The Role of Glyceraldehyde-3-Phosphate Dehydrogenase
Acetylation in Liver Metabolism
By
Simon T Bond (B.Sc, Hons)
Submitted in fulfilment of the requirements for the degree of
Doctor of Philosophy
Deakin University
September, 2015
DEAKIN UNIVERSITY
ACCESS TO THESIS - A
I am the author of the thesis entitled
The Role of Glyceraldehyde-3-Phosphate Dehydrogenase Acetylation in
Liver Metabolism
submitted for the degree of
Doctor of Philosophy (Medicine)
This thesis may be made available for consultation, loan and limited copying
in accordance with the Copyright Act 1968.
'I certify that I am the student named below and that the information provided in the form is
correct'
Full Name: Simon Timothy Bond
Signed:
Date: 22-02-2016
DEAKIN UNIVERSITY
CANDIDATE DECLARATION
I certify the following about the thesis entitled Glyceraldehyde-3-Phosphate
Dehydrogenase Lysine Residues Regulate Gluconeogenesis
submitted for the degree of Doctor of Philosophy
a.
I am the creator of all or part of the whole work(s) (including content and
layout) and that where reference is made to the work of others, due
acknowledgment is given.
b.
The work(s) are not in any way a violation or infringement of any copyright,
trademark, patent, or other rights whatsoever of any person.
c.
That if the work(s) have been commissioned, sponsored or supported by any
organisation, I have fulfilled all of the obligations required by such contract or
agreement.
I also certify that any material in the thesis which has been accepted for a degree or
diploma by any university or institution is identified in the text.
'I certify that I am the student named below and that the information provided in the form is
correct'
Full Name: Simon Timothy Bond
Signed:
Date: 23/02/2016
ACKNOWLEDGEMENTS
To Assoc Prof Sean McGee for his fantastic supervision and direction throughout my
time at Deakin University, and for providing an excellent work environment.
To Prof Ken Walder and Dr Kirsten Howlett for their support, supervision, and
assistance throughout my candidature.
To Dr Shaun Mason, Dr Greg Kowalski, and Dr Clinton Bruce for their work on
fractional contribution to endogenous glucose and gluconeogenesis.
To Dr Victor Streltsov for his assistance with generating a computational structure of
GAPDH.
To my colleagues at Deakin University; Prof Andy Sinclair, Dr Kathryn AstonMourney, Dr Daniel Fraher, Dr Amanda Genders, Dr Lyndal Bayles, Dr Vidhi Gaur,
Dr Juliane Czeczor, Dr Yann Gibert, Sheree Martin, Courtney Swinton, Shona
Morrison, Tim Connor, Kirstie De Jong, Luke Amor, and Adrian Cooper. Thanks for
all the help and laughs along the way.
To my family and friends for their unconditional support.
I
PUBLICATIONS
Published
T.J.Parmenter, M.Kleinschmidt, K.M.Kinross, S.T.Bond, J.Li, M.R.Kaadige, A.Rao,
K.E.Sheppard, W.Hugo, G.M.Pupo, R.B.Pearson, S.L.McGee, G.V.Long,
R.A.Scolyer, H.Rizos, R.S.Lo, C.Cullinane, D.E.Ayer, A.Ribas, R.W.Johnstone,
R.J.Hicks and G.A.McArthur. Response of BRAF-Mutant Melanoma to BRAF
Inhibition Is Mediated by a Network of Transcriptional Regulators of Glycolysis.
Cancer Discovery, 2014; 4(4):423-33. (I.F = 19.453).
In Preparation
S.T.Bond, V.Streltsov, K.R.Walder, K.F.Howlett, S.L.McGee. Mutation of GAPDH
lysine residues to arginine alters glucose metabolism in hepatocytes.
S.T.Bond, G.Kowalski, C.Bruce, K.R.Walder, K.F.Howlett, S.L.McGee. GAPDH
deacetylation in liver affects glucose homeostasis in the mouse.
V.Gaur, T.Connor, A.Sanigorski, S.T.Bond, K.McEwan, C.R.Bruce,
D.C.Henstridge, S.D.Martin, L.Kerr-Bayles, C.Fleming, M.Wu, L.S.Pike-Winer,
D.Chen, K.Baar, M.A.Febbraio, P.Gregorevic, F.M.Pfeffer, K.R.Walder,
M.Hargreaves and S.L.McGee. Disruption of an HDAC-repressor complex induces
exercise-like adaptations in skeletal muscle.
II
Abstracts:
S.T.Bond, K.R.Walder, K.F.Howlett, and S.L.McGee. Regulation of glucose
homeostasis by reversible lysine acetylation of GAPDH. Abstracts of the Annual
Scientific Meeting of the Australian Diabetes Society, Sydney, 2013. Abstract 131.
S.T.Bond, K.R.Walder, K.F.Howlett, and S.L.McGee. Regulation of glucose
homeostasis by reversible lysine acetylation of GAPDH. Abstracts of the World
Diabetes Congress of the International Diabetes Federation, Melbourne, 2013.
Abstract PD-0655.
S.T.Bond, K.R.Walder, K.F.Howlett, and S.L.McGee. The role of GAPDH
acetylation in liver metabolism and type 2 diabetes. Abstracts of the Liver
Metabolism and Non-alcoholic Fatty Liver Disease Keystone Symposia, Whistler,
2015. Abstract.
III
DECLERATION
This thesis summarises my original and previously unpublished work, in the
School of Medicine, Deakin University, Waurn Ponds, Australia. All experimental
work was performed by the author, with the exception of the following:
x
GAPDH structure computational analysis in chapter two was performed by Dr
Victor Streltsov, Principal Scientist at the CSIRO Manufacturing Flagship,
Parkville, Australia.
x
Deuterated water glucose studies were performed by Dr Shaun Mason, Dr Greg
Kowalski, and Dr Clinton Bruce, School of Exercise and Nutrition Sciences,
Deakin University, Burwood, Australia.
x
Glycogen concentration analyses was performed by Dr Kirsten Howlett,
School of Exercise and Nutrition Sciences, and Assoc Prof Sean McGee,
School of Medicine, Deakin University, Waurn Ponds, Australia.
IV
ABSTRACT
Type 2 diabetes (T2D) is a chronic progressive disorder that is growing in
pandemic proportions, which can lead to complications that include cardiovascular
disease, kidney disease and stroke.
Elevated blood glucose levels and insulin
resistance are key factors in T2D which are characterised by altered muscle and liver
metabolism due to defective insulin action.
The liver is the major source of
endogenous glucose, and when hepatic insulin resistance occurs, it causes the liver to
produce excessive glucose which contributes to hyperglycemia.
Hepatic glucose production maintains blood glucose levels during fasting and
periods
of
high
glucose
demand,
through
the
metabolic
pathways
gluconeogenesis/glycolysis and glycogenolysis. Glycolysis and gluconeogenesis are
competing reciprocal pathways regulated by hormones, which in turn regulate the
transcription of gluconeogenic genes. The rate of gluconeogenesis and glycogenolysis
is increased in T2D, however the molecular mechanisms involved in mediating
elevated hepatic glucose production in T2D, are poorly understood.
Metabolic enzymes can be regulated by transcriptional regulation, allosteric
regulation, and post-translational modifications (PTMs). PTMs can regulate signalling
pathways acutely, and provide an elegant feedback mechanism for metabolic pathways
in response to nutrient availability. Protein acetylation is a PTM that has been
identified as playing a key role in regulating several metabolic processes. Reversible
acetylation is highly conserved from bacteria to mammals, and recent proteomic
studies have revealed more than 2000 non-nuclear acetylated proteins. It has also been
established that reversible acetylation of metabolic enzymes can directly affect
metabolic enzyme and pathway activity. Recent studies have demonstrated that the
acetylation
profile
of
the
metabolic
V
enzyme
glyceraldehyde-3-phosphate
dehydrogenase (GAPDH) plays a role in regulating glucose flux in Salmonella
enterica.
These studies demonstrated that the GAPDH acetylation profile was
regulated in response to nutrient availability, where high glucose conditions caused
hyper-acetylation of GAPDH, which favoured glucose breakdown through glycolysis.
They also demonstrated that when bacteria were exposed to citrate or acetate
conditions it resulted in deacetylation of GAPDH, which favoured glucose production
through gluconeogenesis.
Various proteomic studies have also revealed that
metabolic enzymes in the human liver, including GAPDH, are acetylated, but the role
of reversible GAPDH acetylation in mammalian glucose metabolism is unknown.
Altered acetylation profiles of particular proteins is now strongly regarded as
a source of dysfunction in several diseases, and the acetylation profile of GAPDH and
other metabolic enzymes in T2D is unknown.
To date, current approaches to
glycaemic control in many T2D patients are inadequate, and new therapeutical
approaches are required. This study aimed to identify new mechanisms involved in
the pathogenesis of hyperglycemia and T2D by investigating the acetylation profile of
GAPDH.
We hypothesised that reversible GAPDH acetylation regulates
gluconeogenesis, and is reduced in T2D, contributing to elevated hepatic glucose
production. We further propose that increasing GAPDH acetylation could enhance the
effects of metformin and further reduce hepatic glucose production in patients with
T2D, and therefore identification of enzymes controlling GAPDH acetylation could
have important therapeutic implications.
VI
TABLE OF CONTENTS
Page
Acknowledgements
I
Publications
II
Decleration
IV
Abstract
V
List of Figures
XIV
Abbreviations
XVIII
Chapter One: Review of the Literature
1.1 Introduction
1
1.2 Hepatic glucose metabolism
4
1.2.1 Hepatic insulin resistance
8
1.3 Acetylation
11
1.3.1 Metabolic regulation by acetylation
1.4 Glycolysis/gluconeogenesis
15
16
1.4.1
Glyceraldehyde 3-phosphate dehydrogenase (GAPDH)
1.4.2
Regulation of glycolysis/gluconeogenesis by GAPDH acetylation 19
1.4.3
Regulation of GAPDH acetylation
22
1.4.4
GAPDH acetylation in T2D
23
1.5 Summary
18
24
Chapter Two: The Role of GAPDH Lysine Residues in the Regulation
of Glucose Metabolism
2.1 Introduction
27
VII
2.2 Methods
30
2.2.1 Cloning
30
2.2.2 Cell culture
30
2.2.3 Protein extraction
31
2.2.4 Immunoprecipitation
31
2.2.5 Immunoblotting
32
2.2.6 Confocal imaging
32
2.2.7 Computational structural imaging
33
2.2.8 Metabolic profiling
33
2.2.9 Glucose production assay
34
2.2.10 GAPDH activity
35
2.2.11 Oil Red-O
36
2.2.12 RT-PCR
36
2.2.13 Statistical analyses
37
2.3 Results
38
2.3.1 GAPDH residues K115, K160, K225, and K252 are positioned
on the surface of GAPDH
38
2.3.2 Total GAPDH acetylation is not reduced in GAPDH mutants
38
2.3.3 WT GAPDH and 4KR GAPDH are localised in the cytoplasm
38
2.3.4 Lysine residues that are acetylated determine GAPDH
glycolytic/gluconeogenic activity in vitro
39
2.3.5 Basal glucose production is increased in K225R, K252R and
4KR mutant GAPDH cell lines
39
2.3.6 Glycolytic flux is reduced in K225R, K252R and 4KR mutant
GAPDH cell lines
39
VIII
2.3.7 Insulin suppresses glucose production and forskolin increases
glucose production in mutant GAPDH FAO cells
40
2.3.8 Basal and maximum glycolysis is reduced in 4KR GAPDH
cell lines
40
2.3.9 Mitochondrial respiration is unchanged in 4KR GAPDH mutant
FAO cells
41
2.3.10 Glucose and palmitate dependent oxidation is impaired in
4KR GAPDH mutant FAO cells
41
2.3.11 Lipid accumulation is increased in 4KR GAPDH mutant
FAO cells
42
2.4 Discussion
54
Chapter Three: Mediators of GAPDH Acetylation/Deacetylation
3.1 Introduction
63
3.2 Methods
66
3.2.1 Liver protein extraction, immunoprecipitation, and
immunoblotting
66
3.2.2 Cell culture
66
3.2.3 DN HDAC6 construct
66
3.2.4 Transfection
67
3.2.5 Glucose production assay
67
3.2.6 Metabolic profiling
67
3.2.7 Statistical analyses
68
3.3 Results
69
3.3.1 Overexpression of HDAC6 reduces α-tubulin acetylation
IX
69
3.3.2 KDAC inhibitors reduce glucose production in WT GAPDH
FAO cells, but not 4KR GAPDH FAO cells
3.3.3 DN HDAC6 reduces glucose production and increases ECAR
69
69
3.3.4 DN HDAC6 increases OCR, basal mitochondrial respiration,
ATP turn over, and uncoupled respiration
70
3.3.5 DN HDAC6 reduces glucose production in WT GAPDH
co-expressing cells, but not 4KR GAPDH co-expressing cells
70
3.3.6 HDAC6 opposes the insulin response and promotes the
forskolin response
70
3.3.7 Total GAPDH acetylation was not altered by HDAC6
overexpression
71
3.3.8 Metformin reduced glucose production further in WT GAPDH
cells compared to 4KR GAPDH cells
3.4 Discussion
71
79
Chapter Four: Regulation of GAPDH Acetylation
4.1 Introduction
85
4.2 Methods
88
4.2.1 Total GAPDH acetylation assay
88
4.2.2 Permits and ethical approval
88
4.2.3 Mice experimental design
88
4.2.4 Liver protein extraction, immunoprecipitation, and
immunoblotting
89
4.2.5 Statistical analyses
89
4.3 Results
91
X
4.3.1 Total GAPDH acetylation in FAO hepatocytes was unchanged
by glucose, insulin, and forskolin
91
4.3.2 Insulin, glucagon and nor-epinephrine do not regulate GAPDH
acetylation in vivo
91
4.3.3 Total GAPDH acetylation was unchanged by fasting and
re-feeding
91
4.3.4 Hepatic GAPDH acetylation is reduced in T2D
92
4.3.5 Global hepatic lysine acetylation is unchanged in T2D
92
4.3.6 Hepatic GAPDH acetylation is reduced in mice on a high fat diet
92
4.4 Discussion
100
Chapter Five: The Role of Hepatic GAPDH Lysine Residues in the
Regulation of Glucose Metabolism in vivo
5.1 Introduction
106
5.2 Methods
108
5.2.1 Permits and ethical approval
108
5.2.2 Construction of adeno-associated virus vectors serotyped with
AAV8 capsid
108
5.2.3 Animals and experimental design
109
5.2.4 Glucose tolerance tests (GTT)
110
5.2.5 Insulin tolerance test (ITT)
110
5.2.6 Glucagon tolerance test (GnTT)
110
5.2.7 Fasting/re-feed blood glucose test
110
5.2.8 Hepatic glucose production measured by deuterated water
labelling
111
XI
5.2.9 Insulin ELISA
112
5.2.10 Body composition analysis
112
5.2.11 Protein extraction, immunoprecipitation, and immunoblotting
112
5.2.12 GAPDH activity assay
113
5.2.13 Quantitative measurement of percent Haemoglobin A1c in blood
113
5.2.14 Glycogen concentration assay
113
5.2.15 Triglyceride quantification
114
5.2.16 Histology
114
5.2.17 Statistical analyses
114
5.3 Results
115
5.3.1 AAV8 GAPDH expression is liver specific
115
5.3.2 Total hepatic GAPDH acetylation is unchanged in
4KR GAPDH mice
115
5.3.3 Gluconeogenic GAPDH activity is increased in
4KR GAPDH mice
115
5.3.4 Fractional contribution of gluconeogenesis to total endogenous
glucose is elevated in 4KR GAPDH mice
115
5.3.5 Food consumption and body composition were unchanged in
4KR GAPDH mice
116
5.3.6 Weekly blood glucose levels and percentage HbA1c are
unchanged in 4KR GAPDH mice
116
5.3.7 Glucose tolerance was altered in 4KR GAPDH mice
after injection with glucose
116
5.3.8 Glucagon action was altered in 4KR GAPDH mice
after injection with glucagon
117
XII
5.3.9 The post-prandial response after fasting and re-feeding
was unchanged in 4KR GAPDH mice
117
5.3.10 Insulin tolerance is not altered in 4KR GAPDH mice
117
5.3.11 Hepatic glycogen content was elevated in 4KR GAPDH mice
118
5.3.12 Hepatocyte morphology and hepatic triglyceride levels are
unchanged in 4KR GAPDH mice
118
5.4 Discussion
131
Chapter Six: Summary and Conclusions
136
Bibliography
148
Appendix
157
XIII
LIST OF FIGURES
Figure
Page
Figure 1.1:
Insulin and glucagon regulate hepatic glucose metabolism
Figure 1.2:
Hepatic insulin signalling pathway, and fatty acid induced
insulin resistance in T2D
Figure 1.3:
6
9
Elevated hepatic glucose production due to fatty acid
induced insulin resistance
10
Figure 1.4:
Reversible lysine acetylation
12
Figure 1.5:
Glycolysis/gluconeogenesis metabolic pathway, with
acetylated enzymes identified by proteomics, highlighted
in red
17
Figure 1.6:
Proposed mechanism of GAPDH regulation by acetylation
25
Figure 2.1.1:
Rat, mouse, and human GAPDH protein sequence and
pairwise alignment
28
Figure 2.2.1:
GAPDH activity assay pathway
35
Figure 2.3.1:
Crystal structure of one monomer of rabbit muscle
glyceraldehyde-3-phosphate dehydrogenase, generated
from PDB
Figure 2.3.2:
43
GAPDH expression and total GAPDH acetylation in
WT and mutant GAPDH FAO cells
Figure 2.3.3:
FAO WT GAPDH-GFP and FAO 4KR GAPDH-GFP
confocal imaging
Figure 2.3.4:
44
45
FAO WT GAPDH-GFP and FAO mutant GAPDH-GFP
activity
46
XIV
Figure 2.3.5:
FAO WT GAPDH and FAO mutant GAPDH glucose
production assay
Figure 2.3.6:
47
FAO WT GAPDH and FAO mutant GAPDH extra
cellular acidification rate
Figure 2.3.7:
FAO WT GAPDH and FAO mutant GAPDH, insulin and
forskolin treated glucose production assay
Figure 2.3.8:
49
FAO WT GAPDH and FAO 4KR GAPDH basal and
maximum glycolysis and mRNA expression
Figure 2.3.9:
48
50
FAO WT GAPDH and FAO 4KR GAPDH OCR and
mitochondrial respiration
51
Figure 2.3.10: FAO WT GAPDH and FAO 4KR GAPDH substrate
dependent OCR
52
Figure 2.3.11: FAO WT GAPDH and FAO 4KR GAPDH Oil Red-O stain
53
Figure 2.4.1:
4KR GAPDH Metabolic pathway hypothesis
60
Figure 3.2.1:
DN HDAC6 construct
66
Figure 3.3.1
α-tubulin acetylation in WT HDAC6 and DN HDAC6
expressing FAO cells
Figure 3.3.2
72
FAO WT GAPDH and FAO 4KR GAPDH KDAC inhibited
GP assay
73
Figure 3.3.3:
FAO WT HDAC6 and FAO DN HDAC6 energetics
74
Figure 3.3.4:
HDAC6 mutant/GAPDH mutant co-expressing FAO cell
GP assay
Figure 3.3.5:
75
HDAC6 mutant/GAPDH mutant co-expressing FAO cell
forskolin and insulin treated GP assay
XV
76
Figure 3.3.6:
GAPDH acetylation in WT HDAC6 and DN HDAC6
expressing FAO cells
Figure 3.3.7:
77
Metformin treated FAO WT GAPDH and FAO
4KR GAPDH GP assay
Figure 4.3.1:
78
GAPDH acetylation in FAO cells after exposure to high
glucose, insulin and forskolin
Figure 4.3.2:
94
Hepatic GAPDH acetylation in C57BL/6 mice after injection
with either insulin, glucagon, norepinephrine, or vehicle
95
Figure 4.3.3:
GAPDH acetylation in fasted and fed C57BL/6 mice
96
Figure 4.3.4:
Hepatic GAPDH acetylation in db/db +/- and db/db -/- mice
97
Figure 4.3.5:
Global hepatic lysine acetylation and acetylated-tubulin
levels in db/db +/- and db/db -/- mice
Figure 4.3.6:
98
Hepatic GAPDH acetylation in mice fed CHOW versus
mice fed a HFD
99
Figure 5.2.1:
TBG-GAPDH-GFP AAV8 plasmid map
108
Figure 5.2.2:
Animal study timeline
109
Figure 5.3.1:
Western blot for GFP expression
119
Figure 5.3.2:
Hepatic total GAPDH acetylation in 4KR GAPDH and
WT GAPDH mice
Figure 5.3.3:
120
Gluconeogenic/glycolytic GAPDH activity in 4KR GAPDH
and WT GAPDH mice
Figure 5.3.4:
121
Percentage of endogenous glucose derived from
gluconeogenesis and glycogenolysis after fasting
122
Figure 5.3.5:
Body composition and food consumption
123
Figure 5.3.6:
Weekly blood glucose levels and percentage HbA1c
124
XVI
Figure 5.3.7:
Glucose tolerance test
125
Figure 5.3.8:
Glucagon tolerance test
126
Figure 5.3.9:
Re-feed postprandial response
127
Figure 5.3.10: Insulin tolerance test
128
Figure 5.3.11: Muscle and liver glycogen concentration in WT GAPDH
and 4KR GAPDH mice
129
Figure 5.3.12: Liver morphology and triglyceride quantification in
4KR GAPDH and WT GAPDH mice
XVII
130
ABBREVIATIONS
AcK – Acetyl lysine
ADP – Adenosine diphosphate
Akt2/PKB2 – Protein kinase B
ALDB – Aldolase B
AMP – Adenosine monophosphate
ANOVA - Analysis of variance
ATP - Adenosine triphosphate
cAMP – Cyclic adenosine monophosphate
ChREBP – Carbohydrate responsive element binding protein
CoA – Coenzyme A
DAG – Diacylglycerols
DHAP – Dihydroxyacetone phosphate
DIO – Diet induced obesity
DN – Dominant Negative
ECAR – Extracellular acidification rate
ER - Endoplasmic reticulum
FAO – Rat hepatoma cell line
FBPase – Fructose 1,6-bisphosphatease
FCCP – Carbonyl cyanide-4-phenylhydrazone
FOXO – Forkhead box protein O
GAPDH – Glyceraldehyde 3-phosphate dehydrogenase
GLUT2 – Glucose transporter type 2
GnTT – Glucagon tolerance test
G-6-Pase – Glucose-6-phosphatase
XVIII
GP – Glucose production
G3P – Glycerol phosphate
GSK3 – Glycogen synthase kinase 3
GTT – Glucose tolerance test
%HbA1c – Percentage Haemoglobin A1c
HDAC – Histone deacetylase
HDACi – Histone deacetylase inhibitor
H & E – Hematoxylin and eosin stain
HFD – High fat diet
IP – Immunoprecipitation
IRES - Internal ribosome entry site
IRS-2 – Insulin receptor substrate 2
ITT – Insulin tolerance test
KAT – Lysine acetyltransferase
KDAC – Lysine deacetylase
KHB – Krebs-Henseleit buffer
LCFA – Long-chain fatty acids
LDH – Lactate dehydrogenase
L-PK – Liver type pyruvate kinase
NAD+ – Nicotinamide adenine dinucleotide
NEFA – Non-esterified fatty acids
NES – Nuclear export signal
OCR – Oxygen consumption rate
PC – Pyruvate carboxylase
PCAF – P300/CBP-associated factor
XIX
PDB – Protein Data Bank
PDH – Pyruvate dehydrogenase
PEPCK – Phosphoenolpyruvate carboxykinase
PFK-1 – Phosphofructokinase-1
PGK – Phosphoglycerate kinase
PGM – Phosphoglycerate mutase
PK – Pyruvate kinase
PKA – Protein kinase A
PKC-є – Protein kinase C epsilon
PPP – Pentose phosphate pathway
PP2A – Protein phosphatase 2
PTM – Post-translational modification
Ser/Thr – Serine/Threonine
SILAC – Stable isotope labeling by amino acids in cell culture
SIRT – Sirtuin
STAT3 – Signal transducer and activator of transcription 3
TBG – Thyroxine binding globulin
TCA – Tricarboxylic acid cycle, or the citric acid cycle
T2D – Type 2 diabetes
WHO – World Health Organisation
XX
XXI
CHAPTER 1
REVIEW OF THE LITERATURE
1.1
INTRODUCTION
Type 2 diabetes (T2D) is a chronic metabolic disorder that has a significant
impact on health, life expectancy, and quality of life
1,2
. T2D is a complex disease
predominantly caused by the body’s ineffective use of insulin, and is often the result of
excess body weight and physical inactivity1,2. In Australia T2D is the fastest growing
chronic condition, and the 6th leading cause of death1. Lifestyle factors such as obesity
and physical inactivity are believed to play a major role in the development of T2D,
although it is now well established and widely accepted, that there is also a complex
interplay between genetic, nutritional, and environmental factors that lead to the
development of T2D1,3. The World Health Organisation (WHO) forecasts that T2D
and related deaths will double over the next forty years, which correlates with a rise in
obesity, with 80% of individuals with T2D being obese1,2,4. For this reason, T2D is
now emerging as one of the greatest health challenges that humanity will endure in the
21st century5.
In Australia T2D is defined by fasting blood glucose levels above 7.0
mmol/L, and is characterised by insulin deficiency, or resistance in metabolically
active tissues, and an increase in hepatic glucose production1,6. Insulin resistance
leads to chronic hyperglycemia, that disrupts metabolic homeostasis and can cause
dysregulation of fat, carbohydrate, and protein metabolism6,7. Insulin resistance
prevents the uptake of glucose into insulin sensitive tissues such as skeletal muscle
and fat, which accounts for approximately 80% of glucose disposal, and also
stimulates the β-cells of the pancreas to release more insulin in response to the
1
elevated blood glucose levels5,8. In addition, insulin resistance results in elevated
de novo synthesis of glucose from the liver, and an increase in the release of fatty
acids from adipose tissue into the blood stream9. This results in hyperinsulinemia,
hyperglycemia, and an accumulation of fatty acids (hyperlipidemia)10. The outcome
of these combined metabolic deficiencies is a feedback system that further
exacerbates dysfunctional metabolism, and initiates
circulating inflammatory
factors and oxidative stress11. Elevated concentrations of non-esterified fatty acids
(NEFA) inhibit skeletal muscle glucose metabolism while increasing hepatic
gluconeogenesis, and activating inflammatory factors5. Increased levels of circulating
inflammatory cytokines increase adipocyte lipolysis and interfere with the insulin
signalling cascade5. In addition, patients with T2D have a defect in lipid oxidation
and impaired oxidative capacity5.
T2D leads to acute and long term complications that cause a range of
serious health problems. Over time, chronic hyperglycemia can lead to an increased
risk of peripheral arterial and cerebrovascular disease, retinopathy, nephropathy
and neuropathy1,2. The risk of death for individuals with T2D is approximately
double that of individuals who do not have T2D, with 50% of individuals
diagnosed with T2D dying from cardiovascular disease1,2. The inability of the
majority of individuals with T2D to make lifestyle modifications, such as consuming
a healthy diet and participating in regular exercise, means
pharmacological
intervention is required in order to combat the exponential growth of T2D. There is
no cure for T2D, and the drugs available do not adequately control the symptoms
of T2D5.
The impaired metabolism seen in T2D is complicated by the sheer
number and complexity of biochemical reactions and metabolic pathways involved,
2
which further complicates pharmacological intervention. Hepatic glucose production
is increased in T2D, which contributes to the high blood glucose levels that are a
defining feature of this disease12,13. By reducing hepatic glucose production, as seen
in T2D, there is the potential to reduce blood glucose levels in individuals with T2D.
Dysfunctional hepatic gluconeogenesis that occurs in T2D is the major source of
endogenous glucose production in individuals with T2D13. Gluconeogenesis has been
shown to be regulated by the expression of a number of metabolic genes and proteins,
but pharmacological attempts to control the regulation of gluconeogenesis at a
transcriptional level have been unsuccessful13,14. In addition, transcriptional regulation
does not account for the rapid changes that occur in metabolic pathways. Post
translational modifications (PTMs), such as reversible lysine acetylation, are now
widely accepted as a potential mechanism involved in the acute regulation of
metabolism15,16.
Reversible lysine acetylation is a PTM, which is emerging as a major
regulator of cellular processes in response to environmental and nutritional stimuli. In
particular, acetylation of metabolic enzymes in bacteria has recently been implicated
as a fundamental component regulating metabolic flux, rate, and direction17-19.
Acetylation is an abundant PTM, and the acetylation/deacetylation of histones is a
well-known regulator of transcription18,20. Only recently has it been revealed that
acetylation occurs in abundance on non-histone and cytosolic proteins20,21.
In
particular, most metabolic enzymes can undergo acetylation20,21. Recent proteomic
studies carried out to analyse acetylated proteins throughout the whole cell have found
that as many as 95% of metabolic enzymes can undergo acetylation, with up to 85%
of gluconeogenic enzymes found to be acetylated in the human liver19-22. Considering
the large number of enzymes that are acetylated in the gluconeogenic pathway,
3
acetylation could possibly play an important role in the regulation and dysregulation
of this pathway. Recently it was demonstrated in the bacteria Salmonella enterica
(S.enterica), that acetylated GAPDH increases glycolysis, and deacetylated GAPDH
increases gluconeogenesis19,22. It was also shown in S.enterica, that the carbon source
determined the acetylation profile of GAPDH, where GAPDH acetylation increased
in glucose rich media, and GAPDH acetylation levels decreased in acetate or citrate
rich media19. This potentially links the regulation of glucose metabolism to substrate
availability, which can be connected to both diet and exercise. Further investigation
is required to see whether this mechanism of metabolic regulation by reversible
GAPDH acetylation occurs in mammals.
The addition of an acetyl moiety to a lysine residue is catalysed by a family
of enzymes known as the lysine acetyltransferases (KATs)23-25. The removal of an
acetyl moiety from a lysine residue is catalysed by a family of enzymes called the
lysine deacetylases (KDACs)23-25. There are eighteen known human KDACs, which
are further divided into two groups; the eleven Histone deacetylases (HDACs) which
are zinc (Zn2+) dependent, and the seven Sirtuins (SIRTs), which are nicotinamide
adenine dinucleotide (NAD+) dependent23,25. Altered acetylation profiles of proteins
is now strongly regarded as a source of dysfunction in several diseases, however, the
acetylation profile of GAPDH in T2D has not been investigated. The KDACs
responsible for the deacetylation of GAPDH in mammals are only partially known.
Uncovering these proteins, coupled with the knowledge of altered acetylation profiles
in disease, could hypothetically provide new therapeutic targets26.
1.2 HEPATIC GLUCOSE METABOLISM
The liver is the primary source of endogenous glucose, and therefore plays
4
a key role in glucose homeostasis13,14. The liver has the capacity to store and generate
de novo formation of glucose, which is tightly regulated by hormones, in particular
insulin and glucagon (Figure 1.1)14,27. Metabolites such as free fatty acids, along
with 3-carbon compounds such as lactate, alanine, pyruvate and glycerol, also play
a role in regulating hepatic glucose metabolism14,28. The liver stores glucose as
glycogen, which can then be converted back into glucose via glycogenolysis, and
can also produce glucose via gluconeogenesis (Figure 1.1)14. In obese individuals
with T2D, approximately one-third of endogenous glucose is produced by hepatic
glycogenolysis, with the remaining two-thirds of endogenous glucose produced by
hepatic gluconeogenesis29. Gluconeogenesis only occurs in the liver and kidney,
as glucose 6-phosphatase (G-6-Pase, an enzyme that catalyses the final step in
gluconeogenesis)
is
principally found
in
theses
tissues,
however
non-
gluconeogenic tissues, such as skeletal muscle, can produce glucose that is stored
as glycogen30. In order for gluconeogenesis to occur, substrates produced by
peripheral tissues must be transported to gluconeogenic tissues, in particular the
liver9.
Glucose homeostasis is primarily controlled at a hormonal level, where
passive glucose intake through GLUT2 transporters in the pancreas allows the
glucose sensitive α and β cells to regulate the secretion of glucagon and insulin,
respectively27. When blood
glucose levels are low, glucagon is secreted,
stimulating hepatic glucose production. When blood glucose levels are high, insulin
is secreted, which inhibits hepatic glucose production (Figure 1.1)27. Insulin
stimulates glucose uptake by peripheral tissues such as skeletal muscle and fat,
and
increases
glycolysis,
glycogenesis,
lipogenesis
and
protein
synthesis27,31,32. Insulin does not affect glucose uptake by the liver, but rather
5
Glucose secretion
Liver
Glycolysis
Glycogen
Glycogenesis
Gluconeogenesis
Glycogenolysis
Insulin
(High blood glucose)
Glucagon
(Low blood glucose)
α
β
Pancreas
Figure 1.1: Insulin and glucagon regulate hepatic glucose metabolism. Increased blood glucose
levels stimulate insulin secretion resulting in increased glycolysis and glycogenesis, and decreased
glycogenolysis and gluconeogenesis27,33. When blood glucose levels are low insulin secretion is
decreased and the secretion of glucagon is increased, resulting in increased glycogenolysis and
gluconeogenesis, and decreased glycolysis and glycogenesis27.
6
inhibits gluconeogenesis and glycogenolysis27,31,32. Glucose uptake by the liver is
facilitated by the glucose transporter GLUT2, which passively allows glucose
movement across the cell membrane when high concentrations of sinusoidal glucose
are present (Figure 1.2)14. During times of elevated blood glucose levels the liver
metabolises glucose into energy and substrates such as glycogen and lipids14. When
blood glucose levels are low the liver maintains blood glucose levels via the
production of glucose through gluconeogenesis, and the release of glucose from
stored glycogen via the process of glycogenolysis (Figure 1.1)14.
In research studies to date the regulation of gluconeogenesis in the liver
has been primarily defined at a transcriptional level through the regulation of
insulin and
glucagon. Insulin can inhibit gluconeogenesis by preventing the transcription of
Phosphoenolpyruvate carboxykinase (PEPCK) and G-6-Pase, and by suppressing
substrate availability, such as the generation of free fatty acids from adipose
tissue31,32.
PEPCK, G-6-Pase and Fructose 1,6-bisphosphatase (FBPase) are
acknowledged as gluconeogenic rate limiting enzymes, where insulin signalling
results in a decrease in the expression of the PEPCK and G-6-Pase genes (Figure
1.2)34,35. The reduction in gene expression leads to a decrease in protein levels that
limits gluconeogenic flux34-36. When hepatic insulin resistance occurs, it results in
an increase in the expression of gluconeogenic rate limiting enzymes that increase
the rate of gluconeogenesis, and
leads to elevated blood glucose levels.
Dysfunctional hepatic metabolism in individuals with T2D, namely the increased
rate of gluconeogenesis due to insulin resistance, is a key factor involved in the
increase of blood glucose levels in T2D.
7
1.2.1 Hepatic Insulin Resistance
The mechanisms of hepatic insulin resistance from nutrient overload are
incompletely understood, however it is widely accepted that there is an association
with the accumulation of by-products from nutrient overload, specifically lipid
accumulation (Figure 1.2)37. This induces
endoplasmic reticulum (ER) stress,
inflammatory effects, and other humoral factors, which
lead to hepatic insulin
resistance37,38. Hepatic insulin resistance results in excessive hepatic glucose
production, further exacerbating the high blood glucose levels seen in T2D (Figure
1.3)37,38.
The principal cause of excessive hepatic
glucose production is
predominantly due to the increased rate of hepatic gluconeogenesis13. Hepatic insulin
resistance inhibits the suppression gluconeogenesis and glycogenolysis, and reduces
glycogen synthase activation6,7. This results in reduced glycogen synthesis and
storage, and excessive hepatic glucose production.
Several factors have been implicated in the dysregulation of hepatic
gluconeogenesis due to insulin resistance, such as an increase in the expression of
gluconeogenic rate limiting enzymes34-36. There are numerous insults that can cause
insulin resistance, which can also occur independent of changes to the insulin
signalling pathway39.
Traditionally, a build-up of fatty acids activates kinase
pathways that prevent the phosphorylation of proteins involved in the insulin
signalling pathway (Figure 1.2)40. However, research by Hoehn et al suggests that
defects within the insulin signalling pathway that occur in insulin resistance are
unlikely to contribute to the early development of insulin resistance39.
Acetylation is a PTM that has been gaining interest, and is now believed
to be as abundant, and as relevant as phosphorylation. Lysine acetylation is a
reversible PTM that has only recently been associated with metabolic enzymes41,42.
8
Fatty acids
Figure 1.2: Hepatic insulin signalling pathway, and fatty acid induced insulin resistance in T2D.
Insulin stimulates the phosphorylation of IRS-2, PI3-K, and Akt-2 which leads to the phosphorylation
of GSK3 and FOXO. Phosphorylation of GSK3 increases glycogenesis and the storage of glycogen.
Phosphorylation of FOXO inhibits the expression of gluconeogenic rate limiting enzymes PEPCK and
G-6-Pase. Insulin resistance can occur due to increased delivery of fatty acids from the plasma and a
build-up of diacylglycerol (DAG) content, which leads to the activation of protein kinase C (PKC-ε),
decreasing insulin-stimulated IRS-1/IRS-2 tyrosine phosphorylation, PI3K activation, and downstream
insulin signalling. These fatty acid induced flaws in insulin signalling causes a decrease in GSK3 and
FOXO phosphorylation, resulting in a decrease in glycogen synthesis, and an increase in the expression
of gluconeogenic rate limiting enzymes, increasing gluconeogenesis and hepatic glucose production6.
IRS-2 = insulin receptor substrate-2, PI3-K = Phosphoinositide 3-kinase, Akt-2 = protein kinase B-2,
LCFA = long chain fatty acids, GLUT-2 = glucose transporter-2, G6Pase = glucose 6-phosphatase.
9
Insulin
Free fatty acids
Liver
Insulin resistance
Glycogen
Glycogenolysis
Gluconeogenesis
Glucose
Increased glucose production and secretion
Figure 1.3: Elevated hepatic glucose production due to fatty acid induced insulin resistance.
Insulin resistance can occur due to an increase in free fatty acid uptake, and results in a build-up of
gluconeogenic substrates, causing an increase in gluconeogenesis and glycogenolysis33,43. The outcome
of all these factors results in an increase in hepatic glucose production.
10
Recent studies have found that a large number of metabolic proteins can be
acetylated, potentially providing a novel biological mechanism of metabolic
regulation20.
Certain acetylation profiles of metabolic enzymes could be
physiologically significant, and it is possible that the acetylation profile of metabolic
enzymes in diseases, such as T2D, could play a role in dysfunctional metabolism.
1.3 ACETYLATION
Lysine acetylation is a reversible PTM that was first documented by Allfrey
et al in 196444. Reversible acetylation in mammals involves an acetyl group from
acetyl-CoA that is transferred to the epsilon (ε)-amino group of specific lysine
residues to form N-acetyl lysine (Figure
1.4)45,46. Lysine acetylation and
deacetylation modifies the charge present on lysine residues, which modifies the
overall electrostatic charge of the protein, which can have effects on protein stability,
activity, and conformation47. Unmodified lysine residues are positively charged,
which can modulate protein stability and function by way of hydrogen bonding
with other amino acid residues45,48. Recent studies have revealed that a large number
of cellular proteins, in particular metabolic enzymes, are capable of reversible
acetylation20,21. Due to the conservation, and large number of proteins capable of
acetylation, lysine acetylation of proteins now rivals phosphorylation as a potential
regulator of multiple cellular processes45,46,49-51. Acetylation and deacetylation is
facilitated by KATs and KDACs respectively, however acetylation can also occur
independent of enzymes. It has been demonstrated in vitro that physiologic pH and
acetyl-CoA concentrations are sufficient to cause dose and time dependent, but
enzyme independent, acetylation of mitochondrial and non-mitochondrial proteins52.
In 2009 Zhang et al identified a considerable number of lysine acetylated
orthologs between mammals and Escherichia coli (E.coli), implying that lysine
11
Lysine residue
Acetylated-lysine residue
NH3+
CH2
CH2
H
O
N
C
CH3
CH2
Acetyl-CoA
CoA
CH2
CH2
CH2
CH2
CH2
KAT
Protein
Protein
KDAC
Figure 1.4: Reversible lysine acetylation. The positively charged amino acid residue lysine, can be
acetylated/deacetylated, thereby changing the charge on the lysine tail. Biochemically, the є-amino
groups of lysine residues are positively charged, which maintains protein stability via hydrogen bonding
with neighbouring amino acid residues48. This can regulate a proteins propensity to interact with other
residues, or change the activity of an enzyme19. When acetyl groups are attached to the lysine tail, the
charge is neutralised, causing a possible shape/activation shift. Acetyl groups are sourced from the
acetyl-coenzyme A complex, which plays a key role in many metabolic processes.
12
acetylation is an evolutionarily conserved modification, from bacteria to mammals51.
This suggests that lysine acetylation has evolved, and has been conserved, as a
necessary component for a wide variety of cellular events. The number of cellular
pathways in which lysine acetylation plays a role is only just being recognised now,
to the point where lysine acetylation has emerged as a PTM that could be as
physiologically important as phosphorylation46. These revelations have only just
become apparent due to the improved techniques and technology used to
investigate acetylation, which is still in its infancy. As the techniques used to
research acetylation improve and become more readily available, it is predicted that
it will reveal a vast array of biological mechanisms that are regulated by acetylation45.
Acetylation is best known for its role in regulating transcription,
through the acetylation/deacetylation of histones, where acetylation allows DNA to
detach from the histones so that transcription factors and RNA polymerases can
gain access to the DNA16. Acetylation of histones, and its role in regulating
transcription, has been largely examined over the last fifty years, with little
consideration for the acetylation of non-nuclear
proteins49.
Due to new
experimental techniques such as stable isotope labelling by amino acids in cell
culture (SILAC), recent proteomic studies have demonstrated that non-histone
proteins are extensively acetylated, in particular metabolic enzymes21,53-55.
Metabolic enzymes were acetylated in both bacteria and mammals, suggesting an
evolutionary conserved role for acetylation in regulating metabolic direction/flux,
protein stability, and enzyme activity19,22,56-59. It was also revealed that the majority
of metabolic enzymes involved in glycolysis and gluconeogenesis can also undergo
reversible lysine acetylation60.
13
The amount of an enzyme present at any time can be regulated in the long
term at a transcriptional and translational level, by gene and protein expression,
however the mechanisms of acute regulation in metabolism that occur due to external
stimuli have not been completely elucidated34,35. Regulation of proteins and enzymes
by reversible lysine acetylation could provide an acute mechanism of metabolic
regulation in response to internal and external stimuli61.
Acetylation/deacetylation is dependent on the acetyl-CoA/CoA levels, which
is an intermediate metabolite, and links acetylation to carbon sources, and also the
supply of nutrients in the diet56,62. The carbon source has been shown to affect the
acetylation profile of metabolic enzymes, where Wang et al demonstrated in
S.enterica, that cells grown in glucose rich media have elevated acetylated lysine
residues compared to cells grown in acetate or citrate rich conditions19. This potentially
suggests that acetylation is sensitive to nutrient availability, and could play a key role
in metabolic regulation in response to nutritional overload or depletion. Excluding
exogenous sources of acetyl-CoA, such as tryptophan, acetyl-CoA can be synthesised
through several metabolic pathways, which are also dependent on nutrient
availability63. In the mitochondria, pyruvate can be converted to acetyl-CoA by
pyruvate dehydrogenase, and acetyl-CoA can also be generated from fatty acids via βoxidation63,64. In the cytosol, acetyl-CoA synthetase can synthesise acetyl-CoA from
acetate, however the majority of acetyl-CoA is synthesised from citrate, which is
exported out of the mitochondria via the citrate transporter and then converted to
acetyl-CoA by ATP citrate lyase63,64.
Acetyl-CoA can also be generated from
acetoacetate and pyrimidine catabolism63.
Acetylation can modulate the activity of metabolic enzymes, suggesting that
there could be altered metabolic protein acetylation profiles that may contribute to the
14
pathophysiology of metabolic diseases like T2D48,65,66. How the acetylation profile of
enzymes in healthy and diseased pathological states affects enzyme activity and
metabolism remains unclear48. It is currently unknown whether the acetylation of
gluconeogenic rate limiting enzymes is altered in T2D, or whether this mechanism
contributes to elevated hepatic glucose production.
1.3.1 Metabolic Regulation by Acetylation
Transcriptional regulation generally occurs over several hours, and can
only account for long term metabolic responses. There is a time delay for changes in
mRNA levels and the equivalent changes in protein levels at a transcriptional level,
which suggests that there must be other mechanisms that account for the acute
metabolic changes that occur in response to internal and external stimuli67. In
addition, the amount of metabolic enzyme present does not take into account the
active state of metabolic enzymes, so mechanisms that regulate enzyme activity
could also be responsible for the acute metabolic responses to external stimuli64.
Recent studies have found that PTMs play a significant role in metabolic
regulation16,60.
PTMs can occur rapidly, and are now largely thought to be
responsible for the acute metabolic changes that occur in response to external
stimuli16,60. PTMs can influence the activity, stability, and sub-cellular localisation
of enzymes, which could explain the regulation of metabolism independent of
transcriptional regulation. In addition, there have been several recent studies that have
demonstrated that transcriptional expression of PEPCK and G-6Pase does not account
for increased gluconeogenesis12,68,69. Samuel et al demonstrated that PEPCK and G6Pase levels did not change in T2D, or after fasting in control rats68. Ramnanan et al
and Edgerton et al showed that maximum physiological levels of insulin could reduce
hepatic glucose production, and that changing expression levels of PEPCK had no
15
effect on changes in gluconeogenesis12,69. These findings are in contrast to the
prevailing theory of gluconeogenic regulation by expression of the traditional rate
limiting enzymes PEPCK and G-6-Pase, and suggests that other cellular mechanisms
are responsible for the increased hyperglycemia and gluconeogenesis that occur in
T2D.
PTMs of metabolic enzymes could play a key role in regulating metabolism,
and if dysregulation of PTMs of metabolic proteins can be shown in T2D, this could
partially account for the dysregulated metabolism seen in T2D. In particular, if the
acetylation profile of metabolic enzymes involved in gluconeogenesis is altered in
T2D, it could contribute to the increased rate of gluconeogenesis and hepatic glucose
production in T2D.
1.4 GLYCOLYSIS/GLUCONEOGENESIS
Glycolysis is a catabolic anaerobic pathway, consisting of ten reactions
connecting ten intermediate metabolites, transforming glucose into lactate and
pyruvate, releasing energy in the form of adenosine-5'-triphosphate (ATP) and
nicotinamide adenine dinucleotide (NADH), (Figure 1.5)63.
The end product of
glycolysis, pyruvate, is shuttled into the mitochondria to fuel the citric acid cycle
(TCA). Gluconeogenesis is the reverse reaction of glycolysis, resulting in glucose
that is generated from pyruvate63. Gluconeogenesis requires energy in the form of
ATP to proceed, and is generally restricted to the liver and parts of the kidney63.
Glycolysis and gluconeogenesis are reciprocal, where enzymes shared by both
pathways can only allow flux through the glycolytic or gluconeogenic pathway at
any one time. The tightly regulated relationship between glycolysis/gluconeogenesis
and blood glucose levels is dysregulated in T2D, and the molecular mechanisms
16
Glucose
ATP
G-6-Pase
HK
ADP
Glucose-6-phosphate
GPI
Fructose-6-phosphate
FBPase
ATP
PFK-1
ADP
Fructose-1,6-bisphosphate
ALDB
TPI
Dihydroxyacetone phosphate
3-phosphoglyceraldhyde
NAD+ + H+
NAD+ + PI
GAPDH
NADH + H+
1,3-Bisphosphoglycerate
ADP
ADP
PGK
ATP
Glycolysis
Gluconeogenesis
NADH + PI
ATP
3-Phosphoglycerate
PGM
2-Phosphoglycerate
Enolase
Phosphoenolpyruvate
GDP
PEPCK
GTP
ADP
PK
ATP
Pyruvate
PC
Oxaloacetate
ADP
PDH
LDH
ATP
Lactate
Acetyl-CoA
To TCA cycle
Figure 1.5: Glycolysis/gluconeogenesis metabolic pathway, with acetylated enzymes identified by
proteomics, highlighted in red. Arrows indicate the direction of a reaction, some of which are
reversible. Glycolysis converts glucose to pyruvate, and gluconeogenesis is the reverse of this reaction,
producing glucose from pyruvate20. Abbreviations are as follows: G-6-Pase=Glucose-6-phosphatase,
HK=Hexokinase, GPI=Glucose-6-phosphate isomerase, FBPase=Fructose 1,6-bisphosphatease, PFK1=Phosphofruvtokinase-1,
GAPDH=Glyceraldehyde
PGM=Phosphoglycerate
ALDB=Aldolase
3-phosphate
mutase,
B,
dehydrogenase,
PK=Pyruvate
TPI=Triose-phosphate
PGK=
kinase,
PC=
isomerise,
Phosphoglycerate
Pyruvate
kinase,
carboxylase,
PEPCK=Phosphoenolpyruvate carboxykinase, LDH=Lactate dehydrogenase, and PDH=Pyruvate
dehydrogenase.
17
involved have not been completely established70. In individuals with T2D, hepatic
gluconeogenesis is increased and glycolysis is decreased, and this results in elevated
hepatic glucose production2,5.
In addition to the TCA and glycogen pathway, glycolysis/gluconeogenesis
also fuels several other metabolic pathways such as the pentose phosphate pathway
(PPP), the hexose monophosphate pathway, and de novo triacylglycerol synthesis.
Glucose-6-phosphate is the precursor for glycogen synthesis and the PPP, which
synthesises pyrimidines71. Dihydroxyacetone phosphate (DHAP) is the precursor for
glycerol and glycerolipid synthesis72. Disruption to the glycolytic/gluconeogenic
pathway can have a knock on effect that can impair the metabolic pathways
associated with glycolysis/gluconeogenesis.
Several enzymes in the glycolysis/gluconeogenesis pathway are involved
only in glycolysis or gluconeogenesis, however most enzymes are involved in both
glycolysis and gluconeogenesis.
Glyceraldehyde 3-phosphate dehydrogenase
(GAPDH) is one such enzyme involved in both glycolysis and gluconeogenesis, and
could play a role in regulating the balance between these pathways (Figure 1.5).
1.4.1 Glyceraldehyde 3-Phosphate Dehydrogenase
(GAPDH)
GAPDH is a highly expressed protein that catalyses the reversible reaction
of glyceraldehyde 3-phosphate and NAD+ to D-glycerate 1,3-bisphosphate
and
NADH during glycolysis, and vice versa for gluconeogenesis (Figure 1.5)63,73.
GAPDH protein levels generally remain constant for specific tissues, so it is possible
that GAPDH can be altered by PTMs in order to determine GAPDH function and
activity, and the direction it catalyses glycolysis/gluconeogenesis.
GAPDH is not traditionally known as a glycolytic/gluconeogenic rate
18
limiting enzyme, and is commonly thought of as simply a housekeeping gene,
however new research is now revealing that GAPDH is involved in multiple cellular
processes other than metabolism. GAPDH is involved in apoptosis, DNA repair,
endocytosis, microtubule bundling, membrane transport, and several other cellular
processes74. PTMs have been found to play a key role in regulating the various
functions of GAPDH75-78. For example; acetylation of lysine residues on GAPDH
that are regulated by substrate availability allow it to interact dynamically with
microtubules that facilitate membrane transport, and as a transcription factor for S
phase activation77,79. Phosphorylation of a tyrosine residue on GAPDH enables
GAPDH to interact with Rab2, allowing its transport between the golgi apparatus and
the endoplasmic reticulum73.
The subcellular localisation of GAPDH can be
mediated by PTMs, and is a key factor in determining the diverse functions of
GAPDH75-78. Recent research has shown that P300/CBP-associated factor (PCAF)
mediated lysine acetylation of human GAPDH K117, K227 and K251 induces
GAPDH Siah1 binding and translocation to the nucleus77,79. When GAPDH is
localised in the nucleus it can act as a transcriptional activator, and can also mediate
apoptosis77. Recently it has been demonstrated in bacteria that metabolism can be
regulated by the acetylation of metabolic enzymes22. The glycolytic/gluconeogenic
activity of mammalian GAPDH could also be regulated by acetylation.
1.4.2 Regulation of Glycolysis/Gluconeogenesis by GAPDH
Acetylation
Gluconeogenesis has been shown to be regulated at a transcriptional level,
and until recently there has been little consideration for other factors that may acutely
control pathway activity. Insulin is known to initiate phosphorylation of proteins
such as protein kinase B-2 (Akt-2), glycogen synthase kinase-3 (GSK3), and forkhead
19
box protein O (FOXO), which in turn stimulates glycogenesis, and inhibits the
expression of genes that encode gluconeogenic metabolic proteins such as PEPCK
and G-6-Pase (Figure 1.3)31,32.
In the liver, Protein kinase A (PKA) can activate
liver-type pyruvate kinase (L-PK) by phosphorylation, and also regulates glucose flux
by regulating the transcription of the L-PK gene in response to nutrient availability80.
The transcription factor carbohydrate responsive element binding protein (ChREBP),
which is required for L-PK transcription, is activated by high glucose and inhibited by
cAMP, where it is phosphorylated by PKA and de-phosphorylated by protein
phosphatase 2 (PP2A)80. When ChREBP is phosphorylated in response to glucose, it
is shuttled into the nucleus allowing the transcription of L-PK, which increases
carbohydrate metabolism80. However, transcriptional regulation occurs over several
hours, and therefore is unlikely to account for the acute metabolic changes that
occur in response to rapid fluctuations in nutrient availability and external stimuli22,47.
GAPDH protein and mRNA levels can vary between tissues but remain constant within
those tissues, therefore GAPDH activity is likely to be regulated by other mechanisms
such as its subcellular location and PTMs81. Acetylation could provide a novel
mechanism for regulating mammalian GAPDH function and activity.
There are numerous rapid biochemical processes
that regulate protein
activity, such as allosteric regulation and PTMs. PTMs generally involves an
enzyme that facilitates the addition or removal of a chemical moiety, thereby
changing the properties of that protein, influencing interactions with substrates and
regulatory proteins49. Tripodi et al, mapped acetylated and ubiquinated enzymes
involved in central carbon metabolism and identified that the glycolytic pathway has
the highest concentration of modified lysine residues67. Recent studies have shown
that a large number of acetylated proteins in both prokaryotes and eukaryotes are
20
metabolic enzymes, and that nearly all metabolic enzymes can be acetylated21,42.
Gluconeogenic enzymes such as GAPDH could be likely candidates for the
regulation of glycolysis/gluconeogenesis by reversible lysine acetylation. Denu et
al demonstrated that acetylation of the glycolytic enzyme phosphoglycerate mutase 1
(PGM) increased its glycolytic activity, and that under glucose restriction, SIRT 1
levels increased leading to deacetylation of PGM and a reduction in PGM glycolytic
activity82.
It has been demonstrated in S.enterica that acetylation of GAPDH is
regulated by nutrient availability, which determines the direction of the
glycolytic/gluconeogenic metabolic pathway19,22. They demonstrated that when cells
were supplemented with glucose, acetylation was elevated, and when supplemented
in acetate or citrate rich conditions, acetylation was reduced19,22. This mechanism of
regulation by reversible acetylation has not been demonstrated in mammals or
hepatic tissues. It has been observed that human GAPDH is activated by K254
acetylation in response to high glucose83. Lei et al demonstrated that human GAPDH
K254 was acetylated by PCAF and deacetylated by HDAC5 in vitro, and that
deacetylation of K254 reduced tumour growth in vivo83. However, it has recently been
demonstrated that HDAC5 does not possess deacetylase activity84,85. There is now a
large body of evidence that suggests that the class IIa HDACs do not possess
deacetylase activity, and recent studies have demonstrated that the class IIa HDACs
deacetylase activity is actually via the recruitment of other deacetylases84,85. HDAC5
has been shown to interact with the HDAC3 complex to initiate deacetylation of target
substrates, so therefore HDAC5 could deacetylate mammalian GAPDH via the
recruitment of the HDAC3 complex84,85. Acetylation of additional GAPDH lysine
residues could play a role in regulating GAPDH activity, and their role in glucose
21
metabolism has not been investigated.
The role of mammalian GAPDH acetylation in hepatic glucose metabolism is
incomplete, in particular its role in T2D. In addition, the KDACs and KATs
responsible for the deacetylation of GAPDH in mammals remains speculative, and
it is presently controversial as to whether high levels of acetyl-CoA facilitates
acetylation by means of an equilibrium reaction independent of KATs49.
1.4.3 Regulation of GAPDH Acetylation
Previous research has demonstrated that reversible acetylation of metabolic
enzymes, in particular GAPDH, is largely regulated by substrate availability. It has
been shown in bacteria that GAPDH acetylation increases in response to elevated
glucose, and that deacetylation occurs in response to low glucose and increased acetate
or citrate19,83. These changes in substrates could then determine the activity of specific
KATs and KDACs that are responsible for the acetylation and deacetylation of
metabolic enzymes such as GAPDH.
KATs and KDACs are a family of enzymes involved in acetylation and
deacetylation respectively. These enzymes are grouped into classes depending on their
sub-cellular localisation, known functions, and substrate reactivity. Changes in NAD+
and Zn+ substrate availability determines KDAC activity and is directly linked to
metabolism, creating a feedback loop.
GAPDH
that is participating in
glycolysis/gluconeogenesis is predominantly found in the cytosol, and therefore
acetylation of metabolically active GAPDH could be regulated by cytosolic KATs
and KDACs25,86,87. SIRT2 and HDAC6 are the KDACs that are predominantly
localised in the cytoplasm, although the majority of KDACs can shuttle between the
nucleus, cytosol, and organelles25,86,87. In previous research, HDAC6 has been shown
22
to regulate glucocorticoid-induced hepatic gluconeogenesis, by mediating the
glucocorticoid receptor88.
Recent in vitro studies have suggested that human GAPDH K254 is
deacetylated by HDAC5 in response to low glucose83. However, this was not shown
in vivo, and several studies have demonstrated that the class IIa HDACs (HDAC4, 5,
7, and 9) do not possess deacetylation activity84,85. These studies also demonstrated
that HDAC4 and HDAC5 bind to the HDAC3 N-CoR/SMRT complex via their Cterminal domains to initiate deacetylase activity through HDAC384,85. Therefore
HDAC5 could regulate mammalian GAPDH deacetylation, but this would only occur
via the recruitment of the HDAC3 N-CoR/SMRT complex. Furthermore, the activity
of KATs, KDACs, and the acetylation profile of human GAPDH in T2D is unknown.
1.4.4 GAPDH Acetylation in T2D
There is growing potential for the use of KDAC inhibitors to be used in T2D
to treat elevated blood glucose levels. Previous research has demonstrated that during
fasting or calorie restriction SIRT1 is activated and induces hepatic gluconeogenesis
by increasing the transcription of PEPCK, G-6-Pase, and FBPase by deacetylation of
PGC-1, FOXO1, and signal transducer and activator of transcription 3 (STAT3)89. It
has also been demonstrated that hepatic glucose production, and fasting blood glucose
levels, are reduced in SIRT1 anti-sense oligonucleotide rats, providing rationale for
KDAC inhibitors to treat T2D90.
If the acetylation profile of rate limiting metabolic enzymes determines the
direction of glycolysis and gluconeogenesis, then it is also possible that altered
acetylation profiles of these enzymes may contribute to diseases such as T2D
(Figure 1.6). The enzymes that are responsible for the acetylation/deacetylation of
particular metabolic enzymes could be manipulated to
23
rescue the altered
acetylation profiles that might occur in T2D.
If KDACs that deacetylate
gluconeogenic enzymes can be identified, and deacetylation of these enzymes is
shown to increase hepatic glucose production, then specific KDAC inhibitors could
be developed to reduce blood glucose levels in T2D.
1.5 SUMMARY
Impaired hepatic glucose metabolism in T2D is characterised by insulin
resistance and excessive hepatic de novo production of glucose, contributing to
elevated blood glucose levels in T2D.
Dysregulated hepatic metabolism as a
result of insulin resistance, is the primary source of excessive glucose production
that contributes to the elevated blood glucose levels seen in T2D. The mechanisms
that underlie elevated hepatic gluconeogenesis and glucose production in T2D are
not completely understood. If hepatic glucose production in T2D could be reduced,
then it could lower blood glucose levels, and alleviate many of the serious
complications associated with chronic hyperglycemia.
Glycolysis/gluconeogenesis occurs in the cytosol of cells, and therefore
the proteins and enzymes involved are subject to changes in cellular conditions
within the cytosol63,73. Organisms must quickly adapt to these fluctuations in
nutrient availability and environmental variability. PTMs such as lysine
acetylation are now recognised as regulators of multiple cellular processes,
including metabolism, and possibly hepatic glucose production vi a
gluconeogenesis4 5 , 9 1 . Reversible lysine acetylation influences metabolic processes
independent of transcription, and has emerged as a regulator of several cellular
processes such as cell motility, protein degradation, and recently, metabolic enzyme
activity91. Disruption to these PTM signalling pathways, and alterations to the
24
Figure 1.6: Proposed mechanism of human GAPDH regulation by acetylation. Acetylated GAPDH
increases glycolysis, removing glucose from the blood.
Deacetylated GAPDH increases
gluconeogenesis and increases blood glucose levels, which could play a role in the dysfunctional
metabolism that occurs in T2D.
acetylation profile of proteins, could potentially play a role in diseases such as
T2D. Acetylation could be as relevant as phosphorylation and it is likely that
acetylation, along with other PTMs, acts collectively to regulate a wide range of
protein functions and cellular processes48. Advances in molecular techniques will
enable more precise methods for investigating acetylation, and it is likely that there
is cross talk between lysine acetylation and other modifications such as
phosphorylation, which provides an even more complex picture23,92,93.
GAPDH is a multidimensional protein with diverse functions, that could play
25
a role in regulating glycolysis/gluconeogenesis, and there has been little research
done on mammalian GAPDH acetylation and its role in hepatic glucose metabolism.
It has also been shown that there is a link between nutrient availability, acetylation
and metabolic regulation50. However further studies are required to demonstrate
that these mechanisms of metabolic regulation occur in mammals in vivo. The
acetylation profile of metabolic enzymes in disease, such as GAPDH, also remain
unidentified. This project investigated whether altered acetylation profiles of the
metabolic enzyme GAPDH are contributing to increased gluconeogenesis in T2D.
26
CHAPTER 2
THE ROLE OF GAPDH LYSINE RESIDUES IN THE
REGULATION OF GLUCOSE METABOLISM
2.1 INTRODUCTION
T2D is inadequately controlled by the current drugs available, and with
increasing diagnoses of T2D, additional methods of pharmacological intervention are
required94. The aim of this thesis is to identify novel metabolic regulatory mechanisms
in the glycolytic/gluconeogenic pathway that are disrupted in T2D, which could aid in
the development of new pharmacological interventions. GAPDH is a highly expressed
metabolic enzyme, which catalyses glycolytic and gluconeogenic reactions. Previous
studies have shown in bacteria that high glucose supplemented media stimulated
GAPDH acetylation and increased glycolysis19,22. It was also demonstrated in bacteria
that citrate or acetate supplemented media decreased acetylation of GAPDH, thereby
increasing gluconeogenesis19,22. Further work is required to determine whether this
mechanism of metabolic regulation by reversible GAPDH acetylation also occurs in
mammalian hepatocytes. In addition, there are twenty six lysine residues on GAPDH,
and further research is required to determine which lysine sites undergo acetylation
and play a role in regulating glycolysis/gluconeogenesis.
Recent proteomic studies using mouse and human liver have demonstrated
that mouse GAPDH K115, K160, K225, and K252 (human K117, K162, K227 and
K254) can undergo reversible acetylation77,79,83. Lei et al demonstrated that human
GAPDH K254 (equivalent of rat GAPDH K252, Figure 2.1.1) is acetylated acutely in
response to high glucose in vitro, and that GAPDH K254Q in vivo reduced tumour
growth, where lysine residues mutated to glutamine have been traditionally used in
acetylation biology to mimic acetylated lysine83. Other studies have demonstrated that
27
acetylation of rat GAPDH K225 and human GAPDH K117, K227, and K251
(equivalent of rat K115, K225, and K249 respectively, Figure 2.1.1) regulate GAPDH
nuclear translocation and apoptosis77,79. However these studies have not linked
acetylation of rat GAPDH K115, K160, and K225 to glycolysis/gluconeogenesis.
A
B
Figure 2.1.1: Rat, mouse, and human GAPDH protein sequence and pairwise alignment95. (A)
Rat, mouse and human GAPDH protein sequence sourced from NCBI data bank. Rat GAPDH (333aa)
accession: NP_058704.1. Mouse GAPDH (333aa) accession: AAH83149.1. Human GAPDH (335aa)
accession: NP_001276674.1. (B) Pairwise alignment generated with Clustal.
Therefore the aim of this study was to determine whether rat GAPDH
acetylation sites K115, K160, K225, and K252 play a metabolic regulatory role in rat
hepatocytes. Human hepatic GAPDH crystallography has shown that amino acid
28
residues 1-151 and 315-335 are part of the GAPDH NAD+ binding domain, and that
residues 152-314 make up the GAPDH catalytic domain95. Therefore rat GAPDH
lysine residues (K115R, K160R, K225R, and K252R), that have been identified in
proteomic studies, are located in both the NAD+ binding domain and the catalytic
domain19-21. This suggests that the catalytic activity of mammalian GAPDH, and
the ability of mammalian GAPDH to bind NAD+, could be disrupted by mutation of
these lysine residues.
29
2.2 METHODS
2.2.1 Cloning
All plasmids were generated by GenScript, USA, using rat GAPDH in either
pIRES-Hyg3 (Clontech, #631620), pIRES-Neo (Clonetech, #631621), or pEGFP-N1
(Clontech, #6085-1) vectors.
Clones (1μl) were transformed using Subcloning
Efficiency DH5α (Invitrogen, #18265-017) per manufactures instructions. Bacteria
were selected in either 25μg/ml kanamycin or 100μg/ml carbenicillin. DNA plasmid
purification was performed using a GeneJET plasmid miniprep kit (Thermo Scientific,
#K0503) per manufactures instructions.
Single mutations to rat GAPDH residues K115, K160, K225, and K252 to
arginine were generated, and in addition a plasmid was generated that had all four rat
GAPDH residues K115, K160, K225, and K252 mutated to arginine (4KR GAPDH).
Mutation of lysine to arginine has been traditionally used in acetylation biology to
mimic deacetylated lysine. Mutation of lysine to glutamine has been traditionally used
in acetylation biology to mimic acetylated lysine.
Rat GAPDH (accession: NP_058704.1), mouse GAPDH (accession:
AAH83149.1), and human GAPDH (accession: NP_001276674.1) sequences were
sourced from the National Center for Biotechnology Information. Pairwise alignment
was generated using Clustal with the Jalview software program96.
2.2.2 Cell culture
Rat FAO hepatoma cells were used in these studies, as they are a well
characterised in vitro model of immortalised mammalian hepatocytes that have been
frequently used in metabolic studies97. FAO hepatocytes are particularly useful for
investigating the regulation of gluconeogenesis, because glucose production is not
generated from glycogenolysis, since the glycogen content of FAO cells is extremely
30
low under basal conditions97. Rat FAO immortalised hepatoma hepatocytes and
monkey fibroblast-like immortalised kidney cells were grown in 75cm2 flasks at 37°C
and 5% CO2. FAO cells were grown in RPMI 1640 (Life Technologies, #11875),
supplemented with 10% FBS. COS7 cells were grown in DMEM (Life Technologies,
#11995-065) supplemented with 10% FBS. Cells were transfected with the plasmid
pIRES-Hyg3 with no tag, or pEGFP-N1 with N terminal GFP, using lipofectamine
LTX (Life Technologies, #11668019), and then selected in hygromycin 1:200 (70μM).
Transfected cell lines included GAPDH WT, single lysine to arginine mutants GAPDH
K115R, GAPDH K160R, GAPDH K225R, GAPDH K252R, and quad mutant
GAPDH K115, 160, 225, 252R (4KR; GenScript). Assays were carried out when cells
were ‫׽‬ͺͲǦ90% confluent.
2.2.3 Protein extraction
Cells were harvested in 500μl of lysis buffer containing SIRT and HDAC
inhibitors (see appendix A), and homogenised (Kema Keur Pro 200) at 33,000 rpm for
20 seconds. Samples were centrifuged at 12,000 rpm for 10 minutes at 4°C, and the
supernatant was collected and stored at -80°C. Total protein was determined by
Bicinchoninic Acid Assay (BCA – Pierce, #23225) with a xMarkTM spectrophotometer
(BioRad).
2.2.4 Immunoprecipitation
Immunoprecipitation (IP) was performed by diluting 1000μg of protein in
lysis buffer to a total volume of 500μl, and adding 1μg of anti-GAPDH antibody (Cell
Signalling Technology, #2118). The antibody-protein lysate was then incubated at
4°C on a rotating wheel for 24 hours. 100μl of 50% protein-A agarose beads (Thermo
Scientific, #20333) was added to antibody-protein lysate and incubated at room
temperature on a rotating wheel for 4 hours. The protein-antibody-bead complex was
31
then centrifuged at 13,000 rpm for 2 minutes, and the supernatant was collected and
stored at -80°C. The beads were then washed in IP buffer (see appendix A), and
centrifuged at 13,000 rpm for 1 minute and the supernatant discarded. This step was
repeated three times. The beads were then washed in milli-Q water and centrifuged at
13,000 rpm, and the supernatant was discarded.
35μl of 2X loading buffer
(ThermoFisher, #39000) was added to the beads, then heated at 95°C for 5 minutes,
and run on SDS-PAGE.
2.2.5 Immunoblotting
SDS-PAGE was conducted using 1.5mm thick mini 8% Tris-HCl separating
gels (see appendix A) in 1X running buffer (see appendix A), at 100V for 2 hours. A
PageRuler pre-stained protein ladder (Thermo Scientific, #SM0671) was used to
determine molecular weight. Gels were transferred to a polyvinylidene difluoride
membrane (PVDF) in 1X Transfer buffer (see appendix A), at 100 V for 100 minutes
at 4°C. PVDF membranes were then blocked in blocking solution (1% BSA in TBST)
for 1 hour, and probed for 24 hours at 4°C with primary antibody (1:1000). PVDF
membranes were then washed in TBST and probed for 1 hour with secondary antibody
(1:10,000).
Blots were imaged using a Chemidoc XRS (BioRad) and
chemiluminescence (Invitrogen #WP20005).
2.2.6 Confocal imaging
Transfected FAO and COS7 cells were transferred to 35mm glass bottom
culture dishes (MatTek, #P35G-0-14-C). Four hours later, cells were washed in PBS
and stained in 1ml of media with DAPI (300nm), for 1 hour. Live cells were visualized
at 37°C using a FluoView FV10i laser scanning confocal microscope (Olympus) at
X500. DAPI was visualized at an excitation wavelength of 350nm, and an emission
32
wavelength of 470nm. EGFP was visualized at an excitation wavelength of 488nm,
and an emission wavelength of 509nm.
2.2.7 Computational structural imaging
GAPDH structure and lysine residue locations were generated by Dr Victor
Streltsov, CSIRO Manufacturing Flagship, Parkville, Australia. GAPDH structure
was simulated from the closest structure by sequence on the protein data bank (PDB,
http://www.rcsb.org/pdb/home/home.do), the structure of 1J0X ‘Crystal structure of
rabbit muscle glyceraldehyde-3-phosphate dehydrogenase’.
2.2.8 Metabolic profiling
The cellular bioenergetics profile of FAO hepatocytes was assessed using a
Seahorse XF24 Flux Analyser (Seahorse Bioscience). The XF24 Flux Analyser
measures the oxygen consumption rate (OCR), from which substrate dependent
oxidation, basal mitochondrial respiration, ATP turnover, uncoupled respiration,
maximum respiratory capacity, and spare respiratory capacity can be determined. The
XF24 Flux Analyser also measures the extra cellular acidification rate (ECAR), from
which basal and maximum glycolysis can be determined. Cells were seeded into a 24well XF24 cell culture microplate (Seahorse Bioscience) at a density of 50,000 cells
per well, and incubated for 24 hours. Cells were washed and incubated in 600μl of
unbuffered DMEM (containing 25mM glucose, 1mM pyruvate and 1mM glutamate)
pH 7.4, at 37 °C in a non-CO2 incubator (1 hour prior to bioenergetics assessment).
Six basal OCR measurements were performed using the Seahorse analyser, and three
measurements were repeated following injection of oligomycin (1μM), carbonyl
cyanide 4-(trifluoromethoxy) phenylhydrazone (FCCP, 1μM) and antimycin-A
(1μM). ECAR was determined from data collected at basal and oligomycin
measurement points. Following completion of the assay, cell number was determined
33
using the CyQuant® Cell Proliferation Assay kit (Molecular Probes) according to
manufacturer's instructions, or total protein was determined using BCA (Pierce,
#23225).
Analysis of glucose, glutamate, pyruvate and palmitate oxidation was
performed using 24-well Seahorse V7 plates that were seeded with FAO mutant cell
lines as specified above. Cells were washed and then resuspended in assay running
media for glucose oxidation (1 x KHB, 25mM glucose), glutamate oxidation (1 x
KHB, 2 X GlutaMAX, Life Technologies), pyruvate oxidation (1 x KHB, 2mM
pyruvate), and palmitate oxidation (1 x KHB, 200μM palmitate, 500μM carnitine).
Basal oxidative energetics were established after ten measurements.
Following
completion of the assay, cell number was determined using the CyQuant® Cell
Proliferation Assay kit (Molecular Probes) according to manufacturer's instructions,
or total protein was determined using BCA (Pierce, #23225).
2.2.9 Glucose production assay
All treatments of transfected hepatocytes were for 24 or 36 hours in glucose
and serum-free RPMI 1640 media supplemented with 2mM sodium pyruvate, 20mM
sodium l-lactate and 0.1% BSA (glucose production media; GP media). Cells were
treated with either insulin (0.1nM) or forskolin (20μM). Forskolin was used as a
glucagon mimetic in vitro, as FAO cells have resistance to glucagon98. Glucose
produced by the cells was measured by collecting 40μl of media from each well, which
was then combined with 250μl of 0.12M NaH2PO4·2H2O pH 7.0 assay buffer,
(1mg/mL phenol, 0.5mg/mL 4-aminoantipyrine, 1.6 units/mL peroxidise, and
10 units/mL glucose oxidase). This solution was incubated for 25 minutes at 37 °C
and then absorbance was measured at 490nm. Cells were lysed in 100μl 0.03% SDS
and protein was quantified using BCA (Pierce, #23225).
34
2.2.10 GAPDH activity
A colourmetric assay was used to measure GAPDH activity based on the
oxidation of NADH to NAD in the gluconeogenic direction, and vice versa for the
glycolytic direction (Figure 2.2.1). NADH absorbance is measured at 340nm which
declines over time in the gluconeogenic direction, and increases over time in the
glycolytic direction. Protein was extracted from transfected cells (GFP tagged) in lysis
buffer supplied with GAPDH activity assay kit (ScienCell, #8148), and gluconeogenic
GAPDH analysis was performed following the manufactures instructions. Glycolytic
GAPDH activity was measured using the ScienCell assay kit, substituting the
gluconeogenic reagents with the glycolytic precursor’s glyceraldehyde-3phosphate
(50mM), sodium-arsenate (1M), and nicotinamide adenine dinucleotide (NAD+,
10mM).
GLYCOLYTIC ACIVITY
3-GAP
3-GAP
NAD +
ADP
NAD +
ADP
GAPDH
GFP
NADH +
ATP
NADH +
ATP
1,3-BPG
1,3-BPG
3-PGK
3-PGA
GLUCONEOGENIC ACTIVITY
Figure 2.2.1: GAPDH activity assay pathway. The glycolytic/gluconeogenic direction for the
GAPDH activity assay is determined by substrate availability. For every cycle of glycolysis two units
of NADH are produced, and for every cycle of gluconeogenesis two units of NADH are oxidised to 2
units of NAD. NADH absorbance is measured at 340nm.
35
Measurements were made every 3 minutes up to 15 minutes, and then at 60
minutes, at 340nm using a xMarkTM spectrophotometer (BioRad). GAPDH activity
was calculated by normalising to GAPDH quantity determined by GFP tag
fluorescence (excitation wavelength of 488nm, and an emission wavelength of
509nm), and to a GAPDH standard curve using recombinant GAPDH (ScienCell,
#8148h).
2.2.11 Oil Red-O
Cells were grown to confluence in 6-well plates, then washed and fixed in 4%
paraformaldehyde for 1 hour at room temperature. Cells were then washed in 60%
isopropanol and stained with Oil Red-O (see appendix A) for 1 hour. Cells were then
washed in milli-Q water three times and imaged at X200 on an axioskop 2 microscope
(Zeiss). Quantification of neutral lipids was determined using ImageJ following nature
protocols, Imaging of neutral lipids by oil red O for analyzing the metabolic status in
health and disease, 2013, section 15B99.
2.2.12 RT-PCR
RNA was extracted from rat GAPDH mutant cell lines using RNeasy columns
(Qiagen, #74104) according to the manufacturer’s instructions. cDNA was generated
from 1μg of RNA using SuperScript III Reverse Transcription Kit (Life Technologies,
#18080093) according to the manufacturer’s instructions. Gene expression levels
were quantitated using SYBR Green PCR Master Mix (Life Technologies, #4385610),
using the primers listed below. Primer sequences were designed on the National
Center
for
Biotechnology
information
(http://www.ncbi.nlm.nih.gov/tools/primer-blast/) as follows:
36
(NCBI)
website
glucose-6-phosphatase:
glucokinase:
PEPCK:
(FWD) 5’-CTTGTGGTTGGGATACTGG-3’
(REV) 5’-GAGGCTGGCATTGTAGATG-3’
(FWD) 5’-GAGCAGAAGGGAACAACATC-3’
(REV) 5’-TGGCGGTCTTCATAGTAGC-3’
(FWD) 5’-AGCCATGTGCAACTCATGCA-3’
(REV) 5’-CTCGGTGCCACCTGAAACA-3’
Changes in expression were normalized to an Oligreen standard curve (Life
Technologies, #O7582) as per manufacturer’s instructions. 5μl of each standard and
sample was run in triplicate on a black 96 well fluorescence plate, scanned at excitation
485nm and emission 538nm, using a FlexStation II-384 (Molecular Devices).
2.2.13 Statistical analyses
Standard curves were generated using Microsoft Excel 2013, and statistical
analysis was conducted using SPSS 20 and MiniTab 16. A Kolmogorov-Smirnov test
was performed to determine normality and distribution. For data that was not normally
distributed, a non-parametric test (Kruskal–Wallis test) was used to compare means.
To compare means for normally distributed data a two tailed T-Test was used, where
equal variance was assumed if Levene’s significance was greater than 0.05. For
normally distributed data, one-way and two-way ANNOVA was used to compare
groups using Games Howell method (equal variances not assumed), or Fisher's Least
Significant Difference method (equal variances assumed). Results are presented as
mean ± SEM, and P < 0.05 was considered statistically significant.
37
2.3 RESULTS
2.3.1 GAPDH residues K115, K160, K225, and K252 are positioned on the surface
of GAPDH. The structure of GAPDH was generated from crystallography of rabbit
muscle GAPDH.
Human 1U8F overlayed with rat 2VYN and rabbit 1J0X (from
PDB) structures shows that the rat, rabbit and human GAPDH structures are very
similar. Some differences between rat, rabbit and human are mainly in NAD+ binding
domains at residues 1-151 with RSM distance matches of 3.3A, while catalytic
domains 152-314 align very well with RMS of 0.35A. Residues corresponding to 160,
225, and 252 in rat and rabbit GAPDH are at the same positions as in human. Lys 225
is at the entrance to the cleft formed by the tetramer. Rat GAPDH lysine residues 225
and 252 form part of the catalytic region of GAPDH, and are exposed on the surface
where they could play an important role in protein-protein interactions (Figure 2.3.1).
2.3.2 Total GAPDH acetylation is not reduced in GAPDH mutants. Total GAPDH
acetylation and expression in FAO stable transfected pIRES-Hyg3 GAPDH mutant
cells, was determined by immunoprecipitation and immunoblotting. FAO stable
transfected pIRES-Hyg3 rat GAPDH mutant cells were used for all experiments in this
chapter, excluding the GAPDH activity assay, where FAO transient transfection with
pEGFP-N1 rat GAPDH mutants were used.
Immunoblotting of GAPDH stable
mutants showed that GAPDH expression was not changed relative to α-tubulin
expression (Figure 2.3.2 A). Total GAPDH acetylation was unchanged between nontransfected FAO, and across all FAO stable transfected mutants (Figure 2.3.2 B). This
highlights the importance of using site specific acetyl-lysine antibodies, and has
implications for interpreting acetyl-lysine blots using pan acetyl-lysine antibodies.
2.3.3 WT GAPDH and 4KR GAPDH are localised in the cytoplasm. Previous
studies have showed that acetylation of GAPDH can regulate its cellular localisation75-
38
78,100,101
. To determine whether mutation of lysine residues found to be acetylated in
human liver to arginine alters GAPDH localisation, WT GAPDH-GFP and 4KR
GAPDH-GFP expressing FAO and COS7 cells were stained with DAPI (300nM), for
1 hour and visualised at X500 using a confocal microscope. WT GAPDH and 4KR
GAPDH were predominantly located throughout the cytoplasm not the nucleus (Figure
2.3.3).
2.3.4
Lysine
residues
that
are
acetylated
determine
GAPDH
glycolytic/gluconeogenic activity in vitro. To determine whether mutation of lysine
residues alters GAPDH activity, a colourmetric assay with substrates limiting GAPDH
activity to either the glycolytic or gluconeogenic direction was performed. FAO cells
were transiently transfected with pEGFP-N1 GAPDH mutants, so that GAPDH
activity could be normalised to GFP fluorescence to represent units of GAPDH per
well. Glycolytic GAPDH activity was reduced in K115R, K160R, K225R, K252R,
and 4KR GAPDH mutant FAO cells compared with WT GAPDH FAO cells (Figure
2.3.4 B). Gluconeogenic GAPDH activity was increased in K225R, K252R, and 4KR
GAPDH mutant FAO cells compared with K115R, K160R, and WT GAPDH FAO
cells (Figure 2.3.4 C).
2.3.5 Basal glucose production is increased in K225R, K252R and 4KR mutant
GAPDH cell lines. To determine whether glucose production is altered in GAPDH
lysine mutants, cells were seeded into 48 well plates and grown in GP media for 24
hours. Consistent with the GAPDH activity in these mutants, FAO GAPDH mutant
cells K225R, K252R and 4KR produced more glucose than FAO WT cells and FAO
mutant cells K115R and K160R (Figure 2.3.5).
2.3.6 Glycolytic flux is reduced in K225R, K252R and 4KR mutant GAPDH cell
lines. To determine whether reduced GAPDH acetylation impairs glycolysis, ECAR
39
was measured to assess glycolytic flux. Consistent with the reciprocal nature of
glycolysis/gluconeogenesis, ECAR was significantly reduced in the K225R, K252R
and 4KR mutant GAPDH FAO cells compared with K115R, K160R, and WT GAPDH
FAO cells (Figure 2.3.6).
2.3.7 Insulin suppresses glucose production and forskolin increases glucose
production in mutant GAPDH FAO cells. To determine whether hormonal control
of glucose production is altered in GAPDH mutants, cells were seeded into 48 well
plates and grown in GP media treated with insulin (0.1nM) or forskolin (20μM) for 36
hours. Insulin suppressed glucose production in all cell lines compared with basal WT
GAPDH glucose production (Figure 2.3.7). Forskolin increased glucose production
in all cell lines compared with basal WT GAPDH glucose production (Figure 2.3.7).
Cells expressing 4KR GAPDH were hypersensitive to forskolin, resulting in a 2-fold
increase in glucose production compared with WT GAPDH cells that had been treated
with forskolin (Figure 2.3.7).
Increases in gluconeogenesis and decreases in glycolysis appeared to be
maximised in the 4KR GAPDH mutants compared to the single mutation mutants,
where the combination of GAPDH residues K225 and K252 in the catalytic region,
and K115 in the NAD+ binding domain, collectively impair glycolysis and increase
gluconeogenesis. Based on the results to this point, additional characterisation was
performed only in WT GAPDH and 4KR GAPDH mutant cell lines.
2.3.8 Basal and maximum glycolysis is reduced in 4KR GAPDH cell lines. Data
generated to date in this chapter suggests that GAPDH could be rate limiting for
glycolytic/gluconeogenic flux. This is in contrast to dogma in the literature suggesting
that expression of key gluconeogenic genes is rate limiting. Therefore, basal and
maximal glycolytic flux was assessed to determine whether basal and maximum
40
glycolytic capacity is altered in 4KR GAPDH mutants, where ECAR was measured
basally and in the presence of oligomycin (1μM).
Basal ECAR and maximal
glycolytic capacity were reduced in 4KR GAPDH FAO cells compared with WT
GAPDH FAO cells (Figure 2.3.8 A). Maximal glycolytic capacity was significantly
greater than basal glycolytic rates in WT GAPDH cells, while there was no difference
between basal and maximal glycolytic rate in 4KR GAPDH cells (Figure 2.3.8 A).
To determine whether these effects are independent of transcriptional
regulation of key gluconeogenic genes, the mRNA expression of traditional rate
limiting enzymes phosphoenolpyruvate carboxykinase (PEPCK), glucose 6phosphatase (G-6-Pase), and glucokinase (GK) was measured using qRT-PCR. There
was no significant change in PEPCK, G-6-Pase and GK expression levels between
WT GAPDH and 4KR GAPDH mutant FAO cells (Figure 2.3.8 B, C, and D).
2.3.9 Mitochondrial respiration is unchanged in 4KR GAPDH mutant FAO cells.
The oxygen consumption rate (OCR) was measured in the presence of mitochondrial
inhibitors oligomycin, FCCP, and antimycin A, to determine overall mitochondrial
function in WT and 4KR GAPDH cells. OCR was unchanged in 4KR GAPDH
compared with WT GAPDH (Figure 2.3.9 A). Basal mitochondrial respiration, ATP
turnover, uncoupled respiration (proton leak), maximal mitochondrial respiratory
capacity, and spare respiratory capacity was unchanged in 4KR GAPDH mutant FAO
cells compared with WT GAPDH cells (Figure 2.3.9 B).
2.3.10 Glucose and palmitate dependent oxidation is impaired in 4KR GAPDH
mutant FAO cells. To determine whether substrate dependent oxidation is altered in
4KR GAPDH mutants, either pyruvate, palmitate, glutamine or glucose in KHB buffer
were used to determine specific substrate dependent OCR in FAO mutant cells.
Glucose dependent OCR was reduced in 4KR GAPDH mutant cells compared with
41
WT GAPDH cells (Figure 2.3.10 A). Palmitate dependent OCR was reduced in 4KR
GAPDH mutant cells compared with WT GAPDH cells (Figure 2.3.10 B). Glutaminedependent OCR was unchanged in 4KR GAPDH mutant cells compared with WT
GAPDH (Figure 2.3.10 C). As the end product of glycolysis, pyruvate is also a key
substrate for mitochondrial oxidative metabolism.
Oxidation of pyruvate was
unchanged in WT GAPDH and 4KR GAPDH cells (Figure 2.3.10 D).
2.3.11 Lipid accumulation is increased in 4KR GAPDH mutant FAO cells. To
determine whether the reduction in palmitate oxidation was associated with lipid
accumulation with 4KR GAPDH, Oil Red-O stain was used to visualise neutral lipids
in WT GAPDH and 4KR GAPDH cell lines (Figure 2.3.11 A). Lipid deposits were
greater in 4KR GAPDH FAO cells compared with WT GAPDH FAO cells, P<0.05
(Figure 2.3.11 B).
42
Figure 2.3.1: Crystal structure of one monomer of rabbit muscle glyceraldehyde-3-phosphate
dehydrogenase, generated from PDB. Monomer of GAPDH, which forms a tetramer, displaying the
NAD+ binding site and lysine residues 115, 160, 225, and 252.
43
A
GAPDH:
WT
K115R K160R K225R K252R 4KR
GAPDH
α-tubulin
B
nonGAPDH: transfect WT
K115R K160R K225R K252R 4KR
IP:GAPDH/WB:AcK
IP:GAPDH/WB:GAPDH
140
Total GAPDH AcK
(%non-transfected)
120
100
80
60
40
20
0
non-transfected WT GAPDH
K115R
K160R
K225R
K252R
4KR GAPDH
Figure 2.3.2: GAPDH expression and total GAPDH acetylation in WT and mutant GAPDH FAO
cells. (A) GAPDH expression in FAO hepatocytes stable transfected with pIRES-Hyg3 GAPDH
mutants. (B) GAPDH immunoprecipitation representing total GAPDH acetylation in FAO hepatocytes
stable transfected with pIRES-Hyg3 GAPDH mutants. Data represented as %non-transfected, mean ±
SEM, n=8 replicates per group. Results not statistically significant.
44
A
DNA
GAPDH
Bright field
Overlay
WT GAPDH
4KR GAPDH
B
DNA
GAPDH
Bright field
Overlay
WT GAPDH
4KR GAPDH
2.3.3: FAO WT GAPDH-GFP and FAO 4KR GAPDH-GFP confocal imaging (magnification
X500). (A) WT GAPDH-GFP FAO cells and 4KR GAPDH-GFP FAO cells were stained with DAPI
(300nM) and imaged by confocal microscopy at (magnification X500).
(B) WT GAPDH-GFP COS7
cells and 4KR GAPDH-GFP COS7 cells were stained with DAPI (300nM) and imaged by confocal
microscopy at (magnification X500).
45
A
GAPDH: FAO WT 115 160 225 252 4KR
GFP
GAPDH
α-tubulin
Glycolytic GAPDH activity
(%WT)
B
140
120
100
80
*
60
*
*
40
*
*
20
0
Gluconeogenic GAPDH activity
(%WT)
C
WT
115
160
225
252
4KR
700
*
600
500
*
*
225
252
400
300
200
100
0
WT
Figure 2.3.4:
115
160
4KR
FAO WT GAPDH-GFP and FAO mutant GAPDH-GFP activity.
(A) GFP
expression of pEGFP-N1 GAPDH mutants. (B) Glycolytic GAPDH activity. (C) Gluconeogenic
GAPDH activity. All data represented as %WT GAPDH activity. All data represented as mean ± SEM,
n=8 replicates per group. * Significantly different from WT GAPDH, P < 0.05.
46
250
*
Glucose production (%WT)
*
200
*
150
100
50
0
WT
115
160
225
252
4KR
Figure 2.3.5: FAO WT GAPDH and FAO mutant GAPDH glucose production assay. GAPDH
stable cell lines were grown in GP media for 36 hours, and total glucose produced was measured. All
data represented as %WT GAPDH, mean ± SEM, n=8 replicates per group. * Significantly different
from WT GAPDH, P < 0.05.
47
120
ECAR (%WT)
100
80
60
*
40
*
*
20
0
WT
115
160
225
252
4KR
Figure 2.3.6: FAO WT GAPDH and FAO mutant GAPDH extra cellular acidification rate
(ECAR). ECAR was measured in a Seahorse Bio-Analyser using DMEM running buffer. All data
represented as %WT GAPDH, mean ± SEM, n=8 replicates per group. * Significantly different from
WT GAPDH, P < 0.05.
48
Glucose production (%WT)
600
#
500
*
400
Fors 20uM
300
Ins 0.1nM
#
200
#
#
#
#
100
#
#
#
#
#
#
0
GAPDH WT
GAPDH K115R GAPDH K160R GAPDH K225R GAPDH K252R GAPDH 4KR
Figure 2.3.7: FAO WT GAPDH and FAO mutant GAPDH, insulin and forskolin treated glucose
production assay. WT GAPDH and mutant GAPDH cell lines were treated with either 20μM forskolin
or 0.1nM insulin in GP media, and glucose production was measured after 36 hours.
All data
represented as %WT GAPDH basal glucose production, mean ± SEM, n=8 replicates per group. #
Significantly different from WT GAPDH basal glucose production, P < 0.05. * Significantly different
from single GAPDH mutants, P < 0.05.
49
A
180
#
160
ECAR (%WT basal)
140
120
100
WT GAPDH
80
4KR GAPDH
*
*
60
40
20
0
basal glycolysis
C
1.2
D
1.6
1.0
0.8
0.6
0.4
0.2
0.0
WT
4KR
G-6-Pase mRNA (relative to
WT)
1.2
Glucokinase mRNA (relative
to WT)
PEPCK mRNA (relative to
WT)
B
max glycolysis
1.0
0.8
0.6
0.4
0.2
0.0
WT
4KR
1.4
1.2
1.0
0.8
0.6
0.4
0.2
0.0
WT
4KR
Figure 2.3.8: FAO WT GAPDH and FAO 4KR GAPDH basal and maximum glycolysis and
mRNA expression. (A) Basal glycolysis was determined by averaging six measurements of basal
ECAR, and maximum glycolysis was determined by averaging three measurements in the presence of
oligomycin (1μM). Data represented as %WT GAPDH, mean ± SEM, n=10 replicates per group. *
Significantly different from WT GAPDH, P < 0.05. # Significantly different from WT GAPDH basal
glycolysis, P < 0.05. (B) PEPCK mRNA expression normalised to WT GAPDH, (C) Glucokinase
mRNA expression normalised to WT GAPDH, (D) G-6-Pase mRNA expression normalised to WT
GAPDH. All data represented as mean ± SEM, n=10 replicates per group.
50
A
120
OCR (%WT)
100
80
60
40
20
0
WT GAPDH
B
4KR GAPDH
OCR (pmoles/min/mg protein)
2500
2000
1500
WT GAPDH
1000
4KR GAPDH
500
0
basal respiration
ATP turn over
uncoupled
respiration
max respiratory spare respiratory
capacity
capacity
Figure 2.3.9: FAO WT GAPDH and FAO 4KR GAPDH OCR and mitochondrial respiration.
(A) Oxygen consumption rate. (B) Mitochondrial energetics. All data represented as mean ± SEM,
n=10 replicates per group. Results not statistically significant.
51
B
120
Palmitate dependent OCR (%WT)
Glucose dependant OCR (%WT)
A
100
80
*
60
40
20
WT
80
*
60
40
20
4KR
D
120
Pyruvate dependent OCR (%WT)
Glutamate dependent OCR (%WT)
100
0
0
C
120
100
80
60
40
20
0
WT
4KR
WT
4KR
120
100
80
60
40
20
0
WT
4KR
Figure 2.3.10: FAO WT GAPDH and FAO 4KR GAPDH substrate dependent OCR. (A) Glucose
dependent OCR, (B) Palmitate dependent OCR, (C) Glutamate dependent OCR, (D) Pyruvate
dependent OCR. All data represented as %WT GAPDH. All data represented as mean ± SEM, n=10
replicates per group. * Significantly different from WT GAPDH, P < 0.05.
52
A
WT GAPDH
4KR GAPDH
B
600
*
Oil Red-O (%WT)
500
400
300
200
100
0
WT GAPDH
4KR GAPDH
Figure 2.3.11: FAO WT GAPDH and FAO 4KR GAPDH Oil Red-O stain (Magnification X200).
(A) WT GAPDH and 4KR GAPDH cells were fixed and stained with Oil Red-O. (B) Densitometry
quantification of neutral lipids, P<0.05. n=6 replicates per group.
53
2.4 DISCUSSION
Reversible acetylation of metabolic enzymes is now considered to play a key
role in the regulation of metabolism22,45,58,102. Reversible GAPDH acetylation in
bacteria has been shown to regulate glycolysis/gluconeogenesis based on substrate
availability, however the lysine residues involved are not completely known 19,22.
GAPDH lysine residues have also been shown to be acetylated in the liver of humans,
thus the purpose of this study was to determine the role of these GAPDH lysine
residues on aspects of hepatocyte metabolism. Based on previous studies, rat GAPDH
residues K115, K160, K225 and K252 that have been shown to undergo reversible
acetylation, were mutated to arginine to mimic deacetylation19-21,79. This study showed
that mutation of lysine residues on GAPDH to arginine, increased gluconeogenic
GAPDH activity and glucose production, and reduced ECAR and glycolytic GAPDH
activity.
In addition, palmitate and glucose oxidation was impaired, and lipid
accumulation was higher in 4KR GAPDH. These results occurred independent of
transcriptional regulation.
The total acetylation of the 4KR GAPDH and single lysine GAPDH
mutations was not reduced compared to the WT GAPDH mutants (Figure 2.3.2 B).
This suggests that measuring total GAPDH acetylation using pan acetyl-lysine
antibodies is not representative of the four lysine residues that have been mutated. Site
specific acetyl-lysine antibodies are therefore required to accurately determine the
acetylation status of rat GAPDH residues K115, K160, K225, and K252. Site specific
antibodies are currently being generated but were not available for submission of this
thesis. In addition, a point of consideration is that lysine residues undergo several
different types of acyl PTMs, which could be affected by mutation to arginine, and this
needs to be considered when interpreting the data64.
54
The sub-cellular localisation of GAPDH has been shown to play an important
role in determining the functional properties of GAPDH78. GAPDH that participates
in transcription and apoptosis is located in the nucleus and GAPDH that participates
in glycolysis/gluconeogenesis is located in the cytosol78. Previous research has shown
that acetylation of GAPDH can determine its subcellular localisation. Ventura et al
demonstrated that acetylation of human GAPDH residues K117, K227, and K249
(equivalent of rat K115, K225, and K249, Figure 2.1.1) in response to apoptotic stimuli
are required for GAPDH translocation to the nucleus, where acetylation of K160
initiates apoptosis73,76,79. Using confocal microscopy, WT GAPDH and 4KR GAPDH
were primarily located in the cytosol (Figure 2.3.3), demonstrating that 4KR GAPDH
was still localised to the region of the cell where the glycolytic/gluconeogenic
processes occur. The cytosolic localisation of GAPDH appears to be perinuclear and
requires further investigation to determine the nature of this GAPDH cytosolic
localisation.
GAPDH activity in the gluconeogenic direction was increased in K225R,
K252R, and 4KR GAPDH mutants compared with WT GAPDH (Figure 2.3.4 C). This
correlated with the ECAR and GP data that showed a decrease in glycolysis and an
increase in glucose production, respectively, in the K225R, K252R, and 4KR GAPDH
mutants (Figures 2.3.5 and 2.3.6). This suggests that rat GAPDH K225 and K252
acetylation plays a regulatory role in glycolysis/gluconeogenesis. GAPDH activity in
the glycolytic direction was reduced in all mutant GAPDH cell lines compared with
WT GAPDH (Figure 2.3.4 B). This does not correspond with the ECAR data, where
ECAR was not reduced in K115R and K160R mutants. This could be explained by
increased contribution to ECAR from other metabolic pathways, possibly by the TCA
55
cycle which contributes a proton each cycle, thereby contributing to cellular
acidification and therefore ECAR.
Chronic measurement of glucose production showed that K225R, K252R,
and 4KR GAPDH mutants produced more glucose than WT GAPDH (Figure 2.3.5).
The rat GAPDH K115R and K160R GAPDH mutants had no effect on glucose
production compared with WT GAPDH (Figure 2.3.5). In addition, ECAR was
reduced in K225R, K252R, and 4KR GAPDH mutants compared with WT GAPDH
(Figure 2.3.6). ECAR was not reduced in K115R and K160R mutations compared
with WT GAPDH (Figure 2.3.6). Together these results indicate that reversible
acetylation of GAPDH K225 and K252 could play a more significant
glycolytic/gluconeogenic regulatory role then GAPDH K115 and K160.
These findings support the hypothesis that site-specific acetylation of
GAPDH drives glycolysis, and that deacetylated GAPDH drives gluconeogenesis.
Further investigation into the structural changes that occur to GAPDH when these
lysine residues are acetylated/deacetylated could reveal whether these effects alter the
binding of GAPDH with intermediary metabolite substrates. Jenkins and Tanner
showed through crystallography of human hepatic GAPDH that amino acid residues
1-151 and 315-335 are part of the GAPDH NAD+ binding domain, and that residues
152-314 make up the GAPDH catalytic domain103. Therefore residue K115 is within
the NAD+ binding domain and residues K160, K225 and K252 are within the catalytic
domain, so therefore the 4KR GAPDH mutant has mutations within the NAD+ and
catalytic binding domains. Combined with the findings in this study, this suggests that
rat GAPDH K225 and K252, which are located in the catalytic domain and on the
surface of GAPDH, are important for the regulation of glycolysis/gluconeogenesis.
Furthermore, rat GAPDH K225 lies within the cleft formed by the GAPDH tetramer,
56
which appears to have important properties for protein-protein binding. Rat GAPDH
K225 has already been implicated in Siah-1 binding, and alterations to cleft shape and
charge following acetylation/deacetylation of rat GAPDH K225 coupled with K252
could play a role in the interaction of GAPDH with intermediary metabolites77.
Further research will be required to determine whether K225 and K252 acetylation
favours binding of glyceraldehyde 3-phosphate and glycolysis, and whether K225 and
K252 deacetylation favours binding to
D-glycerate 1,3-bisphosphate and
gluconeogenesis. The results so far indicate that the combination of deacetylated lysine
residues in the 4KR GAPDH mutant amplifies the effects of glucose production and
ECAR that are seen in the single mutations of K115R, K160R, K225R and K252R.
This could suggest that multiple deacetylated residues increase the disruption to the
NAD+ and catalytic binding domains. Further research which focuses on the 4KR
mutant is likely to provide greater insight into characterising the effect of reduced
GAPDH acetylation on hepatocyte metabolism.
Glucose production was measured in the presence of insulin and forskolin to
determine the role of GAPDH lysine residues that can be acetylated on the hormonal
regulation of glycolysis/gluconeogenesis. Forskolin is a plant compound that activates
the cAMP pathway, and was used as a glucagon mimetic, as FAO cells are resistant to
glucagon104. Insulin was still able to suppress elevated glucose production in the 4KR
GAPDH mutant, presumably through transcriptional regulation or protein degradation.
However, the 4KR GAPDH mutant was hypersensitive to forskolin compared to WT
GAPDH, increasing glucose production 4-fold (Figure 2.3.7). The sensitivity to
forskolin could be due to the glycolytic inability of 4KR GAPDH, where WT GAPDH
could have both gluconeogenesis and glycolysis occurring alternatively over twenty
four hours, thereby reducing the amount of glucose produced. This is significant in
57
terms of T2D, as research has shown that increased glucagon action is the main driver
behind increased hepatic glucose production105-107.
Basal ECAR reflects the rate at which protons are produced by metabolic
pathways, with the majority of protons produced by glycolysis, therefore ECAR is a
proxy measure of glycolysis. Basal glycolysis was reduced in the 4KR GAPDH
mutants compared with WT GAPDH (Figure 2.3.8 A). Maximum glycolytic capacity
is the maximum rate of lactate produced from glycolysis when the mitochondrial ATP
synthase is inhibited. WT GAPDH had an increase in maximal glycolytic capacity
compared with basal glycolysis, while there was no difference between basal and
maximal glycolysis in 4KR GAPDH (Figure 2.3.8 A). The 4KR GAPDH mutants also
had reduced maximum glycolytic capacity compared with WT GAPDH (Figure 2.3.8
A). This result further indicates that glycolysis is impaired in the 4KR GAPDH
mutant, and suggests that lysine modification of GAPDH is rate limiting for
glycolysis/gluconeogenesis.
Another factor to take into account is the shift in ATP demand and synthesis.
One cycle of the glycolytic pathway generates two ATP, and one cycle of the
gluconeogenesis pathway requires two units of ATP to produce one unit of glucose.
The reduction in glycolysis and the increase in gluconeogenesis in 4KR GAPDH
suggests that 4KR GAPDH cells have an increased demand for ATP compared with
WT GAPDH cells. Therefore other metabolic pathways must be compensating for the
increased demand for ATP that is required to facilitate the increase in gluconeogenesis,
and the reduction in ATP that is generated from glycolysis. Measuring the OCR and
mitochondrial flux showed that there was no change in OCR and basal mitochondrial
respiration between 4KR GAPDH and WT GAPDH (Figures 2.3.9 A and B). ATP
turn over, uncoupled respiration, maximum mitochondrial respiration, and spare
58
respiratory capacity were also unchanged (Figure 2.3.9 B). This data demonstrates
that a reduction in GAPDH acetylation does not have a detrimental effect on
mitochondrial function and oxidative phosphorylation. Oxidative phosphorylation is
the most likely source of ATP to replace the ATP required to support the increased
rate of gluconeogenesis in 4KR GAPDH mutants, however OCR was not increased in
4KR GAPDH cells (Figure 2.3.9 A). Alternative glucose pathways, such as the
pentose phosphate pathway, could be compensating for the increased demand for ATP
by the 4KR GAPDH mutants.
Oil Red-O staining showed that lipid deposits were increased in the 4KR
GAPDH mutants compared with WT GAPDH (Figure 2.3.11). This demonstrates that
there could be either a reduction in lipid oxidation, or an increase in lipid synthesis, or
a combination of both. A reduction in lipid oxidation could be a result of TCA cycle
cataplerosis used to drive gluconeogenesis, reducing intermediary metabolites in the
TCA cycle, which would have a knock on effect limiting TCA cycle reactions and
ultimately β-oxidation. An increase in lipogenesis could also result from glucose being
shunted from glycolysis and GAPDH to dihydroxyacetone phosphate, and into
glycerol, then onto glycerolipids (Figure 2.4.1).
There are metabolic reactions
involved in palmitate oxidation that require ATP to proceed, which could be consumed
by the gluconeogenic pathway in the 4KR GAPDH cells, thereby reducing the ability
of those cells to metabolise palmitate. The increased demand for ATP in 4KR GAPDH
hepatocytes could be depriving other metabolic pathways of ATP, such as palmitate
metabolism, thereby altering or impairing those pathways. Further investigation using
glucose tracers and metabolomics will be required to confirm these mechanisms.
Substrate dependent OCR was used to further characterise metabolic
alterations in 4KR GAPDH cells. Glucose-dependent OCR was reduced in 4KR
59
GAPDH mutants compared with WT GAPDH (Figure 2.3.10 A). This suggests that
the inability of the 4KR GAPDH mutant to proceed in the glycolytic direction has
reduced the amount of intermediary metabolites below GAPDH, such as pyruvate and
acetyl-CoA, causing a reduction in OCR. This can be rescued by substituting glucose
with pyruvate, as pyruvate-dependent OCR was not different between WT and 4KR
GAPDH expressing cells (Figure 2.3.10 D). This provides further evidence that
metabolic function downstream of acetyl-CoA is normal provided there are precursor
substrates available (Figure 2.4.1). However, under normal conditions where pyruvate
Glucose
glycerol
GAPDH
Dihydroxyacetone-P
glycerolipids
Phosphophenolpyruvate
Fatty acid acyl CoA
Acetyl-CoA
Oxaloacetate
α-ketogluterate
Glutamine
Figure 2.4.1: 4KR GAPDH metabolic pathway hypothesis. 4KR GAPDH is blocked in the
glycolytic direction but gluconeogenesis is increased. Lipid metabolism is impaired in 4KR GAPDH
cells.
is not present in supra-physiological amounts, and other substrates are available,
pyruvate OCR in 4KR GAPDH cells could be decreased, as the available pyruvate
60
would likely be required to fuel the increased rate of gluconeogenesis. Further
demonstration that the TCA cycle is functional downstream of acetyl-CoA was shown
where glutamine-dependent OCR was unchanged between 4KR GAPDH and WT
GAPDH (Figure 2.3.10 C). Palmitate-dependent OCR is reduced in 4KR GAPDH
mutants compared with WT GAPDH (Figure 2.3.10 B), which could help explain the
increase in lipids in the 4KR GAPDH mutants. A limitation to this method of
measuring substrate dependent oxidation though is the contribution by endogenous
substrates to the OCR.
To confirm that the increase in gluconeogenesis and decrease in glycolysis is
not attributed to transcriptional regulation of rate limiting enzymes, the mRNA of key
gluconeogenic and glycolytic genes PEPCK, G-6-Pase, and Glucokinase was
determined, and found to be unchanged between 4KR GAPDH and WT GAPDH
(Figure 2.3.8 B, C, and D).
These results indicate that transcription of key
gluconeogenic genes is not affected by GAPDH acetylation, and suggests that the
increase in glucose production and subsequent decrease in glycolysis occurs due to
altered enzymatic activity independent of transcriptional regulation.
In conclusion these results suggest that mutation of four rat GAPDH lysine
residues K115, K160, K225, and K252 to arginine, which cannot be acetylated, leads
to a reduction in glycolysis and subsequent increase in gluconeogenesis independent
of transcriptional regulation. The results suggest that there is no change in oxidative
phosphorylation in the 4KR GAPDH mutant that could be compensating for the
elevated ATP demands required for increased gluconeogenesis. The results also
suggest that lipid metabolism is impaired in the 4KR GAPDH, resulting in increased
lipid accumulation. Glucose/lipid tracer studies and metabolomics would be required
to confirm some of the findings above, in particular to determine where lipid
61
metabolism is impaired, and why lipid accumulates in the 4KR GAPDH mutants.
Tracer studies could also determine the fate of glucose in the 4KR GAPDH mutants.
62
CHAPTER 3
MEDIATORS OF GAPDH
ACETYLATION/DEACETYLATION
3.1 INTRODUCTION
Previous research has established in bacteria that deacetylation of GAPDH
regulates glycolysis in response to nutrient availability, but how GAPDH is
deacetylated in mammals is not completely known19,58,83,108.
The KDACs that
deacetylate mammalian hepatic GAPDH, and the mechanisms that regulate those
KDACs, are currently unknown. The KATs add acetyl groups to lysine residues, and
the KDACs; Sirtuins (SIRTs) and Histone Deacetylases (HDACs), remove acetyl
groups. There are currently eighteen known KDACs and twenty two KATs that have
been identified in human and mouse cells64. The KDACs consist of two distinct
families, the seven NAD+ dependent SIRTs and the eleven Zn2+ dependent HDACs,
which together are known to deacetylate in excess of two thousand proteins25. The
ability of these moonlighting proteins to deacetylate multiple proteins can make them
a difficult therapeutical target, however there has been some success with KDAC
inhibitors and cancer87,109.
The KATs and KDACs of mammalian hepatic GAPDH lysine residues K115,
K160, K225, and K252 are unknown. Previous research has established that human
GAPDH K254 (equivalent of rat K252, Figure 2.1.1) is acetylated by P300/CBPassociated factor (PCAF) and deacetylated by HDAC5 in response to glucose
signalling in vitro, however it is not known whether this mechanism occurs in vivo,
and if PCAF/HDAC5 also acetylates/deacetylates other GAPDH lysine residues83. In
addition, it is now well established that the class IIa HDACs do not possess deacetylase
activity, and actually recruit other KDACs in order to deacetylate proteins84,85. If
63
HDAC5 does regulate the deacetylation of human GAPDH K254 it could do this via
the recruitment of the HDAC3-SMRT/N-CoR complex, however this multi-protein
complex is largely nuclear84,85.
HDACs are a family of eleven deacetylases that are predominantly located in
the nucleus, however HDAC6 is located primarily in the cytoplasm outside of the
mitochondria, where glycolysis/gluconeogenesis occurs110. HDAC6 is a class IIb
HDAC that is localised to the cytosol, and is highly expressed in the heart and the
gluconeogenic tissues liver and kidney, with the highest expression of HDAC6
occurring in the liver88,110,111.
HDAC6 is known to play a significant role in
microtubule-dependent cell motility via deacetylation of tubulin, and is also important
for the degradation of misfolded proteins110,111. HDAC6 localisation in the cytosol
makes it an ideal candidate for GAPDH deacetylation, as GAPDH is predominantly
found in the cytosol, which is the location of glycolysis/gluconeogenesis110-112. This
suggests that HDAC6 could be a potential regulator of glycolysis/gluconeogenesis via
the deacetylation of metabolic enzymes. SIRTs have also been shown to play an
important role in metabolism, in addition, SIRT1 and SIRT2 have cytosolic
localisation and high liver expression15,82,89,90.
In particular, SIRT1 has been
implicated in hepatic glucose production, where SIRT1 knockdown in liver decreases
basal hepatic glucose production and increases hepatic insulin responsiveness in
diabetic rats90. SIRT2 also deacetylates known HDAC6 protein targets such as αtubulin, which suggests that GAPDH could also have multiple deacetylases113. Studies
in this thesis have focused on HDAC6 as a potential GAPDH deacetylase, however
further work is currently underway to determine whether SIRT1 and SIRT2 are
GAPDH deacetylases.
64
Previous research has established that HDAC6 plays a regulatory role in
metabolism88,114-117.
Kintsher et al demonstrated that HDAC6 modified
glucocorticoid-induced hepatic gluconeogenesis by regulating glucocorticoid receptor
nuclear translocation88. It has also been demonstrated that HDAC6 is required for
glucocorticoid-mediated hepatic gene expression88. Therefore the first aim of this
study was to determine whether HDAC6 deacetylase activity regulates
glycolysis/gluconeogenesis, and whether regulation occurs via the deacetylation of
GAPDH.
By using a dominant negative form of HDAC6 (DN HDAC6) with no
deacetylase activity, and a wild type (WT HDAC6), the effect of the deacetylase
activity of HDAC6 on metabolism can be investigated. If HDAC6 is involved in
GAPDH deacetylation, HDAC6 could be implicated in the dysregulation of
metabolism via the deacetylation of GAPDH, resulting in impaired blood glucose
homeostasis and hyperglycaemia in T2D. Specific HDAC6 inhibitors could then be
used in combination with metformin to further reduce blood glucose levels.
For the last fifty years metformin has been the most widely used drug to treat
T2D, but its mechanism of action is controversial, and a source of ongoing debate118.
The current theories for the potential molecular mechanism of metformin action
include; inhibition of the mitochondrial complex I, activation of AMPK, inhibition of
mitochondrial glycerophosphate dehydrogenase, inhibition of glucagon-induced
elevation of cAMP and consequent activation of PKA119-122. Metformin reduces
hepatic glucose production, thereby reducing blood glucose levels in patients with
T2D. However, additional medication is often required, and not all patients can
tolerate metformin. Therefore the second aim of this study was to determine whether
metformin action in 4KR GAPDH cells is impaired in relation to WT GAPDH cells.
65
3.2 METHODS
3.2.1 Liver protein extraction, immunoprecipitation, and immunoblotting
All protein extraction, immunoprecipitation, and immunoblotting were
performed as outlined in Chapter Two.
3.2.2 Cell culture
Rat FAO hepatocytes were grown in 75cm2 flasks at 37°C and 5% CO2. FAO
cells were grown in RPMI 1640 (Life Technologies, #11875), supplemented with 10%
FBS. Assays were carried out when cells were ‫׽‬ͺͲǦ90% confluent.
3.2.3 DN HDAC6 construct
Histidine residues 216 in deacetylase domain one, and 611 in deacetylase
domain two, were mutated to alanine to remove HDAC6 deacetylase activity to
generate dominant negative HDAC6 (Figure 3.2.1).
pEGFP.N1-HDAC6.DC
(catalytic deficient DN HDAC6) was a gift from Tso-Pang Yao (Addgene, #36189)123.
pEGFP.N1-HDAC6 (WT HDAC6) was a gift from Tso-Pang Yao (Addgene,
#36188)123.
Deacetylase activity
NES
Nuclear
exclusion
DD1
DD2
216 bp:
H→A
Cytoplasmic
Anchoring
SE14
611 bp:
H→A
ZnFUBP
Ubiquitin
Binding.
Figure 3.2.1: DN HDAC6 construct. Histidine residues 216 in deacetylase domain one, and 611 in
deacetylase domain two, were mutated to alanine to remove HDAC6 deacetylase activity. NES =
nuclear export signal, DD1 = deacetylase domain 1, DD2 = deacetylase domain 2, SE14 = Ser/Glu
tetradecapeptide domain, ZnF-UBP = zinc-finger ubiquitin binding domain.
66
3.2.4 Transfection
FAO cells were stably transfected and grown in RPMI 1640 supplemented
with 10% FBS (Sigma-Aldrich), and 1:200 hygromycin (70μM) or 1:200 neomycin
(70μM). Cells were incubated at 37°C and 5% CO2. Cells were transfected with the
plasmid pIRES-Hyg3 (Clontech, #631620), or pIRES-neo3 (Clontech, #631621),
using lipofectamine LTX (Life Technologies), and then selected in hygromycin or
neomycin. Transfected cell lines included WT HDAC6, DN HDAC6, WT GAPDH,
single lysine mutants GAPDH K115R, GAPDH K160R, GAPDH K225R, GAPDH
K252R, and quad mutant 4KR GAPDH (Genscript).
3.2.5 Glucose production assay
All treatments of hepatocytes were for 24 or 36 hours in glucose and serumfree RPMI 1640 media supplemented with 2mM sodium pyruvate, 20mM sodium llactate and 0.1% BSA (GP media). Glucose produced by the cells was measured by
collecting 40μl of media from each well, which was then combined with 250μl of
0.12M NaH2PO4·2H2O pH 7.0 assay buffer (1mg/mL phenol, 0.5mg/mL 4aminoantipyrine, 1.6 units/mL peroxidise and 10 units/mL glucose oxidase). This was
incubated for 25 minutes at 37 °C and the absorbance was measured at 490nm. Cells
were lysed in 100μl 0.03% SDS, and protein was quantified using BCA (Pierce,
#23225). Results are expressed as μg glucose per mg protein. Cells were treated with
either insulin (0.1nM), forskolin (20μM), metformin (200μM), NAM (10mM), or TSA
(10μM).
3.2.6 Metabolic profiling
The cellular bioenergetics profile of transfected FAO hepatocytes was
assessed using a Seahorse XF24 Flux Analyser (Seahorse Bioscience). Cells were
67
seeded into a 24-well XF24 cell culture microplate (Seahorse Bioscience) at a density
of 50,000 cells per well, and incubated for 24 hours. Cells were washed and incubated
in 600μl of unbuffered DMEM (containing 25mM glucose, 1mM pyruvate and 1mM
glutamate) pH 7.4, at 37 °C in a non-CO2 incubator (1 hour prior to bioenergetics
assessment). Six basal OCR measurements were performed using the Seahorse
analyser, and measurements were repeated following injection of oligomycin (1μM),
FCCP (1μM) and antimycin-A (1μM). ECAR was determined from data collected at
basal measurement points. Following completion of the assay, cell number was
determined using the CyQuant® Cell Proliferation Assay kit (Molecular Probes)
according to manufacturer's instructions, or total protein was determined using BCA
(Pierce, #23225).
3.2.7 Statistical analyses
Standard curves were generated using Microsoft Excel 2013, and statistical
analysis was conducted using SPSS 20 and MiniTab 16. A Kolmogorov-Smirnov test
was done to determine normality and distribution. For data that was not normally
distributed, a non-parametric test (Kruskal–Wallis test) was used to compare means.
To compare means for normally distributed data a two tailed T-Test was used, where
equal variance was assumed if Levene’s significance > 0.05. For normally distributed
data, one-way and two-way ANOVA was used to compare groups using Games
Howell method (equal variances not assumed) or Fisher's Least Significant Difference
method (equal variances assumed). Results are presented as mean ± SEM, and P <
0.05 was considered significant.
68
3.3 RESULTS
3.3.1 Overexpression of HDAC6 reduces α-tubulin acetylation. HDAC6 is known
to deacetylate α-tubulin, therefore α-tubulin acetylation was used as a proxy measure
of HDAC6 expression and to determine whether DN HDAC6 deacetylase activity had
been inhibited. Over expression of WT HDAC6 reduced α-tubulin acetylation, while
α-tubulin acetylation in cells over expressing DN HDAC6 was similar to nontransfected cells (Figure 3.3.1).
3.3.2 KDAC inhibitors reduce glucose production in WT GAPDH FAO cells, but
not 4KR GAPDH FAO cells. Glucose production was measured after treatment with
KDAC inhibitors to determine whether HDACs, SIRTs, or both regulate glucose
production in a GAPDH dependent manner. WT GAPDH and 4KR GAPDH FAO
cells were treated in GP media with either vehicle (DMSO), NAM (10mM, SIRT
inhibitor), or TSA (10μM, HDAC inhibitor) for twenty four hours, and then the media
was assayed for glucose. Non-specific KDAC inhibitors, NAM and TSA, reduced
glucose production in WT GAPDH cells compared with vehicle, but did not
significantly reduce glucose production in 4KR GAPDH cells compared with vehicle
(Figure 3.3.2).
3.3.3 DN HDAC6 reduces glucose production and increases ECAR. TSA is a
HDAC inhibitor that could not reduce glucose production in 4KR GAPDH cells,
suggesting that HDACs could potentially deacetylate GAPDH.
HDAC6 was
investigated as a potential GAPDH deacetylase based on its subcellular localisation to
the cytosol, where GAPDH that participates in glycolysis/gluconeogenesis is located.
FAO cells were transfected with either WT HDAC6 or DN HDAC6, and then exposed
to GP media for 24 hours to determine whether HDAC6 has an effect on glucose
production. DN HDAC6 cells produced less glucose compared with WT HDAC6 cells
69
(Figure 3.3.3 A). Basal and maximum glycolysis was determined with a Seahorse BioAnalyser using oligomycin (1μM), where DN HDAC6 cells had increased basal and
maximum glycolysis compared with WT HDAC6 cells (Figure 3.3.3 B).
3.3.4 DN HDAC6 increases OCR, basal mitochondrial respiration, ATP turn
over, and uncoupled respiration. To determine whether HDAC6 can regulate
mitochondrial respiration and OCR, DN HDAC6 and WT HDAC6 cells were run on
a Seahorse Bio-Analyser. DN HDAC6 cells had elevated OCR compared with WT
HDAC6 (Figure 3.3.3 C).
Basal mitochondrial respiration, ATP turn over and
uncoupled respiration was increased in DN HDAC6 cells compared with WT HDAC6
cells (Figure 3.3.3 D).
3.3.5 DN HDAC6 reduces glucose production in WT GAPDH co-expressing cells,
but not 4KR GAPDH co-expressing cells. To determine whether HDAC6 could
affect glucose production through GAPDH deacetylation, WT and 4KR GAPDH FAO
cells were co-transfected with WT or DN HDAC6 plasmids, and assayed for glucose
production. Glucose production was reduced in DN HDAC6/WT GAPDH cells
compared with WT HDAC6/WT GAPDH cells (Figure 3.3.4). However, in DN
HDAC6/4KR GAPDH cells glucose production was increased compared with WT
HDAC6/WT GAPDH and DN HDAC6/WT GAPDH (Figure 3.3.4).
3.3.6 HDAC6 opposes the insulin response and promotes the forskolin response.
Glucose production was measured in GAPDH and HDAC6 co-expressing cells after
treatment with insulin and forskolin to determine whether DN HDAC6 could reduce
glucose production in 4KR GAPDH co-expressing cells. WT GAPDH/DN HDAC6
cells treated with insulin had reduced glucose production compared with WT
GAPDH/WT HDAC6 cells treated with insulin (Figure 3.3.5). WT GAPDH/DN
HDAC6 cells treated with forskolin had reduced glucose production compared with
70
WT GAPDH/WT HDAC6 cells treated with forskolin (Figure 3.3.5). WT and DN
HDAC6 cells expressing 4KR GAPDH treated with forskolin had elevated glucose
production compared with WT GAPDH/WT HDAC6 cells treated with forskolin
(Figure 3.3.5).
3.3.7 Total GAPDH acetylation was not altered by HDAC6 overexpression. Total
GAPDH acetylation was measured in WT HDAC6 and DN HDAC6 overexpressing
FAO cells to determine whether HDAC6 is a GAPDH deacetylase. Total GAPDH
acetylation was not altered by DN HDAC6 or WT HDAC6 compared with FAO
control (Figure 3.3.6).
3.3.8 Metformin reduced glucose production further in WT GAPDH cells
compared with 4KR GAPDH cells. WT and 4KR GAPDH FAO cells were treated
with metformin (200μM) in GP media for 24 hours to determine whether GAPDH
acetylation could further reduce glucose production in combination with metformin.
Metformin reduced glucose production in both WT GAPDH and 4KR GAPDH cells
compared with vehicle, independent of GAPDH acetylation (Figure 3.3.7). However,
glucose production was lower in WT GAPDH cells treated with metformin compared
with 4KR GAPDH cells treated with metformin (Figure 3.3.7).
71
WT HDAC6
DN HDAC6
CON
HDAC6
AcK-α-tubulin
α-tubulin
Figure 3.3.1: α-tubulin acetylation in WT HDAC6 and DN HDAC6 expressing FAO cells.
Immunoblot of HDAC6, α-tubulin and acetylated α-tubulin in FAO cells expressing WT HDAC6 and
DN HDAC6 compared to control FAO cells (CON).
72
WT GAPDH
Glucose production (μg glucose/mg protein)
300
4KR GAPDH
*
250
*
*
200
150
100
#
#
50
0
Vehicle
10mM NAM
10μM TSA
Figure 3.3.2: FAO WT GAPDH and FAO 4KR GAPDH KDAC inhibited GP assay. Glucose
production was measured in WT GAPDH FAO cells and 4KR GAPDH FAO cells after 24 hours, in the
presence of either the SIRT inhibitor NAM (10mM), or the HDAC inhibitor TSA (10μM). All data
represented as mean ± SEM, n=8 replicates per group. * Significantly different from WT GAPDH
under the same treatment, P < 0.05. # Significantly different from Vehicle, P < 0.05.
73
A
B
30
*
80
ECAR (mpH/min/50K cells)
Glucose production
(μg glucose/mg protein)
90
70
60
50
40
*
30
20
10
WT HDAC6
15
WT HDAC6
DN HDAC6
10
5
basal glycolysis
DN HDAC6
maximal glycolysis
D
300
400
*
250
OCR (pmoles/min/50K cells)
OCR (pmoles/min/50K cells)
*
20
0
0
C
25
200
150
100
50
350
300
250
200
150
*
*
WT HDAC6
DN HDAC6
100
50
*
0
0
WT HDAC6
basal
ATP turn uncoupled maximal
spare
respiration over respiration respiratory respiratory
capacity capacity
DN HDAC6
Figure 3.3.3: FAO WT HDAC6 and FAO DN HDAC6 energetics. (A) Glucose production was
measured in WT HDAC6 FAO cells and DN HDAC6 FAO cells after 24 hours, n=24 replicates per
group. (B) Basal glycolysis was determined by averaging six measurements of basal ECAR, and
maximum glycolysis was determined by averaging three measurements in the presence of oligomycin
(1μM), represented as ECAR per 50K cells, n=10 replicates per group. (C) OCR was determined by
averaging ten readings over 1.5 hours. Data represented as OCR per 50K cells, n=10 replicates per
group. (D) Mitochondrial energetics. Data represented as OCR per 50K cells, n=10 replicates per
group. All data represented as mean ± SEM. * Significantly different from WT HDAC6, P < 0.05.
74
160
*
Glucose production
(%WT GAPDH/WT HDAC6)
140
120
100
*
80
60
40
20
0
WT GAPDH/WT HDAC6
WT GAPDH/DN HDAC6
4KR GAPDH/DN HDAC6
Figure 3.3.4: HDAC6 mutant/GAPDH mutant co-expressing FAO cell GP assay. HDAC6/GAPDH
mutant FAO co-expressing cells were incubated in GP media for 24 hours and analysed for glucose
produced. Data represented as %WT GAPDH/WT HDAC6, mean ± SEM, n=12 replicates per group.
* Significantly different from WT HDAC6/WT GAPDH, P < 0.05.
75
Glucose production
(%WT GAPDH/WT HDAC6 basal GP)
250
*
#
200
*#
150
100
*
0.1nM insulin
#
50
#
20uM forskolin
#
†
#
#
0
Figure 3.3.5: HDAC6 mutant/GAPDH mutant co-expressing FAO cell forskolin and insulin
treated GP assay. HDAC6/GAPDH mutant co-expressing FAO cells were incubated in GP media with
either insulin (0.1nM) or forskolin (20μM) for 24 hours and analysed for glucose. All data represented
as %WT GAPDH/WT HDAC6 basal glucose production, mean ± SEM, n=6 replicates per group. #
Significantly different from WT HDAC6/WT GAPDH basal glucose production, P < 0.05.
†
Significantly different from WT GAPDH/WT HDAC6 insulin treatment, P < 0.05. * Significantly
different to WT GAPDH/ WT HDAC6 forskolin treatment, P < 0.05.
76
WT HDAC6
DN HDAC6
CON
IP:GAPDH/WB:AcK
IP:GAPDH/WB:GAPDH
Figure 3.3.6: GAPDH acetylation in WT HDAC6 and DN HDAC6 expressing FAO cells.
Immunoprecipitation of GAPDH blotted for acetyl-lysine and GAPDH from FAO cells expressing WT
HDAC6 and DN HDAC6 compared with control FAO cells (CON).
77
35.0
*
Glucose production (ug glucose/mg protein)
30.0
25.0
*
20.0
#
WT GAPDH
4KR GAPDH
15.0
#
10.0
5.0
0.0
Veh
200 uM MET
Figure 3.3.7: Metformin treated FAO WT GAPDH and FAO 4KR GAPDH GP assay. GAPDH
mutant FAO cells were incubated in GP media, with either metformin (MET, 200μM) or vehicle (Veh,
water), for 24 hours and analysed for glucose. All data represented as mean ± SEM, n=12 replicates per
group. * Significantly different from WT GAPDH, P < 0.05. # Significantly different from vehicle, P
< 0.05.
78
3.4 DISCUSSION
Acetylation has emerged as an important, and highly conserved regulatory
PTM, which has recently been shown in bacteria to play a regulatory role in metabolic
flux19. Chapter Two in this thesis demonstrated that mutation of GAPDH residues
K115, K160, K225, and K252 to arginine increased glucose production in vitro. The
acetylation and deacetylation of proteins is facilitated by KATs and KDACs
respectively, and the KATs and KDACs that acetylate and deacetylate mammalian
GAPDH are unknown. In humans there are currently eighteen known KDACs; seven
SIRTs and eleven HDACs. The KDACs are further broken down into four groups;
Class I (HDAC1, 2 and 3), Class IIa (HDAC4, 5, 7 and 9), Class IIb (HDAC6 and 10),
Class III (SIRT1 to 7) and Class IV (HDAC11)25. The majority of these KDACs have
the ability to shuttle between the nucleus and the cytosol, however HDAC6, HDAC10
and SIRT2 are primarily localised to the cytosol25.
The KATs and KDACs that acetylate and deacetylate mammalian hepatic
GAPDH lysine residues K115, K160, K225, and K252 are unknown. Lei et al
demonstrated that HDAC5 deacetylated human GAPDH K254 in vitro, however this
has not been demonstrated in vivo, and the deacetylase activity of HDAC5 is
speculative83. There are now several studies that have shown that the class IIa HDACs
do not have deacetylase activity, and actually deacetylate proteins via the recruitment
of other deacetylases, such as HDAC384,85. It has also been shown that in response to
glucagon, the class IIa HDACs in the liver are rapidly dephosphorylated and
translocated to the nucleus, where they influence glucose production by associating
with promoters of gluconeogenic enzymes124. It was demonstrated that the HDAC4/5
complex recruits HDAC3, resulting in transcriptional induction of genes via
deacetylation and activation of FOXO transcription factors124.
79
KDACs play a role in glucose metabolism via GAPDH deacetylation, as
demonstrated in this chapter, where the KDAC inhibitors NAM and TSA, reduced
glucose production in WT GAPDH cells but not in 4KR GAPDH cells (Figure 3.3.2).
HDAC and SIRT inhibition was unable to significantly reduce glucose production in
4KR GAPDH cells, suggesting that acetylation of K115, K160, K225, and K252 was
required for their impact on glucose metabolism (Figure 3.3.2). This establishes that
KDACs have the ability to influence glucose metabolism, and also demonstrates that
4KR GAPDH glucose production did not decrease in the presence of KDAC inhibitors,
suggesting that GAPDH is potentially deacetylated by both SIRTs and HDACs.
To determine likely candidates for the deacetylation of GAPDH, the cytosolic
localisation of HDAC6, HDAC10, and SIRT2 was considered, and suggests that they
could be GAPDH KDACs based on their sub-cellular location. The reduction in
glucose production in the presence of NAM and TSA suggests that both SIRTs and
HDACs are involved in glucose production, and of the eighteen known KDACs,
SIRT2 and HDAC6 are the predominant cytosolic KDACs. SIRT2 and HDAC6 have
been shown to deacetylate α-tubulin, and both could be GAPDH deacetylases111,113.
HDAC6 was investigated as it is a potential GAPDH deacetylase that has cytosolic
localisation, and has already been implicated in glucose metabolism as a modifier of
glucocorticoid-induced hepatic gluconeogenesis88.
HDAC6 has been shown to
regulate hepatic gluconeogenesis by regulating the glucocorticoid receptor, and has
been shown to be required for mitochondrial metabolic activity88,114,115,125. Therefore
HDAC6 was investigated in this study due to its cytosolic localisation, and previous
studies that have implicated HDAC6 in glucose production. Furthermore, the highest
expression of HDAC6 is in the liver, which further suggests that HDAC6 has a role in
hepatocyte functions88.
80
HDAC6 has been shown to deacetylate α-tubulin, and acetylation of α-tubulin
was increased in DN HDAC6 expressing cells (Figure 3.3.1), but total GAPDH
acetylation was not affected by DN HDAC6 expression (Figure 3.3.6)111. However,
site specific antibodies are required to determine whether the acetylation of the lysine
residues of interest are altered by DN HDAC6 expression. Based on these findings,
this study measured glycolysis and gluconeogenesis in FAO cells in response to
genetic manipulation of HDAC6 deacetylase activity. WT HDAC6 FAO cells had
elevated glucose production compared with DN HDAC6 FAO cells (Figure 3.3.3 A),
in addition, WT HDAC6 FAO cells had reduced basal and maximum glycolysis
compared with DN HDAC6 cells (Figure 3.3.3 B). This suggests that HDAC6 plays
a role in glycolysis/gluconeogenesis, possibly by deacetylating metabolic proteins,
resulting in an increase in gluconeogenesis and glycogenolysis, while also reducing
glycolysis. It is still unknown whether HDAC6 deacetylates GAPDH, however an
association between glucose metabolism and HDAC6 has been demonstrated88.
The OCR was increased in DN HDAC6 FAO cells compared with WT
HDAC6 cells (Figure 3.3.3 C), and basal mitochondrial respiration, ATP turnover, and
uncoupled respiration was also increased in the DN HDAC6 cells compared with WT
HDAC6 cells (Figure 3.3.3 D). This suggests that HDAC6 has a metabolic regulatory
effect, and demonstrates that HDAC6 can reduce mitochondrial respiration. The exact
mechanisms involved in the reduction of mitochondrial respiration by HDAC6
requires further investigation, but based on the results in this thesis it would suggest
that ATP and substrate availability is reduced, resulting in decreased flux through
glycolysis and the TCA cycle. Due to the small number of KDACs and the large
number of proteins that undergo deacetylation, it is probable that HDAC6 has several
81
proteins that it regulates directly, or indirectly via deacetylation, which could include
other glycolytic/gluconeogenic enzymes.
DN HDAC6 suppressed glucose production in cells co-expressing WT
GAPDH compared to WT HDAC6/WT GAPDH cells, but DN HDAC6 did not
suppress glucose production in cells co-expressing 4KR GAPDH (Figure 3.3.4).
These results suggest that there could be an association between HDAC6 and GAPDH,
where removing the deacetylase activity of HDAC6 reduces glucose production in WT
GAPDH cells but not in 4KR GAPDH cells. This suggests that HDAC6 either
regulates glucose production through GAPDH, or that DN HDAC6 is not able to
overcome metabolic alterations that occur in 4KR GAPDH cells. This further supports
Figure 3.3.2 where TSA and NAM had little effect on 4KR GAPDH glucose
production, suggesting that GAPDH is deacetylated by SIRTs and HDACs.
Conversely, HDAC6 could exert these effects independent of GAPDH.
Insulin could still suppress glucose production in HDAC6/GAPDH coexpressing cells, but 4KR GAPDH cells that co-expressed WT HDAC6 and DN
HDAC6 were still hypersensitive to forskolin (Figure 3.3.5). However, glucose
production was lower in WT GAPDH/DN HDAC6 insulin treated cells compared with
WT GAPDH/WT HDAC6 insulin treated cells, suggesting that HDAC6 opposes the
insulin response (Figure 3.3.5). DN HDAC6/WT GAPDH co-expressing cells had
reduced glucose production under forskolin treatment compared with WT
HDAC6/WT GAPDH co-expressing cells treated with forskolin (Figure 3.3.5). In
addition, WT and DN HDAC6 cells co-expressing 4KR GAPDH had significantly
higher glucose production in response to forskolin compared with WT GAPDH/WT
HDAC6 forskolin treated cells (Figure 3.3.5). The decrease in glucose production in
DN HDAC6 forskolin treated cells is lost in 4KR GAPDH co-expressing cells, which
82
suggests that the forskolin response mediated through HDAC6 occurs via GAPDH
deacetylation.
This study has demonstrated that mutation of GAPDH K115, K160, K225,
and K252 to arginine in 4KR GAPDH cells has a similar GP and ECAR phenotype to
FAO cells expressing WT HDAC6. This suggests that HDAC6 plays a regulatory role
in glycolysis/gluconeogenesis, potentially via GAPDH deacetylation. Further work is
required to determine the metabolic pathways and mechanisms that underlie the
metabolic regulation by HDAC6. It may be possible that there are more than one
deacetylase that have the potential to deacetylate GAPDH, which is supported by the
NAM and TSA suppressed glucose production data (Figure 3.3.2), and furthermore
the substrates that regulate these deacetylases require further investigation.
WT GAPDH and 4KR GAPDH FAO cells were treated with metformin for
twenty four hours in GP media to determine whether the efficacy of metformin was
reduced when GAPDH acetylation was reduced. Metformin was able to reduce
glucose production in both WT GAPDH cells and 4KR GAPDH cells, however
glucose production was lower in WT GAPDH cells compared with 4KR GAPDH cells
(Figure 3.3.7). This demonstrates that the effects of metformin are independent of
GAPDH acetylation, and that there could be additional therapeutic value in targeting
mechanisms that regulate GAPDH acetylation in combination with metformin. If this
can be shown in vivo then GAPDH deacetylase inhibitors could be used in combination
with metformin to further lower blood glucose levels in T2D.
These results suggest that HDAC6 regulates glucose metabolism, and further
elucidation of the role of GAPDH acetylation and HDAC6 in hepatic glucose
metabolism could uncover this signalling axis as a novel therapeutic target for the
treatment of T2D in combination with metformin.
83
HDAC6 activity in T2D is
unknown, and it would be interesting to see whether HDAC6 activity is increased in
T2D and contributing to elevated gluconeogenesis. However, HDAC6 deacetylates
several proteins so there must be other factors that interact with HDAC6, which would
allow HDAC6 deacetylase activity towards GAPDH to be increased while HDAC6
deacetylase activity to other proteins, such as α-tubulin, remains normal. Further
investigation is required to determine whether GAPDH is specifically deacetylated by
HDAC6, and if so, which lysine residues are deacetylated by HDAC6. Site specific
antibodies and stable isotope labelling of amino acids in cell culture (SILAC) could be
used to determine which lysine residues are deacetylated under different conditions or
treatments. Further work is also required to determine how GAPDH is acetylated,
whether it is enzymatically acetylated or if it is an un-catalysed mass action reaction.
If it is enzymatically acetylated, then the KAT(s) responsible could be identified, and
then inhibited, or activated, in combination with a KDAC inhibitor to treat T2D.
In conclusion, the enzyme(s) responsible for the deacetylation/acetylation of
GAPDH
are
still
unknown.
However,
WT
HDAC6
can
regulate
glycolysis/gluconeogenesis similar to the regulation seen by 4KR GAPDH, and is a
potential GAPDH deacetylase, although this requires further investigation. Metformin
can reduce glucose production further in vitro when GAPDH is acetylated, and this
now needs to be investigated in vivo. Together these results support the notion that
HDAC6 could deacetylate GAPDH, and that reduced GAPDH acetylation increases
gluconeogenesis.
84
CHAPTER 4
REGULATION OF GAPDH ACETYLATION
4.1 INTRODUCTION
Obesity can cause insulin resistance that results in elevated gluconeogenesis
and hepatic glucose production, which contributes to chronic hyperglycemia and
T2D29,126. This has now become a global challenge, and the pathophysiology behind
the dysregulated metabolism that occurs in T2D is incompletely understood. The
acetylation status of metabolic enzymes in disease is largely unknown, but it is now
emerging as a potential and promising therapeutic target25. Previous studies in this
thesis have shown that mutation of GAPDH lysine residues to arginine, mimicking a
deacetylated state, results in an increase in gluconeogenesis in vitro. Results from
Chapter Two demonstrated that GAPDH is a rate limiting enzyme for
glycolysis/gluconeogenesis in FAO cells through reversible acetylation. WT GAPDH
favoured glycolysis and 4KR GAPDH, which mimics GAPDH deacetylation at
residues K115, K160, K225, and K252, favoured gluconeogenesis. In addition, the
acetylation status of GAPDH in FAO cells appears to play a role in lipid metabolism.
Results from Chapter Three demonstrated that SIRTs and HDACs are potential
GAPDH deacetylases, and that HDAC6 regulates glycolysis/gluconeogenesis,
possibly via deacetylation of GAPDH.
It has been shown in bacteria that GAPDH acetylation is regulated by KDACs
and KATs in response to nutrient availability19,22,57. Bacterial GAPDH acetylation is
increased in response to an increase in glucose, and GAPDH acetylation is reduced in
response to reduced glucose and increased acetate/citrate19,22,57. The factors that
regulate GAPDH acetylation in mammals is unclear, but previous research using
85
human GAPDH in vitro has shown that acetylation of GAPDH K254 increased in
response to an increase in glucose83. Li et al demonstrated in HEK 293T cells that
acetylation of human GAPDH K254 increased after exposure to high glucose (25mM)
compared to low glucose (1.5mM) over several hours (6-24 hours)83. The regulation
of hepatic GAPDH in vivo is unknown, and it is unclear whether the physiological
factors that regulate glucose production in vivo can also regulate hepatic GAPDH
acetylation. Previous research has demonstrated that in response to nitric oxide human
GAPDH lysine residues 117, 227, and 251 become acetylated79. This results in Snitrosylation of GAPDH and Siah-1 binding, abolishing its catalytic activity, which
causes GAPDH translocation to the nucleus where it can induce apoptosis77,78,127. As
the factors that regulate the acetylation of hepatic GAPDH in vivo are unknown, the
first aim of this study was to determine whether GAPDH acetylation in vitro is altered
by glucose, and whether GAPDH acetylation in mice was altered by hormones and
nutrient availability, which are known to regulate hepatic glucose homeostasis in
humans. In humans, insulin, glucagon and glucocorticoids have been shown to
regulate hepatic glucose homeostasis in response to nutrient availability, which
becomes impaired in diseases such as T2D5,7,28,29,128,129.
The acetylation profile of GAPDH in diseases is not established, and the
acetylation profile of GAPDH could be altered in certain diseases where glucose
metabolism is dysregulated, such as cancer and T2D. It has been demonstrated that
human GAPDH acetylated at K254 is important in cell proliferation, and that mutation
of K254 to glutamine or arginine restricted glycolysis and xenografted tumour growth
(A549 cells)83. However the acetylation profile of GAPDH in cancers is unknown,
but it could be important in cancer biology as the majority of cancers overexpress
GAPDH and exhibit the Warburg effect130. Given that the metabolic phenotype of
86
FAO hepatoma cells expressing 4KR GAPDH was similar to aspects of liver
metabolism in T2D, this suggests that there could be an altered GAPDH acetylation
profile in insulin resistant liver compared to normal liver28,29. As the acetylation
profile of hepatic GAPDH in obesity and T2D has not previously been investigated,
the second aim of this study was to determine whether hepatic GAPDH acetylation
was reduced in T2D, and in diet-induced obese mice (DIO).
87
4.2 METHODS
4.2.1 Total GAPDH acetylation assay
Non-transfected FAO cells were grown to 80% confluence in 10cm2 cell
culture dishes, as per Chapter Two. To determine whether glucose altered GAPDH
acetylation, cells were exposed to glucose free media or high glucose media (25mM)
for thirty minutes, and then harvested for immunoprecipitation as per Chapter Two.
To determine whether GAPDH acetylation was altered by hormones that regulate
glucose homeostasis, cells were treated with vehicle (DMSO), insulin (0.1nM), or
forskolin (20μM) for thirty minutes, and then harvested for immunoprecipitation as
per Chapter Two.
4.2.2 Permits and ethical approval
Approval for animal experimentation was obtained from the Deakin
University Animal Welfare Committee (A70-2009, GO4-2013 and G15-2014).
4.2.3 Mice experimental design
All mice were purchased from the Animal Resource Center in Western
Australia, and arrived at Deakin University at eight weeks of age. Mice were contained
in a controlled environment at Deakin University animal house at 22°C, on a twelve
hour light/dark cycle. All mice were administered a diet of standard chow for eight
weeks before dietary intervention, and water was consumed ad libitum throughout all
animal studies. For animal study one, male C57BL6 mice (n=48) were used to
investigate the regulation of hepatic GAPDH acetylation to determine whether
GAPDH acetylation is regulated by hormones or nutrient availability. For animal
study two, male db/db -/- (T2D, n=8) and db/db +/- (control, n=8) mice were used to
investigate whether GAPDH acetylation was altered in T2D. For animal study three,
male C57BL/6 mice that were fed a HFD (43% energy intake from fat, n=8) or standard
88
laboratory chow (n=8) for eight weeks, were used to investigate whether diet induced
obesity (DIO) altered GAPDH acetylation. At sixteen weeks of age mice from animal
studies one, two, and three were killed humanely by cervical dislocation, in the fed
state, and the livers were immediately excised and snap frozen in liquid nitrogen and
stored at -80°C.
For animal study one, male C57BL/6 mice (n=48) maintained on a standard
chow diet were used to determine the effect of fasting (16 hours, n=8), re-feeding (16
hour fast followed by 30 minute re-feed, n=8), insulin (0.75U/Kg, n=8), glucagon
(15μg/Kg, n=8), or norepinephrine (100ng/Kg, n=8) on GAPDH acetylation. Insulin,
glucagon, norepinephrine, and vehicle (saline of comparable volume to hormone
injections) were administered via intraperitoneal injection, and blood glucose was
measured via tail cut and blood spot using an ACCU-CHEK Performa glucose meter
(ROCHE) pre and post feeding/injection. The animals were killed humanely 15
minutes post injection or 30 minutes post re-feeding, and the livers were immediately
excised and snap frozen in liquid nitrogen and stored at -80°C.
4.2.4 Liver protein extraction, immunoprecipitation, and immunoblotting
All protein extraction, immunoprecipitation, and immunoblotting were
performed as outlined in Chapter Two.
4.2.5 Statistical analyses
Standard curves were generated using Microsoft Excel 2013, and statistical
analysis was conducted using SPSS 20 and MiniTab 16. A Kolmogorov-Smirnov test
was performed to determine normality and distribution. For data that was not normally
distributed, a non-parametric test (Kruskal–Wallis test) was used to compare means.
To compare means for normally distributed data a two tailed T-Test was used, where
equal variance was assumed if Levene’s significance > 0.05. For normally distributed
89
data, one-way and two-way ANOVA was used to compare groups using Games
Howell method (equal variances not assumed) or Fisher's Least Significant Difference
method (equal variances assumed). Results are presented as mean ± SEM, and P <
0.05 was considered statistically significant.
90
4.3 RESULTS
4.3.1 Total GAPDH acetylation in FAO hepatocytes was unchanged by glucose,
insulin, and forskolin. To determine whether GAPDH acetylation is regulated by
factors that regulate hepatic glucose production in vitro, FAO cells were exposed for
thirty minutes to insulin (0.1nM), forskolin (10μM), low and high glucose (0mM and
25mM), and then assayed for total GAPDH acetylation. Total GAPDH acetylation
was not altered after exposure to glucose free media or high glucose media for thirty
minutes (Figure 4.3.1 A). There were no differences in total GAPDH acetylation after
treatment with insulin and forskolin compared with vehicle (Figure 4.3.1 B).
4.3.2 Insulin, glucagon and norepinephrine do not regulate GAPDH acetylation
in vivo. Insulin, glucagon, and norepinephrine are key hormones that regulate hepatic
glucose metabolism in mammals. To determine whether hepatic GAPDH acetylation
is regulated by these hormones, C57BL/6 mice were administered insulin (0.75U/Kg,
n=8), glucagon (30μg/Kg, n=8) or nor-epinephrine (100ng/Kg, n=8) in the fed state by
intraperitoneal injection. The livers were excised fifteen minutes post-injection and
analysed for GAPDH acetylation. Akt is involved in insulin signalling and regulation
of gluconeogenesis, and phosphorylation of serine 473-Akt suggested that the insulin
pathway had been activated by these hormones (Figure 4.3.2 A). Phosphorylation of
serine-133 CREB suggested pathways downstream of glucagon and nor-epinephrine
signalling were activated (Figure 4.3.2 A). There was no significant change in total
GAPDH acetylation after insulin, glucagon or nor-epinephrine administration
compared with vehicle (Figure 4.3.2 B).
4.3.3 Total GAPDH acetylation was unchanged by fasting and re-feeding. To
determine whether GAPDH acetylation is altered in more physiological states of
hepatic glucose regulation, C57BL/6 mice (n=16) were fasted overnight (17:00-9:00)
91
and n=8 were killed, and n=8 were re-fed then killed thirty minutes later. The livers
were excised and analysed for glucagon and insulin signalling pathway activity, and
GAPDH acetylation. PKA is involved in glucagon signalling and is a regulator of
gluconeogenesis, and PKA was phosphorylated at threonine-197 in fasted mice, which
suggests activation of the glucagon pathway (Figure 4.3.3 A). Akt phosphorylation
was not detected, suggesting that the insulin pathway was not activated in fasted or fed
mice (Figure 4.3.3 A). Blood glucose levels increased after re-feeding (Figure 4.3.3
B), and there was no difference in GAPDH acetylation in fasted mice compared with
fed mice (Figure 4.3.3 C).
4.3.4 Hepatic GAPDH acetylation is reduced in T2D. To investigate the acetylation
profile of hepatic GAPDH in T2D, db/db mice were used as a model of T2D.
Consistent with this model, whole body mass and fasting blood glucose levels were
increased in db/db -/- compared with db/db +/- mice (Figure 4.3.4 A and B). Hepatic
total GAPDH acetylation was reduced in db/db -/- mice compared with db/db +/- mice
(Figure 4.3.4 C).
4.3.5 Global hepatic lysine acetylation is unchanged in T2D. To determine global
hepatic lysine acetylation, liver lysate from db/db -/- and db/db +/- mice was run on
SDS-PAGE and blotted for acetyl-lysine or acetylated α-tubulin, and then normalised
to total α-tubulin. There was no significant change in global acetylation or acetylated
α-tubulin in liver lysate from db/db +/- mice compared with db/db -/- mice (Figures
4.3.5 A and B).
4.3.6 Hepatic GAPDH acetylation is reduced in mice on a high fat diet. To
investigate the acetylation profile of hepatic GAPDH in obesity independent of T2D,
DIO C57BL/6 mice were used as a model of obesity. Whole body mass was increased
in mice on a HFD compared with mice on a CHOW diet (Figure 4.3.6 A). Blood
92
glucose levels were normal and were not different between mice on a HFD and mice
on a CHOW diet (Figure 4.3.6 B). C57BL/6 mice on a HFD had reduced hepatic
GAPDH acetylation compared with C57BL/6 mice on a standard CHOW diet (Figure
4.3.6 C).
93
AcK GAPDH/GAPDH (%0mM)
A
300
250
200
150
100
50
0
0mM
25mM
Glucose
AcK GAPDH/GAPDH (%veh)
B
250
200
150
100
50
0
Veh
Ins
Fors
Figure 4.3.1: GAPDH acetylation in FAO cells after exposure to high glucose, insulin (Ins) and
forskolin (Fors). Cells were immunoprecipitated for GAPDH and immunoblotted for total acetylation.
(A) FAO GAPDH acetylation in 0mM and 25mM glucose media after thirty minutes. Data expressed
as %0mM. (B) FAO GAPDH acetylation after exposure to vehicle (Veh, DMSO), insulin (0.1nM),
and forskolin (20μM). Data expressed as %Veh. All data represented as mean ± SEM, n=22 replicates
per group. Results not statistically significant.
94
A
B
gluc n-epi veh
ins
IP:GAPDH/WB:AcK
gluc
n.e
veh
IP:GAPDH/WB:GAPDH
ins
pS473-Akt
Akt
140
veh gluc gluc
n.e
pS133-creb
creb
n.e
ins
ins
120
AcK GAPDH (%veh)
veh
100
80
60
40
20
0
Veh
Gluc
n-epi
Ins
Figure 4.3.2: Hepatic GAPDH acetylation in C57BL/6 mice after injection with either insulin
(ins), glucagon (gluc), norepinephrine (n-epi), or vehicle (veh). Fifteen minutes post-injection,
mouse livers were removed and homogenised in lysis buffer containing TSA and NAM. (A) Liver
lysates were western blotted for total and phospho-specific Akt and CREB. (B) Liver lysates were
immunoprecipitated for GAPDH and western blotted for GAPDH and acetyl-lysine. Lysine-acetylated
GAPDH was normalised to total GAPDH and expressed as %veh. All data are represented as mean ±
SEM, n=8 replicates per group.
95
B
A
C
re-fed fasted
IP:GAPDH/WB:AcK
fasted
PKA
Blood glucose (mmol)
Akt
pT197-PKA
12
IP:GAPDH/WB:GAPDH
140
10
120
8
*
6
4
2
0
Re- Fed
Fasted
AcK GAPDH (%fasted)
re-fed
pS473-Akt
100
80
60
40
20
0
Re- Fed
Fasted
Figure 4.3.3: GAPDH acetylation in fasted and re-fed C57BL/6 mice. (A) Liver lysates were
immunoblotted for total and phospho-specific Akt and PKA to confirm that the glucagon pathway was
activiated after fasting. (B) Blood glucose levels were measured in fasted C57BL/6 mice and thirty
minutes post-feeding. (C) Hepatic GAPDH acetylation in fasted and re-fed mice. All data represented
as mean, ± SEM, n=8 replicates per group. * denotes statistically different to fasted mice, P < 0.05.
96
A
B
C
+/-
-/-
+/-
-/-
IP:GAPDH/WB:AcK
IP:GAPDH/WB:GAPDH
30
*
Blood glucose (mmol)
Whole body mass (g)
40
35
30
25
20
15
10
5
0
db/db +/-
db/db -/-
140
*
25
120
AcK GAPDH
(%db/db +/-)
45
20
15
10
100
80
60
*
40
5
20
0
0
db/db +/-
db/db -/-
db/db +/-
db/db -/-
Figure 4.3.4: Hepatic GAPDH acetylation in db/db +/- (control) and db/db -/- (T2D) mice. Liver
lysates from db/db +/- and db/db -/- mice were immunoprecipitated for GAPDH, and then western
blotted for GAPDH and acetyl-lysine. (A) Whole body mass. (B) Fasting blood glucose. (C) Lysineacetylated GAPDH was normalised to total GAPDH and expressed as %db/db +/-. All data are
represented as mean ± SEM, n=8 replicates per group. * denotes statistically different to db/db +/-, P <
0.05.
97
A
B
CON T2D CON T2D
AcK-α-tubulin
CON T2D
CON T2D
160
140
55
WB:AcK
40
35
25
15
α-tubulin
Hepatic global AcK (%CON)
100
70
120
α-tubulin
100
AcK-α-Tubulin (%CON)
MWM
170
130
120
100
80
60
40
20
80
60
40
20
0
0
CON
CON
T2D
T2D
Figure 4.3.5: Global hepatic lysine acetylation and acetylated-tubulin levels in db/db +/- (CON)
and db/db -/- (T2D) mice. Whole liver cell lysates from control and T2D mice were run on an SDSPAGE, and probed using an acetyl-lysine, acetylated α-tubulin, and α-tubulin antibodies. (A) Global
acetylation was normalised to α-tubulin and expressed as %CON. (B) The amount of acetylated αtubulin was normalised to total α-tubulin and expressed as %CON. All data are represented as mean, ±
SEM, n=8 replicates per group.
98
A
B
C
CHOW HFD CHOW HFD
IP:GAPDH/WB:AcK
IP:GAPDH/WB:GAPDH
45
9
30
25
20
15
10
5
AcK GAPDH (%CHOW)
35
Blood glucose (mmol)
Whole body mass (g)
40
120
10
*
8
7
6
5
4
3
2
1
CHOW
HFD
80
*
60
40
20
0
0
0
100
CHOW
HFD
CHOW
HFD
Figure 4.3.6: Hepatic GAPDH acetylation in mice fed CHOW versus mice fed a HFD. Liver lysate
from CHOW fed and HFD fed mice were immunoprecipitated for GAPDH, and then western blotted
for GAPDH and acetyl-lysine. (A) Whole body mass. (B) Blood glucose. (C) Lysine-acetylated
GAPDH was normalised to total GAPDH and expressed as %CHOW. All data represented as mean ±
SEM, n=8 replicates per group. * denotes statistically different to CHOW, P < 0.05.
99
4.4 DISCUSSION
Elevated hepatic de novo glucose production is the primary source of
endogenous glucose in T2D, and a major contributor to hyperglycemia73. Reducing
hepatic blood glucose output could reduce blood glucose levels and alleviate
symptoms of hyperglycemia.
Results from Chapter Two demonstrated that the
acetylation profile of GAPDH can regulate glycolysis/gluconeogenesis in vitro.
Specifically, mutation of GAPDH lysine residues to arginine, to mimic their
deacetylation, increased glucose production and lipid accumulation.
This study has demonstrated that total hepatic GAPDH acetylation was
reduced in db/db -/- (T2D) mice compared to db/db +/- (Control) mice (Figure 4.3.4
C). This is the first study to show that hepatic GAPDH acetylation is altered in T2D,
and together with our previous findings in this thesis suggests that deacetylated hepatic
GAPDH could play a role in the dysregulated metabolism that occurs in T2D. A clear
limitation to these findings is that the acetylation site(s) dysregulated in T2D are
unknown at present, and further work will be required to identify which specific
hepatic GAPDH lysine residues are deacetylated in T2D. It is also unclear from this
data whether the reduced GAPDH acetylation observed in T2D could potentially result
in a pathophysiological effect, such as hyperglycemia. In previous chapters of this
thesis it has been shown that the combined deacetylation of rat GAPDH lysine residues
K115, K160, K225 and K252 resulted in an increase in gluconeogenesis and glucose
production in vitro. However, it was also demonstrated that total GAPDH acetylation
was not reduced in 4KR GAPDH FAO cells compared to WT GAPDH FAO cells,
which further suggests that the reduction in GAPDH acetylation in T2D and DIO, may
be the result of additional deacetylated lysine residues on GAPDH. GAPDH has
twenty six lysine residues, and it is possible that the deacetylation, or other acyl PTMs,
100
of the majority of these lysine residues is required to exert an affect in vivo, which
could result in increased glucose production.
Hepatic GAPDH acetylation was also reduced in DIO mice, which suggests
that hepatic GAPDH acetylation is regulated in vivo by nutrient availability (Figure
4.3.6 C). Chronic exposure to diets high in fat could lead to a reduction in hepatic
GAPDH acetylation, causing an increase in hepatic glucose production, which could
progress to hyperglycemia and T2D. However, the blood glucose levels in the mice
on a HFD were not increased compared to mice on CHOW (Figure 4.3.6 B). This
suggests that reduced GAPDH acetylation in DIO may not result in a measurable
pathophysiological defect. However, it is also possible that the normal blood glucose
levels seen in mice on a HFD compared to control mice, may be a result of a greater
rate of glucose disposal. These results demonstrate that hepatic GAPDH acetylation
is reduced in obesity and T2D in vivo, but further work is required to determine if the
altered acetylation profile of GAPDH is a pathophysiological factor in T2D and DIO.
The reduced in vivo acetylation of hepatic GAPDH in T2D could play a role in elevated
gluconeogenesis and hepatic glucose production. Further work is needed to identify
the specific hepatic GAPDH lysine residue that are deacetylated on a HFD to
determine whether they match the deacetylated residues in T2D, and whether
gluconeogenesis is increased in T2D and mice on a HFD.
Increased mitochondrial acetylation and increased acetylation of histones has
been shown to occur in T2D and DIO, and although hepatic GAPDH acetylation was
reduced in T2D, there was no significant change seen in global hepatic acetylation or
acetylated α-tubulin in T2D (Figure 4.3.6 A and B)103,104,131,132. This demonstrates that
total acetylation is not reduced in T2D and that the hepatic GAPDH deacetylation that
occurs in T2D is independent of global deacetylation/acetylation events. It is likely
101
that there are alterations in acetylation, whether acetylation is reduced or increased,
throughout the various tissues and cellular compartments in diseased states that could
disrupt pathway signalling. Acetylated α-tubulin levels are unchanged in T2D, which
demonstrates that α-tubulin acetylation and the KATs/KDACs responsible for
acetylating/deacetylating α-tubulin, are not altered in T2D. HDAC6, along with
SIRT2, are known α-tubulin deacetylases111,113,133-137. Interpreting the data from this
chapter would suggest that the activity of these α-tubulin deacetylases is not increased
in T2D. However, the eighteen known KDACs are responsible for deacetylating in
excess of 2000 proteins, so there are likely other factors, such as PTMS and protein
complex formations/interactions, which are also involved in determining the
deacetylation of specific proteins by HDAC6 and SIRT220,21,64,65. This would allow
for HDAC6 or SIRT2 activity to be increased, where it could have normal
deacetylation properties for α-tubulin in T2D, but elevated deacetylation of other
proteins in T2D, such as GAPDH.
Previous research has demonstrated that bacterial GAPDH acetylation is
regulated in response to nutrient availability19,58. GAPDH acetylation did not change
significantly after thirty minutes in FAO cells exposed to low and high glucose media,
suggesting that GAPDH acetylation is not regulated by glucose in mammalian cells
(Figure 4.3.1 A). In addition, GAPDH acetylation did not change in FAO cells after
treatment with insulin and forskolin, compared with vehicle (Figure 4.3.1 B). The
studies in this chapter suggest that GAPDH acetylation is not regulated by hormones
in vivo, and remains unchanged during fasting and re-fed states (Figures 4.3.2 and
4.3.3). However the in vivo and in vitro studies in this chapter require further
investigation, and a time course analysis of GAPDH acetylation after feeding or
injection with insulin and glucagon could demonstrate rapid fluctuations in GAPDH
102
acetylation. SILAC has been used in multiple proteomic and acetylation studies, and
can also be used to determine lysine residues that are targeted by specific
KDACs19,45,138. A potential limitation to these results using FAO cells is that they are
a cancer cell line and glycolysis could be clamped, or constitutive to an extent, and is
not regulated by the usual signalling processes. It is possible that GAPDH acetylation
is not regulated by nutrient availability or hormones, and that the deacetylation of
GAPDH in T2D and mice on a HFD is a pathological factor that does not have a
physiological role. Furthermore, these results are based on acute regulation rather than
chronic regulation as seen in DIO mice, or in db/db mice.
There were several limitations to this study, in particular the determination of
total GAPDH acetylation using pan acetyl-lysine antibodies, and the design of the
animal study based on ethical and time restrictions. The current pan acetyl-lysine
antibodies are not reliable, and the epitope recognized by a particular pan acetyl-lysine
antibody may not be compatible with the acetylated lysine residue within the protein
that are under investigation139.
Pan acetyl-lysine antibodies may also show
batch̻to̻batch variability, in terms of both target specificity and enrichment
efficiency that limits experimental reproducibility, and complicates the interpretation
of results64. Site specific acetyl-lysine antibodies are required to confirm whether the
rat GAPDH lysine sites 115, 160, 225, and 252 from Chapter Two are indeed
deacetylated in T2D and DIO. Individual site specific acetyl-lysine antibodies to
GAPDH K115, K160, K225, and K252 are currently being generated for this project.
Site specific antibodies for all twenty six GAPDH lysine residues, and SILAC
procedures, would be useful to determine the acetylation profile of lysine residues in
T2D, and at various time points under the influence of different substrates such as
glucose and insulin. The fluctuations of lysine residues from an acetylated state to a
103
deacetylated state are likely to occur rapidly, and the acetylation profile of lysine
residues cannot be determined conclusively from a single time point. In the case of
T2D and DIO, it may be that the fluctuations between acetylated and deacetylated
lysine are negligible, where a deacetylated lysine state is favoured.
Further
investigation into the KDACs responsible for the deacetylation of lysine residues in
diseased states could potentially explain the reduction in fluctuations of those lysine
residues back to an acetylated state.
In conclusion, the results shown in this chapter demonstrate that total hepatic
GAPDH acetylation is reduced in T2D, and could contribute to the increased rate of
gluconeogenesis and hepatic glucose production seen in individuals with T2D. Results
from this study have also shown that hepatic GAPDH acetylation is reduced in mice
on a HFD, suggesting that hepatic GAPDH acetylation could be regulated at a
nutritional level. Further investigation is required to identify the specific GAPDH
lysine residues that are deacetylated, and the enzymes that deacetylate those lysine
residues. Future work could target the KDACs that deacetylate GAPDH, preventing
GAPDH deacetylation and potentially reducing hepatic gluconeogenesis and glucose
production in T2D.
Cell culture studies in this chapter demonstrated that glucose, insulin, and
forskolin did not affect the acetylation profile of GAPDH, however this contradicts
previous research and requires further investigation with site specific acetyl-lysine
antibodies, and systems biology experiments such as SILAC19,22,46,58,65,140-143.
Previous research has demonstrated that the acetylation of metabolic enzymes and
GAPDH is increased in high glucose media, however this study showed no
relationship between GAPDH acetylation and glucose, insulin, norepinephrine, and
glucagon/forskolin in vitro and in vivo19,83.
104
The factors that regulate GAPDH
acetylation in mammals is still not clear, and further research is required to determine
these factors.
105
CHAPTER 5
THE ROLE OF HEPATIC GAPDH LYSINE RESIDUES
IN THE REGULATION OF GLUCOSE METABOLISM
IN VIVO
5.1 INTRODUCTION
In healthy individuals gluconeogenesis and the breakdown of stored liver
glycogen maintain hepatic glucose production, and subsequent blood glucose levels,
during fasting and exercise144. Hepatic gluconeogenesis and blood glucose levels are
elevated in T2D, and the mechanisms involved are incompletely understood5,28. In
previous chapters of this thesis, it has been demonstrated that mutation of rat GAPDH
lysine residues 115, 160, 225, and 252 to arginine (4KR GAPDH) increased glucose
production in FAO hepatocytes, and that hepatic GAPDH acetylation is reduced in
T2D and mice on a HFD. This demonstrated that acetylation of GAPDH has rate
limiting properties in vitro, which has also been demonstrated in bacteria19,58,140.
However, the regulation of hepatic GAPDH acetylation, and the effects of reversible
GAPDH acetylation on hepatic glucose metabolism in vivo, is not fully understood.
Chapter Four of this thesis investigated the regulation of hepatic GAPDH
acetylation in vivo, and the total acetylation of hepatic GAPDH in T2D and DIO.
These studies found that physiological factors that regulate glucose production in
mammals do not regulate total hepatic GAPDH acetylation (Figures 4.3.2 and 4.3.3).
However, there was a reduction in total hepatic GAPDH acetylation in T2D and in
DIO mice (Figures 4.3.4 and 4.3.6), suggesting that reduced GAPDH acetylation may
only be altered in pathological states. To determine the factors that regulate GAPDH
acetylation, in particular the specific lysine sites that are mutated in the 4KR GAPDH
mutant, further research is required.
106
Previous research has demonstrated that a point mutation of human GAPDH
lysine K254 (equivalent to GAPDH K252 in rat, Figure 2.1.1) to glutamine reduced
xenografted tumour growth in rats83. Mutation of lysine residues to glutamine has
been traditionally used as an acetylation mimetic, but actually differs substantially, and
does not function in the same manner as an acetylated lysine64. It is now considered
that mutations of lysine residues to glutamine does not mimic an acetylated lysine, and
that analyses derived from glutamine mutations is the result of a lysine site that cannot
undergo its normal PTMs64. Tumours exhibit elevated rates of aerobic glycolysis,
known as the Warburg effect, and if human GAPDH K254Q inhibited glycolysis this
could explain the reduction in tumour size. If deacetylation of rat GAPDH lysine
residues K115, K160, K225, and K252 increases gluconeogenesis and blood glucose
levels in mice, then it could provide a new target for the treatment of T2D and certain
cancers.
It remains to be determined whether reduced hepatic GAPDH acetylation
increases gluconeogenesis and blood glucose levels in vivo. Previous chapters in this
thesis have shown that 4KR GAPDH increased gluconeogenesis and reduced
glycolysis in FAO cells, however this needs to be shown in vivo. Cell culture studies
do not accurately reflect the conditions that occur in vivo, so it is necessary that the
results from previous studies in this thesis are validated in an in vivo model. Glucose
pathways, regulation, and metabolism in vivo is a complex system that involves
multiple tissues and signalling pathways, however the cell culture results are based on
autonomous hepatocytes.
Therefore the aim of this study was to determine whether
4KR GAPDH increased hepatic gluconeogenesis and blood glucose levels in vivo. WT
GAPDH and 4KR GAPDH AAV8 vectors, that co-express GFP, were used to express
WT GAPDH and 4KR GAPDH specifically in the liver of C57Bl6/J mice.
107
5.2 METHODS
5.2.1 Permits and ethical approval
Approval for animal experimentation was obtained from the Deakin
University Animal Welfare Committee (G15-2014).
5.2.2 Construction of adeno-associated virus vectors serotyped with AAV8 capsid
WT GAPDH and 4KR GAPDH AAV8 plasmids and viruses were generated
by The Department of Neurology, Hope Center Viral Vectors Core, Washington
University, School of Medicine, Saint Louis, USA. Virus packaging, purification, and
measurement of viral titre were all performed by Washington University.
WT
GAPDH and 4KR GAPDH vectors (Figure 5.2.1) co-expressed GFP via an internal
ribosome entry site (IRES), and contained the liver specific promoter; thyroxinebinding globulin (TBG).
Sma I (43)
Sma I (54)
ColE1-ori
ITR
Eco RI (163)
Asc I (396)
TBG promoter
ApR
HindIII (886)
pAAV-TBG-GAPDA W T-IRES-GFP
6757 bp
GAPDH
Eco RV (1912)
f1+
IRES
Sma I (3852)
Sma I (3841)
ITR
eGFP
bGHpolyA
SV40 polyA
Figure 5.2.1: TBG-GAPDH-GFP AAV8 plasmid map. GAPDH AAV8 co-expressing GFP plasmid
map, generated by The Department of Neurology, Hope Center Viral Vectors Core, Washington
University, School of Medicine, Saint Louis, USA.
108
Rather than tag GAPDH with GFP, GFP was co-expressed independently to
prevent any alterations to GAPDH, or effects that could occur due to GAPDH as a
result of a GFP tag.
5.2.3 Animals and experimental design
C57BL/6 male mice (n=16) were acquired from the Animal Resource Center
in Western Australia, and arrived at Deakin University at four weeks of age. Mice
were contained in a controlled environment at the Deakin University animal house at
22°C, on a twelve hour light/dark cycle. Mice were fed a diet of standard chow for the
duration of this study. At fifteen weeks of age the mice were injected with either WT
GAPDH-GFP AAV8 (1 x 1012vg) or 4KR GAPDH-GFP AAV8 (1 x 1012vg), via tail
vein injection. At seventeen weeks of age weekly measurements of blood glucose,
food consumption, body weight and composition were obtained. Weekly blood
glucose measurements were taken in the fed state. At eighteen weeks of age a blood
glucose fasting/ re-feed experiment was conducted, followed by a glucose tolerance
test (GTT) at nineteen weeks of age. At twenty weeks of age an insulin tolerance test
(ITT) was conducted, followed by a glucagon tolerance test (GnTT) at twenty one
weeks of age. At twenty two weeks of age, mice were administered deuterated water
and seventy two hours later 60μl of blood was taken in the fasted and fed state. Mice
were then humanely killed two days later (Figure 5.2.2).
Age of mice
0wks
15wks
Injection of
AAV vector
17wks
Weekly
measurements
commence
18wks
19wks
20wks
21wks
Fast &
re-feed
GTT
ITT
GnTT
22wks
D2O and
fast/refeed
23wks
Humanely
killed and
tissues
collected
Figure 5.2.2: Animal study time line. Time course for mutant GAPDH AAV8 animal study.
109
5.2.4 Glucose tolerance tests (GTT)
Mice were fasted for five hours and then administered an intraperitoneal
injection of glucose (2g/Kg). Blood samples (30μl) were obtained pre-injection and
then 15, 30, and 60 minutes post injection via tail cut in heparinised tubes for plasma
collection. Blood glucose levels were measured pre-injection and then 15, 30, 45, 60,
and 90 minutes post injection using an ACCU-CHEK Performa glucose meter
(ROCHE). To obtain plasma, blood was spun in a centrifuge at 12,000 rpm, and
plasma was stored at -80°C.
5.2.5 Insulin tolerance test (ITT)
Mice were fasted for five hours and then administered an intraperitoneal
injection of insulin (0.75U/Kg).
Blood glucose levels were measured pre-injection
and then 15, 30, 45, 60, and 90 minutes post injection using an ACCU-CHEK Performa
glucose meter (ROCHE).
5.2.6 Glucagon tolerance test (GnTT)
Mice were fasted for five hours and then administered an intraperitoneal
injection of glucagon (30μg/Kg). Blood samples (30μl) were obtained pre-injection
and then 15, 30, and 60 minutes post injection via tail cut in heparinised tubes for
plasma collection. Blood glucose levels were measured pre-injection and then 15, 30,
45, 60, and 90 minutes post injection using an ACCU-CHEK Performa glucose reader
(ROCHE). To obtain plasma, blood was spun in a centrifuge at 12,000 rpm, and
plasma was stored at -80°C.
5.2.7 Fasting/re-feed blood glucose test
Mice were fasted for fifteen hours (from 17:00 to 08:00) and then refed.
Blood samples were taken pre-refeed and then sixty minutes post-refeed via tail cut in
heparinised tubes for plasma collection. Blood glucose levels were taken pre-refeed
110
and then 15, 30, 45, 60, and 90 minutes post refeed using an ACCU-CHEK Performa
glucose meter (ROCHE). To obtain plasma, blood was spun in a centrifuge at 12,000
rpm, and plasma was stored at -80°C.
5.2.8 Hepatic glucose production measured by deuterated water labelling
2
H2O technique recently adapted by Antoniewicz et al was used to determine
whether hepatic overexpression of 4KR GAPDH had an impact on gluconeogenesis or
glycogenolysis145.
Following the administration of
2
H2O,
2
H atoms become
incorporated into endogenously synthesised glucose. Then by measuring the ratio of
2
H enrichment in plasma glucose at carbon 5 and 2, the proportion of glucose that is
produced from gluconeogenesis can be calculated. The proportion of glucose being
produced from glycogenolysis is determined by subtracting the C5/C2 ratio from 1.
Gluconeogenesis derived from the mitochondria via the TCA cycle was determined by
measuring the enrichment of 2H at carbon 6 vs 2, while gluconeogenesis derived from
the cytosol is estimated by calculating the difference between all sources of
gluconeogenesis.
At twenty two weeks of age, mice were administered 2H2O (20 mL/Kg, >
99.9% enriched, Sigma) by intraperitoneal injection, and cage water was substituted
with 10% 2H2O.
Three days post-injection, mice were fasted for fifteen hours
overnight (from 17:00 to 08:00), then 60μl of blood was taken via tail cut. Mice were
re-fed and 60μl of blood was taken after thirty minutes. Blood was centrifuged at
12,000 rpm, and plasma was separated and stored at -80°C.
The 2H enrichment of glucose from 20μL plasma was determined by
measuring the enrichment in six ions from three different glucose derivatives: glucose
aldonitrile pentapropionate, glucose 1,2,5,6-di-isopropylidene propionate, and glucose
methyloxime pentapropionate. The derivatives were prepared as described by
111
Antoniewicz et al145 and analysed using an Agilent HP 6890 GC system coupled to an
Agilent HP 5973 MSD (Agilent Technologies) in electron ionisation mode with
helium as a carrier gas. The injector insert temperature was 270°C and the GC–MS
transfer line temperature was 250 °C. The oven temperature gradient was set to: 70 °C
(one minute); 70 °C to 295 °C at 12.5 °C/min; 295 °C to 320 °C at 25 °C/min; 320 °C
for two minutes. The mass isotopomer abundances were determined using Mass
Hunter Workstation (Agilent Technologies) and the enrichment of individual glucose
positions was calculated using Matlab software (MathWorks).
5.2.9 Insulin ELISA
Insulin assays were performed using a Mouse Ultrasensitive Insulin ELISA
kit (Alpco), per manufacturer’s instructions, with 5μl of plasma in duplicate. A 5parameter logistic curve and insulin concentration were calculated using ELISA
analysis software (Elisakit, www.elisaanalysis.com/app).
5.2.10 Body composition analysis
Lean mass and fat mass were measured weekly from seventeen weeks of age,
two weeks post AAV injection, and were evaluated by quantitative nuclear magnetic
resonance imaging. Mice were transferred into a plastic tube and analysed with an
EchoMRI-500 (Echo MRI). Lean and fat mass were then normalised to whole body
mass.
5.2.11 Protein extraction, immunoprecipitation, and immunoblotting
All protein extraction, immunoprecipitation, and immunoblotting were
performed as outlined in Chapter Two.
112
5.2.12 GAPDH activity assay
Liver lysate containing no KDAC inhibitors (diluted 1:20) was used to
determine gluconeogenic and glycolytic GAPDH activity using a colourmetric
GAPDH assay following the manufactures instructions as outlined in Chapter Two.
5.2.13 Quantitative measurement of percent haemoglobin A1c in blood
Percent HbA1c was determined by blood spotting via tail cut into a small
glass capillary tube, which was then loaded into a DCA Systems Haemoglobin A1c
Reagent Kit (Siemens). Percent HbA1c was then measured using a DCA Vantage
Analyser (Siemens).
5.2.14 Glycogen concentration assay
Glycogen concentration was determined by hydrolysis of glycogen to glucose
using the acid-hydrolysis method developed by Passonneau and Lauderdale, and
adapted for mouse liver by Zhang146,147. Glucose, the hydrolysis product of glycogen,
can be converted into glucose-6-phosphate (G-6-P) by hexokinase, which is further
converted into 6-phosphogluconic acid by G-6-P dehydrogenase producing
NADPH146,147.
NADPH can be measured spectrophotometrically to determine
glycogen content.
A portion of frozen liver or quadriceps muscle was excised and weighed
(approximately 10mg), and digested in 250μl of 2M HCl at 95-100°C for thirty
minutes. Samples were then homogenised at 33,000 rpm for 20 seconds, and incubated
for two hours at 95-100°C with gentle agitation at 15, 30, 60, and 90 minutes. After
incubation, 750μl of 0.667M NaOH was added and the samples were gently mixed to
neutralise the extract. Glucose concentration was determined using a Glucose (HK)
assay kit (Sigma, #GAHK20-1KT) according to the manufacturer’s instructions.
Absorbance was measured at 340nm, and NADPH concentration was determined from
113
a standard curve. NADPH concentrations were then normalised to samples with a
known glucose concentration, and further normalised to wet tissue weight (mmol
glycogen/kg wet tissue).
5.2.15 Triglyceride quantification
Triglyceride
quantification
was
performed
using
a
Triglyceride
Quantification Colorimetric/Fluorometric kit (BioVision), as per manufactures
instructions. Diluted (1:30) liver lysate (5μl) containing no KDAC inhibitors was used
in the Triglyceride quantification assay.
5.2.16 Histology
Histology and hematoxylin and eosin staining of liver sections were performed
by The Department of Anatomy and Neuroscience, School of Biomedical Sciences,
University of Melbourne, Parkville, Melbourne.
5.2.17 Statistical analyses
Standard curves and incremental area under the curve (iAUC) were generated
using Microsoft Excel 2013, and statistical analysis was conducted using SPSS 20 and
MiniTab 16. A Kolmogorov-Smirnov test was performed to determine normality and
distribution.
For data that was not normally distributed, a non-parametric test
(Kruskal–Wallis test) was used to compare means. To compare means for normally
distributed data a two tailed T-Test was used, where equal variance was assumed if
Levene’s significance > 0.05. For normally distributed data, one-way and two-way
ANOVA was used to compare groups using Games Howell method (equal variances
not assumed) or Fisher's Least Significant Difference method (equal variances
assumed). Results are presented as mean ± SEM, and P < 0.05 was considered
significant.
114
5.3 RESULTS
5.3.1 AAV8 GAPDH expression is liver specific.
Tissue lysates were
immunoblotted for GFP to confirm liver specific expression of exogenous GAPDH
AAV8 vectors that co-express GFP. GFP was expressed in WT GAPDH and 4KR
GAPDH mice compared with mice that were not administered AAVs (Figure 5.3.1 A).
GFP was expressed in liver but not heart, kidney, muscle or fat (Figure 5.3.1 B).
5.3.2 Hepatic total GAPDH acetylation is unchanged in 4KR GAPDH mice. In
order to see whether hepatic total GAPDH acetylation was reduced in 4KR GAPDH
mice, GAPDH acetylation was determined by GAPDH immunoprecipitation and
immunoblotting for GAPDH and acetyl-lysine. GAPDH acetylation was normalised
to total GAPDH. Hepatic total GAPDH acetylation was unchanged in 4KR GAPDH
mice compared with WT GAPDH mice (Figure 5.3.2).
5.3.3 Gluconeogenic GAPDH activity is increased in 4KR GAPDH mice. To
determine whether 4KR GAPDH mice had increased gluconeogenic GAPDH activity,
a GAPDH activity assay was performed using a colourmetric GAPDH activity kit, and
GAPDH activity was normalised to total protein. Gluconeogenic GAPDH activity was
increased in 4KR GAPDH mice compared to WT GAPDH mice (Figure 5.3.3 B).
Glycolytic hepatic GAPDH activity was reduced in 4KR GAPDH mice compared with
WT GAPDH mice (Figure 5.3.3 A).
5.3.4 Fractional contribution of gluconeogenesis to total endogenous glucose is
elevated in 4KR GAPDH mice. Deuterated water was administered to mice in order
to determine the fractional contribution of gluconeogenesis and glycogenolysis to total
endogenous glucose by mass spectrometry.
4KR GAPDH mice had a higher
proportion of endogenous glucose produced from gluconeogenesis, and a reduced
contribution to total endogenous glucose from glycogenolysis, compared with WT
115
GAPDH mice (Figure 5.3.4 A). Gluconeogenesis derived from glycerol was higher in
4KR GAPDH mice compared with WT GAPDH mice (Figure 5.3.4 B), and
gluconeogenesis derived from phosphoenolpyruvate carboxylase (PEP) was not
statistically different in 4KR GAPDH mice compared with WT GAPDH mice (Figure
5.3.4 B).
5.3.5 Food consumption and body composition were unchanged in 4KR GAPDH
mice. To determine whether hepatic 4KR GAPDH expression had an effect on body
composition, whole body, lean, and fat mass were measured weekly by ECHO MRI.
WT GAPDH mice and 4KR GAPDH mice average daily food intake was not different
(Figure 5.3.5 A). Whole body mass was unchanged in 4KR GAPDH mice compared
with WT GAPDH mice (Figure 5.3.5 B). Percent fat mass and percent lean mass,
determined by quantitative nuclear magnetic resonance imaging, were not different in
4KR GAPDH mice compared with WT GAPDH mice (Figure 5.3.5 C and D).
5.3.6 Weekly blood glucose levels and percentage HbA1c were unchanged in 4KR
GAPDH mice. Mutation of GAPDH lysine residues to arginine increased glucose
production in FAO hepatocytes, so to determine whether GAPDH acetylation alters
blood glucose levels in vivo, blood glucose in the fed state was measured weekly, and
percentage HbA1c was measured in week twenty two before 2H2O administration.
Blood glucose levels did not change in 4KR GAPDH mice compared with WT
GAPDH mice (Figure 5.3.6 A). Percentage HbA1c levels were measured at twenty
two weeks of age as a proxy measure of long term glucose homeostasis. Percentage
HbA1c was unchanged in 4KR GAPDH mice compared with WT GAPDH mice
(Figure 5.3.6 B)
5.3.7 Glucose tolerance was altered in 4KR GAPDH mice after injection with
glucose. To determine whether mutation of GAPDH lysine residues affected glucose
116
homeostasis, a GTT was performed. Glucose tolerance was higher in 4KR GAPDH
mice compared with WT GAPDH mice after injection with glucose (Figure 5.3.7 A
and B). Insulin levels were unchanged in 4KR GAPDH mice compared with WT
GAPDH mice after injection with glucose (Figure 5.3.7 C and D).
5.3.8 Glucagon action was altered in 4KR GAPDH mice after injection with
glucagon. To determine whether mutation of GAPDH lysine residues affected glucose
homeostasis in response to glucagon, a GnTT was performed. Mice expressing 4KR
GAPDH were more glucagon tolerant compared with WT GAPDH mice (Figure 5.3.8
A and B). Either the rate of glucose disposal was higher in 4KR GAPDH mice
compared with WT GAPDH mice after injection with glucagon, or the production of
glucose in response to glucagon was reduced in 4KR GAPDH mice compared with
WT GAPDH mice (Figure 5.3.8 A and B). Insulin levels were unchanged in 4KR
GAPDH mice compared with WT GAPDH mice after injection with glucagon (Figure
5.3.8 C and D).
5.3.9 The post-prandial response after fasting and re-feeding was unchanged in
4KR GAPDH mice. To determine whether mutation of GAPDH lysine residues
affected the post-prandial response after feeding, blood glucose levels were measured
over ninety minutes after mice were fasted and re-fed. After a fifteen hour fast blood
glucose levels were unchanged in 4KR GAPDH mice compared with WT GAPDH
mice (Figure 5.3.9 A). After feeding blood glucose levels were not different in 4KR
GAPDH compared with WT GAPDH mice (Figure 5.3.9 A and B).
5.3.10 Insulin tolerance is not altered in 4KR GAPDH mice. To determine whether
mutation of GAPDH lysine residues to arginine affected insulin tolerance, an ITT was
performed. Insulin tolerance was not different in 4KR GAPDH mice compared with
WT GAPDH mice (Figure 5.3.10 A and B).
117
5.3.11 Hepatic glycogen content was elevated in 4KR GAPDH mice. To determine
whether increased glycogen storage was accounting for the normal blood glucose
levels in 4KR GAPDH mice that exhibited increased gluconeogenic GAPDH activity,
a glycogen assay was performed on liver and muscle tissue. Hepatic glycogen
concentrations were increased in 4KR GAPDH mice compared with WT GAPDH
mice (Figure 5.3.11 A). Muscle glycogen concentration in 4KR GAPDH mice was
not different compared with WT GAPDH mice (Figure 5.3.11 B).
5.3.12 Hepatocyte morphology and hepatic triglyceride levels are unchanged in
4KR GAPDH mice. To ensure that mutation of GAPDH lysine residues to arginine
did not affect liver morphology hematoxylin and eosin staining was performed. Liver
morphology was normal in 4KR GAPDH and WT GAPDH mice (Figure 5.3.12 A).
Liver triglyceride concentration was determined using a colourmetric TAG assay kit.
Hepatic triglyceride concentrations were unchanged in 4KR GAPDH mice compared
with WT GAPDH mice (Figure 5.3.12 B)
118
A
WT GAPDH
-ve con con
GFP
GAPDH
α-tubulin
4KR GAPDH
-ve con con
GFP
GAPDH
α-tubulin
B
WT Liver Heart Kidney Muscle
Fat
GFP
GAPDH
α-tubulin
4KR Liver
Heart Kidney Muscle Fat
GFP
GAPDH
α-tubulin
Figure 5.3.1: Western blot for GFP expression. (A) GFP expression from liver in WT GAPDH and
4KR GAPDH mice. (B) GFP expression in WT GAPDH and 4KR GAPDH mice from liver, heart,
kidney, muscle and fat.
119
IP:GAPDH/WB:AcK
IP:GAPDH/WB:GAPDH
160
AcK GAPDH/GAPDH (%WT)
140
120
100
80
60
40
20
0
WT GAPDH
4KR GAPDH
Figure 5.3.2: Hepatic total GAPDH acetylation in 4KR GAPDH and WT GAPDH mice. Liver
tissue was homogenised in lysis buffer containing TSA and NAM.
Lysates were then
immunoprecipitated for GAPDH and western blotted for GAPDH and acetyl-lysine. Lysine-acetylated
GAPDH was normalised to total GAPDH and expressed as %WT GAPDH, mean ± SEM, n=8 replicates
per group. Results were not statistically significant.
120
Gluconeogenic GAPDH activity (%WT)
A
160
*
140
120
100
80
60
40
20
0
WT GAPDH
Glycolytic GAPDH activity (%WT)
B
4KR GAPDH
120
100
80
*
60
40
20
0
WT GAPDH
4KR GAPDH
Figure 5.3.3: Gluconeogenic/glycolytic GAPDH activity in 4KR GAPDH and WT GAPDH mice.
(A) Gluconeogenic GAPDH activity per mg protein. (B) Glycolytic GAPDH activity per mg protein.
All data expressed as %WT GAPDH activity, mean ± SEM, n=8 replicates per group. * Significantly
different from WT GAPDH, P < 0.05.
121
A
120%
Fractional contribution to fasting
endogenous glucose
*
100%
80%
*
Glycogenolysis
Gluconeogenesis
60%
40%
20%
0%
WT GAPDH
B
120%
4KR GAPDH
*
Fractional contribution to
gluconeogenesis
100%
80%
Glycerol
60%
Phosphoenolpyruvic acid
40%
20%
0%
WT GAPDH
4KR GAPDH
Figure 5.3.4: Percentage of endogenous glucose derived from gluconeogenesis and glycogenolysis
after fasting. (A) Fractional contribution of gluconeogenesis and glycogenolysis to total endogenous
glucose. (B) Fractional contribution to gluconeogenesis by glycerol and PEP. All data expressed as
mean ± SEM, n=8 replicates per group. * Significantly different from WT GAPDH, P < 0.05.
122
A
B
35
4.5
WT GAPDH
4KR GAPDH
30
4.0
Body mass (g)
Daily food intake (g/day/mouse)
5.0
3.5
3.0
2.5
2.0
1.5
1.0
25
20
15
10
5
0.5
0
0.0
WT GAPDH
1
4KR GAPDH
2
3
4
5
6
7
8
Week
C
14
WT GAPDH
D
4KR GAPDH
Lean mass (%body mass)
Fat mass (%body mass)
12
10
8
6
4
2
90
WT GAPDH
4KR GAPDH
85
80
75
70
65
60
55
50
0
1
2
3
4
5
6
1
7
2
3
4
5
6
7
Week
Week
Figure 5.3.5: Body composition and food consumption. (A) Average daily food consumption per
mouse. (B) Whole body mass, measured weekly. (C) Percent fat mass, measured weekly. (D) Percent
lean mass, measured weekly. All data represented as mean ± SEM, n=8 replicates per group. Results
not statistically significant.
123
A
B
16
WT GAPDH
6
4KR GAPDH
5
12
HbA1c (%)
Blood glucose (mmol)
14
10
8
6
4
3
2
4
1
2
0
0
1
2
3
4
5
Week
6
7
8
WT GAPDH
4KR GAPDH
Figure 5.3.6: Weekly blood glucose levels and percentage HbA1c. (A) Weekly blood glucose levels
measured in the fed state. (B) Percentage HbA1c levels in WT GAPDH and 4KR GAPDH mice at
twenty two weeks of age, seven weeks post AAV8 injection. All data represented as mean ± SEM, n=8
replicates per group. Results not statistically significant.
124
A
WT GAPDH
30.0
120
*
*
20.0
100
iAUC (%WT)
25.0
Blood glucose (mmol)
B
4KR GAPDH
0.053
15.0
10.0
5.0
*
60
40
20
0
0.0
0
15
30
45
Time (min)
60
90
WT GAPDH
4KR GAPDH
WT GAPDH
4KR GAPDH
D
C
1.80
WT GAPDH
160
4KR GAPDH
1.60
140
iAUC (%WT)
1.40
Insulin (ng/ml)
80
1.20
1.00
0.80
0.60
120
100
80
60
0.40
40
0.20
20
0.00
0
0
15
Time (min)
30
60
Figure 5.3.7: Glucose tolerance test (GTT). (A) Blood glucose time course after intraperitoneal
injection with glucose (2g/Kg). (B) iAUC for blood glucose time course. (C) Insulin ELISA from
plasma taken during GTT. (D) iAUC for insulin ELISA. All data represented as mean ± SEM, n=8
replicates per group. * Significantly different from WT GAPDH, P < 0.05.
125
A
WT GAPDH
20.0
140
18.0
*
*
*
120
16.0
14.0
iAUC (%WT)
Blood Glucose (mmol)
B
4KR GAPDH
*
12.0
10.0
8.0
6.0
80
0.08
60
40
4.0
20
2.0
0.0
0
15
30
45
Time (min)
60
0
90
C
WT GAPDH
4KR GAPDH
WT GAPDH
4KR GAPDH
D
2.50
WT GAPDH
140
4KR GAPDH
120
iAUC (%WT)
2.00
Insulin (ng/ml)
100
1.50
1.00
100
80
60
40
0.50
20
0.00
0
15
Time (min)
30
60
0
Figure 5.3.8: Glucagon tolerance test (GnTT). (A) Blood glucose time course after intraperitoneal
injection with glucagon (30μg/kg). (B) iAUC for blood glucose time course. (C) Plasma insulin during
GnTT. (D) iAUC for insulin ELISA. All data represented as mean ± SEM, n=8 replicates per group. *
Significantly different from WT GAPDH, P < 0.05.
126
A
B
WT GAPDH
4KR GAPDH
140
14.0
120
12.0
100
iAUC (%WT)
Blood glucose (mmol)
16.0
10.0
8.0
6.0
80
60
4.0
40
2.0
20
0.0
0
0
15
30
45
Time (min)
60
90
WT GAPDH
4KR GAPDH
Figure 5.3.9: Re-feed postprandial response. (A) Blood glucose time course after re-feeding, post
fifteen hour fast. (B) iAUC for blood glucose time course. All data represented as mean ± SEM, n=8
replicates per group. Results not statistically significant.
127
A
B
14.0
WT GAPDH
160
4KR GAPDH
140
*
10.0
iAUC (%WT)
Blood glucose (mmol)
12.0
*
8.0
6.0
120
100
80
60
4.0
40
2.0
20
0.0
0
Figure 5.3.10:
20
40
60
Time (min)
90
120
0
WT GAPDH
4KR GAPDH
Insulin tolerance test (ITT). (A) Blood glucose time course after intraperitoneal
injection with insulin (0.75U/Kg). (B) iAUC for blood glucose time course. All data represented as
mean ± SEM, n=8 replicates per group. * Significantly different from WT GAPDH, P < 0.05. Results
not statistically significant.
128
Glycogen concentration (%WT)
A
180
*
160
140
120
100
80
60
40
20
0
Glycogen concentration (%WT)
B
WT GAPDH
4KR GAPDH
WT GAPDH
4KR GAPDH
140
120
100
80
60
40
20
0
Figure 5.3.11: Muscle and liver glycogen concentration in WT GAPDH and 4KR GAPDH mice.
(A) Liver glycogen concentration. (B) Muscle glycogen concentration. All data represented as %WT
GAPDH, mean ± SEM, n=8 replicates per group. * Significantly different from WT GAPDH, P < 0.05.
129
A
X50
X100
X200
X400
WT GAPDH
4KR GAPDH
Triglyceride (nM/mg protein)
B
5.0
4.5
4.0
3.5
3.0
2.5
2.0
1.5
1.0
0.5
0.0
WT GAPDH
4KR GAPDH
Figure 5.3.12: Liver morphology and triglyceride quantification in 4KR GAPDH and WT
GAPDH mice. (A) Hematoxylin and eosin stain (H & E) of liver sections from WT GAPDH and 4KR
GAPDH mice. (B) WT GAPDH and 4KR GAPDH triglyceride concentration. All data represented as
mean ± SEM, n=8 replicates per group. Results not statistically significant.
130
5.4 DISCUSSION
Chapter Two of this thesis demonstrated that FAO hepatocytes expressing rat
GAPDH that has residues K115, K160, K225, and K252 mutated to arginine (4KR
GAPDH) increased gluconeogenesis and reduced glycolysis in vitro. Palmitate and
glucose oxidation was also reduced in 4KR GAPDH FAO hepatocytes, which also had
increased lipid deposits. Based on these results, it was expected that 4KR GAPDH
mice would have elevated blood glucose levels, with increased hepatic
gluconeogenesis and lipid deposits.
Mice were injected, via tail vein, with either AAV8 WT GAPDH-IRES-GFP
or AAV8 4KR GAPDH-IRES-GFP, and expression was demonstrated by western
blotting tissue lysates for GFP expression (Figure 5.3.1 A). AAV8 has high tropism
for the liver, and in addition, a liver specific promoter (TBG) was used to prevent
expression in other tissues148. Western blot for GFP in liver, heart, kidney, muscle,
and fat showed that expression of exogenous GAPDH occurred exclusively in the liver
(Figure 5.3.1 B). There was no significant difference in total hepatic GAPDH
acetylation between WT GAPDH and 4KR GAPDH (Figure 5.3.2), however site
specific acetyl-lysine antibodies are required to determine if rat GAPDH K115, K160,
K225, and K252 are deacetylated. This data is difficult to interpret due to the nature
of the pan acetyl-lysine antibodies, and in addition, endogenous GAPDH could be
affecting the data. Chapter Two of this thesis demonstrated that mutation of lysine
residues in 4KR GAPDH, could not be detected by pan acetyl-lysine antibodies. If
GAPDH activity is not affected by the addition of a protein tag, then
immunoprecipitation of that tag could pull down 4KR GAPDH and WT GAPDH,
which could potentially show reduced acetylation in 4KR GAPDH compared with WT
GAPDH. This study did not incorporate a tag on GAPDH to prevent any loss of
131
activity, or side effects, which might occur as a result of a tag. It is unclear whether
the exogenous GAPDH is outcompeting the endogenous GAPDH, but a GAPDH
knockout model could be used to remove endogenous GAPDH in future research. One
such method is to knockout endogenous liver GAPDH using lox-cre or CRISPR/Cas9
technology, followed by reintroduction of WT GAPDH and 4KR GAPDH.
In FAO cells 4KR GAPDH increased gluconeogenic GAPDH activity and
reduced glycolytic GAPDH activity (Figure 2.3.4 B and C). In 4KR GAPDH mice
gluconeogenic GAPDH activity was increased and glycolytic GAPDH activity was
reduced compared with WT GAPDH mice (Figure 5.3.3 A and B). This demonstrates
that GAPDH activity of 4KR GAPDH is similar in vivo and in vitro. In addition,
analysis of the fractional contribution of gluconeogenesis and glycogenolysis to total
fasting endogenous glucose, revealed that 4KR GAPDH mice had a higher
contribution from gluconeogenesis compared with WT GAPDH mice (Figure 5.3.4
A), and WT GAPDH mice had a higher contribution from glycogenolysis compared
with 4KR GAPDH mice (Figure 5.3.4 A). This further suggests that gluconeogenesis
is elevated in 4KR GAPDH mice compared with WT GAPDH mice. The major
contribution to gluconeogenesis in the 4KR GAPDH mice came from glycerol, which
was unexpected, and further complicates what is occurring in vivo in response to
mutation of GAPDH lysine residues to arginine (Figure 5.3.4 B). The preference of
4KR GAPDH mice to metabolise glycerol suggests that mutation of GAPDH lysine
residues to arginine could in fact impair GAPDH function in both the glycolytic and
gluconeogenic direction. However, both WT GAPDH mice and 4KR GAPDH mice
could metabolise PEP through gluconeogenesis, which suggests that the function of
GAPDH in the gluconeogenic direction is normal (Figure 5.3.4 B).
132
4KR GAPDH and WT GAPDH mice food consumption was not different
(Figure 5.3.5 A), and 4KR GAPDH did not affect whole body mass (Figure 5.3.5 B)
or body composition (Figure 5.3.5 C and D) compared with WT GAPDH mice.
Surprisingly, 4KR GAPDH expression had opposing effects on whole body glucose
homeostasis to those expected based on our previous analyses in vitro. C57BL/6 mice
expressing hepatic 4KR GAPDH did not have elevated blood glucose levels (Figure
5.3.6 A) or increased percentage HbA1c (Figure 5.3.6 B) compared with WT GAPDH
mice. In addition, 4KR GAPDH mice were more glucose tolerant compared with WT
GAPDH mice (Figure 5.3.7 A and B). Insulin levels after injection with glucose were
normal and did not vary significantly between WT GAPDH and 4KR GAPDH mice
(Figure 5.3.7 C and D). 4KR GAPDH mice showed greater control of glucose
homeostasis following glucagon administration than WT GAPDH mice (Figure 5.3.8
A and B). Insulin levels after injection with glucagon were normal and did not vary
significantly between WT GAPDH and 4KR GAPDH mice (Figure 5.3.8 C and D).
Based on the cell culture studies, where 4KR GAPDH cells were hypersensitive to
forskolin, it was predicted that the 4KR GAPDH mice would be hypersensitive to
glucagon. However, glucagon activates the cAMP pathway by binding to the glucagon
receptor which then increases cAMP, and forskolin activates the cAMP pathway by
activating adenylate cyclase which then increases cAMP, so the effects of these two
molecules is similar, but could produce different responses in vivo149,150. A forskolin
tolerance test could be performed to determine whether there are different glucose
homeostasis responses to forskolin compared with glucagon. Blood glucose levels
after feeding were not significantly different in 4KR GAPDH mice compared with WT
GAPDH mice (Figure 5.3.9 A and B), however fasting blood glucose levels tended to
be higher, although not significantly, in WT GAPDH mice (5.3.9 A and 5.3.10 A).
133
This data collectively suggests that mutation of lysine residues may have had
an overall impact on the function of GAPDH, where hepatic glucose production in
4KR GAPDH mice could be elevated, but glucose disposal could also be higher in
4KR GAPDH mice, which maintains normal blood glucose levels. The glucose
disposal pathways may be more efficient in 4KR GAPDH mice in order to reduce
blood glucose levels to normal, compensating for elevated gluconeogenesis.
If
gluconeogenesis and glucose disposal are elevated, then it needs to be determined
where the glucose disposal is occurring. Rather than secreting glucose, elevated
gluconeogenesis could be producing glucose that is immediately synthesised into
glycogen. Analyses of hepatic glycogen content showed that 4KR GAPDH mice had
higher hepatic glycogen content compared with WT GAPDH mice (Figure 5.3.11 A),
while 4KR GAPDH mice muscle glycogen content was not different compared with
WT GAPDH mice (Figure 5.3.11 B). This could account for the normal blood glucose
levels in 4KR GAPDH mice that have elevated gluconeogenic GAPDH activity,
suggesting that excess glucose is synthesised into glycogen, and that glycogenolysis is
down regulated. Excess glucose could also be used in the synthesis of glycerol and
lipids, however TAG quantification did not show elevated TAG levels in 4KR
GAPDH mice (Figure 5.3.12 B). A lipid tolerance test measuring the postprandial
response could also be useful to determine whether lipid storage and utilisation is
disrupted in 4KR GAPDH mice151.
The effects of 4KR GAPDH observed in vivo do not conclusively support the
hypothesis and results from Chapter Two. Isolation of primary hepatocytes expressing
4KR GAPDH could be used to perform similar in vitro experiments that were
performed in Chapter Two to help validate those results. Testing whether acetylated
GAPDH could rescue a T2D phenotype (db/db -/-), by reducing gluconeogenesis and
134
blood glucose levels, could also help validate the regulation of gluconeogenesis by
reversible GAPDH acetylation. Together, the in vivo data from this study and Chapter
Four, suggest that GAPDH deacetylation is a pathological phenotype that perhaps is
not physiologically relevant in terms of metabolic regulation. However, GAPDH
deacetylation on its own may not be sufficient to regulate metabolic pathways, and
other metabolic enzymes might need to be deacetylated/acetylated. Other GAPDH
PTMs like phosphorylation, or PTMs to other metabolic enzymes are probably
required collectively to regulate metabolic pathways in vivo.
In conclusion, mutation of hepatic GAPDH lysine residues 115, 160, 225, and
252 to arginine, increased gluconeogenic GAPDH activity but did not increase blood
glucose levels in mice. The glycogen content and the fractional contribution to
endogenous glucose from gluconeogenesis was higher in 4KR GAPDH mice. Total
hepatic GAPDH acetylation was not reduced in 4KR GAPDH mice, and 4KR GAPDH
did not alter body composition or weight. Glucose homeostasis was altered in 4KR
GAPDH mice after glucose and glucagon were administered, but insulin action was
normal. Further work is required to determine whether hepatic gluconeogenesis and
glucose secretion was actually increased in 4KR GAPDH mice.
135
CHAPTER 6
SUMMARY & CONCLUSIONS
Diabetes is emerging as one of humanity’s greatest medical challenges, and
is predicted to become the world’s 7th leading cause of death by 20302. Total deaths
from diabetes are projected to rise by 50% over the next ten years, where T2D accounts
for 90% of all diabetes worldwide2. T2D is characterised by insulin resistance and
elevated blood glucose levels, and is associated with obesity and physical inactivity.
Increased hepatic glucose production and reduced uptake of glucose by skeletal muscle
are the largest contributors to elevated blood glucose levels in T2D5. Dysfunctional
metabolism, namely increased gluconeogenesis, is the principal cause of increased
hepatic glucose production in T2D5. GAPDH is a metabolic enzyme that catalyses
one of the forward and reverse reactions of glycolysis/gluconeogenesis, and the
acetylation profile of GAPDH has been shown to be important in the regulation of
glycolysis/gluconeogenesis in bacteria and cancer cell lines19,83. This thesis has
demonstrated that hepatic GAPDH acetylation is reduced in T2D and DIO, and has a
regulatory effect on glycolysis/gluconeogenesis in vitro.
The GAPDH residues K115, K160, K225, and K252 that were mutated to
arginine were chosen based on previous proteomic studies that have shown that those
lysine
residues
can
undergo
reversible
acetylation,
and
had
functional
importance19,79,83,103. Computational analysis of crystallography studies showed that
GAPDH K115 is located in the NAD+ binding site, and that K160, K225 and K252
are located in the catalytic domain103. It has also been shown that GAPDH residues
K115, K160, K225, and K252 are located on the surface of GAPDH where KDACs,
KATs and other proteins can interact with them. Recent studies have shown that
GAPDH has multiple and diverse functions, and that the sub-cellular location of
136
GAPDH plays an important role in determining GAPDH function73,76,100. Confocal
imaging demonstrated that 4KR GAPDH and WT GAPDH were both predominantly
localised to the cytosol where glycolytic/gluconeogenic reactions occur.
Results from Chapter Two of this thesis established that mutation of GAPDH
residues K115, K160, K225, and K252 to arginine increased gluconeogenesis and
reduced glycolysis in FAO hepatocytes. Individual mutations of GAPDH residues
K225 and K252 to arginine, which are located in the GAPDH catalytic domain,
affected glucose production and ECAR, but individual mutations to GAPDH residues
K115 and K160 to arginine did not. However, the combination of GAPDH residues
K115, K160, K225, and K252 mutated to arginine (4KR GAPDH) appeared to have
the greatest effect on altering hepatic glucose metabolism. As such, the majority of
this thesis has focussed on characterising 4KR GAPDH. Glucose production and
gluconeogenic GAPDH activity were increased in 4KR GAPDH cells compared with
WT GAPDH cells. ECAR and glycolytic GAPDH activity were reduced in 4KR
GAPDH cells compared with WT GAPDH cells. This suggests that acetylation of
GAPDH residues K115, K160, K225, and K252 have regulatory properties that
determine whether GAPDH participates in glycolytic or gluconeogenic reactions.
GAPDH residues K225 and K252, located in the catalytic domain and the cleft that is
formed by the GAPDH tetramer, could regulate glycolysis/gluconeogenesis by
allowing GAPDH to bind with 3-phosphoglyceraldehyde when acetylated, resulting in
GAPDH reacting in the glycolytic direction. Alternatively, when GAPDH residues
K225 and K252 are deacetylated, GAPDH could bind to 1,3-bisphosphoglycerate,
resulting in GAPDH reacting in the gluconeogenic direction.
Computational
simulations could help support this theory, however the computer technology currently
137
available cannot simulate this mechanism of action with WT GAPDH and 4KR
GAPDH with the corresponding metabolites.
4KR GAPDH also had effects on the TCA cycle in vitro, where palmitate and
glucose dependent oxidation were reduced in 4KR GAPDH cells. Mitochondrial
energetics showed that mitochondrial function was normal, and Oil Red-O analysis
showed that there was increased lipid deposits in 4KR GAPDH cells. This suggests
that 4KR GAPDH lipid metabolism is impaired, but the mechanisms behind this
require further investigation. It is possible that impaired glycolysis is shunting glucose
through dihydroxyacetone phosphate into glycerol and lipid synthesis. The build-up
of lipids in 4KR GAPDH cells could be attributed to an increase in lipid synthesis that
outweighs lipid catabolism, or an impairment in lipid metabolism.
How lipid
metabolism is impaired is difficult to determine based on the data in this thesis, but
impaired TCA anaplerosis could account for the reduction in palmitate and glucose
dependent OCR.
Another factor to take into account is the ATP required for
gluconeogenesis to occur. Glycolysis generates two units of ATP per unit of glucose,
while gluconeogenesis requires two units of ATP to produce one unit of glucose63.
Glucose production in 4KR GAPDH cells is increased three-fold, while glycolysis is
severely impaired, so therefore the 4KR GAPDH cells are not generating ATP through
glycolysis, and require three times as much ATP to sustain the elevated
gluconeogenesis that is occurring. This suggests that the ATP that is utilised for
gluconeogenesis may be depriving other metabolic pathways of ATP, thereby reducing
their metabolic function. However, one cycle of oxidative phosphorylation generates
thirty six units of ATP, which should adequately cover the increased demand for ATP
in the 4KR GAPDH cells, although OCR was not increased in 4KR GAPDH cells.
Further experiments, such as metabolomics could quantify the intermediary
138
metabolites in order to determine if there is impaired TCA anaplerosis, while palmitate
and glucose tracer studies could determine where glucose and lipid metabolism is
impaired, and where the additional ATP is sourced from to sustain the elevated rate of
gluconeogenesis in 4KR GAPDH cells.
Glucagon and insulin are in vivo regulators of hepatic glucose production in
mammals, where hepatic insulin resistance in T2D plays a key role in increased hepatic
gluconeogenesis and glucose production that occurs in T2D. Insulin reduces hepatic
glucose production by inhibiting the expression of key gluconeogenic genes, such as
PEPCK and G-6-Pase. However, there have been several studies that have now shown
that the protein levels of key gluconeogenic proteins PEPCK and G-6-Pase are not
important in the regulation of gluconeogenesis12,68,69. The activity of PEPCK, G-6Pase and other metabolic enzymes are likely regulated by mechanisms such as PTMs.
Chapters Two and Four in this thesis investigated whether insulin and glucagon could
modulate GAPDH acetylation, in vitro and in vivo. Forskolin was used as a glucagon
mimetic in vitro, as FAO cells have resistance to glucagon98. Chapter Two showed
that insulin could suppress glucose production in both WT GAPDH and 4KR GAPDH
cells, however 4KR GAPDH cells were hyper-sensitive to forskolin with a three-fold
increase in glucose production. This suggests that the ability of insulin to suppress
glucose production is occurring independent of GAPDH acetylation, however the
actions of forskolin are amplified by GAPDH deacetylation. This further suggests that
GAPDH acetylation plays an important role in the cAMP pathway response to
glucagon and glucose production. Further experimentation with site specific acetyllysine antibodies is required to monitor the acetylation profile of WT GAPDH cells
over time in response to insulin and forskolin. The objective of Chapter Four was to
validate these findings in vivo, however the total hepatic GAPDH acetylation profile
139
did not change fifteen minutes after glucagon, norepinephrine, insulin and vehicle
were administered. This animal study needs to be repeated with site specific acetyllysine antibodies, and the acetylation profile needs to be monitored over several time
points, as the changes in GAPDH acetylation can occur rapidly152,153. It cannot be
conclusively stated that GAPDH acetylation is not affected by glucose regulating
hormones by effectively taking a snapshot at one time point at fifteen minutes. Chapter
Four did demonstrate that hepatic GAPDH acetylation is reduced in T2D and DIO,
however the relevance of this finding, and its relation to elevated glucose production
in T2D, is not clear.
Although in vivo studies in Chapter Four demonstrated that GAPDH
acetylation is reduced in T2D and DIO, mutation of GAPDH residues K115, K160,
K225, and K252 to arginine did not increase blood glucose levels in mice. However,
there are twenty six lysine residues on GAPDH which can potentially undergo
reversible acetylation, and the reduced GAPDH acetylation seen in T2D could be a
result of other GAPDH lysine residues undergoing deacetylation. The GAPDH lysine
residues that are deacetylated in T2D need to be identified to further test whether their
deacetylation regulates glycolysis/gluconeogenesis. In addition, GAPDH acetylation
analysis revealed that there was no reduction in total GAPDH acetylation in 4KR
GAPDH mice, but site specific acetyl-lysine antibodies are required to confirm
whether the GAPDH residues K115, K160, K225, and K252 are actually deacetylated
in 4KR GAPDH mice, and in T2D and DIO mouse models. It is possible that GAPDH
deacetylation is a pathological phenotype that does not play a role in elevated hepatic
gluconeogenesis and blood glucose levels.
However, it is possible that the
deacetylation of other glycolytic/gluconeogenic enzymes in combination with other
140
PTM events, in addition to deacetylated GAPDH, is required to regulate glucose
metabolism.
Interestingly, the 4KR GAPDH mice had a higher tolerance to glucose after
glucose was administered, and also produced less glucose or had higher glucose
disposal after glucagon was administered. In both glucose and glucagon tolerance tests
the insulin levels were normal and were not significantly different. It is difficult to
conclusively determine why the 4KR GAPDH mice have higher glucose tolerance
compared with WT GAPDH mice from the results in this thesis, however there are
several possibilities that could explain these results.
There was a reduction in
glycolytic GAPDH activity in 4KR GAPDH mice, and gluconeogenic GAPDH
activity was increased in the 4KR GAPDH mice, which suggests that hepatic glucose
production should be increased in the 4KR GAPDH mice, however blood glucose
levels and percentage HbA1c were normal.
Fractional quantification of total
endogenous glucose showed that 4KR GAPDH mice had a higher contribution from
gluconeogenesis and a lower contribution from glycogenolysis compared with WT
GAPDH mice. This supports the GAPDH activity data where 4KR GAPDH mice had
elevated gluconeogenic GAPDH activity, however 4KR GAPDH mice did not have
elevated blood glucose levels. Hepatic glycogen content was higher in 4KR GAPDH
mice, where reduced glycogenolysis and increased glycogenesis could be
compensating for the elevated gluconeogenesis that could be occurring in 4KR
GAPDH mice, thereby maintaining normal blood glucose levels. It is also possible
that the excess glucose could be synthesised into lipids, however TAG quantification
showed no difference in TAG levels between 4KR GAPDH and WT GAPDH mice,
however there could be an increase in lipid metabolism to account for this. In addition,
analysis of different lipid species could reveal specific changes in lipid species
141
between WT GAPDH and 4KR GAPDH. The cell culture studies did show that lipid
metabolism was impaired, although lipid metabolism could be unaffected in vivo, and
the excess glucose could be synthesised into lipids and immediately metabolised in
vivo.
Although it was shown that PEP fuelled gluconeogenesis was not different in
4KR GAPDH mice compared with WT GAPDH mice, glycerol fuelled
gluconeogenesis was higher in 4KR GAPDH mice.
This suggests that
gluconeogenesis in 4KR GAPDH mice is bypassing GAPDH via glycerol metabolism
through dihydroxyacetone phosphate (DHAP), which suggests that mutation of
GAPDH residues K115, K160, K225, and K252 to arginine has removed GAPDH
function. However, gluconeogenic GAPDH activity is elevated in 4KR GAPDH mice,
and PEP fuelled gluconeogenesis is not significantly different compared with WT
GAPDH mice. Glycerol could be required to fuel gluconeogenesis in 4KR mice due
to a depletion of gluconeogenic metabolites, which have not been replenished due to
impaired glycolysis, resulting in gluconeogenesis being shunted through the glycerol
pathway. 4KR GAPDH cells had increased lipid accumulation and reduced palmitate
oxidation, combined with increased glycerol metabolism in 4KR GAPDH mice,
suggests that 4KR GAPDH has altered lipid metabolism in vitro and in vivo. Glycerol
is an important constituent of triglycerides, where it forms the back bone of
triglycerides, and along with palmitate is important in several aspects of lipid
metabolism63. Palmitate, along with dietary fatty acids, are acyl-CoA donors that are
required for the acylation of glycerol 3-phosphate (G3P) which is an intermediary
between DHAP, glycerol, and glycerolipid, and is important for the generation of
phosphatidic acid154. In 4KR GAPDH FAO cells which lack glycogen storage, a
reduction in palmitate oxidation through the TCA cycle could be due to increased
142
demand for palmitate required for the generation of glycerol, which is required for
increased gluconeogenesis. Glucose, glycerol, and palmitic acid tracer studies could
be performed in vitro and in vivo to determine the pathways through which they are
metabolised. Interestingly, total GAPDH acetylation was reduced in T2D and DIO,
and 4KR GAPDH resulted in accumulation of lipids in vitro, and potentially increased
utilisation of lipids via glycerol metabolism in vivo. These effects could be relevant
to the progression of T2D, as it has been demonstrated that an increased accumulation
of fatty acids and dysregulation of lipid metabolism can lead to the development of
T2D5,7,126.
The findings from this research suggest that the reduction in hepatic total
GAPDH acetylation in T2D and DIO are a pathological factor rather than a
physiological factor. It should be noted that T2D and DIO are chronic conditions while
the in vivo study of GAPDH acetylation in response to hormones was an acute
condition. If reduced GAPDH acetylation in T2D and DIO is a pathological factor,
there must be elevated GAPDH deacetylation occurring due to increased activity from
GAPDH KDACs. The enzymes that deacetylate and acetylate GAPDH in vivo are
currently unknown and unsubstantiated.
Lei et al demonstrated that HDAC5
deacetylated GAPDH K254 in vitro, however this study is unsupported and did not
take into account that HDAC5 lacks intrinsic HDAC activity83. Several studies have
now shown that the class IIa HDACS, which includes HDAC4, 5, 7, and 9, do not
possess any deacetylase activity, but rather function via the recruitment of other
KDACs such as HDAC384,85. It is possible that HDAC5 could play a regulatory role
in the deacetylation of GAPDH K254, but this would be via the recruitment of other
KDACs such as HDAC3.
143
Glucose production was reduced in WT GAPDH cells after treatment with
NAM or TSA, while glucose production in 4KR GAPDH cells was not significantly
reduced. This suggests that GAPDH is potentially deacetylated by both HDACs and
SIRTs. The focus of this thesis was on HDAC6, due to its cytosolic localisation and
because it is highly expressed in liver and kidney. Chapter Three, and previous
research, has demonstrated that HDAC6 plays a role in metabolism, and that
deacetylation of proteins by HDAC6 results in increased glucose production and
reduced ECAR and OCR. This demonstrates that HDAC6 has similar effects to 4KR
GAPDH, and suggests that HDAC6 could cause these metabolic alterations via
deacetylation of GAPDH. Further glucose production assays using cells that coexpress HDAC6 and GAPDH mutants showed that DN HDAC6 was able to reduce
glucose production in WT GAPDH cells but not 4KR GAPDH cells. In addition, 4KR
GAPDH is hypersensitive to forskolin, and DN HDAC6 was able to reduce glucose
production in WT GAPDH cells after treatment with forskolin but not 4KR GAPDH
cells.
This suggests that HDAC6 mediates the forskolin response via GAPDH
deacetylation, as the decrease in glucose production in WT GAPDH/DN HDAC6
forskolin treated cells is lost. Together these results demonstrate that HDAC6 could
be a GAPDH deacetylase.
HDAC6 has metabolic regulatory properties, and could potentially be a
GAPDH deacetylase, but the proteins that are actually deacetylated by HDAC6 and
that play a role in elevated glucose production, require further investigation. In
addition, the factors that regulate HDAC6 activity remain largely unknown, although
HDAC6 deacetylase activity is zinc dependent111. Zinc is a co-factor for the activity
and function of up to ten percent of mammalian proteins155. Elevated zinc levels, or a
deficiency in zinc levels, both have catastrophic consequences that have been linked
144
to diabetes and stroke155. Zinc homeostasis, and the proteins responsible for its
transport and their function, regulation, and crosstalk by which they operate, is poorly
understood155.
Defects in the ZnT-8 zinc transporter in the pancreas has been
associated with T2D and β-cell dysfunction, and it is possible that there is dysfunction
in other zinc transporters in T2D which could allow increased HDAC6 activity and
GAPDH deacetylation155.
Studies in this thesis demonstrated that hepatic GAPDH acetylation was
reduced in T2D, and that HDAC6 increased glucose production. The role of GAPDH
acetylation and HDAC6 in hepatic glucose metabolism requires further investigation,
but it could become a signalling axis that is an important therapeutic target for the
treatment of T2D. However there were several limitations to this project. Pull down
of GAPDH from cell lysates after exposure to high glucose was not as efficient as
GAPDH pull down from glucose free and normal glucose cell lysates. This could be
explained by GAPDH translocating to the nucleus to initiate apoptosis in high glucose
conditions156,157.
There were also limitations to immunoprecipitations and
immunoblotting using pan acetyl-lysine antibodies, which are not effective for
investigating individual lysine residue acetylation, and can have variability between
batches. For this reason the use of site specific acetyl-lysine antibodies is now
emerging as an essential requisite for acetylation biology.
Mass spectrometry
techniques, such as SILAC, have also emerged as important tools for acetylation
biology.
The mutation of lysine residues to arginine has been commonly used to mimic
deacetylated lysine residues in acetylation studies, while mutation of lysine residues
to glutamine has been commonly used to mimic acetylated lysine residues. However,
it is now becoming apparent that these mutations do not accurately reflect the
145
acetylated/deacetylated lysine residue, and actually differ substantially in form and
function64. Therefore analyses of results from lysine mutation studies should take this
into account when analysing the functional importance of lysine mutations64. In
addition, lysine residues undergo multiple PTMs by several acyl moieties including:
formylation, succinylation, malonyation, butyrylation, propionylation, glutarylation,
and crotonylation64. Mutation of lysine residues could therefore have an impact on the
modifications listed above, which would make it difficult to attribute any functional
changes derived from such mutations solely to acetylation.
In summary, the conclusions that can be drawn from the results presented in this thesis
are:
1. Total GAPDH acetylation is reduced in T2D and DIO in vivo
2. Mutation of rat GAPDH residues K115, K160, K225, and K252 to arginine
-
increased gluconeogenesis and reduced glycolysis in vitro
-
increased lipid deposits in vitro
-
impaired glucose and palmitate oxidation in vitro
-
did not alter blood glucose levels or percentage HbA1c in vivo
-
affected glucose and glucagon tolerance after injection with glucose or
glucagon
-
increased gluconeogenic GAPDH activity and reduced glycolytic GAPDH
activity in vivo
-
increased hepatic glycogen concentration in vivo
3. WT HDAC6 increased glucose production and reduced glycolysis in vitro
Since its discovery, acetylation has been considered as a modification that
occurred exclusively on histones, until recent proteomic studies have shown that the
146
majority of acetylation events occur on non-nuclear proteins20,21. An interesting
finding from those studies is that acetylation is widespread in the mitochondria, and
considering the lack of phosphorylation events in the mitochondria, acetylation likely
plays an important role in mitochondrial function158,159. Acetylation compared to
phosphorylation is also more abundant in bacteria, and this supports the mitochondrial
endosymbiotic theory of prokaryote origin159. Whether acetylation in mammals plays
a regulatory role outside of the mitochondria that is as important as it is in bacteria is
yet to be proven, and it remains to be seen whether lysine acetylation sites are
conserved in mammals due to their function, or whether this is just a consequence of
their frequent location in conserved structural features64. It is possible that the
acetylation profile of metabolic enzymes collectively can regulate metabolic
pathways, but not individually, as it is probable that there are compensatory
mechanisms in place to prevent that from occurring. However, the outcomes from this
work
could
provide
one
piece
of
the
puzzle
for
the
regulation
of
glycolysis/gluconeogenesis by reversible acetylation of metabolic enzymes. Further
investigation could lead to KDAC inhibitors that can be used in conjunction with other
treatments such as metformin to reduce blood glucose levels in T2D.
147
BIBLIOGRAPHY
1.
2.
3.
4.
5.
6.
7.
8.
9.
10.
11.
12.
13.
14.
15.
16.
17.
18.
Australian Diabetes Council, 2010, Diabetes facts, www.diabetesnsw.com.au.
World Health Originisation, 2011, Type 2 Diabetes: Fact sheet N°312,
http://www.who.int/mediacentre/factsheets/fs312/en/.
Kahn, M, C.R., Vicent, M, D. & Doria, M, Ph. D, A., Genetics of noninsulin-dependent (type-II) diabetes mellitus, Annual review of medicine 47,
509-531 (1996).
World Health Orginisation, 2006, definition and diagnosis of diabetes
mellitus and intermediate hyperglycemia, http://www.who.int/en/.
Stumvoll, M, Goldstein, BJ & Van-Haeften, TW, Type 2 diabetes: principles
of pathogenesis and therapy, Lancet 365, 1333-1346 (2005).
Morino, K, Petersen, KF & Shulman, GI, Molecular mechanisms of insulin
resistance in humans and their potential links with mitochondrial dysfunction,
Diabetes 55, S9-S15 (2006).
Muoio, DM & Newgard, CB, Molecular and metabolic mechanisms of
insulin resistance and β-cell failure in type 2 diabetes, Nature Reviews
molecular cell biology 9, 193-205 (2008).
Pessin, JE & Saltiel, AR, Signaling pathways in insulin action: molecular
targets of insulin resistance, Journal of Clinical Investigation 106, 165-170
(2000).
Herman, MA & Kahn, BB, Glucose transport and sensing in the maintenance
of glucose homeostasis and metabolic harmony, Journal of Clinical
Investigation 116, 1767 (2006).
Greenberg, AS & McDaniel, ML, Identifying the links between obesity,
insulin resistance and β-cell function: potential role of adipocyte-derived
cytokines in the pathogenesis of type 2 diabetes, European Journal of
Clinical Investigation 32, 24-34 (2002).
Dandona, P, Aljada, A & Bandyopadhyay, A, Inflammation: the link between
insulin resistance, obesity and diabetes, Trends in immunology 25, 4-7
(2004).
Edgerton, DS, et al., Effects of insulin on the metabolic control of hepatic
gluconeogenesis in vivo, Diabetes 58, 2766-2775 (2009).
Magnusson, I, Rothman, D, Katz, L, Shulman, R & Shulman, G, Increased
rate of gluconeogenesis in type II diabetes mellitus. A 13C nuclear magnetic
resonance study, Journal of Clinical Investigation 90, 1323 (1992).
Roden, M & Bernroider, E, Hepatic glucose metabolism in humans—its role
in health and disease, Best Practice & Research Clinical Endocrinology &
Metabolism 17, 365-383 (2003).
Shin, J, Zhang, D & Chen, D, Reversible Acetylation of Metabolic Enzymes
Celebration: SIRT2 and p300 Join the Party, Molecular cell 43, 3-5 (2011).
Katada, S, Imhof, A & Sassone-Corsi, P, Connecting Threads: Epigenetics
and Metabolism, Cell 148, 24-28 (2012).
Guan, K & Xiong, Y, Regulation of intermediary metabolism by protein
acetylation, Trends in Biochemical Sciences 36, 108-116 (2011).
Patel, J, Pathak, RR & Mujtaba, S, The biology of lysine acetylation
integrates transcriptional programming and metabolism, Nutrition &
Metabolism 8, 12 (2011).
148
19.
20.
21.
22.
23.
24.
25.
26.
27.
28.
29.
30.
31.
32.
33.
34.
35.
36.
37.
38.
39.
40.
41.
Wang, Q, et al., Acetylation of metabolic enzymes coordinates carbon source
utilization and metabolic flux, Science 327, 1004 (2010).
Kim, SC, et al., Substrate and functional diversity of lysine acetylation
revealed by a proteomics survey, Molecular cell 23, 607-618 (2006).
Choudhary, C, et al., Lysine acetylation targets protein complexes and coregulates major cellular functions, Science's STKE 325, 834 (2009).
Zhao, S, et al., Regulation of Cellular Metabolism by Protein Lysine
Acetylation, Science 327, 1000-1004 (2010).
Kim, GW & Yang, XJ, Comprehensive lysine acetylomes emerging from
bacteria to humans, Trends in biochemical sciences 36, 211-220 (2011).
Thao, S & Escalante-Semerena, JC, Control of protein function by reversible
N-lysine acetylation in bacteria, Current opinion in microbiology (2011).
Yang, X & Seto, E, HATs and HDACs: from structure, function and
regulation to novel strategies for therapy and prevention, Oncogene 26, 53105318 (2007).
Christensen, DP, et al., Histone deacetylase (HDAC) inhibition as a novel
treatment for diabetes mellitus, Molecular Medicine 17, 378 (2011).
Cherrington, AD & Lecture, B, Control of glucose uptake and release by the
liver in vivo, Diabetes New York 48, 1198-1214 (1999).
Alemán, JO, Gluconeogenesis as a system: development of in vivo flux
analysis of hepatic glucose production in Type 2 Diabetes, (Massachusetts
Institute of Technology, 2008).
Gastaldelli, A, et al., Influence of obesity and type 2 diabetes on
gluconeogenesis and glucose output in humans: a quantitative study, Diabetes
49, 1367-1373 (2000).
Van Schaftingen, E & Gerin, I, The glucose-6-phosphatase system, Biochem.
J 362, 513-532 (2002).
White, MF, Insulin signaling in health and disease, Science's STKE 302, 1710
(2003).
Saltiel, AR & Kahn, CR, Insulin signalling and the regulation of glucose and
lipid metabolism, Nature 414, 799-806 (2001).
Le lay, J & Kaestner, KH, The Fox genes in the liver: from organogenesis to
functional integration, Physiological reviews 90, 1-22 (2010).
Brynildsen, M, Wong, W & Liao, J, Transcriptional regulation and
metabolism, Biochemical Society Transactions 33, 1423-1426 (2005).
Desvergne, B, Michalik, L & Wahli, W, Transcriptional regulation of
metabolism, Physiological reviews 86, 465-514 (2006).
Yabaluri, N & Bashyam, MD, Hormonal regulation of gluconeogenic gene
transcription in the liver, Journal of biosciences 35, 473-484 (2010).
Shulman, GI, Cellular mechanisms of insulin resistance, Journal of Clinical
Investigation 106, 171-176 (2000).
Samuel, VT & Shulman, GI, Mechanisms for Insulin Resistance: Common
Threads and Missing Links, Cell 148, 852-871 (2012).
Hoehn, KL, et al., IRS1-independent defects define major nodes of insulin
resistance, Cell metabolism 7, 421-433 (2008).
Barthel, A & Schmoll, D, Novel concepts in insulin regulation of hepatic
gluconeogenesis, American Journal of Physiology-Endocrinology And
Metabolism 285, E685-E692 (2003).
Iyer, A, Fairlie, DP & Brown, L, Lysine acetylation in obesity, diabetes and
metabolic disease, Immunology and Cell Biology 90, 39-46 (2011).
149
42.
43.
44.
45.
46.
47.
48.
49.
50.
51.
52.
53.
54.
55.
56.
57.
58.
59.
60.
61.
Zhang, J, et al., Lysine acetylation is a highly abundant and evolutionarily
conserved modification in Escherichia coli, Molecular & Cellular Proteomics
8, 215-225 (2009).
Kronenberg, HM, Memed, S, Polonsky, KS & Larsen, PR, Williams Textbook
of Endocrinology, Elsevier, 2010.
Allfrey, V, Faulkner, R & Mirsky, A, Acetylation and methylation of histones
and their possible role in the regulation of RNA synthesis, Proceedings of the
National Academy of Sciences of the United States of America 51, 786
(1964).
Norris, KL, Lee, JY & Yao, TP, Acetylation goes global: the emergence of
acetylation biology, Science's STKE 2, pe76 (2009).
Kouzarides, T, Acetylation: a regulatory modification to rival
phosphorylation?, Science's STKE 19, 1176 (2000).
Scott, I, Regulation of cellular homoeostasis by reversible lysine acetylation,
Essays in Biochemistry 52, 13-22 (2012).
Wagner, GR & Payne, RM, Mitochondrial acetylation and diseases of aging,
Journal of aging research (2011).
Close, P, et al., The emerging role of lysine acetylation of non-nuclear
proteins, Cellular and molecular life sciences 67, 1255-1264 (2010).
Guarente, L, The Logic Linking Protein Acetylation and Metabolism, Cell
Metabolism 14, 151-153 (2011).
Zhang, J, et al., Lysine acetylation is a highly abundant and evolutionarily
conserved modification in Escherichia coli, Molecular & Cellular Proteomics
8, 215-225 (2009).
Wagner, GR & Payne, RM, Widespread and enzyme-independent Nϵacetylation and Nϵ-succinylation of proteins in the chemical conditions of the
mitochondrial matrix, Journal of Biological Chemistry 288, 29036-29045
(2013).
Henriksen, P, et al., Proteome-wide analysis of lysine acetylation suggests its
broad regulatory scope in Saccharomyces cerevisiae, Molecular & Cellular
Proteomics 11, 1510-1522 (2012).
Svinkina, T, et al., Deep, quantitative coverage of the lysine acetylome using
novel anti-acetyl-lysine antibodies and an optimized proteomic workflow,
Molecular & Cellular Proteomics, mcp. O114. 047555 (2015).
Hoedt, E, Zhang, G & Neubert, TA, Stable isotope labeling by amino acids in
cell culture (SILAC) for quantitative proteomics, in Advancements of Mass
Spectrometry in Biomedical Research 93-106 Springer, 2014.
Xu, W & Zhao, S, Metabolism is regulated by protein acetylation, Frontiers
in Biology 6, 213-218 (2011).
Gerosa, L & Sauer, U, Regulation and control of metabolic fluxes in
microbes, Current opinion in biotechnology (2011).
Guan, KL & Xiong, Y, Regulation of intermediary metabolism by protein
acetylation, Trends in biochemical sciences 36, 108-116 (2011).
Jones, JD & O'Connor, CD, Protein acetylation in prokaryotes, Proteomics
(2011).
Wellen, KE & Thompson, CB, A two-way street: reciprocal regulation of
metabolism and signalling, Nature Reviews molecular cell biology (2012).
Oliveira, AP & Sauer, U, The importance of post translational modifications
in regulating Saccharomyces cerevisiae metabolism, FEMS Yeast Research
(2012).
150
62.
63.
64.
65.
66.
67.
68.
69.
70.
71.
72.
73.
74.
75.
76.
77.
78.
79.
Metallo, CM & Vander Heiden, MG, Metabolism strikes back: metabolic flux
regulates cell signaling, Genes & development 24, 2717-2722 (2010).
Horton, R, Moran, L, Scrimgeour, G, Perry, M & Rawn, D, Principles of
Biochemistry, fourth edition, Pearson Prentice Hall, New Jersey, 2006.
Choudhary, C, Weinert, BT, Nishida, Y, Verdin, E & Mann, M, The growing
landscape of lysine acetylation links metabolism and cell signalling, Nature
Reviews Molecular Cell Biology 15, 536-550 (2014).
Glozak, MA, Sengupta, N, Zhang, X & Seto, E, Acetylation and
deacetylation of non-histone proteins, Gene 363, 15-23 (2005).
Khan, SN & Khan, AU, Role of histone acetylation in cell physiology and
diseases: An update, Clinica Chimica Acta 411, 1401-1411 (2010).
Tripodi, F, Nicastro, R, Reghellin, V & Coccetti, P, Post-translational
modifications on yeast carbon metabolism: Regulatory mechanisms beyond
transcriptional control, Biochimica et Biophysica Acta (BBA)-General
Subjects 1850, 620-627 (2015).
Samuel, VT, et al., Fasting hyperglycemia is not associated with increased
expression of PEPCK or G6Pc in patients with Type 2 Diabetes, Proceedings
of the National Academy of Sciences 106, 12121-12126 (2009).
Ramnanan, CJ, et al., Molecular characterization of insulin-mediated
suppression of hepatic glucose production in vivo, Diabetes 59, 1302-1311
(2010).
Goh, KP & Sum, CF, Connecting the dots: molecular and epigenetic
mechanisms in type 2 diabetes, Current diabetes reviews 6, 255-265 (2010).
Kruger, NJ & von Schaewen, A, The oxidative pentose phosphate pathway:
structure and organisation, Current opinion in plant biology 6, 236-246
(2003).
Chen, HC & Farese, RV, Inhibition of triglyceride synthesis as a treatment
strategy for obesity - Lessons from DGAT1-deficient mice, Arterioscl Throm
Vas 25, 482-486 (2005).
Seidler, NW, GAPDH: Biological Properties and Diversity Introduction,
(Springer 233 Spring st, NY 10013, United States, 2013).
Sirover, MA, New insights into an old protein: the functional diversity of
mammalian glyceraldehyde-3-phosphate dehydrogenase, Biochimica et
Biophysica Acta (BBA)-Protein Structure and Molecular Enzymology 1432,
159-184 (1999).
Seidler, NW, Basic biology of GAPDH, in GAPDH: Biological Properties
and Diversity 1-36 Springer, 2013.
Tristan, C, Shahani, N, Sedlak, TW & Sawa, A, The diverse functions of
GAPDH: views from different subcellular compartments, Cellular signalling
23, 317-323 (2011).
Hara, MR, et al., S-nitrosylated GAPDH initiates apoptotic cell death by
nuclear translocation following Siah1 binding, Nature cell biology 7, 665-674
(2005).
Sirover, MA, Subcellular dynamics of multifunctional protein regulation:
mechanisms of GAPDH intracellular translocation, Journal of cellular
biochemistry 113, 2193-2200 (2012).
Ventura, M, et al., Nuclear translocation of glyceraldehyde-3-phosphate
dehydrogenase is regulated by acetylation, The International Journal of
Biochemistry & Cell Biology 42, 1672-1680 (2010).
151
80.
81.
82.
83.
84.
85.
86.
87.
88.
89.
90.
91.
92.
93.
94.
95.
96.
Kawaguchi, T, Takenoshita, M, Kabashima, T & Uyeda, K, Glucose and
cAMP regulate the L-type pyruvate kinase gene by
phosphorylation/dephosphorylation of the carbohydrate response element
binding protein, Proceedings of the National Academy of Sciences 98, 1371013715 (2001).
Hsiao, L-L, et al., A compendium of gene expression in normal human
tissues, Physiological genomics 7, 97-104 (2001).
Hallows, WC, Yu, W & Denu, JM, Regulation of glycolytic enzyme
phosphoglycerate mutase-1 by Sirt1 protein-mediated deacetylation, Journal
of Biological Chemistry 287, 3850-3858 (2012).
Li, T, et al., Glyceraldehyde-3-phosphate dehydrogenase is activated by
lysine 254 acetylation in response to glucose signal, Journal of Biological
Chemistry 289, 3775-3785 (2014).
Fischle, W, et al., Enzymatic activity associated with class II HDACs is
dependent on a multiprotein complex containing HDAC3 and SMRT/N-CoR,
Molecular cell 9, 45-57 (2002).
Lahm, A, et al., Unraveling the hidden catalytic activity of vertebrate class
IIa histone deacetylases, Proceedings of the National Academy of Sciences
104, 17335-17340 (2007).
Roth, SY, Denu, JM & Allis, CD, Histone acetyltransferases, Annual review
of biochemistry 70, 81-120 (2001).
Xu, W, Parmigiani, R & Marks, P, Histone deacetylase inhibitors: molecular
mechanisms of action, Oncogene 26, 5541-5552 (2007).
Winkler, R, et al., Histone Deacetylase 6 (HDAC6) Is an Essential Modifier
of Glucocorticoid-Induced Hepatic Gluconeogenesis, Diabetes 61, 513-523
(2012).
Rodgers, JT, et al., Nutrient control of glucose homeostasis through a
complex of PGC-1α and SIRT1, Nature 434, 113-118 (2005).
Erion, DM, et al., SirT1 knockdown in liver decreases basal hepatic glucose
production and increases hepatic insulin responsiveness in diabetic rats,
Proceedings of the National Academy of Sciences 106, 11288-11293 (2009).
Spange, S, Wagner, T, Heinzel, T & Krämer, OH, Acetylation of non-histone
proteins modulates cellular signalling at multiple levels, The International
Journal of Biochemistry & Cell Biology 41, 185-198 (2009).
van Noort, V, et al., Cross-talk between phosphorylation and lysine
acetylation in a genome-reduced bacterium, Molecular Systems Biology
8(2012).
Yang, XJ & Seto, E, Lysine acetylation: codified crosstalk with other
posttranslational modifications, Molecular cell 31, 449-461 (2008).
Charbonnel, B, Karasik, A, Liu, J, Wu, M & Meininger, G, Efficacy and
safety of the dipeptidyl peptidase-4 inhibitor sitagliptin added to ongoing
metformin therapy in patients with type 2 diabetes inadequately controlled
with metformin alone, Diabetes care 29, 2638-2643 (2006).
National Center for Biotechnology Information, U.S. National Library of
Medicine, 8600 Rockville Pike, Bethesda MD, 20894 USA (2015).
Waterhouse, AM, Procter, J.B., Martin, D.M.A, Clamp, M. and Barton, G. J.,
Jalview Version 2 - a multiple sequence alignment editor and analysis
workbench, Bioinformatics25 (29) 1189-1191 doi:
1110.1093/bioinformatics/btp1033 (2009).
152
97.
98.
99.
100.
101.
102.
103.
104.
105.
106.
107.
108.
109.
110.
111.
112.
113.
PRIPǦBUUS, C, et al., Evidence for an impaired longǦchain fatty acid
oxidation and ketogenesis in Fao hepatoma cells, European Journal of
Biochemistry 209, 291-298 (1992).
Fehlmann, M, Crettaz, M & Kahn, CR, Glucagon resistance of hepatoma
cells. Evidence for receptor and post-receptor defects, The Biochemical
journal 214, 845-850 (1983).
Mehlem, A, Hagberg, CE, Muhl, L, Eriksson, U & Falkevall, A, Imaging of
neutral lipids by oil red O for analyzing the metabolic status in health and
disease, Nature protocols 8, 1149-1154 (2013).
Seidler, NW, Compartmentation of GAPDH, in GAPDH: Biological
Properties and Diversity 61-101 Springer, 2013.
Zannat, T, Bhattacharjee, RB & Bag, J, In the absence of cellular poly (A)
binding protein, the glycolytic enzyme GAPDH translocated to the cell
nucleus and activated the GAPDH mediated apoptotic pathway by enhancing
acetylation and serine 46 phosphorylation of p53, Biochemical and
Biophysical Research Communications (2011).
Huang, W, Wang, Z & Lei, Q-Y, Acetylation control of metabolic enzymes
in cancer: an updated version, Acta biochimica et biophysica Sinica 46, 204213 (2014).
Jenkins, JL & Tanner, JJ, High-resolution structure of human Dglyceraldehyde-3-phosphate dehydrogenase, Acta Crystallographica Section
D: Biological Crystallography 62, 290-301 (2006).
Fehlmann, M, Crettaz, M & Kahn, CR, Glucagon resistance of hepatoma
cells. Evidence for receptor and post-receptor defects, Biochem. J 214, 845850 (1983).
Lee, Y, et al., Hyperglycemia in rodent models of type 2 diabetes requires
insulin-resistant alpha cells, Proceedings of the National Academy of
Sciences 111, 13217-13222 (2014).
Cantley, J & Ashcroft, FM, Q&A: insulin secretion and type 2 diabetes: why
do β-cells fail?, BMC biology 13, 33 (2015).
Chang, E & He, L, Antagonistic Effects of Insulin Signaling and Glucagon
Signaling on Controlling Hepatic Gluconeogenic Gene Expression, Pediat
Therapeut 4, 2161-0665.1000200 (2014).
Rajendran, R, et al., Acetylation mediated by the p300/CBP-associated factor
determines cellular energy metabolic pathways in cancer, International
journal of oncology 42, 1961 (2013).
Tang, J, Yan, H & Zhuang, S, Histone deacetylases as targets for treatment of
multiple diseases, Clinical Science 124, 651-662 (2013).
Valenzuela-Fernández, A, Cabrero, JR, Serrador, JM & Sánchez-Madrid, F,
HDAC6: a key regulator of cytoskeleton, cell migration and cell–cell
interactions, Trends in cell biology 18, 291-297 (2008).
Hubbert, C, et al., HDAC6 is a microtubule-associated deacetylase, Nature
417, 455-458 (2002).
Ya-Li, Y & Wen-Ming, Y, Beyond histone and deacetylase: an overview of
cytoplasmic histone deacetylases and their nonhistone substrates, Journal of
Biomedicine and Biotechnology 2011(2010).
North, BJ, Marshall, BL, Borra, MT, Denu, JM & Verdin, E, The Human
Sir2 Ortholog, SIRT2, Is an NAD+-Dependent Tubulin Deacetylase,
Molecular cell 11, 437-444 (2003).
153
114.
115.
116.
117.
118.
119.
120.
121.
122.
123.
124.
125.
126.
127.
128.
129.
130.
131.
132.
133.
Li, Y, Shin, D & Kwon, SH, HDAC6 plays a role as a distinct regulator of
diverse cellular processes, FEBS Journal (2012).
Kovacs, JJ, et al., HDAC6 regulates Hsp90 acetylation and chaperonedependent activation of glucocorticoid receptor, Molecular cell 18, 601-607
(2005).
Chen, S, Owens, GC, Makarenkova, H & Edelman, DB, HDAC6 Regulates
Mitochondrial Transport in Hippocampal Neurons, PloS one 5, e10848
(2010).
Li, Y, Shin, D & Kwon, SH, Histone deacetylase 6 plays a role as a distinct
regulator of diverse cellular processes, FEBS Journal (2012).
Berstein, LM, Metformin in obesity, cancer and aging: addressing
controversies, Aging (Albany NY) 4, 320 (2012).
Rena, G, Pearson, ER & Sakamoto, K, Molecular mechanism of action of
metformin: old or new insights?, Diabetologia 56, 1898-1906 (2013).
Madiraju, AK, et al., Metformin suppresses gluconeogenesis by inhibiting
mitochondrial glycerophosphate dehydrogenase, Nature 510, 542-546 (2014).
Zhou, G, et al., Role of AMP-activated protein kinase in mechanism of
metformin action, Journal of clinical investigation 108, 1167 (2001).
Miller, RA, et al., Biguanides suppress hepatic glucagon signalling by
decreasing production of cyclic AMP, Nature 494, 256-260 (2013).
Gao, Y-s, Hubbert, CC & Yao, T-P, The microtubule-associated histone
deacetylase 6 (HDAC6) regulates epidermal growth factor receptor (EGFR)
endocytic trafficking and degradation, Journal of Biological Chemistry 285,
11219-11226 (2010).
Mihaylova, MM, et al., Class IIa histone deacetylases are hormone-activated
regulators of FOXO and mammalian glucose homeostasis, Cell 145, 607-621
(2011).
Kamemura, K, et al., Depression of mitochondrial metabolism by
downregulation of cytoplasmic deacetylase, HDAC6, FEBS letters (2012).
Boden, G & Shulman, G, Free fatty acids in obesity and type 2 diabetes:
defining their role in the development of insulin resistance and βǦcell
dysfunction, European journal of clinical investigation 32, 14-23 (2002).
Sen, N, et al., Nitric oxide-induced nuclear GAPDH activates p300/CBP and
mediates apoptosis, Nature cell biology 10, 866-873 (2008).
Meyer, C, et al., Abnormal renal and hepatic glucose metabolism in type 2
diabetes mellitus, Journal of Clinical Investigation 102, 619 (1998).
Martin, B, et al., Role of glucose and insulin resistance in development of
type 2 diabetes mellitus: results of a 25-year follow-up study, The Lancet
340, 925-929 (1992).
Colell, A, Green, DR & Ricci, J-E, Novel roles for GAPDH in cell death and
carcinogenesis, Cell Death & Differentiation 16, 1573-1581 (2009).
Hirschey, MD, et al., SIRT3 deficiency and mitochondrial protein
hyperacetylation accelerate the development of the metabolic syndrome,
Molecular cell 44, 177-190 (2011).
Tu, P, et al., Liver histone H3 methylation and acetylation may associate with
type 2 diabetes development, Journal of physiology and biochemistry 71, 8998 (2015).
Haggarty, SJ, Koeller, KM, Wong, JC, Grozinger, CM & Schreiber, SL,
Domain-selective small-molecule inhibitor of histone deacetylase 6
154
134.
135.
136.
137.
138.
139.
140.
141.
142.
143.
144.
145.
146.
147.
148.
149.
150.
151.
(HDAC6)-mediated tubulin deacetylation, Proceedings of the National
Academy of Sciences 100, 4389 (2003).
Serrador, JM, et al., HDAC6 deacetylase activity links the tubulin
cytoskeleton with immune synapse organization, Immunity 20, 417-428
(2004).
Tran, ADA, et al., HDAC6 deacetylation of tubulin modulates dynamics of
cellular adhesions, Journal of cell science 120, 1469-1479 (2007).
Matsuyama, A, et al., In vivo destabilization of dynamic microtubules by
HDAC6-mediated deacetylation, Science's STKE 21, 6820 (2002).
Liu, Y, Peng, L, Seto, E, Huang, S & Qiu, Y, Modulation of Histone
Deacetylase 6 (HDAC6) Nuclear Import and Tubulin Deacetylase Activity
through Acetylation, Journal of Biological Chemistry 287, 29168-29174
(2012).
Sol, EM, et al., Proteomic investigations of lysine acetylation identify diverse
substrates of mitochondrial deacetylase sirt3, PLoS One 7, e50545 (2012).
Ellis, DJ, Yuan, Z & Seto, E, Determination of protein lysine deacetylation,
Current Protocols in Protein Science, 14.12. 11-14.12. 14 (2008).
Weinert, BT, et al., Acetylation dynamics and stoichiometry in
Saccharomyces cerevisiae, Molecular Systems Biology 10(2014).
Jiang, W, et al., Acetylation regulates gluconeogenesis by promoting
PEPCK1 degradation via recruiting the UBR5 ubiquitin ligase, Molecular
cell 43, 33-44 (2011).
Cohen, T & Yao, TP, AcK-knowledge reversible acetylation, Science's STKE
2004, pe42 (2004).
Xiong, Y, Lei, Q, Zhao, S & Guan, K, Regulation of Glycolysis and
Gluconeogenesis by Acetylation of PKM and PEPCK, (Cold Spring Harbor
Laboratory Press, 2011).
Knapik, JJ, et al., Influence of fasting on carbohydrate and fat metabolism
during rest and exercise in men, Journal of Applied Physiology 64, 19231929 (1988).
Antoniewicz, MR, Kelleher, JK & Stephanopoulos, G, Measuring deuterium
enrichment of glucose hydrogen atoms by gas chromatography/mass
spectrometry, Analytical chemistry 83, 3211-3216 (2011).
Zhang, P, Analysis of Mouse Liver Glycogen Content, Bio-protocol 2, e186
(2012).
Passonneau, J & Lauderdale, V, A comparison of three methods of glycogen
measurement in tissues, Analytical biochemistry 60, 405-412 (1974).
Gao, G, et al., Biology of AAV serotype vectors in liver-directed gene
transfer to nonhuman primates, Molecular Therapy 13, 77-87 (2006).
Seamon, K & Daly, JW, Activation of adenylate cyclase by the diterpene
forskolin does not require the guanine nucleotide regulatory protein, Journal
of Biological Chemistry 256, 9799-9801 (1981).
Seamon, KB, Padgett, W & Daly, JW, Forskolin: unique diterpene activator
of adenylate cyclase in membranes and in intact cells, Proceedings of the
National Academy of Sciences 78, 3363-3367 (1981).
Dekker, MJ, et al., New oral fat tolerance tests feature tailoring of the
polyunsaturated/saturated fatty acid ratio to elicit a specific postprandial
response, Applied Physiology, Nutrition, and Metabolism 32, 1073-1081
(2007).
155
152.
153.
154.
155.
156.
157.
158.
159.
Katan-Khaykovich, Y & Struhl, K, Dynamics of global histone acetylation
and deacetylation in vivo: rapid restoration of normal histone acetylation
status upon removal of activators and repressors, Genes & development 16,
743-752 (2002).
Josling, GA, Selvarajah, SA, Petter, M & Duffy, MF, The role of
bromodomain proteins in regulating gene expression, Genes 3, 320-343
(2012).
Foster, DA, Phosphatidic acid and lipid-sensing by mTOR, Trends in
Endocrinology & Metabolism 24, 272-278 (2013).
Sekler, I, Sensi, SL, Hershfinkel, M & Silverman, WF, Mechanism and
regulation of cellular zinc transport, Molecular medicine 13, 337 (2007).
Leisner, TM, Moran, C, Holly, SP & Parise, LV, CIB1 prevents nuclear
GAPDH accumulation and non-apoptotic tumor cell death via AKT and ERK
signaling, Oncogene 32, 4017-4027 (2013).
Gao, X, et al., NleB, a bacterial effector with glycosyltransferase activity,
targets GAPDH function to inhibit NF-κB activation, Cell host & microbe 13,
87-99 (2013).
Weinert, BT, et al., Proteome-wide mapping of the Drosophila acetylome
demonstrates a high degree of conservation of lysine acetylation, Science's
STKE 4, ra48 (2011).
Gnad, F, et al., Evolutionary constraints of phosphorylation in eukaryotes,
prokaryotes, and mitochondria, Molecular & Cellular Proteomics 9, 26422653 (2010).
156
APPENDIX
SOLUTIONS AND REAGENTS
Concentrations in Molarity or Percentage by Weight/Volume.
Reagent
10 X Running Buffer
(SDS-PAGE)
Reagent
10 X Transfer Buffer
(SDS-PAGE)
Reagent
2 X 1.5 mm 8% Tris-HCl
gels
(Separating/Resolving
Gel)
Chemical/Substance
Amount
Tris
Glycine
SDS
Milli-Q H2O
60 g
288 g
20 g
Up to 2 L
Chemical/Substance
Amount
Tris
Glycine
Milli-Q H2O
60 g
288 g
Up to 2 L
(Stacking Gel)
Final
Concentration
0.25 M
1.92 M
-
Chemical/Substance
Amount
Final
Concentration
MQ H2O
8.64 mL
-
4 mL
0.38 M
40% Acryl / 0.8% Bis
3.2 mL
-
20% SDS
80 μL
0.05 mM
10% Aps
80 μL
2.19 μM
12 μL
-
MQ H2O
3.84 mL
-
0.5M Tris-HCl pH 6.8
1.5 mL
0.13 M
0.6 mL
-
20% SDS
30 μL
1.87 μM
10% Aps
30 μL
0.82 μM
TEMED
10 μL
-
1.5M Tris-HCl pH 8.8
TEMED
gels
Final
Concentration
0.25 M
1.92 M
0.04 M
-
40% Acryl / 0.8% Bis
157
Reagent
LB Agar (pH 7.5)
Reagent
LB Broth (pH 7.5)
Reagent
SeaHorse Running
Media (pH 7.4)
Chemical/Substance
Amount
Final
Concentration
Bacto-Trytone
10 g
1%
Yeast extract
5g
0.5%
Sodium Chloride
(NaCl)
10 g
0.17 M
Agar
15 g
1.5%
Sodium Hydroxide
(NaOH)
pH
adjustment
(7.5)
-
Milli-Q H2O
Up to 1 L
-
Chemical/Substance
Amount
Final
Concentration
Bacto-Trytone
10 g
1%
Yeast extract
5g
0.5%
Sodium Chloride
(NaCl)
10 g
0.17 M
Sodium Hydroxide
(NaOH)
pH
adjustment
(7.5)
-
Milli-Q H2O
Up to 1 L
-
Chemical/Substance
Amount
Final
Concentration
Unbuffered DMEM
(SeaHorse BioScience,
# 100965-000)
50 μL
-
100 X Glutamax
500 μL
2 mM
Sodium Pyruvate
500 μL
1 mM
5 M Sodium
Hydrocide (NaOH)
pH
adjustment
(7.4)
-
158
Chemical/Substance
Amount
Final
Concentration
Tris pH 7.4
200 mL
-
Sodium Chloride
(NaCl)
58.44 g
0.05 M
Tween-20
10 mL
-
Milli-Q H2O
Up to 20 L
-
Chemical/Substance
Amount
Final
Concentration
1 M Tris-HCl pH 6.8
2 mL
50 mM
SDS
0.8 g
0.28 M
100% glycerol
4 mL
0.4 M
400 μL
0.59 M
1 mL
12.5 mM
Bromophenol Blue
8 mg
1.19 mM
Reagent
Chemical/Substance
Amount
1% BSA in TBST
1 X TBST
BSA
500 mL
5g
Reagent
Chemical/Substance
Amount
Final
Concentration
Tris-base
108 g
0.9 M
Boric acid
55 g
0.45 M
EDTA
9.3 g
15.9 mM
Milli-Q H2O
Up to 2 L
-
Reagent
1 X TBST
Reagent
4 X Laemmli Buffer (10
mL)
5 X TBE (2 L)
14.7 M
β-mercaptoethanol
0.5 M EDTA
159
Final
Concentration
1%
Reagent
Lysis Buffer
Chemical/Substance
Amount
Final
Concentration
1 M Tris pH 7.5
50 μL
50 mM
0.05 M EDTA
20 μL
1 mM
0.1 M EGTA
10 μL
50% Glycerol
200 μL
1 mM
10% Triton X-100
100 μL
15 μM
0.5 M NaF
100 μL
50 mM
0.05 M Sodium
Pyrophosphate
(Na4P2O7)
100 μL
5 mM
0.1 M sodium
orthovanadate
(Na3VO4)
10 μL
1 mM
0.5 M DTT
2 μL
1 mM
100 X PIC
10 μL
Milli-Q H2O
408 μL
-
Chemical/Substance
Amount
Final
Concentration
Tris-base
605.5 g
5M
Milli-Q H2O
Up to 1 L
HCl
pH
adjustment
(Variable)
1 mM
-
NAM
TSA
Reagent
Tris solutions (5 M)
160
-
Reagent
Amount
Final
Concentration
500 mL
-
FBS
50 mL
10%
Penicillin/streptamycin
5 mL
1%
Sodium Pyruvate
5 mL
1 mM
Glutamax
5 mL
2 mM
Chemical/Substance
Amount
Final
Concentration
#11875 glucose free
RPMI 1640
Up to 100
mL
-
2 mL
2 mM
Na-L-Lactate
0.224 g
20 mM
BSA
0.1 g
0.1%
Chemical/Substance
Amount
Final
Concentration
NaH2PO4.H2O
1.44 g
1.44%
Phenol (detached
crystals)
0.1 g
0.1%
Chemical/Substance
#11995 high-glucose
1 X DMEM
HepG2 DMEM Media
(500 mL)
Reagent
Sodium Pyruvate
GP media (100mL)
Reagent
0.12 M Phosphate
buffer
(100 mL)
100 mM
5 M NaOH
Milli-Q H2O
161
pH
adjustment
(variable)
Up to 100
mL
-
Reagent
DYE
Reagent
POD
Reagent
GOD
Reagent
Glucose production
assay buffer (25 mL)
Reagent
Oil Red-O stock
Chemical/Substance
Amount
Final
Concentration
4-aminoantipyrine
0.625 g
0.025%
0.12 M Phosphate
buffer
2.5 mL
-
Chemical/Substance
Amount
Final
Concentration
Peroxidase
35.43 mg
0.0015%
0.12 M Phosphate
buffer
2.5 mL
-
Chemical/Substance
Amount
Final
Concentration
Glucose Oxidase
0.472 g
0.0095%
0.12 M Phosphate
buffer
5 mL
-
Chemical/Substance
Amount
Final
Concentration
0.12 M Phosphate
buffer
25 mL
-
DYE
50 μL
-
POD
25 μL
-
GOD
125 μL
-
Chemical/Substance
Amount
Final
Concentration
Oil Red-O
0.7 g
-
Isopropanol
200 ml
-
162
Reagent
Chemical/Substance
NaCl
Amount
Final
Concentration
32.5 g
555 mM
KHB buffer 5X
KCl
1.75 g
23.5 mM
(1 litre)
MgSO4
1.2 g
10 mM
Na2HPO4
0.85 g
6 mM
Reagent
Chemical/Substance
Amount
Final
Concentration
IP buffer pH 7.2
NaCl
4.383 g
150 mM
(500 ml)
Tris
1.514 g
25 mM
163