| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Pacific Northwest Research Institute and the Division of Endocrinology (H.Z., T.Z., E.O., J.H., N.T., R.P.R.), Departments of Medicine and Pharmacology, Seattle, Washington 98122; and Departments of Pharmacology and Medicine (R.P.R.), University of Washington, Seattle, Washington 98108
Address all correspondence and requests for reprints to: R. Paul Robertson, Pacific Northwest Research Institute, 720 Broadway, Seattle, Washington 98122. E-mail: rpr{at}pnri.org.
| Abstract |
|---|
|
|
|---|
-cell function by insulin. | Introduction |
|---|
|
|
|---|
-cell, which usually secretes glucagon promptly when blood glucose drops to hypoglycemic levels, places the diabetic patient at risk for prolonged hypoglycemia. This is also true for the diminished epinephrine response (2, 3, 4, 5, 6). Although the mechanism of the defective epinephrine response remains obscure, the mechanism of the defective glucagon response has been postulated to be the lack of intraislet insulin regulation of the
-cell (7, 8, 9). This hypothesis posits that absence of ß-cells abrogates the necessary insulin switch-off signal to the
-cell to secrete glucagon. In contrast to the absent glucagon response to hypoglycemia, the glucagon response to nonglucose-related signals in type 1 diabetes remains intact (10). We reported that restoration of the intraislet insulin switch-off signal during hypoglycemia is sufficient to completely restore glucagon secretion in vivo in Wistar rats rendered diabetic with streptozotocin (11). These studies used an intrapancreatic artery infusion of insulin that was abruptly discontinued at the time the animals glucose levels reached hypoglycemic levels caused by intrajugular vein injections of insulin. Restoration of the glucagon response was not observed if saline was used instead of insulin or if the insulin infusion was not discontinued. Moreover, the response did not occur if the animals were euglycemic or hyperglycemic. We reported similar observations from in vitro experiments using perifused islets obtained from the same streptozotocin-treated Wistar rats (12).
To evaluate the hypothesis that our previous observations using streptozotocin-treated diabetic rats pertain to spontaneous, autoimmune-based insulin-dependent diabetes mellitus, we used the same approach of intrapancreatic artery insulin infusions in the diabetes-prone, lymphopenic BB rat. This animal does not have all of the features of autoimmune type 1 diabetes in humans but is generally accepted as a model of the human disease. We examined: 1) the glucagon response to hypoglycemia in 2-month-old BB rats before the onset of hyperglycemia; 2) the glucagon response to hypoglycemia in 4- to 5-month-old BB rats after the onset of hyperglycemia; 3) whether provision of an insulin switch-off signal via the pancreatic artery would restore the glucagon response to hypoglycemia in diabetic BB rats; and 4) whether changes in
-function might be related to combined effects of diabetes and increased maturity. The control studies related to development were conducted in nondiabetic BB and Wistar rats. It should be noted that these are no studies of aging in the precise sense of the word. Rather, they are studies of rats at two different ages.
| Materials and Methods |
|---|
|
|
|---|
Surgical procedure for BB rats
On the day of the study, approximately 3 wk after arrival at Pacific Northwest Research Institute, overnight-fasted animals were anesthetized by inhalation of 2% isofluorane. The abdomen was shaved and incisions were made through the skin to expose the peritoneum and then through the peritoneum to expose the abdominal contents, as previously described (11). The hepatic artery was identified and two ligatures were placed under the artery. The cephalad ligature was tied, and then a nick was made in the artery with a 21-gauge needle. A cannula was advanced retrograde through this nick so that its opening lodged upstream of the superior pancreaticoduodenal arterial (SPDa) branch. Then the caudad ligature was tied to keep the cannula in place. The right jugular vein was also cannulated. Both cannulae were filled with heparin to maintain patency. After surgery, rats were rested for 30 min, and anesthesia was maintained by 1% isofluorane until the end of the experimental protocol, at which time they were killed.
Study protocols
Two-month-old male BB rats that were not yet hyperglycemic were given an intrajugular infusion of insulin (Humulin 0.5 U/ml, 50 µl/min) to induce hypoglycemia. Jugular vein insulin infusion was turned off when iv blood glucose levels reached less than 60 mg/dl; 0.3-ml blood samples were collected for determination of blood glucose, glucagon, and C-peptide levels. Two months later the same protocol was repeated.
The majority of BB rats developed spontaneous diabetes at 7090 d of age. Jugular vein and SPDa were cannulated in anesthetized 4- to 5-month-old diabetic BB rats. After resting the rats for 30 min, 0.5 U/ml insulin was infused into the jugular vein to decrease blood glucose to approximately 100 mg/dl. After collection of basal blood samples, insulin (0.025 U/min) was infused into the SPDa. Ten minutes after SPDa infusion, another sample was collected. When the glucose level became less than 60 mg/dl, the pancreatic artery insulin and jugular vein insulin infusions were switched off and blood was sampled at 0, 15, 30, 45, 60, and 90 min.
Two- and 4- to 5-month-old male Wistar rats were studied to examine the effect on
-cell function over time. In these studies, the Wistar animals were given iv insulin to examine glucagon responses to hypoglycemia, arginine (0.5 g/kg) to assess glucagon responses to a nonhypoglycemic stimulus, and ip glucose tolerance tests (1 g/kg) to establish the status of carbohydrate tolerance.
Assays
Plasma glucose levels were measured immediately using an Accu-check Advance meter (Roche Diagnostics, Indianapolis, IN). Blood samples were collected into heparin-coated ice-chilled tubes. Trasylol (1000 IU/ml) was added to prevent degradation of glucagon. Plasma C-peptide and glucagon were measured using a rat C-peptide and glucagon RIAs (Linco Research, St. Charles, MO). Plasma insulin levels after ip glucose challenge were measured by rat ELISA kit (Mercodia AB, Uppsala, Sweden). Intraassay and interassay coefficients of variation for the glucose, C-peptide, glucagon, and insulin assays were all 46 and 814%, respectively.
Statistics
Data are presented as mean ± SE. When glucagon data are expressed as delta, peak values minus the average of the two basal values are shown. Significant differences in physiological responses were first identified by ANOVA; only if significance was found, specific points were examined for significant differences using either parametric or nonparametric tests. Results were normally distributed, except for the insulin values in Fig. 6
, which were analyzed by paired, nonparametric (Wilcoxon) testing. The data in Figs. 2
and 5
were analyzed by paired Students t testing. The data in Figs. 1
, 3
, and 4
were analyzed by unpaired Students t tests. P < 0.05 was considered statistically significant. We recognize an analysis of the data using paired testing across all the metabolic tests performed would be the most desirable. Ideally, each BB rat would have been studied at 2 months (to document intact glucagon responses) and 45 months (to demonstrate absent glucagon responses) with an insulin tolerance test and then again studied with the SPDa procedure a few weeks later (to demonstrate restoration of glucagon responses). However, federal regulations strongly recommend that each animal can undergo only one major survival surgery.
|
|
|
|
|
|
| Results |
|---|
|
|
|---|
The 2-month-old BB rats (228 ± 5 g) weighed less (P < 0.001) than the 4- to 5-month-old nondiabetic BB animals (350 ± 15 g). Compared with the 2-month-old BB rats that were not hyperglycemic and that had robust glucagon responses to insulin-induced hypoglycemia, 4- to 5-month-old BB rats that were not diabetic had markedly decreased glucagon responses to hypoglycemia (2-month-old nondiabetic BB = 1038 ± 112, n = 12; 4- to 5-month-old nondiabetic BB = 217 ± 71, n = 7; peak delta glucagon, picograms per milliliter; P < 0.0001; Fig. 3
). The durations of the insulin infusions in the 2-month-old and 4- to 5-month-old nondiabetic BB rats were 41 ± 3 vs. 60 ± 5 min (P < 0.01). Peak insulin values for the nondiabetic 2- and 4- to 5-month-old BB rats were, respectively, 69.9 ± 14.3 and 81.3 ± 37.2 ng/ml, P = ns. The basal C-peptide levels in the 4- to 5-month-old animals were arithmetically, but not significantly, higher, compared with the 2-month-old animals (2 month old = 349 ± 29; 45 month old = 482 ± 139; picomoles, P = ns). There was no significant difference in the glucagon response to hypoglycemia between the 4- to 5-month-old nondiabetic BB rats and the diabetic BB rats of the same age who received the intrapancreatic artery insulin switch-off signal (Fig. 4
).
The 2- and 4- to 5-month-old Wistar rats were aged 64 ± 1 and 138 ± 1 d, respectively. The 2-month-old Wistar rats (277 ± 7 g) weighed less (P < 0.001) than the 4- to 5-month-old animals (499 ± 15 g). Two-month-old nondiabetic Wistar rats had greater glucagon responses to hypoglycemia than 4- to 5-month-old Wistar rats (2-month-old animals = 1173 ± 151, n = 11; 4- to 5-month-old animals = 414 ± 70 n = 11; delta peak glucagon, picograms per milliliter; P < 0.0002; Fig. 4
). The durations of the insulin infusions in the 2- and 4- to 5-month-old Wistar rats were 54 ± 3 vs. 74 ± 8 min, P < 0.05). The basal C-peptide levels in the 4- to 5-month-old animals were significantly higher than in the 2-month-old animals (2 months old = 287 ± 51; 45 months old = 597 ± 71 pM; P < 0.002). Peak insulin values for the 2- and 4- to 5-month-old Wistar rats were, respectively, 92.4 ± 18.9 and 106.7 ± 23.6 ng/ml, P = ns. Interestingly, the 4- to 5-month-old Wistar rats had a significantly greater glucagon response to iv arginine, compared with 2-month-old Wistar rats (142 ± 34, n = 4 vs. 42 ± 10, n = 4; peak delta glucagon, picograms per milliliter, P = 0.03; Fig. 5
). To assess carbohydrate tolerance as a function of aging in both strains of animals, we performed ip glucose tolerance tests. Four- to 5-month-old nondiabetic BB rats and 4- to 5-month-old Wistar rats had diminished ip glucose tolerance, compared with their 2-month-old counterparts (nondiabetic BB rats: 2 month old = 24 ± 10; 45 month old = 60 ± 7; n = 7 pairs: Wistar rats: 2 month old = 58 ± 13; 45 month old = 113 ± 8; 7 pairs; area under the curve; milligrams per deciliter per minute glucose; P < 0.001 in both strains). The 4- to 5-month-old animals also had higher basal insulin values (BB rats: 2 month old = 0.22 ± 0.03; 45 month old = 0.46 ± 0.05, n = 7 pairs: Wistar rats: 2 month old = 0.14 ± 0.02; 45 month old = 0.88 ± 0.38, n = 7 pairs; basal insulin; nanograms per milliliter; both P < 0.01) and higher insulin levels during the glucose tolerance test in BB rats (0.032 ± 0.031 vs. 0.348 ± 0.077, n = 7 pairs; P < 0.003) but not Wistar rats (0.164 ± 0.075 vs. 0.311 ± 0.322, n = 7 pairs, P = ns; area under curve in nanograms per milliliter per minute; Fig. 6
).
| Discussion |
|---|
|
|
|---|
One difficulty in interpreting these studies was to account for the age differences in the nondiabetic and the diabetic BB rats because the onset of diabetes occurs at times up to 3 months of age. We approached this issue by conducting control studies of both
- and ß-cell function in nondiabetic BB and nondiabetic Wistar rats. We observed that 4- to 5-month-old nondiabetic BB rats, as well as 4- to 5-month-old nondiabetic Wistar rats, developed decreased glucose tolerance and higher insulin levels, suggesting diminished ß-cell function and less insulin sensitivity. We also observed that the glucagon response to arginine was markedly greater in the 4- to 5-month-old Wistar animals. This suggests the
-cells in the 4- to 5-month-old animals were less sensitive to insulins tonic suppression of glucagon secretion. In support of the contention that over time animals develop less insulin sensitivity at the level of the
-cell, it is noteworthy that the baseline C-peptide values during the insulin tolerance tests tended to be higher in the 4- to 5-month-old animals, which should have provided a greater ß-cell switch-off signal. Yet the glucagon response was smaller. Thus, it appears that as it develops, this animal model is accompanied by less insulin sensitivity in the
-cell, a phenomenon recently suggested in a study of normal humans (13). Others have reported development of obesity and insulin resistance within the time frame of the animals we studied (14, 15). Unfortunately, because we measured only total body weights in this study, we cannot judge whether this lessening of insulin sensitivity over time is related to increased fat stores because we made no assessment of adipose mass.
These findings provide further support for the intraislet insulin hypothesis as an explanation of defective glucagon secretion during hypoglycemia in diabetes. Extending our experiments from the model of streptozotocin-induced diabetes is important because this model does not involve spontaneous, autoimmune disease and therefore is less relevant to autoimmune diabetes in humans. Provision of the SPDa insulin infusion switch-off signal elicited glucagon responses to hypoglycemia in the diabetic BB rats that were not less than responses observed in nondiabetic, age-matched BB rat controls. The age at which we observed a decrease in glucagon responses to hypoglycemia is consistent with earlier experiments of Jacob et al. (16), who documented the time course of the loss of glucagon responses to hypoglycemia in the BB rat.
The extent to which our results from animal experiments in vivo and islets in vitro (Refs. 11 and 12 and current manuscript) pertain to human subjects with type 1 diabetes remains to be demonstrated. To our knowledge there has been no published demonstration that restoration of an intraislet switch-off signal will restore glucagon secretion during hypoglycemia in type 1 diabetic subjects. This would be virtually impossible using the technique we used in animals because it would require catheterization of the pancreatic arteries of diabetic patients. Safety concerns preclude the use of systemic infusions of insulin to reproduce the switch-off signal because of the high systemically circulating insulin concentrations that would be needed to mimic the levels delivered from the ß-cell to the
-cell. However, Israelian et al. (17) recently published data that may be relevant because they involved insulin-requiring type 2 diabetic patients. In this work glucagon responses during hypoglycemic clamps were studied on two separate occasions in 12 subjects. On one occasion, tolbutamide was infused before the clamp to elevate insulin levels 2-fold so that the decrement in insulin during hypoglycemia was greater. On the other occasion, saline rather than tolbutamide was used. The glucagon response during hypoglycemia was 2-fold greater when the clamp was preceded with tolbutamide, compared with the saline control infusion. This paradigm would not be usable when studying type 1 diabetic subjects because it depends on the ability of tolbutamide to stimulate endogenous insulin secretion.
In conclusion, these studies demonstrate that defective glucagon responses to insulin-induced hypoglycemia can be corrected in an autoimmune model of diabetes by provision of intraislet insulin via a pancreaticoduodenal artery infusion that is switched off at the time of systemic hypoglycemia. We have also observed that glucagon responses to hypoglycemia become strikingly less in more mature rats and suggest that decreased insulin sensitivity at the level of the
-cell might cause this phenomenon.
| Footnotes |
|---|
Abbreviations: ns, Not significant; SPDa, superior pancreaticoduodenal arterial.
This work was supported by National Institutes of Health Grant RO1 DK39994 (to R.P.R.) and American Diabetes Association mentor-based fellowship (to H.Z. and R.P.R.).
Disclosure Statement: The authors have nothing to disclose.
Received October 10, 2006.
Accepted for publication February 8, 2007.
| References |
|---|
|
|
|---|
cell defect. Science 182:171173
-cell function by the ß-cell during hypoglycemia in Wistar rats: the "switch-off" hypothesis. Diabetes 53:14821487
-cell function by the ß-cell in isolated human and rat islets deprived of glucose: the "switch-off" hypothesis. Diabetes 53:14881495
-cell response to hypoglycemia in diabetic BB rats. Metabolism 45:14221426This article has been cited by other articles:
![]() |
X. Fan, Y. Ding, S. Brown, L. Zhou, M. Shaw, M. C. Vella, H. Cheng, E. C. McNay, R. S. Sherwin, and R. J. McCrimmon Hypothalamic AMP-activated protein kinase activation with AICAR amplifies counterregulatory responses to hypoglycemia in a rodent model of type 1 diabetes Am J Physiol Regulatory Integrative Comp Physiol, June 1, 2009; 296(6): R1702 - R1708. [Abstract] [Full Text] [PDF] |
||||
![]() |
I. Quesada, E. Tuduri, C. Ripoll, and A. Nadal Physiology of the pancreatic {alpha}-cell and glucagon secretion: role in glucose homeostasis and diabetes J. Endocrinol., October 1, 2008; 199(1): 5 - 19. [Abstract] [Full Text] [PDF] |
||||
![]() |
P. Bansal and Q. Wang Insulin as a physiological modulator of glucagon secretion Am J Physiol Endocrinol Metab, October 1, 2008; 295(4): E751 - E761. [Abstract] [Full Text] [PDF] |
||||
![]() |
L. S. Farhy, Z. Du, Q. Zeng, P. P. Veldhuis, M. L. Johnson, K. L. Brayman, and A. L. McCall Amplification of pulsatile glucagon counterregulation by switch-off of {alpha}-cell-suppressing signals in streptozotocin-treated rats Am J Physiol Endocrinol Metab, September 1, 2008; 295(3): E575 - E585. [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 |