Vol. 137, No. 8
Printed
IR U.S.A.
Copyright
O-i996
by The Endocrine
Society
Glucagon
Response
Insulin-Independent
Diabetic
Rats*
Z. Q. SHIT,
M. VRANIC
K. S. RASTOGI,
to Hypoglycemia
Is Improved
Restoration
of Normoglycemia
M. LEKAS,
S. EFENDIC,
D. J. DRUCKER$,
by
in
AND
Departments
of Physiology (Z.Q.S., K.S.R., M.L., M.V.), Medicine
(D.J.D., M.V.), and Surgery (Z.Q.S.),
Faculty of Medicine,
lJniversit.y of Toronto, Toronto, Ontario, Canada; and the Department
of
Endocrinology,
Karolinska
In&i&e
(S.E.), Stockholm,
Sweden
ABSTRACT
The aim of this study was to determine
whether
the impaired
glucagon response
to insulin-induced
hypoglycemia
in the diabetic
rat
can be improved
by correction
of hyperglycemia
independent
of insulin. Four groups of age-matched
male Sprague-Dawley
rats (246 2
13 g BW) were studied:
1) normal
controls
(NC; n = 7); 2) diabetic,
untreated
(DU; n = 6); 3) diabetic,
treated
for 5-7 days using sustained release (2-3 U/day)
insulin
implants
(DI; n = 6); and 4) diabetic, treated
for 3-4 days with phlorizin
(0.4 g/kg), given SC twice
daily (DP; n = 7). Diabetes
was induced
by a single
injection
of
streptozotocin
(65 mg/kg).
Basal plasma glucose was 7.4 2 0.3 mM in
NC, but rose to 14.5 5 2.2 mM in DU. Basal hyperglycemia
was
corrected
with phlorizin
and insulin
treatments
(5.5 + 0.5 and 6.7 +
0.8 mM, respectively).
NC rats responded
to insulin-induced
hypoglycemia
with a rapid and marked
increase
in glucagon
(peak, 2059
2 311 pg/ml). The glucagon
response
was blunted
in DU (635 2 180
pg/ml)
and was partially
improved
by prolonged
normalization
of
H
YPOGLYCEMIA is a severe complication of insulindependent diabetes mellitus (IDDM), especially in
those treated with intensive insulin therapy (l-3). Hypoglycemia is associatedwith impaired glucagon release(2-6) and
hepatic glucose production (2, 7). The impairment in glucagon responsivenessmay involve at least a defect in the glucose-sensingmechanism of the pancreatic a-cell (6, 8). It has
been suggested that diabetic pancreatic c-u-cells,similarly to
the p-cells in noninsulin-dependent diabetes mellitus
(NIDDM) (9) become insensitive to changes in glucose concentration (4). Recent studies using the hypoglycemic insulin
clamp technique showed that hypoglycemia greatly amplifies the a-cell response to hyperaminoacidemia in healthy
subjects, but not in IDDM patients (10). In alloxan-induced
diabetic dogs with severe basal hyperglycemia and hyperglucagonemia, glucagon was not suppressed by infusion of
Received
December
19, 1995.
Address
all correspondence
and requests
for reprints
to: 2. Q. Shi,
M.D.,
Ph.D., Department
of Physiology,
Medical
Sciences Building,
Room 3358, University
of Toronto,
Toronto,
Ontario,
Canada M5S lA8.
*This work was supported
by grants from the Medical
Research
Council
of Canada (to Z.Q.S., M.V., and D.J.D.) and from the Juvenile
Diabetes
Foundation
International
and the Canadian
Diabetes
Association (to M.V.).
t Fellow of the Canadian
Diabetes
Association
and presently
supported
by a Scholarship
from the Banting
and Best Diabetes
Center,
University
of Toronto.
$ Scientist of the Medical
Research
Council.
glycemia
in DP (1335 t- 295 pg/ml; P < 0.05). Plasma
somatostatin
levels in all diabetic
groups were 2- to 3-fold higher in the basal state,
but were not different
during
hypoglycemia,
than those in NC rats.
Compared
to levels in NC rats, diabetes resulted
in decreased
insulin,
but elevated
glucagon
and somatostatin
concentrations
in the pancreatic tissue. Treatment
with both insulin
and phlorizin
reversed
the
changes in the pancreatic
content
of both glucagon
and somatostatin.
Pancreatic
proglucagon
messenger
RNA did not show significant
differences
among the four groups in either state. Insulin
treatment
in
the DI group resulted
in a delayed
and much smaller
increase
in the
glucagon
response
(740 + 138 pg/ml)
to hypoglycemia
despite normalization
ofglvcemia.
We. therefore.
conclude
that in strentozotocindiabetic
rats,tge
impaired’glucagon
responsiveness
to hypoglycemia
is significantly
improved
by insulin-independent
correction
of hyperglycemia,
suggesting
the importance
of normoglycemiaper
se in maintaining,
at least in part, the glucose sensitivity
of pancreatic
cY-cells.
Uh&minology
137: 3193-3199,
1996)
a large glucose dose (8). In IDDM, the a-cell defect appears
to be selective, as the glucagon response to arginine was
maintained when its response to hypoglycemia was impaired (4). Defects in both glucagon and catecholamine responseshave been reported in IDDM and NIDDM patients,
as well as in experimental diabetic animals (4, 7, 11, 12).
The mechanism by which the a-cell losesits sensitivity to
glucose has not yet been clarified. Chronic hyperglycemia,
among other factors, may play an important role in desensitizing the a-cells. Correction of hyperglycemia per sewithout insulin normalized the glucagon response to the acute
hyperglycemic challenge with glucose infusion (B), improved hepatic glucose production during moderate hypoglycemia in the alloxan-induced diabetic dogs (7), and normalized tissue insulin sensitivity in diabetic rats (13). It has
also been shown that prolonged hyperglycemia can induce
altered sensitivity of pancreatic CPand p-cells to glucose in
healthy humans (14). However, in IDDM patients, optimal
glycemic control with various insulin therapies could not
restore the impaired glucagon response to hypoglycemia
(15-17). The underlying mechanisms are unclear, although
exogenous insulin therapy may have an inhibitory effect on
glucagon secretion (18-20). Therefore, it needs to be established whether the impaired glucagon response to hypoglycemia can be normalized or improved by prolonged correction of hyperglycemia independent of insulin.
3193
3194
IMPAIRED
GLUCAGON
RESPONSE
An intriguing
aspect of the pancreatic hormonal
response
to changes in circulating
glucose is the somatostatin/
glucagon relationship
in the pancreas and the circulation.
We have
previously
reported increased total pancreatic
somatostatin
content with no change in glucagon in alloxan-diabetic
dogs
(21, 22). However,
acute normalization
of glycemia
using
insulin in these dogs resulted in a marked decrease (5-fold)
in the total pancreatic
glucagon
content, with a relatively
smaller decrease in the somatostatin
content. We hypothesized that the relative abundance
of somatostatin
over glucagon, combined
with the decreased total glucagon
content
in normoglycemia
could explain at least in part the impaired
glucagon
response to hypoglycemia
in diabetes
(21, 22).
These data were obtained
under the combined
effects of
correction
of hyperglycemia
and acute insulin administration. Insulin is known to exert paracrine
and endocrine
effects on islet hormonal
behavior.
Therefore,
the impact of
normalization
of plasma glucose on the somatostatin
/ glucagon relationship
needs to be distinguished
from that of
acute insulin administration.
In the present study, we investigated whether
the impaired
glucagon
response to hypoglycemia can be corrected by normalization
of blood glucose
in the streptozotocin
(STZ)-diabetic
rat with or without
insulin for a prolonged
period.
Materials
Animal
and
Methods
preparation
Male Sprague-Dawley
rats (Charles River, Quebec, Canada),
initially
weighing
between
250-300
g, were individually
housed in metal cages
in a temperatureand humidity-controlled
room for 7 days before use.
All rats were fed rat chow (Ralston Purina Co., St. Louis, MO) and water
ad libitutn. Four groups of age-matched
male Sprague-Dawley
rats (246
+ 13 g BW) were used: 1) normal
controls
(NC; n = 7); 2) diabetic,
untreated
for 2 weeks (DU; n = 6); 3) diabetic,
treated for 5-7 days using
sustained
release (2-3 U/day)
insulin
implants
(DI; n = 6); and 4)
diabetic,
treated for 3-4 days with phlorizin
(0.4 g/kg),
given SC twice
daily (DP; n = 7). For the DU, DP, and DI rats, the total duration
of
diabetes was 2 weeks, including
the last 5-7 days required
for insulin
treatment
and the last 3-4 days for phlorizin
treatment,
before the rats
were subjected
to acute hypoglycemia
experiments.
Diabetes
was induced by a single injection
of STZ (65 mg/kg;
Sigma Chemical
Co., St.
Louis, MO) in sterile saline into the penile vein under pentobarbital
anesthesia
(50 mg/ kg body wt, ip). The rats were given 5% glucose in
their drinking
water for the first 24 h after ST2 injection.
Glycosuria
and
urinary
ketone bodies were determined
once every 2 days. This model
features
moderate
diabetes
with fasting hyperglycemia
and 70% reduced residual plasma insulin
(23). DI rats were treated with Linplant,
a sustained
release bovine
insulin
preparation
(Linshin
Canada,
Inc.,
Ontario,
Canada),
implanted
SC under
pentobarbital
anesthesia
(50
mg/kg
BW, ip). One full-size
implant
releases 2.0 U regular
porcine
insulin/day.
The rationale
and technical
aspects of the insulin
implant
have been detailed previously
(24). DP rats were treated with phlorizin
for 3-4 days using a 40% solution
in propylene
glycol, administered
as
a SC injection
twice daily (0.4 g/kg)
to ensure
continuous
day-long
inhibition
of renal tubular
glucose reabsorption.
The last dose of phlorizin was given 4 h before the experiment.
Any rats with positive urinary
ketone bodies and detected
hypoglycemia
(fasting blood glucose,
c4.0
mM) were precluded
from further
studies.
Three or 4 days before the acute hypoglycemia
experiment,
rats were
anesthetized
with pentobarbital
(50 mg/kg
BW, ip), and indwelling
SILASTIC
brand tubing (length,
3 cm; id, 0.020 in.; od, 0.037 in.; Dow
Corning,
Midland,
MI) connected
to a polyethylene
catheter
(length, 10
cm; PE-50, Clay Adams, Boston, MA) was inserted
into a carotid artery.
A heparin
(1000 U/ml)
solution
in 60% polyvinyl
pyrrolidone
was used
to prime the catheters
to prevent
blood clot formation.
The catheters
Endo.
1996
Vol 137 . No 8
TO HYPOGLYCEMIA
were then tunnelled
SC, exteriorized
chored at the back of the neck.
Experimental
through
a skin
incision,
and an-
protocol
Experiments
were carried out in conscious overnight-fasted
rats. Carotid cannulas
were flushed
with heparinized
saline. At time zero,
regular
insulin was injected
in varying
doses to achieve the same level
of hypoglycemia
(1.60-2.75
mM) within
30-60
min, which was then
maintained
for 45-60 min in all groups.
Blood samples were taken via
carotid cannula before insulin injection
(0 min) for all basal measurments
and at 15- to 30-min intervals
during insulin-induced
hypoglycemia
for
glucose and glucagon
measurements.
Somatostatin
and catecholamine
samples were taken at time zero and at the end of the experiments.
The
concentration
of glucose
was determined
from a sample of 0.01 ml
plasma using a Beckman
glucose analyzer
II (Palo Alto, CA). For measurements
of glucagon
and somatostatin,
blood was collected
in chilled
tubes containing
EDTA and Trasylol
(FBA Pharmaceuticals,
New York,
NY). The packed blood cells, after removal
of plasma, were resuspended
in heparinized
saline (10 U/ml)
and reinfused
after each blood sampling
to prevent
volume
depletion
and anemia. Hematocrit,
determined
every
15 min, was maintained
above 35%. At the end of the experiments,
rats
were killed by decapitation.
All blood samples
were immediately
centrifuged
at 4 C and plasma was stored at -20 C before analysis.
A
separate subgroup
of rats (n = 5-7) from each treatment
protocol
(NC,
DU, DP, and DI) was killed at a time corresponding
to those undergoing
hypoglycemia
experiments
to obtain baseline pancreatic
tissue samples.
Hormone
analysis
assays and proglucagon
messenger
RNA CmRNA)
Immediately
after death, the pancreas was removed,
and part of the
tail portion
was excised for Northern
blot analysis.
The rest of the
pancreas was frozen in liquid nitrogen
and kept at -70 C for subsequent
extraction
and immunoassay
of pancreatic
hormones.
The frozen tissue
was homogenized
with (10 ml/g)
cold acidified
75% ethanol
(7.5 ml
ethanol, 0.15 ml HCI, and 2.5 ml H,O) and centrifuged.
Trasylol
(10,000
kallikrein
inhibitor
units/ml)
was added (20% of the pancreatic
extract
volume)
for hormonal
stability.
Insulin,
glucagon,
and somatostatin
RIAs (22) were carried
out on both plasma samples and on an aliquot
of neutralized
tissue extracts.
Pancreatic
proglucagon
mRNA was determined
for all rats (n = 6-7)
from each group in hypoglycemic
experiments
and for all rats in baseline
control subgroups
(n = 5-7). Total RNA was isolated by homogenization
of pancreatic
tissue in guanidine
isothiocyanate,
centrifugation,
and
acid-ethanol
precipitation
(25). RNA was size-fractionated
on a 1.3%
(wt/vol)
agarose-formaldehyde
gel. The gel was ethidium
bromide
stained to assess the integrity
of the RNA, which was then transferred
to a nylon membrane
and fixed by UV irradiation.
Blots were prehybridized
and hybridized
in 1 X Denhardt’s
solution
(50 X Denhardt’s
solution
= 5 g Ficoll, 5 g polyvinylpyrrolidone,
5 g BSA fraction
V, and
distilled
H,O to 500 ml), 4 X SSC (1 X SSC is 0.15 M NaCl and 0.015 M
Na citrate),
200 kg/ml
salmon sperm DNA, and 40% deionized
formamide in 0.014 M Tris, pH 7.4. Complementary
DNAs for glucagon
and
tubulin
were labeled to a specific activity
of 5 X 10’ cpm/pg
with
[“Pldeoxy-ATP
by a random
priming
method,
and hybridization
was
performed
with 1 X 10h cpm/ml
probe for 24 h at 42 C. Final washing
was carried out in 0.1 X SSC-0.1% (wt/vol)
SDS at 65 C. Kodak X-Omat
film (Eastman
Kodak,
Rochester,
NY) was used for autoradiography
at
-70 C. Densitometry
was employed
for quantitative
assessment
and
comparison
of glucagon
mRNA
transcripts
in all groups.
All results are expressed
as the mean t SEM. Statistical
analysis was
performed
by ANOVA
using the Statistical
Analysis
System package for
personal
computers
(SAS Institute,
Cary, NC). For multiple
comparisons, Tukey’s
significance
test was used.
Results
Table 1 shows mean body weights and basal plasma levels
of glucose, insulin, glucagon,
and somatostatin
in all four
groups. The mean body weights of the untreated
diabetic rats
(272 2 20 g) were significantly
decreased compared
to those
IMPAIRED
TABLE
1. Basal
values
in normal,
GLUCAGON
diabetic,
and phlorizin
Normal
Control
(n = 7)
Body Weight
(gm)
Glucose (mM)
Insulin
(pmol/l)
Glucagon
(pg/ml)
Somatostatin
(pg/ml)
Statistical
significance
333
7.4
and insulin
Diabetic
k 42"
5 0.3**
46
by * (p < 0.05)
-o-
NORMAL
-*-
DIABETIC
-m-
LIIAB
PHLRZN
--A--
DIAB
INS
and
** (p < 0.01)
\
‘\
ii
*
*
3195
rats
Untreated
(n = 6)
Diabetic
2 20
k 2.2
?I 8
Phlorizin
(n = 7)
300
5.5
38
Diabetic
Insulin
(n = 6)
k 17
k 0.5""
I? 8
309
6.7
500
+
k
%
189 i29 +
451 +- 108
k 12
37
in each group
-t 5
as compared
with
**
i
i
diabetic
301 2 61
14 -f 2**
is indicated
TO HYPOGLYCEMIA
treated
272
14.5
62
184 5 32””
141 2 20**
T
‘\
RESPONSE
-”
2500
E
is
s
2000
5
:
the diabetic
untreated
**
T
T
I - - - 6,
00
I
*“,
15
30
I
*
22
0.8**
145""
57
4
rats.
-o-
NORMAL
-*-
DIABETIC
-m-
DIAB
PHLRZN
--.--
DIAB
INS
*
1500
0
3
1000
22
500
e
/0
15
30
45
60
75
i
120
,-
0
0
TIME
FIG. 1. Plasma
45
60
75
120
(min)
glucose levels during
acute hypoglycemia
Insulin
was injected
immediately
after 0 min blood
glycemia
was attained
between
75-120
min, when
killed.
Plasma
glucose
levels in the NC group are
ferent
from those in the DU group
(at 0, 15, 30,
indicated
(**, P < 0.01; *, P < 0.05).
experiments.
sampling.
Hypothe animals
were
significantly
difand 45 min), as
in age-matched
normal control rats (333 2 42 g; P < 0.05). The
body weights of DP (300 t 7 g) and DI (309 5 20 g) rats were
slightly, but not significantly,
lower than those of NC and
higher than those of DU rats. As expected, the mean plasma
glucose values in the overnight-fasted
DU rats were significantly higher than those in NC (14.5 2 2.2 us. 7.4 + 0.3 mM;
P < 0.05). The baseline plasma insulin concentration
was 184
2 32 pmol/liter
in the normal rats and was significantly
lower in the untreated
rats (62 ? 8 pmol/liter;
P < 0.01).
Phlorizin
treatment in the diabetic rats did not significantly
affect insulin levels (38 2 8 pmol/liter).
The SC insulin implants resulted in elevated plasma insulin levels (500 + 145
pmol/liter;
P < 0.01 VS. all other groups).
Phlorizin
and
insulin treatments
normalized
basal plasma glucose levels
(DP, 5.5 2 0.5; DI, 6.7 2 0.8 mM), both of which were significantly
lower than those in DU rats (both P < 0.05).
The changes in plasma glucose during
acute insulin-induced hypoglycemia
are shown in Fig. 1. Before insulin
injection,
the basal plasma glucose concentration
was significantly
higher in DU rats, whereas
in the DI and DP
groups, glucose values were in the normal range. Although
starting from different
basal values, similar hypoglycemic
levels were reached within 75-120 min in all groups (NC, 1.7
+ 0.3; DU, 2.8 ? 0.4; DP, 2.2 & 0.1; DI, 1.6 -t 0.1 mM). The
total
amount
of insulin
required
was higher in all three
diabetic groups (DU, 1.6 ? 0.3; DP, 2.0 + 0.3; DI, 1.0 ? 0.1
U/100 g) than that in the NC group (0.3 U/100 g). Judging
from the different
doses of insulin required
to induce hypoglycemia,
phlorizin
treatment
for 3-4 days had no effect,
whereas insulin treatment moderately
improved
insulin re-
TIME
(min)
FIG. 2. Plasma
glucagon
levels during
acute hypoglycemia
experiments.
Experimental
procedures
are the same as those in Fig. 1.
Plasma
glucagon
levels in the NC and DP groups
are significantly
higher than those in the DU group, as indicated
(**, P < 0.01; *, P <
0.05).
sistance. The difference
may be due to the relatively
short
duration
of the phlorizin
treatment. It has been reported that
phlorizin
treatment
for 4-5 weeks significantly
improved
insulin sensitivity
in diabetic rats (13).
Figure 2 depicts glucagon
values during insulin-induced
hypoglycemia.
In normal rats, with a decrease in the plasma
glucose level, there was a rapid and very marked increase
(14.6-fold) from basal in the peak plasma glucagon response
(2059 2 311 pg/ml,
at 30 min). In DU rats, the glucagon
response was markedly
diminished
and delayed, with a peak
response of 635 2 180 pg/ml
at 45 min. Normalization
of
glycemia by insulin was also associated with a delayed and
attenuated
glucagon
response (740 ? 138 pg / ml, at 60 min).
In contrast, normalization
of glycemia
with phlorizin
partially, but significantly,
improved
the glucagon
response
(P < 0.05 us. DU and DI) during the initial 45 min of hypoglycemia, with a peak of 1335 + 295 pg/ml
at 30 min.
The mean plasma somatostatin
levels in the basal and
hypoglycemic
states are shown in Fig. 3. The initial basal
plasma somatostatin
level in normal control rats (14 + 2
pg/ml)
was lower than that in the DU (45 + 10 pg/ml; P <
O.Ol), DP (37 2 5; P < 0.05), and DI (29 ? 5; P < 0.05) rats.
Plasma somatostatin
decreased in all diabetic groups during
hypoglycemia,
but the decrease was only significant
in the
phlorizin-treated
group (P < 0.05).
There was no significant
difference in norepinephrine
and
epinephrine
levels in the initial basal state (time zero) in all
four groups (Fig. 4). In the hypoglycemic
state, both catecholamines
increased in all groups (P < 0.01 or P < 0.05).
3196
IMPAIRED
GLUCAGON
RESPONSE
TO HYPOGLYCEMIA
Endo.
1996
Vol 137 . No 3
NORMAL
E
E
3
E
70
0
DIABETIC
60
-
DIAB
PWLRZN
DIAB
INS
5oi- I
40
t IA
Z
!-...-
HYPOGLYCEMIA
60
i
30
20
10
0
600
HYPOGLYCEMIA
BASAL
FIG. 3. Plasma somatostatin levels during the basal period (left) and
acute insulin-induced
hypoglycemia (tight). Statistical significance,
obtained from comparison with the DU group in the basal or hypoglycemic state, is indicated (**, P < 0.01; *, P < 0.05).
25
rm
NORMAL
CT]]
DIABETIC
-
DIAB
PHL
DIAB
INS
T
F
T
0
Gi
E
%
E
4
z
F
2
30
20
**
**
0"
2
10
4
0
N
D
DP
DI
FIG. 5. Concentrations of insulin (upper panel), glucagon (middle
panel), and somatostatin (lowerpanel) in the pancreatic tissues ofthe
baseline control rats of each protocol and in the rats subjected to acute
insulin-induced
hypoglycemia. Statistical significance, obtained from
comparison with the DU group in the basal or hypoglycemic state, is
indicated (**, P < 0.01; *, P < 0.05).
BASAL
HYPOGLYCEMIA
FIG. 4. Plasma epinephrine and norepinephrine
levels during the
basal period (left) and acute insulin-induced
hypoglycemia experiments. Statistical significance, obtained from comparison with the
DU group in the basal or hypoglycemic state, is indicated (**, P < 0.01;
*, P < 0.05).
However, the increase in epinephrine in the phlorizintreated group was significantly less(P < 0.05) than those in
the NC and DU groups. This tendency was also observed
with norepinephrine, although statistical significance was
not reached.
Pancreatic tissue concentrations of insulin, glucagon, and
somatostatin, with or without hypoglycemia, are shown in
Fig. 5. The insulin concentration was markedly decreased in
all diabetic groups compared to that in NC rats, and there
was no significant difference in the insulin concentration
before and after insulin-induced hypoglycemia in any group.
The basal pancreatic glucagon concentration was 57% higher
in the DU group than in the NC group (480 2 128 VS.306 +
52 ng/mg protein), although the difference was not statistically significant. Normalization of glycemia resulted in decreases in tissue glucagon concentrations with prolonged
treatment of phlorizin (48%; P < 0.05) and insulin (36%; P =
NS), respectively. Acute hypoglycemia did not change the
profile of pancreatic glucagon concentrations from the basal
values in all groups. The pancreatic tissue concentration of
somatostatin was significantly higher in the untreated diabetic group than in all other groups in both basal and hypoglycemic states.
Figure 6 shows part of a representative Northern blot
obtained from rat pancreata from each subgroup in both
baseline and hypoglycemic states. The pancreatic proglucagon mRNA levels (expressed as densitometric units) were:
IMPAIRED
0
N
DP D DI N
GLUCAGON
RESPONSE
H
DP D DI
T
FIG. 6. Proglucagon
(G) and tubulin (T) mRNA transcripts in the
pancreatic tissues from the baseline control animals (B) and from the
rats subjected to acute insulin-induced
hypoglycemia (H; n = 4 for
each group in either state). No statistical difference was found among
all groups in either state.
NC, 1.79 ? 1.62;DP, 3.37 t 1.10;DU, 2.86 2 0.62; and DI, 2.94
t 1.29 for the baseline samples (n = 4 for each group), and
NC,2.93 + 0.72;DP,4.18 t 1.16;DU,2.94 2 0.74;andDI,2.59
+ 0.55 for the samples taken after acute hypoglycemia experiments (n = 4 for each group). Although the values appeared to be relatively higher with DP (-20-40%) than with
any other group, no statistical difference was found among
all groups in either state.
Discussion
Glucagon plays a primary role in glucose counterregulation during hypoglycemia (1, 26, 27). It has been suggested
that the o-cell unresponsiveness in diabetic humans and
animals may be related to desensitization of the diabetic
pancreatic a-cells (4), and that hyperglycemia plays an important role in the desensitization process (8).
The present study has produced clear evidence that the
glucagon response to insulin-induced hypoglycemia can be
substantially improved by the correction of hyperglycemia,
despite continuing insulin deficiency. Phlorizin normalizes
glycemia by binding to the sodium-dependent active glucose
transporter in the brush border of renal tubular cells, thus
blocking renal glucose reabsorption (28). Our results suggest
that the desensitizing effect of hyperglycemia on the a-cell
glucose-sensingmechanism can be partially circumvented
by an insulin-independent restoration of euglycemia for 3-4
days. Our finding is mirrored by an earlier study by Stark et
al. (8) in hyperglycemic alloxan-diabetic dogs, in which elevated basal glucagon secretion was not inhibited by a progressively increased iv load of glucose at up to 24 mg/
kgmin. Nevertheless, correction of hyperglycemia with
phlorizin for 48 h in these dogs restored the decremental
glucagon response to hyperglycemic challenge (8). Therefore, the current study and that of Stark et al. demonstrate
that prolonged hyperglycemia impairs glucagon responses
to acute hypoglycemia and hyperglycemia, respectively.
Such impairment can be circumvented in part by correction
of hyperglycemia, independent of insulin. Direct perfusion
of phlorizin into the isolated pancreas of nondiabetic dogs
TO HYPOGLYCEMIA
3197
had no effect on the in vitro glucagon responseto glucose (8).
Taken together, these findings point to the relative importance of normoglycemia per sein maintaining the o-cell glucose sensitivity. One can speculate that the improvement of
glucagon responsivenessdue to correction of hyperglycemia
may reflect a reversal of down-regulated glucose transporters at the cellular membrane of the a-cell, a mechanism
comparable to that observed in the skeletal muscle of diabetic
rats (23, 29). However, the glucose transport mechanism of
the diabetic islet a-cell has not yet been established, and
posttransporter mechanisms could be involved.
It is interesting to note that the insulin treatment in STZdiabetic rats resulted in a smaller and delayed increment in
glucagon responseto hypoglycemia than the phlorizin treatment despite a similar degree of normalization of euglycemia. In patients with IDDM, prolonged normoglycemia of
14-20 days with multiple insulin injections could not improve the glucagon response to insulin-induced hypoglycemia (15). A number of other studies have reported similar
results (16,17). In diabetic rats treated with ip insulin osmotic
pumps for more than 2 months, the abnormal glucagon responseto hypoglycemia was only marginally improved (12).
Our findings suggest that the defective glucagon responseto
acute hypoglycemia in diabetes is not a mere consequenceof
insulin deficiency. Furthermore, exogenous insulin is known
to inhibit the islet a-cell (B-20). In the absenceof endogenous insulin, hyperglycemia induced by an oral glucose load
fails to lower plasma glucagon levels, and insulin-induced
hypoglycemia cannot stimulate glucagon secretion. Normalization of glycemia with acute insulin in alloxan-diabetic
dogs resulted in a marked inhibition of the total pancreatic
glucagon content (22). Fanelli ef al. (30, 31) reported that in
IDDM patients, institution of rational intensive insulin therapy with meticulous prevention of hypoglycemia reverses
hypoglycemic unawareness, normalizes glycemic thresholds, and improves neuroendocrine responses during
stepped hypoglycemic clamps. Glucagon secretion was improved, although not normalized (30, 31). Others reported
that maintenance of 24-h normal plasma glucose with an
open loop insulin infusion pump in IDDM corrected someof
the metabolic and endocrine abnormalities (32, 33), but not
the glucagon response to hypoglycemia (15-17,34). It, thus,
appears that although sensitivity of o-cells to glucose is lost,
sensitivity to insulin may still be retained to some extent in
diabetic animals (7,35) and presumably also in patients with
IDDM (36). The slow release insulin implants used in this
study, although capable of restoring basal euglycemia, are
associated with a relative hyperinsulinemia in the fasting
state. Even though the DI rats were selected after discarding
those with detected hypoglycemia, it cannot be excluded that
some rats may develop mild hypoglycemia without being
detected at the time of our blood glucose monitoring. Therefore, in addition to chronic hyperglycemia, a number of other
factors can potentially contribute to the depressed a-cell responsiveness in insulin-treated diabetic animals and
humans.
A deficient epinephrine response to hypoglycemia has
been observed in IDDM patients, particularly in those with
a prolonged course of disease(37,38). In the present study,
there was no difference in epinephrine and norepinephrine
3198
IMPAIRED
GLUCAGON
RESPONSE
levels between
normal and untreated
diabetic rats in both
basal and hypoglycemic
conditions.
This may relate to the
relatively short duration
of diabetes in the experimental
animals. An attenuated
epinephrine
response has been demonstrated
in diabetic rats with prolonged
(>2 months) insulin treatment
(12). In the present
study, there was a
significant
attenuation
in the epinephrine
response and a
somewhat
lower, but insignificant,
decrease in the norepinephrine response to hypoglycemia
in phlorizin-treated
rats.
Similar
observations
of an attenuated
catecholamine
response were made in diabetic dogs receiving phlorizin
treatment (7). Although
the mechanism
of these phlorizin-related
changes in catecholamine
levels is unknown,
it is possible
that a-cell responsiveness
would be improved
to a greater
extent if the attenuation
of the catecholamine
response could
be prevented.
This is because a-cell responsiveness
can be
substantially
suppressed
by functional
adrenalectomy
(by
diversion
of adrenal venous blood from the systemic circulation) (39). In normal humans, catecholamines
play an important counterregulatory
role even in the presence of appropriate
responses
of glucagon
(40). The absence
of
contribution
of adrenergic
mechanisms
to counterregulation
cannot be compensated
by greater increases in glucagon (40).
In IDDM patients, the impairment
in glucagon response renders a greater reliance on catecholamine
secretions (1). The
insulin-treated
diabetic rats in this study had a moderately,
although
not significantly,
attenuated
epinephrine
response
to hypoglycemia.
It has been reported
that an antecedent
hypoglycemic
episode resulted in reduced neuroendocrine
responses, including
that of epinephrine,
to subsequent
hypoglycemia
(41). We cannot rule out the possibility
that due
to the presence of hyperinsulinemia,
some of the DI rats may
have developed
mild hypoglycemia
undetected
by our periodic monitoring.
In the present study, we observed that pancreatic
proglucagon mRNA levels were moderately,
although
not significantly, increased in diabetic rats after treatment
with phlorizin for 3-4 days. It cannot be concluded
from this study
whether
an insulin-independent
restoration
of normoglycemia serves to stimulate
glucagon
synthesis.
However,
in
addition
to a moderate
enhancement
of glucagon
mRNA
transcription,
the moderately
elevated basal glucagon,
the
significantly
improved
glucagon response to hypoglycemia,
and the decreased pancreatic
glucagon
content in the phlorizin-treated
diabetic rats appeared
to suggest an upgrading
of glucagon turnover,
i.e. synthesis and secretion, under the
influence of correction
of hyperglycemia.
Plasma and pancreatic
somatostatin
concentrations
were
higher in the untreated
diabetic rats, probably
due to the
stimulative
effect of hyperglycemia
(42). It has been suggested that an elevated somatostatin
concentration
contributes to the depressed glucagon secretion (21,43). Correction
of hyperglycemia
with phlorizin
treatment resulted in a significant
reduction
of pancreatic
somatostatin
in the basal
state and of plasma somatostatin
during the hypoglycemic
state. Both changes may be favorable for unleashing
a glucagon secretory response from the inhibitory
effect of somatostatin. Even though insulin treatment
was also associated
with decreased basal plasma and pancreatic
somatostatin
concentrations,
its beneficial effect on a-cell responsiveness
Endo.
1996
Vol 137 . No 8
TO HYPOGLYCEMIA
could have been masked by the inhibitory
effect of exogenous insulin. It is yet to be determined
whether such changes
make a significant
contribution
to the improved
glucagon
response to hypoglycemia.
In conclusion,
in STZ-diabetic
rats, impaired
glucagon
responsiveness
to hypoglycemia
is significantly
improved
by
the correction of hyperglycemia
independent
of insulin, suggesting that chronic hyperglycemia
may play a role in the
pathological
process of impaired
pancreatic
a-cell response.
In the present study, normalization
of glycemia by prolonged
insulin treatment
was associated with a delayed and attenuated increase in glucagon levels, which did not significantly
differ from the response in the untreated
diabetic rats. This
is probably
due to the inhibitory
effect of exogenous
insulin
on the islet a-cells in the presence of basal hyperinsulinemia.
The relevance of these findings
in diabetic rats to the impaired glucagon
responsiveness
to hypoglycemia
in IDDM
patients needs to be established.
Acknowledgments
The authors are grateful
to D. Bilinski, M. VanDelangeryt,
and L. Lam
for excellent
technical assistance. We are very grateful
to Dr. Paul Wang,
University
of Toronto,
for the supply
of the Linplant
insulin
implants.
References
1. Cryer PE, Binder
C, Bolli G, Cherrington
AD, Gale EAM, Gerich JE, Sherwin
RS 1989 Hvpoglvcemia
in IDDM.
Diabetes
38:1193-1199
,. -_
2. Orskov
L, Alberti
KGMM,
Mengel
A, Moller
N, Pedersen
0, Rasmussen
0,
Seefeldt
T, Schmitz
0 1991 Decreased
hepatic
glucagon
responses
in type 1
(insulin-dependent)
diabetes
mellitus.
Diabetologia
34:521-526
3. Gardner
LB, Liu Z, Barrett
EJ 1993 The role of glucose-6-phosphatase
in the
action
of insulin
on hepatic
glucose
production
in the rat. Diabetes
42:1614-1620
4. Gerich
JE, Langlois
M, Noacco
C, Karam
JH, Forsham
PH 1973 Lack of
glucagon
response
to hypoglycemia
in diabetes:
ewdence
for an intrinsic
pancreatic
alpha cell defect. Science
182:171-173
5. Maher
TD, Tanenberg
RJ, Greenberg
BZ, Hoffman
JE, Doe RP, Goetz FC
1977 Lack of glucagon
response
to hypoglycemia
in diabetx
autonomic
neuropathy.
Diabetes
26:196-200
6. Ganda 0, Srikanta
S, Gleason
RE, Soeldner
JS, Eisenbarth
GS 1986 Diminished A-cell secretion
in the earlv chase of tvue I diabetes
mellitus.
Metabolism
35:1074-1077
7. Hetenyi
Jr G, Gauthier
C, Byers M, Vranic
M 1989 Phloririn
induced
normoglycemia
partially
restores
glucoregulation
in diabetic
dogs. Am J Physiol
256:E277-E283
8. Stark AS, Grundy
JD, McGarry
JD, Unger
RH 1985 Correction
of hyperglycemia with phlorizin
restores
the glucagon
response
to glucose
in insulin
deficient
dogs: implications
for human
diabetes.
I’roc Nat1 Acad Sci USA
82:1544-1546
9. Cerasi E, Luft R, Efendic
S 1972 Decreased
sensitivity
of the pancreatic
beta
cells to glucose
in prediabetic,
diabetic
subjects.
Diabetes
21:224-234
10. Caprio
S, Tamborlane
WV, Zych K, Gerow
K, Sherwin
RS 1993 Loss of
potentiating
effect of hypoglycemia
on the glucagon
response
to hyperaminoacidemia
in IDDM.
Diabetes
42:550-555
11. Lussier
B, Vranic
M, Kovacevic
N, Hetenyi
Jr G 1986 Glucoregulation
in
alloxan
diabetic
dogs. Metabolism
35:18-24
12. Pate1 DG 1983 Effect of prolonged
msulin
treatment
on blunted
plasma
catecholamine,
glucagon
increase
during
insulin
hypoglycemia
in streptozotocin
diabetic
rats. Metabolism
32:377-382
13. Rossetti
L, Smith D, Shuman
GI, Papachristou
D, DeFronzo
RA 1987 Correction
of hyperglycemia
with phlorizin
normalizes
tissue sensitivity
to insulin
in diabetic
rats. J Clin Invest 79:1510-1515
14. Dimitriadis
G, Cryer P, Gerich
J 1985 Prolonged
hyperglycemia
during
infusion of glucose,
somatostatin
impairs
pancreatic
A- and B-cell responses
to
decrements
in plasma glucose in normal
man: evidence
for induction
of altered
sensitivity
to glucose.
Diabetologia
28:63-69
15. Bolli G, Calabrese
G, De Foe P 1982 Lack of glucagon
response
in glucose
counterregulation
in type I (insulin-dependent)
diabetes:
absence of recovery
after oroloneed
ootimal
insulin
therauv.
Diabetoloria
22:100-105
16. Ens&k
JWY Kanier
RA 1980 Glucag’oh
response
tz hypoglycemia
in type I
diabetic
men after 24.hour
glucoregulation
by glucose-controlled
insulin
infusion.
Diabetes
Care 3:285-289
,
I
,
.
IMPAIRED
GLUCAGON
RESPONSE
17. Ohneda
A, Watanabek
K, Horigoone
K 1978 Abnormal
response
of pancreatic
glucagon
to glycemic
changes
in diabetes
mellitus.
J Clin Endocrinol
Metab
46:504-510
18. Unger
RH 1983 Insulin-glucagon
relationships
in the defence
against
hypoglycemia.
Diabetes
32:575-583
19. Muller
WA, Faloona
GR, Unger RH 1973 The effect of experimental
insulin
deficiency
in glucagon
secretion.
J Clin Invest 50:1991-1999
20. Phil ippe
J 1981 Glucagon
gene transcription
is negatively
regulated
by insulin
in a hamster
islet cell line. J Clin Invest 84:672-677
21. Rastogi
KS, Lickley
L, Jokay
M, Efendic
S, Vranic
M 1990 Paradoxical
reduction
in pancreatic
glucagon
with normalization
of somatostatin,
decrease
in insulin
in normoglycemic
alloxan
diabetic
dogs: a putative
mechanism
of
glucagon
irresponsiveness
to hypoglycemia.
Endocrinology
126:1096-1104
22. Rastogi
KS, Brubaker
PL, Kawasaki
A, Efendic
S, Vranic
M 1993 Increase
in
somatostatin
to glucagon
ratio in islet of alloxan-diabetic
dogs. Can J Physiol
Pharmacol
71:512-517
23. Dimitrakoudis
D, Ramlal
T, Rastogi
S, Vranic
M, Klip
A 1992 Glycemia
regulates
the glucose
transporter
number
in the plasma
membrane
of rat
skeletal
muscle.
Biochem
J 284:341-348
24. Wang PY 1991 Sustained
release implants
for insulin
delivery.
In: Pick-up
JC
(ed) Biotechnology
of Insulin
Therapy.
Blackwell,
Oxford,
pp 42-74
25. Chomczynski
P, Sacchi N 1987 Single-step
method
of RNA isolation
by acid
guanidinium
thiocyanate-phenol-chloroform
extraction.
Anal
Biochem
162:156-159
26. De Feo P, Perriello
G, Torlone
E, Fanelli
C, Ventura
MM, Santeusanio
F,
Brunetti
P, Gerich
JE, Bolli G 1991 Evidence
against important
catecholamine
compensation
for
absent
glucagon
counterregulation.
Am
J Physiol
260:E203-E212
27. Cryer PE 1992 Iatrogenic
hypoglycemia
as a cause of hypoglycemia-associated
autonomic
failure
in IDDM.
Diabetes
41:255-260
28. Silverman
M 1981 Glucose
reabsorption
in the kidney.
Can J Physiol
Pharmacol 59:209-224
29. Klip A, Mar&e
A, Dimitrakoudis
D, Ramlal
T, Giacca A, Shi Z, Vranic
M
1992 Effect of diabetes
on glucoregulation.
From glucose
transporters
to glucose metabolism
irt viva. Diabetes
Care 15:1747-1766
30. Fanelli
C, Pampanelli
S, Epifano
L, Rambotti
AM, Di Vincenzo
A, Modarelli
F, Ciofetta
M, Lepore
M, Annibale
B, Torlone
E, Perriello
G, De Feo P,
Santeusanio
F, Brunetti
P, Bolli G 1994 Long-term
recovery
from unawareness deficient
counterregulation,
lack of cognitive
dysfunction
during
hypoglyceamia
following
institution
of rational
intensive
insulin
therapy
in IDDM.
Diabetologia
37:1265-1276
TO HYPOGLYCEMIA
3199
31. Fanelli
CG, Epifano
L, Rambotti
AM,
Pampanelli
S, Di Vincenzo
A,
Modarelli
F, Lepore
M, Annibale
B, Ciofetta
M, Bottini
P, Porcellati
F,
Scionti
L, Santeusanio
F, Brunetti
P, Bolli G 1993 Meticulous
prevention
of
hypoglycemia
normalizes
the glycemic
thresholds,
magnitude
of most of neuroendocrine
responses
to symptoms
of, cognitive
function
during
hypoglycemia
in intensively
treated
patients
with
short-term
IDDM.
Diabetes
42:1683-1689
32. Raskin
P, Pietri
A, Unger
RH 1979 Changes
in glucagon
levels after four to
five
weeks
of glucoregulation
by portable
infusion
pumps.
Diabetes
28:1033-1035
33. Tamborlane
WV, Sherwin
RS, Koivisto
V, Hendler
R, Gene1 M, Felig P 1979
Normalization
of the growth
hormone,
catecholamine
response
to exercise
in
juvenile-onset
diabetic
subjects treated
with a portable
insulin
infusion
pump.
Diabetes
28:785-788
34. Braaten
JT, Faloona
GR, Unger
RH 1974 The effect of insulin
on alpha-cell
response
to hypoglycemia
in long standing
alloxan
diabetes.
J Clin Invest
53:1017-1021
35. Ito K, Maruyama
H, Hirose
H, Kido K, Koyama
K, Kataoka
K, Saruta T 1995
Exogenous
insulin
dose-dependently
suppresses
glucopenia-induced
glucagon secretion
from perfused
rat pancreas.
Metabolism
44:358-362
36. Liu D, Adamson
UCK, Lins PES, Kollind
ME, Moberg
EAR, Andreasson
K
1992 Inhibitory
effect of circulating
insulin
on glucagon
secretion
during
hypoglycemia
in type I diabetic
patients.
Diabetes
Care 15:59-65
37. Hoeldtke
R, Boden
G, Shuman
C, Owen
C 1982 Reduced
epinephrine
secretion
in hypoglycemia
in insulin-dependent
diabetes
mellitus.
J Clin Invest
69:315-326
38. Hirsch
B, Shamoon
H 1987 Defective
epinephrine
growth hormone
responses
in type I diabetes
are stmulus
specific.
Diabetes
36:20-26
39. Yamaguchi
N, Briand
R, Gaspo R 1990 Effect of functional
adrenalectomy
on
glucagon
secretion,
circulating
catecholamines
during
insulin
hypoglycemia
in
the dog. Can J Physiol
Pharmacol68:1183-1188
40. De Feo P, Perriello
G, Torlone
E, Fanelli
C, Ventura
M, Santeusanio
F,
Brunetti
P, Gerich JE, Bolli G 1991 Contribution
of adrenergic
mechanisms
to
glucose
counterregulation
in humans.
Am J Physiol
261:E724-E736
41. Heller
SR, Cryer PE 1991 Reduced
neuroendocrine,
symptomatic
responses
to
subsequent
hypoglycemia
after 1 episode
of hypoglycemia
in nondiabetic
humans.
Diabetes
40:223-226
42. Klaff
LJ, Taborsky
Jr GJ 1987 Pancreatic
somatostatin
is a mediator
of glucagon inhibition
by hyperglycemia.
Diabetes
36:592596
43. Taborsky
Jr GJ 1983 Evidence
of a paracrine
role for pancreatic
somatostatin
in oiuo. Am J I’hysiol
245:E598-E603