| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
ARTICLES |
Institute of Medical Anatomy and Medical Physiology, University of Copenhagen, The Panum Institute, Copenhagen, Denmark
Address all correspondence and requests for reprints to: Jesper Thulesen, M.D., Institute of Medical Anatomy, Department B, University of Copenhagen, The Panum Institute, Blegdamsvej 3, 2200 Copenhagen N, Denmark. E-mail: J.Thulesen{at}mai.ku.dk
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Insulin therapy has been observed to render ß-cells less vulnerable to immune aggression in diabetes-prone animals (9, 10), and hyperinsulinemia in rats with insulinoma causing hypoglycemia results in reduced expression of GLUT2 on ß-cells (3, 11). Reduced GLUT2 expression might reduce the cellular uptake of streptozotocin into ß-cells. We propose that insulin therapy may affect experimental diabetes induced by streptozotocin, possibly through down-regulation of GLUT2 and a reduced cellular need for NAD+ due to decreased ß-cell activity. Thus, the aim of the present study was to investigate the effect of intensive insulin treatment on the diabetogenic action of streptozotocin.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
): 1)
streptozotocin-induced diabetic (STZ), 2) insulin-treated before and
24 h after streptozotocin injection (INS+STZ), 3) insulin-treated
control (INS), and 4) control (CTRL) rats. All animals were allowed
free access to water and laboratory food (Altromin no. 1314, Altromin
International, Lage, Germany).
|
Insulin treatment
The INS+STZ and INS rats were treated daily for 7 days with a
long acting heat-treated insulin preparation (Ultralente, pH 5.5,
Novo-Nordisk, Bagsvaerd, Denmark). Both groups were treated with 40
IU/rat (200 mU/g BW·day) from days -5 to 0 and then on day 1 with 20
IU/rat if the blood glucose concentration was below 2 mmol/liter, but
otherwise with 40 IU/rat. Insulin doses were 10 times higher than those
recommended for treatment of diabetic rats (12) to ensure diminished
activity of the ß-cells but also survival of the rats despite the
marked hypoglycemia.
Anesthesia
The rats were anesthetized with barbiturate (50 mg/kg, ip;
Brietal, Metohexital, Eli Lilly Co., Indianapolis, IN) before blood and
tissue samples were obtained.
Blood glucose and body weight
The blood glucose concentration was measured in blood from a
tail vein in the morning in STZ rats on day -5 (n = 12), daily
from days 0–3 (n = 12), and on days 5, 7, 14, and 21 (n =
6); in INS+STZ and INS rats daily from days -5 to 3 (n = 12; INS,
n = 8 on day 3) and then on days 5, 7, 10, 14, and 21 (n = 6;
INS, n = 4); and in CTRL rats on day -5 (n = 12) and then at
the same time points as INS+STZ and STZ rats from days 1–21 (n =
8 from day 7) by a One Touch II instrument (Lifescan, Milpitas, CA),
using the glucose oxidase method. The rats were weighed on the same
days.
Tissue and blood samples for RIA
Tissue samples of pancreas were obtained on day 2 (n = 4,
all groups) and 21 (n = 4, all groups) for extraction and
subsequent RIA to measure both the acute and chronic effects of the
treatment. The tissue samples were weighed and then immediately frozen
in dry ice. Blood samples (n = 4, all groups) were obtained on day
21. The blood samples were obtained by cannulation of the left
ventricle of the heart, drained into ice-cold tubes containing 6
mmol/liter EDTA and 500 IU/ml aprotinin, and kept on ice until
centrifugation within 30 min. Plasma was stored at -20 C until
assay.
Tissue extraction
Pancreatic tissue specimens were homogenized and extracted in 4
vol acid-ethanol, as described by Newgard et al. (13). After
2 h at 4 C, the extracts were centrifuged, and the clear
supernatants were neutralized with ammonium hydroxide and then
evaporated on a vacuum centrifuge (Speed-Vac, Heto, Hilleroed,
Denmark). The samples were reconstituted in phosphate buffer (0.04
M; pH 7.4) containing in addition 2.9 g/liter NaCl, 10
mM EDTA, and human serum albumin (Behringwercke, Marburg,
Germany) before analysis of insulin immunoreactivity. Insulin was
measured using antiserum 2004, raised against porcine insulin (but
having strong reactivity against rat insulin), rat insulins I and II
(mixed in the same proportion in which they occur in the pancreas) as
standard, and monoiodinated 125I-labeled human insulin
(generous gift from Novo Nordisk) according to the principles described
by Albano et al. (14).
Assays
Before all plasma measurements, 700 µl plasma/sample·assay
were extracted in 70% ethanol (vol/vol, final concentration). The
supernatant was dried in a vacuum centrifuge and redissolved in
phosphate buffer as described above. All plasma samples were assayed in
duplicate. Insulin was measured as described above. Pancreatic glucagon
was measured using antiserum 4305, monoiodinated
125I-labeled glucagon (generous gift from Novo Nordisk),
and highly purified porcine glucagon as standard (Novo Nordisk) (15, 16). Antiserum 4305 requires the free COOH-terminus of glucagon for
binding. Total glucagon (enteroglucagon and pancreatic glucagon) was
measured using antiserum 4304, which cross-reacts with equal strength
with all peptides that contain the glucagon sequence. The experimental
detection limit for both assays was below 2 pmol/liter. The intra- and
interassay coefficients of variation were below 10% and 20%,
respectively, for insulin at a level of 500 pmol/liter.
Immunohistochemistry
Tissue samples for immunohistochemistry were obtained from the
dorsal region of pancreas on day 0 (n = 4, only INS group), 2
(n = 4, all groups), 7 (n = 4, all groups), and 21 (n =
4, all groups). Tissue samples were immediately fixed by immersion into
ice-cold, freshly prepared, buffered 4% paraformaldehyde. The fixed
tissue samples of pancreas were embedded in paraffin and cut into
10-µm sections using a microtome. The unlabeled antibody
peroxidase-antiperoxidase technique was applied, as described by
Sternberger (17). Insulin antiserum (no. 2004), glucagon antiserum (no.
4304), and somatostatin antiserum (no. 1759–6) were diluted 1:400 and
1:1600 and GLUT2 antiserum (East Acres Biologicals, Southbridge, MA)
was diluted 1:1000 with 0.1 mol/liter phosphate buffer (pH 7.6)
containing 0.1% human serum albumin. The sections were examined by
means of a Zeiss microscope (Carl Zeiss, Oberkocken, Germany). The
examiner of the histological sections was not aware of the origin of
the specimens when evaluating the stained sections. Photographs were
taken with Agfa-pan APX25 film (Agfa-Gevaert AG, Leverkusen,
Germany).
Results and statistics
Results are expressed as the mean and SEM (in Fig. 2, A and B, only mean values are shown). Nonparametric Mann-Whitney U
test (two-tailed) was used for comparison of groups. P
< 0.05 was considered significant.
|
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
) was observed in the STZ
rats within 48 h after administration. Four days after
streptozotocin injection, the mean blood glucose concentration in the
STZ rats was above 20 mmol/liter, and it remained at this elevated
level throughout the 3 weeks of the study. The INS+STZ rats were
hypoglycemic during the insulin treatment, but on day 5, the blood
glucose concentration had returned to a normoglycemic level (day 5,
3.7 ± 1.3 nmol/liter; CTRL, 4.2 ± 0.9 nmol/liter;
P = 0.28). Similar results were observed in the INS
rats. To study whether a shorter period of insulin-induced hypoglycemia could affect the diabetogenic action of streptozotocin to a similar degree as the longer insulin treatment used in the INS+STZ group, we investigated a fifth group of rats (short term hypoglycemia; n = 12). This group was treated with insulin from days -1 to 1 with the same dose as that given to the INS+STZ and INS animals, and it was also injected with streptozotocin (60 mg/kg) on day 0. Blood glucose concentrations were 1.3 ± 0.2 mmol/liter on day 0, 2.2 ± 0.3 mmol/liter on day 3, 14.5 ± 0.6 mmol/liter on day 5, 18.5 ± 1.5 mmol/liter on day 14, and 20.8 ± 0.5 mmol/liter on day 21. These values were significantly (P < 0.01) different from that on day 5 compared to the CTRL group.
Body weight
The body weight (Fig. 2B) of the CTRL rats increased during the 3
weeks of the study from a mean of 212 ± 2 to 243 ± 7 g
(P < 0.01). In contrast, the body weight of the STZ
rats remained constant during the study (day -5, 213 ± 2 g;
day 21, 204 ± 4 g; P = 0.51). The INS and
INS+STZ rats gained more weight than the CTRL rats during the period of
insulin treatment.
RIA
The concentration of insulin in pancreatic tissue samples was
measured on days 2 and 21 (Table 1). The pancreatic
concentration of insulin was significantly reduced (P
< 0.01) in STZ, INS+STZ, and INS rats on day 2 compared to that in the
CTRL group. On day 21, the concentration of insulin in STZ rats was
significantly lower than those in INS+STZ, INS, and CTRL rats
(P < 0.05, P < 0.05, and
P < 0.01, respectively). In the INS+STZ group,
the pancreatic concentration of insulin was reduced to 32% of the
level in CTRL rats, but, on the other hand, it was significantly
(P < 0.05) higher (80 times) than that in STZ rats.
The concentration of insulin in tissue samples from INS rats on day 21
was not significantly different (P = 0.38) from the
CTRL value.
|
) was localized to the cellular membranes of ß-cells
in the pancreatic islets. In INS rats, GLUT2-IR was considerably
reduced on day 0 (Fig. 3B). In the same group of rats, however, on day
7, 6 days after last dose of insulin, the level of GLUT2-IR (Fig. 3C)
was comparable to that in the CTRL group. In the INS+STZ group,
GLUT2-IR was not detectable on day 2 (Fig. 3D); thereafter it
reappeared, but at a lower intensity on day 7 than that in the CTRL
rats (Fig. 3E) and an intensity of GLUT2-IR on day 21 comparable to
that in the CTRL group (Fig. 3F).
|
)
were located centrally in the islets of Langerhans, whereas the
glucagon-IR (Fig. 4B) cells were located in the periphery of the
islets.
|
) was
markedly reduced; in most islets insulin-IR cells were completely
absent. The amount of glucagon-IR cells (Fig. 4D), was increased, and
almost all remaining cells in the islets were 
-cells.
In the INS+STZ rats, a slightly reduced or equal number of insulin-IR
cells (Fig. 4E) and a slightly increased amount of glucagon-IR cells
(Fig. 4F) were found in comparison to the CTRL group.
There were few somatostatin-IR cells in all groups of rats. They were located in the periphery of the islets, and the amount of somatostatin-IR cells was unchanged among the various groups (data not shown).
Plasma insulin concentration
The plasma concentration of insulin (Table 2) on
day 21 was significantly lower (P < 0.05) in STZ rats
than that in CTRL rats. The plasma concentration of insulin in INS+STZ
rats was not significantly different from that in CTRL rats
(P = 0.14), but tended to be higher than that in the
STZ group (P = 0.08). The level in the INS group was
comparable (P = 0.83) to that in the CTRL group.
|
) on
day 21 was significantly increased (P < 0.05) in STZ
rats compared to those in INS+STZ, INS, and CTRL rats. The increased
concentration of total glucagon was due to a significant increment in
enteroglucagon (P < 0.05) and, to a lesser degree, a
significant increment in pancreatic glucagon (P <
0.05). |
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
The exact mechanism of the ß-cytotoxic action of streptozotocin is still conjectural, although it has been intensely investigated. The GLUT2 transporter that mediates the glucose uptake into rodent ß-cells (2) is assumed to also mediate the cellular uptake of streptozotocin (3).
In contrast to GLUT2, GLUT1, an isoform of GLUT2 with very low or no affinity for streptozotocin as a transport substrate (3), is predominately expressed on human ß-cells (18). This may explain why human fetal pancreatic ß-cells are resistant to the toxic effect of streptozotocin (19) and why patients with endocrine tumors treated with streptozotocin do not develop diabetes. Okamoto (5, 6) has shown that the streptozotocin-induced injuries in rodent ß-cells provoke DNA repair mechanisms involving the action of the NAD+-degrading enzyme poly(ADP-ribose) synthetase, and that the subsequent depletion of the cellular NAD+ pool depresses cell functions, leading to degeneration of ß-cells (5, 6, 7, 8, 20). In support of the suggested action of streptozotocin, inhibitors of poly(ADP-ribose) synthetase, e.g. nicotinamide and aminobenzamide, have been shown to prevent streptozotocin-diabetes through protection of the intracellular NAD+ pool (21, 22).
In the present study, short term insulin treatment reduced the pancreatic tissue concentration of insulin, insulin-IR, and GLUT2-IR in pancreatic islets in hypoglycemic rats. Thus, the protection of ß-cells against streptozotocin in the insulin-treated streptozotocin-injected rats might be caused by reduced cellular need for NAD+ in combination with reduced cellular uptake of the drug. A combined effect is likely, as changes in GLUT2 expression on ß-cells in vitro have been shown only to impede the uptake of glucose to a minor degree (23), and this phenomenon could also be true for the cellular uptake of streptozotocin into ß-cells in vivo.
The difference in the intensity of GLUT2-IR between insulin-treated rats on day 0 and insulin-treated streptozotocin-injected rats on day 2 may be caused by the ß-cell toxic effect of streptozotocin, as GLUT2-IR was still present on day 0, although considerably reduced, but it was not detectable 2 days after administration of streptozotocin. Reduced GLUT2-IR on ß-cells has been observed previously in hypoglycemic rats with insulinoma (3, 5) and in experimental type II diabetic rats neonatally induced with streptozotocin (24).
The severity of experimental diabetes and its persistence in rats depends on the dose of streptozotocin used (25). Intraperitoneal injection of streptozotocin in a dose of 45 mg/kg is known to cause a transient state of diabetes (26), and spontaneous recovery from experimental diabetes within 36 weeks in rats injected with streptozotocin in doses up to 50 mg/kg has previously been reported (27). However, low dose insulin treatment of rats injected with 50 mg/kg streptozotocin was previously shown to accelerate the time course for recovery from experimental diabetes to 2 weeks (28). This effect was suggested to be caused by the prevention of hyperglycemia, which was proposed per se to have a harmful effect on ß-cells (28). In contrast, ip injection of 60 mg/kg streptozotocin into rats has been shown to induce a persistent diabetic state (26, 29). We used the high dose of streptozotocin (60 mg/kg) and were still able to significantly protect ß-cells with insulin pretreatment.
The group of insulin-treated streptozotocin-injected rats in the present study was hypoglycemic 3 days before the injection of streptozotocin. To test the significance of short term hypoglycemia, another group of rats was treated with the same dose of insulin from days -1 to 1 and injected with streptozotocin on day 0. These rats were hypoglycemic from days 0–3 and, thus, at the time of streptozotocin injection. However, they became and remained hyperglycemic from day 5, which indicates that insulin pretreatment should be longer than 1 day to prevent the diabetogenic action of streptozotocin.
In the present study, the concentration of immunoreactive insulin in the pancreatic tissue samples from streptozotocin-diabetic rats was markedly reduced on day 21 compared to levels in INS+STZ, INS, and CTRL rats. In the INS+STZ group, the mean pancreatic concentration of insulin on day 21 was one third of that in the CTRL group, but 80 times higher than the pancreatic concentration of insulin in the STZ group. Immunohistochemical examination of the pancreatic islets from the INS+STZ rats revealed a fairly large number of insulin-IR ß-cells, in contrast to the absence of insulin-IR in islets from STZ rats. The amount of glucagon-IR cells was increased in islets from STZ rats compared to those in INS+STZ, INS, and CTRL rats. Likewise, the pancreatic glucagon concentration was significantly higher in diabetic rats. However, the amount of enteroglucagon was even higher in the diabetic rats. This was probably caused by hyperphagia, because an increased amount of nutrients in the distal gut is known to stimulate the secretion of enteroglucagon (30).
Streptozotocin-induced diabetes in laboratory animals has been widely used for research on diabetes and its long term complications. Control animals in these studies are usually injected with citrate buffer solution. However, streptozotocin is known to possess pharmacological effects other than its diabetogenic property (31, 32, 33), and extrapancreatic actions of streptozotocin cannot be excluded. The presence of GLUT2 in liver and kidney might explain the long term complications seen with hepatic and renal tumors in rodents treated with streptozotocin (26, 34). Because of the extrapancreatic effects of streptozotocin, it may be difficult to distinguish effects secondary to diabetes from those secondary to streptozotocin per se. Therefore, we propose that animals protected with high doses of insulin and then injected with streptozotocin, as described in the present work, could be used in future diabetes studies in which streptozotocin is employed. The control group will not become diabetic, yet any extrapancreatic actions of streptozotocin will be comparable in the control and diabetic groups.
In conclusion, the present report suggests that reduced expression of GLUT2, perhaps in combination with reduced cellular activity of ß-cells at the time of the administration of streptozotocin may prevent the diabetogenic action of streptozotocin.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
Received June 5, 1996.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
This article has been cited by other articles:
 |
|
 |
|
  S. A. Hinke Finding GAD: Early Detection of {beta}-Cell Injury Endocrinology, October 1, 2007; 148(10): 4568 - 4571. [Full Text] [PDF]
|
|
|
|
 |
|
 J Thulesen, B Hartmann, C Nielsen, J J Holst, and S S Poulsen Diabetic intestinal growth adaptation and glucagon-like peptide 2 in the rat: effects of dietary fibre Gut, November 1, 1999; 45(5): 672 - 678. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. E. Porter, R. L. Sorenson, P. Dann, A. Garcia-Ocana, A. F. Stewart, and R. C. Vasavada Progressive Pancreatic Islet Hyperplasia in the Islet-Targeted, Parathyroid Hormone-Related Protein-Overexpressing Mouse Endocrinology, September 1, 1998; 139(9): 3743 - 3751. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Endocrinology | Endocrine Reviews | J. Clin. End. & Metab. |
| Molecular Endocrinology | Recent Prog. Horm. Res. | All Endocrine Journals |
