IMedical Hypotheses Mdicoi Hpolh8sax mm) 36.284-288 0 L.cmsmn ckup UJCL4d 1991
Amj4n and Insulin Co-replacement Therapy for Insulin-Dependent (Type I) Diabetes Mellitus G. J. S. COOPER Amylin Corporation, 9373 Towne Centre Drive, Suite 250, San Diego, CA 92721, USA
Abstract - Amylin is a proteinaceous hormone secreted from insulin-producing pancreatic j3-cells following stimulation by food molecules such as glucose and arginine. Amylin decreases insulin-stimulated glucose uptake in skeletal muscle and counteracts the ability of insulin to suppress output of glucose from the liver. Substantial evidence supports the view that amylin is a second glucoregulatory hormone produced from islet p-cells, which can modulate a number of metabolic processes also regulated by insulin. The islet P-cell may therefore transmit a dual message to peripheral tissues through the two hormones, insulin and amylin. Like insulin, amylin is deficient in individuals with autoimmune diabetes mellitus. Since amylin can modulate processes of fuel metabolism in key tissues, amylin deficiency could contribute to the clinical course in patients with autoimmune diabetes. Here, I propose that amylin lack plays a significant role to promote the tendency to hypoglycemia and defective glycemic control characteristic of insulin-treated patients with autoimmune diabetes. Treatment of such diabetics with injections of amylin as well as insulin is being evaluated with the aim of lessening the incidence and severity of hypoglycemia and improving glycemic control.
Introduction Amylin is a newly discovered peptide (Fig.) which was purified from amyloid extracted from the pancreases of human patients with non-insulin-dependent diabetes mellitus (NIDDM) (1, 2, 3). Although the finding of amyloid in the islets of Langerhans was an original clue linking disease of the endocrine pancreas with the cause of diabetes (4), the nature of the protein which forms this amyloid remained hidden until over half a century after the discovery of insulin (5). It now appears that amylin is a second hormone (2,3) produced (1,6-g) and secreted (10) mainly from islet pDate received 27 May 1991 Date accepted 28 June 1991
cells (8.9) which also secrete insulin. Amylin modulates peripheral processes of carbohydrate metabolism regulated by insulin (2, 3, ll-15), and also increases hepatic glucose output in the presence of insulin (14). Co-secretion of amylin with insulin is believed to prevent the primary occurrence of the insulin-induced hypoglycemia which might otherwise occur in normal persons (3). Insulin-dependent diabetes mellitus (IDDM) is caused by destruction of islet p-cells. We have shown that amylin is deficient in the BB/Wor rat (17), a widely accepted animal model for the human syndrome (18). Lack of amylin will produce several ef-
284
AMYLlN AND INSULIN CO-REPLACEMENTTlERAPY FOR IDDM
Fig. The complete stntc&ureof human amylin as determined by protein chemical studies (1.11). and subsequently co&m& by moleculargenetic analysis (9. 20). Both the intramolecular disulfide bridge and the cnrboxy-terminal amide group ate essential for full biological activity of amylin to modulate glucose metabolism in skeletal muscle (3, 7. 9, 11).
fects, including an increased susceptibility to insulininduced hypoglycemia (3). Superior control of blood glucose levels and protection against the hazards of insulin-induced hypoglycemia from correction of the double endocrine deficiency in IDDM, namely the codeficiency of insulin and amylin, is the basis for further research into amylin replacement as an aid in the attainment of an as yet elusive goal of the treatment of IDDM, physiologic blood glucose control (19). Interplay of amylin and insulin The following description of the interplay between
amylin and insulin in the control of fuel metabolism reflects in some measure protein chemical (1 , 11,16), molecular genetic (8, 9, 20), metabolic biochemical (2, 3, 11, 12, 13, 15), physiologic (14), and clinical studies performed over the course of the last 5 years. This brief account summarizes only those aspects of this endocrine system which relate directly to the question of IDDM and amylin replacement therapy. The linkage between over-secretion of amylin and the production of insulin resistance and non-insulin-dependent diabetes and other metabolic disorders, such as obesity, atherosclerosis and hypertension, has been reviewed recently (3, 11,21). Here, I focus on insulin hypersensitivity and the tendency to hypoglycemia from amylin lack, a condition which I believe is en-
hanced when insulin alone is replaced in patients with destroyed islet pcells. Amylin is secreted along with insulin in response to stimulation by typical nutrient secretagogues (glucose and arginine) (10). Major actions of amylin on fuel metabolism reported thus far include: suppression of insulin-stimulated uptake and incorporation of glucose into skeletal muscle glycogen (2,3, 11, 12, 13); decreasing muscle glycogen content (22): stimulation of skeletal muscle glycogenolysis by promoting conversion of glycogen phosphorylase b to a (23, 24); inhibition of muscle glycogen synthase (24); stimulation of liver glycogenolysis (15); stimulation of liver gluconeogenesis (15, 25); stimulation of hepatic glucose output, even in the presence of insulin (14); and elevation of blood lactate and glucose levels in fasted and fed rats (25). Amylin can therefore limit the ability of insulin to (i) reduce hepatic glucose production (14), and (ii) promote transfer of blood glucose into muscle glycogen. By these actions amylin will impact the ability of insulin to cause hypoglycemia. When amylin levels are relatively low, insulin tends to promote storage of glucose as glycogen in skeletal muscle and liver. Amylin does not inhibit insulinstimulated carbohydrate metabolism in adipose tissues (2). Therefore, as amylin levels increase and cause progressive insulin resistance in skeletal muscle (2, 3, 12, 13), there will be a shift in the ultimate site
MEDICALHYF’OTHESES
286 of disposal of carbohydrate. The liver will increasingly tend to become an organ of glucose and lipid synthesis and output, and insulin will then act mainly to promote conversion of carbohydrate into triglyceride for ultimate storage in adipose tissue. The major actions of amylin on glucose metabolism thus enable the body to switch the site of carbohydrate storage away from skeletal muscle glycogen towards triglyceride in adipose tissue (2,3). Amylin, in combination with insulin, serves as a signal which determines the site(s) of a carbohydrate load (3). The operation of this switching mechanism is dependent on the presence of both insulin and amylin. The emerging picture is one that amylin and insulin act together to direct the flux and sites of storage of metabolic fuels. The lack of demonstrated effects of amylin on carbohydrate metabolism in adipose tissue (2) indicates that the effect of certain combinations of insulin and amylin will switch the storage site of substrate carbohydrate into adipose tissue triglyceride. Amylin may be thought of as regulating or switching insulin-stimulated anabolic pathways between the glycogen of liver and particularly skeletal muscle, and adipose tissue triglyceride. The combined effects of amylin and insulin may represent the biochemical expression of the ‘thrifty genotype’ (26). Although the actions of amylin are similar in some respects to those of well known counter-regulatory hormones, particularly glucagon, epinephrine and norepinephrine, there are major differences. Fist, amylin secretion is stimulated by glucose (lo), the same nutrient that is the primary stimulus for insulin secretion. Furthermore, as shown in the studies published thus far, amylin and insulin secretion follow similar time courses (10). Insulin normally acts, therefore, against a backdrop of amylin action. Amylin and insulin tend to be secreted in parallel, although a degree of differential secretion, e.g. amylin secretion increasing relative to insulin secretion, has also been observed (lo), which is not inconsistent for two cooperating but distinct regulators. Thus amylin can serve as a ‘feed-forward’ controller and thereby as a ‘buffer’ or ‘back-stop’ limiting the ability of insulin to promote peripheral glucose disposal and inhibit hepatic glucose output. In contrast, the other known counter-regulatory hormones are brought into play by the occurrence of hypoglycemia (27). While these hormones respond to hypoglycemia (28) and are important in recovery (28). they do not ‘anticipate’ and prevent it. Interestingly, careful studies in normal man following induction of hypoglycemia by small physiologic doses of insulin showed that reversal of insulin-
induced suppression of hepatic glucose production preceded measurable increments in systemic levels of known counter-regulatory hormones by at least 30 min (28). These studies led to the inference that there may exist an unidentified protective physiologic mechanism capable of reversing insulin-mediated inhibition of hepatic glucose output in normal persons, which is active when small doses of insulin are given, and which prevents circulating glucose in normal persons from falling to levels at which release of known counter-regulatory hormones is triggered (28). I suggest that this process is mediated by amylin. Amylin deficiency in IDDM IDDM is caused by lymphocyte-mediated autoimmune destruction of pancreatic p-cells (29). Amylin is mainly contained in these cells, and our data (Koda J E, Cooper G J S et al, in preparation) shows that human patients with IDDM are deficient in amylin as well as insulin. As noted above, evidence now sup ports the view that amylin is primarily produced in and secreted from islet p-cells. Certain studies have demonstrated co-localization of amylin and insulin in pancreatic islet cells (6, 30). Amylin mRNA has also been localized to the islets (8). We have recently shown that pancreatic amylin mRNA levels are reduced about seven-fold when diabetes is fully estab lished in the BB/Wor rat (17), a well accepted animal model of IDDM. Amylin and insulin are also both deficient in the pancreases of rats with streptozotocininduced diabetes (10). In a recent preliminary communication, the absence of detectable amylin levels in plasma from patients with IDDM was report& (31). From a metabolic perspective, IDDM may therefore be re-defined as a syndrome resulting from the deficiency of two hormones, insulin and amylin. Thereapy of IDDM for the past 68 years has relied on insulin replacement (5), which has been one of the most important medical advances of this century. However, even after seven decades of increasing refinement of this treatment, patients with IDDM do not enjoy normal health or life expectancy (32). Ftigorous control of blood glucose improves matters, but steering a passage between the scylla of inadeqtate reduction in blood glucose levels and the charybdis of hazardous hypoglycemia remains difficult. I suggest that a major reason for this problem in IDDM is lack of amylin. Although lack of insulin is primarily responsible for the occurrence of diabetic ketoacidosis, deficiency of amylin will partly determine the response of type 1 diabetics to insulin replacement therapy. Doubtless the degree of amylin lack will vary
287
AhSYLlN AND INSULIN CO-REPLACEMENT THERAPY FOR IDDM
between patients and with progression of disease. Perhaps those with the most difficult-to-control diabetes will have the lowest or most fluctuating amylin levels. Hypoglycemiaand long-term complicationsin insulin-treatedIDDM: the scylla and charybdis of physiologic blood glucose control It is now an accepted clinical goal of diabetes therapy to maintain blood glucose levels within the physiologic range (33, 34). However, euglycemia must be achieved safely and without concomitant hypoglycemia (27, 35). Many of the chronic complications of IDDM, such as retinopathy. neuropathy and nephropathy, are thought to be caused by chronic exposure of tissues to increased blood glucose levels and the associated metabolic abnormalities (27). Unfortunately, it has remained almost impossible to maintain chronically normal blood glucose levels in type I diabetics using currently available and clinically practical methods of insulin therapy (34). This difficulty is caused in substantial measure by insulininduced hypoglycemia (27). There is a definite increase in the risk of hypoglycemia with the most carefully designed insulin regimens aimed at producing near-physiologic blood glucose control (27). The observed biological effects of amylin are consistent with the concept that deficiency of this hormone contributes to the occurrence and severity of hypoglycemia in insulin-treated IDDM. Its actions are also consistent with a role for amylin in preventing hypoglycemia in normal persons, and with amylin deficiency as a significant contributor to the metabolic disturbance in IDDM. The occurrence of insulin-induced hypoglycemia in the setting of insulin replacment therapy may not be accounted for by observed deficiencies of classical counter-regulatory mechanisms, which normally occur late in the course of established IDDM (28). However, defects in these counter-regulatory systems probably do account for differing severity of, and rates of recovery from, hypoglycemia once it actually occurs (27). Expected outcomes of amylin replacement therapy The experimentally observed effects of amylin indicate that amylin deficiency in IDDM will predispose to hypoglycemia during treatment with insulin, and support the concept that amylin replacement therapy will complement insulin therapy and lessen the risk of hypoglycemia. Glycemic control should also be improved because amylin replacment will allow administration of larger doses of insulin to reduce hyper-
glycemia. It is also believed that the combined actions of amylin and insulin to promote triglycetide formation will provide a more normal adipose tissue pattern in thin type 1 diabetics. Finally, a&bough mechanisms remain uncertain, amylin deficiency per se may contribute to some of the long-term complications of IDDM, lending further benefit to amylin replacement. We are currently performing animal studies to explore amylin and insulin co-replacement therapy for IDDM. These studies will be extended to human subjects following preparation of sufficient quantities of chemically pure and biologically active amylin (36). and after requisite safety and pharmacological studies have been completed. Acknowledgements I wish to thank all of my colleagues who have contributed to the various aspects of this work over the past 5 years, pauicularly Ken Reid, Tony Willis, Brmdan Leighton, Anne R-s, Andrew Young, Joy Koda, Laum Lehman de Gaeta. Elisabeth Albrecht, Ming-Wei Wang, and Huei-Jen Su Huang. I also thank Tim Rink for his constructive criticism of the manuscript.
References 1. Cocper G J S, Willis A C, Clark A, Turner R C, Sim R B, Reid K B M. Purification and characterixation of a pep& from amyloid-rich pancreases of type 2 diabetic patients. Proc Nam Acnd Sci USA 84: 8628, 1987. 2. Cooper G J S, Leighton B, Dimitriadis G D, Parry-Billings M, Kowalchuk J M. Howland K. Rothbard J B. Wiis A C. Reid K B M. Amylin found in amyloid deposits in human type 2 diabetes mellitus may be a hormone that regulates glycogen metabolism in skeletal muscle. Proc Nam Acad Sci USA 85: 7763, 1988. 3. Cooper G J S, Day A J, Willis A C, Roberts A N, Reid K B M, Leighton B. Amylin and the amylin gene. Structure. function and relationship to islet amyloid and to diabetes mellitus. Biochim Bioohvs Acta 1014: 247. 1989. 4. Opie E L The-relation of diabetes mellitus to lesions of the pancreas. Hyaline degeneration of the islands of Langemans. J Exp Med 5: 527. 1900-1901. 5. Banting FG, Best CH. The internal secretion of the pancreas. J lab Clin Med 7: 256. 1922. 6. Clark A, Cooper G J S, Lewis C E, Morris J F, Willis A C. Reid K B M, Turner R C. Islet amyloid formed fran diabetes-associated peptide may be pathogenic in type-2 diabetes. Lancet 2: 231, 1987. 7. Cooper G J S, Willis A C, Leighton B. Amylin hormone. Nature 340: 272, 1989. 8. l&fen J D, Newgard C B, Okamoto H, Miibum J L. Luskey K L. Rat amylin: cloning and tissue-specific expression in pancreatic islets. Proc Nam Acad Sci USA 86: 3127, 1989. 9. Roberts A N. Leighton B. Todd J A, Cockbum D, Schofield P N, Sutton R, Holt S, Boyd Y. Day A J, Foot E A, Willis A C. Reid K B M, Cooper G J S. Molecular and functional characterization of amylin. a peptide associated with type 2 diabetes mellitus. Prcc Nam Acad Sci USA 86: 9662, 1989. 10. Ogawa A. Harris V, McCorkle S K. Unger R H. luskey K L Amylin secretion from the rat pancreas and its selective loss
288 tmatment. J Clin Invest 85: 973, 1990. atIer streptoxuocin 11. Cooper G J S. ‘Ibe chamcterixatias of amylin and analysis of its role in diabetes mellitus. D Phil Thesis, University of Oxford, 1989. 12 Leightcn B, Cocper G J S. Pancreatic amylin and mlcitatin gene-related peptide cause resistance to insulin in skeletal muscle in vi&.-Nature 335: 632.1988. 13. Leiahton B. Feat E A. Cooner G J S. Kina J M. Calcita& gene-Alated peptide-l (t&P-l) is a polent mgulator of glycogen metabolism in rat skeletal muscle. FBBS (Fed Em Biochem Sot) Lett 249: 357.1989. 14. Molina J M, Cocper G J S, Leighton B. Olefsky J M. Induction of insulin resistance in vivo by amylin and calcitonin generelated peptide. Diabetes 39: 2&O.1990. 15. Ciaraldi T P. Cocuer G J S. Stolne M. In vitm effects of amylin en c&x&irate metabolism. Diabetes 39 Suppl 1: 149A. 19%. 16. Cooper G J S, Leightan B. Wiis A C. Day A J. The amyhn superfamily: a novel grouping of biologically active polypeptides related to the insulin A-chain. Prog Growth Factor Res 1: 96, 1989. 17. Huang H-JS, Cooper G J S. Young A A, Johnson M J. De.8 ciency of amylin expression in the pancreas of auto-immune BB/Wor diabetic rats. J Cell Biochem Suppl 15B: 67.1991. 18. Monies J P, Dcsemone J, Rossini A A. The BB rat. Diabetes Metab Rev 3: 725.1987. B. ‘lhe physiologic replacement of insulin. An elusive 19. Zian goal. N Engl J Med 321: 363. 1989. 20. Nisbi M, Ghan S J. Nagamatsu S, Bell G I, Steiner D F. Conservation of the sequence of islet amyloid polypcptide in five mammals is consistent with its putative role as an islet hormone. Proc Natn Acad Sci USA 86: 5738.1989. 21. Leighton B. Cooper G J S. The role of amylin in the insulin resistance of non-insulindependcnt diabetes mellitus. Trends B&hem Sci 15: 295. 1990. 22. Young D A, Deems R 0. Deacon R W, McIntosh R H, Foley J E. Fffects of amylin on glucose metabolism and glycogenolysis in vivo and in vitro. Am J Physiol259: E457, 1990. 23. Young A A, Mott D M. Stone K. Couper G J S. Amylin activates glycogen phosphorylase in the isolated soleus muscle
MEDKALHYPtYrHEsEs
of the rat. FEBS (Fed Eur Biochem Sot) Lett 281: 149,199l. 24. Deans R 0, Deaccn R W, Yamg D A. Amylin activates glycogen phosphorylase and inactivates glycogen rynthase by a CAMP-independent mechanism. Biochcsn Biophyr Res Commun 174: 716.1991. 25. Young A A, Wang M-W, Rink T J, Cooper G J S. In vivo activity of amylin. J Physiol, 1991, in press (Pmceedmgs). 26. Neel J V. Diabetes mellitus: a ‘thrifty’ genotype rendered detrimentalby ‘progress’?Am J Hum Genet 14: 353, 1962. 27. Cryer P E. Binder C. Bolli G B, Cherrington A D, Gale E AM, Gerich J E. Sherwin R S. Hvnoalvcemia in IDDM. Diabetes __ __ 38: 1193. i989. 28. Amiel S A, Tamborlane W V, Sacea L. Sherwin R S. Hypoglycemia mtd glucose ccunterregulation in normal and insulindePendem diabetic subjects. Diabetes Metab Rev 4: 7 1.1988. 29. Eisenlxuth G S. lope I diabetes mellitus. A chronic autoimmune disease. N Engl J Med 314: 1360, 1986. 30. Johnson K H. O’Brien T D. Betsholtx C, Westermark R Islet amyloid. islet-amyloid polypeptide, and diabetes mellitus. N Engl J Med 321: 513.1989. 31. Hartter E, Svoboda T. Ludvik B. Schuller M, Lell B, Kucnburg E, Brunnbauer M, Wolosxcmk W, Prager R. Basal and stimulated plasma levels of pancreatic amylin indicate its cosecretion with insulin in humans. Diabetologia 34: 52. 1991. 32. Borch-Johnsen K, Knitter S. Dcckett T. Mortality of type 1 (insulin-dependent) diabetes mellitus in Denmark: a study of relative mortality in 2930 Danish type 1 diabetic patients diaenosed from 1933 to 1972. Diabetoloaia 29: 767.1986. 33. Shafir E, Bergman M. Felig P ‘lhi Endocrine Pancreas: Diabetes Mellitus pp 1125-1129 in Endocrinology and Metabolism. 2nd ed. (Felig P, Baxter J D, Broadus A E. Frohman L A, eds) McGraw-Hill, New Yc+k, 1987. 34. Sonsken P H, Judd S L. Lowy C. Home monitoring of blood glucose. Method for improving diabetic control. Lancct 1: 729. 1978. 35. Gale E A M, Tattersall R B. Unrccognised nocturnal hypoglycemia in insulin-treated diabetics. Lancet 1: 1049. 1979. 36. Albrecht E, Young A A, Cooper G J S, Lehman de Gaeta LS. Synthesis and biological activity of human and canine amylin. Int J Peptide Prot Res 1991, submitted.
Amj4n and Insulin Co-replacement Therapy for Insulin-Dependent (Type I) Diabetes Mellitus G. J. S. COOPER Amylin Corporation, 9373 Towne Centre Drive, Suite 250, San Diego, CA 92721, USA
Abstract - Amylin is a proteinaceous hormone secreted from insulin-producing pancreatic j3-cells following stimulation by food molecules such as glucose and arginine. Amylin decreases insulin-stimulated glucose uptake in skeletal muscle and counteracts the ability of insulin to suppress output of glucose from the liver. Substantial evidence supports the view that amylin is a second glucoregulatory hormone produced from islet p-cells, which can modulate a number of metabolic processes also regulated by insulin. The islet P-cell may therefore transmit a dual message to peripheral tissues through the two hormones, insulin and amylin. Like insulin, amylin is deficient in individuals with autoimmune diabetes mellitus. Since amylin can modulate processes of fuel metabolism in key tissues, amylin deficiency could contribute to the clinical course in patients with autoimmune diabetes. Here, I propose that amylin lack plays a significant role to promote the tendency to hypoglycemia and defective glycemic control characteristic of insulin-treated patients with autoimmune diabetes. Treatment of such diabetics with injections of amylin as well as insulin is being evaluated with the aim of lessening the incidence and severity of hypoglycemia and improving glycemic control.
Introduction Amylin is a newly discovered peptide (Fig.) which was purified from amyloid extracted from the pancreases of human patients with non-insulin-dependent diabetes mellitus (NIDDM) (1, 2, 3). Although the finding of amyloid in the islets of Langerhans was an original clue linking disease of the endocrine pancreas with the cause of diabetes (4), the nature of the protein which forms this amyloid remained hidden until over half a century after the discovery of insulin (5). It now appears that amylin is a second hormone (2,3) produced (1,6-g) and secreted (10) mainly from islet pDate received 27 May 1991 Date accepted 28 June 1991
cells (8.9) which also secrete insulin. Amylin modulates peripheral processes of carbohydrate metabolism regulated by insulin (2, 3, ll-15), and also increases hepatic glucose output in the presence of insulin (14). Co-secretion of amylin with insulin is believed to prevent the primary occurrence of the insulin-induced hypoglycemia which might otherwise occur in normal persons (3). Insulin-dependent diabetes mellitus (IDDM) is caused by destruction of islet p-cells. We have shown that amylin is deficient in the BB/Wor rat (17), a widely accepted animal model for the human syndrome (18). Lack of amylin will produce several ef-
284
AMYLlN AND INSULIN CO-REPLACEMENTTlERAPY FOR IDDM
Fig. The complete stntc&ureof human amylin as determined by protein chemical studies (1.11). and subsequently co&m& by moleculargenetic analysis (9. 20). Both the intramolecular disulfide bridge and the cnrboxy-terminal amide group ate essential for full biological activity of amylin to modulate glucose metabolism in skeletal muscle (3, 7. 9, 11).
fects, including an increased susceptibility to insulininduced hypoglycemia (3). Superior control of blood glucose levels and protection against the hazards of insulin-induced hypoglycemia from correction of the double endocrine deficiency in IDDM, namely the codeficiency of insulin and amylin, is the basis for further research into amylin replacement as an aid in the attainment of an as yet elusive goal of the treatment of IDDM, physiologic blood glucose control (19). Interplay of amylin and insulin The following description of the interplay between
amylin and insulin in the control of fuel metabolism reflects in some measure protein chemical (1 , 11,16), molecular genetic (8, 9, 20), metabolic biochemical (2, 3, 11, 12, 13, 15), physiologic (14), and clinical studies performed over the course of the last 5 years. This brief account summarizes only those aspects of this endocrine system which relate directly to the question of IDDM and amylin replacement therapy. The linkage between over-secretion of amylin and the production of insulin resistance and non-insulin-dependent diabetes and other metabolic disorders, such as obesity, atherosclerosis and hypertension, has been reviewed recently (3, 11,21). Here, I focus on insulin hypersensitivity and the tendency to hypoglycemia from amylin lack, a condition which I believe is en-
hanced when insulin alone is replaced in patients with destroyed islet pcells. Amylin is secreted along with insulin in response to stimulation by typical nutrient secretagogues (glucose and arginine) (10). Major actions of amylin on fuel metabolism reported thus far include: suppression of insulin-stimulated uptake and incorporation of glucose into skeletal muscle glycogen (2,3, 11, 12, 13); decreasing muscle glycogen content (22): stimulation of skeletal muscle glycogenolysis by promoting conversion of glycogen phosphorylase b to a (23, 24); inhibition of muscle glycogen synthase (24); stimulation of liver glycogenolysis (15); stimulation of liver gluconeogenesis (15, 25); stimulation of hepatic glucose output, even in the presence of insulin (14); and elevation of blood lactate and glucose levels in fasted and fed rats (25). Amylin can therefore limit the ability of insulin to (i) reduce hepatic glucose production (14), and (ii) promote transfer of blood glucose into muscle glycogen. By these actions amylin will impact the ability of insulin to cause hypoglycemia. When amylin levels are relatively low, insulin tends to promote storage of glucose as glycogen in skeletal muscle and liver. Amylin does not inhibit insulinstimulated carbohydrate metabolism in adipose tissues (2). Therefore, as amylin levels increase and cause progressive insulin resistance in skeletal muscle (2, 3, 12, 13), there will be a shift in the ultimate site
MEDICALHYF’OTHESES
286 of disposal of carbohydrate. The liver will increasingly tend to become an organ of glucose and lipid synthesis and output, and insulin will then act mainly to promote conversion of carbohydrate into triglyceride for ultimate storage in adipose tissue. The major actions of amylin on glucose metabolism thus enable the body to switch the site of carbohydrate storage away from skeletal muscle glycogen towards triglyceride in adipose tissue (2,3). Amylin, in combination with insulin, serves as a signal which determines the site(s) of a carbohydrate load (3). The operation of this switching mechanism is dependent on the presence of both insulin and amylin. The emerging picture is one that amylin and insulin act together to direct the flux and sites of storage of metabolic fuels. The lack of demonstrated effects of amylin on carbohydrate metabolism in adipose tissue (2) indicates that the effect of certain combinations of insulin and amylin will switch the storage site of substrate carbohydrate into adipose tissue triglyceride. Amylin may be thought of as regulating or switching insulin-stimulated anabolic pathways between the glycogen of liver and particularly skeletal muscle, and adipose tissue triglyceride. The combined effects of amylin and insulin may represent the biochemical expression of the ‘thrifty genotype’ (26). Although the actions of amylin are similar in some respects to those of well known counter-regulatory hormones, particularly glucagon, epinephrine and norepinephrine, there are major differences. Fist, amylin secretion is stimulated by glucose (lo), the same nutrient that is the primary stimulus for insulin secretion. Furthermore, as shown in the studies published thus far, amylin and insulin secretion follow similar time courses (10). Insulin normally acts, therefore, against a backdrop of amylin action. Amylin and insulin tend to be secreted in parallel, although a degree of differential secretion, e.g. amylin secretion increasing relative to insulin secretion, has also been observed (lo), which is not inconsistent for two cooperating but distinct regulators. Thus amylin can serve as a ‘feed-forward’ controller and thereby as a ‘buffer’ or ‘back-stop’ limiting the ability of insulin to promote peripheral glucose disposal and inhibit hepatic glucose output. In contrast, the other known counter-regulatory hormones are brought into play by the occurrence of hypoglycemia (27). While these hormones respond to hypoglycemia (28) and are important in recovery (28). they do not ‘anticipate’ and prevent it. Interestingly, careful studies in normal man following induction of hypoglycemia by small physiologic doses of insulin showed that reversal of insulin-
induced suppression of hepatic glucose production preceded measurable increments in systemic levels of known counter-regulatory hormones by at least 30 min (28). These studies led to the inference that there may exist an unidentified protective physiologic mechanism capable of reversing insulin-mediated inhibition of hepatic glucose output in normal persons, which is active when small doses of insulin are given, and which prevents circulating glucose in normal persons from falling to levels at which release of known counter-regulatory hormones is triggered (28). I suggest that this process is mediated by amylin. Amylin deficiency in IDDM IDDM is caused by lymphocyte-mediated autoimmune destruction of pancreatic p-cells (29). Amylin is mainly contained in these cells, and our data (Koda J E, Cooper G J S et al, in preparation) shows that human patients with IDDM are deficient in amylin as well as insulin. As noted above, evidence now sup ports the view that amylin is primarily produced in and secreted from islet p-cells. Certain studies have demonstrated co-localization of amylin and insulin in pancreatic islet cells (6, 30). Amylin mRNA has also been localized to the islets (8). We have recently shown that pancreatic amylin mRNA levels are reduced about seven-fold when diabetes is fully estab lished in the BB/Wor rat (17), a well accepted animal model of IDDM. Amylin and insulin are also both deficient in the pancreases of rats with streptozotocininduced diabetes (10). In a recent preliminary communication, the absence of detectable amylin levels in plasma from patients with IDDM was report& (31). From a metabolic perspective, IDDM may therefore be re-defined as a syndrome resulting from the deficiency of two hormones, insulin and amylin. Thereapy of IDDM for the past 68 years has relied on insulin replacement (5), which has been one of the most important medical advances of this century. However, even after seven decades of increasing refinement of this treatment, patients with IDDM do not enjoy normal health or life expectancy (32). Ftigorous control of blood glucose improves matters, but steering a passage between the scylla of inadeqtate reduction in blood glucose levels and the charybdis of hazardous hypoglycemia remains difficult. I suggest that a major reason for this problem in IDDM is lack of amylin. Although lack of insulin is primarily responsible for the occurrence of diabetic ketoacidosis, deficiency of amylin will partly determine the response of type 1 diabetics to insulin replacement therapy. Doubtless the degree of amylin lack will vary
287
AhSYLlN AND INSULIN CO-REPLACEMENT THERAPY FOR IDDM
between patients and with progression of disease. Perhaps those with the most difficult-to-control diabetes will have the lowest or most fluctuating amylin levels. Hypoglycemiaand long-term complicationsin insulin-treatedIDDM: the scylla and charybdis of physiologic blood glucose control It is now an accepted clinical goal of diabetes therapy to maintain blood glucose levels within the physiologic range (33, 34). However, euglycemia must be achieved safely and without concomitant hypoglycemia (27, 35). Many of the chronic complications of IDDM, such as retinopathy. neuropathy and nephropathy, are thought to be caused by chronic exposure of tissues to increased blood glucose levels and the associated metabolic abnormalities (27). Unfortunately, it has remained almost impossible to maintain chronically normal blood glucose levels in type I diabetics using currently available and clinically practical methods of insulin therapy (34). This difficulty is caused in substantial measure by insulininduced hypoglycemia (27). There is a definite increase in the risk of hypoglycemia with the most carefully designed insulin regimens aimed at producing near-physiologic blood glucose control (27). The observed biological effects of amylin are consistent with the concept that deficiency of this hormone contributes to the occurrence and severity of hypoglycemia in insulin-treated IDDM. Its actions are also consistent with a role for amylin in preventing hypoglycemia in normal persons, and with amylin deficiency as a significant contributor to the metabolic disturbance in IDDM. The occurrence of insulin-induced hypoglycemia in the setting of insulin replacment therapy may not be accounted for by observed deficiencies of classical counter-regulatory mechanisms, which normally occur late in the course of established IDDM (28). However, defects in these counter-regulatory systems probably do account for differing severity of, and rates of recovery from, hypoglycemia once it actually occurs (27). Expected outcomes of amylin replacement therapy The experimentally observed effects of amylin indicate that amylin deficiency in IDDM will predispose to hypoglycemia during treatment with insulin, and support the concept that amylin replacement therapy will complement insulin therapy and lessen the risk of hypoglycemia. Glycemic control should also be improved because amylin replacment will allow administration of larger doses of insulin to reduce hyper-
glycemia. It is also believed that the combined actions of amylin and insulin to promote triglycetide formation will provide a more normal adipose tissue pattern in thin type 1 diabetics. Finally, a&bough mechanisms remain uncertain, amylin deficiency per se may contribute to some of the long-term complications of IDDM, lending further benefit to amylin replacement. We are currently performing animal studies to explore amylin and insulin co-replacement therapy for IDDM. These studies will be extended to human subjects following preparation of sufficient quantities of chemically pure and biologically active amylin (36). and after requisite safety and pharmacological studies have been completed. Acknowledgements I wish to thank all of my colleagues who have contributed to the various aspects of this work over the past 5 years, pauicularly Ken Reid, Tony Willis, Brmdan Leighton, Anne R-s, Andrew Young, Joy Koda, Laum Lehman de Gaeta. Elisabeth Albrecht, Ming-Wei Wang, and Huei-Jen Su Huang. I also thank Tim Rink for his constructive criticism of the manuscript.
References 1. Cocper G J S, Willis A C, Clark A, Turner R C, Sim R B, Reid K B M. Purification and characterixation of a pep& from amyloid-rich pancreases of type 2 diabetic patients. Proc Nam Acnd Sci USA 84: 8628, 1987. 2. Cooper G J S, Leighton B, Dimitriadis G D, Parry-Billings M, Kowalchuk J M. Howland K. Rothbard J B. Wiis A C. Reid K B M. Amylin found in amyloid deposits in human type 2 diabetes mellitus may be a hormone that regulates glycogen metabolism in skeletal muscle. Proc Nam Acad Sci USA 85: 7763, 1988. 3. Cooper G J S, Day A J, Willis A C, Roberts A N, Reid K B M, Leighton B. Amylin and the amylin gene. Structure. function and relationship to islet amyloid and to diabetes mellitus. Biochim Bioohvs Acta 1014: 247. 1989. 4. Opie E L The-relation of diabetes mellitus to lesions of the pancreas. Hyaline degeneration of the islands of Langemans. J Exp Med 5: 527. 1900-1901. 5. Banting FG, Best CH. The internal secretion of the pancreas. J lab Clin Med 7: 256. 1922. 6. Clark A, Cooper G J S, Lewis C E, Morris J F, Willis A C. Reid K B M, Turner R C. Islet amyloid formed fran diabetes-associated peptide may be pathogenic in type-2 diabetes. Lancet 2: 231, 1987. 7. Cooper G J S, Willis A C, Leighton B. Amylin hormone. Nature 340: 272, 1989. 8. l&fen J D, Newgard C B, Okamoto H, Miibum J L. Luskey K L. Rat amylin: cloning and tissue-specific expression in pancreatic islets. Proc Nam Acad Sci USA 86: 3127, 1989. 9. Roberts A N. Leighton B. Todd J A, Cockbum D, Schofield P N, Sutton R, Holt S, Boyd Y. Day A J, Foot E A, Willis A C. Reid K B M, Cooper G J S. Molecular and functional characterization of amylin. a peptide associated with type 2 diabetes mellitus. Prcc Nam Acad Sci USA 86: 9662, 1989. 10. Ogawa A. Harris V, McCorkle S K. Unger R H. luskey K L Amylin secretion from the rat pancreas and its selective loss
288 tmatment. J Clin Invest 85: 973, 1990. atIer streptoxuocin 11. Cooper G J S. ‘Ibe chamcterixatias of amylin and analysis of its role in diabetes mellitus. D Phil Thesis, University of Oxford, 1989. 12 Leightcn B, Cocper G J S. Pancreatic amylin and mlcitatin gene-related peptide cause resistance to insulin in skeletal muscle in vi&.-Nature 335: 632.1988. 13. Leiahton B. Feat E A. Cooner G J S. Kina J M. Calcita& gene-Alated peptide-l (t&P-l) is a polent mgulator of glycogen metabolism in rat skeletal muscle. FBBS (Fed Em Biochem Sot) Lett 249: 357.1989. 14. Molina J M, Cocper G J S, Leighton B. Olefsky J M. Induction of insulin resistance in vivo by amylin and calcitonin generelated peptide. Diabetes 39: 2&O.1990. 15. Ciaraldi T P. Cocuer G J S. Stolne M. In vitm effects of amylin en c&x&irate metabolism. Diabetes 39 Suppl 1: 149A. 19%. 16. Cooper G J S, Leightan B. Wiis A C. Day A J. The amyhn superfamily: a novel grouping of biologically active polypeptides related to the insulin A-chain. Prog Growth Factor Res 1: 96, 1989. 17. Huang H-JS, Cooper G J S. Young A A, Johnson M J. De.8 ciency of amylin expression in the pancreas of auto-immune BB/Wor diabetic rats. J Cell Biochem Suppl 15B: 67.1991. 18. Monies J P, Dcsemone J, Rossini A A. The BB rat. Diabetes Metab Rev 3: 725.1987. B. ‘lhe physiologic replacement of insulin. An elusive 19. Zian goal. N Engl J Med 321: 363. 1989. 20. Nisbi M, Ghan S J. Nagamatsu S, Bell G I, Steiner D F. Conservation of the sequence of islet amyloid polypcptide in five mammals is consistent with its putative role as an islet hormone. Proc Natn Acad Sci USA 86: 5738.1989. 21. Leighton B. Cooper G J S. The role of amylin in the insulin resistance of non-insulindependcnt diabetes mellitus. Trends B&hem Sci 15: 295. 1990. 22. Young D A, Deems R 0. Deacon R W, McIntosh R H, Foley J E. Fffects of amylin on glucose metabolism and glycogenolysis in vivo and in vitro. Am J Physiol259: E457, 1990. 23. Young A A, Mott D M. Stone K. Couper G J S. Amylin activates glycogen phosphorylase in the isolated soleus muscle
MEDKALHYPtYrHEsEs
of the rat. FEBS (Fed Eur Biochem Sot) Lett 281: 149,199l. 24. Deans R 0, Deaccn R W, Yamg D A. Amylin activates glycogen phosphorylase and inactivates glycogen rynthase by a CAMP-independent mechanism. Biochcsn Biophyr Res Commun 174: 716.1991. 25. Young A A, Wang M-W, Rink T J, Cooper G J S. In vivo activity of amylin. J Physiol, 1991, in press (Pmceedmgs). 26. Neel J V. Diabetes mellitus: a ‘thrifty’ genotype rendered detrimentalby ‘progress’?Am J Hum Genet 14: 353, 1962. 27. Cryer P E. Binder C. Bolli G B, Cherrington A D, Gale E AM, Gerich J E. Sherwin R S. Hvnoalvcemia in IDDM. Diabetes __ __ 38: 1193. i989. 28. Amiel S A, Tamborlane W V, Sacea L. Sherwin R S. Hypoglycemia mtd glucose ccunterregulation in normal and insulindePendem diabetic subjects. Diabetes Metab Rev 4: 7 1.1988. 29. Eisenlxuth G S. lope I diabetes mellitus. A chronic autoimmune disease. N Engl J Med 314: 1360, 1986. 30. Johnson K H. O’Brien T D. Betsholtx C, Westermark R Islet amyloid. islet-amyloid polypeptide, and diabetes mellitus. N Engl J Med 321: 513.1989. 31. Hartter E, Svoboda T. Ludvik B. Schuller M, Lell B, Kucnburg E, Brunnbauer M, Wolosxcmk W, Prager R. Basal and stimulated plasma levels of pancreatic amylin indicate its cosecretion with insulin in humans. Diabetologia 34: 52. 1991. 32. Borch-Johnsen K, Knitter S. Dcckett T. Mortality of type 1 (insulin-dependent) diabetes mellitus in Denmark: a study of relative mortality in 2930 Danish type 1 diabetic patients diaenosed from 1933 to 1972. Diabetoloaia 29: 767.1986. 33. Shafir E, Bergman M. Felig P ‘lhi Endocrine Pancreas: Diabetes Mellitus pp 1125-1129 in Endocrinology and Metabolism. 2nd ed. (Felig P, Baxter J D, Broadus A E. Frohman L A, eds) McGraw-Hill, New Yc+k, 1987. 34. Sonsken P H, Judd S L. Lowy C. Home monitoring of blood glucose. Method for improving diabetic control. Lancct 1: 729. 1978. 35. Gale E A M, Tattersall R B. Unrccognised nocturnal hypoglycemia in insulin-treated diabetics. Lancet 1: 1049. 1979. 36. Albrecht E, Young A A, Cooper G J S, Lehman de Gaeta LS. Synthesis and biological activity of human and canine amylin. Int J Peptide Prot Res 1991, submitted.
