC in diabetic neuropathy is still unclear. For example, defective PKC has
been implicated in impaired Na+/K+-ATPase activity in the sciatic nerve of 4 weeks STZtreated rats (Borghini et al., 1994). On the other hand, PKC inhibition was shown to
normalize the decrease in blood flow in diabetic animals (Cameron et al., 1997b). The
effect of ROS on PKC isoforms is also controversial. Recent evidence suggested that
PKC isoenzymes were subject to inactivation by ROS (e.g. glutathione) through
unidentified oxidative structural modifications (Ward et al., 2000). This author showed
that oxidative inactivation of PKC involves S-thiolation where endogenous thiols, such as
glutathione, react with PKC. Other studies showed that PKC isoforms can be activated
through tyrosine phosphorylation in response to hydrogen peroxide (Konishi et al., 1997;
Li et al., 1999). It is possible that some of the hydrogen peroxide-induced PKC activation
leads to vascular dysfunction in the early stages of diabetes (Konishi et al., 1997).
Activation of PKC under diabetic conditions was suggested to have a modulatory role in
oxidative stress-induced renal injury (Ha and Kim, 1999). Hyperglycaemia-induced
oxidative stress may also mediate the adverse effects of PKC-beta isoform by the
activation of the DAG-PKC pathway (Koya and King, 1998).
Numerous studies have shown a relationship between insulin, ET, and PKC activity. For
example, insulin treatment prevented an increase in PKC activity and DAG levels in the
heart (Inoguchi et al., 1994), aorta (4 weeks) (Okumura et al., 1991), and renal glomeruli
(2 & 4) of diabetic rats (Babazono et al., 1998). Also, excessive PKC activity might
interfere with insulin receptor signalling (Chin et al., 1994) under this disease condition.
It was suggested that ET-1 could also interfere with insulin signalling in smooth muscle
cells by both PKC-dependent and -independent pathways (Chin et al., 1994; Jiang et al.,
1999). In the former study it was found that ET-1 increased the level of receptor (betasubunit & IRS-2) phosphorylation as well as interfering with the PI-3 kinase pathway.
Under diabetic condition or in the presence of high glucose, change in PKC activity can
also regulate vascular permeability. It may increase permeability by direct
phosphorylation of the barriers formed by endothelial cells and skin granulation tissues
112
(Wolf et al., 1991), and / or indirectly by activating the vascular permeability factor
(Koya and King, 1998).
In chapter 6, the role of PKC in modulating skin microvascular blood flow is examined in
early diabetes.
1.9. Wound healing
Wound healing is defective in patients and animals with diabetes mellitus, which may be
attributed to many factors. For example, hyperglycaemia leads to osmotic diuresis and
subsequent decreased oxygenation and perfusion. It also limits polymorphonuclear
leucocyte (PMN) functioning and produces malnutrition by increasing hormones that
cause catabolism (Terranova, 1991). A recent study showed that impairment of the
microcirculation is a major factor causing delayed tissue repair in diabetes (Pham et al.,
1998). Although the exact mechanism underlying microvascular impairment is not
known, dysfunction of the endothelium has emerged over the last decade, as the
prominent abnormality related to vascular disease in diabetes (McMillan, 1997; Pickup
and Williams, 1997).
1.9.1. Skin
1.9.1.1.
Structure / extracellular matrix
The skin is one of the largest organs in the body, acting as a flexible barrier between the
organism and the external environment. It protects the body against physical trauma,
restricts water loss, plays a major role in temperature control, and being rich in nerves it
forms an extensive sensory surface (Wood and Bladon, 1985).
It consists of two distinguishable layers:
The outer epidermis is cellular, non-vascular, and is the most superficial layer of the skin.
It arises from the embryonic ectoderm and forms most of the cutaneous appendages
including sweat and sebaceous glands, hair, and nails. In the human, it is 0.06-0.1mm
thick on average and a highly organized kinetic system in which the cells divide, migrate,
differentiate and eventually die. The epidermis consists of epithelial cells that grow in
layers, the deepest cell layer resting on the dermis and attached to it by a basement
membrane. Cells of the basal layer on the basement membrane continuously divide and
113
daughter cells migrate towards the surface of the skin during which they undergo
differentiation. Ultimately they become packed with inert protein, keratin, and eventually
die to form a layer of flattish, dry, horny cells. This layer is inert and covered with a thin
layer of sebum, an oily, waxy material produced by sebaceous glands, to make it flexible
and water-resistant.
The lower connective tissue layer or dermis forms the bulk of the skin. It is a tough,
flexible, and highly elastic tissue that cushions underlying organs against mechanical
injury. The dermis contains blood vessels, lymphatics, nerve endings, some sweat glands
and hair follicles. It is much thicker than the epidermis (2-4mm), arises from the
embryonic mesoderm and supports and nourishes the cutaneous appendages. In
comparison with the epidermis, the dermis contains relatively few cells, mainly
fibroblasts, mast cells, macrophages, lymphocytes, and melanocytes. The papillary layer
of the dermis is adjacent to the basement membrane of the epidermis and includes ridges
and papillae of fine connective tissue fibres, loops of capillaries, and some nerve endings.
The reticular layer of dermis consists of coarse, dense, and interlacing collagen and
elastic fibres.
Beneath the dermis is a layer of loose connective tissue called the hypodermis, sometimes
recognised as a third layer of the skin. Over most of the body it forms a layer of adipose
tissue which provides thermal insulation and mechanical protection and the fat represents
an energy reserve.
Extracellular matrix
Most cells in a multicellular organism are in contact with an extracellular meshwork of
interacting macromolecules, which constitute the extracellular matrix, and the dermis is
no exception. Protein and polysaccharide molecules forming this matrix are secreted
locally and assemble into an organised meshwork. Long and strong collagen fibres in the
dermis strengthen and organise the matrix, the elastin fibres provide elasticity, while the
aqueous phase of polysaccharide gel known as ground substance permits the diffusion of
nutrients, metabolites, and hormones between the blood and tissue cells.
Collagen by volume is the principal constituent of the dermis. Collagen polypeptides
(alpha-chains) are synthesized on membrane-bound ribosomes (pro-collagen) which pass
114
into the endoplasm reticulum space where each pro-collagen polypeptide combines with
two others to form a hydrogen-bound triple stranded helix. Later, they are removed by
specific enzymes and remaining triple helical sections come together in the extracellular
space to form larger collagen fibrils, except type IV collagen (basement membrane
collagen), which seems to retain its extension peptides. After collagen fibrils have formed
in this space they are considerably strengthened by the formation of covalent cross-links
within and between the constituent collagen molecules.
Elastin appears to be synthesized by connective tissue cells and secreted into the
extracellular space to form filaments and sheets in which molecules are extensively crosslinked to form a network. This network allows material to stretch and recoil like rubber,
giving the skin its flexibility (Millington and Wilkinson , 1983; Wood and Bladon, 1985).
Ground substance where collagen and elastin are embedded is made of a carbohydrate
material known as glycosaminoglycan (GAG) (Millington and Wilkinson , 1983). This
term encompasses a whole group of compounds with similar structures, namely:
hyaluronic acid, dermatan sulphate, chondroitin sulphate, and heparan sulphate.
Basically, they are long, unbranched polysaccharides composed of a repeating
disaccharide unit. GAGs are highly negatively charged because of the presence of
carboxyl and sulphate groups on many of the sugar residues. The high negative charge
causes polymeric chains to adopt a stretched or extended configuration where the latter
gives a high viscosity to the surrounding region. Their function may be related to the
provision of normal suppleness and turgor as they are very hydrophilic and attract large
amounts of water. GAGs also act as filters by allowing salts to move freely but due to
their highly negative charge, they prevent large molecules such as proteins to diffuse into
their domain. Hyaluronic acid is different from the other GAGs because it lacks a
sulphate group, having very long chains, and not being joined to a protein core (Wood
and Bladon, 1985). It is produced in large quantities in tissues through which cells are
migrating during development or wound healing. It is possible that hyaluronic acid
attracts water and swells the matrix, thus facilitating cell migration (Wood and Bladon,
1985).
Dermal ground substance was shown to be susceptible to the action of a number of
hormones such as corticoids, sex hormones, thyroid hormones, anti-diuretics, and insulin
115
(Champion et al., 1970). A positive relationship between their degradation and the
duration of diabetes was found. Other factors such as radiation, ageing, certain infections,
as well as some dietary components (vitamin C and A) were also shown to have an effect
either on synthesis of GAGs or on the level of hyaluronic acid (Champion et al., 1970).
It is believed that negatively charged molecules such as proteoglycans could interact with
cells and under certain circumstances direct their migration, growth, and differentiation.
Degradation and turnover of proteoglycans have not been clearly demonstrated in the
skin, although polysaccharide-degrading enzymes have been found. These enzymes can
react upon hyaluronic acid and chondroitin sulphate, but not on dermatan sulphate or
proteases, which act upon the protein core of proteoglycans. The inability to degrade
proteoglycans has been used to diagnose proteoglycan storage diseases (Pinnell et al.,
1979).
1.9.1.2.
Vascular supply / Endothelium
Blood vessels develop from the somatic mesoderm very early in life. They subsequently
connect to the heart and once the flow is established, differentiation of smaller arteries
and veins occurs. Larger arteries and veins are formed independently and, after foetal life,
new vessels only arise by sprouting of the pre-existing blood vessel endothelium
(Millington and Wilkinson , 1983). The vascular bed is therefore able to develop
according to the requirements of the tissue concerned.
There are 4 classes of vessels in the skin: arteries, veins, capillaries, and lympathics.
Arteries of the skin pass through the deep fascia at definite points and thereafter in some
sites branch to form the first plexus. From this plexus branches pass to the hypodermis
where they give rise to the capillary network between fat cells. Other branches pass
through this layer to form a second plexus in the deep layer of the dermis to form the
capillary network around and through the sweat gland and papillae of the hair follicles.
Further branches rise to give a third plexus in the outer zone of the dermis (sub-papillary)
and to provide arteriovenous anastomoses (AVAs). These are specialised structures that
permit total blood flow to the skin to increase without a change in the capillary flow
(Millington and Wilkinson , 1983). They allow the cutaneous blood supply to perform its
116
dual task of thermoregulation and nutrition and are found in large numbers in the hands,
feet, nose, and ears.
Blood supply of the skin plays a far greater role in regulating body temperature than just
providing nutrients and removing waste products (Wood and Bladon, 1985). Highly
developed nervous control is exercised particularly through the smallest arteries, which
have the highest proportion of muscle in their walls, to control the blood flow in the skin.
Arteriovenous shunts can short-circuit the capillary circulation in the dermis to prevent
heat loss.
In circulation, muscular arteries and arterioles largely control the total blood flow by
converting the irregular pulsation of the heart into a more continuous flow. Smaller
arterioles eventually lead into non-contractile capillaries with variable width of 4 to10
μm. Nutrition occurs only through skin capillaries, and variation in flow from zero up to
2.5 ml/100g tissue/min is required to cover these needs (Champion et al., 1970). The precapillary sphincter is the last portion of a smooth muscle, circularly disposed around
arterioles. It controls the capillary flow and is influenced by local metabolic conditions
rather than by nervous impulses. Larger arteries are more under the control of
sympathetic nerves and less influenced by local conditions. The pre-capillary sphincters
on the other hand, control the capillary flow and are influenced mainly by local metabolic
conditions (e.g. histamine, CO2) rather than by nervous impulses and the pressure
gradient across the vessel wall. Post-capillary venules, although only weakly contractile,
play an important role in the regulation of circulation. Although a slight increase in their
tone will slightly reduce the flow rate, it makes an important difference in the rate of gas
or material exchange between these vessels and a tissue. These vessels largely contribute
to the normal skin colour and are involved in many pathological processes.
One of the major functions of the lymphatic system is to remove excess fluid and plasma
proteins from extracellular space as well as antigenic materials from tissues. Lymph,
which is similar to plasma but with different amounts of protein, moves in the lymph
vessel by the compression of surrounding muscles and the valves that monitor the
direction of flow. In the skin, each dermal papilla has a single lymphatic trunk that drains
into a sub-papillary plexus and then a deeper one with valves. Then lymph pass centrally
to the main lymphatic truncus and eventually into veins in the thoracic region. Flow in
117
the lymphatics is usually fairly slow but greatly increases with inflammation. In wound
healing, lymphatics sprout from pre-existing lymphatics and not blood vessels.
Endothelium
The endothelium is composed of a single layer of flattened squamous cells joined
together edge to edge to form a membrane of cells. It lines the innermost surface of blood
vessels and is uniquely situated to sense and respond to changes in circulation. The
endothelium plays a central role in the regulation of vascular muscle tone and ultimately
arterial tone and blood flow by releasing and / or producing vasoactive substances (both
vasorelaxants and vasoconstrictors) (Vane and Botting, 1994). Some of the vasoactive
substances released by endothelium are EDRF, prostaglandins, kinins, ET, histamine,
ACE, angiotensin II, cyclo-oxygenase (COX)-dependent constricting factor, and
superoxide anions, etc (Bennett and Gardiner, 1996).
Activation of endothelial cells stimulates NO production and its release can lead to
vascular smooth muscle relaxation by activation of guanylate cyclase and an increase in
the cell cGMP. Continuous basal release of EDRF (NO) from most blood vessels,
particularly arteries, contributes to resting vascular tone. Endothelial cell damage or
inhibition of NO synthesis, on the other hand, leads to an increase in vessel tone (Bennett
and Gardiner, 1996). Not all endothelium-dependent vasodilatations can be explained by
the release / action of EDRF (NO).
Endothelial cells are also sources of endothelium-derived constrictor factors (EDCF),
EDHF, and prostaglandins, which can be released following stimulation of endothelial
cells by vasoactive substances or other physiological stimuli such as changes in blood
flow, thrombin, NA, stretch, hypoxia, etc. At least three different classes of EDCFs have
been recognised: 1) metabolites of AA, 2) the polypeptide ET, and 3) a still unidentified
factor released from anoxic / hypoxic endothelial cells (Bennett and Gardiner, 1996).
1.9.1.3.
Innervation
The skin is well supplied with nerves that are constantly sending information about the
environment to the central nervous system. Sensory nerves have a variety of endings,
many being naked (pain, touch), while others end in sensory cells (temperature, pressure).
118
This presumably enables a range of ‘sensations’ to be detected, from light touch to heavy
pressure or pain, heat and cold, and vibrations (Wood and Bladon, 1985). Some areas
such as the fingertips are much richer in nerve endings and consequently more sensitive
than other regions. The mechanism of sensation (e.g. roughness, moistness) is probably
brought about by varying stimuli acting on the ‘sensory unit’ in the skin, sending a
mixture of temporal and spatial pattern of impulses to the cerebral cortex (Wood and
Bladon, 1985). These may then reach the level of consciousness where, together with the
individual’s past knowledge and experience, will be recognized as a particular sensation.
Sympathetic fibres accompany cutaneous nerves and go to sweat glands, blood vessels,
and erector muscles of hairs.
Skin vessels, both arterioles and venules, are under reflex neural control from a rich
sympathetic adrenergic nerve supply. These active vasoconstrictor nerve fibres are the
afferent arms from 4 types of sensory input: a) cutaneous thermoreceptors, b)
baroreceptors of arteries and the cardiac-pulmonary system, c) chemoreceptors, and d)
receptors activated by upright posture and exercise. Sympathetic nerves reflexly control
the volume in veins but their responsiveness is modified by changes in local temperature.
Veins are not responsive to baroreceptors but they do constrict during exercise. There are
no parasympathetic nerves in the skin, only those from the sympathetic axis giving an
adrenergic innervation to the smooth muscle of blood vessels, nipple, and dartos
(Champion et al., 1970).
1.9.2. Mechanism of tissue repair
Upon tissue injury, the complex process of wound repair is initiated which consists of
three phases: inflammation, proliferation, and tissue remodelling (Stadelmann et al.,
1998; Waldorf and Fewkes, 1995). The inflammatory phase is marked by platelet
accumulation, coagulation, and leucocyte migration. The proliferative phase is
characterized by re-epithelialization, angiogenesis, fibroplasia, and wound contraction.
Finally, the remodelling phase takes place over a period of months, during which the
dermis responds to injury with production of the collagen and matrix proteins and then
returns to its pre-injury phenotype. Wound healing is an important area of research but an
in-depth discussion regarding the healing mechanism of skin wounds is beyond the scope
119
of this thesis. Therefore, the mechanism of wound healing in the skin is briefly discussed
and only certain factors relevant to this thesis are mentioned.
Inflammatory phase
Following cutaneous injury, the endothelial integrity is disrupted while antidromic
release of substances in an axon reflex induces a number of vascular changes. This is
followed by active vasodilation that usually becomes most pronounced within 20 min
after the injury and is accompanied by an increase in capillary permeability. Histamine is
believed to be a key chemical mediator of these actions. Platelet aggregation then follows
with concurrent stimulation of the coagulation cascade. Platelets are also a storehouse for
a great number of growth factors and vasoactive substances such as PDGF, TGF-beta,
FGF, platelet-derived angiogenesis factor (PDAF), epidermal growth factor (EGF), betathromboglobulin, serotonin, BK, prostacyclins, thromboxane, and histamine (Greenhalgh,
1996). Platelet aggregation generates a coagulation factor, thrombin, which in turn
recruits additional platelets to release agents such as fibronectin and fibrin to promote
platelet adhesion and to assist with the reflux of monocytes and macrophages.
Fibronectin deposition (24-48 hrs after injury) creates scaffolding on which fibroblasts
can migrate into the wound. It is the fibroblast population that will become the dominant
cell in the healing wound once the inflammatory phase has subsided. A variety of cell
populations such as PMNs and mononuclear leucocytes will enter the trauma site as a
result of increased vascular permeability and initially mature into wound macrophages
and later into lymphocytes. Monocytes are considered to be the most vital component of
the early phase of normal wound healing, however, PMNs are essential for protecting the
wound against infection. Macrophages / PMNs destroy bacteria and remove devitalized
tissue fragments and debris. Activated neutrophils release free radicals and lyzozomal
enzymes to help infection and clean the wound. Mediators such as thrombin and
kallikrein are chemoattractant to circulating leucocytes, as well as being capable of
releasing cytokines, which facilitate subsequent cellular chemotaxis, proliferation, and
extracellular matrix production. The cytokine PDGF is a potent mitogen for fibroblasts
and smooth muscle cells, and a chemoattractant for many other cells (Hosgood, 1993).
120
Proliferation
Fibroblasts appear in the wound within 2-3 days and dominate the wound population of
cells within a week-They migrate to the wound site via scaffolding made by fibronectin
and hyaluronate, the early extracellular matrix. Fibroblasts produce a variety of
substances essential for tissue repair such as GAG and collagen (section 1.9.1.1). GAGs
form ‘ground substance’ and collagen levels rise up to 3 weeks (increasing wound tensile
strength) until a point of homeostasis is achieved. During the first 2-3 days fibroblasts
mainly replicate and migrate rather than synthesize collagen, so very little in wound
strength occurs during this period, often refered to as the ‘lag phase’. However, this term
has been abandoned due to an enormous amount of cellular metabolism and fibroblastic
activity that occurs during this period. Collagen synthesis is a seminal feature of
fibroplasia. Fibroblasts are the major source of collagen and wound connective tissue.
Angiogenesis accompanies this fibroblast phase where new capillary growth
accompanies advancing fibroblasts into the wound to provide for their metabolic needs. If
angiogenesis fails, fibroblast migration will be arrested and wound healing fails to
proceed.
Wound re-epithilialization represents a sequence of steps involving mobilization,
migration, mitosis, and cellular differentiation of epithelial cells. The epithelial cells
immediately adjacent to the wound are stimulated to begin migration after loss of contact
inhibition resulting in their growth away from adjacent intact epithelial cells (to cover the
wound) (Stadelmann et al., 1998).
The phenomenon of wound contraction occurs at about 1 week after injury. At this time a
portion of the wound fibroblasts undergo transformation into specialized cells,
myofibroblasts. These cells are able to form secure intercellular attachments and since
they are adherent to one another as well as the wound margins, the entire granulation bed
contracts, drawing the wound edges closer together (Waldorf and Fewkes, 1995).
Simultaneously, collagen is synthesized, deposited, and cross-linked to form a rigid
scaffold holding the wound in place.
121
Tissue remodelling
After 3 weeks (post-injury), wound remodelling begins and continues for up to 2 years.
Although there is no net increase in collagen content during this period, there is reorganization of collagen fibres into a more organized lattice structure determined by local
mechanical factors. During this phase, the wound continues to increase in tensile
strength. The majority of type III collagen fibres placed down early in the healing process
are replaced by type I. Type I, II, and III collagen are responsible for most of the tensile
strength in the body. During the tissue remodelling phase, GAGs are steadily degraded
and remodelling continues until the GAG concentration and collagen ratio returns to that
found in normal skin.
1.9.2.1.
Role of sensory nerves
An intact nociceptor system of capsaicin-sensitive primary afferent sensory nerves is
important for the initiation of the inflammatory process and successful tissue repair
(Brain, 1997; Holzer, 1991). Unmyelinated afferent nerves with polymodal nociceptors
mediate early components of the neurogenic inflammatory response, namely early
neutrophil migration (Eglezos et al., 1991), plasma extravasation (Khalil and Helme,
1989), local vasodilatation (Khalil and Helme, 1990b), and modulation of cellular
components of the immune response (Eglezos et al., 1991). These are mediated by
neuropeptides (e.g. SP, CGRP) contained in peripheral terminals of capsaicin-sensitive
nerves (Brain, 1997; Khalil and Helme, 1996). Healing of the injured tissue requires
adequate organisational and functional reactions of the primary afferent sensory nerves
and the microcirculatory system.
SP and CGRP are involved in wound healing. A recent immunohistochemical study of
human burn wound healing demonstrated the presence of SP in the viable dermis in early
burns and in a fibrous network of older burns, suggesting the involvement of this peptide
in several steps in the repair process (Dunnick et al., 1997). A previous study also showed
that CGRP increased survival of a musculocutaneous flap, in rats with skin injury, an
effect independent of its vasodilator activity (Kjartansson et al., 1987). These growth
promoting effects of sensory neuropeptides are thought to be involved in wound healing
and contribute to the trophic functions of C-afferent fibres. A recent study has also been
122
supportive of the hypothesis that sympathetic efferents are important for wound healing
(Kim et al., 1998).
NO is a potent vasodilator with the ability to increase vascular permeability and therefore
may be important in inflammation, tissue injury and / or wound healing (Shuler et al.,
1995). On the other hand, NO can also impose deleterious effects on healing process by
inhibiting platelet adherence / aggregation (Radomski et al., 1990) and vascular smooth
muscle proliferation (Garg and Hassid, 1989). The conflicting data produced by previous
investigations on the physiological role of NO in the wound healing could be due to the
major differences in the injury model, the type of animal used, and the timeframe of the
investigation.
1.9.2.2.
Role of growth factors
Growth factors are important signal transducers in the wound microenvironment. They
are known to play a role in cell division, migration, differentiation, protein expression,
and enzyme production (Greenhalgh, 1996). Almost all growth factors are peptides that
bind to a cell surface receptor protein causing intracellular reactions that are not entirely
understood.
There are 7 major families of growth factors: EGF, TGF-beta, IGF, PDGF, FGF, ILs and
colony stimulating factor (CSF); but this is not a complete list of all growth factors
known. They are a group of hormone-like polypeptides that play a central role in
different phases of wound healing. Growth factors attract cells into the wound,
stimulating their proliferation, and have a profound influence on extracellular matrix
formation (Ashcroft et al., 1995; Greenhalgh, 1996; Martin et al., 1992; Okumura et al.,
1996; Stoscheck et al., 1992). Sensory peptides are important for the chemotactic activity
of monocytes, macrophages, and T-lymphocytes in skin wounds and for the subsequent
availability of transforming growth factor (TGF)-alpha, which is essential for healing
(Richards et al., 1997). Sensory neuropeptides and growth factors could have an
overlapping range of actions during the wound healing process.
The event of early wound healing reflects a finely balanced environment in which
proteolytic activity and matrix synthesis occur under tight regulation, leading to
uncomplicated and rapid tissue repair. In chronic wounds, this fine balance is lost causing
123
delayed healing. Since describing roles of all these growth factors is beyond the scope of
this thesis, only a few were discussed. Several studies provide evidence that EGF
stimulates rapid epithelialization of skin wounds with an increase in EGF-receptor
expression in the epidermis during the proliferation and migration of epidermal
keratinocytes (Nanney, 1990; Stoscheck et al., 1992). Basic FGF (bFGF) can also
improve healing largely by enhancing angiogenesis and granulation tissue formation
(Okumura et al., 1996). There is evidence that IGF-1 stimulates GAG biosynthesis
(Cechowska-Pasko et al., 1999) and stimulates keratinocyte proliferation (in vitro)
(Blakytny et al., 2000). PDGF and TGF-beta are both primarily secreted from platelets
but also from activated macrophages and fibroblasts. They not only possess chemotactic
properties but also increase collagen content and strength in wounds in normal and
compromised tissues (Hosgood, 1993).
1.9.3. Effect of diabetes
1.9.3.1.
Structural changes
Wound healing data derived from human and animal models of diabetes revealed a
reduction in contraction of open wounds, decreased number of fibroblasts, and decreased
number of PMNs (Franzen and Roberg, 1995; Hehenberger et al., 1999; Stadelmann et
al., 1998).
A previous study demonstrated diminished wound strength in STZ-induced diabetic rats
accompanied by a reduced collagen content (Verhofstad and Hendriks, 1994). In this
study, the fact that the diabetic skin is thinner than normal individuals was ruled out as a
reason for the decrease in the wound breaking strength (Bitar, 1998). In a different study,
using 4 weeks diabetic rats with circular full-thickness skin wounds, decreased granulated
tissue was suggested to be due to the increase in wound collagenase level (Ramamurthy
et al., 1998). Many local and systemic factors can delay wound repair such as tissue
ischaemia, intra- and extra-vascular fibrin deposition, vasodilatation of non-nutritive
microcirculation, necrosis, infection, impaired migration of the epithelial cells of the
ulcer edge, and inadequate cytokine pattern (Morian and Colen, 1990). For example, in
diabetic mice, decreased neovascularization and less organization of the granulation
124
tissue during the late inflammatory phase of wound healing were suggested to be due to
altered leucocyte infiltration and wound fluid IL-6 levels (Fahey et al., 1991).
A relationship between IGF-1 and GAG biosynthesis and their role in full thickness
wounds in diabetic animals has also been documented (Cechowska-Pasko et al., 1999;
Tsuboi et al., 1995). Also reduction in the synthesis and content of heparan sulphate, a
proteoglycan, in the skin of acute diabetic animals (Cechowska-Pasko et al., 1996b;
Rohrbach et al., 1983) was shown to be inversely related to the level of plasma glucose.
By normalising the glucose level, using insulin treatment, the level of synthesis was
returned to normal in these animals (Rohrbach et al., 1983). Heparan sulphate in the
basement membrane restricts the passage of proteins; its absence could account for the
increased porosity of the basement membrane in diabetes.
1.9.3.2.
Physiological changes
Physiological changes related to the wound healing process observed in diabetic rats are
increased oedema, increased rate of infection, decreased capillary volume, decreased
neovascularization, delayed reappearance, and abnormal reformation of epithelial
basement membrane proteins (Fahey et al., 1991; Morian and Colen, 1990; Nemeth et al.,
1999b; Sato et al., 1999). The role of sensory neuropeptides (SP, CGRP) as trophic
agents was described earlier. Delayed healing may be related to a reduced level of
sensory neuropeptides.
In diabetes mellitus, there is evidence that sensory neuropeptide level is reduced in axons
and cell bodies of primary afferent nerves (Diemel et al., 1992; Tomlinson et al., 1988),
leading to reduced inflammatory responses in these animals (Mathison and Davison,
1993; Nemeth et al., 1999b). The role of sensory neuropeptides in acceleration of wound
healing was previously documented (Baker et al., 1997; Khalil and Helme, 1996; Khalil
and Merhi, 2000). Khalil and Helme (1996) have provided direct evidence for the ability
of sensory neuropeptides (exogenously applied) to accelerate wound healing in aged rats
and indirect evidence to suggest a possible role for sensory peptides in modulating wound
contraction and re-epithelialization. This latter proposition is supported by the Watcher
and Wheeland (1989) study, which provided evidence that topical application of
capsaicin accelerated wound contraction and re-epithelialization of full-thickness
125
wounds. The authors suggested that effects of capsaicin might be mediated via the release
of SP from sensory nerves. Furthermore, Khalil and Helme (1996) provided evidence that
interaction between these peptides can influence their ability to modulate the healing
process. In a study by Khalil and Merhi (2000), tissue repair was accelerated by increased
levels of neuropeptides released from sensory nerves using transcutaneous electrical
nerve stimulation. In the Baker et al (1997) study, again electrical stimulation of sensory
nerves using specific parameters accelerated the healing process in diabetic patients with
ulcers.
Intact fibroblast function is required for normal wound healing. An in vitro study showed
suppression of fibroblast proliferation in cells exposed to high glucose (Hehenberger et
al., 1999) suggested that PKC and hyaluronic acid might play a role as modulators of
fibroblast proliferation during the diabetic state.
A recent study showed a significant decrease in neutrophil counts and phagocytosis in 5
weeks diabetic rats with an incisional skin wound (Canturk et al., 1999). Neutrophils play
a critical role in the host defence mechanism and impairment in its function can cause
healing difficulties in diabetes with or without infections (Stadelmann et al., 1998).
Studies on immune cells necessary for wound healing, such as PMN and fibroblasts,
suggest that there is a delayed response to injury and impaired functioning of immune
cells in diabetes mellitus (Rosenberg, 1990).
Elevated ROS in short-term diabetes (Mullarkey et al., 1990; Traverso et al., 1998)
reduced tissue blood flow by causing deficits in nerve conduction velocity (Cameron et
al., 1991), which can, in turn, affect the release of neuropeptides from sensory nerves
(Nemeth et al., 1999a). ROS also cause antioxidant-preventable vascular endothelium
abnormalities, neutralizing NO-mediated vasodilation, and increased reactivity to
vasoconstrictors (Stohs, 1995). In both in vivo and in vitro settings, insulin increased the
activity of some antioxidants (Cu-Zn-SOD, catalase, glutathione peroxidase) and the
production of hydrogen peroxide in rat macrophages (Pereira et al., 1995). Insulin
improved blood flow and wound matrix formation in burn patients (Pierre et al., 1998).
The former effect may be due to an improvement in antioxidant defence mechanisms as
126
well as production of hydrogen peroxide (Pereira et al., 1995). ROS can influence
collagen remodelling by affecting the collagen structure (Arisawa et al., 1996). NO plays
a significant role in wound reparative collagen deposition (Schaffer et al., 1997) as well
as being involved in inflammatory responses (Bennett and Gardiner, 1996). ROS are
known to destroy NO (Gryglewski et al., 1986; Kamata and Kobayashi, 1996) and in this
way they may contribute to delayed healing. Furthermore, endothelium-derived NO may
play an important role in mediating angiogenesis by supporting endothelial cell migration
(Murohara et al., 1999), in part via an integrin-dependent mechanism.
High levels of ET-1 were previously detected in both diabetic humans and STZ-treated
rats (Makino and Kamata, 1998; Nakamura et al., 1997; Takahashi et al., 1990). An
increase in ET-1 was suggested to be one of the factors responsible for partial necrosis of
the skin flap wound contributing to the genesis of the no-reflow phenomenon (Matzuzaki,
1993). On the other hand, it was reported that ET stimulated transmigration of endothelial
cells and accelerated endothelial wound healing by acting on ET-B receptors (Goligorsky
et al., 1999). The author showed that, in vivo, ET enhances new vessel formation, an
effect that requires functional eNOS.
In chapter 10, the information gained from previous chapters is utilized to examine the
effects of specific treatment protocols (local and systemic) in modulating the repair of
full-thickness skin wounds in 4 weeks diabetic rats. It was of interest to prevent the
effects of a number of factors namely ETs, free radicals, AGEs, and hyperglycaemia on
microvasculature in early diabetic rats, using these treatment protocols, and examine the
progression of tissue repair process.
127
1.10.
Aims and hypotheses
Unmyelinated primary afferent sensory neurones and the microvasculature innervated by
these nerves play a major role in modulating the inflammatory and tissue repair
processes. Alterations in these processes caused by diabetes mellitus can lead to the
impaired blood flow and delayed tissue healing. The mechanisms underlying changes in
skin microvascular blood flow and their effects on tissue repair processes, especially in
the early stages of diabetes, are not clear and require further investigation.
Therefore, the aims of this thesis are:
1. To establish a time frame for the occurrence of skin microvascular changes in the
STZ-treated diabetic rat model. Chapter 3 presents the findings of the investigation
undertaken to address this aim. It was hypothesized that there is a relationship
between the duration of diabetes and the occurrence of these changes.
2. To examine alterations in sensory nerve-mediated events (pre- and post-terminal
levels) and identify the mechanisms underlying these alterations in 4 weeks STZtreated diabetic rats. Once again, chapter 3 presents the findings of the investigations
undertaken to address these aims. It was hypothesized that the contribution of both
pre- and post-terminal mechanisms might contribute to reduced inflammatory
responses in early diabetes.
3. To examine the effect of 4 weeks diabetes on both the superficial and deep
compartments of the rat skin microvasculature. Chapters 3 and 4 present the findings
of the investigations undertaken to address this aim. It was hypothesized that there are
alterations in the function of microvessels that contribute to the reduction in skin
microvascular responses in early diabetes.
128
4. To examine alterations in the function of parasympathetic and sympathetic nerves and
identify mechanisms underlying these changes in 4 weeks STZ-treated diabetic rats.
Chapter 4 and 5 present the findings of the investigation undertaken to address these
aims. It was hypothesized that these mechanisms contribute to alterations in
microvascular responses in early diabetes.
5. To examine the role of local tissue factors in modulating microvascular responses in 4
weeks STZ-treated diabetic rats. Chapter 4 presents the investigation undertaken to
address this aim. It was hypothesized that since local tissue factors could modulate
the function of sensory nerves, any alterations in their function could influence
microvascular blood flow.
6. To examine the role of ETs, free radicals, AGEs, and hyperglycaemia in modulating
microvascular responses in 4 weeks STZ-treated diabetic rats. Chapters 6, 7, 8, and 9
present the results of the investigations undertaken to address this important aim. It
was hypothesized that ETs, free radicals, AGEs, and hyperglycaemia may be
involved in altering skin microvascular function in early diabetes, which may be
inhibited by early intervention.
7. To examine the effects of specific treatment protocols (local and systemic) in
modulating the repair of full-thickness skin wounds in 4 weeks diabetic rats. Chapter
10 presents the findings of the investigation undertaken to address this aim. It was
hypothesized that preventing the deleterious effects of ETs, free radicals, AGEs, and
hyperglycaemia, using specific inhibitors / antagonists, may improve skin
microvascular blood flow and tissue repair processes.
129
CHAPTER 2
Materials and General methods
130
2.1.
Materials
Rat food (GR2 and mouse breeder cubes) was provided by Barastock Stockfeeds Pty Ltd
(Melbourne, Australia). Neuropeptides and chemicals were purchased from different
sources:
1.
Nembutal (Pentobarbitone sodium, 60mg/ml): Boehringer Ingelheim Pty Ltd
(North Ryde, Australia)
2.
Substance P (SP), Endothelin (ET)-A receptor antagonist (BQ-123), Calcitonin
gene-related peptide (CGRP): Auspep Pty Ltd (Melbourne, Australia)
3.
Streptozotocin (STZ) (mixed anomers), Alloxan monohydrate, ET-B receptor
antagonist (BQ-788), Sodium nitroprusside (SNP), Superoxide dismutase (SOD)
and PEG-SOD (both from bovine erythrocytes), Bradykinin (BK), NG-Nitro-Larginine methyl ester (L-NAME), Nicotine, Muscarine chloride, N-acetylcysteine
(NAC), Guanethidine monosulfate, Capsaicin: Sigma Chemical Company (USA)
4.
Aminoguanidine hydrochloride (AG): Sapphire Biosciences Pty Ltd (Alexandria,
Australia)
5.
Bisindolylmaleimide (BIM): Calbiochem-Novabiochem Corp (Melbourne,
Australia)
6.
Amadori products, Advanced glycated end-products (AGEs), anti-RAGE
antibodies (IgG): Peptech (Dee Why, Australia) (See Appendix C)
7.
Tirilazad mesylate (Freedox) (1.5mg/ml): Pharmacia & Upjohn (Rydamere,
Australia)
8.
Protophane penfill (isophane insulin, 100IU/ml): NovoNordisk Pharmaceuticals
Pty Ltd (Melbourne, Australia)
2.2.
Animals
Male outbred Sprague-Dawley rats were used in this study as both alloxan and
streptozotocin can easily cause diabetes in these animals (Dunn and McLetchie, 1943;
Rakienten et al., 1963). They were purchased from Monash University animal services
(Melbourne, Australia) and were delivered to the National Ageing Research Institute
(NARI), where all the experiments were performed, on a regular basis. Upon arrival, rats
were placed in a special room where the temperature was kept at 22°C and the light was
131
on a 12hr dark / light cycle throughout the year. The original weight of rats at the start of
every treatment was around 130-150 g (6-7 weeks old).
Rats were anaesthetised by a single dose of pentobarbitone sodium (65mg/kg) at the start
of every experiment and added doses were administered to ensure rats remained under a
constant state of anaesthesia. The loss of eyelash-reflex was used as a test to ensure the
maintenance of anaesthesia throughout the experiment. Animals were kept under a deep
state of anaesthesia for all experimental (invasive) procedures in order to comply with the
ethic’s regulation. Animals were fully conscious during pre-treatment with alloxan,
streptozotocin, insulin, etc. At the end of each experiment rats were sacrificed with
anaesthetic overdose.
2.3.
Types of food
2.3.1.
Mouse Breeder
This type of food is rich in certain nutrients, which are required for growing young
rodents (Table 2-1). Some vitamins such as A, E, and niacin that are present in this type
of food, seem to play an important role in delaying or preventing diabetes onset (Chapter
1, section 1.7.2.3).
2.3.2. GR2
As shown in table 2-1, the content of GR2 is suitable for slightly older rats and contains
lower levels of vitamin A, E, and niacin among others (Table 2-1). In this study, more
rats develop diabetes (more than 80%), using STZ / alloxan treatments, when rats were
fed GR2 instead of mouse breeder cubes.
132
Content
Mouse breeder
GR2
Vitamin A
15000 IU/kg
9500 IU/kg
Vitamin B1
64.7 mg/kg
10.78 mg/kg
Vitamin B2
60
mg/kg
12.8
mg/kg
Folic acid
11
mg/kg
2
mg/kg
Vitamin K
50
mg/kg
9.4
mg/kg
Pantothenic acid
250
mg/kg
45.5
mg/kg
Pyridoxine
30
mg/kg
5
mg/kg
Niacin
400
mg/kg
80.7
mg/kg
Vitamin E
260
mg/kg
50
mg/kg
2.4.
Table 2-1: Major differences between
the content of Mouse breeder
and GR2.
Induction of diabetes
2.4.1.
Choosing a diabetogenic agent
In animals, diabetes can be induced chemically using a diabetogenic agent at a specific
dose (Abdel-Rahman et al., 1992b; Rakienten et al., 1963). Soon after induction of
diabetes
1. High levels of glucose should be present in animal’s urine and blood
2. Animals should show physical symptoms of diabetes (i.e. weight loss, polydipsia,
polyphagia, polyuria)
3. Animals should respond to insulin treatment by showing a reduction in the level of
glucose in urine and blood and reduction in the physical symptoms of diabetes
In this study, two diabetogenic agents namely STZ and Alloxan were used. Rats (130150g) (6-7 weeks old), fed with GR2, were fasted for 24 hrs before injecting them with
one dose of freshly prepared alloxan or streptozotocin. It was suggested that
hypoglycaemia is severe in fasted rats treated with these chemicals and it can lead to
frequent attacks of convulsions and finally death (Rerup, 1970). For this reason, injected
rats were given 5% sucrose solution for 48hrs after the injection to avoid immediate
death. The chosen diabetogenic doses and conditions resulted in more than 80% success
rate in creating diabetic rats whether they were treated with alloxan or STZ.
133
The classic symptoms of diabetes (polydipsia, polyuria, polyphagia, weight loss) became
generally apparent within 2-6 days after injection of the chemical, depending on which
agent was used, and diabetes was confirmed by urine glucose test (Diastix, Bayer
Australia Ltd). Before each experiment the blood glucose of every rat was also measured
with Reflolux S (Boehringer Mannheim, Ausatralia Pty Ltd) and those with blood
glucose of 27.7 mmol/lit or more were included in the study (section 2.5).
2.4.2.
Control rats
Control rats of similar age / weight (as above) were fasted over night. Then instead of a
diabetogenic agent, they were injected intraperitoneally (i.p.) with a similar volume of
0.1M cold sodium citrate buffer. These rats were kept under the same condition as the
experimental group and fed the same diet (GR2). Control rats were tested at the same day
as the experimental rats. The data representing controls were used throughout the thesis
to facilitate comparison with diabetic data.
2.4.3.
Alloxan treatment
Alloxan (40mg/kg) was dissolved in cold sterile saline and was kept on ice to avoid
decomposition. After 24hrs of fasting, rats were injected with one dose of freshly
prepared (cold) alloxan solution intravenously (i.v.) into the tail vein (Abdel-Rahman et
al., 1992b). Symptoms of diabetes became apparent 4-6 days later and diabetes was
confirmed by urine glucose test. Before each experiment the blood glucose of every rat
was also measured with Reflolux S and those with blood glucose of 27.7 mmol/lit or
more were included in the study (section 2.5). Alloxan-treated rats coped well with the
severity of their physical symptoms and did not require insulin injection during the
experimental period.
2.4.4.
Streptozotocin
STZ (75mg/kg) was dissolved in 0.1M cold sodium citrate buffer (pH 4) and kept on ice
to avoid decomposition (section 2.4.6). Once again, after 24hrs of fasting, rats were
injected (i.p.) with a single dose of freshly prepared (cold) STZ (Rakienten et al., 1963).
Symptoms of diabetes became apparent in these rats earlier than alloxan-treated group
(within 2-3 days) and their diabetic condition was confirmed by urine glucose test. Beside
134
classic symptoms of diabetes, other symptoms such as a change in the condition of their
fur, development of cold and pale paws, severe diarrhoea in some cases, and reduced
physical activity, were also observed in both alloxan- and STZ-treated rats. However,
overall changes in physical symptoms were more severe in the latter group. As a result,
STZ-treated diabetic rats were given one or two insulin injections, when the severity of
symptoms became life threatening, to provide a temporary relief and prevent death
(section 2.4.6). After insulin treatment, apart from obvious physical improvements in
these animals, their survival rate was also enhanced. Rats were sacrificed at any point if
they lost 15% or more of their body mass to comply with the ethic’s regulation.
2.4.5.
STZ vs Alloxan
STZ was chosen as a preferred agent to alloxan in creating an animal model of diabetes
mellitus for this study after careful consideration of both advantages and disadvantages of
the two chemicals.
1. STZ treatment induced diabetes in rats in a shorter period of time (2-3 days)
compared to alloxan treatment (4-6 days).
2. To achieve a high success rate (80% or more) in creating diabetic rats i.p., it was
more economical to apply STZ (75 mg/kg) than alloxan (100mg/kg).
3. Diabetic condition was more severe in STZ than alloxan-treated animals (i.e. blood
glucose above 27mmol/lit).
4. No spontaneous recovery was observed in STZ-treated rats.
5. The mortality rate was slightly higher in STZ than alloxan-treated groups due to
severity of diabetes. However, this was easily prevented if diabetic rats were given
5% sucrose solution for the first 48hours after STZ injection.
2.4.6.
Modifications made to the original procedure
The original adapted procedure for induction of diabetes in rats, using STZ (Rakienten et
al., 1963; Takeuchi et al., 1994) and alloxan (Abdel-Rahman et al., 1992b), was modified
in the following manner to produce the best success rate and reduce the total number of
animals required for experiments in this thesis.
135
1. Food
Previously Mouse breeder cube was the only standard food used in our laboratory. In the
original experiments, STZ treatment failed to cause diabetes in the rats that were on the
Mouse breeder cube diet. Careful analysis of content of this food showed high levels of
vitamin A, E, and niacin. Previous studies suggested that these vitamins could play a role
in inhibiting the diabetogenic action of STZ by protecting pancreatic beta cells (Beales et
al., 1994; Hoorens and Pipeleers, 1999) (see also chapter 1, section 1.6.2.2.1). Changing
from Mouse breeder cube to GR2, which has lower levels of these vitamins, reduced their
effect and STZ injection could induce diabetes in GR2 fed rats within 2-3 days.
2. STZ and alloxan solution
STZ is an unstable compound that dissolves in saline at room temperature but
decomposes within a few minutes by forming a visible gas. Dissolving STZ in a cold
freshly made 0.1 M citrate buffer solution of (pH 4) on ice proved to be successful in
keeping this solution stable for a short period of time before injection (Rerup, 1970).
Alloxan is freely soluble in water and its stability in aqueous form is, once again,
dependent on pH and temperature. Below pH 3 alloxan is fairly stable at room
temperature, whereas at pH 7 the solution has to be kept below 4°C to avoid rapid
decomposition.
3. STZ and Alloxan dose
A diabetic dosage is defined as the amount of an inducing agent which, in 80% of
animals of a given species, produces sustained hyperglycaemia and beta cell necrosis, but
does not cause harm to other organs (Goldner and Gomori, 1944). In rats, a single i.v.
injection of 50mg/kg STZ and 30-70mg/kg alloxan were reported to yield 100% diabetes
(Rerup, 1970). Rerup (1970) reported that if non-i.v. injection was chosen as the route of
administration for these agents, a different dose may be required to achieve the 80% or
more success rate. Hence, 50 mg/kg was originally chosen as a minimum starting dose
for STZ and 30 mg/kg for alloxan since a different route of administration was used.
However, results showed that under our experimental conditions, the diabetogenic dose
for STZ was 75mg/kg (i.p.) and alloxan was 40mg/kg (i.v.).
136
4. Weight
Using 75mg/kg STZ, only 20% of rats became diabetic if they weighed more than 150g
and 100% of rats died within 48hrs after the injection due to severe diabetes if they
weighed under 130g. However, 80-90% of rats in the range of 130g-150g became
severely diabetic with the above dose and no fatality was observed. Similar results were
observed when alloxan, at 40mg/kg, was used.
5. 5% sucrose solution
Sucrose solution was used to overcome the severe hypoglycaemic condition of STZ- /
alloxan-treated rats (Rerup, 1970). Our own original experiments showed that by
providing treated rats with 5% sucrose solution for the first 48hr after injection, compared
to 24hrs, the animals’ survival rate could be improved by 30% or more during the crucial
period after diabetic onset (i.e. the first 10 days).
6. Insulin treatment for survival
In general, insulin treatment was not used to control the diabetic condition, except were
indicated (chapters 9 and 10), but to give regular respite from the catabolic dominance of
the condition while allowing regular and severe hyperglycaemia (Willars et al., 1989).
One or two injections of insulin (Protophane 2IU/100g) were given subcutaneously (s.c.)
to the newly diabetic rats depending on the severity of their physical condition (e.g. little
eating or drinking, weight loss to around 115g, very inactive). However, if they did not
show any improvement or lost more than 15% of their body weight they were sacrificed.
2.4.6.1. Establishing a time frame
The original experiments were purely performed to establish a diabetic rat model with an
appropriate time frame to examine the gradual deleterious effects of diabetes on skin
peripheral microvasculature over time using established techniques. These results were
then presented to the animal’s ethics committee to justify the purpose of the thesis and
proposed experiments. Experimental procedures were also discussed and approval was
obtained for the use of these animals in this study.
137
Changes in microvascular blood flow in STZ-treated rats were observed during the first
10 weeks after diabetes onset using two protocols: 1) the blister model of inflammation
and 2) electrical stimulation of the sciatic nerve (sections 2.6 & 2.7). For each week of
diabetes, a minimum of 12 STZ-treated rats was used (n = 6).
2.5.
Insulin treatment protocol
Insulin treatment was used under two conditions: 1) as an infrequent injection to relieve
the severe life-threatening symptoms of diabetes in experimental rats and extending their
survival rate, and 2) as regular injections for the purpose of controlling diabetes in rats.
The former insulin treatment was available for all diabetic rats used in the thesis, except
those that were used in chapter 10 (group 3 only) and chapter 9 experiments that received
the latter insulin treatment. The dose of injection varied from 2-6IU/100g/day depending
on the daily level of urine glucose, weight loss, and the observed physical condition of
animals (level of activity, appearance, eating / drinking). Urine glucose was tested every
morning, between 9am and midday, using specific test strips for urine glucose assay
(Diastix, Bayer Australia Ltd). For measuring plasma glucose, a drop of blood was taken
from the rat’s artery or vein and the glucose level was measured by Reflolux S
(Boehringer Mannheim Australia Pty Ltd).
The type of insulin was protophane, an intermediate acting insulin with a peak effect of
4-12 hrs, and this was administered by a NovoPen. Insulin was injected (s.c.) on the back
of the rat’s neck, where the skin is loose, every morning between 9am and midday.
2.6.
Blister model of neurogenic inflammation
Neurogenic inflammatory responses can be induced experimentally in a number of ways;
one is by exogenously administering sensory neuropeptides (chapter 1, section 1.7.3). For
this type of experiment, rats were first anaesthetized with pentobarbitone sodium
(65mg/kg, i.p.). Then a vacuum pressure of negative 40 kPa was applied for
approximately 15-30 minutes to the glabrous skin of the hind footpad, using a metal
suction cap heated to 40°C by an attached heating element. The smaller diabetic rats
developed a blister on their footpad faster (15-20 min) than control animals (30 min).
When a blister was established, the surface epidermis was removed to allow direct access
to the blood vessels and nerve endings. Then the foot was fixed in a Perspex chamber
138
with inlet and outlet ports to allow constant perfusion of substances over the blister base.
Perfusion was maintained at 4ml/hr by a peristaltic pump (Microperpex, LKB, Sweden)
and body temperature was maintained at approximately 37°C through out the experiment
(Fig 2-1).
The blood flow in the foot was monitored throughout the experiment via a laser Doppler
flowmetry probe placed in a third port directly above the centre of the blister base. In
every experiment, first a stable base line blood flow was achieved after 20-30 minutes
perfusion of Ringer’s solution over the blister base. In perfusion experiments, SP (1μM)
was then perfused over the blister base for 30 minutes, during which time the perfusate
(plasma extravasation) was collected every 10 minutes (prior, during, and post
stimulation). The concentration of SP used in these experiments was previously
established in our lab as the submaximal concentration that induces a significant plasma
extravasation response (Andrews et al., 1989). The protein content of the perfusate was
measured using Bradford colorimetric assay (Bradford, 1976). After SP perfusion, a 20
minutes washout with Ringer’s solution was allowed to establish the baseline for the
second time. Changes in the smooth muscle-dependent vasodilation were then examined
using SNP (100μM, 10 min), the direct smooth muscle vasodilator. After that, once again
Ringer’s solution was perfused for 10 minutes or until the base line was re-established for
the third time. If applicable, then CGRP (1μM, 10 min) was perfused until the maximum
height of the response was achieved. CGRP perfusion was last in all experiments since its
vasodilator action in the skin, unlike SP, does not undergo desensitization and can
continue for a few hours (Brain, 1996). Depending on the type of laser Doppler
flowmeter used, the results were calculated differently and this is described in section 2.9.
SNP, SP, and CGRP were perfused over the blister base in most experiments throughout
this thesis. The independent smooth muscle vasodilator, SNP, was used to examine
independent changes in smooth muscle reactivity. A sensory neuropeptide, SP, was used
to examine changes in sensory mediated events at both pre- and post-terminal levels.
Another sensory neuropeptide, CGRP, was used to examine changes in sensory mediated
139
events (at the post-terminal level) with mechanisms different to those involved in SP and
SNP responses.
2.7.
Electrical stimulation of sciatic nerve
Among many nerve preparations currently in use, the sciatic and saphenous nerves are
the most common as they are the major nerves of the hind limbs and are easily accessible
(Baker et al., 1979-80). To prepare the sciatic nerve for electrical stimulation (ES), once
again rats were anaesthetised and a blister was formed in their hind footpad as described
in section 2.6. Then the rat’s right leg was shaved at the mid-thigh region and under
anaesthesia and aseptic condition, a cut was made to expose the sciatic nerve. Using blunt
dissection through biceps femoris, the nerve was then separated from the surrounding
tissue and cut as proximally as possible. The distal portion of the cut nerve was placed
over bipolar platinum electrodes and immersed in liquid paraffin warmed to 37°C and
contained in a “bath” that was formed by raising the skin flaps of the wound (Fig 2-1).
The electrodes were fixed in such a position that electrical leakage to adjacent nerve and
tissue structures was minimised.
Grass Electrical stimulator
Fig 2-1: A schematic representation
of an experimental set up for
electrical stimulation of the
sciatic nerve.
Laser Probe
Laser Doppler Flowmeter
Perspex Chamber
140
For examining the vascular response to electrical nerve stimulation, the rat’s sciatic nerve
was stimulated (Grass S48 stimulator, Grass medical instruments, Quincy, U.S.A.) every
20 minutes (20 V, 5 Hz, 2msec for the duration of 1 minute). These parameters have been
previously used to stimulate efferent C-fibre responses (Helme et al., 1985; Lembeck and
Holzer, 1979), to evoke an immediate increase in local blood flow. Ringer’s solution was
perfused during and for at least 20 minutes after stimulation of the nerve at the speed of
4ml/hr. The time elapsed from the start of the blister induction until sciatic nerve
stimulation did not exceed 60 minutes. Again, SNP (same concentration as before) was
used to measure the smooth muscle-dependent vasodilatation in these rats. This was
performed before stimulating the sciatic nerve to devoid our results from the sympathetic
nerve contribution.
2.8.
Neonatal capsaicin pre-treatment
Capsaicin was prepared in normal saline containing 10% Tween 80 and 10% ethanol.
Neonatal rats of 4 days old were then injected (50mg/kg, s.c.) with a single dose of
capsaicin on the back of the neck (Fitzgerald, 1983; Jancso et al., 1977). The
effectiveness of the neonatal pre-treatment (up to 90% degeneration of unmyelinated
primary afferents and destruction of small DRG) was assessed around 2-3 months of age.
Prior to each experiment, a drop of capsaicin solution (0.1% capsaicin in ethanol) was
applied to the cornea of treated animals (Fitzgerald, 1983); if the number of eye wipes
were less than 25% of controls, rats were considered capsaicin denervated. Those that
were not sensory denervated were euthanized (by anaesthetic overdose).
Capsaicin pre-treated rats (130-150g) were then made diabetic by STZ injection using the
same protocol as described before. Two and four weeks later, they were anaesthetized
and a blister was induced on the hind footpad. Once again the foot was secured in a
Perspex chamber and the blood flow was monitored using the Perimed or the Moor laser
Doppler flowmeters (PLDF or MLDF). Since the MLDF has dual ports, the blood flow
was monitored in both feet simultaneously. SP and SNP (same concentrations as before)
were perfused over their blister bases as described before (section 2.6).
141
2.9.
Blood flow measurements using laser Doppler flowmeter (LDF)
Laser Doppler flowmetry is a non-invasive technique that is simple to perform yet
reproducible and reliable. This is suitable for long-term monitoring of the circulation and
examination of the effects of short-term alterations to the circulation e.g. caused by
chemical sensory nerve stimulation. LDFs were developed in the early 1970’s (Halloway
and Watkins, 1977; Stern, 1975) but Gamse and Saria (1987) were first to adopt this
technique as a method for measuring antidromic vasodilation in rat skin.
A laser Doppler flowmeter has a solid-state laser diode as a light source, which is
delivered by a fibre optic line to the probe and tissue. In the skin, light from the probe is
absorbed, scattered or reflected (5-7%) because the epidermal layers act as absorption
filters limiting the depth of the light’s penetration (Anderson and Parish, 1981). Since
short light wavelengths are absorbed and scattered more than longer ones, red and far-red
lasers with wavelengths from 600-1200 nm have been preferred in most LDF
applications. This ensures the light’s maximum depth of penetration into the dermal
layers of the skin (Gush and King, 1991). A portion of backscattered light is then
collected by the LDF where it is analyzed.
The principle of laser Doppler flowmetry is based on the fact that moving red blood cells
induce a Doppler (Frequency) shift of the wavelength. Therefore, the scattered light from
a moving element will have a different frequency to the original light. The output signal
from the laser Doppler is directly proportional to the perfusion of blood cell flux (Oberg,
1990), which is defined as
The number of moving cells x mean velocity = blood flux
Two LDFs have been utilised in this study, the Perimed periflux and Moor DRT4.
2.9.1. Perimed LDF (PLDF)
The Periflux model (PF2B, Perimed, Sweden) uses a laser beam of 633nm (Visible red
light) which penetrates approximately 1mm deep into the skin where superficial
capillaries (around 10μm in diameter) are located. Capillaries are arising from the
papillae of the corneum and returning to subpapillary venous plexus. The blood flux
(Volts) is electronically processed and is plotted directly onto a chart recorder where each
142
unit (2.5 cm in height) represents a magnitude of 1 volt of blood flux. Then the surface
area under the response curve (cm2) is measured using a digital planimeter (section
2.9.1.1.). The area under the response curve is the most accurate measure that represents
the total change in blood flux incorporating not only the height but also the duration of
the response curve.
2.9.1.1. Digital Planimeter
Digital planimeter (Planix, Tamaya & company, Japan) is an instrument that is used to
measure the area under a response curve (changes in blood flux) recorded by the PLDF
on a chart recorder. Each area was measured three times (Area ± 0.2 cm2), and the
average area was calculated.
2.9.2. Moor LDF (MLDF)
The DRT4 model (Moor Instruments Limited, U.K) uses an advanced microprocessorcontrolled Doppler monitor and it offers power and flexibility to investigate
microvascular perfusion. Similar to the Perimed, the Moor also contains solid-state laser
diodes but it operates mainly at 780-820 nm wavelength. These long-range wavelengths
can penetrate 2-3 mm deep into the skin where arterioles, venules, and arteriovenous
anastomoses (AVAs) (around 100μm in diameter), are present. AVAs arise from
arterioles and because of their smaller diameter, the blood flow per unit length is far
greater than in capillaries. The MLDF software calculates perfusion data relating to the
mean red blood cell flux, red blood cell concentration, and mean red blood cell speed and
converts them into percent above basal using the following formula:
(Value-PRL) / (PRL-PZL) x 100
PRL= Percent reference level
PZL= percent zero level
These parameters can be displayed graphically or digitally on a bright LCD display.
Unlike the Perimed, data from the Moor is automatically stored for later analysis (locally
or by computer) and printout (Moor DRT4 User Manual, 1995). Simultaneous use of
both the PLDF and the MLDF, with probes of different wavelengths, could distinguish
blood flow differences between the superficial capillaries and the deeply located vessels
in the skin.
143
2.10.
Sympathetic nerve blockade
Guanethidine monosulfate, a sympathetic neurone blocker, prevents the release of
vasoconstrictors from sympathetic nerves. When a single high dose of guanethidine is
injected (i.p.), it is taken up from the synaptic cleft into sympathetic terminals by means
of the noradrenaline (NA) re-uptake mechanism. The normal vesicular content of the
sympathetic terminals is then depleted, flooding tissues with NA. This NA release leads
to the acute sympathomimetic effect of guanethidine. After uptake, guanethidine replaces
NA in the sympathetic terminals and acts as a “false transmitter”, thus causing short-term
functional nerve blockade. This treatment does not cause any structural damage to the
sympathetic fibres or terminals or to the sensory fibres (Shir and Seltzer, 1991).
Sympathetically released neurotransmitters can modulate sensory nerve function by
inhibiting peptide release when sensory nerves are stimulated. In this study, guanethidine
monosulfate powder was dissolved in saline (50mg/kg) and a single dose was injected
(i.p.) into either control or diabetic (4 weeks) rats one hour before sensory nerve
stimulation. Administration of guanethidine (50 mg/kg) in this manner (i.p.) in adult rats
is known to cause temporary (up to 2 weeks) functional sympathetic nerve blockade (Shir
and Seltzer, 1991).
2.11.
2.11.1.
Wound healing
Induction of thermal burn
The rat fur on the interscapular region was removed with animal clippers and a cosmetic
depilatory agent. Rats were anaesthetized with sodium pentobarbitone (65mg/ml, i.p.),
24-48 hrs after this treatment, and a thermal burn was induced in the treated region by
using a CO2 laser (one stimulation of 25 watts power, 2 seconds duration, and a spot
diameter of 5 mm). This technique delivers standard energy levels over a given area of
skin and therefore gives a similar and reproducible thermal burn. This injury results in a
large circular wound area (4 cm2). The wound area increases to 4.5-5 cm2 by day 2 due to
progressive loss of microcirculation as thermal injury damages capillary endothelial cells.
This shape and size of wound allows for healing primarily by contraction and reepithelialization over approximately 2.5-3 weeks in diabetic rats. It is also known that
144
round wounds do tend to contract more uniformly, rarely deform into two or more
wounds, and are associated with less measurement error (Richey et al., 1989).
Thermal burn was induced in control and diabetic rats (day 28) at the same time. After
induction of wounds, rats were kept in groups of 2 per cage. General health of the
animals and condition of their wounds were monitored carefully, especially diabetic rats.
The cages were cleaned every day for the animal’s comfort and to prevent infection.
2.11.2.
Drug treatments
Among diabetic rats involved in wound healing experiments, one group was not treated
but the rest were treated with:
1. Insulin (2-6IU/100g), daily, s.c. - Commenced on day 3 after diabetes onset and its
confirmation (n = 5-6)
2. Aminoguanidine (AG) (25mg/kg), daily, i.p. - Commenced on day 3 after diabetes
onset and its confirmation (n = 5-6)
3. Tirilazad mesylate (Freedox) (5mg/kg), every 2 days, i.p. - Commenced on day 3 after
diabetes onset and its confirmation (n = 5-6)
4. Combined Superoxide Dismutase-Polyethylene Glycol (PEG-SOD) (100U/ml) + BQ123 (100μM) (150μl, first 5 days after induction of thermal wound, i.d.) -Injection
commenced on day 28 after diabetes onset and its confirmation (n = 5-6)
PEG-SOD (50μl) + BQ-123 (100μl) = 150μl
This solution was injected at 2 sites, 1cm distance to the burn edge daily and at the same
time of the day. Five consecutive days of treatment was chosen based on our own
previous study on wound healing showing the most significant change in wound size
occurs within the first week of its induction (Khalil and Helme, 1996).
2.11.3.
Measurement of healing
The area of burn was traced daily and measured with a digital planimeter. An observer
who was unaware of the treatment protocol measured the wounds. Burns were left
uncovered and observed daily for any sign of infection. Although scab formation is
minimal in full-thickness wounds, when it occurred, the scab was gently removed. The
145
purpose of this action was to keep all wounds comparable, as a previous study by
Snowden et al (1982) showed that scab formation induces a transient decrease in the rate
of wound contraction. It was also easier to accurately measure the open wound area
without the overlying scab. The healing endpoint was determined, as the time when full
wound contraction had occurred. It should be noted that although wound contraction is
only one parameter of wound healing, it accounts for large portions of wound closure in
full-thickness wounds.
2.12.
Protein assay technique
Bradford protein assay technique is a simple colorimetric assay that measures the protein
content of the perfusate, which is a measure of plasma extravasation (Bradford, 1976).
First, a standard curve was established using a series of bovine serum albumin (BSA)
concentrations (e.g. 0-400μg/ml). Then, 100μl of the perfusate was placed in a tube and
5ml of coomassie brilliant blue dye was added and left for 10 minutes. The colour of the
sample was measured at 590 nm with a spectrophotometer (Beckmen DU 530, USA) and
readings were recorded. Using the standard curve, colorimetric readings were converted
to concentrations of protein present in the samples. The concentration of every collection
(prior, during, and after stimulation) was recorded separately for individual animals in a
group then the mean values were calculated.
2.13. Statistical methods of analyses
Results from vasodilatation responses were either calculated as areas under the response
curves or percent above basal or percent of control. Plasma extravasation (PE) responses
are presented as percent above basal. All data are expressed as mean ± standard error of
the mean (SEM). The absolute values and 95% confidence intervals (95%CI) of those
values are quoted in the text. Presentation of 95% confidence limits would allow the
reader to assess both distribution and significance of results.
The statistical analyses employed were chosen according to the type of experiment and
the properties of its dependent and independent variables. The statistical analyses
described in this thesis were performed utilising SPSS for Windows Statistical package
version 9.0.
146
Analysis of Variances (ANOVA) (one-way or two-way) were used to examine whether
there were any group differences or interactions between control and diabetic rats (with
and without treatments). If ANOVA indicated a significant main effect, then post-hoc
comparisons using one or more of the following tests (t-test, Dunnet’s test, Tukey HSD
test) was performed. One-way repeated measure ANOVAs were performed to identify
whether there were significant group differences (control and diabetic) according to the
PLDF and the MLDF, in responses to SP at different time points of perfusion (SP10,
SP20, SP30). Again, where a significant main effect was detected, post hoc independent
sample t-tests were conducted. A repeated measure ANOVA (one-way or two-way) with
priori planned comparisons was performed to investigate whether there were significant
differences in PEs between control and diabetic groups (with and without treatments) at
different time points of collection. Wound healing data were analysed using a
combination of one-way and repeated measure ANOVAs to identify group or treatment
differences followed by either priori-planned or post hoc comparisons. In all cases a
value of p < 0.05 was set to denote statistical significance.
2.14. Limitations of this study
1. It was not possible to examine the longer-term effect of diabetes (over 8 weeks)
without insulin treatment due to ethical reasons.
2. The role of food ingredients such as vitamins in protecting pancreatic beta cells
against diabetogenic agents proved interesting, based on the experience during the
development of the STZ-model, and should be further investigated. In this study, the
role of food was not part of the original plan and therefore time was not allocated to
pursue this line of investigation.
3. Reflolux S, a blood glucose-monitoring device, used by human patients can
accurately measure glucose levels up to 27.7mmol/lit. This posed a limitation in
accurately measuring blood glucose above this level in diabetic rats. In turn, this
limitation affected the dose of insulin used, especially in these experimental groups
where the purpose of insulin treatment was to control the blood glucose level in
diabetic rats (chapter 9 and chapter 10).
147
4. It would have been a great advantage if some of these experiments could have been
performed on a type II model of diabetes. It was considered that by developing a type
II model of diabetes, using STZ in Sprague-Dawley rats, a direct comparison of
results would have been possible. Several attempts were made using an STZ injection
in neonatal rats but this method proved to be very difficult and time consuming. Since
the aim of the thesis was to examine changes in microvascular blood flow in a type I
diabetic rat model, and the fact that time and resource were not originally allocated to
this experiment, further attempts to develop a type II model of diabetes were not
made.
148
CHAPTER 3
Diabetes and sensory nerve function
149
3.1.
Neurogenic inflammation: Part 1 (Post-terminal events)
3.1.1. Introduction
Inflammation is a protective response that is essential for tissue repair. The neurogenic
component of the inflammatory response is mediated via a subset of sensory nerves,
namely unmyelinated primary afferent sensory neurones. Upon stimulation, sensory
peptides such as substance P (SP) and calcitonin gene-related peptide (CGRP) are
released from terminal endings of primary afferent fibres causing dilatation of adjacent
microvessels as well as increasing the permeability of post-capillary venules (plasma
extravasation) (Holzer, 1998).
There are numerous reports showing a significant reduction in the inflammatory
responses in diabetes (Bennett et al., 1998; Gamse and Jancso, 1985; Mathison and
Davison, 1993; Nemeth et al., 1999b; Walmsley and Wiles, 1991). Evidence suggests
that the endothelium is damaged (Huszka et al., 1997) and nitric oxide (NO) production is
reduced in streptozotocin (STZ)-treated diabetic rats at early stages of the disease (Kiff et
al., 1991a; Stevens, 1995). Also, a reduced amount of SP and CGRP was observed in the
axons and cell bodies of the primary afferent nerves in diabetic rats (Karanth et al., 1990;
Rittenhouse et al., 1996a). It was suggested that the function of unmyelinated sensory Cfibres is impaired in diabetes mellitus (Gyorfi et al., 1996), altering the production and /
or release of inflammatory mediators (Nemeth et al., 1999a; Rittenhouse et al., 1996a).
This in turn can lead to inadequate tissue blood supply in this disease condition (Rendell
et al., 1993).
Endothelial cells can also alter the contractile magnitude of the vascular smooth muscle
cells by releasing a number of constricting factors (Bennett and Gardiner, 1996) (chapter
1, sections 1.5.3 & 1.8.4). It has been suggested that in the diabetic condition some of
these factors may play a role in deactivating NO (Gryglewski et al., 1986). It is also
known that intracellular calcium overload can occur in early diabetes. This may be due to
the reduction in the activity of associated enzymes and abnormalities in sarcolemmal
membrane Na+/Ca2+-exchange and calcium pump (Golfman et al., 1998; Ver et al., 1997)
(chapter 1, section 1.8.5.3). An increase in the activity of voltage-dependent calcium
150
channels has also been suggested to play a role in regulating intracellular calcium in
diabetes (Kamata et al., 1992). The role of endothelins (ETs) and free radicals in
intracellular calcium homeostasis in early diabetes could not be ignored (chapter 1,
sections 1.8.4 & 1.5.3). Overall these studies suggest that damage to endothelial cells in
diabetic condition may alter skin microvascular blood flow by affecting the release of
vasoactive peptides (e.g. NO) and altering the contractile magnitude of vascular smooth
muscles.
In this study, the blister model of neurogenic inflammation using the rat hind footpad was
used. This model was developed in our laboratory (Helme and White, 1983; Helme et al.,
1985) to investigate the complex interactions between mediators of neurogenic
inflammation (Khalil et al., 1988; Khalil and Helme, 1989; Khalil and Helme, 1990b;
Khalil and Helme, 1991). By forming a blister on the rat hind footpad, separating the
epidermis from the dermis and then removing the former, easy access to the extracellular
space with intact skin blood vessels can be achieved.
Sensory neurotransmitters (e.g. SP, CGRP), when applied over the blister base, can cause
inflammatory responses such as vasodilatation and / or plasma extravasation (postterminal events) via mechanisms that are mainly post-terminal in nature.
Two different laser Doppler flowmeters (LDFs), with different laser light penetration
capabilities, were used to examine changes in microvascular blood flow in STZ-diabetic
rats at different skin levels (superficial and deep). The Perimed LDF (PLDF) and the
Moor LDF (MLDF) were used respectively to record blood flow at superficial and the
combined superficial and deep compartments of the skin (see also chapter 2, section 2.9).
The simultaneous use of these two LDFs could distinguish differences in blood flow
between the superficial capillaries and the deeply located vessels in the skin (Gush and
King, 1991), thus distinguishing which type of blood vessel is more susceptible to the
effects of early diabetes. Results from these experiments could provide valuable
information regarding early mechanisms that might eventually contribute to the delay in
repair of skin wounds in diabetic patients.
151
The relationship between the duration of diabetes and changes in skin microvascular
blood flow in STZ-treated rats has not been previously studied. Also, the possible
underlying mechanisms for the above changes in diabetic rat skin are not clear.
Therefore, the aims of chapter 3, part 1 are:
1. To examine the changes in the inflammatory response in diabetic rats over time (i.e.
to establish a time frame for the development of microvascular changes).
2. To examine independent changes in smooth muscle reactivity overtime using the
independent smooth muscle vasodilator sodium nitroprusside (SNP).
3. To examine the function of sensory nerves in 4 weeks diabetic rats (as determined
from aim 1 to be the most suitable) in relation to changes at the post-terminal level
using local perfusion of SP and CGRP over the blister base. It was hypothesized that
there may be an alteration in sensory mediated events at the post-terminal level.
4. To examine sensory independent effects of diabetes on the microvasculature using
sensory denervated (capsaicin pre-treated) rats (with or without diabetes). It was
hypothesized that there could be alterations in microvascular function that may be
independent of sensory nerve related changes.
5. To examine the effect of diabetes (4 weeks) on the microvasculature at both
superficial and deep compartments of the rat skin. It was hypothesized that there
could be an alteration in the function of capillaries in early diabetes.
Recent studies suggested that STZ toxicity was not only responsible for pancreatic cell
death and development of diabetes but might also cause neuronal cell damage (Adeghate
and Parvez, 2000; Prickaerts et al., 2000). This raised the possibility that changes in
microvascular blood flow observed in STZ-treated diabetic rats might not be related to
hyperglycaemia but to the neurotoxic effect of STZ. Therefore, changes in microvascular
blood flow in 4 weeks alloxan-treated rat was also examined. It was hypothesized that in
STZ-treated rats, the changes in microvasculature might be caused by a possible
independent neurotoxic effect of STZ other than its hyperglycaemic effect. The possible
neurotoxic effect of STZ was also previously addressed by an insulin treatment study
(Schmidt et al 1983). In chapter 9 this issue was further examined by recording changes
in microvascular blood flow in insulin treated diabetic (STZ treated) rats.
152
3.1.2. Method
Twenty-two groups of rats were used, 4 control, 10 STZ-treated, 2 alloxan-treated, and 6
capsaicin pre-treated diabetic (using STZ) groups. Each group consisted of 6-9 rats (n =
6-9). Control groups were injected i.p. with sodium citrate buffer solution (0.1M, pH 4)
(CB-treated) (chapter 2, section 2.4.1). Other rats were made diabetic with either STZ or
alloxan treatment (chapter 2, sections 2.4.3 & 2.4.4)
For the first set of experiments, ten STZ-treated diabetic groups of rats were tested after 1
to 10 weeks of diabetes (1 group every week) together with controls. In experiment 1
(peptide perfusion), first SP (1μM, 30 min), then SNP (100μM, 10 min), a direct smooth
muscle vasodilator, and finally CGRP (1μM, 10 min) were perfused over the blister base.
The changes in blood flow after SP (30 min), SNP (20 min), and CGRP (20 min)
perfusions were recorded by the PLDF and the areas under the response curves were
measured using a digital planimeter.
Animals
Control
(CB-treated)
Diabetic
(STZ)
Weeks (0, 1, 5, 10)
1-10 weeks
Peptide
perfusion
Peptide
perfusion
Data collected from the MLDF were displayed as percent above basal (see also chapter 2,
section 2.9.2). The results obtained from the PLDF indicated that in STZ diabetic rats,
vascular changes (at superficial skin layer) tend to occur during the first 4 weeks of
diabetes. Hence, the next set of experiments focused on the microvascular changes at 2
and 4 weeks of diabetes using the combined sensory denervated and STZ-treated diabetic
rats. Also, the changes at the deeper level of the microvasculature were examined at the
above time points post-diabetes. For this purpose, six separate groups of capsaicin pretreated rats were used as shown below (see also chapter 2, section 2.8). Four groups were
153
injected with STZ and the other two with 0.1 M citrate buffer solution (CB-treated). Two
or four weeks later, capsaicin pre-treated diabetic and CB-treated groups were
anaesthetized and a blister was induced on the hind footpad. Once again, blood flow was
monitored using both the PLDF and the MLDF. The vascular responses to SP and SNP
(same concentrations and time mentioned before) were recorded using both the PLDF
and the MLDF.
Animals
Capsaicin-treated
Diabetic (STZ)
CB-treated
2 weeks
PLDF
MLDF
PLDF
Peptide
perfusion
4 weeks
MLDF PLDF
Peptide
perfusion
MLDF
Peptide
perfusion
In alloxan-treated rats, after the base line was established, SP, SNP, and CGRP were
perfused over the blister base (same concentrations and time mentioned above) and
perfusion data were collected using the PLDF. Plasma extravasation (PE) was collected
from all peptide perfusion experiments. At the end of each experiment, rats were
sacrificed with pentobarbitone overdose.
Animals
CB-treated
Diabetic
(Alloxan)
4 weeks
Peptide
perfusion
Peptide
perfusion
154
3.1.2.1.
Expression of data and statistical analyses
Vasodilatation responses were calculated as areas under the response curves, percent
above basal, and percent of control. PE responses are presented as percent above basal.
All data are expressed as mean ± standard error of the mean (SEM). The absolute values
and 95% confidence intervals (95%CI) of these values are quoted in the text.
Separate one-way Analyses of Variances (ANOVAs) were performed to investigate
whether there were significant differences in basal blood flow (BBF), weight gain, PE
results and responses to SNP, SP, and CGRP, between control and diabetic rats (1-10
weeks). Where ANOVA indicated a significant main effect, post hoc comparisons were
conducted using Dunnett’s test. Separate one-way repeated measure ANOVAs were
performed to investigate whether there were significant differences in vascular responses
to SP between control and diabetic rats, recorded by PLDF and MLDF, at different times
of perfusion (SP10, SP20, SP30). Where a significant main effect was detected, post hoc
independent sample t-tests were performed. A two-way ANOVA was performed to
investigate whether there were significant differences in BBF and responses to SNP, SP,
and ES, between control and capsaicin pre-treated animals (with or without diabetes).
This test should also show interaction effects between these groups. Where ANOVA
indicated a significant main effect, post hoc comparisons were conducted using
independent sample t-tests. To investigate whether there were significant differences in
PE responses to SP (SP2,
3 & 4)
between control and diabetic rats over time, a one-way
repeated measure ANOVA was performed with priori planned comparisons. Separate
independent sample t-tests were performed to examine other results involving control and
diabetic rats.
3.1.3. Results
Responses in CB-treated control rats were not expected to change with time. Hence,
random CB-treated control groups for weeks 0, 1, 5, and 10 were tested. Results showed
BBF and other vascular responses (SNP, SP, CGRP, and ES) from CB-treated control
rats at weeks 1, 5, and 10 were similar to those at week zero (Appendix A, Table A-1).
Therefore, all control (CB-treated) values presented in this chapter and throughout the
155
thesis are those of week zero. The data representing controls were used througout the
thesis to allow comparison with diabetic data.
The BBF (measured for 20 min) in diabetic rats was generally lower than control
animals. However, this reached a significant level after 4 weeks of diabetes [(16.7 ± 1.3
cm2) (95%CI 14.1-19.2)] when compared to control [(25.1 ± 0.9 cm2) (95%CI 23.3-
Area under the response curve (cm2 )
26.9)], and remained approximately stable at this level for a further 6 weeks (Fig 3-1).
60
Control
Diabetic (weeks 1-10)
50
40
30
20
*
*
*
*
*
*
*
4
5
6
7
8
9
10
10
0
0
1
2
3
Fig 3-1: BBF in rats before and after induction of diabetes (changes over time), presented as mean ±
SEM (n = 6-9). * Denotes significance from control rats (P<0.05)
Responses to SNP, SP, and CGRP were initially calculated as percentage increase above
own basal blood flows as shown in Fig 3-2.
156
Control
Di abetic (4 weeks)
(%) Percent above basal
300
250
200
*
150
*
*
100
50
0
SNP
SP
CGRP
Fig 3-2: Vascular responses to SNP (100 μM), SP, and CGRP (both at 1μM), presented as percent above
basal ± SEM (n = 6-9). * Denotes significance from control rats (P<0.05)
Vasodilatation responses to SNP, SP and CGRP in 4 weeks diabetic rats (filled bars)
reached 99 ± 17.6 %, 93 ± 15.9 % and 129 ± 17.5 % above basal blood flow respectively.
These were significantly different to those of control animals (unfilled bars) where the
responses to SNP, SP, and CGRP reached 256 ± 18.9%, 157 ± 10.2 % and 214 ± 37.4%
above basal blood flow respectively.
Vasodilatation responses were also measured as areas above basal (areas under the
response curve) as shown in Fig 3-3. In 4 weeks diabetic rats, the response to SNP was
significantly reduced [(7.4 ± 0.6 cm2) (95%CI 6.2-8.6)], reaching only 28% of the control
value [(26.1 ± 1.8 cm2) (95%CI 22.6-29.6)]. Similarly, vascular responses to SP [(21.6 ±
3.4 cm2) (95%CI 14.9-28.3)] and CGRP [(10.4 ± 1.3 cm2) (95%CI 7.8-12.9)] were also
significantly reduced in these rats, reaching 42% of their own control groups [(51.4 ± 2.9
cm2) (95%CI 45.7-57.1) and (24.8 ± 4.9 cm2) (95%CI 15.2-34.4) respectively].
157
Area under the response curve (cm2 )
60
Control
Diabetic(4 weeks)
50
40
30
*
20
*
*
10
0
SNP
SP
CGRP
Fig 3-3: Vascular responses to SNP (100 μM), SP, and CGRP (both at1μM), presented as mean ± SEM
(n = 6-9). * Denotes significance from control rats (P<0.05)
Since both methods of measurement (percentage increase above basal blood flow and
area above basal) showed similar significant differences between diabetic and control
rats, all subsequent results for vasodilatation responses are presented as areas above
baseline.
As mentioned above, a general reduction was observed in SNP, SP and CGRP responses
in diabetic rats over time (Figs 3-4, 3-5, 3-6). This decline continued until week 4 where
SNP, SP and CGRP responses reached a minimum followed by a slight increase before
reaching a plateau. Hence, 4 weeks post-diabetes was considered as a suitable time frame
for this and subsequent chapters in this study.
158
Area under the response curve (cm2 )
60
Control SNP
Diabetic SNP (weeks 1-10)
50
40
30
20
*
*
*
*
10
*
*
*
*
8
9
10
*
*
0
0
1
2
3
4
5
6
7
Fig 3-4: Changes in the vascular response to SNP (100 μM) in diabetic rats over time, presented as mean
Area under the response curve (cm2 )
± SEM (n = 6-9) * Denotes significance from control rats (P<0.05)
60
Control SP
Diabetic SP (weeks 1-10)
50
40
*
30
*
*
*
*
*
*
*
*
20
10
0
0
1
2
3
4
5
6
7
8
9
10
Fig 3-5: Changes in the vascular response to SP (1μM) in diabetic rats over time, presented as mean ±
SEM (n = 6-9). * Denotes significance from control rats (P<0.05)
159
Area under the response curve (cm2 )
60
Control CGRP
Diabetic CGRP (weeks 1-10)
50
40
30
20
*
10
0
0
1
2
3
4
5
6
7
8
9
10
Fig 3-6: Changes in the vascular response to CGRP (1μM) in diabetic rats over time presented as mean
± SEM (n = 6-9). * Denotes significance from control rats (P<0.05)
The results also showed a significant difference in average weight gain between control
and diabetic rats over the 10 weeks period. Non-diabetic animals gained an average of
30.1g/week compared to only 8.8 g/week in diabetic rats. Table 3-1 and Figs 3-7 and 3-8
show changes in weight in control and diabetic rats during this period.
Animal
Average weight at week 0
Average weight at week 10
Control
147 ± 2g
448 ± 10.9g
Diabetic
147 ± 2g
235 ± 11.3g
Table 3-1: Average weight of control and diabetic rats at week zero and week ten.
160
500
*
*
450
*
400
Weight (g)
*
*
350
*
*
*
300
*
250
*
200
150
100
50
0
1
0
2
3
4
5
6
7
8
9
10
Weeks
Fig 3-7: Weight in non-diabetic rats over time, presented as mean ± SEM (n = 6-9)
* Denotes significance from week zero (P<0.05)
500
450
400
Weight (g)
350
300
250
*
*
*
7
8
9
*
200
150
100
50
0
0
1
2
3
4
5
6
10
W e e k s afte r ST Z inje c tio n
Fig 3-8: Weight in diabetic rats over time, presented as mean ± SEM (n = 6-9).
* Denotes significance from week zero (P<0.05)
PE was collected for control and 2, 4, 6, and 9 weeks diabetic rats. The PE response to SP
(SP2, 3 & 4) was calculated as percent above basal ± SEM. SEMs were omitted from Fig 39 for clarity but quoted in the text. Basal PE level (R1) was significantly increased in 4, 6,
and 9 weeks diabetic rats when compared to control group (Appendix B, Table B-1).
The PE response to SP, in control rats, was 36 ± 19%, 72 ± 18%, and 89 ± 14% above
basal respectively during the three 10 minutes perfusion time. This increase above basal
was significantly reduced in diabetic rats, reaching 25 ± 15%, 60 ± 23%, and 49 ± 18%
by week 2 and 9 ± 23%, 16 ± 32% and 0.1 ± 23% by week 4 (Fig 3-9).
161
(%) Percent above basal
120
Control
Diabetic (2W)
Diabetic (4W)
Diabetic (6W)
Diabetic (9W)
100
80
60
40
20
0
-20
R1-SP2
R1-SP3
R1-SP4
Fig 3-9: PE in diabetic rats over time, presented as percent above basal (n = 6-9).
Effect of diabetes on microvasculature at both superficial and deep microvascular
compartments
Vasodilatation responses to SNP and SP, using the MLDF, showed a significant
reduction in both 2 and 4 weeks diabetic rats. The areas under the response curves from
the MLDF were automatically calculated as percent above basal for each 10 minutes of
the response, using Moor software. It was simpler, therefore, to convert the PLDF results
into percent above basal blood flow for comparison purposes. For this reason, the total
area under the response curve for SP, measured by the PLDF (for 30 min), was divided
into three 10 minute periods. Then, each period was measured by digital planimeter and
results were calculated as percent above basal blood flow (Figs 3-10 & 3-11).
162
Control SP
Diabetic SP (2 weeks)
Diabetic SP (4 weeks)
(%) Percent above basal
300
250
200
*
150
100
*
*
50
*
*
0
SP10
SP20
SP30
Fig 3-10: Vascular responses to SP (1μM) in control and diabetic rats (2 and 4 weeks) at 10 minute
intervals measured by the PLDF, presented as percent above basal ± SEM (n = 6-9).
* Denotes significance from control rats (same perfusion time) (P<0.05)
(%) Percent above basal
300
Control SP
Di abeti c SP (2 w eeks)
Di abeti c SP (4 w eeks)
250
200
150
100
X
*
50
*
*
*
*
0
SP10
SP20
SP30
Fig 3-11: Vascular responses to SP (1μM) in control and diabetic rats (2 and 4 weeks) at 10 minute
intervals measured by the MLDF, presented as percent above basal ± SEM (n = 6-9).
* Denotes significance from control rats (same perfusion time) (P<0.05)
X Denotes significance from 2 weeks diabetic rats (same perfusion time)(P<0.05)
Using the MLDF and the PLDF together allowed comparison between microvascular
responses measured from all skin compartments (superficial and deep) and those
measured from superficial compartments. The total vascular response to SP in control
rats, calculated as a percent above basal, for the PLDF and the MLDF, was 413 ± 20%
and 179 ± 8% respectively. In diabetic rats, the total response to SP measured by the
163
PLDF and the MLDF was reduced to 223 ± 21% and 89.5 ± 2.9% after 2 weeks and 194
± 10% and 49 ± 3.2% after 4 weeks. The ratio of blood flow at the superficial
compartment, measured by the PLDF, to that of both superficial and deep compartments,
measured by the MLDF, was 2:1 in control animals (i.e. 413 / 179). After 2 weeks of
diabetes, this ratio increased to approximately 3:1 (i.e. 223 / 89.5) and by 4 weeks of
diabetes it reached 4:1 (i.e. 194 / 49).
Effect of diabetes on microvasculature in sensory denervated (capsaicin-treated)
rats measured by PLDF
BBFs were significantly and equivalently reduced in rats treated with either capsaicin or
STZ [(16.6 ± 1.6 cm2) (95%CI 13.5-19.7) and (16.7 ± 1.3 cm2) (95%CI 14.1-19.2)
respectively] reaching 66 ± 6.5% and 67 ± 5.4% of control value [(25.1 ± 0.9 cm2)
(95%CI 23.3-26.9)] respectively (Fig 3-12). When capsaicin rats were treated with STZ,
BBFs were further reduced [(9.6 ± 0.9cm2) (95%CI 7.8-11.4)] reaching only 38.2 ± 3.5%
of control value.
% Percent of control
120
Control BBF
Capsaicin BBF
Diabetic BBF
Diabetic + Capsaicin BBF
100
80
*
*
+
60
X
*
40
20
0
Fig 3-12: BBFs in control, diabetic (4 weeks), capsaicin pre-treated, and capsaicin pre-treated diabetic
rats, presented as percent of control ± SEM (n = 6).
* Denotes significance from control rats (P<0.05)
X Denotes significance from capsaicin pre-treated rats (P<0.05)
+ Denotes significance from diabetic rats (P<0.05)
164
The vascular responses to SNP were slightly but insignificantly reduced in rats treated
with capsaicin [(19.7 ± 1.9 cm2) (95%CI 16-23.4)] but were significantly reduced in
STZ-treated rats [(7.4 ± 0.6 cm2) (95%CI 6.2-8.6)]. These responses reached 75.2 ± 7.5%
and 29.3 ± 3 % of control value [(26.1 ± 1.8 cm2) (95%CI 22.6-29.6)] respectively (Fig
3-13).
Control SNP
Capsaici n SNP
Di abeti c SNP
Diabetic + Capsaicin SNP
(%) Percent of control
120
100
80
60
X
40
*
X
*
20
0
Fig 3-13: Vascular responses to SNP (100 μM) in control, diabetic (4 weeks), capsaicin pre-treated, and
capsaicin pre-treated diabetic rats, presented as percent of control ± SEM (n = 6).
* Denotes significance from control rats (P<0.05)
X Denotes significance from capsaicin pre-treated rats (P<0.05)
When capsaicin rats were treated with STZ, SNP responses were further reduced [(5.2 ±
0.6 cm2) (95%CI 4-6.4)] reaching 20 ± 2.5% of control. These SNP responses were
significantly lower than that observed in rats treated with capsaicin alone. There was no
significant difference between SNP responses in diabetic rats and capsaicin pre-treated
diabetic rats.
The vascular response to SP in capsaicin pre-treated rats was significantly reduced [(27.4
± 3.5 cm2) (95%CI 20.5-32.3)], reaching 53.8 ± 7.1% of control [(51.4 ± 2.9 cm2)
(95%CI 45.7-57.1)]. On the other hand, the response to SP in STZ-treated rats [(21.6 ±
3.4 cm2) (95%CI 14.9-28.3)] (42 ± 6.8% of control), was not different to that of capsaicin
rats. Treatment of capsaicin rats with STZ, however, showed a further significant
165
reduction in SP response [(9.1 ± 1 cm2) (95%CI 7.1-11.1)] reaching 17.6 ± 1.5 % of
control and this was also significantly different from all the other responses (Fig 3-14).
Control SP
Capsai ci n SP
Di abeti c SP
Di abeti c + Capsai ci n SP
(%) Percent of control
120
100
80
*
60
*
+
40
X
*
20
0
Fig 3-14: Vascular responses to SP (1μM) in control, diabetic(4 weeks), capsaicin pre-treated, and
capsaicin pre-treated diabetic rats, presented as percent of control ± SEM (n = 6).
* Denotes significance from control rats (P<0.05)
X Denotes significance from capsaicin pre-treated rats (P<0.05)
+ Denotes significance from diabetic rats (P<0.05)
Effect of diabetes on microvasculature in alloxan-treated rats
Rats injected with alloxan were similar in age to STZ-treated rats. Approximately 60% of
alloxan-treated rats showed a blood glucose level of 27.7mmol/lit compared to the blood
glucose of STZ-treated rats, where around 10% were equal to, and the rest were above
27.7 mmol/lit.
Although alloxan-treated rats (4 weeks) showed physical symptoms of diabetes, they
were heavier in weight and generally in better physical condition than STZ-treated rats
after 4 weeks (Table 3-2).
Treatment
Average weight at week 0
Average weight at week 4
Control
147 ± 2g
297 ± 4.8 g
STZ
147 ± 2g
160 ± 10.6g
Alloxan
147 ± 2g
211 ± 10.4g
Table 3-2: Average weight of STZ- and alloxan-treated rats at week zero and week four
166
Similar to STZ-treated rats, after 4 weeks, there was a significant reduction in BBF in
alloxan-treated rats [(18 ± 1.4 cm2) (95%CI 15.2-20.7)] (Fig 3-15).
Area under the response curve (cm2 )
60
Control BBF
STZ-treated BBF
Alloxan-treated BBF
50
40
30
*
*
20
10
0
Fig 3-15: BBFs in control, STZ-treated, and alloxan-treated diabetic rats, presented as mean ± SEM
(n = 6). * Denotes significance from control rats (P<0.05)
The reduction in vascular responses to SNP and SP in alloxan-treated rats was also
similar to those in STZ-treated animals. The SNP response was reduced to (9.4 ± 1.3
cm2) (95%CI 6.9-11.9) and SP to (24.7 ± 3.4 cm2) (95%CI 18-31.4), reaching 36% and
Area under the response curve (cm2 )
48% of own control response (Figs 3-16 & 3-17).
60
Control SNP
STZ-treated SNP
Alloxan-treated SNP
50
40
30
20
10
*
*
0
Fig 3-16: Vascular responses to SNP (100 μM) in control, STZ-treated, and alloxan-treated diabetic
rats, presented as mean ± SEM (n = 6). * Denotes significance from control rats (P<0.05)
167
Area under the response curve (cm2 )
60
Control SP
STZ-treated SP
Alloxan-treated SP
50
40
30
*
*
20
10
0
Fig 3-17: Vascular responses to SP (1μM) in control, STZ-treated, and alloxan-treated diabetic
rats, presented as mean ± SEM (n = 6). * Denotes significance from control rats (P<0.05)
The vascular response to CGRP was also diminished in alloxan-treated rats [(9.75 ± 1.8
cm2) (95%CI 6.2-13.3)], reaching 39% of control response. This reduction in CGRP
response was similar to the reduction in CGRP response in STZ-treated rats [(10.4 ± 1.3
cm2) (95%CI 7.8-12.9)] (42% of control) (Fig 3-18).
Area under the response curve (cm2 )
60
Control CGRP
STZ-treated CGRP
Alloxan-treated CGRP
50
40
30
20
*
*
10
0
Fig 3-18: Vascular responses to CGRP (1μM) in control, STZ-treated, and alloxan-treated diabetic
rats, presented as mean ± SEM (n = 6). * Denotes significance from control rats (P<0.05)
168
3.1.4. Discussion
Diabetes mellitus is a complex metabolic disorder affecting small nerve fibres and blood
vessels over time leading to severe complications (McMillan, 1997; Vranes et al., 1999;
Westerman et al., 1989). It was suggested that impaired sensory C-fibres in diabetes
could play an important role in inflammation and control of microvascular blood flow
(Gyorfi et al., 1996).
The results demonstrated a significant reduction in diabetic rats’ inflammatory
vasodilatation responses over time. These responses at the superficial skin layer reached a
minimum 4 weeks after induction of diabetes (Figs 3-4, 3-5 & 3-6). Since the changes in
microvascular responses in diabetic rats occurred within the first 4 weeks of the disease
development, this time frame was considered suitable for the following experiments in
this thesis.
Results showed a reduction in BBF and body weight in diabetic rats (Figs 3-1, 3-7 & 38). These data suggest that there is a significant reduction in the skin nutritional blood
flow in the planter surface of diabetic rat paw. Blood flow activity is influenced by many
factors such as local temperature, nervous impulses (e.g. axon reflexes involving sensory
nerves
and
sympathetic
vasoconstrictor
tone),
circulating
hormones
(e.g.
catecholamines), and physical stimuli. Previously it was reported that blood flow at the
planter surface of the rat paw did not change throughout the first 3 months of diabetes
(Rendell et al., 1993). Studies in diabetic patients, however, showed severe reduction in
the skin capillary blood flow, thus supporting the current results (Journeskog et al., 1995;
Yosipovitch et al., 1996). Journeskog et al. (1995), showed that in diabetic patients the
disturbances were mainly seen in the capillaries and the capillary blood flow (in the great
toe) was severely reduced during reactive hyperaemia. In contrast, total skin
microcirculation was normal, indicating that sufficient blood reaches the area but not
enough reaches the capillaries.
Superficial capillary blood flow, in this chapter, was examined using the PLDF with
short-wavelength emitting beams (633 nm) that can only penetrate to a depth of 1 mm.
The current vasodilatation results, using the PLDF, showed that microvascular function at
the superficial compartment of the skin is altered in the skin of diabetic rats, which is in
agreement with a human study by Yosipovich et al (1996) (Fig 3-10). In that study, an
169
abnormal blood flow was observed in the feet of young diabetic patients with and without
microangiopathy and regardless of the presence of peripheral neuropathy. It should be
noted, however, that the reduction in inflammatory vascular responses in this PhD study
was not simply due to the reduction in basal blood flow. This is based on the data
showing that measurement of responses as a percent above basal or areas above baseline
were similarly reduced in diabetic rats and capsaicin pre-treated rats, with and without
diabetes, when compared to control (Figs 3-2, 3-3 & 3-12).
Evidence is provided that the reduction in inflammatory responses after 4 weeks of
diabetes coincides with a 72% reduction in smooth muscle reactivity (Figs 3-4 & 3-5)
(Kamata et al., 1992). One possible reason for this reduction might be related to an
alteration in the intracellular calcium level in the smooth muscle. Na+/K+-ATPase and
Ca2+-stimulated ATPase dysfunction in early diabetes and alteration in Na+/Ca2+exchange and Ca2+-pump in the sarcoplasmic membrane (Golfman et al., 1998) may have
caused calcium overload and smooth muscle contraction. Experiments in chapter 6,
involving bisindolylmaleimide (BIM), examined possible alterations in the intracellular
calcium level in smooth muscle of early diabetic rats.
The reduction in vascular responses to SNP, a NO donor, suggests that NO has alower
efficiency to induce vasodilatation. This could be due to factors like deactivation of NO
(Gryglewski et al., 1986) or reduction in the efficiency of the second messenger system
(cyclic guanosine monophosphate (cGMP)) (Sasaki and Okabe, 1993; Weisbrod et al.,
1993), rather than factors that affect the release or production of NO. Other possibilities
for reduced responses to SNP could include the presence of ET and free radicals (Bassirat
and Khalil, 2000). The roles of these substances are discussed in chapters 6 and 7
respectively.
In rat skin, CGRP activates the cyclic adenosine monophosphate (cAMP) pathway to
induce relaxation of the smooth muscle. The results showed a significant reduction in
CGRP response in 4 weeks diabetic rats, reaching 42% of control response (Fig 3-6),
suggesting that the cAMP pathway might also be inefficient. However, since the
reduction in SNP response (28% of control) was greater than CGRP, it could be argued
that reduction in CGRP could be totally related to the changes in smooth muscle
reactivity. Furthermore, it is also possible that the cAMP system in diabetic rats functions
170
more efficiently to compensate for poor performance by the cGMP system (Chiappe de
Cingolani, 1986).
Vascular responses to SP were examined in diabetic (2 & 4 weeks) and control rats using
both LDFs. Blood flow changes in superficial capillaries in the skin (1 mm in depth) can
be measured by the PLDF. However, the MLDF with longer wavelength emitting beams
(780-820 nm) can penetrate deeper (2-3 mm) into the skin and measure blood flow
changes in both superficial and deeper vessels (arterioles, venules, and arteriovenous
anostomoses (AVAs)). AVAs are specialized structures that are controlled by
thermoregulatory reflexes. The reaction of AVAs to stimuli is different to capillaries;
they can permit total blood flow to the skin to increase without a change in the capillary
flow (Millington and Wilkinson, 1983).
Before analysis, the Perimed results were made comparable to those obtained from Moor
as described in the result section (Table 3-2). When the MLDF results were compared to
those obtained from the PLDF, a greater reduction in SP responses in diabetic rats was
observed, especially after 4 weeks (27% compared to 47% of own total control) (see also
Figs 3-10 & 3-11). The ratio of blood flow in superficial vessels to combined superficial
and deeper vessels was also increased with the duration of diabetes. This ratio increased
from 2:1 in control to 3:1 after 2 weeks and 4:1 after 4 weeks of diabetes. These results
suggest that, in early diabetes, both superficial and deeper vessels in the skin are altered
but deeper vessels might contribute more to the reduction in SP response. The current
results also showed that this reduction can be detected after only 1 week of diabetes,
which is in agreement with previous studies (Brands and Fitzgerald, 1998; Lash and
Bohlen, 1991; Zhu et al., 1999). The response to SP was subject to further reduction with
time (4 weeks).
PE results suggest alterations in the permeability of post-capillary venules in diabetic rats
(Fig 3-9). The protein content of R1 collected from diabetic rats showed an overall
increase with the duration of diabetes (up to 9 weeks) (Appendix B, Table B-1). This
value was increased by 63% after 2 weeks and by 135% after 4 weeks of diabetes. As
diabetic post-capillary venules are already leaky they respond poorly to SP.
Consequently, the ability of SP to induce sufficient PE responses is reduced (Mathison
and Davison, 1993). Post-capillary venules are an important part of the circulation and
171
are affected by inflammation. Elevated R1 in diabetic rats, especially after 4 weeks,
suggest that there are possible structural disturbances in endothelial cells. This is in
accord with a previous study demonstrating that elevated endothelial extravasation is one
of the early vascular changes in diabetes (Chakir and Plante, 1996).
In this chapter, vasodilatation responses were also examined in sensory denervated
(capsaicin pre-treated) rats, with and without diabetes. Capsaicin treatment significantly
reduced SP response to 54% of control but no significant change was observed in smooth
muscle reactivity (Figs 3-13 & 3-14). On the other hand, induction of diabetes caused a
significant reduction in both SP (42% of control) and SNP responses (28% of control).
When capsaicin rats were treated with STZ, both SP and SNP responses were further
reduced reaching 18% and 20% of their own controls respectively (Figs 3-13 & 3-14).
When SP is applied exogenously, it exerts its vasodilatory effect on rat skin in two ways.
The first is directly on blood vessels, causing NO release and vasodilatation (postterminal event). The second is indirectly by releasing histamine from mast cells that
subsequently excites sensory nerves, leading to release of more neurotransmitters from
these nerves (pre-terminal event) (Khalil and Helme, 1989). In the former situation, SP
acts upon neurokinin-1 (NK-1) receptors on the endothelium, initiating the phosphatidylinositol-bisphosphate pathway and activating NO synthase to produce NO, which then
diffuses into smooth muscle causing its relaxation and subsequent vasodilation. In the
latter situation, SP is involved in degranulation of mast cells and release of histamine,
which causes more SP to be released by sensory nerves.
The reduction in SP, in capsaicin rats (54% of control), is mainly related to the direct
effect of capsaicin on sensory C-fibres and dorsal root ganglia (DRG). Capsaicin
treatment of neonatal rats causes reduction in SP synthesis in DRG, impairment of SP
axonal transport, and depletion of 50-60% of SP in the skin (Fitzgerald, 1983; Gamse et
al., 1982). Although neonatal capsaicin treatment can destroy most of the C-fibres, the
remaining fibres are likely to remain normal and function efficiently. In capsaicin pretreated diabetic rats, the further reduction in smooth muscle reactivity (by 9%) alone
cannot explain the further reduction in SP response (by 24%) in these same rats.
Therefore, the possibility was raised that diabetes may have also affected the remaining
172
intact sensory nerves (after capsaicin treatment) and altered their function. This
possibility is further examined in part 2 of this chapter. Hence, the overall reduction in SP
response, in capsaicin pre-treated diabetic rats, could be related to the combined loss of
sensory nerves due to capsaicin treatment and the effect of diabetes on the remaining
intact nerves.
The effect of diabetes on microvasculature in STZ- and alloxan-treated rats
Early pre-clinical studies revealed that STZ is toxic to pancreatic cells, causing
hyperglycaemia and at high concentrations (above the diabetogenic dose), it can also
damage the kidneys (Weiss, 1982). The less common myelosuppression and
hepatotoxicity caused by STZ have also been previously reported (Weiss, 1982).
Evidence exists that the toxic effect of STZ on pancreatic beta cells involves the NO
pathway (Adeghate and Parvez, 2000). It was also hypothesized that STZ may cause
neuronal damage by liberating NO from its nitroso moiety (Prickaerts et al., 2000).
However, Prickaerts et al showed that administration of NO (using SNP), in vivo, did not
lead to any behavioural and / or neuronal deficits.
In this chapter, alloxan-treated rats showed a significant reduction in the SP response
(48% of control), which could be mainly due to the reduction in smooth muscle reactivity
(36% of control) (Figs 3-16 & 3-17). The CGRP response was also reduced reaching
39% of control response (Fig 3-18). Similar responses were observed in STZ-treated rats
showing significant reductions in SP, SNP, and CGRP responses (42%, 28% & 42% of
own control values respectively). Collectively, the data from both STZ- and alloxantreated rats support the hypothesis that the observed changes in diabetic microvascular
responses in STZ-treated rats after 4 weeks is most likely related to the effect of
hyperglycaemia. However, future investigation may be required to further clarify this
issue. It should be noted that since STZ and alloxan were administered via diferent
routes, a separate set of control rats (CB-treated (i.v.)) were used for alloxan-treated
animals.
173
In conclusion, the changes observed in vasodilation and PE results in diabetic rats support
the notion that there are functional and structural disturbances in endothelial cells. It was
demonstrated that the observed reduction in neurogenic inflammatory responses (SP &
CGRP) is mainly due to a reduction in smooth muscle reactivity. It was suggested that a
common mechanism could be involved in reducing SNP, SP, and CGRP responses in
early diabetes. Inactivation of NO and / or alterations in the efficiency of cGMP and
cAMP pathways were raised as possible causes for this reduction. Evidence was provided
that reduction in sensory nerve function in capsaicin pre-treated diabetic rats is largely
independent of smooth muscle reactivity. The results also supported the hypothesis that
the observed changes in the microvasculature of STZ-induced rats are most likely related
to the hyperglycaemic effect of STZ rather than a possible independent neurotoxic effect
of this drug.
In chapter 3, part 1 (above), the effect of diabetes on sensory-mediated events at the postterminal level was investigated. In chapter 3, part 2, the effect of diabetes on sensorymediated events at the pre- and post-terminal levels was studied. By utilizing both sets of
results (parts 1 & 2), changes in microvascular responses at the pre-terminal level can
also be examined.
174
3.2. Neurogenic inflammation: Part 2 (Pre-terminal and Post-terminal events)
3.2.1. Introduction
In part 1, a neurogenic inflammatory response was induced in diabetic rats by perfusing
sensory peptides exogenously over a blister base. In this section (part 2), neurogenic
inflammatory responses were induced in diabetic rats using electrical nerve stimulation
(ES), whereby the sciatic nerve is stimulated antidromically (at parameters known to
activate C-fibres) to mimic the axon reflex phenomenon, leading to the release of sensory
neurotransmitters (chapter 2, section 2.7). Previously, the release of SP and CGRP from
isolated tracheas of 4 weeks diabetic rats was shown to be substantially diminished in
response to electrical field stimulation (Nemeth et al., 1999a). Also, a reduction in
content and transport of SP and CGRP was observed in sciatic nerves of 4 weeks diabetic
rats (Diemel et al., 1992; Donadoni et al., 1995; Karanth et al., 1990). These changes in
sensory neuropeptides may be responsible for the weak neurogenic inflammatory
response observed in diabetic animals (Nemeth et al., 1999b).
The first aim of (chapter 3) part 2 is to examine changes in the vascular response to ES in
control and diabetic rats over time. The second aim is to examine the function of sensory
nerves in 4 weeks diabetic rats in relation to changes at the pre-terminal level. It was
hypothesized that there is an alteration in sensory mediated events at the pre-terminal
level.
Electrical stimulation of the unmyelinated C-fibres of the sciatic nerve can provide
information related to sensory mediated events at both pre-and post-terminal levels.
Although the ES technique relies on an invasive procedure in terms of nerve preparation,
the main advantage of antidromic stimulation is that the observed vascular responses are
induced by the endogenous release of peptides (e.g. SP, CGRP, etc).
Sensory denervated (capsaicin pre-treated) animals and alloxan-treated rats are also used
(different groups to part 1) to determine sensory independent and hyperglycaemic
independent mediated effects of diabetes on the microvasculature respectively.
175
3.2.2. Method
Eleven groups of rats were used in this study: four control, five diabetic, and two
capsaicin pre-treated groups. Four control groups and one of the capsaicin pre-treated
groups were injected with citrate buffer solution (CB-treated) (chapter 2, section 2.4.1).
One group of rats was made diabetic by alloxan injection (chapter 2, section 2.4.3). The
other groups, including one capsaicin pre-treated group, were made diabetic by STZ
injection (see the below diagram) (see also chapter 2, sections 2.8 & 2.4.4).
Rats were anaesthetized and a blister was induced on the hind footpad. The rat sciatic
nerve was prepared for ES, under aseptic conditions, as described in chapter 2 (section
2.7). Once the basal blood flow was established, first SNP (100μM, 10 min) was perfused
to measure smooth muscle reactivity. Then the basal blood flow was re-established using
Ringer’s solution (20 min) followed by activation of sensory nerve fibres (20V, 5Hz,
2msec for the duration of 1 minute) using Grass stimulator (chapter 2, section 2.7). In
control and diabetic rats, changes in the skin microvascular blood flow over time were
recorded by the PLDF.
Animals
Diabetic (STZ)
Control
(CB-treated)
(Weeks 0, 1, 5, 10)
2W
SNP
ES
SNP
4W
ES SNP
6W
ES SNP
9W
ES SNP
ES
Then, in capsaicin pre-treated rats (control and 4 weeks diabetic) and alloxan-treated
animals (4 weeks), the sciatic nerve was stimulated, using the above parameters, and
again the skin microvascular blood flows were recorded by the PLDF.
176
Animals
CB-treated
Diabetic
(alloxan, 4 W)
Capsaicin-treated
CB-treated
SNP
ES
SNP
ES
SNP
STZ-treated
(4 W)
ES SNP
ES
3.2.2.1. Expression of data and statistical analyses
Results were calculated as areas under the response curves and percent of control. All
data are expressed as mean ± standard error of the mean (SEM). The absolute values and
95% confidence intervals (95%CI) of these values are quoted in the text.
Separate one-way Analyses of Variances (ANOVAs) were performed to investigate
whether there were significant differences in responses to SNP and ES between control
and diabetic rats over time. Where ANOVA indicated a significant main effect, post hoc
comparisons were conducted using either Dunnett’s or independent sample t-tests. A twoway ANOVA was also performed to investigate whether there were significant
differences in responses to ES between control and capsaicin pre-treated rats (with and
without diabetes) and interaction between these groups. Where ANOVA indicated a
significant main effect, post hoc comparisons were conducted using independent sample
t-tests. Separate independent sample t-tests were performed to examine other results
involving control and diabetic rats.
3.2.3. Results
Originally random control groups for weeks 0, 1, 5, and 10 were tested. Once again,
results showed the vascular responses to SNP and ES from control rats at weeks 1, 5, and
10 were similar to those at week zero (Appendix A, Table A-1). Hence, the results of
week zero were reported as controls in this section of the study.
177
Responses of diabetic rats to SNP and ES were all significantly reduced compared to
Area under the response curve (cm2 )
their own control groups (Figs 3-19 & 3-20).
60
Control SNP
Diabetic SNP (2 weeks)
Diabetic SNP (4 weeks)
Diabetic SNP (6 weeks)
Diabetic SNP (8 weeks)
50
40
30
20
*
X
*
*
*
10
0
Fig 3-19: Changes in the vascular response to SNP (100 μM) in diabetic rats over time, presented as
Area under the response curve (cm2 )
mean ± SEM (n = 6-9).
* Denotes significance from control rats (P<0.05)
X Denotes significance from 2 weeks diabetic rats (P<0.05)
60
Control ES
Diabetic ES (2 weeks)
Diabetic ES (4 weeks)
Diabetic ES (6 weeks)
Diabetic ES (8 weeks)
50
40
30
20
*
10
*
*
*
0
Fig 3-20: Changes in the vascular response to ES (20V, 5Hz, 2msec for 1 minute) in diabetic rats over
time, presented as mean ± SEM (n = 6-9). * Denotes significance from control rats (P<0.05)
178
At week 4, both SNP and ES responses reached 28% of their own controls respectively.
Diabetes and changes in sensory nerve function in capsaicin pre-treated rats
In capsaicin pre-treated rats, the vasodilator response to ES [(9.7 ± 1.5 cm2) (95%CI 6.812.6)] was significantly reduced, reaching 55 ± 8.4% of control [(17.6 ± 1.3 cm2) (95%CI
15-20.1] (Fig 3-21).
Control ES
Capsaicin ES
Diabeti c ES
Diabeti c + Capsaicin ES
(%) Percent of control
120
100
80
*
60
X
40
*
*
20
0
Fig 3-21: Vascular responses to ES (20V, 5Hz, 2msec for 1 minute) in control, diabetic, capsaicin
pre-treated, and capsaicin pre-treated diabetic rats, presented as percent of control ±
SEM (n = 6).
* Denotes significance from control rats (P<0.05)
X Denotes significance from capsaicin pre-treated rats (P<0.05)
This reduction in this response was similar to the reduction in SP response in these
animals (53.8 ± 7.1%) (Fig 3-14). The smooth muscle reactivity in capsaicin rats [(19.7 ±
1.9 cm2) (95%CI 16-23.4)] was not significantly different from control (Fig 3-13).
In capsaicin pre-treated diabetic rats, however, the vascular response to ES was further
reduced [(5.9 ± 1 cm2) (95%CI 3.9-7.9)], reaching 33.5 ± 5.7% of control which was
similar to the ES response in diabetic rats (28.9 ± 4.8% of control) (Fig 3-21). The
response to SNP in capsaicin pre-treated diabetic rats [(5.2 ± 0.6cm2) (95%CI 4-6.4)]
reached 20 ± 2.5% of control which was similar to those in diabetic rats [(7.4 ± 0.6 cm2)
(95%CI 6.2-8.6)] (29.3 ± 3% of control) (Fig 3-13).
179
Diabetes and changes in sensory nerve function in alloxan-treated rats
The vascular response to ES in alloxan-treated rats was reduced (8.2 ± 1.8 cm2) (95%CI
4.7-11.7) reaching 46% of control ES. This was greater than the ES response in STZ-
Area under the response curve (cm2 )
treated rats [(5.1 ± 2 cm2) (95%CI 1.2-9)] (Fig 3-22).
60
Control ES
STZ-treated ES
Alloxan-treated ES
50
40
30
20
*
10
*
0
Fig 3-22: Comparing the response to ES (20V, 5Hz, 2msec for 1 minute) in control, STZ-treated, and
alloxan-treated diabetic rats, presented as mean ± SEM (n = 6).
* Denotes significance from control rats (P<0.05)
3.2.4. Discussion
The results showed a significant reduction in ES responses of diabetic rats over time (Fig
3-20). However, these reductions were similar to the reductions in the smooth muscle
reactivity of the corresponding diabetic rats (Fig 3-19). For example, in 4 weeks diabetes,
the responses to ES and SNP both reached 28% of their own controls. This suggests that
at 4 weeks of diabetes the observed reduction in ES responses could be due to reduction
in smooth muscle reactivity rather than damage to the C-fibres by diabetes, which further
supports the results in chapter 3, part 1.
The role of sympathetic nerves in reducing ES responses in diabetic rats cannot be ruled
out and is investigated in chapter 5. Capsaicin pre-treatment of neonatal rats causes
reduction in SP synthesis in DRG, its transportation across the nerve, as well as its
release from nerve terminals (Fitzgerald, 1983; Gamse et al., 1982). This may explain the
reduction in ES response (55% of control) with no significant change in the smooth
180
muscle reactivity in capsaicin pre-treated rats. The ES and SNP responses, in capsaicin
pre-treated diabetic rats, were further reduced (33.5% & 20% of own controls
respectively) when compared to these responses in capsaicin pre-treated rats. This
suggests that diabetes reduced SNP response in these rats by an action independent of
sensory nerves. Therefore, the reduction in microvascular blood flow in 4 weeks diabetic
rats may be due to changes in smooth muscle reactivity rather than an action on sensory
nerves.
The vascular response to ES in alloxan-treated rats (46% of control) was less reduced
when compared to those in STZ-treated animals (29% of control). In part 1, it was
reported that the blood glucose level in alloxan-treated rats was approximately
27.7mmol/lit (while 90% of the STZ-treated rats were above this figure) and alloxantreated animals were in better physical health than STZ-treated animals. These
observations suggest that although both groups of rats showed severe diabetic symptoms,
the alloxan-treated rats may be less diabetic. The blood glucose-monitoring device that
was employed in this study, was originally designed to be used by diabetic patients and
was capable of accurately measuring the blood glucose levels up to 27mmol/lit (chapter
2, section 2.4). Therefore, it was not possible to report the exact blood glucose level in
diabetic rats since they were equal to or above 27.7mmol/lit. The severity of diabetes can
be related to the concentration of alloxan used in this study. The minimum concentration
of alloxan required for producing diabetes in 80% of rats was 40mg/kg. On the other
hand, 75mg/kg of STZ proved to be the minimum concentration required for producing
the same number of diabetic rats. Due to ethical issues, higher concentrations of alloxan
were not used. Therefore, direct comparison of results between these two groups of rats
was not possible (see also chapter 2, section 2.4.5).
In conclusion, the results further confirmed that post-terminal changes in microvascular
function are more apparent than the pre-terminal changes in early stages of diabetes. The
data from part 1 and part 2 of this chapter, collectively, suggest that other factors
associated with diabetes may be involved in reducing the inflammatory responses in 4
weeks diabetic rats, acting mostly at the post-terminal level (see also chapters 6 & 7). It
was also demonstrated that in experimental diabetes, the type and / or the dose of the
181
diabetogenic agent are important factors to consider and that interpretation of
comparative results should be made with caution.
182
CHAPTER 4
Diabetes and modulation of microvascular blood flow:
Role of local tissue factors and parasympathetic
neurotransmitters
183
4.1.
Introduction
Changes in the vascular response to acetylcholine (ACh) with diabetes
The interaction between blood vessels and perivascular nerves in the periphery is
important for many physiological responses including neurogenic inflammation and
blood flow regulation. A number of neurotransmitters have been identified in
perivascular nerves and some are co-localized in: sensory nerves (substance P (SP),
calcitonin gene-related peptide (CGRP) and neurokinin A (NKA)), parasympathetic
nerves (ACh and vasoactive intestinal peptide (VIP)), and sympathetic nerves
(noradrenaline (NA) and neuropeptide Y (NPY)), (Bennett and Gardiner, 1996). The
close apposition between these nerve varicosities and endothelial cells in capillaries
raises the possibility of direct interactions between nerves and endothelial cells in the
microvasculature. Vasoactive substances released from these nerves and endothelial cells
play a role in the control of microvascular tone, either independently or collectively.
Smooth muscle is innervated by noradrenergic, and in some instances by cholinergic,
nerve fibres and hence their reactivity is also important in the modulation of
microvascular tone.
In diabetes, previous studies showed that the basal release of ACh was not altered at 1
week but was significantly reduced after 5 weeks (Welsh and Wecker, 1991). Diabetes
may also be associated with an early alteration (2 weeks) to muscarinic receptor density
and function (Soulie et al., 1992; Tong et al., 1999; Wald et al., 1998). The relaxation
response to ACh in diabetic animals was different depending on the duration of diabetes
and type of vascular bed under study (Chan et al., 2000; Pieper, 1999) (chapter 1, section
1.5.4.2). It was proposed that ACh-induced dilatation is mediated by multiple
mechanisms that may include the endothelial-dependent production of nitric oxide (NO)
and endothelial-derived hyperpolarizing factor (EDHF) (Edwards et al., 1998; Quilley et
al., 1997).
Reduced production of NO (Pieper and Dondlinger, 1997) and decreased activity of
superoxide dismutase in diabetes (Kobayashi and Kamata, 1999b) were suggested to be
two possible causes for impaired ACh response. Evidence exists that in normal rats, free
radicals can depress endothelium-dependent relaxations mediated by activation of
184
muscarinic receptors (Prasad and Bharadwaj, 1996; Rubanyi and Vanhoutte, 1986). In
diabetes, free radicals have also been implicated in altering the function of vascular
smooth muscle (Fleischhacker et al., 1999; Sakata et al., 1998), leading to development
of circulatory dysfunction (chapter 1, section 1.5.3).
Previous studies in early diabetic animals did not assess the role of the parasympathetic
neurotransmitters in modulation of blood flow in the skin tissue. Therefore, in chapter 4,
changes in skin microvascular blood flow in diabetic rats are examined using exogenous
perfusion of ACh and two related receptor agonists, namely nicotine and muscarine
chloride. The role of NO in ACh-induced vascular responses in early diabetes is also
examined. Furthermore, the relative contribution of the superficial and deeper vessels to
vascular activity of ACh in diabetic rats (4 weeks) is determined using two laser Doppler
flowmeters (LDFs), Perimed and Moor (PLDF & MLDF).
Changes in the vascular response to bradykinin (BK) with diabetes
Several endogenous local tissue factors appear to be involved in the modulation of the
peripheral release of neuropeptides from sensory nerves. An important local tissue factor
is BK, which is produced during inflammation and tissue injury and can both sensitize
and directly stimulate sensory nerve terminals (Dray and Perkins, 1993). In rats, BK was
shown to induce neurogenic inflammatory responses (Cambridge and Brain, 1995; Khalil
and Helme, 1992; Miao et al., 1996), and play a role in activation of mast cells (Zhao et
al., 1996). This local tissue factor also has diverse actions on vascular endothelium,
smooth muscle, and cellular function (chapter 1, Figs 1-11 & 1-12).
BK actions are mediated directly via B1 and B2 receptors (Hall, 1997) on target tissues
and indirectly by release or amplification of NO or other inflammatory mediators
including neuropeptides and prostaglandins (chapter 1, section 1.8.5.2). Based on
previous studies, including one study on rat skin, the general conclusion has been made
that acute activation by BK of sensory neurones is mediated through B2 receptors (Dray
and Perkins, 1993; Jones et al., 1995). However, where there is a significant
inflammatory process, B1 receptors are expressed and contribute to the action of BK on
sensory nerves (Farmer et al., 1991; Hall, 1997; Perkins et al., 1993).
185
The vascular response to BK in diabetic rats, once again, seems to depend on the vascular
tissue under study and the duration of diabetes (chapter 1, section 1.8.5.2). In early
diabetes (1-10 days) an increase in BK density was demonstrated in both in vivo and in
vitro situations (Pheng et al., 1997; Velarde et al., 1997; Zuccollo et al., 1996). Also,
alteration in B2 receptor activity was reported on the smooth muscle cells in STZdiabetic rats (4 weeks) (Wang et al., 1998).
The role of BK in modulating microvascular blood flow was previously studied in
different diabetic vascular beds (hindquarter, mesenteric artery, tail artery, cremaster
muscle arterioles) (3-4 weeks) but not skin (Kiff et al., 1991b). Therefore, in this chapter,
changes in skin microvascular responses to BK are examined in 4 weeks STZ-diabetic
rats. The role of NO in BK-induced vascular response in early diabetic rats is also
assessed.
Changes in NO involvement in microvascular responses in diabetes
The vascular endothelium synthesises a number of vasoactive substances that modulate
vascular tone and smooth muscle proliferation. NO was shown to mediate endotheliumdependent relaxation (Furchgott and Zawadzki, 1980). Endothelial and neuronal nitric
oxide synthase (eNOS and nNOS) are constitutively expressed and, upon activation (e.g.
by ACh, BK), can generate NO within a short period of time (Khalil and Helme, 1992;
Ralevic et al., 1992). Inducible NOS (iNOS), however, is not normally expressed but it
can be activated by pro-inflammatory agents, leading to the generation of large quantities
of NO (Moncada and Higgs, 1993).
NO plays an important role in cell communication, defence, and injury. Increasing
evidence indicates that NO may play a role in inflammation (Ialenti et al., 1992). In rat
skin, NO has a role in the vasodilatation and inflammatory response to BK (Khalil and
Helme, 1992), SP and ACh (Khalil et al., 1994; Ralevic et al., 1995b). NG-nitro-Larginine methyl ester (L-NAME), an inhibitor of NO synthase, attenuated vasodilatation
and plasma extravasation (PE) responses induced by SP (Ralevic et al., 1995b), BK
(Khalil and Helme, 1992) and the vasodilatation response to ACh (Ralevic et al., 1992).
186
Experimental evidence suggested that any one or combination of the following factors
might occur in diabetes, which could contribute to vascular complications: a reduction in
NO release (Steinberg et al., 1994) and its bioactivity (Bucala et al., 1991; Pieper et al.,
1997b), possible defect in secondary messenger system (cyclic guanosine monophosphate
(cGMP)) (Trovati et al., 1996), and defective vascular smooth muscle response to NO
(Kamata et al., 1992).
In this chapter, the role of NO in modulating microvascular blood flow in 4 weeks
diabetic rats is examined using a NO synthase inhibitor, L-NAME. The aim is to examine
the contribution of NO to SP response in 4 weeks diabetic rats.
4.2.
Method
A total of 16 groups of rats (n = 6-9) were used in this study, of which 14 groups were
equally divided into control (treated with 0.1M sodium citrate buffer) and STZ-treated
animals (chapter 2, sections 2.4.1 & 2.4.4). The following substances were perfused over
the footpad blister base of both control and diabetic rats.
Group 1: ACh (100μM, 30min)
sodium nitroprusside (SNP, 100μM, 10min)
Group 2: Nicotine (100μM, 30min)
SNP (100μM, 10min)
Group 3: Muscarine Chloride (100μM, 30min)
Group 4: BK (10μM, 30min)
SNP (100μM, 10min)
SNP (100μM, 10min)
Group 5: L-NAME (100μM, 10min)
[L-NAME (100μM) + BK (10μM)] (30 min)
SNP (100μM, 10min)
Group 6: SP (1μM, 30 min)
SNP (100μM, 10min)
Group 7: L-NAME (100μM, 10min)
[L-NAME (100μM) + SP (1μM)] (30 min)
SNP (100μM, 10min)
The remaining two groups (one diabetic and one control) were used in an experiment
using exogenous perfusion of ACh and blood flow measurement using the MLDF.
Induction of diabetes and peptide perfusion has been described in detail in chapter 2,
sections 2.4 and 2.6. PE was collected from all experiments and at the end of each
experiment, rats were sacrificed with pentobarbitone overdose.
187
4.2.1. Expression of data and statistical analyses
Results from vasodilatation responses were calculated as areas under the response curves.
In some instances responses were calculated as percent above basal or percent of control
to facilitate comparisons between groups. PE responses are presented as percent above
basal. All data are expressed as mean ± standard error of the mean (SEM). The absolute
values and 95% confidence intervals (95% CI) of these values are quoted in the text.
Separate one-way Analyses of Variances (ANOVAs) were performed to investigate
whether there were significant differences in responses to BK and BK + L-NAME or SP
and SP + L-NAME, between control and diabetic rats. Where ANOVA indicated a
significant main effect, post hoc comparisons were conducted using the Tukey HSD test.
Separate one-way repeated measure ANOVAs were performed to investigate group
differences (control and diabetic) according to the PLDF and the MLDF in response to
ACh at different times of perfusion (ACh10, ACh20, ACh30). Where a significant main
effect was detected, post hoc independent sample t-tests were performed. To investigate
whether there were significant differences in PE responses to BK, BK + L-NAME, SP,
SP + L-NAME, nicotine, and muscarine chloride between control and diabetic rats, a
one-way repeated measure ANOVA with priori planned comparisons was performed.
Separate independent sample t-tests were performed to examine other results involving
control and diabetic rats.
4.3.
Results
Changes in the vascular response to ACh with diabetes
The average basal blood flow (BBF) was (24.7 ± 1.1 cm2) (95%CI 22.5-26.8) in control
rats and (14.8 ± 1.3 cm2) (95%CI 12.2-17.3) in diabetic rats.
188
The vascular response to SNP showed a significant reduction in diabetic rats [(6.1 ± 0.8
cm2) (95%CI 4.5-7.7)], reaching 25% of control [(24.5 ± 1.9 cm2) (95%CI 20.8-28.2)]
Area under the response curve (cm2 )
(Fig 4-1).
60
50
Control SNP
Diabetic SNP
40
30
20
10
*
0
Fig 4-1: Vascular responses to SNP (100μM) in control and diabetic rats, presented as mean ± SEM
(n = 6-7). * Denotes significance from control rat (P<0.05)
There was no difference between vascular responses to ACh [(41.1 ± 3.3 cm2) (95%CI
34.6-47.6)], nicotine [(42.4 ± 5.2 cm2) (95%CI 32.2-52.6)], and muscarine chloride [(40.6
Area under the response curve (cm2 )
± 4.3 cm2) (95%CI 32.2-49)] in control animals (Fig 4-2).
Control ACh
Control Nicotine
Control Muscarine chloride
60
50
40
30
20
10
0
Fig 4-2: Vascular responses to ACh, Nicotine, and Muscarine chloride (all 100μM) in control rats,
presented as mean ± SEM (n = 6-7).
189
However, in diabetic rats, all these responses were significantly reduced, reaching 2736% of their own control responses. Responses to ACh reached (11 ± 1.8 cm2) (95%CI
7.5-14.5), to nicotine reached (15.3 ± 1.9 cm2) (95%CI 11.6-19), and to muscarine
chloride reached (13.7 ± 2.1 cm2) (95%CI 9.6-17.8) in these animals (Fig 4-3).
Area under the response curve (cm2 )
60
Diab ACh
Diab Nicotine
Diab Muscarine chloride
50
40
30
20
10
0
Fig 4-3: Vascular responses to ACh, Nicotine, and Muscarine chloride (all 100μM) in diabetic (Diab)
rats, presented as mean ± SEM (n = 6-7).
The profile of the vasodilator response to ACh was characterized by a tachyphylaxis
where there was a gradual reduction in response despite continuous perfusion. Similar
profiles were observed using both the PLDF and the MLDF instruments (Figs 4-4 & 4-5).
A C h
10
20
30
Fig 4-4: Response profile to ACh (100μM, 30 min) in control rat skin microvasculature, as recorded by
the PLDF.
190
S N P
A C h
400
Fig 4-5: Response profils to SNP (100μM, 10 min) and ACh (100μM, 30 min) in control rat skin
microvasculature,as recorded by the MLDF.
Using the MLDF and the PLDF together enables a comparison of microvascular
responses to ACh measured from all skin compartments (superficial and deep) to those
measured from superficial compartments. The total vascular response to ACh calculated
as a percent above basal was similar between Perimed and Moor in control animals (165
± 10% & 128.5 ± 14%) but was different in diabetic rats (127.5 ± 9% & 80.5 ± 6.2%)
(Table 4-1). The ratio of blood flow at the superficial compartment measured by the
PLDF to that of both superficial and deep compartments measured by the MLDF, was 1:1
in control animals but increased to about 2:1 after 4 weeks of diabetes.
The results from the PLDF showed the most reduction in response to ACh in diabetic rats
(by 30%) was during the first 10 minutes of perfusion (Table 4-1, Fig 4-6). This was
followed by 2.5% and 5% reductions during the next 20 minutes of perfusion (ACh20
and ACh30 respectively).
PLDF (%)
MLDF (%)
RAT
ACh10
ACh20
ACh30
ACh10
ACh20
ACh30
Control
92 ± 12
53 ± 12
20 ± 7
59 ± 17
41.5 ± 15
28 ± 13.5
Diabetic
62 ± 12
50.5 ± 11
15 ± 2
36 ± 7
26 ± 6
18.5 ± 6
Table 4-1: Vascular responses to ACh (100μM) in control and diabetic rats (4 weeks) measured by
the PLDF and the MLDF, presented as percent above basal ± SEM (n = 6).
191
(%) Percent above basal
120
Control ACh
Di ab ACh
100
80
60
40
20
0
ACh10
ACh30
ACh20
Fig 4-6: Vascular responses to ACh (100μM) in control and diabetic (Diab) rats (4 weeks) at 10 minutes
intervals measured by the PLDF, presented as percent above basal ± SEM (n = 6).
Results measured by the MLDF, on the other hand, showed similar reduction during the
first 10 minutes (by 23%) but more reduction in the following 20 minutes of ACh
perfusion (15.5% & 9.5%) (Fig 4-7).
(%) Percent above basal
120
Control ACh
Diab ACh
100
80
60
40
20
0
ACh10
ACh20
ACh30
Fig 4-7: Vascular responses to ACh (100μM) in control and diabetic (Diab) rats (4 weeks) at 10
minutes intervals measured by the MLDF, presented as percent above basal ± SEM (n = 6).
192
Changes in the vascular response to BK and NO involvement with diabetes
The average BBF was (24.1 ± 2.1 cm2) (95%CI 20-28.2) in control animals and (15.8 ±
0.9 cm2) (95%CI 14-17.6) in diabetic rats.
The vascular response to SNP showed a significant reduction in diabetic rats [(8.2 ± 0.7
cm2) (95%CI 6.8-9.6)], reaching only 32% of control [(25.6 ± 1.3 cm2) (95%CI 23-28.1)]
Area under the response curve (cm2 )
(Fig 4-8).
60
Control SNP
Diabetic SNP
50
40
30
20
*
10
0
Fig 4-8: Vascular responses to SNP (100μM) in control and diabetic rats, presented as mean ± SEM
(n = 6-9). * Denotes significance from control rats (P<0.05)
In control rats, perfusion of L-NAME significantly decreased the vascular response to
BK [(38 ± 2.9 cm2) (95%CI 32.3-43.7)], reaching 64.4 ± 5% of control response [(58.3 ±
1.3 cm2) (95%CI 55.7-60.8)]. The vascular response to BK was also diminished in
diabetic rats [(27.9 ± 2.5 cm2) (95%CI 23-32.8)], reaching 48 ± 4.2% of control response.
Compared to control animals, L-NAME was more potent in reducing the BK response in
diabetic rats [(11.8 ± 1.2 cm2) (95%CI 9.4-14.1)], reaching only 20 ± 2% of control
response (Fig 4-9).
193
Control BK
Control BK + L-NAME
Diabetic BK
Diabetic BK + L-NAME
(%) Percent of control
120
100
80
*
60
*
40
‹
x
20
0
Fig 4-9: Vascular responses to BK (10μM) in control and diabetic rats, before and after L-NAME
(100μM) perfusion, presented as percent of control ± SEM (n = 6-9).
* Denotes significance from BK in control rats (P<0.05)
X Denotes significance from BK + L-NAME in control rats (P<0.05)
‹ Denotes significance from BK in diabetic rats (P<0.05)
L-NAME perfusion reduced SP response in control rats [(51.4 ± 2.9 cm2) (95%CI 45.757.1)] to 42.5 ± 5.5% of control value [(20.9 ± 3 cm2) (95%CI 15-26.8)] (Fig 4-10).
120
Control SP
Control SP + L-NAME
Diabetic SP
Diabetic SP + L-NAME
(%) Percent of control
100
80
60
#
#
40
20
∇
+
0
Fig 4-10: Vascular responses to SP (1μM) in control and diabetic rats, before and after L-NAME
(100μM) perfusion, presented as percent of control ± SEM (n = 6-9).
# Denotes significance from SP in control rats (P<0.05)
+ Denotes significance from SP + L-NAME in control rats (P<0.05)
∇ Denotes significance from SP in diabetic rats (P<0.05)
194
The vascular response to SP in diabetic rats [(21.6 ± 3.4 cm2) (95%CI 14.9-28.8)] was
further more reduced by this NO inhibitor [(7.6 ± 1.4 cm2) (95%CI 4.8-10.3)], reaching
15 ± 3% of control response.
PE was collected from control and diabetic rats. The PE response to SP (SP2,
3 & 4)
was
calculated as percent above basal ± SEM. SEMs were omitted from Fig 4-11 for clarity
but quoted in the text. The basal PE (R1) did not change in either control or diabetic rats
after L-NAME perfusion (Appendix B, Table B-2). However, in control rats, L-NAME
caused a 20 ± 22% reduction in PE response during the first 10 minutes of SP perfusion.
This reduction became significant during the following 20 minutes reaching 45 ± 31% at
SP3 and 53 ± 19% at SP4. In diabetic rats, there was no significant reduction in PE
responses to SP at any time after L-NAME was perfused (Appendix B, Table B-2).
In control rats, PE responses to nicotine (20 ± 9%, 52 ± 17%, 64 ± 22%) were weaker
than SP, whereas in diabetic rats (4 ± 9%, 9 ± 8%, -9 ± 10%) these responses were
similar (Fig 4-11).
Control BK
Diab BK
Control SP
Diab SP
Control Nicotine
Diab Nicotine
Control Mus.Cl
Diab Mus.Cl
(%) Percent above basal
120
100
80
60
40
20
0
R1-SP2
R1-SP3
R1-SP4
Fig 4-11: PE in control {SP (1μM), BK (10μM)}, diabetic (Diab) {SP (1μM), BK (10μM)}, control after
Nicotine (100μM) or Muscarine chloride (100μM), and diabetic rats after Nicotine (100μM)
or Muscarine chloride (100μM), presented as percent above basal (n = 6-9).
195
PE responses to BK were similar to SP in both control [(35 ± 8%, 86 ± 13%, 114 ± 30%)
& (36 ± 19%, 72 ± 18%, 89 ± 14%) respectively) and diabetic rats [(3 ± 5%, 12 ± 6%, 7
± 5%) & (9 ± 23%, 16 ± 32%, 0.1 ± 23%) respectively).
4.4.
Discussion
It is known that blood vessels from humans and animals with experimental diabetes
exhibit changes to vasoactive agents (Vane and Botting, 1994). This may lead to vascular
dysfunction and diabetes-associated complications (Cooper et al., 1997). In this chapter,
the microvascular response to ACh was examined in early diabetes. In addition, the
changes in the contribution of NO to BK and SP responses were assessed in 4 weeks
diabetic rats with a focus on the relative contribution of the superficial and deeper vessels
to vascular reactivity to ACh in the utilized model.
In general, BBFs in control and diabetic rats (4 weeks) showed similarity to those
observed in chapter 3. There was a reduction in BBF in diabetic rats when compared to
controls. The similarity between SP response in control rats after L-NAME and SP
response in diabetic rats suggests, in absolute terms, that the contribution of NO to this
response is reduced (Fig 4-10). It is possible that NO synthesis is not reduced but the
bioactivity of NO is diminished possibly due to oxidative stress. The above results also
suggests that since the relative contribution of NO to SP response is not altered after 4
weeks of diabetes, any change in BBF may be related to increase in sensitivity of blood
vessels to vasoconstrictors (see chapter 6). In chapter 5, the role of sympathetic nerves in
reducing basal blood flow in diabetic animals is examined.
Changes in the vascular response to ACh with diabetes
Vasodilator responses to nicotine and muscarine chloride were similar to that of ACh in
both control and diabetic rats with the latter showing significant reductions in all
responses (Figs 4-2 & 4-3). A previous study in rat skin microvasculature showed that the
vasodilator response to ACh is partially mediated by NO release when muscarinic (M3)
receptors on the endothelium are directly stimulated (Ralevic et al., 1992). The other part
of the response is mediated by sensory neurotransmitters (e.g. SP, CGRP) released by
activation of nicotinic receptors on sensory nerves (Ralevic et al., 1992). The vasodilator
196
response to nicotine and muscarine chloride is also mediated by the indirect and direct
effect of these substances on nicotinic and muscarinic receptors respectively, releasing
NO, which may explain the similarity between these three responses.
In diabetic rats, the vascular response to ACh, nicotine, and muscarine chloride were
similarly reduced, reaching 27%, 36% and 34% of their own controls respectively (Fig 43). Since all these vasoactive agents are known to cause the release of NO, it is possible
that in diabetes, the released NO is deactivated by free radicals such as superoxide anions
(Gryglewski et al., 1986), causing reduced vasodilation. Previous studies supported this
notion by showing an increase in free radical concentration and a decrease in the level of
antioxidants such as SOD (Ha and Kim, 1999; Kobayashi and Kamata, 1999b; Traverso
et al., 1998). SOD is known to be capable of reducing the concentration of superoxide
anions (Diederich et al., 1994). The vascular response to SNP (a NO donor) in diabetic
rats was reduced to 25% of its own control, providing support for this proposition (see
also section 1.5.4.2). The effect of free radicals on microvascular responses in 4 weeks
diabetic rats is examined in chapter 7. The other possibility could be alterations in
muscarinic receptor function. In diabetes, damage to the endothelium could alter
muscarinic (M3) receptor function that mediates vasodilation (Huszka et al., 1997). This
may explain the reduction in response to muscarine chloride in 4 weeks diabetic rats (Fig
4-3). The similarity between reduction in response after ACh and muscarine chloride in
diabetic rats provides further support for this proposition.
A previous study by Kobayashi showed no alteration in ACh response in aorta of 4 weeks
diabetic rats in vitro (Kobayashi and Kamata, 1999b). The discrepancy between the
experimental results in this study and that of Kobayashi could be related to the difference
in the vascular bed and the fact that the results in this study were from an in vivo
preparation.
The relative contribution of different skin compartments to the vasodilator response to
ACh was examined in this study. In diabetic rats, vascular responses to ACh, recorded by
the PLDF and the MLDF, during the first 10 minutes of perfusion, were equally reduced
(by 30% & 23% respectively) (Figs 4-6 & 4-7). During the following 20 minutes of the
peptide perfusion, however, the MLDF showed a further 25% reduction in response
197
compared to only 7.5% reduction recorded by the PLDF. These results suggest that
around 17.5% of the reduction in ACh response is mostly related to changes in
microvascular function of deeper skin compartments (e.g. AVAs).
In normal and diabetic rats respectively, total ACh response measured by the PLDF and
calculated as a percent above basal was greater (165 ± 31% & 128.5 ± 45.5%) than the
MLDF (127.5 ± 25% & 80.5 ± 19%) (Table 4-1). Also, the ratio of superficial to
combined superficial and deeper vessel’s blood flow was doubled in the diabetic
condition. Overall, these results suggest that both superficial and deeper vessels are
affected by diabetes. Also, the greater percent reduction in ACh response using the
MLDF compared to the PLDF might suggest an additional involvement of deeper vessels
in diabetes, further supporting the results in chapter 3.
Changes in the vascular response to BK and NO involvement with diabetes
In control rats, although perfusion of L-NAME significantly reduced the vascular
response to BK (65% of control), it showed more potency in reducing SP response in
these animals (40% of control) (Figs 4-9 & 4-10). The reduction in BK and SP responses
in control rats after L-NAME perfusion is mainly attributed to inhibition of NO synthase
activity and unavailability of NO. The greater effect of L-NAME on SP response can be
related to a greater contribution of NO to SP-mediated vasodilation in rat skin.
A previous study showed that a significant part of the vasodilator response to BK is
mediated through an excitatory action on peripheral nociceptive C-fibres where sensory
neurotransmitters (e.g. SP, CGRP) are released (Khalil and Helme, 1992). SP,
exogenously applied, exerts its vasodilatory effect on rat skin in two ways. The first is
directly on blood vessels, causing NO release, and vasodilatation. The second is
indirectly by degranulating mast cells with subsequent release of mediators that excite
sensory nerves, leading to release of more neurotransmitters (Khalil and Helme, 1989).
Therefore, when SP is used as a vasodilator agent, more NO is released and hence LNAME is capable of exerting a greater inhibitory effect.
In diabetic rats, vascular responses to SP and BK were significantly diminished reaching
42% and 48% of own controls respectively (Figs 4-9 & 4-10). These reductions could be
partially related to a decrease in smooth muscle reactivity (28% of control) (Fig 4-8).
198
Perfusion of L-NAME, in diabetic rats, caused a further (similar) reduction in SP and BK
responses (15% & 20% of own controls respectively). Results from both BK and SP
responses after L-NAME perfusion suggests that endothelial cells, in 4 weeks diabetic
rats, are still capable of synthesizing NO. The similarity between SP response in control
rats after L-NAME (40% of control) and SP response in diabetic rats (42% of control)
might suggest that the relative contribution of NO to the response is similar. The overall
reduction in vascular responses to SP, BK, and SNP, therefore, could be related to rapid
deactivation of NO. The possibility was raised that the presence of certain substances in
diabetic condition may contribute to this deactivation (Bucala et al., 1991; Pieper et al.,
1997b).
Plasma extravasation and diabetes
The results of this study showed similarity in PE responses to BK and SP in control rats
(Appendix B, Table B-2). This could be due to a similarity between mechanisms
involved in inducing PE response by these agents (Khalil and Helme, 1989; Khalil and
Helme, 1992). In rat skin, BK is known to stimulate sensory nerves to release SP (Khalil
and Helme, 1992). Therefore, the first 10 minutes of an inflammatory response to SP or
BK is possibly mediated by an action of SP at the post-capillary venules through its Cterminal, an action which is independent of histamine or prostaglandins (Khalil and
Helme, 1989). Both SP and BK are capable of stimulating mast cells to release histamine
(Khalil and Helme, 1992; Lee and Pearce, 1990; Zhao et al., 1996). This could contribute
to the PE response after the first 10 minutes interval (Khalil and Helme, 1989).
Once again, in diabetic rats, neither BK nor SP induced a significant change in the
general PE (Fig 4-11). These data confirm the results in chapter 3 and raise the notion
that the ability of SP or BK to induce PE is masked by diabetes-induced endothelial
changes that results in high basal PE.
In conclusion, evidence was provided that NO plays an important role in acute
inflammation in early diabetes and that the reduced skin inflammatory responses in
diabetic rats could be related to inactivation of NO. Evidence was provided to show that
the bioactivity of NO is diminished after 4 weeks of diabetes. It was hypothesized that the
199
release of certain agents in early diabetes could affect NO activity, contributing to the
reduced microvascular responses. This hypothesis will be further investigated in chapters
6 & 7. It was also demonstrated that both superficial and deeper vessels of the rat skin are
affected by diabetes after 4 weeks. However, deeper vessels may have an additional
involvement in reducing microvascular blood flow in these animals. The possibility was
raised that the overall reduction in endothelial-mediated responses in early diabetes might
be related to the fate of NO once it is released.
200
CHAPTER 5
Diabetes and modulation of microvascular blood flow:
Role of sympathetic nerves
201
5.1.
Introduction
Blood vessels in the skin are innervated not only by sensory but also sympathetic nerve
fibres and both sets of nerves release vasoactive mediators. It has been proposed that
sympathetic nerve fibres strongly influence blood flow in skin microvasculature (Kurvers
et al., 1997; Pinter et al., 1997), particularly small arteries, arterioles and arteriovenous
anastomoses (AVAs), which are richly supplied with these fibres (Norberg and
Hamberger, 1964).
It is known that stimulation parameters that activate the small unmyelinated sensory
nerves can also activate the sympathetic efferents, with the release of neurotransmitters
that will act on the microvasculature in an actively opposing manner (chapter 1, section
1.7.4). Most of the sympathetic nerves, which innervate blood vessels, release the potent
vasoconstrictor noradrenaline (NA), which acts on alpha (α-) and beta (β-)
adrenoreceptors to mediate its biological response (chapter 1, section 1.8.3). These
receptors are found on smooth muscle cells and may also exist on unmyelinated nerve
axons (Bennett and Gardiner, 1996; Fuder and Selbach, 1993; Nicholas et al., 1993).
Neuropeptide Y (NPY) and adenosine triphosphate (ATP) could also be released from
sympathetic nerves by high frequency nerve stimulation (Bennett and Gardiner, 1996;
Fuder and Selbach, 1993; Kennedy et al., 1997; Nicholas et al., 1993). NPY has been
shown to regulate the cutaneous microcirculation in rat skin and acts as a neuromodulator
controlling the release and action of NA and ATP (Pinter et al., 1997).
Previous studies have shown that NA concentration is diminished in the skin of rats with
early diabetes (4 weeks) (Ahlgren and Levine, 1993). The effect of diabetes on the
vascular response to NA is strongly related to the type of tissue under investigation and
the duration of diabetes (chapter 1, sections 1.8.3 & 1.5.4.1). A previous study in
diabetic heart showed a reduction in the number of alpha-adrenoreceptors and a receptor
supersensitivity to alpha-agonists (Wald et al., 1988). The authors hypothesized that the
observed supersensitivity could be due to high activity of phospholipase C (PLC) (with
an increase in diacylglycerol (DAG) production). A reduction in the number of betaadrenoreceptors in this tissue was also observed in early diabetes (1-3 weeks) (Bitar and
DeSouza, 1990; Ramanadham and Tenner, 1987).
202
Both sensory and sympathetic nerves control microvascular blood flow in the skin. This
chapter is designed to investigate the role of sympathetic nerves in the modulation of skin
microvascular blood flow in early diabetic rats. It was hypothesized that the function of
sympathetic nerves may be altered in early diabetes, contributing to the reduced
microvascular responses observed in these animals. Changes in microvascular responses
to sodium nitroprusside (SNP), substance P (SP), and electrical stimulation (ES) of the
sciatic nerve are examined in diabetic rats (4 weeks) treated with a short-term
sympathetic neurone blocker guanethidine monosulfate (chapter 2, section 2.10).
5.2.
Method
Eight groups of rats (n = 4-10) were used, four control and four streptozotocin (STZ)treated diabetic groups. Control rats were treated with 0.1M sodium citrate buffer
solution (CB-treated) (chapter 2, section 2.4.1). On day 28, rats (diabetic or control) were
anaesthetized and a blister was induced on the hind footpad. Then rats were further
prepared depending on the type of experimental protocol (i.e. ES or peptide perfusion)
(chapter 2, sections 2.6 & 2.7). One hour before nerve stimulation, four groups of rats (2
control and 2 diabetic) were given a single injection of guanethidine monosulfate
(50mg/kg, i.p) (chapter 2, section 2.10) and others were non-treated (NT).
Animals
Diabetic
Control (CB-treated)
NT
Peptide ES
perfusion
NT
Guanethidine-treated
Peptide ES
perfusion
Peptide ES
perfusion
203
Guanethidine-treated
Peptide
perfusion
ES
In experiment 1 (peptide perfusion), SP (1μM, 30 min) and SNP (100μM, 10 min) were
perfused over the blister base and plasma extravasation (PE) was collected. In experiment
2 (ES), first SNP (100μM, 10 min) was perfused over the blister base and later sensory
nerve fibres were activated by a Grass stimulator (20V, 5Hz, 2 msec, for 1 minute)
(chapter 2, sections 2.6 & 2.7). At the end of each experiment rats were sacrificed with
pentobarbitone overdose.
5.2.1. Expression of data and statistical analyses
Results from vasodilatation responses were calculated as areas under the response curves.
PE responses are presented as percent above basal. All data are expressed as mean ±
standard error of the mean (SEM). The absolute values and 95% confidence intervals
(95% CI) of these values are quoted in the text.
A two-way Analysis of Variance (ANOVAs) was performed to investigate whether there
were significant differences in basal blood flow (BBF) and responses to SNP, SP, and
ES, between control and diabetic rats, based on whether or not these rats received
guanethidine treatment. Where ANOVA indicated a significant main effect, post hoc
independent sample t-tests were performed. A two-way repeated measure ANOVA with
priori planned comparisons was performed to investigate whether there were significant
differences in PE responses to SP at different times of perfusion (SP2,
3& 4),
between
control and diabetic rats, based on whether or not these rats received guanethidine
treatment.
5.3.
Results
Guanethidine treatment insignificantly increased the BBF in control rats [(24 ± 0.8 cm2)
(95%CI 22.4-25.6)] (111% of non-treated control) when compared to non-treated animals
[(21.6 ± 1.3 cm2) (95%CI 19.05-24.1)]. Guanethidine treatment did not change BBF in
diabetic rat [(17.4 ± 0.9 cm2) (95%CI 15.6-19.2)] when compared to non-treated diabetic
animals [(16.7 ± 1.3 cm2) (95%CI 14.1-19.2)] (Fig 5-1).
204
Area under the response curve (cm2 )
60
Non-treated BBF
Guanethidine-treated BBF
50
40
30
+
*
20
10
0
Control
Diabetic
Fig 5-1: BBFs in control and diabetic rats before and after guanethidine treatment (50mg/kg, i.p),
presented as mean ± SEM (n = 4-10).
* Denotes significance from control rats (P<0.05)
+ Denotes significance from guanethidine-treated control rats (P<0.05)
In control rats, there was no difference between guanethidine-treated SNP responses
[(25.1 ± 1.3 cm2) (95%CI 22.6-27.6)] and those of non-treated animals [(26.1 ± 1.8 cm2)
Area under the response curve (cm2 )
(95%CI 22.6-29.6)] (Fig 5-2).
60
Non-treated SNP
Guanethidine-treated SNP
50
40
30
X
20
+
*
10
0
Control
Diabetic
Fig 5-2: Vascular responses to SNP (100μM) in control and diabetic rats before and after guanethidine
treatment (50mg/kg, i.p), presented as mean ± SEM (n = 4-10).
* Denotes significance from control rats (P<0.05)
+ Denotes significance from guanethidine-treated control rats (P<0.05)
X Denotes significance from diabetic rats (P<0.05)
205
However, in diabetic rats, SNP responses in the treated groups were significantly higher
(182% of non-treated diabetic) [(13.5 ± 1 cm2) (95%CI 11.5-15.5)] than their non-treated
counterparts [(7.4 ± 0.6 cm2) (95%CI 6.2-8.6)].
Guanethidine treatment did not change the vascular response to SP in control animals
[(46.5 ± 3.2 cm2) (95%CI 40.2-52.8)] but improved the response to SP in diabetic rats to
185% of non-treated diabetic (Fig 5-3). This response reached (39.9 ± 5.6 cm2) (95%CI
28.9-50.9), which was statistically significant only from SP in non-treated diabetic
animals [(21.6 ± 3.4 cm2) (95%CI 14.9-28.3)].
Area under the response curve (cm2 )
60
Non-treated SP
Guanethidine-treated SP
50
X
40
30
*
20
10
0
Control
Diabetic
Fig 5-3: Vascular responses to SP (1μM) in control and diabetic rats before and after guanethidine
treatment (50mg/kg, i.p), presented as mean ± SEM (n = 4-10).
* Denotes significance from control rats (P<0.05)
X Denotes significance from diabetic rats (P<0.05)
206
The response to ES in control rats [(17.6 ± 1.3 cm2) (95%CI 15-20.1)] was increased
after treatment with guanethidine (119% of non-treated control) [(20.9 ± 1.1 cm2)
(95%CI 18.7-23)] but did not reach statistical significance (Fig 5-4). However, a far
greater improvement in ES response was observed in diabetic rats after guanethidine
treatment (212% of non-treated diabetic) [(10.8 ± 2.2 cm2) (95%CI 6.5-15.1)] when
compared to ES in non-treated diabetic rats [(5.1 ± 2 cm2) (95%CI 1.2-9)].
Area under the response curve (cm2 )
60
Non-treated ES
Guanethidine-treated ES
50
40
30
x
20
+
*
10
0
Control
Diabetic
Fig 5-4: Vascular responses to ES (20V, 5Hz, 2msec, for 1 minute) in control and diabetic rats before
and after guanethidine treatment (50mg/kg, i.p), presented as mean ± SEM (n = 4-8).
* Denotes significance from control rats (P<0.05)
+ Denotes significance from guanethidine-treated control rats (P<0.05)
X Denotes significance from diabetic rats (P<0.05)
Basal PE (R1) in control or diabetic rats was not changed after guanethidine treatment
(Appendix B, Table B-3). The PE response to SP (SP2,
3 & 4)
was calculated as percent
above basal ± SEM. SEMs were omitted from Fig 5-5 for clarity but quoted in the text.
The PE response to SP, in control rats (36 ± 19%, 72 ± 18%, 89 ± 14%), were
significantly different to diabetic rats (9 ± 23%, 16 ± 32%, 0.1 ± 23%). Guanethidine
treatment did not change the PE response in control rats (29 ± 10%, 44 ± 14%, 70 ±
207
26%). Although this treatment increased the PE response in diabetic rats (28 ± 19%, 39 ±
17%, 23 ± 5%), it did not reach statistical significance (Fig 5-5).
(%) Percent above basal
120
C Non-treated
C Guanethidine-treated
D Non-treated
D Guanethidine -treated
100
80
60
40
20
0
-20
R1-SP2
R1-SP3 R1-SP4
Fig 5-5: PE in control (C) and diabetic (D) rats before and after guanethidine treatment
(50mg/kg, i.p), presented as percent above basal (n = 4-10).
5.4.
Discussion
Sympathetic nerve fibres that innervate skin blood vessels can influence microvascular
blood flow via the release of sympathetic neurotransmitters like NA and NPY. A
reduction in rat hindquarter blood flow, after 6 days of diabetes, was suggested to be a
vasoconstrictor response and not due to impairment in endothelium-mediated
vasorelaxation or responsiveness to nitric oxide (NO) (Brands and Fitzgerald, 1998). The
results in chapter 3 showed a reduction in the BBF levels and the vascular inflammatory
responses in diabetic rats over time. In this chapter, possible contribution of sympathetic
efferents to these responses in early diabetes is examined.
The results showed that guanethidine treatment did not alter the BBF in 4 weeks diabetic
rats (Fig 5-1), suggesting that the observed reduction in BBF in diabetic rats (77% of
control) is not due to sympathetically mediated vasoconstriction. This raised the
possibility that in early diabetes, certain endothelial-mediated factors may play a role in
reducing BBF in skin microvascular bed. This possibility is further examined in chapter
6. The fact that guanethidine did not alter BBF in control or diabetic rats suggests that
208
sympathetic nerves play a minor role in regulating basal blood flow under the current
experimental conditions (anaesthesia and blister induction).
This study, however, demonstrates that sympathetically mediated vasoconstriction
contributes to the reduction in smooth muscle reactivity in 4 weeks diabetic rats (28% of
control) (Fig 5-2). The mechanisms underlying this sympathetic effect on smooth muscle
reactivity might be related to alterations in the NA level and the sensitivity of alpha-1,
alpha-2, and beta-2, adrenoreceptors located on smooth muscle cells. NA acts on both
alpha 1- and alpha-2 adrenoreceptors causing smooth muscle contraction. Low NA levels
in the skin tissue (Ahlgren and Levine, 1993) and reduced NA release from the spinal
cord were previously shown to occur in early diabetes (Bitar et al., 1999; Ohtani et al.,
1997). This could lead to the development of adrenoreceptor supersensitivity, causing
enhanced contraction in the smooth muscle by increasing intracellular calcium (Kamata
et al., 1988). Guanethidine treatment might prevent this contraction by inhibiting NA
release and activation of alpha-adrenoreceptors on the smooth muscle. NA also acts on
beta-2 adrenoreceptors to mediate vasodilatation. Beta-adrenoreceptor stimulation can
lead to smooth muscle relaxation by regulating the intracellular calcium concentration. It
was reported that the population size and the affinity of these receptors were reduced in
early diabetes (Bitar and DeSouza, 1990; Ramanadham and Tenner, 1987) (chapter 1,
section 1.8.3). This might lead to the observed reduction in smooth muscle relaxation in 4
weeks diabetic rats, which is in agreement with a previous study (Kamata et al., 1992). A
reduction in NA level, however, might also cause supersensitivity to develop in betaadrenoreceptors and this may reduce some of the contraction caused by alphaadrenoreceptors.
Guanethidine treatment did not affect SP response in control rats. However, in diabetic
rats, this treatment enhanced the vascular response to SP (185% of non-treated diabetic),
which could be due to improvement in smooth muscle reactivity (182% of non-treated
diabetic) (Figs 5-2 & 5-3).
209
The vascular response to ES in control rats showed a slight increase (119% of non-treated
control) after guanethidine treatment (Fig 5-4). This suggests that ES technique
(preparation of the sciatic nerve) could have resulted in some sympathetic nerve
activation that caused the reduction in this response. The vascular response to ES in
diabetic rats was significantly enhanced after guanethidine treatment. The enhancement
in ES in diabetic rats after this treatment (212% of non-treated diabetic) was greater than
that of control, which indicates that sympathetic nerve contribution in diabetic rats is
greater under this experimental condition. Part of the enhancement in ES response after
guanethidine treatment, in diabetic rats, could be explained by the improvement in SNP
response (182% of non-treated control) (post-terminal effect). As mentioned before,
adrenoreceptors in diabetic rats might become supersensitive when circulating NA is
reduced and this may explain the remaining 30% enhancement in the ES response after
the treatment (pre-terminal changes).
NPY and ATP release from sympathetic nerves can vary in proportion depending on the
tissue, the species, and on the parameter of nerve stimulation (Burnstock, 1986) and NPY
resting plasma levels are low in most species (Pernow, 1988). The relationship between
NPY and diabetes has been investigated mainly in the brain tissues (see also chapter 1,
section 1.8.3). It was reported that the distribution or relative density of NPY
immunoreactive nerve fibres was not changed after 2, 4, 8 or 12 weeks of diabetes in the
rat skin (Karanth et al., 1990).
As for ATP, evidence exists that in early diabetes there was no significant change in ATP
concentration or the release of ATP between control and diabetic rats (2 and 8 weeks
respectively) (Belai et al., 1991; Carlsson and Arnqvist, 1981). It was previously shown
that application of low frequency electrical nerve stimulation reduces the chance of NPY
/ ATP release from sympathetic nerves (Pernow, 1988). Therefore, the roles of NPY and
ATP in early diabetes were not examined in this thesis.
210
In conclusion, this chapter provides evidence for an increase in the contribution of
sympathetic efferents in regulating microvascular blood flow in early diabetes. An
increase in sympathetically mediated vasoconstrictor effect controlling smooth muscle
reactivity was also demonstrated. Finally, it was proposed that adrenoreceptor
supersensitivity might play a role in increasing smooth muscle contraction and reducing
the inflammatory response to ES of the sciatic nerve.
211
CHAPTER 6
Diabetes and modulation of microvascular blood flow:
Role of endothelial-derived constrictor factors and
Protein Kinase C
212
6.1.
Introduction
Role of Endothelins (ETs)
Endothelin-1 (ET-1) is a potent vasoconstrictor polypeptide produced by many cell types,
upon demand (e.g. in an emergency and / or in a defensive event) and its production is
regulated at the level of mRNA transcription and translation (Nakamura et al., 1990). ET1 has a wide range of biological actions but its physiological importance is still uncertain
(see also chapter 1, section 1.8.4 & Table 1-5).
Elevated level of glucose has been shown to modulate ET-1 synthesis and receptor
expression in cultured endothelial and vascular smooth muscle cells (chapter 1, section
1.8.4.1). Circulating (plasma) levels of ET-1 have been previously reported to increase
(2-10 fold) in diabetic patients (Battistini et al., 1993). However, in animals with type I
diabetes, there are conflicting reports regarding the plasma level of ET-1 (chapter 1,
section 1.8.4.1). This might be explained by the pre-existing duration of diabetes and the
vascular bed under examination (Hopfner et al., 1999).
It was shown that high glucose levels could also modulate endothelium-derived relaxing
factor (EDRF) production and action (Hopfner and Gopalakrishnan, 1999). The authors
suggested that formation of substances such as advanced glycated end-products (AGEs)
and free radicals, by damaged endothelium, could be involved in inhibiting EDRF and
regulating ET-1 production and action (chapter 1, section 1.8.4.1, Fig 1-10). It was
proposed that under pathological conditions, ET-1 might also inhibit EDRF (Barton et al.,
1998). Furthermore, insulin appears to be a modulator of ET-1, ET receptors, and ET-1
mediated responses in both in vitro and in vivo settings (Hopfner and Gopalakrishnan,
1999). The role of insulin in diabetes is discussed in chapter 9.
ET-1 mediates its biological responses mainly by binding to ET-A and ET-B receptors
(chapter 1, section 1.8.4 & Fig 1-8). Under normal conditions, the in vivo response to ET1 varies depending on the vascular bed and the expression / density of ET-A and ET-B
receptors (Hopfner and Gopalakrishnan, 1999). Various pathological conditions,
including diabetes, can modify the density of these receptors (Nayler, 1990).
The binding of ET to its receptors activates a variety of signal transduction processes but
the activation of phospholipase C (PLC) and the subsequent release of intracellular
213
calcium appears to be the major signal transduction pathway involved (Reid and La,
1995) (see also chapter 1, section 1.8.4, Fig 1-9).
There is conflicting evidence regarding the function of ET in microvascular permeability
and it seems that the route of administration and the species involved are important
factors (Brain et al., 1989; Lawrence et al., 1995).
The results in chapter 3 and 4 led to the hypothesis that the presence of certain substances
may be responsible for the observed reduction in microvascular inflammatory responses
in rats with early diabetes. The literature has associated a link between ETs and vascular
complications in certain disorders such as diabetes (Battistini et al., 1993; Haak et al.,
1992; Takahashi et al., 1990). A few studies have used skin to assess the haemodynamic
action of ET-1 in rats (Brain et al., 1989; Chander et al., 1988; Lawrence et al., 1995) but
none to date have investigated the role of ET-1 in the modulation of skin microvascular
blood flow in early diabetes.
This chapter is designed to determine the quantitative contribution of ET-A and ET-B
receptors to the observed reduction in microvascular responses in 4 weeks streptozotocin
(STZ)-treated diabetic rats. It was hypothesized that ET-1 may be involved in reducing
microvascular blood flow in early diabetes and its effects may be inhibited by early
intervention. BQ-123 (a specific ET-A receptor antagonist) and BQ-788 (a specific ET-B
receptor antagonist) are used to examine the role of these receptors in a) modulating
endothelial and smooth muscle cell mediated vascular responses and b) modulating the
inflammatory response to ES of the sciatic nerve in diabetic rats.
Role of protein kinase C (PKC)
PKC, a family of kinases, play a major role in intracellular signalling. High extracellular
glucose activates PKC but the identity of the isoforms involved and their location in the
cell where they operate is unclear (chapter 1, section 1.8.5.4). Evidence exists that a
number of vascular alterations in diabetes are caused by activation of PKC, including a
decrease in the activity of Na+/K+-ATPase and an increase in extracellular matrix,
cytokines, permeability, contractility, and cell proliferation (Park et al., 1999).
214
Enhanced PKC activity by hyperglycaemia might be tissue specific. Enhanced PKC
activity was previously noted in the heart (Inoguchi et al., 1992; Liu et al., 1999) and
glomeruli (Awazu et al., 1991) but may not occur in the brain and peripheral nerves
(Koya and King, 1998). Enhancement in PKC activity was attributed to an increase in
diacylglycerol (DAG), a major endogenous activator of PKC (Ishii et al., 1998).
Recently, it was shown that enhanced ET-1 expression in diabetes was partially due to
PKC activity (Park et al., 2000). Under physiological conditions in human endothelial
cells, ET-1 induces endothelial NOS (eNOS) gene expression by activating ET-A
receptors and PKC regulates this induction (Marsen et al., 1999). This shows that
regulation of eNOS at the genomic level occurs via post-transcriptional mechanisms.
The role of PKC in vascular alterations associated with diabetes has been shown (chapter
1, section 1.8.4). However, the role of this enzyme in modulating skin microvascular
blood flow in early diabetes has not been previously investigated. In this chapter, the role
of PKC in STZ-induced diabetic rat model is examined using bisindolylmaleimide
(BIM), a specific PKC inhibitor.
6.2.
Method
Ten groups of rats (n = 4-12) were used in this study, five of which were made diabetic
with STZ and the other five were used as controls. Control rats were injected with 0.1M
sodium citrate buffer solution (CB-treated) (chapter 2, section 2.4.1).
Animals
Control (CB-treated)
BIM
peptide
perfusion
BQ-123
BQ-788
peptide perfusion
& ES
Diabetic
BQ-123
BQ-788
peptide perfusion
& ES
215
BIM
peptide
perfusion
Control and diabetic rats were then prepared for experiment 1 (peptide perfusion) alone
or together with experiment 2 (electrical stimulation (ES) of the sciatic nerve). In
experiment 1, after the baseline was established, one of the following substances, i.e. BQ123, BQ-788 (both at 10 μM), or BIM (1μM) was perfused for 10 minutes prior to, and
together with, substance P (SP, 1μM, 30 min). This was followed by separate perfusion
of Ringer’s solution (20 minutes) and sodium nitroprusside (SNP, 100μM, 10 min) over
the blister base respectively. In experiment 2, first SNP was perfused for 10 minutes
followed by Ringer’s solution. Then either BQ-123 or BQ-788 was perfused for 10
minutes followed by electrical stimulation of the sciatic nerve. Perfusion of the antagonist
was continued during the stimulation period and for 20 min after that. Plasma
extravasation (PE) was collected from all peptide perfusion experiments. At the end of
each experiment rats were sacrificed with a pentobarbitone overdose.
6.2.1. Expression of data and statistical analyses
Vasodilatation responses were calculated as areas under the response curves. PE
responses are presented as percent above basal. All data are expressed as mean ± standard
error of the mean (SEM). The absolute values and 95% confidence intervals (95% CI) of
these values are quoted in the text.
Two-way Analyses of Variances (ANOVAs) were performed to investigate whether there
were significant differences in basal blood flow (BBF) and responses to SNP, SP, and
ES, between control and diabetic rats, according to whether rats were exposed to BQ-123
or BQ-788. This test could also show interaction effects between these groups. Where
ANOVA indicated a significant main effect, post hoc independent sample t-tests were
performed. The same analysis and post hoc comparisons were also performed to
investigate whether there were significant differences in BBF and responses to SNP and
SP, between control and diabetic rats, based on whether rats were exposed to BIM or not.
A two-way repeated measure ANOVA with priori planned comparisons was performed
to investigate whether there were significant differences in PE responses between control
and diabetic rats before and after BQ-123, BQ-788, and BIM perfusions.
216
6.3.
Results
Effect of endothelin antagonists on BBF, SNP, and SP
The BBF in diabetic rats [(16.7 ± 1.3 cm2) (95% CI 14.1-19.2)] (measured for 20
minutes), was significantly lower than control animals [(25.1 ± 0.9 cm2) (95% CI 23.326.9)] (Fig 6-1 & 6-2).
Area under the response curve (cm2 )
60
Control BBF
Control BBF + BQ-123
Control BBF + BQ-788
50
40
30
20
10
0
Fig 6-1: BBF in control rats before and after BQ-123 and BQ-788 (both at 10μM), presented as mean ±
SEM (n = 4-9).
Area under the response curve (cm2)
60
Diab BBF
Diab BBF + BQ-123
Diab BBF + BQ-788
50
40
30
20
10
0
Fig 6-2: BBF in diabetic (Diab) rats before and after BQ-123 and BQ-788 (both at 10μM), presented as
mean ± SEM (n = 4-10).
217
Perfusion of BQ-123 or BQ-788 did not significantly change the BBF in control rats (Fig
6-1). However, the BBF in diabetic rats was improved after BQ-123 (20.8 ± 2.1 cm2)
(95% CI 16.7-24.9), reaching 83% of control (Fig 6-2). This response was not altered by
BQ-788 perfusion [(16 ± 0.6 cm2) (95% CI 14.8-17.2)]. The improvement in diabetic
BBF after BQ-123 perfusion did not reach statistical significance.
In control rats, BQ-123 and BQ-788 perfusions did not have any significant effect on
Area under the response curve (cm2 )
vascular responses to SNP and SP (Figs 6-3 & 6-4).
60
50
Control SNP
Control SNP + BQ-123
Control SNP + BQ-788
40
30
20
10
0
Fig 6-3: The vascular response to SNP (100μM) in control rats before and after BQ-123 and BQ-788
(both at 10μM), presented as mean ± SEM (n = 8-12).
Area under the response curve (cm2)
60
50
Control SP
Control SP + BQ-123
Control SP + BQ-788
40
30
20
10
0
Fig 6-4: The vascular responses to SP (1μM) in control rats before and after BQ-123 and BQ-788 (both
at 10μM), presented as mean ± SEM (n = 4-9).
218
On the other hand, the SNP response in diabetic rats was equally and significantly
improved after both BQ-123 [(17.9 ± 2.1 cm2) (95% CI 13.8-22)] and BQ-788 [(18.8 ±
1.1 cm2) (95% CI 16.6-20.9)] perfusions, reaching approximately 70% of control value
Area under the response curve (cm2 )
[(26.1± 1.8 cm2) (95% CI 22.6-29.6)] (Fig 6-5).
60
Diab SNP
Diab SNP + BQ-123
Diab SNP + BQ-788
50
40
30
20
*
*
10
0
Fig 6-5: The vascular response to SNP (100μM) in diabetic (Diab) rats before and after BQ-123 and
BQ-788 (both at 10μM), presented as mean ± SEM (n = 8-10).
* Denotes significance from diabetic rats (P<0.05)
The vascular response to SP in diabetic rats was partially improved after perfusing BQ123 [(32.4 ± 2.7 cm2) (95%CI 27.1-37.7)]. Although this response reached 63% of the
control SP response [(51.4 ± 2.9 cm2) (95%CI 45.7-57.1)], it did not reach statistical
significance (Fig 6-6).
219
Area under the response curve (cm2)
60
50
*
Diab SP
Diab SP + BQ-123
Diab SP + BQ-788
40
30
20
10
0
Fig 6-6: The vascular response to SP (1μM) in diabetic (Diab) rats before and after BQ-123 and BQ-788
(both at 10μM), presented as mean ± SEM (n = 4-10).
* Denotes significance from diabetic rats (P<0.05)
On the other hand, BQ-788 was more potent [(44.4 ± 4.1 cm2) (95%CI 36.4-52.4)] and
improved this response to 86% of control.
Effect of endothelin antagonists on ES
In control rats, the response to ES was significantly enhanced after BQ-123 perfusion
[(28.5 ± 2.5 cm2) (95% CI 23.6-33.4)], reaching 162% of control response (Fig 6-7). This
response was also increased after BQ-788 perfusion [(27.8 ± 5 cm2) (95% CI 18-37.6)]
(158% of control) but did not reach statistical significance.
In diabetic rats, BQ-123 perfusion partially improved the vascular response to ES [(9.6 ±
2.2 cm2) (95%CI 5.3-13.9)], reaching approximately 55% of control response [(17.6 ± 1.3
cm2) (95%CI 15.1-20.1)] (Fig 6-8). BQ-788, on the other hand, proved to be more potent
than BQ-123 and significantly improved the ES response in diabetic animals to 84% of
its control value [(14.8 ± 0.9 cm2) (95%CI 13-16.6)].
220
Area under the response curve (cm2 )
60
Control ES
Control ES + BQ-123
Control ES + BQ-788
50
40
*
30
20
10
0
Fig 6-7: The vascular response to ES (20V, 5Hz, 2msec, for I minute) in control rats before and after
BQ-123 and BQ-788 (both at 10μM), presented as mean ± SEM (n = 8-12).
* Denotes significance from control rats (P<0.05)
Area under the response curve (cm2 )
60
Diab ES
Diab ES + BQ-123
Diab ES + BQ-788
50
40
30
20
*
10
0
Fig 6-8: The vascular response to ES (20V, 5Hz, 2msec, for I minute) in diabetic (Diab) rats before and
after BQ-123 and BQ-788 (both at 10μM), presented as mean ± SEM (n = 8-10).
* Denotes significance from diabetic rats (P<0.05)
221
Effect of PKC inhibitor on BBF, SNP, and SP
BIM perfusion did not change the BBF in control [(26 ± 0.9 cm2) (95% CI 24.2-27.8)] or
Area under the response curve (cm2 )
diabetic rats [(15.6 ± 0.7 cm2) (95% CI 14.2-17)] (Fig 6-9).
60
Control BBF
Control BBF + BIM
Diab BBF
Diab BBF + BIM
50
40
30
20
*
*
10
0
Fig 6-9: BBFs in control, control after BIM, diabetic (Diab), and diabetic rats after BIM (1μM),
presented as mean ± SEM (n = 5). * Denotes significance from control rats (P<0.05)
BIM perfusion did not alter the vasodilator response to SNP in control rats [(26.8 ± 1.8
cm2) (95%CI 23.3-30.3)] but significantly enhanced this response in diabetic rats [(14.7 ±
2 cm2) (95%CI 10.8-18.6)], reaching 56% of control response (Fig 6-10). BIM perfusion
did not have an effect on the vascular response to SP in control [(50 ± 3.2 cm2) (95%CI
56.3-43.7)] or diabetic rats [(21.7 ± 1.1 cm2) (95%CI 19.5-23.8)] (Fig 6-11).
222
Area under the response curve (cm2)
60
Control SNP
Control SNP + BIM
Diab SNP
Diab SNP + BIM
50
40
30
X
20
*
*
10
0
Fig 6-10: Vascular responses to SNP (100μM) in control, control after BIM, diabetic (Diab), and
Area under the response curve (cm2 )
diabetic rat after BIM (1μM), presented as mean ± SEM (n = 5).
* Denotes significance from control rats (P<0.05)
X Denotes significance from diabetic rats (P<0.05)
60
Control SP
Control SP + BIM
Diab SP
Diab SP + BIM
50
40
30
*
*
20
10
0
Fig 6-11: Vascular responses to SP (1μM) in control, control after BIM, diabetic (Diab), and diabetic
rats after BIM (1μM), presented as mean ± SEM (n = 5).
* Denotes significance from control rats (P<0.05)
Effect of endothelin antagonists and PKC inhibitor on PE
PE in control rats was not altered after BQ-123 or BQ-788 perfusions (Appendix B,
Table B-4). However, in diabetic rats, both receptor antagonists reduced basal PE (R1)
but only BQ-123 reduced this value to near normal control level. BQ-123 also improved
the ability of SP to induce a PE response.
223
The PE response to SP (SP2, 3 & 4) was calculated as percent above basal ± SEM. SEMs
were omitted from Fig 6-12 for clarity but quoted in the text. BQ-123 perfusion increased
the PE response to SP in diabetic rats by 43 ± 15%, 75 ± 13%, and 66 ± 11%. BQ-788
also increased the above response by 28 ± 12%, 74 ± 25%, and 93 ± 27% respectively
(Fig 6-12).
BIM perfusion did not change R1 in diabetic rats (Appendix B, Table B-4). However, it
improved the PE response to SP only during the last 20 minutes of the response (5 ±
14%, 27 ± 17%, 34 ± 14%), when compared to non-perfused diabetic animals (9 ± 23%,
16 ± 32%, 0.1 ± 23%). However, these responses did not reach the control value (36 ±
19%, 72 ± 18%, 89 ± 14%) (Fig 6-12).
(%) Percent above basal
120
Control
Di abetic
Di abetic + BQ-123
Di abetic + BQ-788
Di abetic + BIM
100
80
60
40
20
0
-20
R1-SP2 R1-SP3 R1-SP4
Fig 6-12: PE in control, diabetic, and diabetic rats before and after BQ-123, BQ-788 (both
at 10μM), and BIM (1μM), presented as percent above basal (n = 4-10).
6.4.
Discussion
Related to ETs
Previous studies have shown that changes in plasma glucose level in the diabetic
condition is directly involved in regulation of ET-1 release, modulation of receptor
expression, and response to the peptide (chapter 1, section 1.8.4.1, Figs 1-10 & section
1.8.5.3, Fig 1-14). Many studies have proposed that potent vasoconstrictors such as ET-1
224
could be involved in the development of vascular complications associated with diabetes
(chapter 1, section 1.8.4).
In this chapter, the relative contribution of ET receptors to reduced skin microvascular
blood flow in 4 weeks diabetic rats was examined, an issue which has not been
previously investigated.
In diabetic rats, BBF levels did not change after BQ-788 perfusion but were partially
increased after BQ-123 perfusion (83% of control) (Fig 6-2). The latter result suggests
that endothelial mediated factors may be involved in reducing BBF in these animals. This
notion was further supported in chapter 5 by showing that the reduction in diabetic BBF
is not related to sympathetic vasoconstriction.
The reduced vascular response to SNP, a NO donor, in diabetic rats suggests that NO has
a lower efficiency to induce vasodilatation (Fig 6-5). This could be due to deactivation of
NO or inefficiency in the second messenger (cyclic guanosine monophosphate (cGMP))
pathway. ET could be a possible mediator for both of these actions and the results in this
chapter support this proposition. The data showed that both BQ-123, an ET-A receptor
antagonist and BQ-788, an ET-B receptor antagonist, significantly improved smooth
muscle-dependent vasodilatation in diabetic rats to about 70% of control values (Fig 6-5).
It is therefore reasonable to postulate that apart from NO deactivation, ET may either
inhibit guanylate cyclase, thus reducing cGMP, or activate PKC (Barton et al., 1998;
Khalil et al., 1996). The possible role of free radicals in deactivating NO donated by SNP
cannot be excluded.
There is evidence that both ET receptors (ET-A & ET-B) exist in the rat skin (Lawrence
et al., 1995). With respect to their activities, a number of studies in vivo demonstrated
that both receptors were capable of mediating vascular smooth muscle contraction (Soe et
al., 1994) but the relative contribution of each receptor is dependent on both the species
and the vascular bed under study. In this study, both BQ-123 and BQ-788 perfusions
increased the vascular response to SNP in 4 weeks diabetic rats (Fig 6-5). The increase in
SNP response after BQ-123, in 4 weeks diabetic rats, suggests that ET-A receptors may
be up regulated, contributing to the reduction in SNP responses. A previous study is
225
supportive of this proposition (Deng et al., 1999). On the other hand, it was suggested
that increase in ET-1 could down-regulate ET-B receptors (Masao et al., 1999).
Reduction in vasodilator responses mediated by ET-B receptors may exacerbate the
observed vasoconstriction mediated by ET-A receptors in diabetic rats.
The current data suggests equal involvement of both ET-A and ET-B receptors in
inhibiting smooth muscle-dependent vasodilation in diabetic rats at the level of skin
microvasculature. In contrast, ET-B receptor activation plays a greater role in inhibiting
the vascular responses to SP and ES of the sciatic nerve in these animals. Both SP and ES
responses in diabetic rats reached approximately 85% of their own control responses after
BQ-788 perfusion, compared to 63% and 55% of their own controls after BQ-123 (Figs
6-6 & 6-8). This might be related to an alteration in ET-B receptor activity after 4 weeks
of diabetes. It should be noted here that whether SP is perfused exogenously over the
blister base or released endogenously upon ES of the sciatic nerve, it induces the release
of NO from endothelial cells via activating neurokinin-1 (NK-1) receptors (Andrews et
al., 1989). In this experiment, ET may inhibit both the action and release of NO from
endothelial cells. The greater involvement of ET-B receptors in the vascular responses to
SP and ES raised the possibility that activation of ET-B receptors in 4 weeks diabetic rats
can play a significant role in inhibiting NO release from endothelial cells. In diabetes,
there is a relationship between hyperglycaemia and / or a low concentration of insulin and
an increase in ET levels, which in turn could be involved in inhibition of NO release from
endothelial cells (chapter 1, sections 1.8.4.1).
Glucose
NO
Feed back
reaction
226
ET
ET may also indirectly damage endothelium over time, leading to a further increase in
ET-1 production (Huszka et al., 1997; Takahashi et al., 1990; Vermes et al., 1993).
Hence, it is possible that a feed back mechanism is involved and an increase in ET-1 may
further inhibit NO release and cause an increase in production of ET-1 in turn. This was
supported by a previous study showing that, in STZ-rats, changes in response to ET-1 and
its plasma concentrations are duration dependent and occur in parallel with changes in
endothelial function (Hopfner et al., 1999).
The other possibility is that a reduction or lack of insulin may affect the production and
release of NO in early diabetes leading to an increase in ET-1 production, which again
can inhibit NO release by a possible feed back mechanism (chapter 1, section 1.8.5.3).
Previous studies demonstrating the relationship between insulin, NO, and ET under
physiological conditions support this hypothesis. One of these studies showed that insulin
could affect ET levels by direct stimulation of intact endothelial cells (Hu et al., 1993). In
diabetes, damage to endothelial cells may reduce the production of ET by this direct
action. Others demonstrated that insulin could also increase the release of NO (Steinberg
et al., 1994) under physiological conditions leading to a decrease in ET production
(Kourembanas et al., 1993). In diabetes, this action of insulin might be altered.
A previous study, using an atherosclerosis model, suggested that ET-1 may inhibit NO by
interacting with ET-A receptors (Barton et al., 1998). In the current study, however, the
results raise the possibility that in 4 weeks diabetic rats, ET-1 may inhibit NO release via
an ET-B mediated mechanism (chapter 1, section 1.8.4 & Fig 1-8). The greater
effectiveness of BQ-788 in regaining 85% of the inflammatory response to SP and ES as
opposed to regaining 70% of the response to SNP raised the possibility that ET-B
receptors may be involved in inhibiting NO release.
In diabetic rats, the response to SP increased from 42% to 63% of control after BQ-123
perfusion (Figs 6-6). This increase could be totally due to improvement in smooth muscle
reactivity (68% of control) (Fig 6-5). The increase in diabetic SP response after BQ-788
(86% of control) is partially due to improvement in smooth muscle reactivity in these rats
(72% of control). The remaining 14% increase in SP response in diabetic rats may be
related to a mechanism involving NO release.
227
In control rats, both ET receptor antagonists did not alter BBF (Fig 6-1) or responses to
SNP (Fig 6-3) or SP (Fig 6-4). However, a significant increase in response to ES (by
62%) after BQ-123 perfusion was previously observed without altering resting blood
flow (Khalil et al., 1996), supporting the current results (Figs 6-1 & 6-7). The data,
therefore, suggest that under control conditions, preparation of the sciatic nerve for ES
experiment involves the release of ETs. That BQ-788 perfusion in diabetic rats increased
the vascular response to ES by 88%, versus 62% in control rats, suggests a greater
involvement of ET-B receptors in inhibiting the ES response in diabetic rats. ET-B
antagonist did not alter any of the inflammatory responses in control animals, suggesting
that its involvement in these responses in diabetic rats could not be related to the
experimental procedure used.
The results showed BQ-123 and BQ-788 perfusions to be equally potent in normalizing
basal PE (R1) in diabetic rats and improving the ability of SP to induce a PE response
(Fig 6-12) (Appendix B, Table B-4). The normalization of R1, in particular, suggests that
permeability changes in endothelial cells in diabetes are a function of ET-1 acting on both
ET-A and ET-B receptors and are reversible up to 4 weeks after induction of diabetes
(experimental period). That the PE response to SP was also increased after perfusion of
BQ-123 and BQ-788 suggests that the effect of ET is not permanent, as blocking both ET
receptors resulted in restoration of endothelial cell permeability to near normal levels
(when calculated as percent above basal).
Related to PKC
Alterations in PKC activity have been associated with abnormalities in the eye, kidney
and, nerves of diabetic animals and patients. Therefore, the role of PKC in modulating
skin microvascular blood flow in early diabetes was investigated.
Perfusion of BIM over the blister base of control or diabetic rats did not have any effect
on BBF (Fig 6-9). The lack of BIM effect on BBF in 4 weeks diabetic rats suggests that
PKC does not play a role in modulating skin BBF in these animals.
228
BIM perfusion did not have an effect on the vasodilator response to SNP in control rats
but improved this response in 4 weeks diabetic rats from 28% to 56% of control (Fig 610). This improvement in smooth muscle reactivity supports the hypothesis that free
radicals or ET may alter the activity of PKC in early diabetes and contribute to vascular
dysfunction (Ha and Kim, 1999; Konishi et al., 1997). The current results suggest that
reduction in smooth muscle reactivity in diabetic rats may be partly related to a
temporary and reversible effect of ET and PKC on skin microvasculature.
In diabetes there is an established relationship between ET-1, PKC activity, and changes
in intracellular calcium concentration. It was suggested that activation of PKC by
hyperglycaemia could also enhance ET-1 (Park et al., 2000), which can lead to alterations
in the level of intracellular calcium (Masao et al., 1999). Hence, it is possible that
increase in PKC activity in 4 weeks diabetic rats could have increased extracellular
calcium influx and caused vascular contraction (Abebe and MacLeod, 1990; Legan,
1989) (see also chapter 1, section 1.5.4.1). This effect might have been reduced by BIM
and resulted in improvement in SNP response (Fig 6-10).
It was previously reported that enhanced PKC activity in diabetes might be tissue
specific. BIM enhanced the SNP response in diabetic rats to 56% of control (Fig 6-10).
The lack of improvement in microvascular response to SP after BIM perfusion was
somewhat unexpected, particularly in view of the significant improvement in smooth
muscle reactivity (Fig 6-11). An independent smooth muscle reactivity response clearly
showed that PKC is involved. However, this effect was masked when the response to SP
was examined, indicating that the involvement of other factors predominate under the
diabetic condition. The exact role of PKC in reducing microvascular blood flow in early
diabetes cannot be determined from the experiments undertaken in this chapter and
therefore requires further examination.
PKC is also known to be capable of regulating vascular permeability under the diabetic
condition or in the presence of high glucose (Wolf et al., 1991) (chapter 1, section
1.8.5.4). The current results showed that R1 in diabetic rats after BIM perfusion was not
different from those of non-treated diabetic rats (Fig 6-12). The improvement in the PE
229
response to SP in these rats after BIM perfusion did not reach statistical significance.
This indicates that the temporary effects of PKC do not contribute to disruption in
endothelial cell function.
In conclusion, evidence was provided for the involvement of ET in modulating peripheral
endothelial and smooth muscle-dependent vasodilatation in 4 weeks diabetic rats. The
data suggests that ET plays a role in endothelial cell permeability changes in diabetes. It
was demonstrated that PKC does not contribute to disruption in endothelial cell function
in 4 weeks diabetic rats. It was also shown that there are subtle differences in the relative
contribution of the two ET receptor subtypes (ET-A & ET-B) to early changes in diabetic
vascular responses. It was demonstrated that local pharmacological manipulation at the
level of skin microcirculation, using ET receptor antagonists (BQ-123 & BQ-788), could
normalize these early changes, which could be clinically significant.
Publication: Bassirat, M. and Khalil Z. Endothelin and free radicals modulate
microvascular responses in streptozotocin-induced diabetic rats. Microvascular Research,
59:88-98, March 2000.
230
CHAPTER 7
Diabetes and modulation of microvascular blood flow:
Role of reactive oxygen species (ROS)
231
7.1.
Introduction
Reactive oxygen species (ROS) are generated naturally in living organisms and they have
potential to produce high levels of tissue damage (chapter 1, section 1.5.3 & Fig 1-4).
Endothelial-generated ROS (Bonnardel-Phu and Vicaut, 2000a) could modify vascular
endothelium by direct and indirect means. These include interference with nitric oxide
(NO) availability (Jourd'heuil et al., 1997; Munzel et al., 1997), membrane lipid
peroxidation (Hayakawa and Raij, 1999), and formation of advanced glycated endproducts (AGEs) (Stehouwer et al., 1997; Schiekofer et al., 2000) (see also chapter 1,
section 1.5.3).
ROS are important in physiological activities of smooth muscles and in the pathogenesis
of various diseases where the function of smooth muscle is altered (Bauer et al., 1999).
Previous studies showed that ROS can contribute to cell dysfunction by interfering with
calcium homeostasis (Dreher and Junod, 1995; Suzuki et al., 1991). For example, it was
demonstrated that hydroxyl radicals are capable of potentiating the voltage-dependent
influx of calcium (Chang et al., 1993). Lipid peroxidation by-products, on the other hand,
may interfere with calcium homeostasis by inhibiting key enzymes involved in calcium
mobilization (Wells et al., 1997). A different study showed that superoxide anions could
contribute to impaired endothelial function by modifying calcium channels (Chang et al.,
1993).
The body’s natural antioxidants (e.g. superoxide dismutase (SOD), catalase, glutathione
redox enzymes) are capable of neutralizing oxygen free radicals and similar agents
(Thomas, 1995). An acute increase in plasma glucose concentration can enhance free
radical production (Mullarkey et al., 1990; Traverso et al., 1998) and attenuate
antioxidant enzyme activities (Kamata and Kobayashi, 1996; Mak et al., 1996). It was
suggested that alterations in free radical / antioxidant balance might be responsible for
altered endothelial function and other complications of diabetes (Giugliano et al., 1996;
Stehouwer et al., 1997). Also, in diabetes, the efficiency of the body’s antioxidant
system, may not only be related to the duration of this disease but also to some posttranslational modifications (Kakkar et al., 1996). Hyperglycaemia or insulin deficiency
232
was also suggested to be partially responsible for an increase in endothelin-1 (ET-1)
production in diabetic rats, via a ROS-mediated mechanism (Chen et al., 2000).
N-acetylcysteine (NAC), a powerful hydroxyl radical scavenger (Aruoma et al., 1989),
has been widely used in in vitro and in vivo studies (chapter 1, section 1.5.3). Pretreatment with NAC was shown to protect sensory nerves (Sagara et al., 1996) and
prevent endothelial dysfunction (Pieper and Siebeneich, 1998) in streptozotocin (STZ)induced diabetic rats. Tirilazad mesylate (Freedox), a clinically proven antioxidant,
prevents lipid peroxidation and was shown to be cytoprotective by preserving post-injury
calcium homeostasis (Saniova, 1997).
No studies to date have investigated the role of free radicals / ROS in modulating skin
microvascular blood flow in early diabetes. This chapter examines the quantitative
contribution of superoxide anions and hydroxyl radicals and the effect of lipid
peroxidation to the documented (see results in chapter 3) reduced microvascular
responses in 4 weeks diabetic rats. It was hypothesized that free radicals may be involved
in reducing microvascular blood flow in early diabetes and their effects may be inhibited
by early intervention. The role of ROS in modulating endothelial (inflammatory) and
smooth muscle vascular responses are examined by using three antioxidants: SOD, NAC,
and tirilazad mesylate. In addition, the proposal by Chen et al 2000 that hyperglycaemia
or insulin deficiency could increase ET-1 production in diabetic rats via a ROS-mediated
mechanism, raised the possibility of interaction between ET and free radicals. This
possibility is also examined using the combined treatment of BQ-123, an ET-1
antagonist, and SOD, a superoxide anion scavenger.
7.2.
Method
A total of 18 groups of rats (n = 4-10) were used in this study, 8 controls and 10 diabetic
rats. Control rats were injected with 0.1M sodium citrate buffer solution (CB-treated)
(chapter 2, section 2.4.1). Among diabetic rats, 2 groups were pre-treated with tirilazad
mesylate (5mg/kg, i.p., every 2nd day) and 2 groups were left untreated.
In peptide perfusion experiments, once the base line was established, first substance P
(SP, 1μm, 30 min), then sodium nitroprusside (SNP, 100μM, 10 min), and finally CGRP
233
(1μm, 10 min) were perfused over the blister base. In ES experiments, after the base line
was established, SNP was perfused followed by Ringer’s solution and stimulation of the
sciatic nerve (20V, 5Hz, 2msec for 1 minute).
The peptide perfusion and ES experiments were performed on 2 control groups, 2
diabetic groups, and 2 tirilazad mesylate-treated diabetic groups (see diagram below).
Other groups of rats were first equally divided between control and diabetics. Then in
separate experiments, after the base line was established, SOD (100U/ml), NAC (100μM)
or combined BQ-123 (10μM) + SOD (100U/ml) was perfused over the blister base of rats
for 10 minutes prior to, and then together with, the inflammatory stimulus (i.e. ES or SP
& SNP perfusion). Plasma extravasation (PE) was collected from all perfusion
experiments. At the end of each experiment, rats were sacrificed with pentobarbitone
overdose.
Animals
Animals
Control (CB-treated)
Tirilazad
mesylate
SOD NAC SOD+BQ-123
Peptide
(SP, SNP)
Perfusion Perfusion
& ES
& ES
Diabetic
SOD NAC
SOD+BQ-123
Peptide
Perfusion
& ES
(SP & SNP)
perfusion
& ES
7.2.1. Expression of data and statistical analyses
Results from vasodilatation responses were calculated as areas under the response curves.
PE responses are presented as percent above basal. All data are expressed as mean ±
standard error of the mean (SEM). The absolute values and 95% confidence intervals
(95% CI) of these values are quoted in the text.
Two-way Analyses of Variances (ANOVAs) were performed to investigate whether there
were significant differences in basal blood flow (BBF) and responses to SNP, SP, and
ES, between control and diabetic rats, according to whether the rats were exposed to or
234
treated with any of the following drugs (SOD, NAC, Tirilazad mesylate, BQ-123 +
SOD). This test also shows interaction effects between these groups. Where ANOVA
indicated a significant main effect, post hoc comparisons were conducted using the
Tukey HSD test. A two-way repeated measure ANOVA with priori planned comparisons
was performed to investigate whether there were significant differences in PE responses,
between control and diabetic rats, according to whether the rats were exposed to or
treated with any of the following drugs (SOD, NAC, Tirilazad mesylate, BQ-123 +
SOD). To examine other results involving control and diabetic rats, separate independent
sample t-tests were performed.
7.3.
Results
The BBF in control rats [(25.1 ± 0.9 cm2) (95% CI 23.3-26.9)] did not significantly
change after perfusing SOD, NAC or the combined SOD + BQ-123 (Fig 7-1). The results
in chapter 6 also showed no change in control BBF after perfusing BQ-123 alone
Area under the response curve (cm2 )
(Chapter 6, Fig 6-1).
60
50
Control
Control
Control
Control
BBF
BBF + SOD
BBF + NAC
BBF + BQ123+SOD
40
30
20
10
0
Fig 7-1: BBFs in control rat and control rats after SOD (100U/ml), NAC (100μM), and combined
[BQ-123 (10μM) + SOD (100U/ml)], presented as mean ± SEM (n = 4-10).
In diabetic rats, some improvements in BBF were observed following NAC perfusion
[(19.9 ± 1.9 cm2) (95% CI 16.1-23.6)], tirilazad mesylate treatment [(18.3 ± 1.3 cm2)
(95% CI 15.7-20.8)], and the combined BQ-123 + SOD perfusion [(19.3 ± 1.4 cm2) (95%
CI 16.6-22)] (between 73-79% of control) but not after perfusing SOD [(16.6 ± 1.2 cm2)
235
(95% CI 14.2-18.9)] (Fig 7-2). In chapter 6, a partial increase in diabetic BBF was also
shown after BQ-123 perfusion [(20.8 ± 2.1 cm2) (95% CI 16.7-24.9)] (83% of control)
(Chapter 6, Fig 6-2). The observed increase in diabetic BBF after perfusion or treatment
with the above drugs did not reach statistical significance.
Area under the response curve (cm2 )
60
50
40
Diab BBF
Diab BBF + SOD
Diab BBF + NAC
Diab BBF + Freedox
Diab BBF + BQ123+SOD
30
20
10
0
Fig 7-2: BBFs in diabetic (Diab) rats, diabetic rats after SOD (100U/ml), NAC (100μM), combined
[BQ-123 (10μM) + SOD (100U/ml)], and diabetic rats treated with Freedox (Tirilazad mesylate)
(5mg/kg/i.p), presented as mean ± SEM (n = 4-10).
In control rats, the vascular response to SNP did not significantly change after perfusing
SOD, NAC, the combined BQ-123 + SOD, and BQ-123 (Fig 7-3 & Chapter 6, Fig 6-3).
236
Area under the response curve (cm2 )
60
Control
Control
Control
Control
50
SNP
SNP + SOD
SNP + NAC
SNP + BQ123+SOD
40
30
20
10
0
Fig 7-3: The vascular response to SNP (100μM) in control rats and control rats after SOD (100U/ml),
NAC (100μM), and combined [BQ-123 (10μM) + SOD (100U/ml)], presented as mean ± SEM
(n = 4-10).
In diabetic rats, the response to SNP [(7.4 ± 0.6 cm2) (95% CI 6.2-8.6)] was significantly
Area under the response curve (cm2 )
improved after SOD and NAC perfusions and treatment with tirilazad mesylate (Fig 7-4).
60
Diab SNP
Diab SNP + SOD
Diab SNP + NAC
Diab SNP + Freedox
Diab SNP + BQ123+SOD
50
#
+
*
40
30
*
*
*
20
10
0
Fig 7-4: The vascular response to SNP (100μM) in diabetic (Diab) rats, diabetic rats after SOD
(100U/ml), NAC (100μM), combined [BQ-123 (10μM) + SOD (100U/ml)], and diabetic rats
treated with Freedox (Tirilazad mesylate) (5mg/kg/i.p), presented as mean ± SEM (n = 4-10).
* Denotes significance from diabetic rats (P<0.05)
+ Denotes significance from diabetic rats after SOD (P<0.05)
# Denotes significance from diabetic rats after NAC (P<0.05)
237
They reached 76%, 81%, and 90% of control response [(26.1 ± 1.8 cm2) (95% CI 22.629.6)] respectively. SNP responses were (19.9 ± 3.3 cm2) (95% CI 13.4-26.4) after SOD
perfusion, (21.2 ± 1.2 cm2) (95% CI 18.8-23.6) after NAC perfusion, and (23.5 ± 1.7
cm2) (95% CI 20.2-26.8) after tirilazad mesylate treatment. The combined BQ-123 +
SOD perfusion normalized this response [(29.1 ± 1.7 cm2) (95% CI 25.8-32.4)] in
diabetic rats. This response was also significantly improved after perfusing BQ-123 alone
in these animals [(17.9 ± 2.1 cm2) (95% CI 13.8-22)] (Chapter 6, Fig 6-5).
In control rats, the vascular response to SP did not significantly change after perfusing
SOD, NAC, combined BQ-123 + SOD, and BQ-123 (Fig 7-5 & Chapter 6, Fig 6-4).
However, in diabetic rats, the vascular response to SP [(21.6 ± 3.4 cm2) (95% CI 14.928.3)] showed a slight improvement after SOD perfusion (from 42% to 50% of control),
Area under the response curve (cm2 )
reaching (25.9 ± 2.6 cm2) (95% CI 20.8-31) (Fig 7-6).
60
50
Control
Control
Control
Control
SP
SP + SOD
SP + NAC
SP + BQ123+SOD
40
30
20
10
0
Fig 7-5: The vascular response to SP (1μM) in control rats and control rats after SOD (100U/ml), NAC
(100μM), and combined BQ-123 (10μM) + SOD (100U/ml), presented as mean ± SEM
(n = 4-10).
238
Area under the response curve (cm2)
50
Diab SP
Diab SP + SOD
Diab SP + NAC
Diab SP + Freedox
Diab SP + BQ123+SOD
+
*
60
+
*
+
*
40
30
20
10
0
Fig 7-6: The vascular response to SP (1μM) in diabetic (Diab) rats, diabetic rats after SOD (100U/ml),
NAC (100μM), combined [BQ-123 (10μM) + SOD (100U/ml)], and diabetic rats treated with
Freedox (Ttirilazad mesylate) (5mg/kg/i.p), presented as mean ± SEM (n = 4-10).
* Denotes significance from diabetic rats (P<0.05)
+ Denotes significance from diabetic rats after SOD (P<0.05)
On the other hand, the combined SOD + BQ-123 perfusion managed to normalize this
response [(49.9 ± 4.5 cm2) (95% CI 41.1-58.7)]. The vascular response to SP in diabetic
rats was partially increased by BQ-123 perfusion alone [(32.4 ± 2.7 cm2) (95%CI 27.137.7)], reaching 63% of control SP (Chapter 6, Fig 6-6). Other substances significantly
improved the SP response in diabetic rats but tirilazad mesylate treatment showed more
potency [(45.1 ± 2.3 cm2) (95% CI 40.6-49.6)] than NAC [(40.7 ± 1.8 cm2) (95% CI
37.2-44.2)]. These responses reached 88% and 79% of control SP [(51.4 ± 2.9 cm2) (95%
CI 45.7-57.1)] respectively (Fig 7-6).
The vascular response to CGRP in diabetic rats [(10.4 ± 1.3 cm2) (95% CI 7.8-12.9)] was
significantly reduced, reaching 42% of control response [(24.8 ± 4.6 cm2) (95% CI 15.833.8)]. This response was normalized [(29.2 ± 4.6 cm2) (95% CI 20.2-38.2)] after
tirilazad mesylate treatment (Fig 7-7).
239
Area under the response curve (cm2 )
60
Control CGRP
Diab CGRP
Diab CGRP + Freedox
50
40
X
30
20
*
10
0
Fig 7-7: The vascular response to CGRP (1μM) in control, diabetic (Diab), and diabetic rats treated with
Freedox (Ttirilazad mesylate) (5mg/kg/i.p), presented as mean ± SEM (n = 4-10).
* Denotes significance from control rats (P<0.05)
X Denotes significance from diabetic rats (P<0.05)
The vascular response to ES in control rats [(17.6 ± 1.3 cm2) (95% CI 15-20.1)] was
enhanced by all antioxidants, reaching (22.2 ± 1.9 cm2) (95% CI 18.5-25.9) after SOD
and (19.6 ± 2.9 cm2) (95% CI 13.9-25.3) after NAC perfusions (126% & 111% of control
respectively). However, the increase in this response after perfusing the combined BQ123 + SOD perfusion [(26.4 ± 3.1 cm2) (95% CI 20.3-32.5)] and BQ-123 alone [(28.5 ±
2.5 cm2) (95% CI 23.6-33.4)] reached statistical significance (150% & 162% of control
response respectively) (Fig 7-8 & Chapter 6, Fig 6-7).
The ES response in diabetic rats [(5.1 ± 2 cm2) (95% CI 1.2-9)] was partially improved
after perfusing BQ-123 [(9.6 ± 2.2 cm2) (95%CI 5.3-13.9) (55% of control ES)] and SOD
[(12.1 ± 2.8 cm2) (95% CI 6.6-17.8) (70% of control ES)] (Chapter 6, Fig 6-8 & Fig 7-9).
However, neither of these responses reached statistical significance. The ES response in
diabetic rats was normalized after perfusing NAC [(22.6 ± 2.7 cm2) (95% CI 17.3-27.9)],
combined BQ-123 + SOD [(24.6 ± 3.3 cm2) (95% CI 18.1-31.1)], and the treatment with
tirilazad mesylate [(21.7 ± 2.5 cm2) (95% CI 16.8-26.6)].
240
Area under the response curve (cm2 )
60
Control
Control
Control
Control
50
ES
ES + SOD
ES + NAC
ES + BQ123+SOD
40
*
30
20
10
0
Fig 7-8: The vascular response to ES (20V, 5 Hz, 2 msec for 1 minute) in control rats and control rats
Area under the response curve (cm2 )
after SOD (100U/ml), NAC (100μM), and combined [BQ-123 (10μM) + SOD (100U/ml)],
presented as mean ± SEM (n = 5-10). * Denotes significance from control rats (P<0.05)
60
Diab ES
Diab ES + SOD
Diab ES + NAC
Diab ES + Freedox
Diab ES + BQ123+SOD
50
40
30
*
*
*
20
10
0
Fig 7-9: The vascular response to ES (20V, 5 Hz, 2 msec for 1 minute) in diabetic (Diab) rats, diabetic
rats after SOD (100U/ml), NAC (100μM), combined [BQ-123 (10μM) + SOD (100U/ml), and
diabetic rats treated with Freedox (Ttirilazad mesylate) (5mg/kg/i.p), presented as mean ± SEM
(n = 5-10). * Denotes significance from diabetic rats (P<0.05)
All the substances used in this study improved basal PE (R1) in diabetic rats but only the
increase in R1 after perfusing SOD, BQ-123, and the combined SOD + BQ-123 was
statistically significant (Appendix B, Table B-5).
241
The PE response to SP (SP2, 3 & 4) was calculated as percent above basal ± SEM. SEMs
were omitted from Fig 7-10 for clarity but quoted in the text. SOD, NAC and tirilazad
mesylate improved the PE response in diabetic rats upon SP perfusion. The PE response
to SP, after NAC perfusion (10 ± 17%, 49 ± 11%, 54 ± 5%) was similar to those in
tirilazad mesylate-treated rats (19 ± 13%, 50 ± 8%, 58 ± 6%) and both were greater than
those after SOD perfusion (11 ± 6%, 36 ± 5%, 40 ± 3%) (Fig 7-10). However, BQ-123
and the combined BQ-123 + SOD proved to be more potent and normalized the PE
response to SP in diabetic rats (9 ± 23%, 16 ± 32%, 0.1 ± 23%). These responses were
(43 ± 15%, 75 ± 13%, 66 ± 11%) and (23 ± 30%, 74 ± 25%, 90 ± 32%) respectively
when compared to control (36 ± 19%, 72 ± 18%, 89 ± 14%).
(%) Percent above basal
120
100
80
Control
Di abeti c
Di abeti c
Di abeti c
Di abeti c
Di abeti c
+
+
+
+
SOD
NAC
F reedox
B Q123+SOD
60
40
20
0
-20
R1-SP 2
R1-SP 3
R1-SP 4
Fig 7-10: PE in control rats, diabetic rats, diabetic rats after SOD (100U/ml), NAC (100μM), combined
[BQ-123 (10μM) + SOD (100U/ml)], and diabetic rats pre-treated with Freedox (Tirilazad
mesylate) (5mg/kg/i.p), presented as percent above basal (n = 4-10).
7.4.
Discussion
Numerous studies have shown that there is an increase in production and release of free
radicals with diabetes and it is believed that they play a role in diabetes-induced vascular
complications (chapter 1, section 1.5.3). It was shown that an increase in free radicals
could contribute to peripheral nerve and endothelial cell dysfunction (chapter 1, section
1.5.3). Antioxidants have been used extensively in experimental diabetes to reduce or
242
reverse the effect of free radicals. This chapter examines the role of ROS in modulating
skin microvascular responses in the 4 weeks STZ-diabetic rat model.
Results in chapter 6, showed an improvement in the vasodilatation responses to SNP, SP,
and ES in diabetic rats (68%, 63% & 55% of own controls respectively) after perfusing
BQ-123 alone. The current study demonstrated an improvement in these responses after
perfusing SOD (76%, 50% & 70% of control). It was reported that free radicals increase
ET-1 levels in 4 weeks diabetic kidney (Chen et al., 2000). Hence, the combined
perfusion of BQ-123 + SOD was used to examine the possible interaction between ET
and free radicals in modulating microvascular responses in diabetic rat skin. It should be
noted that since BQ-788 and NAC, at the concentrations tested, individually improved SP
vasodilation responses in diabetic rats to approximately 80-90% of the control, the
combined treatment of these two drugs was not persued (Fig 7-6 & chapter 6, Fig 6-6).
An increase in diabetic BBF was observed after perfusing NAC, BQ-123 + SOD, and
tirilazad mesylate treatment but not after SOD perfusion alone (Fig 7-2). The increase in
BBF after NAC perfusion (79% of control) suggests possible invovlement of other free
radicals other than superoxide anions, such as hydroxyl radicals and hydrogen peroxide
in contributing to early changes in microvascular blood flow in diabetes. NAC, a
powerful scavenger of hydroxyl radicals, reacts slowly with hydrogen peroxide and does
not react with superoxide anions (Aruoma et al., 1989). Lack of increase in diabetic BBF
after perfusing SOD is supportive of this view.
In diabetic rats, enhancement in BBF after perfusing the combined BQ-123 + SOD (77%
of control) was similar to those after BQ-123 perfusion alone (83% of control) but greater
than SOD alone (66% of control) (see also chapter 6). This raised the possibility that ET1 might have contributed more to the reduction in BBF in 4 weeks diabetic rats than
superoxide anions, since the blood flow was improved after blocking ET-A receptors by
BQ-123. Based on the results from the Chen et al (2000) study, endogenously or
exogenously derived ROS can also enhance ET-1 production in 4 weeks diabetic rats.
This may explain the greater increase in diabetic BBF after perfusing the combined BQ123 + SOD compared to an increase in BBF after SOD perfusion alone. The results
243
involving tirilazad mesylate treatment suggest that oxidative damage may occur in early
diabetes (4 weeks) (Figs 7-2, 7-4, 7-6 & 7-9), which could be inhibited by treatment with
tirilazad mesylate leading to an increase in microvascular blood flow.
The vascular response to SNP, a NO donor, was not altered by SOD, NAC, BQ-123 +
SOD perfusions in control rats (Fig 7-3). However, these agents and tirilazad mesylate
treatment significantly improved SNP responses in diabetic rats to 76%, 82%, 111% and
90% of control respectively (Fig 7-4). This suggests that the reduction in SNP response in
diabetic rats (28% of control) might be related to deactivation of NO by free radicals or
ETs (Gryglewski et al., 1986).
In diabetic rats, improvement in SNP responses after NAC perfusion and tirilazad
mesylate treatment suggests several possible inhibitory roles for free radicals. They may
deactivate NO or interfere with the secondary messenger pathway, preventing
vasodilatation, or induce contraction of the smooth muscle directly. An example of the
former situation could be when hydroxyl radicals oxidize low-density lipoprotein. The
product (oxLDL) can selectively inhibit vascular smooth muscle relaxation by decreasing
cyclic nucleotides (Galle et al., 1992) and the normalization of the response to CGRP
after tirilazad mesylate supports this possibility (Fig 7-7). Another possibility is the direct
involvement of hydroxyl radicals in smooth muscle contraction by altering membrane
conductance and / or interfering with calcium homeostasis and dependent processes.
There is evidence supporting the proposition that hydroxyl radicals and oxLDL can
increase cytosolic calcium, possibly through both membrane channels and reticular
release (Sasaki and Okabe, 1993; Wells et al., 1997).
In diabetic rats, improvement in the SNP response after SOD perfusion also raised the
possibility of NO deactivation by superoxide anions (Diederich et al., 1994; Gryglewski
et al., 1986). Superoxide anions also play an important role in lipid peroxidation, a
reaction that can interfere with calcium homeostasis (Wells et al., 1997). High levels of
superoxide anions can lead to formation of hydroxyl radicals, which are capable of
potentiating the voltage-dependent influx of calcium (Chang et al., 1993). This could
explain the improvement in the SNP response after NAC perfusion in diabetic rats.
244
When the SNP response after SOD perfusion was compared to that after perfusing BQ123 (68% of control), in diabetic rats, it was apparent that superoxide anions and ET-1
similarly contribute to the inhibition of this response. The current results also showed that
the combined effect of BQ-123 + SOD in improving vascular responses in diabetic rats is
greater than the effect of SOD or BQ-123 alone, which is suggestive of an additive
action. Since the response to CGRP, in diabetic rats, was also normalized by antioxidant
(Tirilazad mesylate) treatment, it is most probable that deactivation of cyclic neucleotides
by ET-1 and superoxide anions play a major role in vascular alterations in early diabetes.
In control rats, results showed no change in SP responses after SOD, NAC, and BQ-123
+ SOD perfusions (Fig 7-5) and this is in agreement with a previous study (Khalil et al.,
1999). In diabetic rats, however, these agents improved vascular responses to SP (50% &
79% of control response) but only perfusion of the combined BQ-123 + SOD normalized
it (Fig 7-6).
Previous studies in our laboratory showed that tirilazad mesylate treatment does not
affect responses in normal young animals (Khodr et al., 2001). This treatment however,
proved to be the second best agent in improving SP responses in diabetic rats, reaching
88% of control. This could be mainly due to the effect of preventing oxidative damage
(Hayakawa and Raij, 1999). It is possible that part of the improvement in SP responses in
diabetic rats after tirilazad mesylate treatment is related to preventing NO deactivation.
The results also showed greater improvement in the diabetic inflammatory response to SP
after perfusing NAC (79% of control) compared to SOD (50% of control) (Fig 7-6). This
suggests that hydroxyl radicals could play a major role in endothelial-mediated vascular
responses measured under the current experimental conditions (Pieper and Siebeneich,
1998).
It should be noted that SOD is a large molecule and will only act extracellularly when
perfused over the blister base. Therefore, intracellular superoxide anions would not be
scavenged with this treatment regime. This may be a reason for the relatively small
effects of SOD observed in this study.
Nerve dysfunction in diabetes is associated with increased oxidative stress and the results
in this chapter showed improvement in the response to ES in diabetic rats using
245
antioxidants (Fig 7-9). SOD perfusion increased the ES response in control rats by 26%
and in diabetic rats by 40% (from 29% to 69% of control value) (Figs 7-8 & 7-9). It is,
therefore, reasonable to propose that at least 26% of the increase in ES response in
diabetic rats could be related to scavenging free radicals that are released independent of
the diabetic condition (i.e. due to experimental conditions).
Unlike SOD, NAC perfusion did not alter the ES response in control rats but significantly
enhanced this response in diabetic rats by 343%. This indicates that the involvement of
hydroxyl radicals in inhibiting the ES response in diabetic rats is mainly related to
diabetes rather than to the preparation of the nerve for the ES technique or blister
induction.
In diabetic rats, the improvement in vascular responses to ES after tirilazad mesylate
treatment (123% of control) was similar to that after NAC perfusion (128% of control).
This suggests that part of the inhibitory action of hydroxyl radicals on the ES response in
these animals might be caused by oxidative damage. It should be noted that since NAC
was only perfused for a short period of time, the full potential of this drug in this
experiment cannot be extrapolated and therefore a true comparison of the effects of NAC
and tirilazad mesylate cannot be made. Lipid peroxidation caused by ROS is known to
damage the cell membrane and / or alter its conductance (Saniova, 1997). Tirilazad
mesylate can exert its protective property by inhibiting the lipid peroxidation process and
preserving the vitamin E content in the cell membrane (Saniova, 1997). Vitamin E has
been shown to have neuroprotective properties in diabetes (Cameron and Cotter, 1999).
The normalization of microvascular responses to SNP and SP, in diabetic rats, after
perfusing BQ-123 + SOD suggests that ET and superoxide anions may have an additive
effect in reducing these responses. In control rats, the response to ES was enhanced after
the combined BQ-123 + SOD perfusion (150% of control). The data in chapter 6 showed
that BQ-123 perfusion also increased the ES response in control rats (162% of control)
without altering BBF. This suggests that under this condition, the technique used to
prepare the sciatic nerve (in ES experiments) results in the release of ET, which acts on
ET-A receptors to reduce the vascular response. The current results indicate that the ES
technique may also be responsible for the release of free radicals such as superoxide
246
anions. Therefore, superoxide anions and ET (acting on ET-A receptors) released by this
technique per se, could have substantiated the effect of the combined BQ-123 and SOD
on the ES response in diabetic rats (enhanced by 382%).
In diabetes, ROS were shown to be responsible for an increase in microvascular
permeability (Bonnardel-Phu and Vicaut, 2000b). In this study, evidence was provided
that ROS increases vascular permeability in STZ-treated rats as early as 4 weeks and this
could be inhibited by treatment with antioxidants. Four weeks after induction of diabetes,
the basal PE was increased by 135%. This response was also improved after NAC
perfusion and tirilazad mesylate treatment (both by approximately 70%). However, while
basal PE after SOD perfusion was significantly improved in diabetic rats (by 30%), this
was only normalized after the combined perfusion of BQ-123 + SOD. The latter result
suggests that permeability changes in endothelial cells are reversible up to 4 weeks of
diabetes (experimental period). In chapter 6, the data showed that basal PE was also
normalized after BQ-123 perfusion alone in diabetic rats. This suggests that ETs play a
greater role than free radicals in increasing vascular permeability. The ability of SP to
induce a PE response was also normalized after BQ-123 + SOD perfusion. This suggests
that the effect of ET, via ET-A receptors, is not permanent, since blocking this receptor
resulted in restoration of endothelial cell permeability to the normal level. Among other
antioxidants, tirilazad mesylate treatment showed the greatest potency in improving the
PE response, followed by NAC and SOD perfusions. This could be related to the ability
of tirilazad mesylate to protect the cell membrane by reducing fluidity (Saniova, 1997).
In conclusion, it was demonstrated that free radicals are involved in modulating
peripheral endothelial and smooth muscle-dependent vasodilatation in 4 weeks diabetic
rats. In addition, the actions of ROS and ETs in inhibiting microvascular blood flow are
additive. The data from this section and those obtained in chapter 6 collectively suggest
that ETs play a greater role than free radicals in altering endothelial cell permeability in
diabetes. There are subtle differences in the relative contributions of different free
radicals and local pharmacological manipulation at the level of the peripheral
microcirculation could normalize diabetic vascular responses. The clinical significance of
this study is that it raises the possibility that changes in microvascular blood flow and
247
endothelial cell permeability that occur during early diabetes can be reversed using free
radical scavengers and ET antagonists.
Publication: Bassirat, M. and Khalil Z. Endothelin and free radicals modulate
microvascular responses in streptozotocin-induced diabetic rats. Microvascular Research,
59:88-98, March 2000.
248
CHAPTER 8
Diabetes and modulation of microvascular blood flow:
Role of advanced glycated end-products (AGEs) (shortand long-term effects)
249
8.1.
Introduction
The glycation process is a natural biochemical event occurring between aldose sugars and
proteins and lipids (chapter 1, section 1.5.2, Fig 1-2). This process occurs at a slow rate
under normal ambient sugar concentrations but is enhanced in the diabetic condition
(Lalla et al., 1998; Nawroth et al., 1999). AGE-modified proteins are normally either
repaired or replaced and degraded in vivo by interaction with RAGE, a cell surface
receptor for AGEs, which is found on many cells (Thornalley, 1998).
Vascular and neuronal abnormalities have been linked to glycated products (Schiekofer et
al., 2000). For example, it was postulated that formation of AGEs could contribute to the
development of a number of diabetic complications (Chibber et al., 1997; Makita et al.,
1995; Wautier and Guillausseau, 1998). AGEs formation on the matrix component of the
vessel wall can cause structural damage (Bierhaus et al., 1998). Functionally, AGERAGE interaction can lead to the generation of a pro-inflammatory environment, causing
vascular damage and functional impairment (Lalla et al., 1998; Nawroth et al., 1999).
AGEs also induce the expression of endothelin-1 (ET-1), altering endothelial function
(Bierhaus et al., 1998; Quehenberger et al., 1995). It has been hypothesized that the
interaction of glycated macromolecules with RAGE might inhibit transmembrane
calcium signalling in the presence of high glucose (Mene et al., 1999). Also, evidence
exists that both soluble RAGE (sRAGE) and the antibody against RAGE (anti-RAGE
IgG) can block AGE-RAGE interaction and ultimately prevent related vascular
complications (Wautier et al., 1996).
Glycated products can generate free radicals either themselves or by interacting with
RAGE and this can lead to endothelial dysfunction (Bucala et al., 1991). In diabetes,
AGE-RAGE interaction also causes depletion of the cellular antioxidant defence
mechanism (Kashiwagi et al., 1996; Mak et al., 1996).
Aminoguanidine (AG), a hydrazine compound, inhibits ongoing AGE formation but is
unable to remove irreversibly bound AGEs from connective tissues and matrix
compounds that have occurred over time. AG was shown to retard or prevent the
development of AGE-related diabetic complications in streptozotocin (STZ)-treated rats
(Vlassara et al., 1992). AG, in vivo, acts as an antioxidant, quenching hydroxyl radicals
250
and lipid peroxidation in cells, and preventing oxidant-induced apoptosis (Giardino et al.,
1998).
No studies to date have examined the role of AGEs in modulating skin microvascular
blood flow in early diabetes. Therefore, in this chapter, short- and long-term effects of
amadori and AGEs (early and late products of glycation process) (see chapter 1, section
1.5.2, Fig 1-2 & Appendix C) in modulating skin microvasular blood flow in 4 weeks
diabetic rats are examined. It was hypothesized that AGEs may be involved in reducing
microvascular blood flow in early diabetes and its effect may be inhibited by early
intervention. Microvascular responses in diabetic rats are examined after short-term
exposure of their microvasculature to amadori and AGEs and after long-term pretreatment with AG to prevent the formation of AGEs.
8.2.
Method
A total of 14 groups of rats (n = 4-10) were used, of which 9 groups were made diabetic
with STZ as described in chapter 2, section 2.4.3. Control rats were injected with 0.1M
sodium citrate buffer solution (CB-treated) (chapter 2, section 2.4.1). Four groups of
diabetic rats were equally divided and two sets were treated with AG (25mg/kg, i.p.) for a
duration of 2 weeks (AG (2W)) and the others were treated for 4 weeks (AG (4W)). The
AG treatment was either commenced on day 14 for the duration of 2 weeks (late
intervention) or on day 3 (after diabetes was confirmed) for the duration of 4 weeks
(early intervention).
Animals
Animals
Control (CB-treated)
AG
2W 4W
AGEs
Diabetic
Amadori
AGEs
Amadori
Anti-RAGE
Peptide
IgG
perfusion
SP & SNP
SP & SNP
& ES
Peptide perfusion
& ES
251
Anti-RAGE
IgG
In the peptide perfusion protocol, first substance P (SP, 1μM, 30 min), then sodium
nitroprusside (SNP, 100μM, 10 min), and finally calcitonin gene-related peptide (CGRP,
1μM, 10 min) were perfused over the blister base. In electrical stimulation (ES)
experiments, first SNP (100μM, 10 min) was perfused over the blister base and then the
sciatic nerve was stimulated (20V, 5Hz, 2 msec for 1 minute).
Peptide perfusion and ES experiments were performed on 2 control groups, 2 diabetic
groups, and 4 AG-treated groups of rats (control and diabetic).
In the amadori (1mg/ml) and AGEs (10mg/ml) groups (2 control and 2 diabetic), after
establishing the baseline, either of these substances were perfused for 5 minutes followed
by 30 minutes incubation where the peristaltic pump was stopped. This was to allow
longer exposure of the microvasculature to these substances. Then a further 10 minutes
perfusion of these substances was performed. Immediately after that, SP was perfused
followed by SNP as described in chapter 2, section 2.6. For the remaining groups (1
control and 1 diabetic), anti-RAGE IgG (100μg/ml) was perfused for 10 minutes prior to
and together with SP, followed by Ringer’s solution and SNP. Plasma extravasation (PE)
was collected from all perfusion experiments. At the end of each experiment, rats were
sacrificed with pentobarbitone overdose.
8.2.1. Expression of data and statistical analyses
Results from vasodilatation responses were calculated as areas under the response curves.
PE responses are presented as percent above basal. All data are expressed as mean ±
standard error of the mean (SEM). The absolute values and 95% confidence intervals
(95% CI) of these values are quoted in the text.
Separate two-way Analyses of Variances (ANOVAs) were performed to investigate
whether there were significant differences in basal blood flow (BBF) and responses to
SNP, SP, CGRP, and ES, between control and diabetic rats, according to whether the rats
were exposed to or treated with any of the following drugs (amadori, AGEs, anti-RAGE
IgG, AG (2W), AG (4W)). This test also shows interaction effects between these groups.
Where ANOVA indicated a significant main effect, post hoc comparisons were
conducted using the Tukey HSD test. Separate two-way repeated measure ANOVA with
252
priori planned comparisons was performed to investigate whether there were significant
differences in PE responses, between control and diabetic rats, according to whether the
rats were exposed to or treated with any of the following drugs (amadori, AGEs, antiRAGE IgG, AG (2W), AG (4W)).
8.3.
Results
BBFs in control [(25.1 ± 0.9 cm2) (95% CI 23.3-26.9)] or diabetic rats [(16.7 ± 1.3 cm2)
(95% CI 14.1-19.2)] did not significantly change after amadori, AGEs, and anti-RAGE
Area under the response curve (cm2)
IgG perfusions or AG treatments (Figs 8-1 & 8-2).
100
90
80
Control
Control
Control
Control
BBF
BBF + amadori
BBF + AGEs
BBF + anti-RAGE
70
60
50
40
30
20
10
0
Fig 8-1: BBFs in control rats and control rats after amadori (1mg/ml), AGEs (10mg/ml), and anti-RAGE
IgG (100μg/ml), presented as mean ± SEM (n = 4-8).
253
Area under the response curve (cm2)
100
Diab BBF
Diab BBF + amadori
Diab BBF + AGEs
Diab BBF + anti-RAGE
Diab BBF + AG (2 W)
Diab BBF + AG (4W)
90
80
70
60
50
40
30
20
10
0
Fig 8-2: BBFs in diabetic (Diab) rats, diabetic rats after amadori (1mg/ml), AGEs (10mg/ml), and antiRAGE IgG (100μg/ml), and diabetic rats treated with AG (25mg/kg, i.p) (2 or 4 weeks),
presented as mean ± SEM (n = 4-10).
Short-term exposure to amadori and AGEs, in control rats, increased vascular responses
to SNP and SP. Amadori significantly enhanced the vasodilator responses to SNP and SP,
reaching 150% and 173% of own control responses respectively (Table 8-1, Figs 8-3 &
8-4). Although exposure to AGEs also increased these responses in control rats (143% &
162% of own control responses respectively), only the response to SP reached statistical
significance. Perfusion of anti-RAGE IgG did not have any effect on SNP or SP
responses in control rats.
Modulator
Control SNP
Control SP
None
(26.1 ± 1.8 cm2)
(51.4 ± 2.9 cm2)
(7.4 ± 0.6 cm2)
(21.6 ± 3.4 cm2)
(95% CI 23-30)
(95% CI 46-57)
(95% CI 6-9)
(95% CI 15-28)
(39.4 ± 4.8 cm )
(89.1 ± 9.4 cm )
(14.3 ± 1.6 cm )
(38.2 ± 5.3 cm2)
(95% CI 30-49)
(95% CI 71-107.5)
(95% CI 11-17)
(95% CI 28-49)
(37.3 ± 5.8 cm )
(83.5 ± 8.8 cm )
(8 ± 2 cm )
(20.05 ± 2.5 cm2)
(95% CI 26-49)
(95% CI 66-101)
(95% CI 4-12)
(95% CI 15-25)
(26.4 ± 6.3 cm )
(43.1 ± 2 cm )
(29 ± 1.8 cm )
(37 ± 2.6 cm2)
(95% CI 14-39)
(95% CI 39-47)
(95% CI 25.5-32.5)
(95% CI 32-42)
2
Amadori
2
AGEs
Anti-RAGE
2
2
2
2
Diabetic SNP
Diabetic SP
2
2
2
Table 8-1: Vascular responses to SP (1μM) and SNP (100μM) in control and diabetic rats before
and after amadori (1mg/ml), AGEs (10mg/ml), and anti-RAGE IgG (100μg/ml) (n = 4-10).
Data are presented as mean ± SEM and their 95% CIs.
254
Area under the response curve (cm2 )
100
Control
Control
Control
Control
90
80
SNP
SNP + amadori
SNP + AGEs
SNP + anti-RAGE
70
60
50
*
40
30
20
10
0
Fig 8-3: The vascular response to SNP (100μM) in control rats and control rats after amadori (1mg/ml),
Area under the response curve (cm2 )
AGEs (10mg/ml), and anti-RAGE IgG (100μg/ml), presented as mean ± SEM (n = 4-8).
* Denotes significance from control rats (P<0.05)
100
*
Control
Control
Control
Control
*
90
80
SP
SP + amadori
SP + AGEs
SP + anti-RAGE
70
60
50
#
X
40
30
20
10
0
Fig 8-4: The vascular response to SP (1μM) in control rats and control rats after amadori (1mg/ml),
AGEs (10mg/ml), and anti-RAGE IgG (100μg/ml), presented as mean ± SEM (n = 4-8).
* Denotes significance from control rats (P<0.05)
X Denotes significance from control rats after amadori (P<0.05)
# Denotes significance from control rats after AGEs (P<0.05)
In diabetic rats, exposure to amadori caused a significant improvement in vasodilation
responses to both SNP and SP, reaching 55% and 74% of their own control values (Table
8-1, Figs 8-5 & 8-6). In diabetic rats, AGEs did not have an effect on either SNP or SP
255
responses, whereas anti-RAGE IgG normalized SNP and improved SP responses to 72%
Area under the response curve (cm2 )
of control response.
100
Diab SNP
Diab SNP + amadori
Diab SNP + AGEs
Diab SNP + anti-RAGE
Diab SNP + AG (2W)
Diab SNP + AG (4W)
90
80
70
60
50
#
X
40
+
#
+
30
20
#
+
+
X
10
0
Fig 8-5: The vascular response to SNP (100μM) in diabetic (Diab) rats, diabetic rats after amadori
Area under the response curve (cm2)
(1mg/ml), AGEs (10mg/ml), anti-RAGE IgG (100μg/ml), and diabetic rats treated with AG
(25mg/kg, i.p) (2 or 4 weeks), presented as mean ± SEM (n = 4-10).
+ Denotes significance from diabetic rats (P<0.05)
X Denotes significance from diabetic rats after amadori (P<0.05)
# Denotes significance from diabetic rats after AGEs (P<0.05)
Diab SP
Diab SP + amadori
Diab SP + AGEs
Diab SP + anti-RAGE
Diab SP + AG (2W)
Diab SP + AG (4W)
100
90
80
70
60
50
+
#
+
40
30
#
+
#
+
X
20
10
0
Fig 8-6: The vascular response to SP (1μM) in diabetic (Diab) rats, diabetic rats after amadori (1mg/ml),
AGEs (10mg/ml), anti-RAGE IgG (100μg/ml), and diabetic rats treated with AG (25mg/kg, i.p)
(2 or 4 weeks), presented as mean ± SEM (n = 4-10).
+ Denotes significance from diabetic rats (P<0.05)
X Denotes significance from diabetic rats after amadori (P<0.05)
# Denotes significance from diabetic rats after AGEs (P<0.05)
256
Treatment of diabetic rats with AG, after 2 weeks and 4 weeks (late and early onset
treatments), significantly improved vasodilatation responses to SNP (both about 80% of
control) and SP (both about 70% of control) (Table 8-2, Figs 8-5 & 8-6).
Stimuli
SNP
SP
CGRP
ES
Non-treated
AG-treated
AG-treated
(4 weeks)
(2 weeks)
(4 weeks)
(7.4 ± 0.6 cm )
(21.4 ± 2.5 cm )
(20.9 ± 2.6 cm2)
(95% CI 6-9)
(95% CI 16.5-26)
(95% CI 16-26)
(21.6 ± 3.4 cm2)
(37.4 ± 3.2 cm2)
(37.6 ± 1.9 cm2)
(95% CI 15-28)
(95% CI 31-44)
(95% CI 34-41)
(10.4 ± 1.3 cm2)
(22.1 ± 1.5 cm2)
(21.6 ± 3 cm2)
(95% CI 8-13)
(95% CI 19-25)
(95% CI 16-27.5)
(5.1 ± 2 cm2)
(12.4 ± 1.3 cm2)
(17.6 ± 2.6 cm2)
(95% CI 1-9)
(95% CI 10-15)
(95% CI 12.5-23)
2
2
Table 8-2: Vascular responses to SNP (100μM), SP (1μM), CGRP (1μM), and ES (20V, 5Hz, 2msec,
for 1 minute) in diabetic rats treated with AG (25mg/kg, i.p) (2 or 4 weeks) (n = 4-10).
Data are presented as mean ± SEM and their 95% CIs.
The vascular response to CGRP in diabetic rats was improved to nearly 90% of control
[(24.8 ± 4.6 cm2) (95% CI 15.8-33.8)] by both AG treatments (Fig 8-7). The vascular
response to ES, on the other hand, was improved to 70% of its own control [(17.6 ± 1.3
cm2) (95% CI 15-20.1)] after late onset AG treatment (2 weeks) but normalized after
early onset treatment with AG (4 weeks) (Fig 8-8).
257
Area under the response curve (cm2 )
100
Control CGRP
Diab CGRP
Diab CGRP + AG (2W)
Diab CGRP + AG (4W)
90
80
70
60
50
40
30
20
+
+
*
10
0
Fig 8-7: The vascular response to CGRP (1μM) in control, diabetic (Diab), and diabetic rats treated
Area under the response curve (cm2 )
with AG (25mg/kg, i.p) (2 or4 weeks), presented as mean ± SEM (n = 4-10).
* Denotes significance from control rats (P<0.05)
+Denotes significance from diabetic rats (P<0.05)
100
Control ES
Diab ES
Diab ES + AG (2W)
Diab ES + AG (4W)
90
80
70
60
50
40
30
+
*
20
10
+
*
0
Fig 8-8: The vascular response to ES (20V, 5Hz, 2msec for 1 minute) in control, diabetic (Diab), and
diabetic rats treated with AG (25mg/kg, i.p) (2 or 4 weeks), presented as mean ± SEM (n = 4-8).
* Denotes significance from control rats (P<0.05)
+Denotes significance from diabetic rats (P<0.05)
The basal PE (R1) in diabetic rats was improved after all perfusions but did not reach
statistical significance (Appendix B, Table B-6). There was no change in R1 after AGEs
perfusion. On the other hand, R1 was significantly improved after AG treatments,
258
decreasing from 235% of control basal value before treatment to 170% after late onset (2
weeks) and 142% after early onset (4 weeks) treatments.
The PE response to SP (SP2,
3 & 4)
was calculated as percent above basal ± SEM. SEMs
were omitted from Fig 8-9 for clarity but quoted in the text. The most effective treatment
was anti-RAGE IgG perfusion, where the PE response to SP showed the greatest increase
above basal (64 ± 28%, 98 ± 27%, 86 ± 15%) (Fig 8-9). The second-most effective
treatment was early onset treatment with AG, enhancing this response to 52 ± 22%, 86 ±
15%, and 82 ± 17% above basal. These responses were similar to PE response in control
rats (36 ± 19%, 72 ± 18%, 89 ± 14%). This indicates that both anti-RAGE IgG perfusion
and early onset AG treatment could significantly improve vascular permeability in
diabetic rats. Amadori perfusion and late onset AG treatment also enhanced the PE
response in diabetic rats but the former perfusion was more effective (30 ± 17%, 45 ±
15%, 48 ± 20%) than the latter treatment (5 ± 16%, 32 ± 18%, 42 ± 22%). The PE
response in diabetic rats (9 ± 23%, 16 ± 32%, 0.1 ± 23%) was not altered after AGEs
perfusion (13 ± 10%, 20 ± 9%, 4 ± 15%).
(%) Percent above basal
120
Control
Diabetic
Diabetic +
Diabetic +
Diabetic +
Diabetic +
Diabetic +
100
80
60
40
amadori
AGEs
anti-RAGE
AG (2W)
AG (4W)
20
0
-20
R1-SP2
R1-SP3
R1-SP4
Fig 8-9: PE in control rats, diabetic rats, diabetic rats after amadori (1mg/ml), AGEs (10mg/ml, antiRAGE IgG (100μg/ml), and diabetic rats treated with AG (25mg/kg, i.p) (2 or 4 weeks),
presented as percent above basal (n = 4-10).
259
8.4.
Discussion
AGEs formation and deposition are much enhanced in diabetes and their presence has
been linked to secondary complications, especially microvascular disease (chapter 1,
section 1.5.2). The mechanisms through which AGEs exert their effects are complicated
and involve many factors, including the duration of the disease. In this chapter, the role of
short- and long-term effects of early and late glycated products in modulating
inflammatory responses in 4 weeks STZ-induced diabetic rats was examined.
The results showed no change in BBFs in control and diabetic rats after exposure to
amadori, AGEs, anti-RAGE IgG or treatments with AG (Figs 8-1 & 8-2). This suggests
that in the diabetic condition exogenous or endogenous glycated products do not play a
major role in modulating vascular tone.
In control rats, the vascular response to SP was significantly increased after amadori and
AGEs perfusions (173% & 162% of control response) (Fig 8-4). This could be partially
due to a concomitant increase in smooth muscle reactivity after perfusing these
substances in control rats (150% & 143% of control respectively) (Fig 8-3). The release
of hydrogen peroxide by amadori (Elgawish et al., 1996; Mossine et al., 1999) may also
play a role in the observed increase in the SP response in control rats in the presence of
amadori products. Based on the results from a previous study (Boyer et al., 1995), it is
possible that hydrogen peroxide may induce endothelial vasodilation independent of
calcium and via a specific signal-reponseive phospholipase A2. In addition, a proinflammatory environment generated by AGE-RAGE interaction (Lalla et al., 1998)
could have contributed to the increase in these responses in control rats.
Overall, it appears that control inflammatory responses are enhanced in the presence of
amadori and AGEs. It should be mentioned that in control animals, since the antioxidant
system is fully functional, superoxide anions generated by amadori (Bucala et al., 1991)
could be removed from the system.
In diabetic rats, short-term exposure of the microvasculature to amadori significantly
enhanced the vascular response to SNP (55% of control) and SP (74% of control) but no
260
changes in these responses was observed after perfusing AGEs (Figs 8-5 & 8-6). The
increase in these responses after amadori suggests that amadori is capable of modifying
vascular responses in diabetic rats. This modification, however, may depend on the
duration of diabetes. In early diabetes, more amadori than AGEs may be present
producing more superoxide anions and hydrogen peroxide than other free radicals.
A previous study, in vitro, showed that smooth muscle cells were damaged after longterm incubation (2 weeks) with early glycated products (Sakata et al., 1998). The authors
suggested that the generation of hydrogen peroxide was responsible for this damage. The
current results showed an enhancement in the SNP response in diabetic rats after 45
minutes incubation and perfusion of amadori. This suggests that short-term exposure to
these products might not have generated hydrogen peroxide in high enough concentration
to cause extensive damage to smooth muscle cells.
The lack of AGEs effect on SP and SNP responses in diabetic rats raises the possibility
that endogenous AGEs are already involved in modulating these responses and therefore
exogenous AGEs, used in this study, may not be capable of further modification. Studies
by Rumble et al (Rumble et al., 1997) and Hill et al (Hill and Ege, 1994) support this
proposition by showing the presence of AGEs in early diabetes.
Exposure of the microvasculature to anti-RAGE IgG did not change SNP or SP responses
in control rats. However, it improved the SP response to 72% of control and normalized
the SNP response in diabetic rats. This result provides further support for the proposition
that endogenous AGEs are involved in the modulation of microvascular inflammatory
responses in diabetic rats. Furthermore, RAGE receptors may play a role in modulating
microvascular blood flow in 4 weeks diabetic rats. It is possible that anti-RAGE IgG
prevented endogenous AGE-RAGE interaction in diabetic rats and reduced the
deleterious effects of endogenous AGEs in the system. For example, anti-RAGE IgG may
have improved vasodilatation, in diabetic rats, by preventing NO inhibition directly and
indirectly (Bucala et al., 1991; Ceriello, 1999; Galle et al., 1998; Kashiwagi et al., 1996)
(see also chapter 1, section 1.5.2).
It should be noted that, in this study, since only a single dose of anti-RAGE IgG was used
(100μM/ml, 40 min perfusion over the blister base), this could not reflect the true
261
contribution of endogenous AGEs to the vascular responses in 4 weeks STZ-treated
diabetic rats. Endogenous AGEs true contribution can only be extrapolated from a doseresponse effect of anti-RAGE IgG on the vascular response, an experiment that was not
performed in the current study. The alternative approach was to examine the long-term
role of endogenous AGEs using long-term treatment with AG (late and early onset
treatments) to prevent AGEs formation.
A previous study reported that AG prevented vascular defects (hyperpermeability and
unresponsiveness to vasodilatory agents) induced by (short-term) AGEs in control rats
(Vlassara et al., 1992). In diabetic animals, on the other hand, AG treatment decreased
AGEs accumulation in large arteries and improved abnormalities of the diabetic
peripheral nerves (Hujiberts et al., 1993; Kihara et al., 1991).
In diabetic rats, late and early onset treatment with AG (2 and 4 weeks), significantly
improved vascular responses to SP (both 70% of control) and CGRP (both about 90% of
control) (Table 8-2, Figs 8-6 & 8-7). These results could be partially explained by the
similar significant enhancement in their smooth muscle-dependent vasodilations after
both treatments (both about 80% of control) (Fig 8-5). It is possible that in diabetic rats,
AG inhibited AGEs formation and hence protected NO from inactivation by these
products. One mechanism through which AG may have exerted this protective effect is
by inhibiting lipid peroxidation (Giardino et al., 1998), a process that can cause oxidative
stress. The other possible mechanism may be related to inhibition of transmembrane
calcium signalling by AGE-RAGE interaction by AG (Mene et al., 1999).
As shown in the results, SP and CGRP responses in diabetic rats were similarly improved
regardless of the onset and duration of AG treatments. Early onset AG treatment for 4
weeks did not result in a full recovery of any of the post-terminal vascular responses.
This suggests that other factors, apart from AGEs, might be responsible for some of the
changes in vascular responses in 4 weeks STZ-treated diabetic rats. For example, the
production of substances by AG that have similar action and potency to free radicals may
be involved and other studies are supportive of this proposition (Ou and Wolff, 1993;
Skamarauskas et al., 1996).
Tilton and co-workers (1993) showed that AG is a potent inhibitor of nitric oxide
synthase (NOS) activity. This raises the possibility that AG treatment may inhibit NO
262
synthesis in diabetic rats. Previous studies, using both healthy and 4 weeks diabetic rats,
showed a significant reduction in the vasodilator responses to BK and SP after perfusion
of a NOS inhibitor, NG-nitro-L-arginine-methyl ester (L-NAME) (100μg/ml) (Khalil and
Helme, 1992; Ralevic et al., 1992) (see also results in chapter 4). In diabetic rats, vascular
responses to BK and SP were also significantly reduced (see results in chapter 4). The
current results showed partial enhancement of microvascular responses in diabetic rats
after acute AG treatment. These results raised the possibility that an inhibitory action of
AG on NOS may have partially masked the effect of AG to increase these responses in
diabetic animals through the prevention of AGEs formation.
Improvement in diabetic vascular responses to ES, after both AG treatments (70% of
control after late onset and normalization after early onset treatment), could be related to
the ability of AG to preserve sensory nerve function in these rats (Fig 8-8). This was
supported by Ryle et al (1997) who showed that, in early diabetic rats, low levels of
AGEs were present in the sciatic nerve and spinal cord. Defects in axonal transport were
also reported in the nerves of short-term diabetic rats (Tomlinson et al., 1988). In
addition, a significant increase in tubulin glycation was observed in the rat sciatic nerve
after 4 weeks of diabetes (Cullum et al., 1991). Cullum et al suggested that AG may have
interfered with the tubulin glycation process and as a result prevented deficits to occur in
nerve axonal transport. Results showed a further 30% improvement in the ES response
after early onset AG-treated rats, when compared to late onset treated animals, which is
independent of the smooth muscle reactivity. This raises the possibility that sensory
nerves are sensitive to the damaging effects of AGEs and that early onset AG treatment
could protect these nerves from such damage. This is supported by Soulis-Liparota et al
(1996), who demonstrated that the duration of AG treatment plays an important role in
protecting diabetic tissues against damage.
In chapter 3, it was demonstrated that the vascular permeability of post-capillary venules,
in 4 weeks diabetic rats, was altered (Appendix B, Table B-1). Also, there was an
increase in basal perfusate (R1) and poor SP response in diabetic vessels. In the current
study, both R1 and SP2, 3, & 4 were normalized in 4 weeks diabetic rats after anti-RAGE
IgG perfusion and early onset AG treatment. Although R1 was improved more after
263
amadori perfusion than late onset treatment with AG, the PE response to SP was similar
after perfusion of both substances (Appendix B, Table B-6 & Fig 8-9). Anti-RAGE
perfusion restored the PE response in 4 weeks diabetic rats. Results after AG treatments
and anti-RAGE IgG perfusion suggests that permeability changes in endothelial cells is
temporary and reversible up to 4 weeks of diabetes (experimental period) (Vlassara et al.,
1992; Wautier et al., 1996). Anti-RAGE IgG may restore diabetic endothelial cell
function back to it original form and early onset AG treatment may protect diabetic
endothelial cells from functional and structural disturbances exerted by AGEs over the 4
week period of diabetes.
In conclusion, in this chapter evidence was provided that short-term exposure of the
microvasculature to amadori products and AGEs could modulate microvascular function
in control rats. In addition, amadori enhanced, while exogenous AGEs did not alter,
inflammatory vascular responses in early diabetes. It was proposed that early products of
the glycation process, namely amadori, could provide a respite for the microvasculature
and hence, improve the inflammatory responses in diabetic rats. It was demonstrated that
AGEs receptors are involved in modulating the inflammatory responses in diabetic rats.
The improvement in vascular responses in early diabetic rats after short-term exposure to
anti-RAGE IgG is encouraging. Hence, further investigation is required to establish the
possible clinical use of this antibody. In addition, it was suggested that short-term AG
treatment can improve the inflammatory responses in 4 weeks diabetic rats and that the
duration and onset of treatment can play an important role with reference to the extent of
preventing neurovascular damage. The time frame beyond which anti-RAGE IgG and
AG treatment may not be capable of reversing diabetic vascular changes also needs to be
established.
Publication: Bassirat, M. and Khalil Z. Short- and long-term modulation of
inflammation in streptozotocin-induced diabetic rats by glycosylated products. Diabetes
(Submitted) (2001).
264
CHAPTER 9
Hyperglycaemia and changes in microvascular blood
flow: Role of insulin
265
9.1.
Introduction
Previous studies in animal models and patients with diabetes have demonstrated that
strict glycaemic control, using both short- and long-term insulin treatment, can assist in
slowing the progression of diabetes-induced complications (Amthor et al., 1994; Havas,
1999; Tack et al., 1998). Insulin deficiency, in type I diabetes, often leads to neuronal
deficits. It was shown that early insulin therapy could prevent alterations in vasoreactivity
and endothelial cell function (chapter 1, section 1.8.5.3). It was suggested that these
beneficial effects might be related to glycaemic control (Biessels et al., 1996).
Insulin can regulate intracellular calcium levels by influencing the activity of calcium
channels and the enzymes involved in vascular smooth muscle cells but high intracellular
calcium was shown to impair insulin sensitivity (Cleland et al., 1998; Zemel, 1995).
Vasodilatation induced by insulin in skeletal muscle is dependent on Na+/K+-ATPase at
the level of the endothelium (Tack et al., 1996). It was suggested that, in diabetes,
alteration in the activity of this enzyme may affect the function of peripheral sensory
nerves (Greene and Lattimer, 1983; King et al., 1994; Ver et al., 1997).
Under physiological conditions, insulin is also a modulator of Endothelin-1 (ET-1), ET
receptors, ET-1 mediated responses, and a stimulator of nitric oxide (NO) activity
(chapter 1, section 1.8.5.3, Fig 1-14). Insulin deficiency or hyperglycaemia could also be
partly responsible for the enhanced ET-1 production induced by reactive oxygen species
(ROS) (Chen et al., 2000). Previous studies demonstrated that, in diabetes, insulin
inhibited lipid peroxidation (Rungby et al., 1992) and advanced glycation process (Turk
et al., 1999). Hence, interaction of insulin with ROS, ET-1, NO, and glycation products
may be important in the modulation of blood flow in the diabetic condition.
In this chapter, the role of insulin in the modulation of skin microvascular blood flow in
early diabetes is investigated. The aim was to examine the relationship between time of
commencement and duration of insulin treatment and development of skin microvascular
alterations in 4 weeks diabetic rats. It was hypothesized that hyperglycaemia may be
involved in reducing microvascular blood flow in early diabetes and its effect may be
inhibited by early intervention.
266
9.2.
Method
A total of ten groups of rats (n = 4-10) were used for this study, 2 control groups, 2
diabetic groups (2 weeks), and 6 diabetic groups (4 weeks) post-onset. Control rats were
injected with 0.1M sodium citrate buffer solution (CB-treated) (see chapter 2, section
2.4.1). Four diabetic groups (4 weeks) were treated with insulin (2-6IU/100g/day, s.c.) for
2 different periods of time. Half of them were treated for 4 weeks, commencing soon
after diabetes onset (early intervention) and the other half were treated for only 2 weeks,
commencing at week 2 (late intervention) and continuing until the day of the experiment.
The protocol for insulin treatment is described in detail in chapter 2, section 2.5. The
remaining two diabetic groups (4 weeks) were not treated (NT).
Animals
Diabetic
(4 weeks)
Diabetic
Control
(CB-treated) (2 weeks)
Insulin-treated
NT
2 weeks
(late onset
intervention)
4 weeks
(early onset
intervention)
Peptide perfusion
& ES
Peptide
perfusion & ES
In experiment 1 (peptide perfusion), first substance P (SP, 1μM, 30 min), then sodium
nitroprusside (SNP, 100μM, 10 min), and finally calcitonin gene-related peptide (CGRP,
1μM, 10 min) were perfused over the blister base. In experiment 2 (electrical stimulation
(ES) of the sciatic nerve), first SNP (100μM, 10 min) was perfused over the blister base
and then the sciatic nerve was stimulated (20V, 5Hz, 2 msec for 1 minute). Plasma
extravasation (PE) was collected from all perfusion experiments. At the end of each
experiment, rats were sacrificed with pentobarbitone overdose.
267
9.2.1. Expression of data and statistical analyses
Results from vasodilatation responses were calculated as areas under the response curves.
PE responses are presented as percent above basal. All data are expressed as mean ±
standard error of the mean (SEM). The absolute values and 95% confidence intervals
(95% CI) of these values are quoted in the text.
A two-way Analysis of Variance (ANOVA) was performed to investigate whether there
were significant differences in weight gain, basal blood flow (BBF), and responses to
SNP, SP, CGRP, and ES, between control, diabetic, and diabetic rats treated with insulin
(late and early intervention). Where a significant main effect was detected, post hoc
comparisons were conduced using the Tukey HSD test. A two-way repeated measure
ANOVA with priori planned comparisons was performed to investigate whether there
were significant differences in PE responses between control, diabetic, and diabetic rats
treated with insulin (late and early intervention).
9.3.
Results
The glucose level in diabetic rats was kept close to the normal range throughout the
experimental period, except for the late onset insulin-treated groups that were
hyperglycaemic for the first 2 weeks after the disease onset. The average urine glucose
level for insulin treated (2 & 4 weeks) groups of diabetic rats is shown in Appendix D.
The blood glucose of insulin- treated diabetic rats on the day of the experiment was
between 3.0-4.8 mmol/lit.
The BBF in diabetic rats [(16.7 ± 1.3 cm2) (95%CI 14.1-19.2)] was increased after 4
weeks insulin treatment [(21.8 ± 2 cm2) (95%CI 17.9-25.7)], reaching 87% of control
response [(25.1 ± 0.9 cm2) (95%CI 23.3-26.9)] (Fig 9-1).
268
Area under the response curve (cm2)
70
Control BBF
Diab BBF (2W)
Diab BBF (4W)
Diab BBF (4W) + late insulin intervention
Diab BBF (4W) + early insulin intervention
60
50
40
30
20
*
*
*
10
0
Fig 9-1: BBFs in control rats, diabetic (Diab) (2 & 4 weeks) rats, and diabetic rats (4 weeks) treated with
insulin (2-6IU/100g/day) (late and early interventions), presented as mean ± SEM (n = 4-10).
* Denotes significance from control rats (P< 0.05)
Rats with 4 weeks diabetes showed a significant reduction in weight [(202 ± 6.7g)
(95%CI 189-215)] compared to control rats of the same age (approximately 10 weeks)
[(398 ± 6.1g) (95%CI 386-410)]. Both insulin treatments increased the average weight of
diabetic rats. They increased to (238 ± 6.6 g) (95%CI 225-251) after late insulin
intervention and to (300 ± 7.1g) (95%CI 286-313.9) after early insulin intervention, with
the latter being more effective (Fig 9-2).
269
Control
Diab (2W)
Diab (4W)
Diab (4W) + late insulin intervention
Diab (4W) + early insulin intervention
450
400
#
*
Weight (g)
350
300
*
250
200
*
*
150
100
50
0
Fig 9-2: Weigh in control rats, diabetic (Diab) (2 & 4 weeks) rats, and diabetic rats (4 weeks)
treated with insulin (2-6IU/100g/day) (late and early interventions), presented as mean ± SEM
(n = 4-7).
* Denotes significance from control rats (P<0.05)
# Denotes significance from diabetic rats (both 2 and 4 weeks) (P<0.05)
The vascular response to SNP in rats was significantly decreased after 2 weeks [(11.6 ±
1.1 cm2) (95%CI 9.4-13.7)] and 4 weeks of diabetes [(7.4 ± 0.6 cm2) (95%CI 6.2-8.6)]
Area under the response curve (cm2 )
(44% & 28% of control respectively) (Fig 9-3).
70
Control SNP
Diab SNP (2W)
Diab SNP (4W)
Diab SNP (4W) + late insulin intervention
Diab SNP (4W) + early insulin intervention
60
50
40
+
30
20
10
*
*
*
0
Fig 9-3: Vascular responses to SNP (100μM) in control rats, diabetic (Diab) (2 & 4 weeks) rats, and
diabetic rats (4 weeks) treated with insulin (2-6IU/100g/day) (late and early interventions)
(n = 4-10).
* Denotes significance from control rats (P<0.05)
+ Denotes significance from diabetic rats (4 weeks) (P<0.05)
270
This response was significantly improved after rats received late onset insulin treatment
[(15.1 ± 0.9 cm2) (95CI 13.3-16.9)], reaching 58% of control value [(26.1 ± 1.8 cm2)
(95%CI 22.6-29.6)]. However, the response to SNP was normalized in diabetic rats that
received early onset insulin treatment [(25.4 ± 1.9 cm2) (95%CI 21.7-29.1)]. The vascular
response to SP in rats also showed a significant reduction after 2 weeks [(26.2 ± 4.9 cm2)
(95%CI 16.6-35.8)] and 4 weeks of diabetes [(21.6 ± 3.4 cm2) (95%CI 14.9-28.3)].
In diabetic rats, late insulin intervention did not improve the response to SP [(19.1 ± 1.5
cm2) (95%CI 16.2-22)] but early insulin intervention normalized it [(57 ± 4.3 cm2)
Area under the response curve (cm2 )
(95%CI 48.6-65.4)] (Fig 9-4).
Control SP
Diab SP (2W)
Diab SP (4W)
Diab SP (4W) + late insulin intervention
Diab SP (4W) + early insulin intervention
70
+
60
50
40
*
30
*
*
20
10
0
Fig 9-4: Vascular responses to SP (1μM) in control rats, diabetic (Diab) (2 & 4 weeks) rats, and diabetic
rats (4 weeks) treated with insulin (2-6IU/100g/day) (late and early interventions), presented as
mean ± SEM (n = 4-10).
* Denotes significance from control rats (P<0.05)
+ Denotes significance from diabetic rats (4 weeks) (P<0.05)
271
The CGRP response was reduced in rats after 2 weeks of diabetes [(13.5 ± 3.1 cm2)
(95%CI 7.4-19.6)], reaching 54% of the control value [(24.8 ± 4.6 cm2) (95%CI 15.8-
Area under the response curve (cm2)
33.8)] (Fig 9-5).
70
Control CGRP
Diab CGRP (2W)
Diab CGRP (4W)
Diab CGRP (4W) + late insulin intervention
Diab CGRP (4W) + early insulin intervention
60
50
40
+
30
20
*
10
0
Fig 9-5: Vascular responses to CGRP (1μM) in control rats, diabetic (Diab) (2 & 4 weeks) rats, and
diabetic rats (4 weeks) treated with insulin (2-6IU/100g/day) (late and early interventions),
presented as mean ± SEM (n = 4-6).
* Denotes significance from control rats (P<0.05)
+ Denotes significance from diabetic rats (4 weeks) (P<0.05)
Late onset insulin treatment did not improve this response in diabetic rats [(14.2 ± 1.5
cm2) (95%CI 11.3-17.5)], when compared to non-treated animals [(10.4 ± 1.3 cm2)
(95%CI 7.8-12.9)] (Fig 9-5). This lack of improvement in CGRP response was similar to
the SP response in diabetic rats after late onset insulin treatment. Once again, early onset
insulin treatment normalized the response to CGRP in diabetic rats [(25.5 ± 4.2 cm2)
(95%CI 17.3-33.7)] (Fig 9-5).
The response to ES was more reduced in 4 weeks diabetic rats [(5.1 ± 2 cm2) (95%CI
1.2-9)] than 2 weeks diabetic animals [(8.6 ± 1.4 cm2) (95%CI 5.8-11.3)]. Late onset
insulin treatment significantly improved the ES response in diabetic rats [(14.3 ± 1 cm2)
(95%CI 12.3-16.3)], reaching 81% of control [(17.6 ± 1.3 cm2) (95%CI 15-20.1)],
whereas early onset insulin treatment normalized it (23.7 ± 2.2 cm2) (95%CI 19.4-28)
(Fig 9-6).
272
Area under the response curve (cm2 )
70
Control ES
Diab ES (2W)
Diab ES (4W)
Diab ES (4W) + late insulin intervention
Diab ES (4W) + early insulin intervention
60
50
40
30
+
20
10
x
*
*
0
Fig 9-6: Vascular responses to ES (20V, 5 Hz, 2 msec, for 1 minute) in control rats, diabetic (Diab) (2 &
4 weeks) rats, and diabetic rats (4 weeks) treated with insulin (2-6IU/100g/day)(late and early
interventions), presented as mean ± SEM (n = 4-8).
* Denotes significance from control rats (P<0.05)
+ Denotes significance from diabetic rats (4 weeks) (P<0.05)
X Denotes significance from diabetic rats (2 weeks) (P<0.05)
The results show that basal PE level (R1) was improved after both insulin treatments but
early intervention was more effective (Appendix B, Table B-7). Before any insulin
treatment, R1 in 4 weeks diabetic rats was 235% of control. This was decreased to 192%
after 2 weeks and 119% after 4 weeks of insulin treatment.
The PE response to SP (SP
2, 3 & 4)
was calculated as percent above basal ± SEM. SEMs
were omitted from Fig 9-7 for clarity but quoted in the text. The PE response to SP in
diabetic rats (9 ± 23%, 16 ± 32%, 0.1 ± 23%) was improved after late onset insulin
treatment (23 ± 15%, 27 ± 14%, 41 ± 10%) (Fig 9-7). However, this response was
normalized (53 ± 6%, 103 ± 16%, 117 ± 18%) and became relatively higher than control
(36 ± 19%, 72 ± 18%, 89 ± 14%) by early onset insulin treatment.
273
Control
Diab (2W)
Diab (4W)
Diab (4W) + late insulin intervention
Diab (4W) + early insulin intervention
(%) percent above basal
120
100
80
60
40
20
0
R1-SP2 R1-SP3
R1-SP4
Fig 9-7: PE in control rats, diabetic (Diab)(2 &4 weeks) rats, and diabetic rats (4 weeks) treated with
insulin (2-6IU/100g/day) (late and early interventions), presented as percent above basal ± SEM
(n = 4-10).
9.4.
Discussion
Evidence is accumulating to indicate a link between the adequate control of
hyperglycaemia in diabetes and the prevention of development and progression of
diabetic vascular complications (chapter 1, section 1.8.5.3). The role of hyperglycaemia
in diabetic animals has been examined previously in different tissues (i.e. sciatic nerve
and aorta) using a daily insulin treatment regimen (Bennett et al., 1998; Diemel et al.,
1992; Turk et al., 1999). Recently, Bennett et al 1999 investigated the role of insulin in
cutaneous microvascular responses in diabetic rats (Bennett et al., 1998; Diemel et al.,
1992; Turk et al., 1999). There are several differences between Bennett’s study and this
study. Two major differences are that in Bennett’s study a lower dose of STZ (50mg/kg,
i.p.) was used to induce diabetes in rats and the insulin treatment (4IU/day, using s.c.
insulin implants) was commenced 4 weeks after induction of diabetes.
In this study, a higher dose of STZ (75mg/kg) was used to induce diabetes in rats and
insulin treatment was introduced from day 1 or two weeks later at a concentration of 26IU/100g/day. This study was designed to examine the role of insulin treatment in the
modulation of skin microvascular blood flow in early (4 weeks) diabetes.
274
Late onset insulin treatment (2 weeks) did not improve the BBF in diabetic rats but early
onset treatment with insulin (4 weeks) improved BBF from 66% to 87% of control (Fig
9-1). The latter result raises the possibility that early onset insulin treatment may increase
NO production and / or prevents its deactivation. A previous study supports this
proposition by showing that insulin modulates endothelial nitric oxide synthase (eNOS)
expression in microvessels (Kuboki et al., 2000) (see also chapter 1, section 1.8.5.3, Fig
1-14). This treatment may also prevent NO deactivation by ROS. It was shown that
adequate control of blood glucose level with insulin treatment is capable of preventing
ROS production (Turk et al., 1999) and ROS are known to be able to inactivate both NO
and insulin activity (Paolisso and Giugliano, 1996). In addition, a previous study
suggested that hyperglycaemia or insulin deficiency might also be partly responsible for
the enhanced ET-1 production induced by ROS (Chen et al., 2000) (see also chapter 1,
section 1.8.4.1, Fig 1-10).
The difference in weight gain between early and late onset insulin treatment groups
reflects poor glycaemic control in the group of animals that received the later treatment
(Fig 9-2, Appendix D).
It was demonstrated that late insulin intervention improved the smooth muscle reactivity
in diabetic rats to 58% of control response, whereas early onset treatment normalized it
(Fig 9-3). An increase in SNP response after both treatments could be partially due to an
increase in cyclic guanosine monophosphate (cGMP) via NO mediated stimulation of
guanylate cyclase, which is supported by previous studies (Kuboki et al., 2000; Trovati et
al., 1996). Other possible mechanisms may be involved in increasing this response. For
example, two important factors, namely ROS and advanced glycated end-products
(AGEs), were shown to be involved in reducing smooth muscle reactivity in diabetic rats
by inactivating NO (Gryglewski et al., 1986; Suzuki et al., 1991). Early onset insulin
treatment may prevent the undesired effect of these factors by inhibiting ROS formation
and retarding the advanced glycation process (Rungby et al., 1992; Turk et al., 1999)
before it occurs. In addition, insulin treatment (<1 hour) might prevent ET-1-induced
vasoconstriction by attenuating intracellular calcium. There is evidence showing that
insulin reduces both voltage- and receptor- mediated calcium influx and stimulates both
275
the transcription and activity of Ca2+-ATPase in vascular smooth muscle cells (Hopfner
and Gopalakrishnan, 1999; Zemel, 1995). Early onset insulin treatment may either inhibit
excess production of ET-1 or down-regulate endothelin-A (ET-A) receptors on these cells
by enhancing subsequent intracellular calcium responses to ET-1 (chapter 1, section
1.8.4.1, Fig 1-11). An increase in messenger ribonucleic acid (mRNA) encoding Na+/K+ATPase was observed after short-term insulin treatment (1-6 hrs) of cultured smooth
muscle cells (Tirupattur et al., 1993). This may also explain the increase in the SNP
response in diabetic rats after insulin treatment in this study.
The absence of any improvement in the vascular response to SP in diabetic rats after late
onset insulin treatment (37% of control) was somewhat unexpected, suggesting that 2
weeks diabetes may have caused irreversible endothelial changes in these animals (Fig 94). In chapters 6, 7, and 8, however, it was demonstrated that ET-B receptor antagonist
(BQ-788), Tirilazad mesylate (Freedox), the combined ET-A receptor antagonist (BQ123) + superoxide dismutase (SOD), and early onset treatment with aminoguanidine
(AG) (4 weeks) can fully reverse the changes in microvascular responses induced by
diabetes. This suggests that substances such as free radicals, excess ETs, and AGEs are
produced during the first two weeks of diabetes and may well be involved in causing
vascular alterations. Hence, it is possible that once these factors are involved, correction
of hyperglycaemia per se might not be the most beneficial approach to solve this
problem. The best approach is to prevent the action of these factors or to prevent their
involvement from the beginning. The current results suggest that early onset insulin
treatment can protect microvessels against the effects of these agents and prevent
microvascular alterations in early diabetes.
Normalization of SP response after 4 weeks of insulin treatment (Fig 9-4) supports the
notion that early onset insulin treatment could prevent microvascular alterations in
diabetic rats by inhibiting the effects of ETs, free radicals, and AGEs. For example,
protection of smooth muscle cells against the effects of these factors could partially
explain the improvement in the SNP response in diabetic rats after 4 weeks treatment
with insulin. Protection of the vascular endothelium against the effect of ETs, free
radicals, and AGEs and preventing NO deactivation could be other outcomes of this
276
treatment. The increase in SP response above normal in 4 weeks insulin-treated rats
(110% of control) might be due to an enhancement in NO synthesis and cGMP level by
insulin treatment (Trovati et al., 1996).
The vascular responses to CGRP in diabetic rats after both early and late onset insulin
treatments showed a similar trend to those of SP (Fig 9-5). The normalization of the
CGRP response in diabetic rats after early onset insulin treatment, once again, could be
partially due to the normalization of the smooth muscle reactivity and partially due to an
increase in cyclic adenosine monophosphate (cAMP) by direct insulin action. There is
evidence showing that insulin increases the level of cAMP in smooth muscle cells
(Trovati et al., 1996).
The vascular response to ES in diabetic rats was improved to 81% of control after late
insulin intervention but was normalized after early insulin intervention (Fig 9-6). Part of
the improvement in ES response after either insulin treatments could be related to the
improvement in the smooth muscle reactivity in these animals. The greater ES response
after 4 weeks insulin treatment may be due to prevention of nerve ischaemia over time
partially by effective glycaemic control and this is supported by previous studies
(Biessels et al., 1996; Stevens et al., 1994).
In chapter 7, the ES response in control rats after perfusing SOD was increased to 126%
of control. It was suggested that the 26% increase in response above control could be
related to the ES technique and possible release of superoxide anions (Bassirat and
Khalil, 2000). In this study, the ES technique could also have caused the release of free
radicals such as superoxide anions. However, the early onset insulin treatment may
protect the antioxidant system in diabetic rats and encourage efficient removal of free
radicals and hence improvement in the ES response (135% of control).
Insulin plays a role in regulating vascular permeability in diabetic rats. The PE results
showed that 2 weeks treatment with insulin improved the R1 in diabetic rats and this
response was normalized only after 4 weeks insulin treatment (Appendix B, Table B-7).
This suggests that effective insulin treatment could restore PE towards normal in diabetic
277
rats, which is supported by a previous study (Bennett et al., 1998). The PE response to SP
(SP
2, 3 & 4)
in late onset insulin treated diabetic rats showed some improvements (Fig 9-
7). This suggests that after 2 weeks of diabetes some other factors such as ETs, free
radicals, and AGEs have control over endothelial function and that correcting for
hyperglycaemia per se is not sufficient to restore normal endothelial function. In support
of this proposition, the PE response to SP in early onset insulin treated diabetic rats was
improved to well above normal levels [(53%, 103%, 117%) compared to those in control
rats (36%, 72%, 89%)]. This improvement in response may be due to excess NO release
as a result of insulin’s direct effect on sensory nerves (Diemel et al., 1992).
Reversal of STZ effects through maintenance of near normoglycaemia levels with insulin
treatment also suggests that STZ effects are not due to a direct neurotoxic action of STZ
further support the results in chapter 3 (Part 1).
In conclusion, this study provided evidence that early onset insulin treatment could have
an important role in preventing skin microvascular damage by diabetes. The possibility
was raised that adequate control of hyperglycaemia at the early stages of diabetes may
reduce the long-term impact on the microvasculature. It was also demonstrated that
alteration in diabetic vascular permeability is reversible up to 4 weeks after the disease
onset (experimental period).
278
CHAPTER 10
Wound healing in diabetic rats
279
10.1.
Introduction
The process of wound healing has been described in detail in chapter 1, section 1.9.
Unmyelinated afferent nerves with polymodal nociceptors play an important role in
initiating the wound repair process (chapter 1, section 1.9.2.1). They mediate both early
and late components of the neurogenic inflammatory responses (vasodilatation (VD),
plasma extravasation (PE), and immune responses) (Brain, 1997; Eglezos et al., 1991;
Khalil and Helme, 1989; Khalil and Helme, 1990b). Among sensory neuropeptides that
mediate these effects, substance P (SP) and calcitonin gene-related peptide (CGRP) are
held to be the most important. Previous studies in wound healing have proposed a trophic
role for SP and CGRP (chapter 1, sections 1.9.2.1 & 1.9.2.2). Sensory peptides are also
important for the chemotactic activity of immune cells in skin wounds and for the
subsequent availability of certain growth factors essential for healing (Richards et al.,
1997).
Poor wound healing in diabetes has been documented (chapter 1, section 1.9.3). Previous
studies have suggested that fibroblast proliferation could be affected by lack of growth
factors (Doxey et al., 1995), dysfunctional immune cells (Rosenberg, 1990), and high
glucose levels (Hehenberger et al., 1999) This in turn could partially explain delayed
wound healing in 4 weeks diabetic rats (Ramamurthy et al., 1998).
Nitric oxide (NO) has been shown to play an important role in inflammation and / or
wound healing (Willerson, 1995). In 2 weeks diabetic rat skin, the level of woundinduced NO was decreased and this was paralleled with impaired wound reparative
collagen deposition (Schaffer et al., 1997).
The plasma level of endothelin-1 (ET-1) in diabetic rats and in normal rats with skin
wounds is increased (Makino and Kamata, 1998; Matzuzaki, 1993) and a role for
endothelins (ETs) in wound repair processes has been previously reported (Goligorsky et
al., 1999).
Insulin-like growth factor (IGF)-1 is a stimulator of glycosaminoglycan (GAG)
biosynthesis (Cechowska-Pasko et al., 1999) and is involved in wound healing in diabetic
rats (Tsuboi et al., 1995). It was shown that the reduction in IGF-1 and GAG levels in
diabetic rat skin could be due to insulin deficiency (Cechowska-Pasko et al., 1996a).
280
Free radicals have been linked with delayed wound healing in diabetes (Hallberg et al.,
1996). It has been suggested that insulin could improve repair processes by increasing the
activity of some antioxidants and production of hydrogen peroxide in rat macrophages
(Pereira et al., 1995), increasing NO production (Steinberg et al., 1994), and stimulation
of NO and ET activities (Cardillo et al., 1999).
It was suggested that aminoguanidine (AG) could prevent the inhibitory effect of diabetes
on angiogenesis and granulation tissue formation (Teixeira et al., 1999). In diabetes,
advanced glycated end-products (AGEs) could disregulate cytokine production and
increase tissue destruction (Lalla et al., 1998). After tissue has been damaged, AG may
prevent changes induced by the glycation process and AGEs, which include alterations to
macrophage phenotype and the impairment of cellular and molecular communications
(Iacopino, 1995).
Previous studies have focused mainly on histological and morphological changes
associated with poor wound healing, while this study is focusing on the mechanisms
underlying changes in wound repair process in early diabetes. Chapter 10 was designed
based on the information obtained in previous chapters. So far, results from the previous
chapters showed that certain treatment protocols could prevent the deleterious effects of
hyperglycaemia, free radicals, AGEs, and ETs on skin microvasculature in 4 weeks
diabetic rats. In the previous chapters insulin, Tirilazad mesylate (Freedox), AG, and the
combination of ET-A receptor antagonist (BQ-123) and a superoxide anion scavenger
(superoxide dismutase (SOD) were shown to be effective in improving microvascular
blood flow in diabetic rats. This current chapter examines the effect of those treatments
on tissue repair processes involved in full-thickness skin wounds in early diabetes. It was
hypothesized that since microvascular blood flow was improved by these treatments,
wound repair will be also improved in these animals.
281
10.2. Method
6 groups of rats were used (n = 5-6), one control, one diabetic, and 4 drug treated diabetic
groups. Control rats were treated with 0.1M sodium citrate buffer solution (CB-treated) at
the same time as other groups were treated with streptozotocin (STZ) (chapter 2, sections
2.4.1 & 2.4.3).
Animals
Control (CB-treated)
Diabetic (4 weeks)
NT
NT
PEG-SOD+BQ-123
AG
Tirilazad
Insulin
mesylate
Diabetes was induced in 5 groups of rats and after three days, when the onset of the
disease was confirmed, drug treatments were commenced in 3 of those groups (groups 3,
4, & 5). The control group and one of the diabetic groups (groups 1 & 2) received no
further treatments (NT). On day 28, a thermal wound was induced on the back of the
neck of all rats (see chapter 2, section 2.11.1). These wounds were monitored in all
groups until they were fully contracted (healed). Rats in groups 3, 4, and 5 continued to
receive their treatments (started on day 3) until wounds were healed. Rats in group 6
received five days treatment starting on day 28 when the wound was induced. The last
treatment (5th) for this group was on day 32 (see chapter 2, section 2.11.2).
Group 1: Control, no further treatment (NT)
Group 2: Diabetic, no further treatment (NT)
Group 3: Diabetic, treated with insulin
Group 4: Diabetic, treated with AG
Group 5: Diabetic, treated with Tirilazad mesylate (Freedox)
Group 6: Diabetic, treated with PEG-SOD (100U/ml) + BQ-123 (100μM)
Induction of thermal wound and measurement of healing are described in chapter 2,
sections 2.11.1 and 2.11.3.
282
10.2.1. Expression of data and statistical analyses
Wound sizes were traced and then calculated as areas (in cm2) using a digital planimeter.
These values are expressed in the text as mean ± standard error of the mean (SEM).
For statistical analysis, first a general two-way (group x day) repeated measures Analysis
of Variance (ANOVA) with a priori planned contrasts was performed to investigate
whether there were any significant group (control and diabetic) or treatment (insulin, AG,
Tirilazad mesylate, BQ-123 + SOD) differences. Then results were divided into 3 sets of
data, based on the information related to different stages of wound healing. The first stage
of the tissue repair process, the inflammatory stage, could occur during the first 2 days
after wound induction. This is followed by the proliferation and wound contraction stage,
between days 2-7. Finally, events leading to tissue remodelling and full wound closure
could occur between days 7-17.
For the first (days 1 & 2 of all groups) and second sets of data (days 2 to 7 of all groups)
a repeated measure ANOVA was performed. Where a significant main effect was
detected, post hoc comparisons were conducted using the Tukey HSD test. For the third
set of data (days 7 to 18 of all groups), first the average of every two days (days 7-9, 911, etc) were calculated and then a repeated measure ANOVA was performed. Where a
significant main effect was detected, again, the Tukey HSD test was used to perform post
hoc comparisons. The average of every two days was calculated for the third set of data,
as it was necessary to reduce the total number of results so they can be analyzed by the
statistical computer software used in this study.
In addition, to examine the overall rate of healing between groups, a one-way ANOVA
was performed. Where a significant main effect was detected, post hoc comparisons were
performed using the Tukey HSD test. The department of Maths and Statistics at the
University of Melbourne was consulted for the best way of analyzing these data and the
following results were obtained and presented based on their advice.
283
10.3.
Results
General observations
The time required for complete healing was monitored in 6 groups of rats: control,
diabetic (NT), and 4 diabetic treated groups. Due to the complexity of the statistical
analyses of the results, graphical representation of the significant differences was not
shown on the graphs in this chapter. Therefore, in each results section, the first paragraph
describes the statistically significant differences between treated, non-treated, and control
groups, based on the time required for full wound closure. Significant differences
between each treatment on individual days are also described in the results.
Figure 10-1 represents the general pattern of wound healing in these animals.
5.0
Control
Diab (4weeks)
Diab + insulin
Diab + AG
Diab + Freedox
Diab + BQ-123 + SOD
4.5
4.0
Wound size (cm2 )
3.5
3.0
2.5
2.0
1.5
1.0
0.5
0.0
0
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
Time (days)
Fig 10-1: Wound healing profile in control, diabetic (Diab) non-treated, and diabetic rats treated with
insulin, AG, Freedox (Tirilazad mesylate), and combined (BQ-123 + SOD), presented as mean
± SEM (n = 5- 6).
None of the treatments alter blood glucose level except in group 3 where insulin
treatment was used to control hyperglycaemia. In group 3 the average blood glucose level
was around 2-5mmol/lit and in other groups it was above 27mmol/lit during the
experimental period (approximately 7 weeks). In control rats (group 1) the number of
284
days for complete wound closure (14 ± 0.4 days) was significantly shorter than nontreated diabetic rats (group 2) (18 ± 0.2 days) (Fig 10-1). In insulin-treated diabetic rats
(group 3) the healing time was similar to AG-treated diabetic animals (group 4) (16 ± 0.3
& 16 ± 0.2 days respectively) and these were not significantly different to group 2. In
tirilazad mesylate-treated diabetic rats (group 5) wounds were fully closed after 13 ± 0.2
days, which was similar to those in BQ-123 + SOD-treated diabetic rats (group 6) (13 ±
0.3 days). The time required for full wound contraction in both groups 5 and 6 was
significantly reduced compared to group 2.
On day 1, there was no significant difference between average wound sizes of all groups.
All rats showed an early increase in wound sizes, which reached its maximum by day 2,
followed by a general sharp reduction in size after that day. All treated rats showed
similarity in their wound sizes especially on day 7. Based on the knowledge related to
different stages of wound healing, the results were divided into 3 separate sets of data as
mentioned earlier.
Wound healing in normal and non-treated diabetic rats (Groups 1 & 2)
The time required for full wound closure in control rats (14 ± 0.4 days) was significantly
shorter than non-treated diabetic animals (18 ± 0.2 days) (Fig 10-2).
285
5.0
Control
Diab (4weeks)
4.5
Wound size (cm2 )
4.0
3.5
3.0
2.5
2.0
1.5
1.0
0.5
0.0
0
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
Time (days)
Fig 10-2: Wound healing profile in control and diabetic (Diab) rats (both non-treated), presented as
mean ± SEM (n = 6).
On day 2, the average wound size in control rats [(4.95 ± 0.1 cm2) (95% CI 4.75-5.15)]
was significantly bigger than non-treated diabetic rats [(4.05 ± 0.3cm2) (95% CI 3.54.6)].
Generally, control rats showed steeper healing profiles when compared to non-treated
diabetic rats. In group 1, overall wounds were reduced more in size (by 2.5 cm2) between
days 2 and 7 compared to group 2 (by 1.3 cm2). However, the average wound sizes on
individual days were not significantly different, except on day 2.
After day 6, wounds in control rats continued to decrease until full closure was achieved
(on day 14). The average wound size in control rats was significantly reduced (by 40%)
between day 7 [(2.5 ± 0.05 cm2) (95% CI 2.4-2.6)] and day 9 [(1.45 ± 0.1 cm2) (95% CI
1.25-1.65)]. Wound contraction in non-treated diabetic rats was slow between days 2-6
and after day 14.
286
Wound healing in insulin-treated diabetic rats (group 3)
The time required for complete wound closure in insulin-treated diabetic rats (16 ± 0.3
days) was not significantly different from non-treated diabetic (18 ± 0.2 days) or control
animals (14 ± 0.4 days) (Fig 10-3).
5.0
Control
Diab (4weeks)
Diab + insulin
4.5
Wound size (cm2 )
4.0
3.5
3.0
2.5
2.0
1.5
1.0
0.5
0.0
0
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
Time (days)
Fig 10-3: Wound healing profile in control, diabetic (Diab), and insulin (2-6IU/100g)-treated diabetic
rats, presented as mean ± SEM (n = 5).
In group 3, an increase in average wound size from day 1 [(4.1 ± 0.3 cm2) (95% CI 3.54.7] to day 2 [(4.5 ± 0.2 cm2) (95% CI 4.1-4.9] was not significant. This increase in size
was more in control (by 0.55 cm2) and less in non-treated diabetic animals (by 0.15 cm2).
Between days 2-7, wounds in diabetic animals treated with insulin were reduced more in
size (by 2.1cm2) than those in group 2 (non-treated diabetic rats) (by 1.3cm2) but less
than those in group 1 (control) (by 2.5 cm2). The average wound size in insulin-treated
rats on day 3 [(3.1 ± 0.3 cm2) (95% CI 2.3-3.7)] and 4 [(3.2 ± 0.2 cm2) (95% CI 2.8-3.6)]
were smaller than their counterparts in non-treated diabetic rats [(3.8 ± 0.2 cm2) (95% CI
3.2-4.4) and (3.4 ± 0.2 cm2) (95% CI 3-3.8) respectively]. These results did not reach
statistical significance. However, on days 3 and 4, the wounds in insulin-treated rats were
significantly smaller than controls.
287
There was a significant reduction in wound size in group 3 animals between days 9 [(2 ±
0.1 cm2) (95% 1.8-2.2) and 11 [(1 ± 0.1 cm2) (95% CI 0.8-1.2)]. The time for full wound
contraction was reduced in insulin-treated animals from day 12 onwards.
Wound healing in AG-treated diabetic rats (group 4)
The time required for complete wound closure in AG-treated diabetic rats (16 ± 0.2 days)
was similar to the insulin-treated group (group 3) (16 ± 0.3 days) and was not
significantly different from control (14 ± 0.4 days) or non-treated diabetic group (18 ±
0.2 days) (Fig 10-4).
5.0
Control
Diab (4weeks)
Diab + AG
4.5
4.0
Wound size (cm2 )
3.5
3.0
2.5
2.0
1.5
1.0
0.5
0.0
0
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
Time (days)
Fig 10-4: Wound healing profile in control, diabetic (Diab), and AG (25mg/kg)-treated diabetic rats,
presented as mean ± SEM (n = 6).
The increase in wound size in group 4 animals, between days 1 and 2 (by 0.1 cm2), was
similar to non-treated diabetic rats (by 0.15 cm2).
The reduction in wound size, between days 2 and 7, in AG-treated rats (by 1.9 cm2) was
also greater and faster than the reduction in group 2 (non-treated diabetic rats) (by 1.3
cm2). Compared to non-treated diabetic rats, group 4 animals showed smaller wound
sizes on days 3 [(3.5 ± 0.2 cm2) (95% CI 3.1-3.9)] and 4 [(3.2 ± 0.3 cm2) (95% CI 2.6-
288
3.8)], which again did not reach statistical significance. These wounds, on the other hand,
were significantly smaller than those in control animals on days 3 and 4.
The profile of healing in group 4 animals, between days 5-12, resembled control rats. The
time required for full wound contraction was also reduced in these rats after day 12.
Wound healing in Tirilazad mesylate (Freedox)-treated diabetic rats (group 5)
The time required for full wound contraction in tirilazad mesylate-treated diabetic rats
(13 ± 0.2 days) was shorter than control (14 ± 0.2 days), insulin-treated, and AG-treated
diabetic rats. This required time was significantly shorter in tirilazad mesylate-treated
group when compared to non-treated diabetic group (18 ± 0.2 days) (Fig 10-5).
5.0
Control
Diab (4weeks)
Diab + Freedox
4.5
4.0
Wound size (cm2)
3.5
3.0
2.5
2.0
1.5
1.0
0.5
0.0
0
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
Time (days)
Fig 10-5: Wound healing profile in control, diabetic (Diab), and Freedox (Tirilazad mesylate) (5 mg/kg)treated diabetic rats, presented as mean ± SEM (n = 6).
On day 2, there was no significant difference between average wound size in this group
[(4.2 ± 0.2 cm2) (95%CI 3.8-4.6)] and group 1 (control animals) [(4.4 ± 0.1 cm2) (95% CI
4.2-4.6)] and group 2 (non-treated diabetic rats) [(3.9 ± 0.1cm2) (95% CI 3.7-4.1)].
However, from day 1 to day 2, wounds in group 5 animals increased more in size (by 0.4
cm2) than group 2 rats (by 0.15 cm2).
289
Between days 2-7, wounds in diabetic animals treated with tirilazad mesylate were
reduced more in size (by 2.25cm2) than those in group 2 (non-treated diabetic rats) (by
1.3cm2). In group 5, wounds on day 3 [(3.4 ± 0.2 cm2) (95% CI 3-3.8)] and 4 [(3.1 ± 0.2
cm2) (95% CI 2.7-3.5)] were similar in size to those in group 2 but were significantly
smaller than those in group 1 rats.
The profile of healing in group 5 animals, between days 5-9, resembled control animals.
Tirilazad mesylate-treated diabetic rats showed a significant reduction in wound size
between day 7 [(2.35 ± 0.2 cm2) (95% CI 2-2.7)] and day 9 [(1.4 ± 0.1 cm2) (95% CI 1.21.6)]. The time required for full wound contraction in these rats was also reduced after
day 11.
Wound healing in BQ-123 and SOD-treated diabetic rats (group 6)
The time required for full wound contraction in (BQ-123 + SOD)-treated diabetic rats (13
± 0.3 days) was similar to tirilazad mesylate-treated animals (group 5) (13 ± 0.2 days)
and shorter than other groups. However, compared to that in non-treated diabetic rats
(group 2) (18 ± 0.2 days) this required time was significance reduced (Fig 10-6).
5.0
Control
Diab (4weeks)
Diab + BQ-123 + SOD
4.5
4.0
Wound size (cm2 )
3.5
3.0
2.5
2.0
1.5
1.0
0.5
0.0
0
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
Time (days)
Fig 10-6: Wound healing profile in control, diabetic (Diab), and combined (BQ-123 + SOD) (100μM +
100U/ml)- treated diabetic rats, presented as mean ± SEM (n = 6).
290
Wounds in group 6 animals increased the most in size (by 0.75 cm2) from day 1 [(4 ± 0.2
cm2) (95% CI 3.6-4.4)] to day 2 [(4.75 ± 0.1 cm2) (95% CI 4.55-4.95)], compared to
other groups and reached statistical significance. The wound size on day 2 in group 6,
however, was not significantly different to those in group 1 (control animals) [(4.95 ± 0.1
cm2) (95% CI 4.75-5.15)] and group 2 (non-treated diabetic rats) (4.05 ± 0.3 cm2) (95%
CI 3.5-4.6)].
Between days 2-7, the extent of wound contraction in BQ123 + SOD-treated diabetic rats
was similar to control animals. Wounds were reduced by 2.35 cm2 in size in group 6,
during this period, which was similar to controls (by 2.5 cm2). In group 6 rats, the
average wound size on days 3 [(3.3 ± 0.3 cm2) (95% CI 2.7-3.9)] and 4 [(2.85 ± 0.2 cm2)
(95% CI 2.45-3.2)] were again smaller than those in non-treated diabetic animals but still
significantly different from those in control rats. The profile of healing in BQ123 + SODtreated diabetic rats, between days 5-9, resembled control animals.
Group 6 animals also showed a significant reduction in their wound size between days 711 (by 1.9 cm2). On day 9, wounds were significantly smaller [(1.46 ± 0.1 cm2) (95% CI
1.3-1.65)] than those on day 7 [(2.42 ± 0.1 cm2) (95% CI 2.2-2.6)]. However, by day 11,
wounds were even more reduced in size reaching only (0.55 ± 0.1 cm2) (95% CI 0.350.75). Group 6 also showed a significantly reduction in wound size after day 11.
291
Brief comparison of wound healing between diabetic treated groups (groups 3,4,5,6)
The number of days required for full wound closure in groups 5 (Tirilazad mesylatetreated rats) and 6 ((BQ-123 + SOD)-treated rats) was greater than 3 (insulin-treated
animals) and 4 (AG-treated rats) (Fig 10-7).
5.0
Diab + insulin
Diab + AG
Diab + Freedox
Diab + BQ-123 + SOD
4.5
4.0
Wound size (cm2 )
3.5
3.0
2.5
2.0
1.5
1.0
0.5
0.0
0
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
Time (days)
Fig 10-7: Wound healing profile in diabetic (Diab) rats treated with insulin, AG, Freedox (Tirilazad
mesylate), and combined (BQ-123 + SOD) (same concentrations as before), presented as mean
± SEM (n =5-6).
From day 1 to day 2, group 6 animals showed the most increase in wound size (by 0.75
cm2). This was followed by groups 5 and 3, both showing 0.4 cm2 increase in their
wound sizes, and finally group 4 with the least wound expansion (by 0.1 cm2).
Between days 2-7, wounds in group 6 animals contracted the most (by 2.35 cm2) and in
shorter period of time than any other treated diabetic group. Overall, the extent of wound
contraction in group 5 was similar to group 3 and 4. Wounds were reduced in group 5 by
2.25 cm2 during this period, which was similar to those in group 3 (by 2.1 cm2) and 4 (by
1.9 cm2).
Between days 7-9, groups 6 and 5 both showed a significant reduction in wound size (by
1 cm2 and 0.94 cm2 respectively), which resembled control in size (by 1 cm2) and the time
292
required for full wound contraction. Wounds on days 9-11 were also significantly
reduced in groups 3 (by 1cm2) and 6 (by 0.9 cm2). The wound contraction was
significantly less after day 12 in groups 3 and 4, and after day 11 in groups 5 and 6.
10.4.
Discussion
Wound repair is a complex process that involves inflammation, granulation, and tissue
remodeling. Inflammation is a protective response, which initiates the process of tissue
repair and intact unmyelinated primary afferent sensory nerves play a major role in this
process (chapter 1, section 1.9.2). It has been documented that diabetic individuals have
poor wound healing, among other complications (Nathan, 1993; Schaffer et al., 1997).
The focus of this study was to examine the mechanisms underlying poor tissue repair in
early diabetes, an area that has not been previously explored. In earlier chapters, the
involvement of hyperglycaemia, AGEs, reactive oxygen species (ROS), and ET in
reducing inflammatory responses in early diabetes was established. It was demonstrated
that treatment with insulin, AG, tirilazad mesylate, and perfusion of the combined BQ123 + SOD over the blister base can improve these responses in 4 weeks diabetic rats. In
this chapter the role of hyperglycaemia, AGEs, ROS, and ET in the repair of fullthickness skin wounds in diabetic condition was examined using the above agents. The
changes in wound size were monitored over time until full wound closure occurred.
Full-thickness wounds, including the thermal burn wound, are known to heal by reepithelialization and contraction (Champion et al., 1970). It should be noted that fullthickness burn wounds in humans could be complicated by excessive connective tissue
remodeling and permanent scar formation (Wood and Bladon, 1985). In this chapter,
wound size and time for complete wound closure were measured following a standard
thermal injury (chapter 2, section 2.11). The change in wound size, as monitored by the
open wound area, was used as the outcome measure of wound healing. It is believed that
these measurements, documenting the change in surface wound area over time, could be
representative of two aspects of wound healing, namely, re-epithelialization and
contraction. However, because quantitative histological and immunohistochemical
techniques were not applied in this study, it is conceded that assessment of wound size as
293
reported here might not exactly reflect the true extent of wound re-epithelialization and
contraction (Khalil and Helme, 1990b).
The results demonstrated that, in general, diabetic animals do not show a significant
wound expansion on day 2 (Fig 10-1). Within hours of skin injury, there is an acute
(non-proliferative) inflammation characterized by vasodilatation, oedema, and neutrophil
and mononuclear cell extravascular accumulation (coupled with death and dissolution of
previously injured cells and extracellular fibrils) (Wood and Bladon, 1985). Shortly after,
a band of polymorphonuclear cells forms while epidermal cells (within 24hrs) start to
migrate from the margin of the wound towards the injured area. Delayed wound
expansion in diabetic rats may be due to a delayed response to injury and impaired
functioning of the immune cells in diabetes, which is supported by previous studies on
polymorphonuclear leucocytes (PMNs) and fibroblasts (Hehenberger et al., 1999;
Rosenberg, 1990).
In diabetes, impaired sensory C-fibres could alter the production and / or release of
inflammatory mediators (e.g. SP, CGRP) (Garrett et al., 1996; Nemeth et al., 1999a;
Rittenhouse et al., 1996a). It was previously reported that in diabetic rats, inflammatory
responses are reduced (Nemeth et al., 1999b) and the results in chapter 3 are in agreement
with this report. Hence, it is possible that the delay in response to injury in diabetic rats is
related to a reduction in inflammatory responses and sensory neuropeptides. This is
supported by previous studies showing that sensory neuropeptides are important in
vasodilatation (Khalil and Helme, 1990b), plasma extravasation (Khalil and Helme,
1989), skin fibroblasts proliferation (Nilsson et al., 1985), neutrophil migration, and
modulation of immune response (Eglezos et al., 1991). The delayed response to injury in
diabetes may also be related to the role of SP in facilitating the arrival of important cells
to the injured site by increasing vessel permeability and dilatation (Holzer, 1998). A
reduction in vascular permeability (Mathison and Davison, 1993), due to a reduced NO
synthesis in early diabetic wounds (Schaffer et al., 1997), might inhibit the arrival of
important cells and / or substances into the surrounds of the injured site (Shuler et al.,
1995). A reduction in blood flow, due to a vasoconstrictive environment created by
reduced NO, and the loss of the effects of the endogenous inhibitors of platelet
294
aggregation and vasoconstriction at the site of endothelial injury (Willerson, 1995) may
also contribute to this delay.
The results in this chapter showed a general delay in wound contraction in non-treated
diabetic rats when compared to control animals (Fig 10-1), which is in agreement with
previous studies (Franzen and Roberg, 1995; Ramamurthy et al., 1998). The data also
showed that all wounds in non-treated diabetic rats, between days 2 and 7, contracted by
only 1.3 cm2 when compared to a contraction of 2.5 cm2 in control rats. This period of
time could represent the proliferative phase of the repair process. At the reepithelialization stage, epidermal mitotic activity is increased, just beyond the edge of the
wounded zone, producing new cells until the wound is covered with epidermis. New
blood vessels also form within the tissue with the aid of growth factors and correct
communication between the cells involved. It is possible that in diabetes the reduction in
growth factors, such as basic fibroblast growth factor (bFGF), and changes in post
terminal-effects mediated by neuropeptides, such as SP, could diminish reepithelialization and neovascularization and hence reduce wound contraction. This
proposition is supported by a previous study showing acceleration in wound healing by
bFGF in diabetic mice (Okumura et al., 1996), via a mechanism that involves enhancing
angiogenesis and granulation tissue formation.
It has also been suggested that a decrease in neovascularization and less organization at
the late inflammatory phase of wound healing in diabetic mice could be due to alterations
in leucocyte infiltration and interleukin (IL)-6 levels (Fahey et al., 1991). It is also known
that SP can stimulate cytokine production by oesinophils and macrophages (Lotz et al.,
1988) as well as being required for the chemotactic activity of monocytes, macrophages,
and T-lymphocytes in skin wounds (Richards et al., 1997). These studies raised the
possibility that alterations in SP content might not only reduce the production of growth
factors required for this stage of wound healing but also interfere with macrophage
function. This possibility is supported by the study of Fahey et al 1991.
Changes in SP mediated events in diabetes involving platelet-derived growth factor
(PDGF) could lead to a delay in neovascularization. This is supported by previous studies
295
showing an initial (first 5 days) increase in PDGF to be absent in wounded tissue of
diabetic rats due to a decrease in its cellular production (Doxey et al., 1995).
Reduction in NO in early diabetic wounds (Schaffer et al., 1997) could interfere with
angiogenesis by effecting endothelial cell migration (Murohara et al., 1999). This may
explain how changes in post-terminal events mediated by SP in early diabetes could
interfere with angiogenesis. The results in chapter 4 suggested that NO synthesis (by
NOS) is not altered in 4 weeks diabetic rats but the release of certain substances during
this disease condition may compromise its activity. The reduction in NO activity could
ultimately play a role in impaired wound healing in diabetes.
After day 7, wound contraction in diabetic rats was generally slower than control
animals. A greater reduction in wound size was observed in control rats between days 7-9
(Fig 10-2). The time for full wound contraction was also significantly reduced in diabetic
rats after day 14. The final stage of healing is the remodelling phase, during which the
dermis responds to injury with the production of collagen and matrix proteins. In the
current study, slow healing and wound contraction after day 7 in diabetic rats suggests
that the remodelling phase of the healing process, including production of collagen and
matrix proteins, might be compromised in these animals. This proposition is in agreement
with previous studies (Ramamurthy et al., 1998; Sato et al., 1999).
Growth factors have a profound influence on extracellular matrix formation (Greenhalgh,
1996). Hence, it is possible that the reduction in the levels of sensory peptides in these
animals may indirectly influence all stages of matrix formation. Impaired healing in 4
weeks diabetic rats may also be related to reduced IGF-1 biosynthesis in the skin
(Blakytny et al., 2000).
NO is involved in collagen deposition and it was reported that this stage of healing is
compromized in diabetes (Schaffer et al., 1997). The results in chapter 4 suggest that, in
early diabetes, the activity of NO is affected. These raise the possibility that slow healing
and wound contraction in diabetic rats, after day 7, might be related to impaired collagen
deposition, due to inactivation of NO by the released substances in the diabetic condition.
296
Role of insulin
From day 1 to 2, wounds in group 3 (insulin-treated animals) increased more in size (by
0.4 cm2) than group 2 (non-treated diabetics) (by 0.15 cm2). A significant wound
contraction (by 2.1cm2) was also observed in insulin-treated diabetic rats between days 27 (Fig 10-3). This suggests that insulin treatment could play a role in improving the
inflammatory phase of the tissue repair process (e.g. VD & PE ). The results in chapter 9,
showing normalization of the vascular responses to SP, CGRP, ES and the basal PE
response (R1) in early onset insulin treated diabetic rats, is supportive of this notion.
The vasodilation ability of insulin has been previously documented (Baron, 1999). It was
also shown that insulin could increase NO production (Steinberg et al., 1994) by
regulating endothelial nitric oxide synthase (eNOS) gene expression (Kuboki et al.,
2000). Insulin could also cause vasodilation by increasing the production of hydrogen
peroxide in rat macrophages (Pereira et al., 1995). Hydrogen peroxide was previously
shown to have vasodilator ability (Boyer et al., 1995). Furthermore, a daily insulin
treatment regime similar to that used in this study was shown to be able to prevent
alterations in vasoreactivity and endothelial cell function in 4 weeks diabetic rats
(Stevens et al., 1993). All these effects could create a vasorelaxant environment at the site
of injury in diabetic rats. Insulin treatment can restore neurogenic oedema towards
normal (Bennett et al., 1998) and this may improve the transport of inflammatory /
immune cells to the injured site and improve the response to tissue injury in diabetic rats.
The number of days required for complete wound closure in insulin-treated diabetic rats
(group 3) (16 ± 0.3 days) was shorter than non-treated diabetic rats (group 2) (18 ± 0.2
days) (Fig 10-3). Also, the level of blood glucose measured daily in 4 weeks diabetic rats,
during the experimental period, indicated that the insulin treatment managed to keep the
blood glucose level close to the normal range (2-5mmol/lit) in these animals. One
possible explanation for increased wound contraction in these animals could be related to
the level of glucose. Insulin treatment and a near normal level of glucose in rats may
improve fibroblast proliferation (Hehenberger et al., 1999) by influencing cell surface
integrins (Wertheimer et al., 1998). There is evidence showing that an increase in NO by
insulin (Steinberg et al., 1994) could improve angiogenesis by supporting endothelial cell
migration via an integrin-dependent mechanism. Insulin was also shown to be able to
297
increase the activity of the antioxidant system (Cu-Zn-SOD, catalase, glutathione
peroxidase) in diabetic rats (Pereira et al., 1995). An increase in endogenous antioxidants
may protect NO against deactivation and, together with an increase in NO production,
could improve angiogenesis in insulin-treated animals. The results in chapter 7 showing
an improvement in SP response in 4 weeks diabetic rats with antioxidants (e.g. SOD, Nacetylcysteine (NAC), and Tirilazad mesylate) supports the possible involvement of ROS
in deactivating NO. Both an increase in fibroblast proliferation and an improvement in
angiogenesis may explain the increase in wound contraction on days 2-7 in insulintreated diabetic rats.
Between days 7-11, full wound contraction in group 3 animals (insulin-treated animals)
occurred in a shorter period of time than non-treated diabetic rats but slightly longer than
control animals. There was also a similarity between the extent of wound contraction in
groups 3 and 6 (combined BQ-123 + SOD-treated animals) between days 9-11 (Figs 10-3
& 10-6). This may be partially related to insulin capable of improving the efficiency of
the antioxidant system in diabetic rats and hence protecting NO against deactivation by
ROS (Pereira et al., 1995).
It is known that collagen is the only protein susceptible to fragmentation by superoxide
anions (Monboisse and Borel, 1992). Therefore, in group 3, part of the improvement in
wound repair after day 9 might be related to an increased activity of SOD by insulin
(Pereira et al., 1995). An increase in antioxidant activity in general may also reduce
levels of hydroxyl radicals, as they are known to denature collagen (Arisawa et al., 1996).
Insulin was previously shown to improve matrix formation (Pierre et al., 1998) and
reparative collagen deposition in diabetic animals (Schaffer et al., 1997). Insulin may
have caused this improvement by increasing GAG biosynthesis (Cechowska-Pasko et al.,
1999) and reducing its susceptibility to degradation (Cechowska-Pasko et al., 1996a) by
an increase in wound collagenase (Ramamurthy et al., 1998). This may contribute to the
faster wound contraction in group 3 (16 days) compared to non-treated animals (18 days)
(Fig 10-3).
Overall, the current results suggests that one of the major roles of insulin in improving
microvascular blood flow and wound contraction in diabetic rats could be related to the
reduction in glucose concentration in these rats.
298
Role of AG
The number of days required for full wound contraction in AG-treated diabetic rats
(group 4) was reduced (16 ± 0.2 days) when compared to non-treated diabetic rats (group
2) (18 ± 0.2 days) (Fig 10-4). From day 1 to 2, the increase in wound size in group 4 (by
0.1 cm2) was similar to group 2 animals (by 0.15 cm2). This suggests that other factors,
besides AGEs, may be involved, which could exert a greater effect on the inflammatory
phase of the tissue repair process in diabetic rats.
Between days 2-7, the reduction in wound size in group 4 (AG-treated animals) (by 1.9
cm2) was greater than those in non-treated diabetics (by 1.3 cm2). Structurally, AGEs are
known to decrease vessel elasticity, increase its thickness, increase its rigidity, and
narrow the vessel lumen (Bierhaus et al., 1998), which could affect blood flow in diabetic
rats. AG may reduce these effects and contribute to improvement in blood flow in
diabetic rats by inhibiting AGEs formation. Unmyelinated afferent sensory nerves and
their peptides play an important role in initiating the wound repair process (chapter 1,
section 1.9.2). Previous studies reported an increase in sciatic nerve tubulin glycation in 2
weeks diabetic rats (Cullum et al., 1991) and a reduction in axonal nerve transport
(Tomlinson et al., 1988). AG may preserve sensory nerve function by interfering with the
glycation process (involving tubulins) and in turn may influence blood flow and
angiogenesis (Teixeira et al., 1999). In chapter 8, evidence was provided that the reduced
inflammatory responses (e.g. SP, ES, PE) in 4 weeks diabetic rats can be normalized by
early onset AG treatment.
AGEs are also known to affect monocyte and macrophage phenotypes. Monocytes /
macrophages are chemotactic to AGEs and are involved in removal of AGEs from the
system and replacement of vascular matrix macromolecules that cross-link by exposure
to glucose, hence participating in normal tissue turnover (Nawroth et al., 1999). AGEs
formation in diabetes may alter the macrophage / monocyte phenotype by disregulating
their cytokine production, resulting in increased inflammatory tissue destruction (Grossi
and Genco, 1998). Growth factors are known to influence all stages of the healing
process, hence any alterations in their levels could cause delayed wound healing
(Greenhalgh, 1996). Platelet glycation might be another reason for delayed wound
contraction at the early stages of the repair process. In addition, factors released by
299
platelets have been shown to be capable of stimulating fibroblast contraction (Goto et al.,
1998).
Inhibition of AGEs by AG may have also improved the proliferative effect of NO (Hogan
et al., 1992) and its vasodilator activity (Bucala et al., 1991). Results in chapter 8 showed
a significant improvement in SP responses in 4 weeks diabetic rats after treatment with
AG, which supports the above studies.
It was suggested that AG can create a vasorelaxant environment by quenching hydroxyl
radicals and inhibiting ET-1 induced by AGEs (Bierhaus et al., 1998). Free radicals
produced by AGEs (Taniguchi et al., 1996) might also encourage the ET-1 effect in this
situation by deactivating NO activity. Evidence was presented in chapters 6 and 7 that
ET-1 and hydroxyl radicals are involved in reducing microvascular blood flow in 4
weeks diabetic rats. Therefore, one can speculate that a possible underlying mechanism
for impaired tissue healing in early diabetic rats may involve free radicals and ET-1. AG
treatment could reduce the effect of these agents and improve microvascular blood flow
and ultimately tissue repair processes, by preventing AGEs formation.
Another study suggested that impaired wound healing in diabetic organisms are due, at
least in part, to the enhanced interaction of AGEs with cellular receptors for AGE
(RAGE) (Lalla et al., 1998). The data in chapter 8 showed that short-term exposure of
diabetic (4 weeks) microvasculature to anti-RAGE IgG could significantly improve (e.g.
response to SP) or normalize vasodilatation (e.g. response to SNP). AGE-RAGE
interaction not only generates free radicals but also causes a reduction in the cellular
antioxidant defence mechanism (Kashiwagi et al., 1996; Mak et al., 1996; Mullarkey et
al., 1990). Both of these effects could lead to NO inactivity and in turn, affect blood flow
and angiogenesis in diabetic rats (Teixeira et al., 1999). Prevention of AGEs formation by
AG may explain the accelerated wound contraction between days 2-7 in 4 weeks diabetic
rats treated with AG.
Wound contraction in AG-treated rats between days 5-12 was faster than non-diabetic
rats. Alteration in the macrophage / monocyte phenotypes could enhance inflammatory
tissue destruction and connective tissue degradation (Grossi and Genco, 1998).
Prevention of AGEs formation by AG might reduce this effect and explain the observed
accelerated wound contraction in AG-treated rats between those days. AG could also
300
accelerate protein turn over by making glycated molecules more susceptible to
proteolysis (Skamarauskas et al., 1996) and increases granulation tissue formation, a
process that is inhibited in diabetes (Teixeira et al., 1999).
ROS can inhibit NO in diabetes (Bucala et al., 1991) and could be generated by AGEs
(Taniguchi et al., 1996). Both reduction in NO and presence of ROS in diabetes have
been implicated in impairment in collagen deposition (Arisawa et al., 1996; Schaffer et
al., 1997). Results in chapter 8 showed that short-term perfusion of anti-RAGE IgG and
long-term treatment with AG could significantly improve vasodilatation in early diabetes.
Therefore, the possibility was raised that this improvement might be related to the
reduction of some deleterious effects of endogenous AGEs. These include inhibition of
NO, generation of free radicals, and depletion of antioxidant defence system (Bucala et
al., 1991; Ceriello, 1999; Galle et al., 1998; Kashiwagi et al., 1996).
Overall, it can be speculated that the effectiveness of the AG treatment in reducing the
time required for full wound closure (16 ± 0.2 days) could be related to inhibiting AGEs
formation and protecting the microvasculature against their deleterious effects.
Role of Tirilazad mesylate(Freedox)
In tirilazad mesylate-treated animals (group 5) the number of days required for full
wound closure (13 ± 0.2 days) was shorter than controls (group 1) (14 ± 0.2 days) (Fig
10-5). In the first two days, wound expansion in group 5 (by 0.4 cm2) showed similarity
to those in group 1 (by 0.55 cm2). Between days 2-7, the extent of wound contraction in
group 5 was similar to group 1. This might be the result of an improvement in blood
flow, endothelial cell migration, and wound reparative collagen deposition induced by
tirilazad mesylate treatment. It is known that tirilazad mesylate preserves vitamin E (an
antioxidant) content in cell membranes and prevents ROS production in diabetes (Aragno
et al., 2000). Both of these could protect cells against damage (cytoprotective property).
Furthermore, tirilazad mesylate could improve inflammatory responses by scavenging
free radical intermediates to protect NO against deactivation (Hayakawa and Raij, 1999;
Munzel et al., 1997; Saniova, 1997). The results in chapter 7 provided further support for
this possible role of tirilazad mesylate in 4 weeks diabetic rats.
301
The similarity between the vascular response to ES in diabetic rats after tirilazad
mesylate treatment and after NAC perfusion (results in chapter 7) suggest that part of the
inhibitory action of hydroxyl radicals on the ES response could be prevented by tirilazad
mesylate treatment. The neuroprotective action of tirilazad mesylate (Cameron and
Cotter, 1999) might play a role in improving microvascular blood flow in diabetic rats
with skin wounds. It was shown that preventing alterations in sensory nerve function
could improve neutrophil migration (Eglezos et al., 1991) and skin fibroblast and
endothelial cell proliferation (Ziche et al., 1990). This may explain significant differences
in average wound size on days 2 and 7 between tirilazad mesylate-treated diabetic rats
and non-treated diabetic rats.
The profile of healing in group 5, between days 5-9, resembled control animals. The
significant reduction in wound size that was observed in group 5, between days 7-9, (by
0.94 cm2) was also similar to those in control rats (by 1 cm2). A previous study showed
that the reduction in NO levels in 2 weeks diabetic rats skin was paralleled with impaired
wound reparative collagen deposition (Schaffer et al., 1997). Furthermore, it was
suggested that tirilazad mesylate might protect NO against ROS and products of lipid
peroxidation (Munzel et al., 1997; Saniova, 1997). This action of tirilazad mesylate might
affect collagen deposition in the treated diabetic rats and contribute to an increase in their
wound contraction. Tirilazad mesylate might have also protected collagen against the
deleterious effects of ROS. This is supported by earlier studies demonstrating collagen
denaturation by hydroxyl radicals (Arisawa et al., 1996) and collagen susceptibility to
fragmentation by superoxide anions (Monboisse and Borel, 1992). In addition, it has been
proposed that tirilazad mesylate inhibits the reaction between products of lipid
peroxidation produced in diabetes and GAG (Muller et al., 1986; Pieper et al., 1997a) and
improves protein synthesis (e.g. exracellular matrix components). These effects of
tirilazad mesylate might also play a role in improving tissue repair in diabetic animals.
Overall, the current results suggest that the major role of tirilazad mesylate in improving
microvascular blood flow and wound contraction in diabetic rats could be mainly related
to its natural protective property, which is inhibiting oxidative damage.
302
Role of BQ-123+SOD
Results in chapters 6 and 7 showed that the local perfusion of either BQ-123 or SOD over
the blister base could improve microvascular blood flow in 4 weeks diabetic rats.
However, only the combined perfusion of BQ-123 + SOD normalized all microvascular
responses in these animals. It was demonstrated that perfusion of combined BQ-123 +
SOD over the blister base caused a greater improvement in microvascular blood flow
than tirilazad mesylate treatment (chapter 7).
In this study, the number of days required for full wound closure in group 6 (BQ-123 +
SOD-treated rats) (13 ± 0.3 days) was similar to group 5 (tirilazad mesylate-treated rats)
(13 ± 0.2 days) (Figs 10-5 & 10-6). These results raise the possibility that both ROS and
ETs could play a major in tissue repair processes.
From days 1-2, the combined BQ-123 + SOD treatment normalized wound expansion in
diabetic rats. Between days 2-7, wounds in group 6 reduced the most in size (by 2.35
cm2) compared to all other treated groups. This result suggests that the proliferative stage
of healing in 4 weeks diabetic rats may be improved. A possible explanation for these
effects is best described by examining the roles of individual substances used in this
combined treatment. For example, an increase in ET-1 was shown to down-regulate
endothelin-B (ET-B) receptors via intracellular calcium immobilization in diabetes
(Masao et al., 1999). BQ-123 can block ET-A receptors, thus inhibiting calcium
immobilization and vasoconstriction (Goto et al., 1998). Therefore, it is possible that BQ123 not only prevented vasoconstriction by inhibiting ET-1 activity but also increased
vasodilation by blocking ET-A receptors and encouraging NO release by activating ET-B
receptors (Fozard and Part, 1992). Evidence was provided in chapter 6 for the
involvement of both ET-A and ET-B receptors in diabetes and the alteration in their
activity after 4 weeks duration.
It is known that the level of ROS is increased in diabetes and the efficiency of the
antioxidant defence system is compromized (Kamata and Kobayashi, 1996; Mak et al.,
1996). In group 6, SOD might prevent the chain reaction that leads to further free radical
formation in these rats, allowing the antioxidant system to function more efficiently to
prevent further tissue damage. This is supported by a previous study showing SOD can
neutralize superoxide anions as well as inhibiting formation of other free radicals
303
(hydroxyl radical, peroxynitrite) (Thomas, 1995). The effect of BQ-123 + SOD on NO
activity might accelerate wound closure by one or all of the following mechanisms:
increasing vasodilatation (Ralevic et al., 1992); reducing activity of vasoconstrictors
(Stohs, 1995); improving vascular permeability (Shuler et al., 1995); stimulating
transmigration of endothelial cells (Goligorsky et al., 1999), stimulating angiogenesis
(Murohara et al., 1999). In chapter 6, the involvement of ETs (e.g. ET-1) and free
radicals (e.g. superoxide anions) in modulating inflammatory responses (VD & PE) in 4
weeks diabetic rats was demonstrated. Hence, it is possible that ET-1 and superoxide
anions inhibit NO activity, altering microvascular blood flow in 4 weeks diabetic rats
with skin wounds. The study by Cameron et al (1991) suggested that SOD could affect
nerve conduction velocity (Cameron et al., 1991), which may also play a positive role by
improving diabetic microvascular blood flow and skin wound repair in the current study.
The profile of healing in group 6, between days 5-9, resembled the control group (Fig 106). Between days 7-9, the reduction in wound size in group 6 (by 1 cm2) resembled the
control group (by 1cm2) and tirilazad mesylate-treated animals (group 5) (by 0.94 cm2).
Between days 9-11, a significant reduction in wound size was also observed in group 6
animals (by 0.9 cm2), which was greater than group 5 (by 0.8 cm2) and far greater than
controls (by 0.45 cm2). These results might be explained by the joint effects of combined
BQ-123 + SOD in increasing matrix production and reducing its degradation. A previous
study showed that ET-1 is involved in extracellular matrix turnover through ET-A
receptors (Gutierrez et al., 1996). This raised the possibility that BQ-123, by blocking
ET-1 activity, might have improved proteoglycan metabolism. ET-1 is also capable of
increasing collagen production (Gutierrez et al., 1996) but this could be inhibited by
products of lipid peroxidation (Muller et al., 1986).
Free radicals are known to cause lipid peroxidation (Stohs, 1995), generating products
that can react with proteins and GAG, thus reducing enzyme activity and protein
synthesis (Muller et al., 1986). SOD could play an important role in protecting
intracellular matrix components against degradation by free radicals (Monboisse and
Borel, 1992). Hence, a possible role for SOD in improving reparative wound collagen
deposition by preventing NO deactivation cannot be ignored (Schaffer et al., 1997). Since
in the diabetic condition the level of free radicals and ETs and the activity of antioxidants
304
are altered (Chapter 1, sections 1.5.3. & 1.8.4.1), the above factors might negatively
affect the final stage of the tissue repair process and contribute to delayed wound
contraction in this study.
Between days 2-7, the similarity in the extent of wound contraction between group 6 and
group 5 suggests that ET-1 and free radicals might equally inhibit the proliferative stage
of the tissue repair process in 4 weeks diabetic rats. However, between days 7-11, the
extent of wound contraction in group 6 was greater than group 5 suggesting that ETs
could exert a greater inhibitory effect on the repair process, especially at the later stages.
In conclusion, this study showed that all treatments used have contributed, to varying
degrees, to different stages of wound repair and to the overall improvement in repair
processes in diabetic rats. The most effective treatments were tirilazad mesylate and the
combined BQ-123 + SOD while AG and insulin were the least effective treatments.
Insulin and AG improved tissue repair processes that were mainly observed towards the
proliferative stage, with only insulin playing a role in the tissue remodelling stage.
tirilazad mesylate and the combined BQ-123 + SOD treatments affected both the
proliferative and tissue remodelling stages of the repair process. Furthermore,
pharmacological manipulation at the level of the microvasculature (e.g. local injection of
BQ-123 + SOD) and systemic treatments (e.g. insulin, AG, and Tirilazad mesylate) were
shown to improve tissue repair processes in early diabetes. This work raised the exciting
possibility that early intervention could prevent microvascular alterations and this could
be of clinical importance. It also indicates that further research is needed to determine the
impact of early intervention on decelerating late diabetic complications.
Publication: Bassirat M and Khalil Z. Factors modulating microvascular blood flow in
streptozotocin-induced diabetic rats: Relevance to tissue repair (In preparation) (2001)
305
CHAPTER 11
General discussion
306
11.1. General overview and Methodology
Delayed tissue repair is a serious complication of long-term diabetes mellitus. The effect
of this disease condition on skin microvasculature has not been widely studied and no
clear evidence exists regarding the mechanism(s) involved in the modulation of skin
blood flow and tissue repair processes during early diabetes. The research undertaken
within this thesis aimed to investigate the effect of early diabetes in modulating
microvascular blood flow, directly and indirectly, via modulating sensory nerve activity.
The effect of these changes on tissue repair processes in the skin of 4 weeks
streptozotocin (STZ)-induced diabetic rats was also examined.
The neurovascular system, comprising of unmyelinated primary afferent sensory
neurones and the microvasculature innervated by these nerves both play major roles in
modulating the inflammatory and tissue repair processes. Sensory nerve terminals
respond to injury via the release of sensory neuropeptides, which mediate inflammation
and tissue repair. Diabetes is known to impact on the function of the sensory nerves and
skin microvasculature, leading to a reduction in the inflammatory responses (Gamse and
Jancso, 1985; Mathison and Davison, 1993) and delay in tissue repair (Fahey et al., 1991;
Schaffer et al., 1997).
The primary aim of this research was to examine the involvement of a number of factors,
namely, hyperglycaemia, endothelins (ETs), free radicals, and advanced glycated endproducts (AGEs) in the changes induced during early diabetes in the skin microvascular
blood flow and, ultimately, tissue repair processes. The literature has suggested that these
factors play a role in the development of long-term diabetic vascular complications
(chapter 1, sections 1.5.2, 1.5.3 & 1.8.4).
Chapter 11 provides an overview of the methodology employed and the methodological
limitations encountered, together with the findings from the studies undertaken in this
thesis. In addition, directions for future research, including the use of other animal
models of diabetes, are proposed.
307
Chapter 1 presented a review of the literature describing diabetes mellitus and its
complications, inflammation, processes of tissue repair, and the relationship between
diabetes and repair processes. The mechanisms involved in long-term diabetes
complications have also been described in chapter 1. The focus of chapter 1 was to
present a review of the current knowledge regarding alterations in the microvasculature
and tissue repair processes in the diabetic condition. Review of the literature revealed that
information regarding changes in skin microvascular function and the effect of these
changes on tissue repair processes, especially in the early stages of diabetes, is very
limited and requires further research. Any new information concerning this issue could be
of clinical importance and could play a role in devising early intervention approaches to
prevent microvascular alterations and decelerate late diabetic complications.
Based on this review, two hypotheses were raised and examined in this study:
First, that factors implicated in long-term diabetic vascular damage could play a role in
altering skin microvascular function in early diabetes (4 weeks streptozotocin (STZ)induced diabetic rats).
Second, that preventing the deleterious effects of these factors could improve skin
microvascular blood flow and skin repair processes in early diabetes.
Specifically, the thesis examined the involvement of hyperglycaemia, ETs, free radicals,
and AGEs in modulating skin microvascular blood flow and their effects on tissue repair
processes in early diabetes.
The second chapter described the general methodology including:
•
The induction of diabetes with STZ and alloxan (in Sprague-Dawley rats)
•
Methods used to induce an inflammatory response
•
Methods used to measure two specific parameters of this response, namely
vasodilatation (VD) (using laser Doppler flowmetry) and plasma extravasation (PE)
•
Procedures to electrically stimulate sensory C-fibres
•
Procedures to induce a full-thickness wound
308
11.2. Neuronal and local tissue factors modulating microvascular blood flow in
early diabetes
According to the literature, the presence of high glucose levels in vivo can alter a number
of biochemical processes (Hopfner and Gopalakrishnan, 1999). This in turn may affect
the function of sensory nerves, the balance of peptides released from these nerves, and
the microvasculature (Cosentino et al., 1997; Guillon et al., 1998; Zochodne et al., 1996).
The first empirical study, described in chapter 3 (parts 1 & 2), investigated the effect of
diabetes on sensory nerve function as determined by the ability of these nerves to mount
an inflammatory response. One of the aims of this chapter was to examine changes in the
inflammatory response with time (i.e. to establish a working time frame). Inflammatory
responses were induced in STZ-treated animals either chemically, using substance P (SP,
1μM) and calcitonin gene-related peptide (CGRP, 1μM) (post-terminal events) (Chapter
3, part 1) or electrically, via selectively stimulating C-fibres (combined pre- and postterminal events) (chapter 3, part 2). Sodium nitroprusside (SNP, 100μM), a direct smooth
muscle vasodilator, was used to assess changes in smooth muscle reactivity.
Results in (chapter 3) part 1 demonstrated an overall reduction in neurogenic
inflammatory responses (VD & PE) over 10 weeks of diabetes, reaching minimum
responses 4 weeks after induction of diabetes.
In diabetic rats, the basal blood flow (BBF) was reduced, reaching 66% of control BBF
after 4 weeks. A significant reduction in vasodilator responses to SP and CGRP (both
42% of own control responses), in 4 weeks diabetic rats, was partly attributed to the
significant reduction in smooth muscle reactivity (28% of control response). The
reduction in SNP, a nitric oxide (NO) donor, suggests that NO is unable to induce
vasodilation. A number of possibilities were raised:
•
Deactivation of NO (e.g. by free radicals) (see chapter 7 for further discussion)
•
Alterations to the efficiency of the second messenger system by certain factors (e.g.
free radicals). Based on the results, the possibility was raised that in early diabetes the
cyclic adenosine monophosphate (cAMP) may function more efficiently to
compensate for the poor performance by the cyclic guanosine monophosphate
(cGMP) system (chapter 1, section 1.5.3)
309
•
Alteration in the intracellular calcium level in smooth muscle due to dysfunction of
Na+ / K+-ATPase, Ca2+-stimulated ATPase, Na+/Ca2+-exchanger and Ca2+-pump in
the sarcoplasmic membrane (chapter 1, section 1.5.3)
•
Reduction in vasodilator effect (e.g. by free radicals and ETs (chapters 6 & 7)
In chapter 3, part 1, the effect of diabetes (4 weeks) on the microvasculature at both
superficial and deep compartments of the rat skin was examined. For this purpose, two
different laser Doppler flowmeter (LDF) machines were used, Perimed (PLDF) and Moor
(MLDF), each with different laser light penetration capabilities (chapter 2, section 2.9).
Results showed that microvascular blood flow at both superficial and deeper
compartments of the skin is reduced after 4 weeks of diabetes. The reduction in the
response to SP was greater using the MLDF compared to the PLDF, which suggests an
additional involvement of deeper vessels. Furthermore, this data showed that a reduction
in the microvascular blood flow can be detected after 2 weeks of diabetes and is subject
to further reduction with time.
The sensory-independent effects of diabetes on the microvasculature were also examined
in chapter 3, part 1, using sensory denervated (capsaicin pre-treated) animals (with and
without diabetes) (Chapter 2, section 2.8). Results showed that, in control rats, capsaicin
treatment alone significantly reduced the inflammatory response to SP (53.8% of control
response) but in capsaicin pre-treated diabetic rats this response was even further reduced
(17.6% of control response). The reduction in SP response in capsaicin-treated diabetic
rats was attributed to the effect of diabetes on the remaining functional sensory C-fibres
that might not have been destroyed by neonatal capsaicin treatment.
Recent studies have suggested that STZ toxicity might be involved in neuronal cell
damage (Adeghate and Parvez, 2000; Prickaerts et al., 2000). Therefore, in (chapter 3)
part 1, the hyperglycaemic effect of diabetes on the microvasculature, independent of
possible neurotoxic effects of STZ, was also examined via the use of alloxan-induced
diabetic rats. The suitability of alloxan was based on the knowledge that it can destroy
pancreatic beta cells in a similar manner to STZ (Thulesen et al., 1997; Yamamoto et al.,
1981) (chapter 1, section 1.6.2 & chapter 2, section 2.4). The changes in microvascular
310
responses in STZ-treated rats were similar to those in alloxan-treated rats. This suggests
that the reduction in these responses in STZ-treated rats is mainly due to the
hyperglycaemic effect rather than a possible independent neurotoxic effect of STZ.
In diabetic rats, PE was also increased up to week 4 after induction of diabetes (Appendix
B, Table B-1). These results suggest alterations in the permeability of post-capillary
venules in 4 weeks diabetic rats.
In chapter 3, part 2, changes in the sensory nerve function with duration of diabetes was
examined by selective electrical stimulation (ES) of the unmyelinated C-fibres of the
sciatic nerve using specific parameters (20V, 5Hz, 2 msec for 1 minute). Once again, a
similar significant reduction in the ES and SNP responses was observed in diabetic rats
over time (10 weeks), reaching a maximum at week 4 (both 28% of own control
responses). Since the maximum reduction in all inflammatory responses (SP, CGRP, ES)
were observed at 4 weeks after induction of diabetes, this time period was chosen as the
working time frame for all subsequent experiments in this study.
There are reports showing that in the diabetic condition, sensory neuropeptide content
and transport in the sciatic nerve are reduced and the release of these peptides from
diabetic tissue is also diminished (chapter 1, section 1.7.3). It was, therefore, important to
establish how much of the reduction in inflammatory responses, observed in early
diabetic rats, is caused by changes in sensory nerves at the pre-terminal level.
The results suggested that, at 4 weeks of diabetes, the reduction in the inflammatory
response to ES is due to a reduction in smooth muscle reactivity (both 28% of own
control responses) rather than possible damage to C-fibres affecting its function.
Capsaicin and STZ treatments significantly reduced ES responses to 55% and 28.9% of
control respectively. However, only a small further reduction in this response was
observed when capsaicin pre-treated animals were made diabetic (from 55% to 33.5% of
control response). The latter result suggested that the direct effect of diabetes on sensory
nerves under the current experimental condition might be minimal since most of these
fibres were already destroyed by capsaicin pre-treatment.
The ES response in alloxan-treated rats was less reduced when compared to STZ-treated
rats (46% compared to 28.9% of own control responses). This could be due to the
311
difference in severity of diabetes between these animals (chapter 2, section 2.4.5 &
chapter 3, part 2 discussion).
Overall, the results in chapter 3 support the notion that there are disturbances in
endothelial cells and smooth muscle cell function in skin microvasculature in early
diabetes. It was also demonstrated that post-terminal changes in microvascular function
are more apparent than actual pre-terminal changes at this early stage of diabetes.
The second empirical study, described in chapter 4, investigated the role of
parasympathetic nerves and local tissue factors in the modulation of microvascular blood
flow in early diabetes. To investigate the former, changes in the microvascular response
to acetylcholine (ACh), which is released by parasympathetic nerves, was examined in 4
weeks diabetic rats. The function of ACh receptors (nicotinic and muscarinic) was also
examined in these animals using two different receptor agonists, nicotine and muscarine
chloride. These receptors are located on sensory nerves and endothelial cells respectively.
BBFs in control and diabetic rats in chapter 4 were similar to those in chapter 3. The
results showed a significant and similar reduction in the vasodilator responses to ACh,
nicotine, and muscarine chloride in diabetic rats (27%, 36% & 34% of own control
responses respectively). The similarity in the reduction of these responses suggested the
involvement of the following possible mechanisms:
•
Deactivation of NO (chapter 7)
•
Generation of endothelium-vasoconstrictor prostanoids and alteration in the function
of muscarinic receptors (chapter 1, section 1.8.5.1)
•
Damage to endothelium and alteration in the function of muscarinic receptors
(chapter 1, section 1.8.5.1)
The vasodilator responses to ACh in 4 weeks diabetic rats were measured using both the
MLDF and the PLDF. The data suggested that vessels at both superficial and the deeper
level of the skin were affected by early diabetes, supporting the results in chapter 3. Since
MLDF recorded a greater reduction in the response compared to PLDF, again the
possibility of additional involvement of deeper vessels was raised (chapter 4, Table 4-1).
312
Bradykinin (BK) is an important local tissue factor produced during inflammation and
tissue injury. In chapter 4, the role of BK in the modulation of microvascular blood flow
in early diabetes was examined. Previous studies have shown that NO, a potent
vasodilator, can modulate the action of important peptides such as SP, ACh, BK (Khalil
and Helme, 1992; Ralevic et al., 1992; Ralevic et al., 1995b). Therefore, the role of NO
in modulating microvascular blood flow in early diabetes was also examined in this
chapter, using a NO synthase (NOS) inhibitor, NG-nitro-L-arginine methyl ester (LNAME). The aim was to assess the relative contribution of NO to BK and SP responses
in 4 weeks diabetic rats.
Results from this part of the study showed a reduction in SP and BK responses (42% &
48% of own control responses) in diabetic rats, which could be partially explained by a
decrease in smooth muscle reactivity (30% of control response). In control rats, perfusion
of L-NAME significantly decreased the vascular response to BK and SP (65% & 40% of
own control responses). In diabetic rats, however, L-NAME similarly reduced SP and BK
responses (15% & 20% of own controls). The response to SP in diabetic rats, in the
presence or absence of L-NAME, suggested that in these rats NO synthesis is not altered.
Therefore, changes in endothelial-mediated responses in diabetic rats could be either
related to the concentration of released sensory neurotransmitters or the fate of NO once
it is released.
No significant change was observed in diabetic PE after nicotine, muscarine chloride, or
BK suggesting that other factors may be involved in altering microvascular permeability
in diabetic rats.
Overall, the results in chapter 4 indicated the presence of certain factors in the diabetic
condition that could contribute to the reduced microvascular blood flows at the early
stages of the disease. The roles of these factors were examined in chapters 6, 7, 8, and 9.
The third empirical study, described in chapter 5, was devoted to investigation of
sympathetic nerve involvement in modulating microvascular blood flow in early diabetes.
The changes in microvascular responses to vasoactive substances were examined in
313
diabetic rats treated with a short-term sympathetic neurone blocker, guanethidine
monosulphate (Chapter 2, section 2.10).
The BBF in control rats was slightly increased after treatment with guanethidine (111%
of control response) but this treatment did not change the BBF in diabetic rats. The
former result suggests that sympathetic nerves play a minor role in regulating basal tone
under inflammatory conditions. The latter result raised the possibility that certain
endothelial-mediated factors may be involved in reducing BBF in early diabetes.
In control rats, no significant change was observed in vascular responses to SNP, SP, or
ES after guanethidine treatment. However, in diabetic rats, a significant enhancement was
observed in all these responses after this treatment (182%, 185% & 212% of own nontreated diabetics respectively). This suggests that sympathetic efferents play a major role
in reducing the above responses in 4 weeks diabetic rats. The reported alterations in the
sensitivity of adrenoreceptors in the diabetic condition have been proposed as a possible
underlying mechanism.
No significant changes in PE were detected in control or diabetic rats after guanethidine
treatment. The latter result suggests that sympathetic efferents are not involved in
microvascular permeability changes in early diabetes.
The roles of neuropeptide Y (NPY) and adenosine triphosphate (ATP), both released by
the sympathetic nerves, were not examined in this chapter due to a number of factors:
•
The chance of NPY and ATP release by low frequency ES used in this study was
limited (Pernow, 1988).
•
The resting level of NPY is low in most species and the density of its immunoreactive
fibres in the skin is unchanged up to 12 weeks after diabetes onset (Karanth et al.,
1990; Pernow, 1988).
•
In early diabetes (2 & 8 weeks), there is no significant change in ATP concentration
or the release of ATP between control and diabetic rats (Belai et al., 1991; Carlsson
and Arnqvist, 1981).
Overall, the findings in this chapter demonstrated that the contribution of sympathetic
efferents in regulating microvascular blood flow is increased in early diabetes.
314
11.3. Endothelial-mediated factors modulating microvascular blood flow in early
diabetes
Endothelin-1 (ET-1) is a potent vasoconstrictor that is produced by many cells during
inflammation and wound repair. Numerous studies have provided evidence that there are
alterations in ET-1 levels, ET receptor (e.g. ET-A, ET-B) activities, and vascular
responses to this peptide in diabetic animals and humans (chapter 1, section 1.8.4).
The fourth empirical study, described in chapter 6, investigated the role of this potent
vasoconstrictor on the modulation of skin microvascular blood flow in early diabetes.
The aim was to examine the relative contribution of two ET receptors, ET-A and ET-B,
to the observed reduction in microvascular responses in 4 weeks diabetic rats (see results
in chapter 3). For this purpose, BQ-123 and BQ-788, ET-A and ET-B receptor
antagonists were used.
Results showed an increase in BBF in diabetic rats after BQ-123 perfusion (from 66% to
83% of control response), supporting the results in chapter 5 that endothelial-mediated
factors rather than sympathetically-derived neurotransmitters might be involved in
reducing diabetic microvascular responses.
The significant improvement in SNP response in 4 weeks diabetic rats (from 28% to 70%
of control response), after perfusing BQ-123 and BQ-788, supported a role for ET-1 in
reducing smooth muscle reactivity and the following possible mechanisms of action were
proposed:
•
Inhibition of NO by ET-1 (see results in chapter 6)
•
Alterations in the efficiency of the (cGMP) pathway by ET-1-induced enhancement
in protein kinase C (PKC) activity (chapter 1, section 1.8.4)
The above results also suggested that, at the level of the microvasculature, ET-A and ETB receptors were equally involved in inhibiting smooth muscle reactivity.
The ET receptor antagonists used in this study also enhanced the vascular responses to SP
and ES in diabetic rats. Perfusion of BQ-123 and BQ-788 improved both responses to SP
(63% & 86% of own control responses) and ES (55% & 84% of own control responses)
315
respectively. These results suggested that ET-B receptors are more involved in inhibiting
the above responses than ET-A receptors. The possibility that ET may affect NO activity
and / or its release from endothelial cells was raised. It was suggested that damage to
endothelial cells by elevated glucose might lead to an increase in ET-1 release and
indirect inhibition of NO (chapter 1, section 1.8.4.1, Fig 1-10).
In control rats, PE was not altered after perfusing BQ-123 and BQ-788. However, these
agents showed equal potency in improving PE in diabetic rats, suggesting that both ET-A
and ET-B receptors are involved in microvascular permeability changes in early diabetes.
The role of PKC in vascular alterations associated with diabetes and a relationship
between enhanced ET-1 synthesis, diabetes, and PKC activity have been demonstrated
(chapter 1, section 1.8.5.4). However, the role of this enzyme in modulating skin
microvascular blood flow in early diabetes has not been previously investigated. The
second aim of chapter 6 was to examine the role of this enzyme in 4 weeks diabetic rats,
using a specific PKC inhibitor, bisindolylmaleimide (BIM).
BIM did not improve the SP response in diabetic rats but significantly enhanced the
response to SNP (56% of control response). The enhancement in SNP response, after
perfusing BIM in these animals, supports the hypothesis that PKC might be involved in
altering the level of intracellular calcium and hence contribute to the reduction in
endothelial-mediated vascular responses in 4 weeks diabetic rats. The lack of
improvement in the SP response was unexpected, raising the possibility that a preterminal event might have masked the involvement of PKC in the modulation of postterminal events.
Overall, this chapter provided evidence that ETs are involved in altering microvascular
blood flow and endothelial cell permeability in early diabetes. It was shown that there are
subtle differences in the relative contribution of the two ET receptor subtypes (ET-A &
ET-B) to changes in vascular responses in early diabetes. It was also demonstrated that
local pharmacological manipulation at the level of peripheral microcirculation, using ET
receptor antagonists (BQ-123 & BQ-788), could normalize these early changes. The
316
exact role of PKC in the modulation of skin blood flow in early diabetes requires further
investigation, since the results in this part of the study were inconclusive.
Reactive oxygen species (ROS) are capable of tissue damage and previous studies have
shown their involvement in long-term vascular complications associated with diabetes
(chapter 1, section 1.5.3). The fifth empirical study, described in chapter 7, investigated
the role of ROS in modulating skin microvascular blood flow in early diabetes. The aim
was to assess the quantitative contribution of superoxide anions and hydroxyl radicals
and the effect of oxidative damage on the observed reduction in microvascular responses
in 4 weeks diabetic rats (see results in chapter 3). For this purpose, three antioxidants
were used namely, superoxide dismutase (SOD), N-acetylcysteine (NAC), and tirilazad
mesylate (Freedox). In order to examine the possible interaction effects between free
radicals and ETs, a combination of SOD with an ET receptor antagonist (BQ-123) was
also used in this study.
In control rats, perfusion of SOD, NAC, BQ-123 + SOD, and treatment with tirilazad
mesylate did not change BBF. In 4 weeks diabetic rats, however, some improvement in
BBF was observed using NAC, BQ-123 + SOD, and tirilazad mesylate (from 66% to
79%, 77% & 73% of control BBF respectively) but not SOD alone. These results suggest
that hydroxyl radicals and ET-1 contribute more to the reduction in microvascular blood
flow in diabetic rats than superoxide anions (chapter 7, Fig 7-2).
These agents did not change the SNP response in control animals. A significant
enhancement in smooth muscle reactivity, however, was observed in diabetic rats using
SOD, NAC, and tirilazad mesylate (76%, 81% & 90% of control response respectively).
This response was normalized after perfusion of the combined BQ-123 + SOD. These
improvements suggest that, in early diabetes, ROS may play a role in deactivating NO or
reducing the efficiency of its action rather than changing the level of NO production.
Once again, these agents did not significantly change the vascular response to SP in
control rats. In diabetic rats, however, the SP response was slightly enhanced after
perfusing SOD (from 42% to 50% of control response) but significantly improved after
perfusing NAC and tirilazad mesylate-treatment (79% & 88% of control response
317
respectively). This response was again normalized after perfusing the combined BQ-123
+ SOD. Treatment with tirilazad mesylate also normalized the vascular response to
CGRP in diabetic rats. The improvement in these responses was suggestive of a greater
involvement of hydroxyl radicals, than other ROS, and ET in reducing endothelialmediated responses in diabetic rats under the current experimental condition.
The following are possible mechanisms that might underlie the effects of free radicals in
4 weeks rats with early diabetes:
•
Chemical reaction between superoxide anions and NO could generate peroxynitrite
and ultimately lead to the formation of hydroxyl radicals. This could result in
deactivation of NO and / or reduction in NO concentration (chapter 1, section 1.5.3).
•
Free radicals might interfere with the secondary messenger system pathway
(decreasing cyclic nucleotides), thus preventing vasodilatation or inducing smooth
muscle contraction by interfering with calcium homeostasis and dependent processes
(chapter 1, section 1.5.3)
•
The results involving SOD and BQ-123 + SOD in diabetic rats suggested that ETs
were more involved than superoxide anions in reducing smooth muscle reactivity. It
appears that these agents exert their effect by deactivating NO. Enhanced ET-1
production by ROS (e.g. superoxide anions, hydrogen peroxide) in early diabetic rats
and its inhibition by SOD and catalase, was previously reported (chapter 1, section
1.5.3).
It should be noted that since BQ-788 and NAC perfusions, at the concentrations tested,
individually improved SP responses to approximately 80-90% of the control value, the
combined treatment of these two drugs was not persued.
In control rats, perfusion of SOD and NAC enhanced the vascular response to ES (126%
& 111% of control) and the increase in this response after perfusing the combined BQ123 + SOD (150% of control response) reached statistical significance. The vascular
response to ES, in diabetic rats was improved after perfusing SOD (70% of control
response) but normalized after using NAC, tirilazad mesylate, and the combined BQ-123
+ SOD. These results suggest that ROS, especially hydroxyl radicals, are involved in
microvascular changes in early diabetes. The similarity between ES responses after NAC
perfusion and tirilazad mesylate treatment also suggests that part of the inhibitory action
318
of hydroxyl radicals could be related to oxidative damage. Since NAC was only perfused
for a short period of time, the full potential of this drug in this experiment could not be
extrapolated and therefore a true comparison of the effects of NAC and tirilazad mesylate
could not be made.
The PE results showed the greatest improvement after perfusing combined BQ-123 +
SOD, followed by treatment with tirilazad mesylate, and the least improvement was after
SOD and NAC perfusions. These results suggest that free radicals are involved in altering
endothelial permeability in 4 weeks diabetic rats.
Overall, results of investigations performed within chapter 7 revealed that free radicals
play a role in modulating diabetic microvascular blood flow at this early stage (4 weeks
post-diabetes). In addition, evidence was provided to suggest that the inhibitory effects of
ETs and free radicals on microvascular blood flow were additive. Results in chapter 6
and 7, collectively, showed that ETs play a greater role than free radicals in altering
endothelial cell permeability. Using different free radical scavengers, there were subtle
differences in the relative contribution of different free radicals in altering microvascular
blood flow. The results again showed that local pharmacological manipulation at the
level of the peripheral skin microcirculation could normalize diabetic vascular responses.
The glycation process, a natural biochemical event, is enhanced in diabetes and can lead
to excess formation of advanced glycated end-products (AGEs). It was postulated that
AGEs are involved in the development of vascular damage and abnormalities in
microvascular function observed in diabetes (chapter 1, section 1.5.2). The sixth
empirical study, described in chapter 8, investigated the role of glycated products in
modulating skin microvascular blood flow in early diabetes. The aims of this chapter
were:
First, to examine the short- and long-term effects of amadori and AGEs (early and late
products of glycation process) in modulating microvascular blood flow in 4 weeks
diabetic rats. For this purpose, the diabetic microvasculature was examined after shortterm exposure to amadori and AGEs and long-term treatment with aminoguanidine (AG).
319
Second, to examine the possible involvement of AGE receptors (RAGE) in altering
microvascular blood flow in animals with early diabetes, using anti-RAGE antibody
(IgG).
BBFs were not significantly changed in control or diabetic rats after amadori, AGEs, and
anti-RAGE IgG perfusions, or after AG treatments. The results in diabetic rats suggest
that in this disease condition, glycated products do not play a major role in modulating
vascular tone.
In control rats, short-term exposure to amadori and AGEs increased vascular responses to
SNP and SP. Amadori significantly enhanced the vasodilator responses to SNP and SP,
reaching 150% and 173% of own control responses respectively. Although exposure to
AGEs also increased these responses in control rats (143% & 162% of own control
responses), only the response to SP reached statistical significance. Perfusion of antiRAGE IgG did not have any effect on SNP or SP in control rats. It appears that under
inflammatory conditions, control microvasculature enhances vascular responses in the
presence of amadori and AGEs.
In diabetic rats, exposure to amadori enhanced the vasodilator responses to SNP and SP
(55% & 74% of own control responses) but no changes in these responses were observed
after perfusing AGEs. This increase in responses after amadori suggested that this
product is capable of modifying vascular response in diabetic rats and this modification
may be time-dependent. The lack of AGEs effect on SP and SNP responses, in diabetic
rats, raised the possibility that endogenous AGEs are already involved in modulating
these responses and, therefore, exogenous AGEs, used in this study, might not be capable
of further modification. This result also suggests that modification of vascular responses
by AGEs may depend on the duration of diabetes.
Anti-RAGE IgG did not have any effect on SP or SNP responses in control rats but
improved SP responses to 72% of control value and normalized SNP responses in
diabetic rats. The latter result not only provided further support for the proposition that
AGEs are involved in the modulation of microvascular responses in diabetic rats but also
suggest that RAGE may play a role in this modulation. Anti-RAGE IgG could improve
vasodilatation by inhibiting AGE-RAGE interaction, which could prevent NO inhibition,
320
generation of free radicals, and depletion of the antioxidant defence system (chapter 1,
sections 1.5.2 & 1.5.3).
It should be noted that since only a single dose of anti-RAGE IgG (100 μM/ml) was used,
the results in this study could not reflect the true contribution of endogenous AGEs on the
vascular response in 4 weeks diabetic rats. Endogenous AGEs true contribution can only
be extrapolated from a dose-response effect of anti-RAGE IgG on the vascular response,
an experiment that was not performed in the current study. The alternative approach was
to examine the long-term role of endogenous AGEs using long-term treatment with AG
(late and early onset treatments) to prevent AGEs.
Results showed that both 2 and 4 weeks (late and early onset) treatments with AG could
significantly enhance the response to SP (both around 70% of control response), and
CGRP (both around 90% of control response) in diabetic rats. These results could be
partially due to enhancement in SNP responses (both about 80% of control response). By
inhibiting AGEs formation, AG could improve these responses by protecting NO against
deactivation. Early onset treatment with AG did not cause normalization of any of the
post-terminal vascular responses in diabetic rats, suggesting that factors other than AGEs
may be involved in reducing microvascular responses in early diabetes.
ES responses in diabetic rats were improved to 70% of control response after late onset
and normalized after early onset treatment with AG. This may be related to preservation
of sensory nerve function by AG and may depend on the duration of diabetes.
Amadori and late onset AG treatment improved only basal PE (R1), whereas anti-RAGE
IgG and early onset AG treatment normalized R1 and the PE response to SP (SP2, 3 & 4) in
diabetic rats. The improvement in PE after anti-RAGE IgG and early onset AG treatment
suggests the involvement of AGEs in altering endothelial cell permeability in early
diabetes.
Overall, the findings presented in chapter 8 demonstrate that short-term exposure of
control microvasculature to early (e.g. amadori) and late (e.g. AGEs) glycation products
could modulate vascular function. Furthermore, amadori products exerted a positive
enhancing effect on microvascular function in early diabetes. It was proposed that in
321
these rats, amadori products could provide a respite for the microvasculature and hence,
improve the inflammatory response. When anti-RAGE IgG was used, the data showed
that AGE receptors are involved in modulating the inflammatory responses in rats with
early diabetes and the duration of treatment could play an important role with reference to
preventing neurovascular damage.
Human and animal studies have demonstrated that strict glycaemic control, using both
short- and long-term insulin treatments, can slow the progression of complications
associated with diabetes (chapter 1, section 1.8.5.3). Previous studies also proposed a
relationship between insulin and free radicals, ET-1, NO and AGEs (chapter 1, section
1.8.5.3).
The seventh empirical study, described in chapter 9, investigated the role of insulin in the
modulation of skin blood flow in early diabetes. The aim was to examine the relationship
between the time of commencement and duration of insulin treatment and the
development of skin microvascular alterations. For this purpose, diabetic rats received
insulin treatment either for 4 weeks, commencing shortly after diabetes onset (early onset
insulin treatment) or for 2 weeks, commencing on day 14 after diabetes onset (late onset
insulin treatment).
Early onset insulin treatment significantly improved BBF in diabetic rats (from 66% to
87% of control BBF), suggesting that this treatment may protect microvascular damage
in diabetes.
Late onset insulin treatment significantly improved SNP responses (58% of control) but
did not improve SP or CGRP responses in diabetic rats. Early onset insulin treatment, on
the other hand, normalized all these responses in diabetic animals. The general
improvement in smooth muscle reactivity in diabetic rats after both early and late onset
insulin treatments might be partially due to an increase in cGMP via NO-mediated
stimulation of guanylate cyclase (chapter 1, section 1.8.5.3). The partial increase in SNP
response after late insulin intervention may be related to the effect of insulin on ET
concentration and / or activity. Insulin may inhibit excess ET-1 production or downregulate ET-A receptors on smooth muscle cells (chapter 1, section 1.8.5.3), preventing
322
NO deactivation by ET-1 (see results in chapter 6) or improve the efficiency of the cGMP
pathway (chapter 1, section 1.8.4).
The lack of improvement in SP response, in late onset insulin-treated diabetic rats,
suggests the involvement of other factors that are affecting microvascular responses
independent of the hyperglycaemic effect. Results in chapters 6, 7, and 8 demonstrated
that ET receptor antagonists, antioxidants, their combination, and preventing AGEs
formation could improve or normalize microvascular responses in 4 weeks diabetic rats.
These results raised the possibility that production of substances such as free radicals,
ETs, and AGEs might be enhanced during the first 2 weeks of diabetes and alter the
microvascular function in these animals. It is possible that early onset insulin treatment
could protect diabetic microvessels against the effects of these substances. Normalization
of SP responses in diabetic rats using late onset insulin treatment was supportive of this
notion. Early insulin intervention might protect NO against inactivation by inhibiting
ROS formation and retarding the formation of AGEs and hence causing normalization of
SP responses. Normalization of CGRP responses in diabetic rats after early onset insulin
treatment could be partially due to an improvement in smooth muscle reactivity and
partially due to an increase in cAMP levels in smooth muscles (chapter 1, section
1.8.5.3).
Late onset insulin treatment improved the ES response in diabetic rats to 81% of control
response but early onset treatment normalized it. Part of the improvement in ES response
could be related to an improvement in the smooth muscle reactivity. Early onset insulin
treatment might also prevent nerve ischaemia, partially due to effective glycaemic control
over time.
PE was normalized in diabetic rats only after early onset insulin treatment, demonstrating
that effective treatment could restore normal endothelial function in diabetic rats (see
chapter 1, section 1.8.5.3 & introduction in chapter 9).
Overall, the data presented in this study demonstrated that early onset insulin treatment
could have an important role in preventing early skin microvascular alterations caused by
diabetes. The possibility was raised that adequate control of hyperglycaemia at the early
323
stages of diabetes may reduce the long-term impact on the microvascular and tissue
repair processes.
11.4. Factors involved in tissue repair process
Sensory nerves, especially unmyelinated afferent nerves, play an important role in
initiating the wound repair process by mediating both early and late components of the
neurogenic inflammatory responses (VD, PE & modulation of immune responses). It is
generally known that in the diabetic condition, sensory nerve function is altered,
inflammatory responses are reduced, and wound healing is impaired.
In previous empirical chapters, it was established that certain factors such as ETs, free
radicals, AGEs, and hyperglycaemia are involved in modulating skin microvascular
blood flow in early diabetes. It was also demonstrated that preventing the deleterious
effects of these factors, using specific inhibitors / antagonists, could normalize diabetic
vascular responses.
The final empirical study (chapter 10) was desighned to examine the effects of certain
treatment protocols (local and systemic), used in previous chapters, in modulating the
repair of full-thickness skin wounds in 4 weeks diabetic rats. The aim was to establish if
these treatment protocols could also improve tissue repair processes involved in fullthickness skin wounds. For this purpose, a thermal wound was induced in a control group
and a number of test groups of diabetic rats. These animals then received several different
treatments and changes in their wound sizes were monitored over time until full wound
closure was achieved (see method in chapter 10).
The results showed that different stages of tissue repair processes were affected in early
diabetes. A lack of a significant wound expansion in diabetic rats suggested a delay in
response to injury, reflecting an impairment of the first stage of the tissue repair process
(i.e. the inflammatory phase). This could be due to an alteration in post-terminal
mechanisms, leading to inadequate inflammatory responses in 4 weeks diabetic rats (see
results in chapter 3).
Delayed wound contraction, between days 2-7, suggested that the proliferative phase (reepithelialization, neovascularization, and wound contraction) of the repair process might
324
also be impaired. Changes in post-terminal effects mediated by sensory neuropeptides
such as SP could play an important role in this stage of the repair process.
Retarded wound contraction after day 7 in diabetic rats suggested that the final stage of
the repair process (i.e. tissue remodelling) could also be compromised. Changes in the
action of sensory peptides and NO synthesis could affect extracellular matrix formation
and collagen deposition (chapter 1, section 1.9.3).
The total number of days required for full wound contraction in diabetic rats was reduced
in insulin-treated (16 ± 0.3 days) compared to non-treated animals (18 ± 0.2 days).
Within the first 2 days after induction, wounds in insulin-treated diabetic rats expanded
more (by 0.4 cm2) than non-treated diabetic rats (by 0.15 cm2). This suggests that
hyperglycaemia is involved in impairing the inflammatory phase of the tissue repair
process in diabetic rats. It was hypothesized that insulin could contribute to the
improvement of this stage of the process by:
•
It’s direct effect as an endothelium-dependent vasodilator (chapter 1, section 1.8.5.3)
•
Inhibiting alterations in vascular permeability and response to vasoactive agents (see
results in chapter 9)
Between days 2-7, reduction in wound size in insulin-treated diabetic rats (by 2.1 cm2)
was significantly greater than non-treated diabetic animals (by 1.3 cm2). Similarly,
between days 9-11, wound contracted more in insulin-treated animals (by 1 cm2)
compared to non-treated diabetic rats (by 0.8 cm2). These results suggest that
hyperglycaemia is also involved in impairing proliferative and tissue remodelling stages
of the repair process in diabetic rats. It was hypothesized that insulin may contribute to
the improvement of these stages of the repair process by:
•
Adequate control of hyperglycaemia (i.e. blood glucose level close to normal range
during the experimental period) (see chapter 9 results)
•
Preventing NO inactivation (see chapter 9 results)
These results suggest that insulin is involved in improving both proliferative and
remodelling stages of the tissue repair process in diabetic rats, but its effect is greater on
the former stage.
325
The total number of days required for full wound contraction in diabetic rats was reduced
in AG-treated (16 ± 0.2 days) compared to non-treated rats (18 ± 0.2 days). Within the
first 2 days after induction, wound expansion in AG-treated diabetic rats (by 0.1 cm2) was
similar to non-treated diabetic animals (by 0.15 cm2). This suggests that factors other
than AGEs may be involved and they may exert a greater effect on the inflammatory
phase of the tissue repair process in diabetic rats. Between days 2-7, wounds in AGtreated diabetic rats were significantly contracted (by 1.9 cm2) when compared to those in
non-treated diabetic animals (by 1.3 cm2). This suggests that AGEs are involved in the
proliferative stage of the tissue repair process. It was hypothesized that AG treatment
could contribute to the improvement of this stage of the repair process by preventing
AGEs-induced changes in post-terminal events (see results in chapter 8).
Between days 9-11, the reduction in wound size in AG-treated diabetic rats (by 0.75 cm2)
was not different from non-treated diabetic rats (by 0.8 cm2). Again, this suggests that
factors other than AGEs may be involved, exerting a greater effect on the tissue
remodelling stage of the repair process in diabetic rats.
These results suggest that AG is involved mainly in improving the proliferative stage of
the tissue repair process in diabetic rats.
The time required for full wound closure in tirilazad mesylate-treated diabetic rats (13 ±
0.2 days) was shorter than control animals (14 ± 0.2 days) but only reached statistical
significance when compared to the period of time required by non-treated diabetic rats
(18 ± 0.2 days). Within the first 2 days after induction, wound expansion in tirilazad
mesylate-treated diabetic rats (by 0.4 cm2) was greater than AG-treated diabetic animals
(by 0.1 cm2) but similar to those in the control group (by 0.55 cm2). Between days 2-7,
contraction in wound size in tirilazad mesylate-treated diabetic rats (by 2.25 cm2) showed
a similarity to those in the control group (by 2.5 cm2) but was greater than those observed
in AG-treated animals (by 1.9 cm2). These results suggest that free radicals play a role in
impairing both inflammatory and proliferative stages of the tissue repair process in
diabetic rats. In addition, free radicals are more involved than AGEs in impairing the
proliferative stage of this process. It was hypothesized that tirilazad mesylate may
326
contribute to the improvement of these stages of the tissue repair process by preventing
free radical-induced changes in post-terminal events (see results in chapter 7).
Between days 7-9, contraction in wound size in tirilazad mesylate-treated diabetic rats
(by 0.9cm2) was similar to those in the control group (by 1cm2). This suggests that free
radicals are also involved in impairing tissue remodelling stage of the repair process in
diabetic rats. Tirilazad mesylate may contribute to the improvement of this stage of the
process by scavenging free radicals and so protecting the microvasculature against
damage by these agents.
These results suggest that tirilazad mesylate is involved in improving both proliferative
and remodelling stages of the tissue repair process in diabetic rats, but its effect is greater
on the latter stage.
The total number of days required for full wound contraction in diabetic rats treated with
combined BQ-123 + SOD (13 ± 0.3 days) was similar to tirilazad mesylate-treated
diabetic rats (13 ± 0.3 days). This raised the possibility that free radicals play a more
important role than ETs in tissue repair processes in diabetes. Within the first 2 days after
induction, the greatest wound expansion was observed in combined BQ-123 + SODtreated diabetic rats (by 0.75 cm2). Between days 2-7 and among other treated diabetic
rats, wounds contracted the most in rats treated with combined BQ-123 + SOD (by 2.35
cm2), which was similar to those observed in control animals (by 2.5 cm2). These results
suggest that ETs and free radicals together could cause a greater impairment in the
inflammatory and proliferative stages of the tissue repair process than individually in
diabetes (additive effect). It was hypothesized that BQ-123 and SOD may improve these
stages of the repair process by preventing changes in post-terminal events induced by ET1 (involving both ET-A and ET-B receptors) and free radicals (see results in chapters 6 &
7).
Between days 9-11, a significant reduction in wound size was observed in combined BQ123 + SOD-treated diabetic rats (by 0.9 cm2), which was greater than those in tirilazad
mesylate-treated diabetic rats (by 0.8 cm2) and far greater than the control group (by 0.45
cm2). This suggests that ETs and free radicals also play an important role in impairing
327
the tissue remodelling stage of the repair process. The combined BQ-123 + SOD may
contribute to the final stage of the repair process by:
•
Inhibiting ET-1 activity
•
Preventing NO deactivation
These results suggest that ETs play a greater role than free radicals in inhibiting the tissue
remodelling stage of the repair process.
Overall, results of the investigation in this chapter demonstrated that all treatments used
have contributed positively, to varying degrees, to different stages of wound repair and to
the overall improvement in repair processes in diabetic rats. Furthermore,
pharmacological manipulations at the level of the microvasculature (e.g. local injection of
combined BQ-123 + SOD) and systemic treatments (e.g. insulin, AG, Tirilazad mesylate)
were shown to improve repair processes in early diabetes.
11.5. Concluding remarks and future directions
The primary aim of this thesis was to investigate mechanisms underlying changes in
microvascular blood flow with relevance to tissue repair in early diabetes. Based on the
findings presented in this thesis, it is reasonable to conclude that a number of factors that
are known to cause long-term vascular complications also contribute to early changes in
skin microvascular blood flow in early diabetes.
This thesis demonstrated that by reducing the effects of ETs, free radicals, AGEs and
hyperglycaemia in early diabetes, using specific treatment protocols, the skin
microvascular blood flow was significantly improved (chapters 6, 7, 8, and 9). When the
same treatments were applied to diabetic rats with skin wounds, overall tissue repair
processes were also enhanced. This work raised the exciting possibility that early
intervention can prevent microvascular alterations in diabetes, which could decelerate the
onset of long-term diabetic complications.
The overall information in this thesis could have significant clinical implications for
treating diabetic patients with long-term vascular complications. The principle behind the
use of non-invasive and easy to apply treatment protocols, which achieve a significant
328
improvement in tissue repair processes, was refreshing and worthy of further
investigation. Future research is needed to determine the impact of early intervention on
decelerating late diabetic complications.
There are a number of issues relevant to the topic of this thesis that were not persued due
to limitations of time and resources. However, investigating some of these issues
(mentioned below) could be of great benefit in further expanding our knowledge in the
areas of diabetes, skin microvasculature, and wound repair, including:
1. Investigating mechanisms underlying changes in skin microvascular blood flow in
•
rats with longer-duration of diabetes
•
type II model of diabetic rats
•
genetic model of type I diabetes
•
short- and long-term treatment with natural antioxidants (e.g. Vitamin E)
2. Examining structural changes in the microvasculature of diabetic rats in relation to
the duration of diabetes, levels of hyperglycaemia, AGEs, and presence of excess free
radicals, with or without treatment.
3. Investigating the potential clinical application of the treatment protocols used in
chapter 10 in improving skin repair processes in diabetic patients with compromized
microvascular blood flow.
329
CHAPTER 12
Bibliography
330
Abdel-Rahman, M. S., Elrakhawy, F. I., and Iskandar, F. A. (1992a). Protection of Bcells against the effect of alloxan. Toxicology Letters 63, 155-164.
Abdel-Rahman, M. S., Elrakhawy, F. L., and Iskandar, F. A. (1992b). "Endocrine
pancreas in the postnatal offspring of alloxan-diabetic rats". .
Abebe, W., Harris, K. H., and MacLeod, K. M. (1994). Role of extracellular Ca2+in the
selective enhancement of contractile responses of arteries from diabetic rats to
noradrenaline. Canadian Journal of Physiology and Pharmacology 72, 15441551.
Abebe, W., and MacLeod, K. M. (1990). Protein Kinase C-mediated contractile
responses of arteries from diabetic rats. British Journal of Pharmacology 101,
465-471.
Adeghate, E., and Parvez, S. H. (2000). Nitric oxide and neuronal and pancreatic beta cell
death. Toxicology 153, 143-156.
Ahlgren, S. C., and Levine, J. D. (1993). Mechanical hyperalgesia in streptozotocindiabetic rats is not sympathetically maintained. Brain Research 616, 171-175.
Akiyama, N., Okumura, K., Watanabe, Y., Hashimoto, H., Ito, T., Ogawa, K., and
Satake, T. (1989). Altered acetylcholine and norepinephrine concentrations in
diabetic rat hearts. Role of parasympathetic nervous sytem in diabetic
cardiomyopathy. Diabetes 38, 231-236.
Aljada, A., and Dandina, P. (2000). Effect of insulin on human aortic endothelial nitric
oxide synthase. Metabolism 49, 147-150.
Alsip, N. L., Schuschke, D. A., and Miller, F. N. (1996). Microvascular responses in the
skeletal muscle of diabetic rats. Journal of Laboratory and Clinical Medicine 128,
429-437.
Amthor, K. F., Dahl-Jorgensen, K., Berg, T. J., Shard-Heier, M., Sandvik, L., Aagenaes,
O., and Hanssen, K. F. (1994). The effect of 8 years of strict glycaemic control on
peripheral nerve function in IDDM patients: the Oslo study. Diabetologia 37,
579-584.
Anderson, R. R., and Parish, J. A. (1981). The optics of human skin. Journal of the
Investigative Dermatology 77, 13-19.
Anderson, S., and Vora, J. P. (1995). Current concepts of renal hemodynamics in
diabetes. Journal of Diabetes and its Complications 9, 304-307.
Andersson, D., Brunkwall, J., Bergqvist, D., and Edvinsson, L. (1992). Diminished
contractile response to neuropeptide Y of arteries from diabetic rabbits. Journal of
Autonomic Nervous System 37, 215-222.
Andrews, P. V., Helme, R. D., and Thomas, K. L. (1989). NK1 receptor mediation of
neurogenic plasma extravasation in rat skin. British Journal of Pharmacology 97,
1232-1238.
Angulu, J., Sanchez-Ferrer, C. F., Peiro, C., Marin, J., and Rodriguez-Manas, L. (1996).
Impairment of enothelium-dependent relaxations by increasing percentages of
glycosylated human haemoglobin. Possible mechanisms involved. Hypertension
28, 583-592.
Apfel, S. C., Arezzo, J. C., Brownlee, M., Federoff, H., and Kessler, J. A. (1994). Nerve
growth factor administration protects against experimental diabetic sensory
neuropathy. Brain Research 634, 7-12.
331
Aragno, M., Parola, S., Tamagno, E., Brignardello, E., Manti, R., Danni, O., and
Boccuzzi, G. (2000). Oxidative derangement in rat synaptosomes induced by
hyperglycaemia: restorative effect of dehydroepiandrosterone treatment.
Biochemical Pharmacology 60, 389-395.
Ari, G., Vardi, Y., and Finberg, J. P. (1999). Nitric oxide and penile erection in
streptozotocin-diabetic rats. Clinical Science 96, 365-371.
Arisawa, S., Arisawa, T., Ohashi, M., Nitta, Y., Ikeya, T., and Asai, J. (1996). Effect of
hydroxyl radical on fibroblast-mediated collagen remodelling in vitro. Clinical
and Experimental Pharmacology and Physiology 23, 222-228.
Arison, R. N., and Feudale, E. L. (1967). Induction of renal tumor by streptozotocin in
rats. Nature 214, 1254-1260.
Armstrong, D. G., and Lavery, L. A. (1998). Diabetic foot ulcers: prevention, diagnosis
and classification. American Family Physician 57, 1325-1332.
Arora, S., Veves, A., Caballaro, A. E., Smakowski, P., and LoGerfo, F. W. (1998).
Estrogen improves endothelial function. Journal of Vascular Surgery 27, 11411146.
Aruoma, O. I., Halliwell, B., Hoey, B. M., and Butler, J. (1989). The antioxidant action
of N-acetylcysteine: its reaction with hydrogen peroxide, hydroxyl radical,
superoxide, and hypochlorous acid. Free Radical, Biology and Medicine 6, 593597.
Ashcroft, G. S., Horan, M. A., and Ferguson, M. W. (1995). The effects of ageing on
cutaneous wound healing in mammals. Journal of Anatomy 187, 1-26.
Askar, M. A., and Baquer, N. Z. (1994). Changes in the activity of NADH-oxidase in rat
tissues during experimental diabetes. Biochemistry and Molecular Biology
International 34, 909-914.
Awazu, M., Parker, R. E., Ha rvie, B. R., Ichikawa, I., and Kon, V. (1991). Downregulation of endothelin-1 receptors by protein kinase C in streptozotocin-diabetic
rats. Journal of Cardiovascular Pharmacology 17 (Supple 7), S500-S502.
Azadzoi, K. M., and De Tejada, I. (1992). Diabetes mellitus impairs neurogenic and
endothelium-dependent relaxation of rabbit corpus cavernosum smooth muscle.
The Journal of Urology 148, 1587-1591.
Babazono, T., Kapor-Drezgic, J., Dlugosz, J. A., and Whiteside, C. (1998). Altered
expression and subcellular localization of diacylglycerol-sensitive protein kinase
C isoforms in diabetic rat glomerular cells. Diabetes 47, 668-676.
Baker, H. J., Lindsey, J. R., and Weisbroth, S. H. (eds.) (1979-80). "The laboratory rat".
Academic Press, New York.
Baker, L. L., Chambers, R., DeMuth, S. K., and Villar, F. (1997). Effects of electrical
stimulation on wound healing in patients with diabetes ulcers. Diabetes Care 20,
405-412.
Balabolkin, M. I., Mamaeva, G. G., and Troshina, E. A. (1994). Use of laser Doppler
flowmeter for the early diagnosis of diabetic microangiopathy. Problemy
Endokrinologii 40, 19-20.
Baron, A. D. (1999). Vascular Reactivity. American Journal of Cardiology 84, 25J-27J.
Barton, M., Haudenschild, C. C., d'uScio, L. V., Shaw, S., Munter, K., and Luscher, T. F.
(1998). Endothelin ET-A receptor blockade restores NO-mediated endothelial
function and inhibits atherosclerosis in apolipoprotein E-deficient mice.
332
Proceedings of the National Academy of Sciences of the United States of America
95, 14367-14372.
Bassirat, M., and Khalil, Z. (2000). The role of endothelin and superoxide anions in
modulating inflammation in streptozotocin-induced diabetic rat model.
Microvascular Research 59, 88-98.
Battistini, B., D'Orleans-Juste, P., and Sirois, P. (1993). Biology of disease. Endothelins:
circulating plasma levels and presence in other biological fluids. Laboratory
Investigation 68, 600-628.
Bauer, V., Sotnikova, R., Machova, J., Matyas, S., Pucovsky, V., and Stefek, M. (1999).
Reactive oxygen species induced smooth muscle responses in the intestine,
vessels and airways and the effect of antioxidants. Life Sciences 65, 1909-1917.
Baumgartner-Parzer, S., Wagner, O., Waldhausl, W., Roth, T., and Lorenzi, M. (1996).
Failure of high ambient glucose to affect endothelin-1 synthesis or release by
cultured human endothelial cells. Hormone and Metabolic Research 28, 610-612.
Bayliss, W. M. (1901). On the origin from the spinal cord of the vasodilator fibres of the
hind limb, and on the nature of these fibres. Journal of Physiology 26, 173-209.
Beales, P. E., Williams, A. J. K., Albertini, M. C., and Rozzilli, P. (1994). Vitamin E
delayes diabetes onset in the non-obese diabetic mouse. Hormone and Metabolic
Research 26, 450-452.
Bean, B. P. (1989). Neurotransmitter inhibition of neuronal calcium currents by changes
in channel voltage dependence. Nature 340.
Becker, B. F., Heindl, B., Kupatt, C., and Zahler, S. (2000). Endothelial function and
homeostasis. Zeitschrift fur Kardiologie 89, 160-167.
Beebe, D. K., and Walley, E. (1992). Diabetic diarrhea: an undiagnosed complication.
Postgraduate Medicine 91, 179-186.
Belai, A., Lefebvre, R. A., and Burnstock, G. (1991). Motor activity and neurotransmitter
release in the gastric fundus of streptozotocin-diabetic rats. European Journal of
Pharmacology 194, 225-234.
Bell, D. S. (1995). Diabetic cardiomyopathy. A unique entity or a complication of
coronary artery disease? Diabetes Care 18, 708-714.
Bennett, G. C., Garret, N. E., Diemel, L. T., Brain, S. D., and Tomlinson, D. R. (1998).
Neurogenic cutaneous vasodilatation and plasma extravasation in diabetic rats:
effects of insulin and nerve growth factor. British Journal of Pharmacology 124,
1573-1579.
Bennett, T., and Gardiner, S. M. (eds.) (1996). "Nervous control of blood vessels".
Harwood academic publishers GmbH, Amesterdam.
Benzeroual, K., Pandey, S. K., Srivastava, A. K., Van de Werve, G., and Haddad, P. S.
(2000). Insulin-induced Ca2+ entry in hepatocytes is important for PI3-kinase
activation, but not for insulin receptor and IRS-1 tyrosine phosphorylation.
Biochimica et Biophysica Acta 1495, 14-23.
Bertello, P., Vedlio, F., Pinna, G., Gurioli, L., Molino, P., Alban, S., and Chiandussi, L.
(1994). Plasma endothelin in NIDDM patients with or without complications.
Diabetes Care 17, 574-577.
Bierhaus, A., Chevion, S., Chevion, M., Hofmann, M., Quenhenberger, P., Illmar, T.,
Luther, T., Brentshtein, E., Tritschler, H., Muller, M., Wahl, P., Ziegler, R., and
Nawroth, P. P. (1997). Advanced glycation end-product-induced activation of
333
Nf-kappa B is suppressed by alpha-lipoic acid in cultured endothelial cells.
Diabetes 46, 1481-1490.
Bierhaus, A., Hofmann, M. A., Ziegler, R., and Nawroth, P. P. (1998). AGEs and their
interactions with AGE receptors in vascular disease and diabetes mellitus: AGE
concept. Cardiovascular Research 37, 586-600.
Biessels, G. J., Cristino, N. A., Rutten, G. J., Hamers, F. P., and Erkelens, D. W. (1999).
Neurophysiological changes in the central and peripheral nervous system of
streptozotocin-diabetic rats. Course of development and effects of insulin
treatment. Brain 122, 757-768.
Biessels, G. J., Stevens, E. J., Mahmood, S. J., Gispen, W. H., and Tomlinson, D. R.
(1996). Insulin partially reverses deficits in peripheral nerve blood flow and
conduction in experimental diabetes. Journal of the Neurological Sciences 140,
12-20.
Bitar, M. S. (1998). Glucocorticoid dynamics and impaired wound healing in diabetes
mellitus. American Journal of Pathology 152, 547-554.
Bitar, M. S., Bajic, K. T., Farrok, T., Thomas, M. I., and Pilcher, C. W. (1999). Spinal
cord noradrenergic dynamics in diabetic and hypercortisolaemic states. Brain
Research 830, 1-9.
Bitar, M. S., and DeSouza, E. B. (1990). Diabetes-related changes in brain betaadrenoreceptors in rats as assessed by quantitative autoradiography: relationship
to hypothalamic norepinephrine metabolism and pituitary-gonadal hormone
secretion. Journal of Pharmacology and Experimental Therapeutics 254, 781785.
Blakytny, R., Jude, E. B., Martin Gibson, J., Boulton, A. J., and Ferguson, M. W. (2000).
Lack of insulin-like growth factor-1 (IGF-1) in the basal keratinocyte layer of
diabetic skin and diabetic foot ulcers. Journal of Pathology 190, 589-594.
Boder, C. W., Lake, D., Savage, M. W., and Williams, G. (1999). Hand vein responses to
noradrenaline in normotensive patients with insulin-dependent diabetes mellitus
and microalbuminuria: effects of alpha-adrenoreceptor blockade with doxazosin.
Current Medical Research Opinion 15, 169-176.
Bonini, J. A., Colca, J., and Hofmann, C. (1995). Altered expression of insulin signalling
components in streptozotocin-treated rats. Biochemical and Biophysical Research
communications 212, 933-938.
Bonnardel-Phu, E., and Vicaut, E. (2000a). Reactive oxygen species and acute
modulation of albumin microvascular leakage in the microcirculation of diabetic
rats in vivo. Journal of Vascular Research 37, 32-38.
Bonnardel-Phu, E., and Vicaut, E. (2000b). Role of oxidative stress in permeability
changes observed in microcirculation of diabetic rats in vivo. Journal des
Maladies Vascularies 25, 115-121.
Bonnardel-Phu, E., Wautier, J., and Vicaut, E. (2000). Advanced glycation end-products
are involved in permeability changes observed in microcirculation of diabetic rats
in vivo. Journal Des Maladies Vasculaires 25, 122-127.
Booth, A. A., Khalifah, R. G., Todd, P., and Hudson, B. G. (1997). In vitro kinetic
studies of formation of antigenic advanced glycation end-products (AGEs): novel
inhibition of post-amadori glycation pathways. Journal of Biological Chemistry
272, 5430-5437.
334
Borghini, I., Ania-Lahuerta, A., Regazzi, R., Ferrari, G., Gjinovci, A., Wollheim, C. B.,
and Pralong, W. F. (1994). alpha, beta I, beta II, delta and epsilon protein kinase
C isoforms and compound activity in the sciatic nerve of normal and diabetic rats.
Journal of Neurochemistry 62, 686-696.
Bouloumie, A., Bauersachs, J., Linz, W., Scholkens, B. A., Wiemer, G., Fleming, I., and
Busse, R. (1997). Endothelial dysfunction coincides with an enhanced nitric oxide
synthase expression and superoxide anion production. Hypertension 30, 934-941.
Boyer, C. S., Bannenberg, G. L., Neve, E. P., Ryrfeldt, A., and Moldeus, P. (1995).
Evidence for the activation of the signal-responsive phospholipase A2 by
exogenous hydrogen peroxide. Biochemical Pharmacology 50, 753-761.
Bradford, M. M. (1976). A rapid and sensitive method for the quantitation of microgram
quantities of protein utilizing the principal of protein-dye binding. Annals of
Biochemistry 72, 248-254.
Brain, S. D. (1996). In “Neurogenic Inflammation” (P. Geppetti and P. Holzer, eds.), pp.
229-244. CRC press, Boca Raton, FL.
Brain, S. D. (1997). Sensory neuropeptides: their role in inflammation and wound
healing. Immunopharmacology 37, 133-152.
Brain, S. D., Crossman, D. C., Buckley, T. L., and Williams, T. J. (1989). Endothelin-1:
demonstration of potent effects on the microcirculation of human and other
species. Journal of Cardiovascular Pharmacology 13 (Suppl 5), S147-S149.
Brain, S. D., and Williams, T. J. (1988). Substance P regulates the vasodilator activity of
calcitonin-gene releated peptide. Nature 335, 73-75.
Brands, M. W., and Fitzgerald, S. M. (1998). Acute endothelium-mediated vasodilation is
not impaired at the onset of diabetes. Hypertension 32, 541-547.
Bredt, D. S., and Synder, S. H. (1990). Localization of nitric oxide synthase indicating a
neuronal role for nitric oxide. Nature 373, 768-770.
Brownlee, M. (1994). Glycation and diabetic complications. Diabetes 43, 836-841.
Brownlee, M. (2000). Negative consequences of glycation. Metabolism 49 (2 Suppl 1),
9-13.
Brownlee, M., Vlassara, H., Kooney, A., Urlich, P., and Cerami, A. (1986).
Aminoguanidine prevents diabetes-induced arterial wall protein cross-linking.
Science 232, 1629-1632.
Brunelleschi, S., Vanni, L., Ledda, F., Giotti, A., Maggi, C. A., and Fantozzi, R. (1990).
Tachykinins activate guinea pig alveolar macrophages: involvement of NK2 and
NK1 receptors. British Journal of Pharmacology 100, 417-420.
Bucala, R., Tracey, K. J., and Cerami, A. (1991). Advanced glycosylation products
quench nitric oxide and mediate defective endothelium-dependent vasodilatation
in experimental diabetes. Journal of Clinical Investigation 87, 432-438.
Burant, C. F., Treutelaar, M. K., and Buse, M. G. (1986). Diabetes-induced functional
and structural changes in insulin receptors from rat skeletal muscle. Journal of
Clinical Investigation 77, 260-270.
Burgess, P., and Pearl, E. R. (1967). Myelinated afferent fibres responding specifically to
the noxious stimulation of the skin. Journal of Physiology 190, 541-562.
Burnstock, G. (1986). In “Coexistence of neuronal messengers: A new principal in
chemical transmission. Progress in Brain research” (T. Hokfelt, K. Fuxe, and B.
Pernow, eds.), Vol. 68, pp. 193-203. Elsevier, Amesterdam.
335
Burnstock, G. (1988). Regulation of local blood flow by neurohumoral substances
released from privascular nerves and endothelial cells. Acta Physiologica
Scandinavica 133 (Suppl 571), 53-59.
Bursell, S. E., Takagi, C., Clermont, A. C., Tagagi, H., Mori, F., Ishii, H., and King, G. L.
(1997). Specific retinal diacylglycerol and protein kinase C beta isoform
modulation mimics abnormal hemodynamics in diabetic rats. Investigative
Ophthalmology and Visual Science 38, 2711-2720.
Butler, A. R., Flitney, F. W., and Williams, D. L. H. (1995). Nitric oxide, nitosonium
ions, nitroxide ions, nitrosothiols and iron-nitrosyls in biology: a chemist's
prospective. Trends in Pharmacological Sciences 16, 18-22.
Cambridge, H., and Brain, S. D. (1995). Mechanism of bradykinin-induced plasma
extravasation in the rat knee joint. British Journal of Pharmacology 115, 641-647.
Cameron, N. E., and Cotter, M. A. (1999). Effects of antioxidants on nerve and vascular
dysfunction in experimental diabetes. Diabetes Research and Clinical Practice
45, 137-146.
Cameron, N. E., Cotter, M. A., Basso, M., and Hohman, T. C. (1997a). Comparison of
the effects of inhibitors of aldose reductase and sorbitol dehydrogenase on
neurovascular function, nerve conduction, and tissue polyol pathway metabolites
in streptozotocin-diabetic rats. Diabetologia 40, 271-281.
Cameron, N. E., Cotter, M. A., Dines, K., and Love, A. (1992). Effects of
aminoguanidine on peripheral nerve function and polyol pathway metabolites in
streptozotocin-diabetic rats. Diabetologia 35, 946-950.
Cameron, N. E., Cotter, M. A., Lai, K., and Hohman, T. C. (1997b). Effect of protein
kinase C inhibition on nerve function, blood flow and Na+-K+-ATPase defects in
diabetic rats. Diabetes 46 (Suppl 1), 31A.
Cameron, N. E., Cotter, M. A., and Low, P. A. (1991). Nerve blood flow in early
experimental diabetes in rats: relation to conduction deficits. American Journal of
Physiology 261, E1- E8.
Canturk, N. Z., Vural, B., Esen, N., Canturk, Z., Oktay, G., Kirkali, G., and Solakoglu, S.
(1999). Effects of granulocyte-macrophage colony-stimulating factor on
incisional wound healing in an experimental diabetic rat model. Endocrine
Research 25, 105-116.
Cardillo, C., Nambi, S. S., Kilcoyne, C. M., Choucair, W. K., Katz, A., Quon, M. J., and
Panza, J. A. (1999). Insulin stimulates both endothelin and nitric oxide activity in
the human forearm. Circulation 100, 820-825.
Carlsson, M., and Arnqvist, H. J. (1981). Influence of exogenous substrates on adenosine
triphosphate (ATP) concentrations in normal and diabetic rat aorta in vitro. Acta
Physiologica Scandinavica 111, 31-34.
Cechowska-Pasko, M., Palka, J., and Bankowski, E. (1996a). Alterations in
glycosaminoglycans in wounded skin of diabetic rats. A possible role for IGF-I,
IGF-binding proteins and proteolytic activity. Acta Biochemica Polonica 43, 557565.
Cechowska-Pasko, M., Palka, J., and Bankowski, E. (1996b). Decrease in the
glycosaminoglycan content in the skin of diabetic rats. The role of IGF-I, IGFbinding proteins and proteolytic activity. Molecular and Celluar Biochemistry
154, 1-8.
336
Cechowska-Pasko, M., Palka, J., and Bankowski, E. (1999). Decreased biosynthesis of
glycosaminoglycans in the skin of rats with chronic diabetes mellitus.
Experimental and Toxicologic Pathology 51, 239-243.
Ceriello, A. (1999). Hyperglycaemia: the bridge between non-enzymatic glycation and
oxidative stress in the pathogenesis of diabetic complications. Diabetes, Nutrition
and Metabolism 12, 42-46.
Chahl, L. A. (1990). Morphine produces release of substance P-like immunoreactivity
from guinea pig central nervous system. Neuroscience Letters 118, 88-90.
Chakir, M., and Plante, G. E. (1996). Endothelial dysfunction in diabetes mellitus.
Prostaglandins, Leukotrienes and Essential Fatty acids 54, 45-51.
Champion, R. H., Gillman, T., Rook, A. J., and Sims, R. T. (eds.) (1970). "An
introduction to the biology of of the skin". Blackwell scientific publications,
Oxford and Edimburugh.
Chan, N. N., Vallance, P., and Colhoun, H. M. (2000). Nitric oxide and vascular
responses in type I diabetes. Diabetologia 43, 137-147.
Chander, C. L., Moore, A. R., Howat, D., and Willoughby, D. A. (1988). The local
modulation of vascular permeability by endothelial cells-derived products.
Journal of Pharmacy and Pharmacology 40, 745-746.
Chang, K. C., Chung, S. Y., Chong, W. S., Suh, J. S., Kim, S. H., Noh, H. K., Seong, B.
W., Ko, H. J., and Chun, K. W. (1993). Possible superoxide radical-induced
alteration of vascular reactivity in aortas from streptozotocin-treated rats. Journal
of Pharmacology and Experimental Therapeutics 266, 992-1000.
Chang, K. S. K., and Stevens, W. C. (1992). Endothelium-dependent increase in vascular
reactivity to phenylephrine in long-term streptozotocin-diabetic rat aorta. British
Journal of Pharmacology 107, 983-990.
Chappey, O., Dosquet, C., Wautier, M. P., and Wautier, J. L. (1997). Advanced glycation
end-products, oxidant stress and vascular lesions. European Journal of Clinical
Investigation 27, 97-108.
Chase, H. P., Jackson, W. E., Hoops, S. L., Cockerham, R. S., Archer, P. G., and O'Brien,
D. (1989). Glucose control and the renal and retinal complications of insulindependent diabetes. JAMA 261, 1155-1160.
Chen, H. C., Guh, J. Y., Shin, S. J., Tsai, J. H., and Lai, Y. H. (2000). Reactive oxygen
species enhances endothelin-1 production of diabetic rat glomeruli in vitro and in
vivo. Journal of Laboratory and Clinical Medicine 135, 309-315.
Chiappe de Cingolani, G. E. (1986). Cyclic AMP, adenylate cyclase and cyclic AMPphosphodiesterase activities in diabetic rat adipocytes. Acta Physiologica et
Pharmacologica Latinoamericana 36, 39-46.
Chibber, R., Molinatti, P. A., Rosatto, N., Lambourne, B., and Kohner, E. M. (1997).
Toxication of advanced glycation end-products on cultured retinal capillary
pericytes and endothelial cells: relevance to diabetic retinopathy. Diabetologia 40,
156-164.
Chin, J. E., Liu, F., and Roth, R. A. (1994). Activation of protein kinase C alpha inhibits
insulin-stimulated tyrosine phosphorylation of insulin receptor substrate-1.
Molecular Endocrinology 8, 51-58.
337
Christensen, J. E. J., Varnek, L., and Gregerson, G. (1985). The effect of an aldose
reductase inhibitor (sorbinil) on diabetic neuropathy and neural function of the
retina: a double blind study. Acta Neurologica Scandinavica 71, 164-167.
Christlieb, A. R., Janka, H. U., Kraus, B., Gleason, R. E., Icasas-Cabral, E. A., Aiello, L.
M., Cabral, B. V., and Solano, A. (1976). Vascular reactivity to angiotensin II and
to norepinephrine in diabetic subjects. Diabetes 25, 268-274.
Cipolla, M. J., Porter, J. M., and Osol, G. (1997). High glucose concentrations dilate
cerebral arteries and diminish myogenic tone through an endothelial mechanism.
Stroke 28, 410-411.
Civan, M. M., Marano, C. W., Matschinsky, F. W., and Peterson-Yantorno, K. (1994).
Prolonged incubation with elevated glucose inhibits the regulatory response to
shrinkage of cultured human retinal pigment epithelial cells. Journal of
Membrane Biology 139, 1-13.
Clark, C. M., and Lee, D. A. (1995). Prevention and treatment of the complications of
diabetes mellitus. The New England Journal of Medicine 332, 1210-1217.
Cleland, S. J., Petrie, J. R., Ueda, S., Elliot, H. L., and Connell, J. M. (1998). Insulin as a
vascular hormone: implications for the pathophysiology of cardiovascular disease.
Clinical and Experimental Pharmacology and Physiology 25, 175-184.
Cooper, M. E., Gilbert, R. E., and Jerums, G. (1997). Diabetic vascular complications.
Clinical and Experimental Pharmacology and Physiology 24, 770-775.
Corbett, J. A., Tilton, R. G., Chang, K., Hasan, K. S., Ido, Y., Wang, J. L., Sweetland, M.
A., Lancaster, J. R., Williamson, J. R., and McDaniel, M. L. (1992).
Aminoguanidine, a novel inhibitor of nitric oxide formation, prevents diabetic
vascular dysfunction. Diabetes 41, 552-556.
Cosentino, F., Hishikawa, K., Kaustic, Z. S., and Luscher, T. F. (1997). High glucose
increases nitric oxide synthase expression and superoxide anions generation in
human aortic endothelial cells. Circulation 96, 25-28.
Cosentino, F., and Luscher, T. F. (1998). Endothelial dysfunction in diabetes mellitus.
Journal of Cardiovascular Pharmacology 32 (Suppl 3), S54-S61.
Cullum, N. A., Mahon, J., Stringer, K., and McLean, W. G. (1991). Glycation of rat
sciatic nerve tubulin in experimental diabetes mellitus. Diabetologia 34, 387-389.
Dahl-Jorgensen, K. (1998). Diabetic microangiopathy. Acta Paediatric Supplement 425,
31-34.
Dahlof, C., Dahlof, P., and Lundberg, J. M. (1986). α2-adrenoreceptor-mediated
inhibition of nerve stimulation-evoked release of neuropeptide Y (NPY)-like
immunoreactivity in the pithed rat. European Journal of Pharmacology 109, 289292.
Dall'Ago, P., Fernandes, T. G., Machado, U. F., Bello, A. A., and Irigoyen, M. C. (1997).
Baroreflex and chemoreflex dysfunction in streptozotocin-diabetic rats. Brazilian
Journal of Medicine and Biological Research 30, 119-124.
Davidson, M. B. (1998). "Diabetes mellitus: diagnosis and treatment". W.B. Saunders
company, Los Angeles, California.
DCCT. (1993). The effect of intensive treatment of diabetes on the development and
progression of long-term complications in insulin-dependent diabetes mellitus.
The New England Journal of Medicine 329, 977-986.
338
De Luca, A., Li, C. G., Rand, M. J., Reid, J. J., Thaina, P., and Wong-Dusting, H. K.
(1990). Effects of ω−conotoxin GVIA on autonomic neuroeffector transmission
in various tissues. British Journal of Pharmacology 101, 437-447.
Deng, D., Evans, T., Mukherjee, K., Dawney, D., and Chakrabarti, S. (1999). Diabetesinduced vascular dysfunction in the retina: role of endothelins. Diabetologia 42,
1228-1234.
Dennis, T., Fournier, A., Cadieux, A., Pomerleau, F., Jolicoeur, F. B., St Pierre, S., and
Quirion, R. (1990). hCGRP8-37, a calcitonin gene-related peptide antagonist
revealing calcitonin gene-related receptor heterogeneity in brain and periphery.
Journal of Pharmacology and Experimental Therapeutics 254, 123-128.
Di Giulio, A. M., Lesma, E., and Gorio, A. (1995). Diabetic neuropathy in the rat: 1.
Alcar augments the reduced levels of axoplasmic transport of substance P.
Journal of Neuroscience Research 40, 414-419.
Dianzani, M. U. (1990). In “Sedentary life and nutrition” (F. Fabris, L. Pernigotti, and E.
Ferrario, eds.), pp. 27-37. Raven Press, NewYork.
Diederich, D., Skopec, J., Diederich, A., and Dai, F. X. (1994). Endothelial dysfunction
in mesenteric arteries of diabetic rats: role of free radicals. American Journal of
Physiology 266, H1153-H1161.
Diemel, L. T., Stevens, E. J., Willars, G. B., and Tomlinson, D. R. (1992). Depletion of
substance P and calcitonin gene-related peptide in sciatic nerve of rats with
experimental diabetes; effect of insulin and aldose reductase inhibition.
Neuroscience Letters 137, 253-256.
Donadoni, M. L., Gavezzotti, R., Borella, F., Di Giulio, A. M., and Gorio, A. (1995).
Experimental diabetic neuropathy. Inhibition of protein mono-ADP-ribosylation
prevents reduction of substance P axonal transport. Journal of Pharmacology and
Experimental Therapeutics 274, 570-576.
Donatelli, M., Colletti, I., Bucalo, M. L., Russo, V., and Verga, S. (1994). Plasma
endothelin levels in NIDDM patients with macroangiopathy. Diabetes Research
25, 159-164.
D'Orleans-Juste, P., DeNucci, G., and Vane, J. R. (1989). Kinins act on B1 or B2
receptors to release conjointly endothelium-derived factor and prostacyclin from
bovine aortic endothelial cells. British Journal of Pharmacology 96, 920-926.
Doxey, D. L., Ng, M. C., Dill, R. E., and Ianopino, A. M. (1995). Platelet-derived growth
factor levels in wounds of diabetic rats. Life Sciences 57, 1111-1123.
Dray, A., and Perkins, M. (1993). Bradykinin and inflammatory pain. Trends in
Neuroscience 16, 99-104.
Dreher, D., and Junod, A. F. (1995). Differential effects of superoxide, hydrogen
peroxide, and hydroxyl radical on intracellular calcium in human endothelial
cells. Journal of Cell Physiology 162, 147-153.
Dulin, W. E., Wyse, B. M., and Kalamazoo, M. S. (1969). Studies on the ability of
compounds to block the diabetogenic activity of streptozotocin. Diabetes 18, 459465.
Dunbar, J. C., Ergene, E., Anderson, G. F., and Barraco, R. A. (1992). Decreased
cardiorespiratory effects of neuropeptide Y in the nucleus tractus solitarius in
diabetes. American Journal of Physiology 262, R865-R871.
339
Dunn, J. S., and McLetchie, N. G. B. (1943). Experimental alloxan diabetes in the rat.
Lancet 11, 384-387.
Dunnick, C. A., Gibran, N. S., and Heimbach, D. M. (1997). Substance P has a role in
neurogenic mediation of human burn wound healing. Journal of Burn Care and
Rehabilitation 17, 390-396.
Dymshitz, J., and Vasko, M. R. (1994). Endothelin enhances capsaicin-induced peptide
release and cGMP accumulation in cultures of rat sensory neurons. Neuroscience
Letters 167, 128-132.
Edwards, G., Dora, K. A., Gardener, M. J., Garland, C. J., and Weston, A. H. (1998). K+
is an endothelium-derived hyperpolarizing factor in rat arteries. Nature 396, 269272.
Eglezos, A., Andrews, P. V., Boyd, R. L., and Helme, R. D. (1991). Modulation of the
immune response by tachykinins: a review. Immunology and Cell Biology 69,
285-294.
Elgawish, A., Glomb, M., Friedlander, M., and Monnier, V. (1996). Involvement of
hydrogen peroxide in collagen cross-linking by high glucose in vitro and in vivo.
Journal of Biological Chemistry 271, 12964-12971.
Ellenberg, M., and Rifkin, H. (1983). "Diabetes Mellitus: theory and practice". Medical
Examination Publishing Co,.Inc., U.S.A.
Elliott, T. G., Cockcroft, J. R., Groop, P. H., Viberti, G. C., and Ritter, J. M. (1993).
Inhibition of nitric oxide synthesis in forearm vasculature of insulin-dependent
diabetic patients: blunted vasoconstriction in patients with microalbuminuria.
Clinical Science 85, 687-693.
Emori, T., Hirata, Y., and Marumo, F. (1990). Specific receptors for endothelin-3 in
cultured bovine endothelial cells and its cellular mechanism of action. FEBS
Letters 263, 261-264.
Enderle, M. D., Benda, N., Schmuelling, R. M., Haering, H. U., and Pfohl, M. (1998).
Preserved endothelial function in IDDM patients, but not in NIDDM patients,
compared with healthy subjects. Diabetes Care 21, 271-277.
Endo, K., Abiru, T., Machida, H., Kasuya, Y., and Kamata, K. (1995). Endotheliumderived hyperpolarizing factor does not contribute to the decrease in endotheliumdependent relaxations in the aorta of streptozotocin-induced diabetic rats. General
Pharmacology 26, 149-153.
Ernster, L., and Dallner, G. (1995). Biochemical, physiological and medical aspects of
ubiquinone function. Biochemical and Biophysical Acta 1271, 195-204.
Evans, J. S., Gerritsen, G. C., Mann, K. M., and Owens, S. P. (1965). Antitumor and
hyperglycaemic activity of streptozotocin (NSC37917) and its co-factor
(U15774). Cancer Chemotherapy Report 48, 1-10.
Faerman, I., Glocer, L., Fox, D., Jadzinsky, M. N., and Rapaport, M. (1974). Impotence
and diabetes. Histological studies of the autonomic nervous fibres of the corpora
cavernosa in impotence diabetic males. Diabetes 23, 971-979.
Fahey, T. J., Sadaty, A., Jones, W. G., Barber, A., Smoller, B., and Shires, G. T. (1991).
Diabetes impairs the late inflammatory response to wound healing. Journal of
Surgical Research 50, 308-313.
340
Fallgren, B., Arlock, P., and Edvinsson, L. (1993). Neuropeptide Y potentiates
noradrenaline-evoked vasoconstriction by an intracellular calcium-dependent
mechanism. Journal of Autonomic Nervous System 44, 151-159.
Farmer, S. G., McMillan, B. A., Meeker, S. N., and Burch, R. M. (1991). Induction of
vascular smoth muscle bradykinin B1 receptors in vivo during antigen arthritis.
Agents and Actions 34, 191-193.
Fazan, R. J., Dias da Silva, V. J., Ballejo, G., and Salgado, H. C. (1999). Power spectra of
arterial pressure and heart rate in streptozotocin-induced diabetes in rats. Journal
of Hypertension 17, 489-495.
Fein, F. S., and Sonnenblick, E. H. (1994). Diabetic cardiomyopathy. Cardiovascular and
Drugs Therapy 8, 65-73.
Ferri, C., Pittoni, V., and Piccoli, A. (1995). Insulin stimulates endothelin-1 secretion
from human endothelial cells and modulates its circulating levels in vivo. Journal
of Clinical Endocrinology and Metabolism 80, 829-835.
Finegold, D., Lattimer, S. A., Nolle, S., Bernstein, M., and Greene, D. A. (1983). Polyol
(sorbitol) pathway activity and myo-inositol metabolism, a suggested relationship
in the pathogenesis of diabetic neuropathy. Diabetes 32, 988-993.
Fitzgerald, M. (1983). Capsaicin and sensory neurones: a review. Pain 15, 109-130.
Fleischhacker, E., Esenabhalu, V. E., Spitaler, M., Holzmann, S., Skrabal, F., Koidl, B.,
Kostner, G. M., and Graier, W. F. (1999). Human diabetes is associated with
hyperreactivity of vascular smooth muscle cells due to altered subcellular Ca2+
distribution. Diabetes 48, 1323-1330.
Flyvbjerg, A., Nielsen, B., Skjaerbaek, C., Frystyk, J., Gronbaek, H., and Orskov, H.
(1994). In “The kidney and hypertension in diabetes mellitus” (C. E. Mrgensen,
ed.), pp. 233-243. Kluwer Academic Publishers.
Fozard, J. R., and Part, M. L. (1992). The role of nitric oxide in the regional vasodilator
effects of endothelin-1 in the rat. British Journal of Pharmacology 105, 744-750.
Frank, H. J., Levin, E. R., Hu, R. M., and Pedram, A. (1993). Insulin stimulates
endothelin binding and action on cultured vascular smooth muscle cell.
Endocrinology 133, 1092-1097.
Franzen, L. E., and Roberg, K. (1995). Impaired connective tissue repair in
streptozotocin-induced diabetes shows ultrastructural signs of impaired
contraction. Journal of Surgical Research 58, 407-414.
Friedman, E. A. (1999). Advanced glycated end-products and hyperglycaemia in the
pathogenesis of diabetic complications. Diabetes Care 22 (Suppl 2), B65-B71.
Fu, M. X., Wells-Knecht, K. J., Blackledge, J. A., Lyons, T. J., Thorp, S. R., and Baynes,
J. W. (1994). Glycation, glycoxydation and cross-linking of collagen by glucose.
Kinetics, mechanisms and inhibition of late stages of the Maillard reaction.
Diabetes 43, 676-683.
Fuder, H. (1994). Functional consequences of pre-junctional receptor activation or
blockade in the iris. Journal of Ocular Pharmacology 10, 109-123.
Fuder, H., and Selbach, M. (1993). Characterization of sensory neurotransmission and its
inhibition via alpha 2B-adrenoreceptors and via non-alpha 2-receptors in rabbit
iris. Naunyn-Schmiedebergs Archives of Pharmacology 347, 394-401.
Fukao, M., Hattori, Y., Kanno, M., Sakuma, I., and Kitabatake, A. (1997). Alterations in
endothelium-dependent hyperpolarization and relaxation in mesenteric arteries
341
from streptozotocin-induced diabetic rats. British Journal of Pharmacology 121,
1383-1391.
Fukui, T., Ishizaka, N., and Rajagopalan, S. (1997). p22phox mRNA expression and
NADPH oxidase activity are increased in aortas from hypertensive rats.
Circulation Research 80, 45-51.
Fulton, D. J. R., Hodgson, W. C., Sikorski, B. W., and King, R. G. (1991). Attenuated
responses to endothelin-1, KCl and Ca Cl2, but not noradrenaline, of aortae from
rats with streptozotocin-induced diabetes mellitus. British Journal of
Pharmacology 104, 928-932.
Furchgott, R. F., and Zawadzki, J. V. (1980). The obligatory role of endothelial cells in
the relaxation of arterial smooth muscle by acetylcholine. Nature 288, 373-376.
Gabbay, R. A., Siconolfi-Baez, L., and Lebovitz, H. E. (1990). Calcium-dependent
protein kinase activity is decreased in diabetic rat sciatic nerve. Endocrinology
127, 1194-1198.
Gale, E. A. (1996). Molecular mechanisms of beta-cell destruction in IDDM: the role of
nicotinamide. Hormone Research 45 (Suppl 1), 39-43.
Galle, J., Bauersachs, J., Busse, R., and Bassenge, E. (1992). Inhibition of cyclic AMPand cyclic GMP-mediated dilatations in isolated arteries by oxidised low density
lipoproteins. Arteriosclerosis and Thrombosis 12, 180-186.
Galle, J., Lenhmann-Bodem, C., Hubner, U., Heinloth, A., and Wanner, C. (2000). Cy A
and OXLDL cause endothelial dysfunction in isolated arteries through endothelinmediated stimulation of O2- formation. Nephrology, Dialysis, Transplantation 15,
33-46.
Galle, J., Schneider, R., Winner, B., Lehmann-Boden, C., Schinzel, R., Munch, G.,
Conzelmann, E., and Wanner, C. (1998). Glyco-oxidized LDL impair endothelial
function more potently than oxidized LDL: role of enhanced oxidative stress.
Atherosclerosis 138, 65-77.
Gamse, R., Holzer, P., and Lembeck, F. (1980). Decrease of SP in primary afferent
neurones and impairment of neurogenic plasma extravasation by capsaicin.
British Journal of Pharmacology 68, 207-213.
Gamse, R., and Jancso, G. (1985). Reduced neurogenic inflammation in streptozotocindiabetic rats due to microvascular changes but not to substance P depletion.
European Journal of Pharmacology 118, 175-180.
Gamse, R., Petsche, U., Lembeck, F., and Jancso, G. (1982). Capsaicin applied to
peripheral nerve inhibits axoplasmic transport of substance P and somatostatin.
Brain Research 239, 447-462.
Gamse, R., and Saria, A. (1987). Antidromic vasodilatation in the rat hindpaw measured
by laser Doppler flowmetery: pharmacological modulation. Journal of the
Autonomic Nervous System 19, 105-111.
Garcia, V. M., Ochoa, J. E., and Elia, M. M. (1999). Effects of early stage of
experimental diabetes on vascular functions of isolated perfused kidneys. Journal
of Autonomic Pharmacology 19, 97-103.
Garg, U. C., and Hassid, A. (1989). Nitric oxide-generating vasodilators and 8-bromocyclic guanosine monophosphate inhibit mitogenesis and proliferation of cultured
rat vascular smooth muscle cells. Journal of Clinical Investigations 83, 17741777.
342
Garland, C. J., and Mcpherson, G. A. (1992). Evidence that nitric oxide does not mediate
the hyperpolarization and relaxation to acetylcholine in the rat small mesenteric
artery. British Journal of Pharmacology 105, 429-435.
Garrett, N. E., Kidd, B. L., Cruwys, S. C., and Tomlinson, D. R. (1996). Effect of
streptozotocin-diabetes on knee joint inflammation-induced changes in substance
P and nerve growth factor in the rat. Brain Research and Molecular Brain
Research 42, 272-278.
Garthwaite, J. (1991). Glutamate, nitric oxide and cell-cell signalling in the nervous
system. Trends in Neuroscience 14, 60-67.
Gebremedhin, D., Koltai, M. Z., Pogatsa, G., Magyar, K., and Hadhazy, P. (1999).
Influence of experimental diabetes on the mechanical responses of canine
coronary arteries. Cardiovascular Research 22, 537-544.
Gellai, M., Fletcher, T., Pullen, M., and Nambi, P. (1996). Evidence for the existence of
endothelin-B receptor subtypes and their physiological roles in the rat. American
Journal of Physiology 271, R254-R261.
Giardino, I., Edelstein, D., and Brownlee, M. (1996). BCL-2 expression or antioxidants
prevent hyperglycemia-induced formation of intracellular advanced glycation
end-products in bovine endothelial cells. Journal of Clinical Investigation 97,
1422-1428.
Giardino, I., Fard, A. K., Hatchell, D. L., and Brownlee, M. (1998). Aminoguanidine
inhibits reactive oxygen species formation, lipid peroxidation, and oxidantinduced apoptosis. Diabetes 47, 1114-1120.
Giugliano, D., Ceriello, A., and Paolisso, G. (1996). Oxidative stress and diabetic
vascular complications. Diabetes Care 19, 257-267.
Gold, M. S., Dastmalchi, S., and Levine, J. D. (1997). Alpha 2-adrenergic receptor
subtypes in rat dorsal root and superior cervical ganglion neurones. Pain 69, 179190.
Goldner, M. G., and Gomori, G. (1944). Alloxan diabetes. Proceedings of the American
Diabetes Association 35, 158.
Golfman, L., Dixon, I. M., Takeda, N., Lukas, A., Dakshinamurti, K., and Dhalla, N. S.
(1998). Cardiac sarcolemmal Na+-Ca2+exchange and Na+-K+-ATPase activities
and gene expression in alloxan-induced diabetic in rats. Molecular and Cellular
Biochemistry 188, 91-101.
Goligorsky, M. S., Budzikowski, A. S., Tsukahara, H., and Noiri, E. (1999). Co-operation
between endothelin and nitric oxide in promoting endothelial cell migration and
angigenesis. Clinical and Experimental Pharmacology and Physiology 26, 269271.
Gonen, B., Bochenek, W., and Harati, Y. (1991). The effect of withdrawl of tolrestat,
aldose reductase inhibitor on signs, symptoms, and nerve function in diabetic
neuropathy. Diabetologia 34 (Suppl 1), A153.
Goto, T., Yanaga, F., and Ohtsuki, I. (1998). Studies on the endothelin-1-induced
contraction of rat granulation tissue pouch mediated by myofibroblasts.
Biochimica et Biophysica Acta 1405, 55-66.
Greene, D. A., and Lattimer, S. A. (1983). Impaired rat sciatic nerve sodium-potassium
adenosine triphosphatase in acute streptozotocin-diabetes and its correction by
343
dietary myo-inositol supplementation. Journal of Clinical Investigation 72, 10581063.
Greene, D. A., Lattimer, S. A., and Sima, A. A. F. (1987). Sorbitol, phosphoinositides,
and sodium-potassium-ATPase in the pathogenesis of diabetic complications. The
New England Journal of Medicine 316, 599-606.
Greenhalgh, D. G. (1996). The role of growth factors in wound healing. Journal of
Trauma 41, 159-167.
Grossi, S. G., and Genco, R. J. (1998). Periodontal disease and diabetes Mellitus: a two
way relationship. Annals of Periodontology 3, 51-61.
Grunfeld, J. A., Tiedemann, G. J., and Westerman, R. A. (1991). Chronic nicotine
exposure enhances cutaneous axon reflexes in the rat. Neuroreport 2, 421-424.
Gryglewski, R. J., Palmer, R. M., and Moncada, S. (1986). Superoxide anion is involved
in the breakdown of endothelium-derived vascular relaxing factor. Nature 320,
454-456.
Guillon, J. M., Thiry, C., Roach, A. G., and Cavero, I. (1998). Preferential reduction in
vascular responses to endothelin-1 in rats with streptozotocin-induced diabetes.
Journal of Cardiovascular Pharmacology 31 (Suppl 1), S133-S137.
Gush, R. J., and King, T. A. (1991). Discrimination of capillary and arterio-venular blood
flow in skin by laser Doppler flowmetry. Medical and Biological Engineering and
Computing 29, 387-392.
Gutierrez, S., Palacios, I., Egido, J., Gomez-Garre, D., Hernandez, P., Gonzalez, E., and
Herrero-Beaumont, G. (1996). Endothelin-1 induces loss of proteoglycans and
enhances fibronectin and collagen production in cultured rabbit synovial cells.
European Journal of Pharmacology 302, 191-197.
Gyorfi, A., Fazekas, A., Feher, E., Ender, F., and Rosivall, L. (1996). Effects of
streptozotocin-induced diabetes on neurogenic inflammation of gingivomucosal
tissue in rat. Journal of Periodontal Research 31, 249-255.
Ha, H., and Kim, K. H. (1999). Pathogenesis of diabetic nephropathy: the role of
oxidative stress and protein kinase C. Diabetes Research and Clinical Practice
45, 147-151.
Haak, T., Jungmann, E., Felber, A., Hillmann, V., and Usadel, K. H. (1992). Increased
plasma levels of endothelin in diabetic patients with hypertension. American
Journal of Hypertension 5, 161-166.
Hall, J. M. (1997). Bradykinin receptors. General Pharmacology 28, 1-6.
Hallberg, C. K., Trocme, S. D., and Ansari, N. H. (1996). Acceleration of corneal wound
healing in diabetic rats by the antioxidant trolox. Research Communications in
Molecular Pathology and Pharmacology 93, 3-12.
Haller, H., Baur, E., Quass, P., Behrend, M., Lindschau, C., Distler, A., and Luft, F. C.
(1995). High glucose concentrations and protein kinase C isoforms in vascular
smooth muscle cells. Kidney International 47, 1057-1067.
Halloway, G. A. J., and Watkins, D. W. (1977). Laser Doppler measurement of cutaneous
blood flow. Journal of Investigative Dermatology 69, 306-309.
Hammes, H., Martin, S., Federlin, K., Geisen, K., and Brownlee, M. (1991).
Aminoguanidine treatment inhibits the development of experimental diabetic
retinopathy. Proceedings of the National Academy of Sciences of the United
States of America 88, 11555-11558.
344
Harada, K., Ohmori, M., Kitoh, Y., Sugimoto, K., and Fujimura, A. (1999). Impaired
beta-adrenoreceptor-mediated venodilatation in patients with diabetes mellitus.
British Journal of Pharmacology 47, 427-431.
Hardin, N. J. (1996). The myocardial and vascular pathology of diabetes. Coronary
Artery Disease 7, 99-108.
Harrison, H. E., Stribling, D., Armstrong, F. M., and Perking, C. M. (1989). Aldose
reductase in the etiology of diabetic comploications: I. Introduction. Journal of
Diabetes and its Complications 3, 6-11.
Hattori, Y., Kasai, K., Nakamura, T., Emoto, T., and Shimoda, S. (1991). Effects of
glucose and insulin on immunoreactive endothelin-1 release from cultured
porcine aortic endothelial cells. Metabolism 40, 165-169.
Hattori, Y., Kawasaki, H., Fukao, M., Gando, S., Akaishi, Y., and Kanno, M. (1996).
Diminishment of contractions associated with depolarization-evoked activation of
Ca2+channels in diabetic aorta. Journal of Vascular Research 33, 454-462.
Havas, S. (1999). Educational guidelines for achieving tight control and minimizing
complications of type 1 diabetes. American Family Physician 60, 1985-1992,
1997-1998.
Hayakawa, H., and Raij, L. (1999). Relationship between hypercholesterolaemia,
endothelial dysfunction and hypertension. Journal of Hypertension 17, 611-619.
Hehenberger, K., Hansson, A., Heilborn, J. D., Abdel-Halim, S. M., Ostensson, C. G.,
and Brismar, K. (1999). Impaired proliferation and increased L-lactate production
of dermal fibroblasts in the GK-rat, a spontaneous model of non-insulin
dependent diabetes mellitus. Wound Repair Regeneration 7, 65-71.
Hellweg, R., and Hartung, H. D. (1990). Endogenous nerve growth factor (NGF) are
altered in experimental diabetes mellitus: a possible role for NGF in the
pathogenesis of diabetic neuropathy. Journal of Neuroscience Research 26, 258267.
Helme, R. D., and White, D. M. (1983). In “Substance P” (P. Skrabanek and D. Powel,
eds.), pp. 186-187. Boole Press, Dublin.
Helme, R. D., White, D. M., and Andrews, P. V. (1985). Neurogenic inflammation in
skin blisters. Experimental Brain research 59, 382-387.
Herr, R. R., Eble, T. E., Bergy, M. E., and Jahnke, H. K. (1959). Isolation and
characterisation of streptozotocin. Antibiotic Annuals , 236-243.
Heygate, K. M., Lawrence, I. G., and Bennett, M. A. (1995). Impaired endotheliumdependent relaxations in isolated resistance arteries of spontaneously diabetic rats.
British Journal of Pharmacology 116, 3251-3259.
Hicks, K. K., Seifen, E., Stimers, J. R., and Kennedy, R. H. (1998). Effects of
streptozotocin-induced diabetes on heart rate, blood pressure and cardiac
autonomic nervous control. Journal of Autonomic Nervous System 69, 21-30.
Highsmith, R. F., Blackburn, K., and Schmidt, D. J. (1992). Endothelin and calcium
dynamics in vascular smooth muscle. Annual Review of Physiology 54, 257-277.
Hill, M. A., and Ege, E. A. (1994). Active and passive mechanical properties of isolated
arterioles from STZ-induced diabetic rats: effect of aminoguanidine treatment.
Diabetes 43, 1450-1456.
345
Hill, M. A., Meininger, G. A., and Granger, H. J. (1985). Altrered skeletal muscle
microvascular hemodynamics after one week of streptozotocin-induced diabetes.
Microcirculation, Endothelium, Lymphatics 2, 687-704.
Hodgson, W. C., Sikorski, B. W., and King, R. G. (1992). Cardiovascular sensitivity
changes to eicosanoids in rats with experimentally-induced diabetes mellitus.
Clinical and Experimental Pharmacology and Physiology 19, 9-15.
Hogan, M., Cerami, A., and Bucala, R. (1992). Advanced glycation end-products block
antiproliferative effect of nitric oxide in the vascular and renal complications of
diabetes mellitus. Journal of Clinical Investigation 90, 1110-1115.
Hokfelt, T., Vincent, S., Dalsgaard, C. J., Skirboll, L., Johansson, O., Schultzberg, M.,
Lundberg, J. M., Rosel, S., Pernow, B., and Jancso, G. (1981). In “Substance P in
the nervous system” (R. Porter and M. O'Connor, eds.), pp. 84-106. Pitman
Books, London.
Holzer, P. (1991). Capsaicin: cellular targets, mechanisms of action, and selectivity for
thin sensory neurones. Pharmacological Reviews 43, 143-201.
Holzer, P. (1992). Peptidergic sensory neurones in the control of vascular functions:
mechanisms and significance in the cutaneous and splanchnic vascular beds.
Reviews of Physiology, Biochemistry and Pharmacology 121, 49-146.
Holzer, P. (1998). Neurogenic vasodilatation and plasma leakage in the skin. General
Pharmacology 30, 5-11.
Holzer, P., and Jocic, M. (1994). Cutaneous vasodilatation induced by nitric oxideevoked stimulation of afferent nerves in the rat. British Journal of Pharmacology
112, 1181-1187.
Holzer, P., Wachter, C., Heinemann, A., Jocic, M., Lippe, I. T., and Herbert, M. K.
(1995). Sensory nerves, nitric oxide and NANC vasodilatation. Archives
Internationales de Pharmacodynamie et de Therapie 329, 67-79.
Honing, M. L., Morrison, P. J., Banga, J. D., Stroes, E. S., and Rabelink, T. J. (1998).
Nitric oxide availability in diabetes mellitus. Diabetes / Metabolism Reviews 14,
241-249.
Hoorens, A., and Pipeleers, D. (1999). Nicotinamide protects human beta cells against
chemically-induced necrosis, but not against cytokine-induced apoptosis.
Diabetologia 42, 55-59.
Hopfner, R. L., and Gopalakrishnan, V. (1999). Endothelin: emerging role in diabetic
vascular complications. Diabetologia 42, 1383-1394.
Hopfner, R. L., Hasnadkda, R. V., McNeill, J. R., Wilson, T. W., and Gopalakrishnan, V.
(1998a). Differential regulation of endothelin-1 action by insulin and vanadate in
vascular smooth muscle. Journal of Cardiovascular Pharmacology 31 (Suppl 1),
S196-S198.
Hopfner, R. L., McNeill, J. R., and Gopalakrishnan, V. (1999). Plasma endothelin levels
and vascular responses at different stages of streptozotocin diabetes. European
Journal of Pharmacology 374, 221-227.
Hopfner, R. L., Misurski, D., Wilson, T. W., McNeill, J. R., and Copalakrishnan, V.
(1998b). Insulin and vanadate restore decreased plasma endothelin concentrations
and exaggerated vascular responses to normal in the streptozotocin diabetic rats.
Diabetologia 41, 1233-1240.
346
Hosgood, G. (1993). Wound healing. The role of platelet-derived growth factor and
transforming growth factor beta. Veterinary Surgery 22, 490-495.
Hotta, N. (1997). New concepts and insights on pathogenesis and treatment of diabetic
complications: polyol pathway and its inhibition. Nagoya Journal of Medical
Science 60, 89-100.
Hotta, N., Koh, N., and Sakakibara, F. (1996). Effect of propionyl-L-carnitine on motor
nerve conduction, autonomic cardiac function, and nerve blood flow in rats with
streptozotocin-induced diabetes: comparison with an aldose reductase inhibitor.
Journal of Pharmacology and Experimental Therapeutics 276, 49-55.
Hu, R. M., Levin, E. R., Pedram, A., and Frank, H. J. (1993). Insulin stimulates
production and secretion of endothelin from bovine endothelial cells. Diabetes 42,
351-358.
Hu, Y., and Dunbar, J. C. (1997). Regional vascular and cardiac responses to systemic
neuropeptide Y in normal and diabetic rats. Peptides 18, 809-815.
Hughes, S. R., and Brain, S. D. (1994). Nitric oxide-dependent release of vasodilator
quantities of calcitonin gene-related peptide from capsaicin-sensitive nerves in
rabbit skin. British Journal of Pharmacology 111, 425-430.
Hughes, S. R., Williams, T. J., and Brain, S. D. (1990). Evidence that endogenous nitric
oxide modulates oedema fromation induced by substance P. European Journal of
Pharmacology 191, 481-484.
Hujiberts, M. S. P., Wolffenbuttle, B. R. H., Boudier, H. A., Crijns, F. R., Kruseman, A.
C., Poitevin, P., and Levy, B. I. (1993). Aminoguanidine treatment increases
elasticity and decreases fluid filteration of large arteries from diabetic rats.
Journal of Clinical Investigation 92, 1407-1411.
Huszka, M., Kaplar, M., Rejto, L., Tornai, I., Palatka, K., Lszlo, P., and Udvardy, M.
(1997). The association of reduced endothelium-derived relaxing factor (NO)
production with endothelial damage and increased in vivo platelet activation in
patients with diabetes mellitus. Thrombosis Research 86, 173-180.
Huvers, F. C., De Leeuw, P. W., and Houben, A. J. (1999). Endothelium-dependent
vasodilatation, plasma markers of endothelial function, and adrenergic
vasoconstrictor responses in type I diabetes under near normoglycemic
conditions. Diabetes 48, 1300-1307.
Iacopino, A. M. (1995). Diabetic periodontitis: possible lipid-induced defect in tissue
repair through alteration of macrophage phenotype and function. Oral Diseases 1,
214-229.
Ialenti, A., Ianaro, A., Moncada, S., and Di Rosa, M. (1992). Modulation of acute
inflammation by endogenous nitric oxide. European Journal of Pharmacology
211, 177-182.
Ido, Y., Kilo, C., and Williamson, J. R. (1996). Interaction between the sorbitol pathway,
non-enzymatic glycation, and diabetic vascular dysfunction. Nephrology,
Dialysis, Transplantation 11 (Suppl 5), 72-75.
Ido, Y., McHowat, J., Chang, K. C., Arrogori-Martelli, E., Orfaliam, Z., Kilo, C., Corr, P.
B., and Williamson, J. R. (1994). Neural dysfunction and metabolic imbalances in
diabetic rats: prevention by acetyl-carritin. Diabetes 43, 1469-1477.
Ignarro, L. J., Byrns, R. E., Buga, G. M., and Wood, K. S. (1987). Endothelium-derived
relaxing factor from pulmonary artery and vein possess pharmacologic and
347
chemical properties identical to those of nitric oxide radical. Circulation Research
61, 866-879.
Inazu, M., Sakai, Y., and Homma, I. (1991). Contractile responses and calcium
immobalization in renal arteries of diabetic rats. European Journal of
Pharmacology 203, 79-84.
Inoguchi, T., Battan, R., Handler, E., Sportsman, J. R., Heath, W., and King, G. L.
(1992). Preferential elevation of protein kinase C isoform beta II and
diacylglycerol levels in the aorta and heart of diabetic rats: deferential
reversibility to glycaemic control by islet cell transplantation. Proceedings of the
National Academy of Sciences of the United States of America 89, 11059-11063.
Inoguchi, T., Xia, P., Kunisaki, M., Higashi, S., Feener, E. P., and King, G. L. (1994).
Insulin's effect on protein kinase C and diacylglycerol induced by diabetes and
glucose in vascular tissue. American Journal of Physiology 267, E369-E379.
Ishii, H., Koya, D., and King, G. L. (1998). Protein kinase C activation and its role in the
development of vascular complications in diabetes mellitus. Journal of Molecular
Medicine 76, 21-31.
Iwase, M., Nunoi, K., Wakisaka, M., Kikuchi, M., Maki, Y., Sadoshima, S., and
Fujishima, M. (1991). Spontaneous recovery from non insulin-dependent diabetes
mellitus induced by neonatal streptozotocin treatment in spontaneously
hypertensive rats. Metabolism 40, 10-14.
Jancso, G., Kiraly, E., and Jancso-Gabor, A. (1977). A pharmacologically induced
selective degeneration of chemosensitive primary sensory neurones. Nature 270,
741-743.
Jiang, Z. Y., Zhou, Q. L., Chatterjee, A., Feener, E. P., Myers, M. G., White, M. F., and
King, G. L. (1999). Endothelin-1 modulates insulin signalling through
phosphatidylinositol 3-kinase pathway in vascular smooth muscle. Diabetes 48,
1120-1130.
Johnson, A. R., and Erdos, E. G. (1973). Release of histamine from mast cells by
vasoactive peptides. Proceedings of the Society for Experimental Biology and
Medicine 142, 1252-1256.
Johnson, E. I., Dunlop, M. E., and Larkins, R. G. (1999). Increased vasodilatory
prostaglandin production in diabetic rat retinal vasculature. Current Eye Research
18, 79-82.
Jones, S., Brown, D. A., Milligan, G., Willer, E., Buckley, N. J., and Caufield, M. P.
(1995). Bradykinin excites rat sympathetic neurons by inhibition of M-current
through a mechanism involving B2 receptors and Gαq/11. Neuron 14, 399-405.
Jorns, A., Tiedge, M., Lenzen, S., and Munday, R. (1999). Effect of superoxide
dismutase, catalase, chelating agents, and free radical scavengers on the toxicity
of alloxan to isolated pancreatic islets in vitro. Free radicals, Biology and
Medicine 26, 1300-1304.
Jourd'heuil, D., Kang, D., and Grisham, M. B. (1997). Interaction between superoxide
and nitric oxide: implications in DNA damage and mutagenesis. Frontiers of
Bioscience 2, D189-D196.
Journeskog, G., Brismar, K., and Fagrell, B. (1995). Skin capillary circulation severely
impaired in toes of patients with IDDM, with and without late diabetic
complications. Diabetologia 38, 474-480.
348
Kakkar, R., Mantha, S. V., Kalra, J., and Prasad, K. (1996). Time course study of
oxidative stress in aorta and heart of diabetic rats. Clinical Science 91, 441-448.
Kakkar, R., Mantha, S. V., Radhi, J., Prasad, K., and Kalra, J. (1998). Increased oxidative
stress in rat liver and pancreas during progression of streptozotocin-induced
diabetes. Clinical Science 94, 623-632.
Kamata, K., and Kobayashi, T. (1996). Changes in superoxide dismutase mRNA
expression by streptozotocin-induced diabetes. British Journal of Pharmacology
119, 583-589.
Kamata, K., Miyata, N., Abiru, T., and Kasuya, Y. (1992). Functional changes in
vascular smooth muscle and endothelium of arteries during diabetes mellitus. Life
Sciences 50, 1379-1387.
Kamata, K., Miyata, N., and Kasuya, Y. (1988). Mechanisms of increased response of the
aorta to alpha-adrenoreceptor agonists in streptozotocin-induced diabetic rats.
Journal of Pharmacobiodynamics 11, 707-713.
Kamata, K., Miyata, N., and Kasuya, Y. (1989). Functional changes in potassium
channels in aortas from rats with streptozotocin-induced diabetes. European
Journal of Pharmacology 166, 319-323.
Kameyama, M., Fushimi, H., and Udaka, F. (1994). Diabetes mellitus and
cerebrovascular disease. Diabetes Research and Clinical Practice 24 (Suppl 1),
S205-S208.
Kanda, M., Eto, K., Tanabe, N., Sugiyama, A., Hashimoto, K., and Ueno, A. (1997).
Effects of ONO-2235, an aldose reductase inhibitor, on muscarinic receptors and
contractile response of the urinary bladder in rats with streptozotocin-induced
diabetes. Japanese Journal of Pharmacology 73, 221-228.
Kaneto, H., Kajimoto, Y., Miyagawa, J., Matsuoka, T., and Hori, M. (1999). Beneficial
effects of antioxidants in diabetes: possible protection of pancreatic B-cells
against glucose toxicity. Diabetes 48, 2398-2406.
Kang, N., Alexander, G., Park, J. K., Maasch, C., Buchwalow, I., Luft, F. C., and Haller,
H. (1999). Differential expression of protein kinase C isoforms in streptozotocininduced diabetic rats. Kidney International 56, 1737-1750.
Karanth, S. S., Springall, D. R., Francavilla, S., Mirrlees, D. J., and Polak, J. M. (1990).
Early increase in CGRP- and VIP-immunoreactive nerves in the skin of
streptozotocin-induced diabetic rats. Histochemistry 94, 659-666.
Kashiwagi, A., Asahina, T., Nishio, Y., Ikebuchi, M., Tanaka, Y., Kikkawa, R., and
Shigeta, Y. (1996). Glycation, oxidative stress and scavenger activity. Glucose
metabolism and radical scavenger dysfunction in endothelial cells. Diabetes 45
(Suppl 3), S84-S86.
Kaustic, Z. S., and Vanhoutte, P. M. (1989). Superoxide anion is an endothelium-derived
contracting factor. Cardiovascular Research 257, H33-H37.
Kawasaki, H. (1997). Pharmacological studies on alterations in contractile reactivity in
aortas isolated from experimental diabetic rats. Hokkaido Igaku Zasshi 72, 649665.
Kawasaki, H., Nuki, C., Saito, A., and Takasaki, K. (1990). Role of calcitonin generelated peptide-containing nerves in the vascular adrenergic neurotransmission.
Journal of Pharmacology and Experimental Therapeutics 252, 403-409.
349
Kennedy, B., Shen, G. H., and Ziegler, M. G. (1997). Neuropeptide Y-mediated pressor
responses following high frequency stimulation of the rat sympathetic nervous
system. Journal of Pharmacology and Experimental Therapeutics 281, 291-296.
Keogh, R. J., Dunlop, M. E., and Larkins, R. G. (1997). Effect of inhibition of aldose
reductase on glucose flux, diacylglycerol formation, protein kinase C, and
phospholipase A2 activation. Metabolism 46, 41-47.
Khalil, Z., Andrews, P. V., and Helme, R. D. (1988). VIP modulates substance P plasma
extravasation in vivo. European Journal of Pharmacology 151, 281-287.
Khalil, Z., Chen, H. B., and Helme, R. D. (1996). Mechanisms underlying the vascular
activity of β−amyloid protein fragment ( βA4 25-35) at the levels of skin
microvasculature. Brain Research 736, 206-216.
Khalil, Z., and Helme, R. D. (1989). Sequence of events in substance P-mediated plasma
extravasation in rat skin. Brain Research 500, 256-262.
Khalil, Z., and Helme, R. D. (1990a). Modulation of a peripheral inflammatory response
to substance P by locally administered opioid receptor antagonists. Neuropeptides
17, 45-53.
Khalil, Z., and Helme, R. D. (1990b). Serotonin modulates substance P-induced plasma
extravasation and vasodilatation in rat skin by an action through capsaicinsensitive primary afferent. Brain Research 527, 292-298.
Khalil, Z., and Helme, R. D. (1991). Effects of serotonin and opioid agonists on
tachyphylaxis to SP-induced vasodilatation. Annal of the New York Academy of
Sciences 632, 418-420.
Khalil, Z., and Helme, R. D. (1992). The quantitative contribution of nitric oxide and
sensory nerves to bradykinin-induced inflammation in rat skin microvasculature.
Brain Research 589, 102-108.
Khalil, Z., and Helme, R. D. (1996). Sensory neuropeptides as neuromodulators of
wound healing in aged rats. Journal of Gerontology and Biological Sciences 51,
B354-B361.
Khalil, Z., Liu, T., and Helme, R. D. (1999). Free radicals contribute to the reduction in
peripheral vascular responses and the maintenance of thermal hyperalgesia in rats
with chronic construction injury. Pain 79, 31-37.
Khalil, Z., and Merhi, M. (2000). Effects of aging on neurogenic vasodilator responses
evoked by transcutaneous electrical nerve stimulation: relevance to wound
healing. Journals of Gerontology. Series A, Biological Sciences and Medical
Sciences 55, B257-B263.
Khalil, Z., Ralevic, V., Bassirat, M., Dusting, G. L., and Helme, R. D. (1994). Effect of
ageing on sensory nerve function in rat skin. Brain Research 641, 265-272.
Khasar, S. G., Green, P. G., Chou, B., and Levine, J. D. (1995). Peripheral nociceptive
effects of alpha 2-adrenergic receptor agonists in the rat. Neuroscience 66, 427432.
Khodr, B., Khalil, Z. (2001). Modulation of inflammation by reactive oxygen species:
implications for aging and tissue repair. Free radicals Biology and Medicine 30,
1-8.
Kiff, R., Gardiner, S., Compton, A., and Bennet, T. (1991a). The effects of endothelin-1
and N-nitro-L-arginine methyl ester on regional haemodynamics in conscious rats
350
with streptozotocin-induced diabetes mellitus. British Journal of Pharmacology
103, 1321-1326.
Kiff, R. J., Gardiner, S. M., Compton, A. M., and Bennet, T. (1991b). Selective
impairment of hindquarters vasodilator responses to bradykinin in conscious
wistar rats with streptozotocin-induced diabetes mellitus. British Journal of
Pharmacology 103, 1357-1362.
Kihara, M., Mitsui, M. K., Mitsui, Y., Okuda, K., Nakasaka, Y., Takahashi, M., and
Schmelzer, J. D. (1999). Altered vasoreactivty to angiotensin II in experimental
diabetic neuropathy. Muscle and Nerve 22, 920-925.
Kihara, M., Schmelzer, J. D., Poduslo, J. F., Curran, G. L., Nicklander, K. K., and Low,
P. A. (1991). Aminoguanidine effects on nerve blood flow, vascular permeability,
electrophysiology and oxygen free radicals. Proceedings of the National Academy
of Sciences of the United States of America 88, 6107-6111.
Kim, L. R., Whelpdale, K., Zurowski, M., and Pomeranz, B. (1998). Sympathetic
denervation impairs epidermal healing in cutaneous wounds. Wound Repair
Regeneration 6, 194-201.
King, G. L., Shiba, T., Oliver, J., Inoguchi, T., and Bursell, S. (1994). Cellular and
molecular abnormalities in the vascular endothelium of diabetes mellitus. Annual
Review of Medicine 45, 179-188.
Kinoshita, J. H., and Nishimura, C. (1988). Aldose reductase in diabetic complications.
Diabetes / Metabolism Reviews 4, 323-337.
Kishi, Y., Nickander, K. K., Schmelzer, J. D., and Low, P. A. (2000). Gene expression of
antioxidant enzymes in experimental diabetic neuropathy. Journal of Peripheral
Nervous System 5, 11-18.
Kjartansson, J., Dalsgaard, C. J., and Jonsson, C. E. (1987). Decreased survival of
experimental critical flaps in rats after sensory denervation with capsaicin. Plastic
and Reconstructive Surgery 79, 218-221.
Kobayashi, T., and Kamata, K. (1999a). Effect of insulin treatment on smooth muscle
contractility and endothelium-dependent relaxation in rat aorta from established
STZ-induced diabetes. British Journal of Pharmacology 127, 835-842.
Kobayashi, T., and Kamata, K. (1999b). Relationship among cholesterol, superoxide
anion and endothelium-dependent relaxation in diabetic rats. European Journal of
Pharmacology 367, 213-222.
Konishi, H., Tanaka, M., Takemura, Y., Matsuzaki, H., Ono, Y., Kikkawa, U., and
Nishizuka, Y. (1997). Activation of protein kinase C by tyrosine phosphorylation
in response to H2O2. Proceedings of the National Academy of Sciences of the
United States of America 94, 11233-11237.
Kontos, H. A., and Wei, E. P. (1993). Hydroxyl-dependent inactivation of guanylate
cyclase in cerebral arterioles by methylene blue and by LY83583. Stroke 24, 427434.
Kourembanas, S., McQuillian, L. P., Leung, G. K., and Faller, D. V. (1993). Nitric oxide
regulates the expression of vasoconstrictors and growth factors by vascular
endothelium under both normoxia and hypoxia. Journal of Clinical Investigations
92, 99-104.
Kowalski, M. L., Sliwinska-kowalska, M., and Kaliner, M. A. (1990). Neurogenic
inflammation, vascular permeability, and mast cells.2. Additional evidence
351
indicating that mast cells are not involved in neurogenic inflammation. Journal of
Immunology 145, 1214-1221.
Koya, D., and King, G. L. (1998). Protein kinase C activation and the development of
diabetic complications. Diabetes 47, 859-866.
Kuboki, K., Jiang, Z. Y., Takahara, N., Ha, S. W., Igarashi, M., Yamauchi, T., Feener, E.
P., Herbert, T. P., Rhodes, C. J., and King, G. L. (2000). Regulation of endothelial
constitutive nitric oxide synthase gene expression in endothelial cells and
microvessels in vivo: a specific vascular action of insulin. Circulation 101, 676681.
Kumar, J. S. S., and Menon, V. P. (1993). Effect of diabetes on levels of lipid peroxides
and glycolipids in rat brain. Metabolism 42, 1435-1439.
Kummer, W., Fischer, A., Mundel, P., Mayer, B., Hoba, B., Philippin, B., and Preissler,
U. (1992). Nitric oxide synthase in VIP-containing vasodilator nerve fibres in the
guinea pig. Neuroreport 3, 653-655.
Kurvers, H. A., Tangelder, G. J., De Mey, J. G., Slaff, D. W., Beuk, R. J., van den
Wildenberg, F. A., Kitslaar, P. J., Reneman, R. S., and Jacobs, M. J. (1997). Skin
blood flow abnormalities in a rat model of neuropathic pain: result of decreased
sympathetic vasoconstrictor outflow? Journal of Autonomic Nervous System 63
19-29.
Lalla, E., Lamster, I. B., and Schmidt, A. M. (1998). Enhanced interaction of advanced
glycation end-products with their cellular receptor RAGE: implications for the
pathogenesis of accelerated periodontal disease in diabetes. Annals of
periodontology 3, 13-19.
Lam, H. C., Lee, J. K., Chiang, H. T., Koh, S. J., Han, T. M., Lu, C. C., Yang, C. Y., and
Hao, L. J. (1995). Does endothelin play a role in the pathogenesis of early diabetic
nephropathy? Journal of Cardiovascular Pharmacology 26 (Suppl 3), S479S481.
Lash, J. M., and Bohlen, H. G. (1991). Structural and functional origins of supressed
acetylcholine vasodilation in diabetic rat intestinal arterioles. Circulation
Research 69, 1259-1268.
Lawrence, E., and Brain, S. D. (1992). Altered microvascular reactivity to endothelin-1,
endothelin-3 and NG-nitro-L-arginine methyl ester in streptozotocin-induced
diabetes mellitus. British Journal of Pharmacology 106, 1035-1040.
Lawrence, E., Siney, L., Wilsoncroft, P., Knock, G. A., Terengi, G., Polak, J. M., and
Brain, S. D. (1995). Evidence for ET-A and ET-B receptors in rat skin and an
investigation of their function in the cutaneous microvasculature. British Journal
of Pharmacology 115, 840-844.
Lazarow, A. (1947). Protection agaist alloxan diabetes. Anatomical Record 97, 353-362.
Le Doux, S. P., Hall, C. R., Forbes, P. M., Patton, N. J., and Wilson, G. L. (1988).
Mechanisms of nicotinamide and thymidine protection from alloxan and
streptozotocin toxicity. Diabetes 37, 1015-1019.
Lee, P. Y., and Pearce, F. L. (1990). Histamine secretion from mast cells stimulated with
bradykinin. Agents and Actions 30, 67-69.
Legan, E. (1989). Effects of streptozotocin-induced hyperglycemia on agonist-stimulated
phosphoinositol turnover in rat aorta. Life Sciences 45, 371-378.
352
Lembeck, F., and Holzer, P. (1979). Substance P as neurogenic mediator of antidromic
vasodilatation and neurogenic plasma extravasation. Naunyn-Schmiedebergs
Archives of Pharmacology 310, 175-183.
Lenzen, S., and Munday, R. (1991). Thiol-group reactivity, hydrophilicity, and stability
of alloxan, its reduction products and its N-methyl derivatives and a comparison
with ninhydrin. Biochemical Pharmacology 42, 1385-1391.
Lenzen, S., and Panten, U. (1988). Alloxan: history and mechanism of action.
Diabetologia 31, 337-342.
Lewis, T. (ed.) (1927). "Blood vessels of human skin and their responses". Shaw,
London.
Li, C. (1992). Study on capillary permeability of myocardium in diabetic rats. Chung
Hua Ping Li Hsueh Tsa Chih 21, 183-185.
Li, P. F., Maasch, C., Haller, H., Dietz, R., and von Harsdorf, R. (1999). Requirement for
protein kinase C in reactive oxygen species-induced apoptosis of vascular smooth
muscle cells. Circulation 100, 967-973.
Lieu, A. T., and Ried, J. J. (1994). Changes in the responsiveness to endothelin-1 in
isolated atria from diabetic rats. European Journal of Pharmacology 261, 33-42.
Lind, H., Erlinge, D., Brunckwall, J., and Edvinsson, L. (1995). Selective attenuation of
neuropeptide Y-mediated contractile response in blood vessels from patients with
diabetes mellitus. Clinical Autonomic Research 5, 191-197.
Lindsay, R. M., Lockett, C., Sternberg, J., and Winter, J. (1989). Neuropeptide
expression in cultures of adult sensory neurons: modulation of substance P and
calcitonin gene-related peptide levels by nerve growth factor. Neuroscience 33,
53-65.
Lipton, S. A. (1999). Neuronal protection and destruction by NO. Cell Death and
Differentiation 6, 943-951.
Liu, X., Wang, J., Takeda, N., Biaglia, L., Panagia, V., and Dhalla, N. S. (1999). Changes
in cardiac protein kinase C activities and isozymes in streptozotocin-induced
diabetes. American Journal of Physiology 277, E798-E804.
Lockett, M. J., and Tomlinson, D. R. (1992). The effects of dietary treatment with
essential fatty acids on sciatic nerve conduction and activity of the Na+/K+ pump
in streptozotocin-diabetic rats. British Journal of Pharmacology 105, 355-360.
Lodge, N. L., Zhang, R., Halaka, N. N., and Moreland, S. (1995). Functional role of
endothelin ET-A and ET-B receptors in venous and arterial smooth muscle.
European Journal of Pharmacology 287, 279-285.
Lorenzi, M., and Cagliero, E. (1991). Pathobiology of endothelial and other vascular cells
in diabetes mellitus. Diabetes 40, 653-659.
Lotz, M., Vaughan, J. H., and Carson, D. A. (1988). Effects of neuropeptides on
production of inflammatory cytokines by human monocytes. Science 241, 12181221.
Lynn, B. (1989). In “Pharmacology of the skin” (M. W. Greaves and S. Shuster, eds.),
Vol. 1, pp. 89-116. Springer, Berlin.
Macgoub, M. A., and Abdel-Fattah, A. S. (1998). Diabetes mellitus and cardiac function.
Molecullar and Cellular Biochemistry 180, 59-64.
353
MacLeod, K. M., and McNeill, J. H. (1985). The influence of chronic experimental
diabetes on contractile responses of rats isolated blood vessels. Canadian Journal
of Physiology and Pharmacology 63, 52-57.
Maeda, K., Fernyhough, P., and Tomlinson, D. R. (1996). Regenerating sensory neurones
of diabetic rats express reduced levels of mRNA for GAP-43, gammapreprotachykinin and nerve growth factor receptors, trkA and p75NGFR. Brain
Research 37, 166-174.
Maggi, C. A. (1991). The pharmacology of the efferent function of sensory nerves.
Journal of Autonomic Pharmacology 11, 173-208.
Maggi, C. A. (1995). Tachykinins and calcitonin gene-related factor (CGRP) as cotransmitters released from peripheral endings of sensory nerves. Progress in
Neurobiology 45, 1-98.
Maillard, L. C. (1912). Action des acides amines sur les sucres; formation des
melanoidines par voie Methodique. Comptes Rendus Hebdomadires des Seances
del Academie des Sciences. D: Sciences Naturelles 154, 66-68.
Mak, D. H., IP, S. P., Li, P. C., Poon, M. K., and Ko, K. M. (1996). Alterations in tissue
glutathione antioxidant system in streptozotocin-induced diabetic rats. Molecular
and Cellular Biochemistry 162, 153-158.
Makimattila, S., Mantysaari, M., and Groop, P. H. (1997). Hyperreactivity to
nitrovasodilators in forearm vasculature is related to autonomic dysfunction in
insulin-dependent diabetes mellitus. Circulation 95, 618-625.
Makino, A., and Kamata, K. (1998). Elevated plasma endothelin-1 level in
streptozotocin-induced diabetic rats and responsiveness of the mesenteric arterial
bed to endothelin-1. British Journal of Pharmacology 123, 1065-1072.
Makino, A., Ohuchi, K., and Kamata, K. (2000). Mechanisms underlying the attenuation
of endothelium-dependent vasodilatation in the mesenteric arterial bed of the
streptozotocin-induced diabetic rats. British Journal of Pharmacology 130, 549556.
Makita, Z., Yanagisawa, K., Kuwajima, S., Yoshioka, N., Atsumi, T., Hasunuma, Y., and
Koike, T. (1995). Advanced glycation end-products and diabetic nephropathy.
Journal of Diabetes and its Complications 9, 265-268.
Malhotra, A., and Sanghi, V. (1997). Regulation of contractile proteins in diabetic heart.
Cardiovascular Research 34, 34-40.
Mankovsky, B. N., Metzger, B. E., Molitch, M. E., and Biller, J. (1996). Cerebrovascular
disorders in patients with diabetes mellitus. Journal of Diabetes and its
Complications 10, 228-242.
Mansford, K. R., and Opie, L. (1968). Comparison of metabolic abnormalities in diabetes
mellitus induced by streptozotocin or by alloxan. Lancet 1, 670-677.
Marino-Buslje, C., Martin-Martinez, M., Mizuguchi, K., Siddle, K., and Blundell, T. L.
(1999). The insulin structure: from protein sequence to structure. Biochemical
Society Transactions 27, 715-726.
Markle, R. A., and Walker, D. K. (1996). Effects of streptozotocin-induced diabetes and
insulin treatment on substance P of the rat arterial wall. Life Sciences 58, 11231129.
354
Marsen, T. A., Egink, G., Suckau, G., and Baldamus, C. A. (1999). Tyrosine-kinasedependent regulation of the nitric oxide synthase gene by endothelin-1 in human
endothelial cells. Pflugers Archives 438, 538-544.
Marshall, S. M. (1998). In “The Diabetes Annual” (S. M. Marshall, P. D. Home, and R.
A. Rizza, eds.), Vol. 11, pp. 169-194. Elsevier, Amesterdam.
Martin, P., Hopkinson-Woolley, J., and Mccluskey, J. (1992). Growth factors and
cutaneous wound repair. Progress in Growth Factor Research 4, 25-44.
Martinez-Nieves, B., and Dunbar, J. C. (1999). Vascular dilatory responses to sodium
nitroprusside (SNP) and alpha-adrenergic antagonism in female and male normal
and diabetic rats. Proceedings of the Society for Experimental Biology and
Medicine 222, 90-98.
Masao, K., Yasunobu, H., Hiroshi, H., Akihiro, T., Daisuki, N., Etsu, S., Kenjiro, K.,
Atuso, G., and Masao, O. (1999). Effects of hypertension, diabetes mellitus and
hypercholesterolemia on endothelin type-B receptor-mediated nitric oxide release
from rat kidney. Circulation 99, 1242-1248.
Mathison, R., and Davison, J. S. (1993). Attenuated plasma extravasation to sensory
neuropeptides in diabetic rats. Agents and Actions 38, 55-59.
Matzuzaki, K. (1993). Effect of skin flaps on plasma endothelin-1 levels. Annals of
Plastic Surgery 31, 499-503.
Mayer, J. H., and Tomlinson, D. R. (1983). Prevention of defects of axonal transport and
nerve conduction velocity by oral administration of myo-inositol or an aldose
reductase inhibitor in streptozotocin-diabetic rats. Diabetologia 25, 433-438.
Mayer, T. O., and Messina, E. J. (1996). Depressed arteriolar responsiveness to
norepinephrine in streptozotocin-induced diabetes in the rat. Prostaglandins 52,
415-430.
Mayhan, W. G. (1994). Responses of cerebral arterioles to activation of betaadrenoreceptors during diabetes mellitus. Stroke 25, 141-146.
Mayhan, W. G., and Patel, K. P. (1998). Treatment with dimethylthiourea prevents
impaired dilatation of basilar artery during diabetes mellitus. American Journal of
Physiology 274, H1895-H1901.
Mayhan, W. G., Simmons, L. K., and Sharpe, G. M. (1991). Mechanism of impaired
responses to cerebral arterioles during diabetes mellitus. American Journal of
Physiology 260, H319-H326.
McGinty, A., Scholfield, C. N., Liu, W. H., Anderson, P., Hoey, D. E., and Trimble, E.
R. (1999). Effect of glucose on endothelin-1 induced calcium transients in
cultured bovine retinal pericytes. Journal of Biological Chemistry 274, 2525025253.
McKay, M. K., and Hester, R. L. (1996). Role of nitric oxide, adenosine, and ATPsensitive potassium channels in insulin-induced vasodilatation. Hypertension 28,
202-208.
McMillan, D. E. (1997). Development of vascular complications in diabetes. Vascular
Medicine 2, 132-142.
Meller, S. T., and Gebhart, G. F. (1993). Nitric oxide and nociceptive processing in the
spinal cord. Pain 52, 127-136.
Mendez, J. D., and Ramos, H. G. (1994). Animal models in diabetes research: review.
Archives of Medical Research 25, 367-375.
355
Mene, P., Pascale, C., Teti, A., Bernardini, S., Cinotti, G. A., and Pugliese, F. (1999).
Effects of advanced glycation end-products on cytosolic calcium signalling of
cultured human mesangial cells. Journal of the American Society of Nephrology
10, 1478-1486.
Merhi, M., Helme, R. D., and Khalil, Z. (1998). Age-related changes in sympathetic
modulation of sensory nerve activity in rat skin. Inflammation Research 47, 239244.
Miao, F. J., Khasar, S. G., Green, P. G., and Levine, J. D. (1996). Tachyphylaxis
develops to bradykinin-induced plasma extravasation in the rat. Neuroscience
Letters 208, 143-145.
Millington, P. F., and Wilkinson , R. (eds.) (1983). "Skin, biological structure and
function". Cambridge University Press, Cambridge.
Mione, M. C., Ralevic, V., and Burnstock, G. (1990). Peptides and vasomotor
mechanisms. Pharmacology and Therapeutics 46, 429-468.
Mombouli, J. V., and Vanhoutte, P. M. (1995). Kinins and endothelial control of vascular
smooth muscle. Annual Reviews of Pharmacology and Toxicology 35, 679-705.
Monboisse, J. C., and Borel, J. P. (1992). Oxidative damage to collagen. EXS 62, 323327.
Moncada, S., and Higgs, A. (1993). The L-arginine-Nitric oxide pathway. The new
England Journal of Medicine 329, 2002-2012.
Moncada, S., Palmer, R. M. J., and Higgs, E. A. (1991). Nitric oxide: physiology,
pathophysiology, and pharmacology. Pharmacological Reviews 43, 109-142.
Morff, R. J. (1990). Microvascular reactivity to norepinephrine at different arteriolar
levels and durations of streptozotocin-induced diabetes. Diabetes 39, 354-360.
Morian, W. D., and Colen, L. B. (1990). Wound healing in diabetes mellitus. Clinics in
Plastic Surgery 17, 493-501.
Morise, T., Takeuchi, Y., Kawano, M., Koni, I., and Takeda, R. (1995). Increased plasma
levels of immunoreactive endothelin and von Willebrand factor in NIDDM
patients. Diabetes Care 18, 87-89.
Mossine, V. V., Linetsky, M., Glinsky, G. V., Ortwerth, B. J., and Feather, M. S. (1999).
Superoxide free radical generation by amadori compounds: the role of acyclic
forms and metal ions. Chemical Research in Toxicology 12, 230-236.
Mullarkey, C. J., Edelstein, D., and Brownlee, M. (1990). Free radical generation by
early glycation products: a mechanism for accelerated atherogenesis in diabetes.
Biochemical and Biophysical Research Communications 173, 932-939.
Muller, G., Fruhauf, A., and Mathias, B. (1986). Thiobarbituric acid positive substances
as indicators of lipid peroxidation. Zeitschrift fur die Gesamte Innere Medizin und
Ihre grenzgebiete 41, 673-676.
Munday, R., Ludwig, K., and Lenzen, S. (1993). The relationship between the
physicochemical properties and the biological effects of alloxan and several Nalkyl substituted alloxan derivatives. Journal of Endocrinology 139, 153-163.
Munzel, T., Heitzer, T., and Harrison, D. G. (1997). The physiology and pathophysiology
of the nitric oxide / superoxide anions. Herz 22, 158-172.
Murat, N., Kalkan, S., and Gidener, S. (1999). Effect of varapamil on responses to
endothelin-1 in aortic rings from streptozotocin-induced diabetic rats.
Pharmacological Research 40, 37-40.
356
Murohara, T., Witzenbichler, B., Spyridopoulos, I., Asahara, T., Ding, B., Sullivan, A.,
Losordo, D. W., and Isner, J. M. (1999). Role of endothelial nitric oxide synthase
in endothelial cell migration. Arteriosclerosis, Thrombosis, and Vascular Biology
19, 1156-1161.
Musabayane, C. T., Ndhlovu, C. E., and Balment, R. J. (1995). Renal fluid and
electrolyte handling in streptozotocin-diabetic rats. Renal Failure 17, 107-116.
Mutafova-Yambolieva, V. N., and Westfall, D. P. (1998). Inhibitory and facilitatory
presynaptic effects of endothelin on sympathetic cotransmission in the rat isolated
tail artery. British Journal of Pharmacology 123, 136-142.
Nakamura, J., Kato, K., Hamada, Y., Nakayama, M., Chaya, S., Nakashima, E., Kasuya,
Y., Mizubayashi, R., Miwa, K., Yasuda, Y., and Hotta, N. (1999). A protein
kinase C-beta-selective inhibitor ameliorates neural dysfunction in streptozotocininduced diabetic rats. Diabetes 48, 2090-2095.
Nakamura, L., Saito, M., Fukumoto, Y., Yoshida, M., Nishi, K., Weiss, R. M., and
Latifpour, J. (1997). Experimental diabetes up-regulates the expression of
uretereral endothelin receptor. Peptides 18, 1091-1093.
Nakamura, S., Naruse, M., Naruse, K., Demura, H., and Uemura, H. (1990).
Immunohistochemical localisation of endothelin in cultured bovine endothelial
cells. Histochemistry 94, 475-477.
Nakamura, T., Ebihara, I., Fukui, M., Tomuno, Y., and Koide, H. (1995). Effects of
specific endothelin receptor A antagonist on mRNA levels for extracellular matrix
components and growth factors in diabetes glomeruli. Diabetes 44, 895-899.
Nankane, M., Schmidt, H. H. H. W., Pollock, J. S., Forstermann, U., and Murad, F.
(1993). Cloned human brain nitric oxide synthse is highly expressed in skeletal
muscle. FEBS letters 316, 175-180.
Nanney, L. B. (1990). Epidermal and dermal effects of epidermal growth factor during
wound repair. Journal of Investigative Dermatology 94, 624-629.
Nathan, D. M. (1993). Long-term complications of diabetes mellitus. The New England
Journal of Medicine 328, 1676-1685.
Nava, P., Collados, M. T., Masso, F., and Guarner, V. (1997). Endothelin mediation of
insulin and glucose-induced changes in vascular contractlity. Hypertension 30,
825-829.
Nawroth, P. P., Bierhaus, A., Vogel, G. E., Hofmann, M. A., Zumbach, M., Wahl, P., and
Ziegler, R. (1999). Non-enzymatic glycation and oxidative stress in chronic
illness and diabetes mellitus. Medizinische Klinik 94, 29-38.
Nayler, W. G. (1990). "The endothelins". Springer-Verlag, ISBN 0-387-52856-3.
Nemeth, J., Szilvassy, Z., Than, M., Oroszi, G., Sari, R., and Szolcsanyi, J. (1999a).
Decreased sensory neuropeptide release from trachea of rats with streptozotocininduced diabetes. European Journal of Pharmacology 369, 221-224.
Nemeth, J., Than, M., Sari, R., Peitl, B., Oroszi, G., Farkas, B., Szolcsanyi, J., and
Szilvassy, Z. (1999b). Impairment of neurogenic inflammatory and antiinflammatory responses in diabetic rats. European Journal of Pharmacology 386,
83-88.
Newby, D. E., Sciberras, D. G., Ferro, C. J., Gertz, B. J., Sommerville, D., Majumdar, A.,
Lowry, R. C., and Webb, D. J. (1999). Substance P-induced vasodilatation is
357
mediated by the neurokinin type 1 receptor but does not contribute to basal
vascular tone in man. British Journal of Clinical Pharmacology 48, 336-344.
Nicholas, A. P., Pieribone, V., and Hokfelt, T. (1993). Distribution of mRNAs for alpha2 adrenergic receptor subtypes in rat brain: an in situ hybridization study. Journal
of Comparative Neurology 328, 575-594.
Nilsson, J., Von Euler, A. M., and Dalsgaard, C. J. (1985). Stimulation of connective
tissue cell growth by substance P and substance K. Nature 315, 61-63.
Nishio, E., and Watanabe, Y. (1996). Glucose-induced down-regulation of NO
production and inducible NOS expression in cultured rat aortic vascular smooth
muscle cells: role of protein kinase C. Biochemical and Biophysical Research
Communications 229, 857-863.
Nishio, Y., Kashiwagi, A., Kida, Y., Kodama, M., Abe, N., Saeki, Y., and Shigeta, Y.
(1988). Deficiency of cardiac beta-adrenergic receptor in streptozotocin-induced
diabetic rats. Diabetes 37, 1181-1187.
Norberg, K. A., and Hamberger, B. (1964). The sympathetic adrenergic neuron: some
characteristic revealed by histochemical studies on the interneuronal distribution
of the transmitter. Acta Physiologica Scandinavica Supplementum 238, 1-42.
Nugent, A. G., McGurk, C., Hayes, R., and Johnson, D. (1996). Impaired
vasoconstriction to endothelin-1 in patients with NIDDM. Diabetes 45, 105-107.
Nukada, H. (1993). The suceptibility of rat diabetic nerve to ischemia: increased or
decreased? Journal of the Neurological Sciences 119, 162-168.
Oberg, P. A. (1990). Laser Doppler flometry. Biomedical Engineering 18, 125-163.
Obrosova, I. G., Fathallah, L., Lang, H. J., and Greene, D. A. (1999). Evaluation of a
sorbitol dehydrogenase inhibitor on diabetic peripheral nerve metabolism: a
prevention study. Diabetologia 42, 1187-1194.
O'Bryan, G. T., and Hostetter, T. H. (1997). The renal hemodynamic basis of diabetic
nephropathy. Seminars in Nephrology 17, 93-100.
Ochodnicka, E., Ochdnicky, M., Belej, K., Fusekova, E., and Bosel'ova, L. (1995).
Quantitative analysis of myelinated nerve fibers of peripheral nerve in
streptozotocin-induced
diabetes
mellitus.
Molecular
and
Chemical
Neuropathology 25, 225-233.
Odetti, P., Aragno, I., Rolandi, R., Garibaldi, S., Valentini, S., Cosso, L., Traverso, N.,
Cottalasso, D., Pronzato, M. A., and Marinari, U. M. (2000). Scanning force
microscopy reveals structural alterations in diabetic rat collagen fibrils: role of
protein glycation. Diabetes Metabolism Research and Review 16, 74-81.
O'Driscoll, G., Green, D., Rankin, J., Stanton, K., and Taylor, R. (1997). Improvement in
endothelial function by angiotensin converting enzyme inhibition in insulindependent diabetes mellitus. Journal of Clinical Investigation 100, 678-684.
Oh, U., Hwang, S. W., and Kim, D. (1996). Capsaicin activates a non-selective cation
channel in cultured dorsal root ganglion neurones. Journal of Neuroscience 16,
1659-1667.
Ohkuwa, T., Sato, Y., and Naoi, M. (1995). Hydroxyl radical formation in diabetic rats
induced by streptozotocin. Life Sciences 56, 1789-1798.
Ohshima, H., Oguchi, S., Adachi, H., Iida, S., Suzuki, H., Sugimura, T., and Esumi, H.
(1992). Purification of nitric oxide synthase from bovine brain: immunological
358
characterization and tissue distribution. Biochemical and Biophysical Research
Communications 183, 238-244.
Ohtani, N., Ohta, M., and Sugano, T. (1997). Microdialysis study of modification of
hypothalamic neurotransmitters in streptozotocin-diabetic rats. Journal of
Neurochemistry 69, 1622-1628.
Okumura, K., Nishiura, T., Awaji, Y., Kondo, J., Hashimoto, H., and Ito, T. (1991). 1, 2diacylglycerol content and its fatty acid composition in thorasic aorta of diabetic
rats. Diabetes 40, 820-824.
Okumura, M., Okuda, T., Nakamura, T., and Yajima, M. (1996). Acceleration of wound
healing in diabetic mice by basic fibrobalst growth factor. Biological and
Pharmaceutical Bulletin 19, 530-535.
Onkamo, P., Vaananen, S., Karvonen, M., and Tuomilehto, J. (1999). Worldwide
increase in incIdence of type I diabetes- the analysis of the data on published
incidence trends. Diabetologia 42, 1395-1403.
Orie, N. N., and Aloamaka, C. P. (1993). Duration-dependent variability in the responses
of diabetic rat aorta to noradrenaline and 5-hydroxytryptamine. General
Pharmacology 24, 243-246.
Ou, P., and Wolff, S. P. (1993). Aminoguanidine: a drug proposed for the prophylaxis in
diabetes inhibits catalase and generates hydrogen peroxide in vitro. Biochemical
Pharmacology 43, 1139-1144.
Oyama, Y., Kawasaki, H., Hattori, Y., and Kanno, M. (1986). Attenuation of
endothelium-dependent relaxations in aorta from diabetic rats. European Journal
of Pharmacology 132, 75-78.
Pa, V., and Tomlinson, D. R. (1997). Effects of adrenergic stimulation on sciatic nerve
blood flow in diabetic and control rats. European Journal of Pharmacology 329,
147-152.
Palmer, A. M., Gopaul, N., Dhir, S., Thomas, C. R., Poston, L., and Tribe, R. M. (1998).
Endothelial dysfunction in streptozotocin-diabetic rats is not reversed by dietary
probucol or simvastatin supplementation. Diabetologia 41, 157-164.
Paolisso, G., and Giugliano, D. (1996). Oxidative stress and insulin action: is there a
relationship? Diabetologia 39, 357-363.
Park, J. Y., Ha, S. W., and King, G. L. (1999). The role of protein kinase C activation in
the pathogenesis of diabetic vascular complications. Peritoneal Dialysis
International 19 (Suppl 2), S222-S227.
Park, J. Y., Takahara, N., Gabriele, A., Chou, E., Naruse, K., Suzuma, K., Yamauchi, T.,
Ha, S. W., Meier, M., Rhodes, C. J., and King, G. L. (2000). Induction of
endothelin-1 expression by glucose: an effect of protein kinase C activation.
Diabetes 49, 1239-1248.
Patel, K. P., Zhang, K., Hein, M., and Mayhan, W. G. (1997). Peripheral noradrenergic
turnover in streptozotocin-induced diabetic rats. Diabetes Research and Clinical
Practice 35, 1-9.
Pereira, B., Rosa, L. F., Safi, D. A., Bechara, E. J., and Curi, R. (1995). Hormonal
regulation of superoxide dismutase, catalase, and glutathione peroxidase activities
in rat macrophages. Biochemical Pharmacology 50, 2093-2098.
359
Perkins, M. N., Campbell, E., and Dray, A. (1993). Antinociceptive activity of the
bradykinin B1 and B2 receptor antagonists, des-Arg9,[Leu8]-BK and HOE 140,
in two models of persistent hyperalgesia in the rat. Pain 53, 191-197.
Pernow, J. (1988). Co-release and functional interactions of neuropeptide Y and
noradrenaline in peripheral sympathetic vascular control. Acta Physiologica
Scandinavica 134 (Suppl 568), 1-56.
Pertovaara, L., Kaipainen, A., and Mustonen, T. (1994). Vascular endothelial growth
factor is induced in response to transforming growth factor-beta in fibroblastic
and epithelial cells. Journal of Biological Chemistry 269, 6271-6274.
Petley, A., Macklin, B., Renwick, A. G., and Wilkin, T. J. (1995). The pharmacokinetics
of nicotinamide in human and rodents. Diabetes 44, 152-155.
Pfaffman, M. A., Ball, C. R., Darby, A., and Hilman, R. (1982). Insulin reversal of
diabetes-induced inhibition of vascular contractility in the rat. American Journal
of Physiology 242, H490-H495.
Pham, H. T., Economides, P. A., and Veves, A. (1998). The role of endothelial function
on the foot. Microcirculation and wound healing in patients with diabetes. Clinics
in Podiatric Medicine and Surgery 15, 85-93.
Pheng, L. H., Nguyen-Le, X. K., Nsa Allogho, S., Gobeil, F., and Regoli, D. (1997).
Kinin receptors in the diabetic mouse. Canadian Journal of physiology and
Pharmacology 75, 609-611.
Pickup, J. C., and Williams, G. (eds.) (1997). "Text book of diabetes". Blackwell Science
Pty Ltd, London.
Pieper, G. (1998). Review of alterations in endothelial nitric oxide production in diabetes.
Hypertension 31, 1047-1060.
Pieper, G. M. (1999). Enhanced, unaltered and impaired nitric oxide-mediated
endothelium-dependent relaxation in experimental diabetes mellitus: importance
of disease duration. Diabetologia 42, 204-213.
Pieper, G. M., and Dondlinger, L. A. (1997). Plasma and vascular tissue arginine are
decreased in diabetes: acute arginine supplementation restores endotheliumdependent relaxation by augmenting cGMP. Journal of Pharmacology and
Experimental Therapeutics 283, 684-691.
Pieper, G. M., Jordan, M., and Roza, A. M. (1997a). Chronic treatment with the 21aminosteriod U74389F, an inhibitor of lipid peroxidation, does not prevent
diabetic endothelial dysfunction. Cardiovascular and Drugs Therapy 11, 435440.
Pieper, G. M., Langenstroer, P., and Siebeneich, W. (1997b). Diabetic-induced
endothelial dysfunction in rat aorta: role of hydroxyl radical. Cardiovascular
Research 34, 145-156.
Pieper, G. M., and Peltier, B. A. (1995). Amelioration by L-arginine of a dysfunctional
arginine / nitric oxide pathway in diabetic endothelium. Journal of
Cardiovascular Pharmacology 25, 397-403.
Pieper, G. M., and Siebeneich, W. (1998). Oral administration of antioxidant, Nacetylcysteine, aborgates diabetes-induced endothelial dysfunction. Journal of
Cardiovascular Pharmacology 32, 101-105.
360
Pierre, E. J., Barrow, R. E., Hawkins, H. K., Nguyen, T. T., Sakurai, Y., Desai, M.,
Wolfe, R. R., and Herndon, D. N. (1998). Effects of insulin on wound healing.
Journal of Trauma 44, 342-345.
Pinnell, S. R., Eisen, A. Z., Graham Smith, J., Bauer, E. A., Fleischmajer, R., Silbert, J.,
and Uitto, J. J. (1979). Connective tissue diseases and aging. Journal of
Investigative Dermatology 73, 485-490.
Pinter, E., Helyes, Z., Petho, G., and Szolcsanyi, J. (1997). Noradrenergic and peptidergic
sympathetic regulation of cutaneous microcirculation in the rat. European Journal
of Pharmacology 325, 57-64.
Pollock, J. S., Forstermann, U., Mitchell, J. A., Warner, T. D., Schmidt, H. H. H. W.,
Nankane, M., and Murad, F. (1991). Purification and characterization of
particulate endothelium-derived relaxing factor synthase from cultured and native
bovine aortic endothelial cells. Proceedings of the National Academy of Sciences
of the United States of America 88, 10480-10484.
Prasad, K., and Bharadwaj, L. A. (1996). Hydroxyl radical: a mediator of acetylcholineinduced vascular relaxation. Journal of Molecular and Cellular Cardiology 28,
2033-2041.
Prickaerts, J., De Vente, J., Honig, W., Steinbusch, H., Ittersum, M. M. V., Blockland,
A., and Steinbusch, H. W. (2000). Nitric oxide synthase does not mediate
neurotoxicity after an i.c.v. injection of streptozotocin in the rat. Journal of
Neural Transmission 107, 745-766.
Qian, M., and Eaton, J. W. (2000). Glycochelates and etiology of diabetic peripheral
neuropathy. Free radical, Biology and Medicine 28, 652-666.
Quehenberger, P., Greten, J., v-Riedesel, J., Ziegler, R., Wahl, P., Speiser, W., Bierhaus,
A., and Nawroth, P. P. (1995). Endothelin-1 is under the control of the
transcription factors p50/p65 (NF-κB) and AP-1. Annals of Hematology 70, 211
(A 53) (abstr.).
Quilley, J., Fulton, D., and McGiff, J. C. (1997). Hyperpolarizing factors. Biochemical
Pharmacology 54, 1059-1070.
Radomski, M. W., Palmer, R. M., and Moncada, S. (1990). Characterization of the Larginine: nitric oxide pathway in human platelets. British Journal of
Pharmacology 101, 325-328.
Raidoo, D. M., and Bhoola, K. D. (1998). Pathophysiology of kallikrein-kinin system in
mammalian nervous system. Pharmacology and Therapeutics 79, 105-127.
Rakienten, N., Rakieten, M. L., and Nadkarni, M. V. (1963). Studies on the diabetogenic
action of streptozotocin (NSC-37917). Cancer Chemotherapy Report 29, 91-98.
Ralevic, V., Belai, A., and Burnstock, G. (1995a). Effects of streptozotocin-diabetes on
sympathetic nerve, endothelial and smooth muscle function in the rat mesenteric
arterial bed. European Journal of Pharmacology 286, 193-199.
Ralevic, V., Dusting, G. J., Khalil, Z., and Helme, R. D. (1992). In “Biology of nitric
oxide” (S. Moncada, M. A. Marletta, J. B. Higgs, and E. A. Higgs, eds.), pp. 204206. Portland press, Colchester.
Ralevic, V., Khalil, Z., Helme, R. D., and Dusting, G. (1995b). Role of nitric oxide in the
actions of substance P and other mediators of inflammation in rat skin
microvasculature. European Journal of Pharmacology 284, 231-239.
361
Ramakrishnan, R., and Nmasivayam, A. (1995). Norepinephrine and epinephrine levels
in the brain of alloxan diabetic rats. Neuroscience Letters 186, 200-202.
Ramamurthy, N. S., Kucine, A. J., McClain, S. A., McNamara, T. F., and Golub, L. M.
(1998). Topically applied CMT-2 enhances wound healing in streptozotocin
diabetic rat skin. Advances in Dental Research 12, 144-148.
Ramanadham, S., Lyness, W. H., and Tenner, T. E. (1984). Alterations in aortic and tail
artery reactivity to agonists after streptozotocin treatment. Canadian Journal of
Physiology and Pharmacology 62, 418-423.
Ramanadham, S., and Tenner, T. E. (1987). Alterations in the myocardial betaadrenoreceptor system of streptozotocin-diabetic rats. European Journal of
Pharmacology 136, 377-389.
Ramanathan, M., Jaiswal, A. K., and Bhattacharya, S. K. (1999). Superoxide dismutase,
catalase and glutathione peroxidase activities in the brain of streptozotocininduced diabetic rats. Indian Journal of Experimental Biology 37, 182-183.
Rees, D. D., Palmer, R. M. J., and Moncada, S. (1989). Role of endothelium-derived
nitric oxide in the regulation of blood pressure. Proceedings of the National
Academy of Sciences of the United States of America 86, 3375-3378.
Reichard, P., Nilsson, B. Y., and Rosenqvist, U. (1993). The effect of long-term
intensified insulin treatment on the development of microvascular complications
of diabetes mellitus. The New England Journal of Medicine 329, 304-309.
Reid, J. J., and La, M. (1995). Review: Endothelin-1 and the regulation of vascular tone.
Clinical and Experimental Pharmacology and Physiology 22, 315-323.
Reid, J. J., Wong-Dusting, H. K., and Rand, M. J. (1989). The effect of endothelin on
noradrenergic transmission in rat and guinea pig atria. European Journal of
Pharmacology 168, 93-96.
Renard, C., Chappey, O., Wautier, M. P., Nagashima, M., Lundh, E., Morser, J., Zhao,
L., Schmidt, A. M., Scherrmann, J. M., and Wautier, J. L. (1997). Recombinant
advanced glycation end-product receptor pharmacokinetics in normal and diabetic
rats. Molecular Pharmacology 52, 52-64.
Rendell, M. S., Finnegan, M. F., Haely, J. C., Lind, A., Milliken, B. K., Finney, D. F.,
and Bonner, R. F. (1998). The relationship of laser-Doppler skin blood flow
measurements to the cutaneous microvascular anatomy. Microvascular Research
55, 3-13.
Rendell, M. S., Kelly, S. T., Finney, D., Luu, K., McIntyre, S. F., and Terando, J. V.
(1993). Decreased skin blood flow early in the course of streptozotocin-induced
diabetes mellitus in the rat. Diabetologia 36, 907-911.
Rerup, C. C. (1970). Drugs producing diabetes through damage of the insulin secreting
cells. Pharmacological Reviews 23, 485-518.
Richards, A. M., Mitsou, J., Floyd, D. C., Terenghi, G., and McGrouther, D. A. (1997).
Neural innervation and healing. Lancet 350, 339-340.
Richey, K. J., Engrav, L. H., Pavlin, E. G., Murray, M. J., Gottileb, J. R., and
Walkinshaw, M. D. (1989). Topical growth factors and wound contraction in the
rat: part 1> Literature review and definition of the rat model. Annals of Plastic
Surgery 23, 159-165.
362
Rittenhouse, P. A., Marchand, J. E., Chen, J., Kream, R. M., and Leeman, S. E. (1996a).
Streptozotocin-induced diabetes is associated with altered expression of peptideencoding mRNAs in rat sensory neurons. Peptides 17, 1017-1022.
Rittenhouse, P. A., Markle, R. A., Marchand, J. E., Kream, R. M., and Leeman, S. E.
(1996b). Streptozotocin-induced diabetes produces a decrease in pitutary
substance P content and preprotachykinin mRNA. Neuroscience Letters 211, 7780.
Roberts, R. E., and McLean, W. G. (1997). Protein kinase C isozyme expression in
sciatic nerves and spinal cords of experimentally diabetic rats. Brain Research
754, 147-156.
Rohrbach, D. H., Wagner, C. W., Star, V. L., Martin, G. R., Brown, K. S., and Yoon, J.
W. (1983). Reduced synthesis of basement membrane heparan sulfate
proteoglycan in streptozotocin-induced diabetic mice. Journal of Biological
Chemistry 258, 11672-11677.
Rongen, G. A., Smits, P., and Thein, T. H. (1994). Endothelium and regulation of
vascular tone with emphasis on the role of nitric oxide. Netherlands Journal of
Medicine 44, 26-35.
Rosenberg, C. S. (1990). Wound healing in the patient with diabetes mellitus. Nursing
Clinics of North America 25, 247-261.
Rowell, L. B. (1993). "Human cardiovascular control". Oxford University Press, New
York.
Rubanyi, G. M. (1988). Vascular effects of oxygen-derived free radicals. Free Radical,
Biology and Medicine 4, 107-120.
Rubanyi, G. M., and Vanhoutte, P. M. (1986). Oxygen-derived free radicals,
endothelium, and responsiveness of vascular smooth muscle. American Journal of
Physiology 250, H815-H821.
Ruda, M. A., Besse, D., Inagaki, S., Deleon, M., and Ren, K. (1994). Nitric oxide
expression and regulation in the dorsal root ganglion and spinal cord. Annals of
the New York Academy of Science 738, 181-190.
Ruffolo, R. R. J. (ed.) (1995). "Endothelin receptors: from the gene to the human". CRC
press.
Rumble, J. R., Cooper, M. E., Soulis-Liparota, T., Cox, A., Wu, L., Youssef, S., Jasik,
M., Jerum, G., and Gilbert, R. E. (1997). Vascular hypertrophy in experimental
diabetes: role of advanced glycation end-products. Journal of Clinical
Investigation 99, 1016-1027.
Rungby, J., Flyvbjerg, A., Anderson, H. B., and Nyborg, K. (1992). Lipid peroxidation in
early experimental diabetes in rats: effects of diabetes and insulin. Acta
Endocrinologica 126, 378-380.
Ryle, C., and Donaghy, M. (1995). Non-enzymatic glycation of peripheral nerve proteins
in human diabetics. Journal of the Neurological Sciences 129, 62-68.
Ryle, C., Leow, C. K., and Donaghy, M. (1997). Nonenzymatic glycation of peripheral
and central nervous system proteins in experimental diabetes mellitus. Muscle and
Nerve 20, 577-584.
Saenz De Tejada, I., Goldstein, I., Azadzoi, K., Krane, R. J., and Cohen, R. A. (1989).
Impaired neurogenic and endothelium-mediated relaxation of penile smooth
363
muscle from diabetic men with impotence. The New England Journal of Medicine
320, 1025-1030.
Sagara, M., Satoh, J., Wada, R., Yagihashi, S., Takahashi, K., Fukuzawa, M., Muto, G.,
Muto, Y., and Toyota, T. (1996). Inhibition of development of peripheral
neuropathy in streptozotocin-induced diabetic rats with N-acetylcysteine.
Diabetologia 39, 263-269.
Saito, M., Nishi, K., Foster, H. E., Weiss, R. M., and Latifpour, J. (1996). Effect of
experimental diabetes on rat prostate endothelin receptors. European Journal of
Pharmacology 310, 197-200.
Sakai, Y., Inazu, M., Shamoto, A., Zhu, B., and Homma, I. (1994). Contractile
hyperreactivity and alterations of PKC activity in gastric fundus smooth muscle of
diabetic rats. Pharmacology, Biochemistry and Behavior 49, 669-674.
Sakata, N., Miyamoto, K., Meng, J., Tachikawa, Y., Imanaga, Y., Takebayashi, S., and
Furukawa, T. (1998). Oxidative damage of vascular smooth muscle cells by the
glycated protein-cupric ion system. Atherosclerosis 136, 263-274.
Saniova, B. (1997). Tirilazad mesylate (Freedox)- an effective inhibitor of lipid
membrane peroxidation. Bratislavske Lekarske Listy 98, 339-342.
Sarges, R., and Oates, P. J. (1993). Aldose reductase inhibitors: recent developments.
Progress in Drug Research 40, 99-161.
Sasaki, H., and Okabe, E. (1993). Modification of hydroxyl radicals of functional
reactivity in rabbit lingual artery. Japanese Journal of Pharmacology 62, 305314.
Sato, N., Nakamura, M., Chikama, T., and Nishida, T. (1999). Abnormal deposition of
laminin and type IV collagen at corneal epithelial basement membrane during
wound healing in diabetic rats. Japanese Journal of Ophthalmology 43, 343-347.
Savage, M. W., Bodmer, C. W., Walker, A. B., Buchan, I. E., Masson, E. A., and
Wiliams, G. (1995). Vascular reactivity to noradrenaline and neuropeptide Y in
the streptozotocin-induced diabetic rat. European Journal of Clinical
Investigation 25, 974-979.
Schaan, B. D., Maeda, C. Y., Timm, H. B., Medeiros, S., Moraes, R. S., Ferlin, E.,
Fernandes, T. G., Riebeiro, J. P., Schmid, H., and Irigoyen, M. C. (1997). Time
course of changes in heart rate and blood pressure variability in streptozotocininduced diabetic rats treated with insulin. Brazilian Journal of Medicine and
Biological Research 30, 1081-1086.
Schaffer, M. R., Tantry, U., Efron, P. A., Ahrendet, G. M., Thornton, F. J., and Barbul,
A. (1997). Diabetes-impaired healing and reduced wound nitric oxide synthesis: a
possible pathophysiologic correlation. Surgery 121, 513-519.
Scherrer, U., Owlya, R., and Lepori, M. (1996). Body fat and sympathetic nerve activity.
Cardiovascular and Drugs Therapy 10 (suppl 1), 215-222.
Schiekofer, S., Balletshofer, B., Andrassy, M., Bierhaus, A., and Nawroth, P. P. (2000).
Endothelial dysfunction in diabetes mellitus. Seminars in Thrombosis and
Hemostasis 26, 503-511.
Schlicker, E., Erkens, K., and Gothert, M. (1988). Probable involvement of vascular
angiotensin II formation in the β2-adrenoreceptor-mediated facilitation of the
neurogenic vasopressor response in the pithed rat. Naunyn-Schmiedebergs
Archives of Pharmacology 338, 536-542.
364
Schmidt, A. M., Hori, O., Brett, J., Yan, S. D., Wautier, J. L., and Stern, D. (1994).
Cellular receptors for advanced glycation end-products: implication for induction
of oxidant stress and cellular dysfunction in the pathogenesis of vascular lesions.
Arteriosclerosis and Thrombosis 14, 1521-1527.
Schmidt, A. M., Hori, O., Cao, R., Yan, S. D., Brett, J., Wautier, J. L., Ogawa, S.,
Kuwabara, K., Matsumoto, M., and Stern, D. (1996). RAGE: a novel cellular
receptor for advanced glycation end-products. Diabetes 45 (Suppl 3), S77-S80.
Schmidt, A. M., Hori, O., Chen, J. X., Li, J. F., Crandall, J., Zhang, J., Cao, R., Yan, S.
D., Brett, J., and Stern, D. (1995). Advanced glycation end-products interacting
with their endothelial receptor induce expression of vascular cell adhesion
molecule-1(VCAM-1) in cultured human endothelial cells and in mice: a potential
mechanism for the accelerated vasculopathy of diabetes. Journal of Clinical
Investigation 96, 1395-1403.
Schmidt, H., Forman, L. A., Cao, X., Sherman, P. S., and Stevens, M. J. (1999).
Heterogenous cardiac sympathetic denervation and decreased myocardial nerve
growth factor in streptozotocin-induced diabetic rats. Diabetes 48, 603-608.
Schmidt, R.E., Plurad, S.B., Olaok, B.J., Scharp, D.W. (1983). The effect of pancreatic
islet transplantation and insulin therapy on experimental diabetic autonomic
neuropathy. Diabetes 32, 532-540.
Schnedl, W. J., Ferber, S., Johnson, J. H., and Newgard, C. B. (1994). STZ transport and
cytotoxicity: specific enhancement in GLUT2-expressing cells. Diabetes 43,
1326-1333.
Schwieler, J. H., Kahan, T., Nussberger, J., Johansson, M. C., and Hjemdahl, P. (1992).
Influence of angiotensin II, α− and β−adrenoreceptors on peripheral
noradrenergic neurotransmission in canine gracilis muscle in vivo. Acta
Physiologica Scandinavica 145, 333-343.
Sechi, L. A., Griffin, C. A., Grady, E. F., Grunfeld, C., Kalinyak, J. E., and Schambelan,
M. (1992). Tissue-specific regulation of insulin receptor mRNA levels in rats with
STZ-induced diabetes mellitus. Diabetes 41, 1113-1118.
Sememkovich, C. F., and Heinecke, J. W. (1997). The mystery of diabetes and
atherosclerosis: time for a new plot. Diabetes 46, 327-334.
Sensi, M., Pricci, F., Andreani, D., and Di Mario, U. (1991). Advanced nonenzymatic
glycation end-products(AGE): their relevance to aging and the pathogenesis of
late diabetic complications. Diabetes Research 16, 1-9.
Sentman, M. L., Jonsson, I. M., and Marklund, S. I. (1999). Enhanced alloxan-induced
beta cell damage and delayed recovery from hyperglycaemia in mice lacking
extracellular superoxide dismutase. Free Radical, Biology and Medicine 27, 790796.
Service, F. J., Rizza, R. A., Daube, J. R., O'Brian, P. C., and Dyck, P. J. (1985). Near
normoglycaemia improved nerve conduction and vibration in diabetic neuropathy.
Diabetologia 28, 722-727.
Shafrir, E. (1990). "Diabetes in animals". Elsevier, NewYork.
Shin, S., Lee, Y., Lin, S., Lia, Y., and Tsai, J. (1995). Decrease of renal endothelin-1
content and gene expression in diabetic rats with moderate hyperglycemia.
Nephron 70, 486-493.
365
Shir, Y., and Seltzer, Z. (1991). Effects of sympathectomy in a model of causalgiform
pain produced by partial sciatic nerve injury in rats. Pain 45, 309-320.
Shuler, R. L., Laskin, D. L., Gardner, C. R., Feder, L. S., and Laskin, J. D. (1995).
Lymphocyte-mediated nitric oxide production by rat endothelial cells. Journal of
Leukocyte Biology 57, 116-121.
Shyu, B. C., Olausson, B., Huang, K. H., Widerstorm, E., and Andersson, S. A. (1989).
Effects of sympathetic stimulation on C-fibre responses in rabbit. Acta
Physiologica Scandinavica 137, 73-84.
Simon, O. R., and West, M. E. (1992). Unstable diabetic state produced by a small dose
of streptozotocin in rats. West Indian Medical Journal 41, 146-149.
Siney, L., and Brain, S. D. (1996). Involvement of sensory neuropeptides in the
development of plasma extravasation in rat dorsal skin following thermal injury.
British Journal of Pharmacology 117, 1065-1070.
Singh, S., and Evans, T. W. (1997). Nitric oxide, the biological mediator of the decade:
fact or fiction? European Respiratory Journal 10, 699-707.
Skamarauskas, J. T., McKay, A. G., and Hunt, J. V. (1996). Aminoguanidine and its prooxidant effects on an experimental model of protein glycation. Free Radical,
Biology and Medicine 21, 801-812.
Smulders, R. A., Stehouwer, C. D., and Olthof, D. G. (1994). Plasma endothelin levels
and vascular effects of intravenous L-arginine infusion in subjects with
uncomplicated insulin-dependent diabetes mellitus. Clinical Science 87, 37-43.
Snowden, J. M., Kennedy, D. F., and Cliff, W. J. (1982). Wound contraction: the effects
of scab formation and the nature of the wound bed. Australian Journal of
Experimental Biology and Medical Science 60, 73-82.
Soe, B., Oemar, B. S., Siebenmann, R., von Segesser, L., and Luscher, T. F. (1994). Both
ET-A and ET-B receptors mediate contraction to endothelin-1 in human blood
vessel. Circulation 89, 1203-1208.
Soulie, M. L., Cros, G., Serrano, J. J., and Bali, J. P. (1992). Impairment of contractile
response to carbachol and muscarinic receptor coupling in gastric antral smooth
muscle cells isolated from diabetic streptozotocin-treated rats and db/db mice.
Molecular and Cellular Biochemistry 109, 185-188.
Soulis-Liparota, T., Cooper, M., Papazoglou, D., Clark, B., and Gerums, G. (1991).
Retardation by aminoguanidine of development of albuminuria, mesangial
expansion, and tissue fluorescence in streptozotocin-induced diabetic rat.
Diabetes 40, 1328-1334.
Soulis-Liparota, T., Cooper, M., Vranes, D., Bucala, R., and Jerums, G. (1996). Effects of
aminoguanidine in preventing experimental diabetic nephropathy are related to
the duration of treatment. Kidney International 50, 627-634.
Stadelmann, W. K., Digenis, A. G., and Tobin, G. R. (1998). Physiology and healing
dynamics of chronic cutaneous wounds. American Journal of Surgery 176
(Suppl), 26S-38S.
Stehouwer, C. D. A., Lambert, J., Donker, A. J. M., and van Hinsbergh, V. W. M. (1997).
Review: Endothelial dysfunction and pathogenesis of diabetic angiopathy.
Cardiovascular Research 34, 55-68.
Steinberg, H. O., Brechtel, G., Johnson, A., Fineberg, N., and Baron, A. D. (1994).
Insulin-mediated skeletal muscle vasodilation is nitric oxide dependent. A novel
366
action of isulin to increase nitric oxide release. Journal of Clinical Investigation
94, 1172-1179.
Stern, M. (1975). In vivo evaluation of microcirculation by coherent light scatterings.
Nature 254, 56-58.
Stevens, E. J., Carrington, A. L., and Tomlinson, D. R. (1994). Nerve ischaemia in
diabetic rats: time-course of development, effect of insulin treatment plus
comparison of streptozotocin and BB models. Diabetologia 37, 43-48.
Stevens, E. J., and Tomlinson, D. R. (1995). Effects of endothelin receptor antagonism
with bosentan on peripheral nerve function in experimental diabetes. British
Journal of Pharmacology 115, 373-379.
Stevens, E. J., Willars, G. B., Lidbury, P., House, F., and Tomlinson, D. R. (1993).
Vasoreactivity and prostacyclin release in streptozotocin-diabetic rats: effects of
insulin or aldose reductase inhibition. British Journal of Pharmacology 109, 980986.
Stevens, M. J. (1995). Nitric oxide as a potential bridge between the metabolic and
vascular hypotheses of diabetic neuropathy. Diabetic Medicine 12, 292-295.
Stohs, S. J. (1995). The role of free radicals in toxicity and disease. Journal of Basic and
Clinical Physiology and Pharmacology 6, 205-228.
Stoscheck, C. M., Nanney, L. B., and King, L. E. (1992). Quantitative determination of
EGF-R during epidermal wound healing. Journal of Investigative Dermatology
99, 645-649.
Strandell, E., Eizirik, D. L., and Sandler, S. (1989). Survival and B-cell function of
mouse pancreatic islets maintained in culture after concomitant exposure to
streptozotocin and nicotinamide. Experimental and Clinical Endocrinology 93,
219-224.
Stricker, S. (1876). Untersuchungen uber die gefasswurzeln des ischiadicus. Sitz-Ber
Kaiserl Akad Wiss 3, 173-185.
Su, E. N., Yu, D. Y., Alder, V. A., Yu, P. K., and Cringle, S. J. (1995). Altered
vasoactivity in the early diabetic eye: measured in the isolated perfused rat eye.
Experimental Eye Research 61, 699-711.
Suzuki, S., Kaneko, M., Chapman, D. C., and Dhalla, N. S. (1991). Alterations in cardiac
contractile proteins due to oxygen free radicals. Biochimica et Biophysica Acta
1074, 95-100.
Swett, J. E., Torigoe, Y., Elie, V. R., Bourassa, C. M., and Miller, P. G. (1991). Sensory
neurons of the rat sciatic nerve. Experimental Neurology 114, 82-103.
Szallasi, A., and Blumbeeg, P. M. (1996). Vanilloid receptors: new insights enhance
potential as a therapeutic target. Pain 68, 195-208.
Szolscany, J. (1984). In “Antidromic vasodilatation and neurogenic inflammation” (L. A.
Chahl, J. Szolscany, and F. Lembeck, eds.), pp. 7-25. Akademiai Kiado,
Budapest.
Szolscanyi, J., Helyyes, Z., Oroszi, G., Nemeth, J., and Pinter, E. (1988). Release of
somatostatin and its role in the mediation of the anti-inflammatory effect induced
by antidromic stimulation of sensory fibres of rat sciatic nerve. British Journal of
Pharmacology 123, 936-942.
367
Tack, C. J., Lenders, J. W., Goldstein, D. S., Lutterman, J. A., Smits, P., and Thein, T.
(1998). Haemodynamic actions of insulin. Current Opinion in Nephrology and
Hypertension 7, 99-106.
Tack, C. J., Lutterman, J. A., Vervoort, G., Thien, T., and Smits, P. (1996). Activation of
the sodium-potassium pump contributes to insulin-induced vasodilatation in
humans. Hypertension 28, 426-432.
Takahashi, K., Ghatei, M. A., Lam, H. C., O'Halloran, D. J., and Bloom, S. R. (1990).
Elevated plasma endothelin in patients with diabetes mellitus. Diabetologia 33,
306-310.
Takahashi, K., Suda, K., Lam, H. C., Ghatei, M. A., and Bloom, S. R. (1991).
Endothelin-like immunoreactivity in rat models of diabetes mellitus. Journal of
Endocrinology 130, 123-127.
Takeuchi, K., Ueshima, K., Ohuchi, T., and Okabe, S. (1994). Induction of gastric lesions
and hypoglycaemic response by food deprivation in streptozotocin-diabetic rats.
Digestive Diseases and Sciences 39, 626-634.
Tam, E. S., Furguson, D. G., Bielefeld, D. R., Lorenz, J. N., Cohen, R. M., and Pun, R.
Y. (1997). Norepinephrine-mediated calcium signalling is altered in vascular
smooth muscle of diabetic rats. Cell Calcium 21, 143-150.
Taniguchi, N., Kaneto, H., Asahi, M., Takahashi, M., Wenyi, C., Higashiyama, S., Fujii,
J., Suzuki, K., and Kayanoki, Y. (1996). Involvement of glycation and oxidative
stress in diabetic macroangiopathy. Diabetes 45 (suppl 3), S81-S83.
Tatsuki, R., Satoh, K., Yamamoto, A., Hoshi, K., and Ichihara, K. (1997). Lipid
peroxidation in the pancreas and other organs in streptozotocin-rats. Japanese
Journal of Pharmacology 75, 267-273.
Tausk, F., and Undem, B. (1995). Exogenous but not endogenous substance P release
histamine from isolated human skin fragments. Neuropeptides 29, 351-355.
Taylor, P. D., Graves, J. E., and Poston, L. (1995). Selective impairment of acetylcholinemediated endothelium-dependent relaxation in isolated resistance arteries of
streptozotocin-induced diabetic rats. Clinical Science 88, 519-524.
Teixeira, A. S., Caliari, M. V., Rocha, O. A., Machado, R. D., and Andrade, S. P. (1999).
Aminoguanidine prevents impaired healing and deficient angiogenesis in diabetic
rats. Inflammation 23, 569-581.
Teraka, K., Coppey, L. J., Davidson, E. P., Dunlap, J. A., Gutterman, D. D., and Yorek,
M. A. (1999). Acetylcholine-induced arteriolar dilation is reduced in
streptozotocin-induced diabetic rats with motor nerve dysfunction. British Journal
of Pharmacology 128, 837-843.
Terenghi, G., Chen, S. T., Carrington, A. L., Polak, J. M., and Tomlinson, D. R. (1994).
Changes in sensory neuropeptides in dorsal root ganglion and spinal cord of
spontaneously diabetic BB rats. Acta Diabetologica 31, 198-204.
Terranova, A. (1991). The effects of diabetes mellitus on wound healing. Plastic Surgical
Nursing 11, 20-25.
Tesafemariam, B., Brown, M. L., Deykin, D., and Cohen, R. A. (1990). Elevated glucose
promotes generation of endothelium-derived vasoconstrictor prostanoids in rabbit
aorta. Journal of Clinical Investigation 85, 929-932.
368
Tesfamariam, B., Brown, M. L., and Cohen, R. A. (1991). Elevated glucose impairs
endothelium-dependent relaxation by activating protein kinase C. Journal of
Clinical Investigation 87, 1643-1648.
Tesfamariam, B., and Cohen, R. A. (1992). Free radicals mediate endothelial cell
dysfunction caused by elevated glucose. American Journal of Physiology 263,
H321-H326.
Thomas, M. J. (1995). The role of free radicals and antioxidants: how do we know that
they are working? Critical Reviews in Food Sciences and Nutrition 35, 21-39.
Thornalley, P. J. (1998). Cell activation by glycated proteins. AGE receptors, receptor
recognition factors and functional classification of AGEs. Cellular and Molecular
Biology 44, 1013-1023.
Thorp, S. R., and Baynes, J. W. (1996). Role of Maillard reaction in diabetes mellitus and
diseases of aging. Drugs and Aging 9, 69-77.
Thulesen, J., Orskov, C., Holst, J. J., and Poulsen, S. S. (1997). Short-term insulin
treatment prevents the diabetogenic action of streptozotocin in rats.
Endocrinology 138, 62-68.
Tilton, R. G., Chang, K., Hasan, K. S., Smith, S. R., Petrash, M. J., Misko, T. P., Moor,
W. M., Currie, M. G., Corbet, J. R., McDaniel, M. L., and Williamson, J. R.
(1993). Prevention of diabetic vascular dysfunction by guanidines: inhibition of
nitric oxide synthase versus advanced glycation end-product formation. Diabetes
42, 221-232.
Tirupattur, P. R., Ram, J. L., Standley, P. R., and Sowers, J. R. (1993). Regulation of
Na+- K+-ATPase gene expression by insulin in vascular smooth muscle cells.
American Journal of Hypertension 6, 626-629.
Tomita, M., Mukae, S., Geshi, E., Umetsu, K., Nakatani, M., and Katagiri, T. (1996).
Mitochondrial respiratory impairment in streptozotocin-induced diabetic rat heart.
Japanese Circulation Journal 60, 673-682.
Tomlinson, D. R. (1993). Aldose reductase inhibitors and the complications of diabetes
mellitus. Diabete et Medicine 10, 214-230.
Tomlinson, D. R., and Mayer, J. H. (1984). Defects of axonal transport in diabetes
mellitus: a possible contribution to the aetiology of diabetic neuropathy. Journal
of Autonomic Pharmacology 4, 59-72.
Tomlinson, D. R., Robinson, J. P., Willars, G. B., and Keen, P. (1988). Deficient axonal
transport of substance P in streptozotocin-induced diabetic rats. Diabetes 37, 488493.
Tong, Y. C., Chin, W. T., and Cheng, J. T. (1999). Alterations in urinary bladder M2muscarinic receptor protein and mRNA in 2 weeks streptozotocin-induced
diabetic rats. Neuroscience Letters 277, 173-176.
Tong, Y. C., Hung, Y. C., Lin, S. N., and Cheng, J. T. (1996). Alterations in urinary
bladder synaptosomal neurotransmitter concentrations in two-weeks
streptozotocin-induced diabetic rats. Neuroscience Letters 15, 165-168.
Took, J. E. (1996). Microvascular function in human diabetes. A physiological
prespective. Diabetes 44, 721-726.
Traverso, N., Menini, S., Cosso, L., Odetti, P., Albano, E., Pronzato, M. A., and Marinari,
U. M. (1998). Immunological evidence for increased oxidative stress in diabetic
rats. Diabetologia 41, 265-270.
369
Trovati, M., Anfossi, G., Massucco, P., Mattielo, L., Costamagna, C., Piretto, V.,
Mularoni, E., Cavalot, F., Bosia, A., and Ghigo, D. (1997). Insulin stimulates
nitric oxide synthesis in human platelets and, through nitric oxide, increases
platelets concentrations of both guanosine-3',5'-cyclic monophosphate and
adenosine-3',5'-cyclic monophosphate. Diabetes 46, 742-749.
Trovati, M., Massucco, P., Mattiello, L., Cavalopt, F., Mularoni, E. M., Hahn, A. W., and
Anfossi, G. (1996). Studies on the influence of insulin on cyclic adenosine
monophosphate in human vascular smooth muscle cells: dependence on cyclic
guanosine monophosphate and modulation of catecolamine effects. Diabetologia
39, 1156-1164.
Tschope, C., Reinecke, A., Seidl, U., Yu, M., Gavriluk, V., Riester, U., Gohlke, P., Graf,
K., Bader, M., Hilgenfeldt, U., Pesquero, J. B., Ritz, E., and Unger, T. (1999a).
Functional, biochemical and molecular investigations of renal kallikrein-kinin
system in diabetic rats. American Journal of Physiology 277, H2333-H2340.
Tschope, C., Walther, T., Reinecke, A., Koch, M., Heringer, S. B., Pesquero, J. B.,
Bader, M., Schultheiss, H., and Unger, T. (1999b). Myocardial expression of rat
bradykinin receptors and two tissue kellikrein genes in experimental diabetes.
Immunopharmacology 44, 35-42.
Tsuboi, R., Shi, C. M., Cox, G. N., and Ogawa, H. (1995). Co-transmission of insulinlike growth factor (IGF)-1 and IGF-binding protein-1 stimulates wound healing in
animal models. Journal of Investigative Dermatology 104, 199-203.
Turk, Z., Misur, I., Turk, N., and Benko, B. (1999). Rat tissue collagen modified by
advanced glycation: correlation with duration of diabetes and glycemic control.
Clinical Chemistry and Laboratory Medicine 37, 813-820.
Umpierrez, G. E., Casals, M. M. C., and Gebhar, S. S. P. (1995). Diabetic ketoacidosis in
obese african americans. Diabetes 44, 790-795.
Ungvari, Z., Pacher, P., Kecskemeti, V., Papp, G., Szollar, L., and Koller, A. (1999).
Increased myogenic tone in skeletal muscle arterioles of diabetic rats. Possible
role for increased activity of smooth muscle Ca2+ channels and protein kinase C.
Cardiovascular Research 43, 1018-1028.
Ustundag, B., Canatan, H., Cinkilinc, N., Halifeoglu, I., and Bahcecioglu, I. H. (2000).
Angiotensin converting enzyme (ACE) activity levels in insulin-independent
diabetes mellitus and effect of ACE levels on diabetes patients with nephropathy.
Cell Biochemistry and Function 18, 23-28.
Ustundag, B., Cay, M., Naziroglu, M., Dilsiz, N., Crabbe, M. J., and Ilhan, N. (1999).
The study of renin-angiotensin-aldosterone in experimental diabetes mellitus. Cell
Biochemistry and Function 17, 193-198.
Utkan, T., Sarioglu, Y., and Yildirim, S. (1998). Impaired contraction and relaxation in
the aorta of streptozotocin-diabetic rats. Pharmacology 56, 207-215.
Vallance, P., Collier, J., and Moncada, S. (1989). Effects of endothelium-derived nitric
oxide on peripheral arteriolar tone in man. Lancet 2, 997-1000.
Vallejo, S., Angulu, J., Peiro, C., Nevado, J., Sanchez-Ferrer, A., Petidier, R., SanchezFerrer, C. F., and Rodriguez-Manas, L. (2000). Highly glycated oxyhaemoglobin
impairs nitric oxide relaxations in human mesenteric microvessels. Diabetologia
43, 83-90.
370
Vane, J. R., and Botting, R. M. (1994). Regulatory mechanisms of the vascular
endothelium: an update. Polish Journal of Pharmacology 46, 499-521.
Vasan, S., Zhang, X., Kapurniotu, A., Bernhagen, J., Teichberg, S., Basgen, J., Wagle,
D., Shih, D., Terlecky, I., Bucala, R., Cerami, A., Egan, J., and Ulrich, P. (1996).
An agent cleaving glucose-derived protein cross-links in vitro and in vivo. Nature
382, 275-278.
Velarde, V., Silver, R., Mayfield, R., and Jaffa, A. (1997). Bradykinin stimulates matrix
expression via induction of TGF-β in vascular smooth muscle cells: effects of
high glucose. Diabetes Supplement , 54A (Abs).
Ver, A., Szanto, I., Banyasz, T., Csermely, P., Vegh, E., and Somogyi, J. (1997). Changes
in the expression of Na+- K+-ATPase isoenzymes in the left ventricle of diabetic
rat hearts: effect of insulin treatment. Diabetologia 40, 1255-1262.
Verhofstad, M. H., and Hendriks, T. (1994). Diabetes impairs the development of early
strength, but not the accumulation of collagen, during intestinal anastomotic
healing in the rat. British Journal of Surgery 81, 1040-1045.
Vermes, I., Spooren, P. F., Kalsbeek-Batenburg, E. M., and Haanen, C. (1993). In
addition to von Willebrand factor and urinary albumin excretion, plasma
endothelin is an indicator of endothelial dysfunction in diabetes mellitus.
Diabetologia 36, 472-473.
Vesci, L., Mattera, G. G., Tobia, P., Corsico, N., and Calvani, M. (1995). Cardiac and
renal endothelin-1 binding sites in streptozotocin-induced diabetic rats.
Pharmacological Research 32, 363-367.
Vianna, E. O., and Garcia-Leme, J. (1995). Allergen-induced airway inflammation in
rats. Role of insulin. American Journal of Respiration and Critical Care Medicine
151, 809-814.
Vinik, A. L. (1999). Diabetic neuropathy: pathogenesis and therapy. American Journal of
Medicine 107, 17S-26S.
Vlassara, H. (1996). Advanced glycation end-products and atherosclerosis. Annals of
Medicine 28, 419-426.
Vlassara, H., Fuh, H., Makita, Z., Krungkrai, S., Cerami, A., and Bucala, R. (1992).
Exogenous advanced glycosylation end-products induce complex vascular
dysfunction in normal animals: a model for diabetic and aging complications.
Proceedings of the National Academy of Sciences of the United States of America
89, 12043-12047.
Vlassara, H., Stricker, L. J., Teichberg, S., Fuh, H., Li, Y. M., and Steffes, M. (1994).
Advanced glycation end-products induce glomerular sclerosis and albuminuria in
normal rats. Proceedings of the National Academy of Sciences of the United
States of America 91, 11704-11708.
Vo, P. A., and Reid, J. J. (1995). Reactivity to nitric oxide in the caudal artery from
diabetic rats. Endothelium 2, 287-295.
Vranes, D., Cooper, M. E., and Dilley, R. J. (1999). Cellular mechanisms of diabetic
vascular hypertrophy. Microvascular Research 57, 8-18.
Wahlestedt, C., Yanaihara, N., and Hakanson, R. (1986). Evidence for different pre- and
post-junctional receptors for neuropeptide Y and related peptides. Regulatory
Peptides 13, 307-318.
371
Wald, M., Borda, E. S., and Sterin-Borda, L. (1988). Alpha-adrenergic supersensitivity
and decreased number of alpha-adrenoreceptors in heart from acute diabetic rats.
Canadian Journal of Physiology and Pharmacology 66, 1154-1160.
Wald, M. R., Borda, E. S., and Sterin-Borda, L. (1998). Participation of nitric oxide and
cyclic GMP in the supersensitivity of acute diabetic rat myocardium by
cholinergic stimuli. Biochemical Pharmacology 55, 1991-1999.
Waldorf, H., and Fewkes, J. (1995). Wound healing. Advanced Dermatology 10, 77-97.
Walker, D., Carrington, A., Cannan, S. A., Sawicki, D., Sredy, J., Boulton, A. J., and
Malik, R. A. (1999). Structural abnormalities do not explain the early functional
abnormalities in the peripheral nerves of the streptozotocin diabetic rat. Journal of
Anatomy 195, 419-427.
Walmsley, D., and Wiles, P. G. (1991). Early loss of neurogenic inflammation in the
human diabetic foot. Clinical Science 80, 605-610.
Walsh, D. T., Weg, V. B., Williams, T. J., and Nourshargh, S. (1995). Substance Pinduced inflammatory responses in guinea pig skin: the effect of specific NK1
receptor antagonists and the role of endogenous mediators. British Journal of
Pharmacology 114, 1343-1350.
Wang, M. M., Raynaud, D., and Pace-Asciak, C. R. (1999). In vivo stimulation of 12(s)lipoxygenase in the rat skin by bradykinin and platelet activating factor: formation
of 12(s)-HETE and hepxillins, and actions on vascular permeability. Biochimica
et Biophysica Acta 36, 354-362.
Wang, Z., Wu, L., and Wang, R. (1998). Kinin B2 receptor-mediated contraction of tail
arteries from normal or streptozotocin-induced diabetic rats. British Journal of
Pharmacology 125, 143-151.
Ward, N. E., Stewart, J. R., Ioannides, C. G., and O'Brian, C. A. (2000). Oxidant-induced
S-glutathiolation inactivates protein kinase C-alpha (PKC-alpha) : a potential
mechanism of PKC isoenzyme regulation. Biochemistry 39, 10319-10329.
Waring, J. V., and Wendt, I. R. (2000). Effects of streptozotocin-induced diabetes
mellitus on intracellular calcium and contraction of longitudinal smooth muscle
from rat urinary bladder. Journal of Urology 163, 323-330.
Wautier, J. L., and Guillausseau, P. J. (1998). Diabetes, advanced glycation end-products
and vascular disease. Vascular Medicine 3, 131-137.
Wautier, J. L., and Wautier, M. P. (1999). Erythrocyte adhesion to the vascular
endothelium. Transfusion Clinique et Biologique 6, 397-402.
Wautier, J. L., Wautier, M. P., Schmidt, A. M., Anderson, M. G., Hori, O., Zoukourian,
C., Capron, L., Chappey, O., Yan, S. D., Brett, J., Guillausseau, P. J., and Stern,
D. (1994). Advanced glycation end-products (AGEs) on the surface of diabetic
erythrocytes bind to the vessel wall via a specific receptor inducing oxidant stress
in the vasculature: a link between surface-associated AGEs and diabetic
complications. Proceedings of the National Academy of Sciences of the United
States of America 91, 7742-7746.
Wautier, J. L., Zoukourian, C., Chappey, O., Wautier, M. P., Guillausseau, P. J., Cao, R.,
Hori, O., Stern, D., and Schmidt, A. M. (1996). Receptor-mediated endothelial
cell dysfunction in diabetic vasculopathy. Soluble receptor for advanced glycation
end-products blocks hyperpermeability in diabetic rats. Journal of Clinical
Investigation 97, 238-243.
372
Weisbrod, R. M., Brown, M. L., and Cohen, R. A. (1993). Effect of elevated glucose on
cyclic GMP and eicosaniods produced by porcine aortic endothelium.
Arteriosclerosis and Thrombosis 13, 915-923.
Weiss, R. B. (1982). Streptozotocin: a review of its pharmacology, efficacy and toxicity.
Cancer Treatment Reports 66, 427-438.
Wells, K. E., Miguel, R., and Alexander, J. J. (1997). Transmembrane calcium flux
regulates LDL oxidation by arterial smooth muscle cells. Journal of Surgical
Research 67, 126-131.
Welsh, B., and Wecker, L. (1991). Effects of streptozotocin-induced diabetes on
acetylcholine metabolism in rat brain. Neurochemical Research 16, 453-460.
Wertheimer, E., Taylor, S. I., and Tennenbaum, T. (1998). Insulin receptor regulation of
cell surface integrins: a possible mechanism contributing to the development of
diabetic complications. Proceedings of the Association of American Physicians
110, 333-339.
Westerman, R. A., Delaney, C., Phillips, A., Horowitz, M., and Roberts, A. (1989).
Concordance between different measures of small sensory and autonomic fibre
neuropathy in diabetes mellitus. Clinical and Experimental Neurology 26, 51-63.
White, D. M., and Helme, R. D. (1985). Release of substance P from peripheral nerve
terminals following electrical stimulation of the sciatic nerve. Brain Research
336, 27-31.
White, R. E., and Carrier, G. O. (1990). Vascular contraction induced by activation of
membrane calcium ion channels is enhanced in streptozotocin-diabetes. Journal
of Pharmacology and Experimental Therapeutics 253, 1057-1062.
Willars, G. B., Calcutt, N. A., Compton, A. M., and Tomlinson, D. R. (1989). Substance
P in peripheral nerve, skin, atrial myocardium and gastrointestinal tract of rats
with long-term diabetes mellitus. Journal of the Neurological Sciences 91, 153164.
Willerson, J. T. (1995). Conversion from chronic to acute coronary heart disease
syndromes. Role of platelets and platelet products. Texas Heart Institute Journal
22 (1), 13-19.
Williams, B., Tsai, P., and Schrier, R. W. (1992). Glucose-induced down-regulation of
angiotensin II and arginine vasopressin receptors in cultured rat aortic vascular
smooth muscle cells. Role of protein kinase C. Journal of Clinical Investigation
90, 1992-1999.
Williams, S. K., Howarth, N. L., Devenvyj, J., and Bitensky, M. W. (1982). Structural
and functional consequences of increased tubulin glycosylation in diabetes
mellitus. Proceedings of the National Academy of Sciences of the United States of
America 79, 6546-6550.
Wolf, B. A., Williamson, J. R., Easom, R. A., Chang, K., Sherman, W. R., and Turk, J.
(1991). Diacylglycerol accumulation and microvascular abnormalities induced by
elevated glucose levels. Journal of Clinical Investigation 87, 31-38.
Wolffenbuttel, B. H., Boulanger, C. M., Crijns, F. R., Huijberts, M. S., Poitevin, P.,
Swennen, G. N., Vasan, S., Egan, J. J., Ulrich, P., Cerami, A., and Levy, B. I.
(1998). Breakers of advanced glycation end-products restore large artery
properties in experimental diabetes. Proceedings of the National Academy of
Sciences of the United States of America 95, 4630-4634.
373
Wong, K. K. (1996). A significant increase of aortic response to a combination of
norepinephrine and KCl in aorta isolated from 2 month diabetic rats induced by
streptozotocin. Artery 22, 164-171.
Wong, K. K., and Tzeng, S. F. (1992). Norepinephrine-induced contractile responses in
isolated rat aortae from different duration of diabetes. Artery 19, 1-13.
Wood, E. J., and Bladon, P. T. (1985). "The human skin". Edward Arnolds (Publishers)
Ltd, London.
Wu, S. Q., and Tang, F. (1998). Impaired paracrine effect of ET-1 on vascular smooth
muscle in streptozotocin-diabetic rats. Cardiovascular Research 39, 651-656.
Xie, Q. W., Cho, H. J., Calaycay, J., Numford, R. A., Swiderek, K. M., Lee, T. D., Ding,
A., Troso, T., and Nathan, C. (1992). Cloning and characterization of inducible
nitric oxide synthase from mouse macrophages. Science 256, 225-228.
Yamagishi, S., Fujimori, H., Yonekura, H., Yamamoto, Y., and Yamamoto, H. (1998).
Advanced glycation end-products inhibit prostacyclin production and induce
plasminogen activator inhibitor-1 in human microvascular endothelial cells.
Diabetologia 41, 1435-1441.
Yamamoto, H., Uchigata, Y., and Okamoto, H. (1981). Streptozotocin and alloxan induce
DNA strand-breaks and poly (ADP-Rribose) synthetase in pancreatic islets.
Nature 294, 284-286.
Yamauchi, T., Ohnaka, K., Takayanagi, R., Umeda, F., and Nawata, H. (1990). Enhanced
secretion of endothelin-1 by elevated glucose levels from cultured bovine aortic
endothelial cells. FEBS Letters 267, 16-18.
Yan, S. D., Schmidt, A. M., Anderson, M. G., Zhang, J., Brett, J., Zou, Y. S., Pinsky, D.,
and Stern, D. (1994). Enhanced cellular oxidant stress by the interaction of
advanced glycation end-products with their receptors / binding proteins. Journal
of Biological Chemistry 269, 9889-9897.
Yanagisawa, M., Inoue, A., Takuwa, Y., Mitsui, Y., Kobayashi, M., and Masaki, T.
(1989). The human preproendothelin-1 gene: possible regulation by endothelial
phosphoinositide turnover signalling. Journal of Cardiovascular Pharmacology
13 (Suppl 5), S13-S17.
Yanagisawa, M., Kurihara, H., Kimura, S., Tomobe, Y., Kobayashi, M., Mitsui, Y.,
Yazakin, Y., Goto, K., and Masaki, T. (1988). A novel potent vasoconstrictor
peptide produced by vascular endothelial cells. Nature 332, 411-415.
Yanagisawa, M., and Masaki, T. (1989). Endothelin: a novel endothelium-derived
peptide. Biochemical Pharmacology 38, 1877-1883.
Yki-Jarvinen, H., and Koivisto, V. A. (1986). Natural course of insulin resistance in type
I diabetes. The New England Journal of Medicine 315, 224-230.
Yokokawa, K., Tahara, H., and Kohno, M. (1991). Hypertension associated with
endothelin-secreting malignant haemangioendotheioma. Annals of Internal
Medicine 114, 213-215.
Yosipovitch, G., Schneiderman, J., Van Dyk, D. J., Chetrit, A., Milo, G., and Boner, G.
(1996). Impairment of the postural venoarteriolar response in young type 1
diabetic patients. A study by laser Doppler flowmetry. Angiology 47, 687-691.
Yu, P. H., and Zuo, D. M. (1997). Aminoguanidine inhibits semicarbazide-sensitive
amine oxidase activity: implications for advanced glycation and diabetic
complications. Diabetologia 40, 1243-1250.
374
Yu, Z., Quamme, G. A., and McNeill, J. H. (1994). Depressed [Ca2+]i response to
isoproterenol and cAMP in isolated cardiomyocytes from experimental diabetic
rats. American Journal of Physiology 266, H2334-H2342.
Zembowics, A., Hatchett, R., Jakubowski, A. M., and Gryglewski, R. J. (1993).
Involvement of nitric oxide in the endothelium-dependent relaxation induced by
hydrogen peroxide in the rabbit aorta. British Journal of Pharmacology 110, 151158.
Zemel, M. B. (1995). Insulin resistance vs. hyperinsulinemia in hypertension: insulin
regulation of Ca2+-transport and Ca2+-regulation of insulin sensitivity. Journal
of Nutrition 125 (Suppl 6), 1738S-1743S.
Zenere, B. M., Arcaro, G., Saggiani, F., Rossi, L., Muggeo, M., and Lechi, A. (1995).
Noninvasive detection of functional alterations of the arterial wall in IDDM
patients with and without microalbuminuria. Diabetes Care 18, 975-982.
Zhang, X., Ji, R. R., Nilsson, S., Villar, M., Ubink, R., Ju, G., Wiesenfeild-Hallin, Z., and
Hokfelt, T. (1995). Neuropeptide Y and galanin binding sites in rat and monkey
lumbar dorsal root ganglia and spinal cord and effect of peripheral axotomy.
European Journal of Neuroscience 7, 367-380.
Zhao, Q. E., Mihara, T., Sugimoto, Y., and Kamei, C. (1996). Mechanism of bradykinininduced histamine release from rat peritoneal mast cells. Biological and
Pharmaceutical Bulletin 19, 237-240.
Zhu, B. H., Guan, Y. Y., and Lin, M. J. (1999). Erigeron breviscapus prevents defective
endothelium-dependent relaxations in diabetic rat aorta. Life Sciences 65, 15531559.
Zhuang, H. X., Snyder, C. K., Pu, S. F., and Ishii, D. N. (1996). Insulin-like growth
factors reverse or arrest diabetic neuropathy: effects of hyperalgesia and impaired
nerve regeneration in rats. Experimental Neurology 140, 198-205.
Ziche, M., Morbidelli, L., Pacini, M., Geppetti, P., Alessandri, G., and Maggi, C. A.
(1990). Substance P stimulates neovascularization in vivo and proliferation of
cultured endothelial cells. Microvascular Research 40, 264-278.
Ziogas, J., and Story, D. F. (1991). Angiotensin II generation in the rat vena cava:
stimulation of local synthesis by β-adrenoreceptor activation. NaunynSchmiedebergs Archives of Pharmacology 343, 31-36.
Zochodne, D., Cheng, C., and Sun, H. (1996). Diabetes increases sciatic nerve
susceptibility to endothelin-induced ischemia. Diabetes 45, 627-632.
Zochodne, D. W., and Ho, L. T. (1993). Diabetes mellitus prevents capsaicin from
inducing hyperaemia in the rat sciatic nerve. Diabetologia 36, 493-496.
Zuccollo, A., Navarro, M., and Catanzaro, O. (1996). Effects of B1 and B2 kinin receptor
antagonists in diabetic mice. Canadian Journal of Physiology and Pharmacology
74, 586-589.
375
APPENDICES
376
Appendix A:
Responses in control rats over time (Chapter 3, part 1)
Control (week zero) = 4-6 weeks old rats
Control (week 1) = 5-7 weeks old rats
Control (week 5) = 9-11 weeks old rats
Control (week 10) = 14-16 weeks old rats
Control
rats
Week 0
Week 1
Week 5
Week 10
BBF
95% CI
SNP
95% CI
SP
95% CI
CGRP
95% CI
ES
95% CI
(25.1 ± 0.9 cm2)
(23.3-26.9)
(26.7 ± 2.2 cm2)
(22.4-31)
(23.2 ± 1.4 cm2)
(20.4-25.9)
(23.3 ± 1.8 cm2)
(19.8-26.8)
(26.1 ± 1.8 cm2)
(22.6-29.6)
(28.4 ± 4.2 cm2)
(20.2-36.6)
(29.2 ± 3.5 cm2)
(22.3-36.1)
(23.6 ± 4 cm2)
(15.7-31.4)
(51.4 ± 2.9 cm2)
(45.7-57.1)
(50.1 ± 3.7 cm2)
(42.8-57.3)
(50.9 ± 2.5 cm2)
(46-55.8)
(52.3 ± 2.4 cm2)
(47.6-57)
(24.8 ± 4.9 cm2)
(15.2-34.4)
(25.1 ± 4.8 cm2)
(15.7-34.5)
(24.4 ± 4.6 cm2)
(15.4-33.4)
(23.3 ± 2.1 cm2)
(19.2-27.4)
(17.6 ± 1.3 cm2)
(15-20.1)
( 17.7 ± 1.6 cm2)
(14.6-20.8)
( 17.5 ± 2.4 cm2)
(12.8-22.2)
(16.4 ± 0.4 cm2)
(15.6-17.2)
Table A-1: Basal blood flow (BBF) and vascular responses to SNP (100μM), SP, CGRP (both 1μM)
and ES (20V, 15Hz, 2 mSec, for 1 min) in control rats over time (n = 6-9). Data are
presented as mean ± SEM and their 95% confidence intervals (95% CI).
Appendix B:
Plasma extravasation results
B1:
Chapter 3, part 1
Rat
Control
Diabetic
(2 weeks)
Diabetic
(4 weeks)
Diabetic
(6 weeks)
Diabetic
(9 weeks)
R1 (μg/ml)
SP2 (μg/ml)
(65 ± 12.9)
(86.7 ± 13.1)
(95%CI 40-90)
(95%CI 61-112)
(106 ± 12.5)
(133 ± 13.6)
(95%CI 81.5-130.5) (95%CI 106-160)
(153 ± 17.7)
(166.6 ± 36)
(95% CI 118-188)
(95%CI 96-237)
(150 ± 23.6)
(186.3 ± 9.6)
(95%CI 104-196) (95%CI 167.5-205)
(159.6 ± 5)
(170.5 ± 6.2)
(95%CI 150-169)
(95%CI 158-183)
SP3 (μg/ml)
SP4 (μg/ml)
(111.7 ± 12)
(95%CI 100-135)
(169.7 ± 21.3)
(95%CI 128-211)
(177.4 ± 49.7)
(95%CI 80-275)
(192 ± 13.4)
(95%CI 166-218)
(186 ± 15)
(95%CI 157-215)
(123 ± 9.2)
(95%CI 105-134)
(158.4 ± 21.4)
(95%CI 116.5-200)
(152.4 ± 35.6)
(95%CI 83-222)
(173.1 ± 21.2)
(95%CI 131.5-215)
(184.3 ± 6.2)
(95%CI 172-196)
Table B-1: PE in diabetic rats over time (n = 6-9). Data are presented as mean ± SEM and their 95%
confidence intervals (95% CI).
377
B2:
Chapter 4
Animals
SP (C)
SP (D)
SP (C) +
L-NAME
SP (D) +
L-NAME
BK (C)
BK (D)
BK (C) +
L-NAME
BK (D) +
L-NAME
Muscarine
chloride (C)
Muscarine
chloride (D)
Nicotine (C)
Nicotine (D)
R1 (μg/ml)
SP2 (μg/ml)
SP3 (μg/ml)
SP4 (μg/ml)
(65 ± 12.9)
(95%CI 40-90)
(153 ± 17.7)
(95%CI 118-188)
(63 ± 8.2)
(95%CI )
(141 ± 14.2)
(95%CI )
(58.3 ± 6.2)
(95%CI 46-70)
(156.5 ± 12.1)
(95%CI 133-180)
(53 ± 9.2)
(95%CI 35-71)
(159 ± 16.7 )
(95%CI 126-192)
(73 ± 15)
(95%CI 44-102)
(158.3 ± 17.2)
(95%CI 125-192)
(73.9 ± 2.7)
(95%CI 69-79)
(166.7 ± 16.9)
(95%CI 133-200)
(86.7 ± 13.1)
(95%CI 61-112)
(166.6 ± 36.1)
(95%CI 96-237)
(69.5 ± 8.1)
(95%CI )
(156 ± 14.1)
(95%CI )
(79 ± 4.6)
(95%CI 70-88)
(161 ± 8.4)
(95%CI 145-177)
(60 ± 6.9)
(95%CI 47-73.5)
(169 ± 11.3)
(95%CI 147-191)
(90.3 ± 11.4)
(95%CI 68-113)
(170 ± 14.7)
(95%CI 141-199)
(89.2 ± 6.9)
(95%CI 76-103)
(173.3 ± 15.5)
(95%CI 143-204)
(111.7 ± 12)
(95%CI 100-135)
(177.4 ± 49.7)
(95%CI 80-275)
(66.7 ± 7.2)
(95%CI )
(167 ± 13.6)
(95%CI )
(109 ± 7.5)
(95%CI 94-124)
(174.8 ± 9.2)
(95%CI 157-193)
(64 ± 10.2)
(95%CI 44-84)
(170.4 ± 9.7)
(95%CI 151-189)
(121.8 ± 11.1)
(95%CI 100-144)
(182.7 ± 20)
(95%CI 143-222)
(112.4 ± 12.7)
(95%CI 87-137)
(182.6 ± 12.8)
(95%CI 157-208)
(123 ± 9.2)
(95%CI 105-134)
(152.4 ± 35.6)
(95%CI 83-222)
(58.1 ± 2.5)
(95%CI )
(143 ± 14.4)
(95%CI )
(125 ± 17)
(95%CI 92-158)
(167.2 ± 8.4)
(95%CI 151-184)
(62 ± 10)
(95%CI 42-82)
(161 ± 12.4)
(95%CI 137-185)
(135.1 ± 16.9)
(95%CI 102-168)
(154.2 ± 16.2)
(95%CI 122-186)
(121.5 ± 14.7)
(95%CI 93-150)
(150.4 ± 16.3)
(95%CI 118-182)
Table B-2: PE in control and diabetic rats after {SP (1μM), SP (1μM) + L-NAME (100μM)}, control
and diabetic rats after {BK (10μM), BK (10μM) + L-NAME (100μM}, control rats after
Nicotine (100μM) or Muscarine chloride (100μM), and diabetic rats after Nicotine (100μM)
or Muscarine chloride (100μM) (n = 6-9). Data are presented as mean ± SEM and their95%
confidence intervals (95% CI).
C=Control
D= Diabetic
B3:
Chapter 5
Animal
Control (N)
Control (T)
Diabetic (N)
Diabetic (T)
R1 (μg/ml)
SP2 (μg/ml)
SP3 (μg/ml)
SP4 (μg/ml)
(65 ± 12.9)
(95%CI 40-90)
(77 ± 3.8)
(95%CI 70-84)
(153 ± 17.7)
(95% CI 118-188)
(154.7 ± 33.9)
(95% CI 88-221)
(86.7 ± 13.1)
(95%CI 61-112)
(99.5 ± 8)
(95%CI 84-115)
(166.6 ± 36.1)
(95%CI 96-237)
(197.4 ± 29.9)
(95%CI 139-256)
(111.7 ± 12)
(95%CI 100-135)
(111.2 ± 9.7)
(95%CI 92-130)
(177.4 ± 49.7)
(95%CI 80-275)
(215.7± 25.9)
(95%CI 165-266.5)
(123 ± 9.2)
(95%CI 105-134)
(131 ± 20.2)
(95%CI 91-171)
(152.4 ± 35.6)
(95%CI 83-222)
(190.9 ± 10.1)
(95%CI 171-211)
Table B-3: PE in control and diabetic rats before and after guanethidine treatment (50mg/kg, i.p)
(n = 4-10). Data are presented as mean ± SEM and their 95% confidence intervals (95% CI).
N = Guanethidine non-treated
T = Guanethidine-treated
378
B4:
Chapter 6
Animal
Control
Control + BQ-123
Control + BQ-788
Diabetic
Diab+BQ-123
Diab+BQ-788
Diab+BIM
R1 (μg/ml)
SP2 (μg/ml)
SP3 (μg/ml)
SP4 (μg/ml)
(65 ± 12.9)
(95%CI 40-90)
( 70 ± 9.7)
(95%CI 51-89)
(66 ± 17.4)
(95%CI 32-100)
(153 ± 17.7)
(95%CI 118-188)
(50.5 ± 7.4)
(95%CI 36-65)
(110 ± 24)
(95%CI 63-157)
(153.9 ± 22)
(95%CI 111-197)
(86.7 ± 13.1)
(95%CI 61-112)
(84.4 ± 8.9)
(95%CI 66.9-102)
(98 ± 11.4)
(95%CI 75.6-120)
(166.6 ± 36.1)
(95%CI 96-237)
(72.2 ± 7.4)
(95%CI 58-87)
(140.7 ± 13.7)
(95%CI 114-168)
(161.5 ± 21)
(95%CI 120-203)
(111.7 ± 12)
(95%CI 100-135)
(106.3 ± 18.9)
(95%CI 69.2-143)
(102.6 ± 11.6)
(95%CI 80-125)
(177.4 ± 49.7)
(95%CI 80-275)
(88.4 ± 6.8)
(95%CI 75-102)
(191.7 ± 27.2)
(95%CI 138-245)
(196 ± 26)
(95%CI 145-247)
(123 ± 9.2)
(95%CI 105-134)
( 128.5 ± 24.5)
(95%CI 80-176.5)
(131.7 ± 14.2)
(95%CI 104-159)
(152.4 ± 35.6)
(95%CI 83-222)
(83.7 ± 5.7)
(95%CI 72.5-95)
(212.2 ± 30)
(95%CI 153-271)
(207 ± 21.9)
(95%CI 164-249)
Table B-4: PE in control, diabetic (Diab,) control and diabetic rats after perfusing BQ-123, BQ-788
(both at 10μM), and diabetic rats after perfusing BIM (1μM)(n = 4-10). Data are presented
as mean ± SEM and their 95% confidence intervals (95% CI).
B5:
Chapter 7
Animal
Control
Diabetic
Diab+SOD
Diab+NAC
Diab+BQ-123
Diab+BQ-123
+SOD
Diab+Freedox
R1 (μg/ml)
(65 ± 12.9)
(95%CI 40-90)
(153 ± 17.7)
(95%CI 118-188)
(85 ± 7.8)
(95%CI 70–100)
(107.6 ± 17.2)
(95%CI 74-141)
(50.5 ± 7.4)
(95%CI 36-65)
(74.8 ± 11.1 )
(95%CI 53-97)
(108.6 ± 14.1 )
(95%CI 81-136)
SP2 (μg/ml)
(86.7 ± 13.1)
(95%CI 61-112)
(166.6 ± 36.1)
(95%CI 96-237)
(94.4 ± 5)
(95%CI 85-104)
(118.2 ± 17.9)
(95%CI 83-153)
(72.2 ± 7.4)
(95%CI 58-87)
(91.9 ± 22.7 )
(95%CI 47-137)
(129.8 ± 14.2 )
(95%CI 102-158)
SP3 (μg/ml)
(111.7 ± 12)
(95%CI 100-135)
(177.4 ± 49.7)
(95%CI 80-275)
(115.3 ± 4)
(95%CI 107.5-123)
(160.7 ± 12.1)
(95%CI 137-184)
(88.4 ± 6.8)
(95%CI 75-102)
(130.4 ± 18.8 )
(95%CI 94-167)
(162.8 ± 8.6 )
(95%CI 146-180)
SP4 (μg/ml)
(123 ± 9.2)
(95%CI 105-134)
(152.4 ± 35.6)
(95%CI 83-222)
(119.3 ± 2.7)
(95% CI 114-125)
(165.8 ± 5.2)
(95%CI 156-176)
(83.7 ± 5.7)
(95%CI 72.5-95)
(142 ± 24 )
(95%CI 95-189)
(172.2 ± 7 )
(95%CI 158.5-186)
Table B-5: PE in control, diabetic (Diab), diabetic rats after perfucing SOD (100U/ml), NAC
(100μM),BQ-123 (10μM), combined [BQ-123 (10μM) + SOD (100U/ml)], and diabetic rats
treated with Freedox (Tirilazad mesylate) (5mg/kg/i.p)(n = 4-10). Data are presented as mean
± SEM and their 95% confidence intervals (95% CI).
379
B6:
Chapter 8
Animals
Control
Diabetic
Diab +
amadori
Diab +
AGEs
Diab +
anti-RAGE
Diab +
AG (2W)
Diab +
AG (4W)
R1 (μg/ml)
SP2 (μg/ml)
SP3 (μg/ml)
SP4 (μg/ml)
(65 ± 12.9)
(95%CI 40-90)
(153 ± 17.7)
(95%CI 118-188)
(124 ± 2.3)
(95%CI 119-128)
(161.8 ± 19.2)
(95%CI 124-199)
(131.6 ± 22.2)
(95%CI 88-175)
(110.2 ± 12.9)
(95%CI 85-135)
(92.5 ± 17.9)
(95%CI 57-128)
(86.7 ± 13.1)
(95%CI 61-112)
(166.6 ± 36.1)
(95%CI 96-237)
(162 ± 20.9)
(95%CI 121-203)
(182.3 ± 15.7)
(95%CI 151-213)
(216.7 ± 36.3)
(95%CI 146-288)
(115.9 ± 17.3)
(95%CI 82-150)
(140.7 ± 20.6)
(95%CI 100-181)
(111.7 ± 12)
(95%CI 100-135)
(177.4 ± 49.7)
(95%CI 80-275)
(179.6 ± 18.6)
(95%CI 143-216)
(194.6 ± 14.8 )
(95%CI 166-224)
(261 ± 36)
(95%CI 190-332)
(145.8 ± 19.7)
(95%CI 107-184)
(172.3 ± 14.3)
(95%CI 144-200)
(123 ± 9.2)
(95%CI 105-134)
(152.4 ± 35.6)
(95%CI 83-222)
(184 ± 25)
(95%CI 135-233)
(168.2 ± 23.8)
(95%CI 122-215)
(245 ± 19.9)
(95%CI 206-284)
(156 ± 23.9)
(95%CI 109-203)
(168.2 ± 15.6)
(95%CI 138-199)
Table B-6: PE in control, diabetic (Diab), diabetic rats after perfusing amadori (1mg/ml), AGE
(10mg/ml), anti-RAGE (100μg/ml), and diabetic rats treated with AG (25mg/kg, i.p) (2 or 4
weeks). Data are presented as mean ± SEM and their 95% confidence intervals (95% CI).
B7:
Chapter 9
Animals
Control
Diab (2 W)
Diab (4 W)
Diab (4 W) +
late insulin
Intervention
Diab (4 W) +
early insulin
intervention
R1 (μg/ml)
SP2 (μg/ml)
SP3 (μg/ml)
SP4 (μg/ml)
(65 ± 12.9)
(95%CI 40-90)
(106.4 ± 12.5)
(95%CI 82-131)
(153 ± 17.7)
(95%CI 118-188)
(125 ± 5)
(95%CI 115-135)
(86.7 ± 13.1)
(95%CI 61-112)
(133 ± 13.6)
(95%CI 106-160)
(166.6 ± 36.1)
(95%CI 96-237)
(154.2 ± 18.5)
(95%CI 118-190)
(111.7 ± 12)
(95%CI 100-135)
(169.7 ± 21.3)
(95%CI 128-211)
(177.4 ± 49.7)
(95%CI 80-275)
(158.5 ± 18)
(95%CI 123-194)
(123 ± 9.2)
(95%CI 105-134)
(158.4 ± 21.4)
(95%CI 116-200)
(152.4 ± 35.6)
(95%CI 83-222)
(176 ± 12.6)
(95%CI 151-201)
(77.4 ± 10.4)
(95%CI 57-98)
(119 ± 5)
(95%CI 109-129)
(157.1 ± 12.2)
(95%CI 133-181)
(168.3 ± 14.1)
(95%CI 141-196)
Table B-7: PE in control, diabetic (Diab) (2 &4 weeks), and diabetic rats (4 weeks) treated with insulin
(2-6IU/day) (late and early interventions) (n = 4-10). Data are presented as mean ± SEM
and their 95% confidence intervals (95% CI).
380
Appendix C:
Commercial Amadori and AGEs (Produced by Peptech, Australia)
C1: The production of amadori and AGEs
For making AGEs, glucose-6-phosphate (G-6-P) and a five week incubation time were used,
and for amadori, D-glucose and only one week incubation were used, otherwise they were
made by the same protocol provided by Peptech as followed for both G-6-P and glucose
respectively.
The method was carried out in a sterile environment. Glucose was filter-sterilized using a
0.2μm filter. Working in a tissue cultured hood, the weighed out bovine serum albumin
(BSA) powder was dispensed gradually into a sterile beaker of 200ml sterile phosphate
buffer saline (PBS) and stirred into solution. 200μl trasylol, 800μl phenylmethylsulfonyl
fluoride (PMSF), 4ml ethylenediaminetetraacetic acid (EDTA), and 200mg ampicillin was
added and sterile filtered with 0.2μm filter. 5ml of filtered 2M glucose was dispensed into
three 50ml tubes (set up in triplicates) and subsequently added up to 50 ml with the rest
of the filtered ingredients. A fourth 50ml tube was prepared without any glucose, i.e.
BSA only. Samples were incubated in a 50°C waterbath for required time according to G6-P or D-glucose respectively. 500μl samples were taken every few days to establish the
time course of amadori and AGEs formation. Samples were frozen and stored at -20°C.
C2: The purification and analysis
The samples were dialyzed against distilled water and subsequently dialyzed against
100mM glycine pH 9.2. Dialysis removes free G-6-P, glucose, and schiff bases. The
500μl samples were tested in a periodate assay [Ahmed, 1991 #627] and the glycine
dialyzed samples were run on a phenyl boronate agarose (PBA) column in order to purify
amadori. The periodate assay was used to identify whether there were any amadori
products present.
381
The PBA affinity chromatography purification of amadori products is based on the
amadori products’ co-planar cis-hydroxyl groups which enables them to bind strongly to
the column. BSA and AGEs lack this structure, and therefore do not bind to the column.
The samples were applied directly after dialysis into the PBA column equiliberated in
glycine buffer. The samples were washed through with glycine buffer at a flow rate of 1
ml/min. Non-glycosylated BSA and AGEs were in the flow through fraction. The
amadori were eluted by lowering the pH, making PBA unable to bind the cis-diol groups
of the amadori. This was done by pumping through tri-sodium citrate or citrate buffer
(100mM, pH 5.8-6).The BSA alone-fraction and the amadori were then dialyzed
extensively in 10mM tri-sodiaum citrate. The amdori were then further analyzed with
PBS to remove any tri-sodium citrate and then quantified for amadori and total protein
concentration using a periodate and PBA protein assay respectively.
A one week incubation at 50°C of D-glucose and BSA gave amadori products which
were purified by column chromatography. The G-6-P counterpart showed no evidence of
the presence of amadori products as found in the periodate assay. The failure to detect
amadori during time course experiment and after enrichment procedures on PBA
columns, strongly suggests that amadori is not a long lived intermediate of nonenzymatic glycosylation when G-6-P is the reducing sugar.
382
Appendix D:
Insul i n tr eated
r ats (4 w ee ks)
Insul i n-tr eate d
rats (2 w e eks)
Urine Glucose Level
(mmol/lit)
120
105
90
75
60
45
30
15
0
0
3
6
9 12 15 18 21 24 27 30
Days afte r induction of diabe te s
The average urine glucose level in insulin-treated (2 and 4 weeks) diabetic rats over time.
Urine glucose levels indicated by Diastix strips:
Negative
Trace (5mmol/lit)
+
15 (15mmol/lit)
++
30 (30 mmol/lit)
+++
60 (60 mmol/lit)
++++ 110 (110 mmol/lit)
383