Lack of Effect of Islet Amyloid Polypeptide on Hepatic Glucose Output in the In Situ-Perfused Rat Liver Susumu Nishimura, Tokio Sanke, Kazuo Machida, Hiroto Bessho, Tadashi Hanabusa, Kazuhiko Nakai, and Kishio Nanjo

Islet amyloid polypeptide

(IAPP), a novel peptide isolated from islet amyloid deposits in patients with insulinoma and non-insulin-dependent diabetes mellitus (NIDDM), has been reported to be cosecreted with insulin from pancreatic p cells and to inhibit glucose uptake and glycogen synthesis in muscle tissue in vitro. We investigated the effects of the synthesized, rat-amidated form of IAPP on hepatic glucose output, and IAPP extraction, using an in situ flow-through perfusion system in rats to elucidate the actions of IAPP on the liver. The IAPP (IO-’ mol/L) alone had no effects on the hepatic glucose release. Infusion of 6 x 10-l’ mol/L glucagon alone resulted in an expected elevation in glucose production (30.0 f 1.7 pm01135 min/g liver). Insulin (3 x IO-” mol/L) submaximally decreased the glucagon-stimulated glucose production to 73% (from 30.0 f 1.7 to 22.0 + 1.4 pmol/35 min/g liver; n = 7, P < .Ol). A simultaneous infusion of IO-* mol/L IAPP did not influence the inhibition of glucagonglucagon-stimulated glucose production (27.6 & 1.2 Cmol/35 min/g liver) or the insulin-dependent stimulated glucose production (22.6 f 1.3 pm01135 min/g liver). IAPP extraction by the liver in a single passage was minimal, in contrast to awroximatelv 50% heoatic insulin extraction. These results indicate that IAPP does not play any important role in .. modulating glycogen metabolism in the liver. Copyright 0 1992 by W. E. Saunders Company

I

SLET AMYLOID poiypeptide (IAPP or amylin), a 37-amino acid peptide, has recently been identified from islet amyloid deposits in patients with insulinoma’ and non-insulin-dependent diabetes mellitus (NIDDM).’ Recent immunohistochemical studies3*4 and characterizations of IAPP cDNA5z6 suggest that IAPP may be a normal islet hormone and is colocalized with insulin within the p-cell secretory granules. Some in vitro7-9 and in vivo9-‘* studies have shown that IAPP has insulin antagonistic effects, particularly in muscle tissue. These findings suggest that IAPP may be involved in the pathogenesis of NIDDM. However, another study has shown that exogenously infused human IAPP has no effects on the plasma levels of glucose, insulin, and glucagon following intravenous glu-

cose injection in normal subjects.13 Recently, some in vitro’4.‘5 and in vivo’6-‘8studies have shown that IAPP is cosecreted with insulin from pancreatic p cells into the portal vein via glucose and nonglucose stimuli. This study was therefore designed to investigate the direct metabolic effects of IAPP on hepatic glycogen metabolism, using perfused rat liver.

MATERIALS

AND METHODS

Animals Male Wistar rats (180 to 230 g) were maintained in a temperaturecontrolled and air-conditioned room, under a light-dark cycle, fed Oriental Laboratory chow (Oriental Yeast, Tokyo, Japan), and given water ad libitum.

electrophoresis, and amino acid analysis. Other chemicals used were of reagent grade and were from commercial sources. Liver Perjksion The start of each experiment was sometime between 9:00 AM and 10:00 AM, to assume sufficient reserve of glycogen in the liver. After intraperitoneal anesthesia with pentobarbital (50 mgikg), the experiments were performed in a 37°C chamber. The perfusion of the liver was performed by a modified technique described by Mortimore.19 The liver was perfused in situ with an oxygenated Krebs-Ringer bicarbonate buffer (pH 7.40) containing 2% BSA and 25% (vol/vol) washed human erythrocytes in a noncirculating system. The flow rate was kept at 8 mL/min, and hepatic venous samples were collected every minute. After a 20-minute equilibration, peptides were infused into the portal vein for the indicated periods. Extraction of Rat IAPP in the Perfusate The perfusates for IAPP assay were collected into chilled tubes containing 500 Kalliginogen Inactivator Units Aprotinin. IAPP in the perfusate was extracted using Sep-Pak Cl8 cartridges (Waters, Milford, MA) by the method described previous1y.‘8,mThe eluates from the Sep-Pak cartridges were evaporated and resolved by the radioimmunoassay (RIA) buffer (0.1 mol/L sodium phosphate buffer, pH 7.4, containing 0.1% Triton X, 0.05 mol/L NaCl, 0.1% BSA, and 0.01% NaN,) and were subjected to IAPP RIA.‘* Ten samples were extracted simultaneously, using Syringe Infusion Pump 22 (Model 2400-006, Harvard Apparatus, South Natick, MA), and the recovery (89.5% * 2.8%, n = 3) was assessed for each extraction. The serial dilution curve of the extract was parallel to a standard curve.

Materials Bovine serum albumin (BSA, fraction V) was purchased from Armour Pharmaceutical, Chicago, IL. Human insulin and porcine glucagon were obtained from Novo Industry, Copenhagen, Denmark. Carboxy-terminus, rat-amidated IAF’P assembled by a solidphase procedure using an Applied Biosystems model 430A peptide synthesizer was kindly supplied by Dr K. Nakajima (Protein Research Foundation Peptide Institute, Osaka, Japan). The homogeneity of the synthesized rat IAPP was confirmed by reversephase high-pressure liquid chromatography (HPLC), capillary Metabolism, Vol41, No 4 (April),

1992: pp 431-434

From the First Department of Medicine, Wakayama University of Medical Science, Wakqama, Japan. Supported by a Grant-in-Aid for Scientific Research (No. 01480294) from the Minis&y of Education, Science, and Culture, Japan. Address reprint requests to Tokio Sanke, MD, First Department of Medicine, Wakayama University of Medical Science, 27 Nanabancho, Wakayama 640, Japan. Comrigh t 0 1992 by W.B. Saunders Company 0026-0495/92/4104-0015$03.00/0 431

NISHIMURA ET AL

Fig 1.

Time

of

Ferfusion

(min)

Analytical Procedure Glucose in the effluents was measured by a glucose oxidase system with a Glucose Analyzer (Yellow Springs Instruments, Yellow Springs, OH). Insulin was determined by RIA, using Phadeseph Insulin RIA Kits (Pharmacia Diagnostics, Uppsala, Sweden). IAPP was measured by an RIA, using Human Amylin RIA Kits (Peninsula Laboratory, Belmont, CA). The synthesized, rat-amidated form of IAPP was used as an RIA standard, instead of the kit standard. The detection limit of the assay was 2 fmolkube. StatisticalAnalysis

The data were presented as the mean k SEM. The statistical analysis was performed with the Mann-Whitney U test, and a P value less than .05 was considered significant. RESULTS

Efiects of Glucagon or lAPP on Hepatic Glucose Output

After a 20-minute equilibration, the rate of glucose output from perfused rat liver was stabilized at a level of 0.56 ~fr0.06 Fmol/min/g liver (n = 5), and it continued for an experimental period in the absence of glucose in the perfusate (Fig l[A]). When glucagon (6 x 10-l’ mol/L) alone was infused for 5 minutes into the portal vein, an expected increase in hepatic glucose output was observed (Table 1 and Fig 1A). The glucose release under this Table 1. Effect of Rat IAPP-NH, on Glucose Output From Perfused Rat Liver

Concentration Condition

(1) Saline (control) (2) Rat IAPP-NH,

Change in Glucose Output

(nmol/L) 10.0

(ZZApmol

minCza55,g liver-‘)

0 2 0.1 0 + 0.1

(3) Glucagon

0.06

30.021.7

(4) Glucagon

0.06

22.0 k 1.4

Insulin (5) Glucagon Rat IAPP-NH, (6) Glucagon Insulin Rat IAPP-NH,

0.3 0.06

1

0.06

t*

NS NS

27.6 r 1.2 * 22.6 + 1.3 I

:

condition represents mainly the glycogenolytic activity of glucagon, because fed-rat liver was used and no substrates affecting the glucose metabolism were included in the perfusate.” On the other hand, a lo-minute infusion of IAPP (1 x 10e8mol/L) alone showed no significant changes in hepatic glucose output (Table 1 and Fig 1B). Effects of ZAPP on Glucagon-Induced Hepatic Glucose output

To investigate the interaction of IAPP and glucagon, effects of IAPP on the hepatic glucose output induced by glucagon (6 x lo-” mol/L) were studied. When IAPP ( 10e8 mol/L) was infused for 5 minutes before and during the glucagon infusion (total 10 minutes), it had no influence on the glucagon-induced glucose output (Table 1 and Fig 2A). The same results were obtained using lo-’ to approximately lo-‘* mol/L IAPP. Effects of iAPP on Insulin-Dependent Inhibition of Glucagon-Induced Hepatic Glucose Output

To elucidate the influences of IAPP on hepatic insulin actions, the effects of IAPP on the insulin-dependent inhibition of glucagon-induced hepatic glucose output were studied. When insulin (3 x lo-” mol/L) was infused for 5 minutes before and during the glucagon (6 x lo-” mol/L) infusion (total 10 minutes), the hepatic glucose production stimulated by glucagon was suppressed by 27% (Table 1 and Fig 2B). The simultaneous infusion of IAPP (1 x 10m8 mol/L) did not influence the insulin-dependent inhibition of glucagon-induced hepatic glucose production (Table 1 and Fig 2B). The same results were obtained using 10e9 to approximately lo-” mol/L IAPP. Hepatic Extraction of Insulin and k4PP

.

t

10.0

Effects of glucagon or the

synthesized, rat-amidated form of IAPP (IAPP-NHJ on hepatic glucose outputfrom perfused rat liver. Livers were perfused in a flowthrough system with perfusion media as described in the Methods. Twenty minutes after the beginning of the perfusion, 6 x lo-” mol/L glucagon(A) or lo-’ mol/L rat IAPP-NH, (6) was infused into the pottel vein. Values are shown as means + SEM of seven independent experiments.

.

0.3 10.0

To estimate the extraction of insulin and IAPP in a single transhepatic passage, the insulin and IAPP levels in the effluents were measured. When insulin (5 x lo-” mol/L) and IAPP (5 x 10-l’ mol/L) were infused simultaneously for 20 minutes into the portal vein, IAPP extraction by the liver was minimal, in contrast to approximately 50% hepatic insulin extraction (Fig 3).

NOTE. Hormones were infused into the portal vein as indicated in Figs 1 and 2. Glucose output was calculated by integrating the area

DISCUSSION

under the curve above basal for glucose output from the beginning of hormone infusion to minute 55. Significance of the difference : lP < .05; tP < .Ol; NS, not significant.

Although definite physiological actions of IAPP have not yet been elucidated, some in vitro”’ and in vivo9-” studies

433

LACK OF EFFECT OF IAPP ON GLYCOGENOLYSIS

Fig 2. Effects of rat IAPP-NH, on the gfucagon-stimulated glucase output (A) and on the insulindependent inhibitiin of glucagonstimulated glucose output(B). Livers were perfused as described in Fig 1. (A) Rat IAPP-NH, (lo-’ mol/L) was added 5 minutes beforethe glucagon infusion. (B) Insulin alone (3 X 1O-‘0 mol/L) or insulin (3 x 1O-‘0 mol/L) and rat IAPPNH, (lo-’ mol/L) were added 5 minutes before the glucagon infu sion. Values are shown as means + SEM of seven experiments, respectively.

3 u 2

17 s OE ‘2 2

M*SEM

A 2

2

1

I-

P

I

I

30

40

Time

suggest that IAPP might cause insulin resistance, particularly in the muscle tissue, which is one of the characteristics of NIDDM.“.” Since IAPP is cosecreted with insulin from pancreatic l3 cells into the portal vein,‘4s’sit is important to investigate the direct metabolic effects of IAPP on the liver. Our results showed that synthesized, rat-amidated IAPP alone has no influence on the hepatic glucose output, and that it does not affect the glucagon-induced hepatic glucose release or the insulin-dependent inhibition of glucagoninduced hepatic glucose release. Molina et al” and Koopmans et al” observed that human or rat IAPP caused hepatic insulin resistance in vivo in anesthetized or conInsulin

5X10-‘“M

o-_o Insulin M

IAPP-NH2 5 X lo-“M

IAPP-NH2 Mean f

SEM

++m n= 3

I_

I-

t-

I-

I-

0

1

5

I

10 Time

I

50

I

15

1

I

20

25

(min>

Fig 3. Fractional hepatic extraction of insulin and rat IAPP-NH,, after simultaneous Infusion of those peptldes. Insulin (5 x lo-” mol/L) and rat IAPP-NH~ (5 x lo-” mol/L) ware infused simubaneously for 20 mlmrtes into the portel vein. The percent of hepetic extraction was calculated by the difference irconoentration of those peptides in perfusate and effluent. Values are shown as means 2 SEM of three experiments.

T c fh

60

20

of Perfusion

I

1

I

30

40

50

l

60

(min)

scious rats. The reasons for the discrepancy between their in vivo results and our in vitro data are unclear. One possible explanation is that as we focused on the direct effects of IAPP on hepatic glycogenolysis in vitro, our data did not exclude an effect of IAPP on hepatic gluconeogenesis by supply of glucogenic metabolites due to peripheral insulin resistance in vivo. Another possible explanation is that the effect of IAPP in vivo may be mediated by unknown factors, such as catecholamine-induced hepatic insulin resistance. Although Molina et al” excluded the influence of catecholamine on IAPP-induced hepatic insulin resistance, we have evidence that the impaired glucose tolerance induced by IAPP infusion in vivo in rats can be partially improved in the presence of adrenergic blockade (data not shown). Recently, Young et al9 reported that the increase in blood glucose levels during IAPP infusion in Is-hour fasted rats could not have been due to an increase in hepatic glucose production, from their results that the overall rate of glucose appearance did not change during IAPP infusion, although the metabolic clearance rate of glucose decreased. Stephens et al” also showed that calcitonin gene-related peptide (CGRP)/IAPP receptors occurred only on the nonparenchymal cells of rat liver and did not directly regulate liver glucose metabolism. These findings are consistent with our present data. The negative data of the IAPP action on hepatic glycogenolysis obtained in the present experiments are not due to the peptide used, which is biologically inactive, because the synthesized IAPP (lo-’ mol/L or more) was confirmed, using an in situ rat lower-limb perfusion system, to have an inhibitory effect on insulin-induced glucose uptake in muscle tissue, which is widely accepted as the action of IAPP (data not shown); however, the molecular weight of the peptide was not checked by mass spectrometry. In addition, we have found that IAPP (lo-’ mol/L or more) stimulated cyclic adenosine monophosphate (CAMP) production without increasing glucose production in the same liver perfusion system, but 10m9mol/L or less IAPP did not stimulate CAMP production (data not shown). The discrepant data between CAMP and glucose release from the perfused liver are consistent with the in vitro results reported by Stephens et al.” Estimation of the hepatic extraction of IAPP has not been reported. It has been established that approximately

434

NISHIMURA

40% to 60% of the basal secretion of insulin is extracted in a single transhepatic passage.XX26In contrast, C-peptide, which is secreted in equimolar quantities with insulin, but has no biological effects on the liver, has been reported not to be extracted by the liver.27,28Our recent study’* has clarified that the peripheral IAPP concentration in normal subjects and in patients with NIDDM is approximately lo-” to lo-l2 mol/L, which is consistent with other reports,‘6Z’7.2qand that IAPP concentration represents 3% to 14% of insulin, on a molar basis. Thus, the experimental condition used in the extraction study should be physiological. In contrast to approximately 50% hepatic insulin extraction, a physiological dose of IAPP was minimally extracted by the liver. Our

finding of reduced

hepatic

the actual proportion

extraction

of IAPP suggests that

of IAPP to insulin released

might be much less than the IAPP/insulin peripheral IAPP

does

plasma,

and also supports

not play any important

glycogen metabolism In summary, insulin-modulated

by l3 cells

molar ratio in

our conclusion role

that

in modulating

in vitro in the liver.

IAPP

had no influence

hepatic

glycogenolysis

liver, and IAPP was minimally results indicate

ET AL

extracted

on the glucagonl in perfused

that IAPP has no impact on hepatic

sensitivity

and does

regulation

of hepatic glycogenolysis.

rat

by the liver. These

not play any important

insulin

role in the

REFERENCES 1. Westermark P, Wernstedt C, Wilander E, et al: Atnyloid fibrils in human insulinoma and islets of Langerhans of the diabetic cat are derived from a neuropeptide-like protein also present in normal islet cells. Proc Nat1 Acad Sci USA 84:3881-3885,1987 2. Cooper GJS, Willis AC, Clark A, et al: Purification and characterization of a peptide from amyloid-rich pancreases of type 2 diabetic patients. Proc Nat1 Acad Sci USA 84:8628-8632,1987 3. Johnson KH, O’Brien TD, Hayden DW, et al: Immunolocalization of islet amyloid polypeptide (IAPP) in pancreatic beta cells by means of peroxidase-antiperoxidase (PAP) and protein A-gold techniques. Am J Path01 130:1-8,1988 4. Lukinius A, Wilander E, Westermark GT, et al: Colocalization of islet amyloid polypeptide and insulin in the B cell secretory granules of the human pancreatic islets. Diabetologia 32:240-244,1989 5. Sanke T, Bell GI, Sample C, et al: An islet amyloid peptide is derived from an 89-amino acid precursor by proteolytic processing. J Biol Chem 263:17243-17246,1988 6. Leffert JD, Newgard CB, Okamoto H, et al: Rat amylin: Cloning and tissue-specific expression in pancreatic islets. Proc Nat1 Acad Sci USA 86:3127-3130,1989 7. Cooper GJS, Leighton B, Dimitriadis GD, et al: Amylin found in amyloid deposits in human type 2 diabetes melhtus may be a hormone that regulates glycogen metabolism in skeletal muscle. Proc Nat1 Acad Sci USA 85:7763-7766,1988 8. Leighton B, Cooper GJS: Pancreatic amylin and calcitonin gene-related peptide cause resistance to insulin in skeletal muscle in vitro. Nature 335:632-6351988 9. Young DA, Deems RO, Deacon RW, et al: Effects of amylin on glucose metabolism and glycogenolysis in vivo and in vitro. Am J Physiol259:E457-E461,1990 10. Sowa R, Sanke T, Hirayama J, et al: Islet amyloid polypeptide amide causes peripheral insulin resistance in vivo in dogs. Diabetologia 33:118-120,199O 11. Molina JM, Cooper GJS, Leighton B, et al: Induction of insulin resistance in vivo by amylin and calcitonin gene-related peptide. Diabetes 39:260-265,199O 12. Koopmans SJ, van Mansfeld ADM, Jansz HS, et al: Amylininduced in vivo insulin resistance in conscious rats: The liver is more sensitive to amylin than peripheral tissues. Diabetologia 34:218-224, 1991 13. Bretherton-Watt D, Gilbey SG, Ghatei MA, et al: Failure to establish islet amyloid polypeptide (amylin) as a circulating beta cell inhibiting hormone in man. Diabetologia 33:115-117,199O 14. Ogawa A, Harris V, McCorkle SK, et al: Amylin secretion from the rat pancreas and its selective loss after streptozotocin treatment. J Clin Invest 85:973-976, 1990

15. Kahn SE, D’Alessio DA, Schwartz MW, et al: Evidence of co-secretion of islet amyloid polypeptide and insulin by S-cells. Diabetes 39:634-638,199O 16. Mitsukawa T, Takemura T, Asai J, et al: Islet amyloid polypeptide response to glucose, insulin, and somatostatin analogue administration, Diabetes 39:639-642,199O 17. Butler PC, Chou J, Carter WB, et al: Effects of meal ingestion on plasma amylin concentration in NIDDM and nondiabetic humans. Diabetes 39:752-756,199O 18. Sanke T, Hanabusa T, Nakano Y, et al: Plasma islet amyloid polypeptide (amylin) levels and their responses to oral glucose in Type 2 (non-insulin-dependent) diabetic patients. Diabetologia 34:129-132, 1991 19. Mortimore GE: Effect of insulin on release of glucose and urea by isolated rat liver. Am J Physiol204:699-704,1963 20. Sanke T, Hanabusa T, Nakano Y, et al: Plasma islet amyloid polypeptide (IAPP/Amylin) concentrations and their responses during OGTT in patients with NIDDM and obesity. Excerpta Medica (in press) 21. Friedmann N, Rasmussen H: Calcium, manganese and hepatic gluconeogenesis. Biochim Biophys Acta 222:41-52,197O 22. Olefsky JM: Pathogenesis of non-insulin dependent (type II) diabetes, in DeGroot LJ, Cahill GH, et al (eds): Endocrinology (ed 2). Philadelphia, PA, Saunders, pp 1369-1388, 1989 23. DeFronzo RA: Lilly Lecture 1987: The triumvirate: p-cell, muscle, liver: A collusion responsible for NIDDM. Diabetes 371667-687, 1988 24. Stephens TW, Heath WF, Hermeling RN: Presence of liver CGRPiAmylin receptors in only nonparenchymal cells and absence of direct regulation of rat liver glucose metabolism by CGRPiAmylin. Diabetes 40:395-400,199l 25. Rojdmark S, Bloom G, Chou MCY, et al: Hepatic extraction of exogenous insulin and glucagon in the dog. Endocrinology 102:806-813,1978 26. Harding P, Bloom G, Field JB: Effect of infusion of insulin into portal vein on hepatic extraction of insulin in anesthetized dogs. Am J Physiol228:1580-1588,1975 27. Kitabchi AE: The biological and immunological properties of pork and beef insulin, proinsulin and connecting peptides. J Clin Invest 49:979-987,197O 28. Hagen C, Faber OK, Binder C, et al: Lack of metabolic effect of c-peptide in normal subjects and juvenile diabetic patients. Acta Endocrinol85:29,1977 (suppl209) 29. Hartter E, Svoboda T, Ludvik B, et al: Basal and stimulated plasma levels of pancreatic amylin indicate its co-secretion with insulin in humans. Diabetologia 34:52-54,199l