8/12/2019 O'Brien, 2002

1/7

Re v iew

Pathogenesis of feline diabetes mellitus

T.D. OBrien *Department of Veterinary Diagnostic Medicine, College of Veterinary Medicine, Veterinary Diagnostic Laboratory, Uni v ersity of Minnesota, 1333

Gortner A v enue, St. Paul, MN 55108, USA

Abstract

The common form of spontaneous diabetes mellitus that occurs in domestic cats bears close resemblance clinically and

pathologically to human type 2 diabetes mellitus (T2DM). For example, the typical diabetic cat is obese and middle-aged, and haslow but detectable circulating insulin le v els. Howe v er, the most striking similarity is the occurrence of islet amyloidosis (IA) in nearlyall diabetic cats and in o v er 90% of humans with T2DM. IA in both humans and cats is deri v ed from islet amyloid polypeptide(IAPP, or amylin) which is a hormone produced and secreted along with insulin by the pancreatic b cells. Since all cats and humansnormally produce IAPP, additional factors must be in v oked in order to explain the de v elopment of IA. Se v eral lines of e v idencesupport the concept that IA is caused by chronically increased stimulus for b cells to secrete IAPP (and insulin). For example,peripheral insulin resistance such as in chronic obesity results in increased IAPP and insulin secretion. A recent study, in whichdiabetes mellitus was induced in cats, demonstrated that IAPP hypersecretion was induced by treatment with a sulfonylurea drugand resulted in 4/4 cats in this group de v eloping IA. In contrast, cats treated with insulin had low IAPP secretion and minimal IAde v eloped in 1/4 cats. Se v eral human-IAPP transgenic mouse models, in which there is IAPP o v erexpression, also support the notionthat prolonged high expression of IAPP leads to IA. In v itro models of IAPP o v erexpression also support this mechanism for IAformation and by demonstrating an association between IA formation and b cell toxicity, suggest a linkage between IA formationand loss of b cells in T2DM. A recent study has indicated that intermediate-sized IAPP-deri v ed amyloid fibrils can disrupt cellmembranes and therefore, may be in v ol v ed in the destruction of b cells. Striking parallels between the pathogenesis of IA and b-amyloid plaque formation in Alzheimers disease suggest possible parallel pathogenetic mechanisms of cell death and pro v idepotential a v enues for future studies into the pathogenesis of IA.# 2002 Else v ier Science Ireland Ltd. All rights reserved.

Keywords: Diabetes mellitus; Amyloidosis; Islet amyloid polypeptide

1. Summary

Diabetes mellitus in the domestic cat closely resembleshuman type 2 diabetes mellitus (T2DM) clinically andpathologically. Islet amyloidosis (IA) is an almost

inv

ariant feature of feline diabetes (FDM) and T2DM,and is associated with significant loss of b cells in thepancreatic islets. Islet amyloid in cats has been shown tobe deri v ed from islet amyloid polypeptide (IAPP), as hasislet amyloid in humans. E v idence from in v itro studieshas demonstrated that fibrillar forms of IAPP arecytotoxic and can trigger apoptosis, thus pro v iding apotential pathogenetic link between IA and b cell loss inFDM. E v idence from transgenic mouse models also

strongly supports a role for IAPP-deri v ed amyloid in thepathogenesis of FDM and T2DM. Further studies areneeded to delineate the mechanisms by which IAPP-deri v ed amyloid triggers cell death and to unequi v ocallydemonstrate these mechanisms in spontaneous FDM.

The most common form of diabetes mellitus in thedomestic cat bears close clinical and pathologicalresemblance to human type 2 diabetes mellitus(T2DM) ( Johnson et al., 1986 ). This re v iew will focuson this form of feline diabetes mellitus (FDM) and willnot consider other less common forms of diabetes in thecat such as type 1-like diabetes or secondary forms of diabetes such as may occur with pancreatitis.

Clinical similarities between FDM and T2DM includeclinical onset in middle age; FDM occurs in cats greaterthan 6 years of age with the peak incidence occurringbetween 9 and 13 years of age, which corresponds tomiddle age in the domestic cat ( Johnson et al., 1986,

* Tel.: /1-612-625-8175; fax: /1-612-624-8707E-mail address: obrie004@tc.umn.edu (T.D. OBrien).

Molecular and Cellular Endocrinology 197 (2002) 213 /219www.else v ier.com/locate/mce

0303-7207/02/$ - see front matter # 2002 Else v ier Science Ireland Ltd. All rights reserved.PII: S 0 3 0 3 - 7 2 0 7 ( 0 2 ) 0 0 2 6 5 - 4

mailto:obrie004@tc.umn.edumailto:obrie004@tc.umn.edu

8/12/2019 O'Brien, 2002

2/7

1989b; OBrien et al., 1993; Panciera et al., 1990 ). Otherclinical similarities include obesity, resistance to ketoa-cidosis, low but measurable fasting serum insulin con-centration, absent or attenuated first phase insulinsecretion, and exaggerated or absent second phaseinsulin secretion ( OBrien et al., 1985 ). All of thesefeatures of FDM closely parallel those of T2DM andtherefore suggest a common pathogenesis. Howe v er, themost striking and pro v ocati v e similarities betweenT2DM and FDM are the lesions occurring in thepancreatic islets, namely islet amyloidosis (IA) ( Fig. 1 )and partial loss of b-cells (Johnson et al., 1986, 1989b;OBrien et al., 1993 ).

Islet amyloidosis occurs in o v er 90% of cats withdiabetes and occurs in a similar proportion of humanswith T2DM ( Johnson et al., 1986, 1989b; OBrien et al.,1993 ). Furthermore, IA in FDM is associated with amean loss of b-cells of approximately 50% ( OBrien et

al., 1986 ). Sev eral similar studies of T2DM ha v e showna similar loss of up to 50% of b-cell mass ( Rahier et al.,1983; Saito et al., 1979; Westermark, 1972 ). Non-diabetic cats with IA also show a partial, but less se v ere,loss of b-cells, further supporting a link between IA andloss of b-cells (OBrien et al., 1986 ). Additional e v idencelinking IA and diabetes has also been demonstrated intype 2-like diabetes occurring in macaque species(OBrien et al., 1996; de Koning et al., 1993; Hansenand Bodkin, 1986; Howard, 1986 ). In macaques IAde v elopment has been shown to precede the onset of diabetes, and IA is significantly more extensi v e in

diabetic macaques than in age-matched non-diabeticcontrols. Yet another parallel situation is seen in humanpatients with cystic fibrosis. IA is found in approxi-mately 67% of diabetics but in only 27% of age-matchednon-diabetic CF patients ( Couce et al., 1996b ). Further-more, the IA in CF is associated with a 50% loss of b-cells. Thus, there is a strong link between IA and loss of b-cells in se v eral similar forms of diabetes mellitus.

The understanding of the pathogenesis of feline andhuman islet amyloidosis was greatly ad v anced by thedisco v ery that the precursor protein of this form of amyloid was a pre v iously unknown hormone which wenamed islet amyloid polypeptide (IAPP, also known asamylin) ( Westermark et al., 1987a,b; Cooper et al.,1987 ). The pancreatic islets are the predominant site of IAPP production and, within the pancreatic islets of all

species thus far studied, IAPP immunoreacti v ity ispredominantly located in the pancreatic b-cells, and atleast in some species, the d-cells (Johnson et al., 1988;Lukinius et al., 1989 ). Ultrastructurally, IAPP immu-noreacti v ity resides predominantly in the b-cell secretoryv esicle (Johnson et al., 1988; Lukinius et al., 1989 ). Insitu hybridization studies of the rat pancreatic islets alsoindicate that IAPP mRNA is predominantly located inthe b-cells (Leffert et al., 1989 ). The localization of IAPPimmunoreacti v ity in the b-cell secretory v esicles pro-v ided the first e v idence that IAPP was co-secreted withinsulin. As the morphologic studies predicted, IAPP andinsulin secretion are qualitati v ely and temporally simi-lar, consistent with the concept that they are co-secreted(Butler et al., 1990; Fehmann et al., 1990; Inoue et al.,1991; OBrien et al., 1991 ).

IA occurs in only a limited number of species (e.g.human beings, macaques, and cats), usually in conjunc-

tion with diabetic syndromes associated with aging, anddoes not occur in rats or mice ( Johnson et al., 1989b;OBrien et al., 1993 ). This obser v ation prompted thecomparison of IAPP structures in these species todetermine whether primary or secondary structuraldifferences within or between species might be asso-ciated with these obser v ations. Such comparisons of mammalian species re v ealed highly conser v ed regions inthe amino-terminal region (residues 1 /19) and in thecarboxy-terminal region (residues 30 /37) (Betsholtz etal., 1989a ). The inter v ening region (residues 20 /29)showed notable sequence v ariations. Secondary struc-

tural predictions of this latter region indicated apropensity for beta pleated sheet configuration forhuman IAPP whereas this was not present in mouse orrat ( Betsholtz et al., 1989b ). Since b-sheet secondarystructure is often linked to amyloid fibril formation, thisregion was further in v estigated for the formation of amyloid fibrils. Indeed, peptides corresponding to hu-man IAPP20 /29 readily formed amyloid fibrils whereasmouse and rat sequence did not ( Westermark et al.,1990 ). E v idence gathered so far indicates that nomutation in the IAPP coding region is required for thede v elopment of IA in NIDDM and in animal models(Westermark et al., 1987a,b; Sanke et al., 1988; Bet-sholtz et al., 1989b, 1990; Nakazato et al., 1990 ).Howe v er, it has recently been shown that a mutantform of human IAPP, the S20G mutant, is associated

Fig. 1. (A) Pancreatic islet from a cat showing lesions characteristic of feline diabetes mellitus including extensi v e islet amyloid deposits (arrows) andmarkedly reduced islet cell mass (arrowheads). H & E stain. (B) Pancreatic islet from a diabetic cat immunohistochemically stained for insulin. Noteextensi v e amyloid deposits which show no insulin immunoreacti v ity (arrows) and few remaining b-cells showing insulin immunoreacti v ity (arrowheads). A v idin /biotin immunohistochemistry, AEC chromogen, Meyers hematoxylin counterstain. (C) Adjacent section of pancreatic islet in B,immunohistochemically stained for IAPP. Note IAPP immunoreacti v ity of amyloid deposits (arrows) and of remaining b-cells (arrow heads).A v idin /biotin Immunohistochemistry, AEC chromogen, Meyers Hematoxylin counterstain. (D) Pancreatic islet from a normal cat demonstratingnormal cellular content and structure. H & E stain. (E) Insulin immunoreacti v ity (arrow) in a normal cat pancreatic islet demonstrating thepredominance of b -cells. A v idin /biotin immunohistochemistry, AEC chromogen, Meyers hematoxylin counterstain. (F) IAPP immunoreacti v ity(arrow) in a normal cat pancreatic islet, demonstrating the localization of IAPP in the b-cells. A v idin /biotin immunohistochemistry, AECchromogen, Meyers hematoxylin counterstain.

T.D. OBrien / Molecular and Cellular Endocrinology 197 (2002) 213 / 219214

8/12/2019 O'Brien, 2002

3/7

Fig. 1

T.D. OBrien / Molecular and Cellular Endocrinology 197 (2002) 213 / 219 215

8/12/2019 O'Brien, 2002

4/7

with early onset T2DM ( Sakagashira et al., 1996; Lee etal., 2001 ).

Simply ha v ing an IAPP sequence capable of formingamyloid fibrils is ob v iously not sufficient for IA tode v elop, because if this were the case, the pre v alence of IA in human beings, cats, and macaques wouldapproach 100% in the general population. It thereforeappears likely that abnormalities in IAPP synthesis,processing, trafficking, secretion, or degradation by b-cells must play a role in the pathogenesis of IA. Thenotion that increased IAPP secretion predisposes to IAformation is supported by studies in which increasedIAPP secretion has been associated with obesity, animportant predisposing factor for the de v elopment of T2DM. In further support of this concept, it has beenshown that the b-cells of cats with impaired glucosetolerance ha v e increased IAPP immunoreacti v ity com-pared to normal controls ( Johnson et al., 1989a; Ma et

al., 1998 ). This suggests that there is an imbalancebetween IAPP production and secretion/degradation inthese cats. The importance of the increase in IAPPcontent in cats with impaired glucose tolerance in thepathogenesis of IA is also supported by the obser v ationthat these cats also ha v e an increased incidence of IAPP-deri v ed IA as compared to normal control cats ( Johnsonet al., 1989a ).

Since increased and/or altered IAPP synthesis orsecretion may be important in the pathogenesis of IAPP-deri v ed IA, studies in v estigating the synthesisand secretion of IAPP under v arying circumstancesmay pro v ide additional clues as to the sequence of ev ents leading to IA. E v idence in support of thishypothesis was recently demonstrated in an experimen-tal cat model ( Hoenig et al., 2000 ). Partially ( /50%)pancreatectomized cats (all IA negati v e) had stablediabetes induced by treatment with dexamethasoneand growth hormone. This was followed by treatmentwith either glipizide or insulin for 18 months. Glipizide-treated cats had significantly increased basal andglucose-stimulated serum IAPP concentrations v ersusinsulin-treated cats. Furthermore, 4/4 glipizide-treatedcats de v eloped IA while only 1/4 insulin-treated cats haddetectable IA. Se v eral in v itro studies ha v e shown that

islets in culture under v arying conditions can show anincrease in the relati v e amounts of IAPP secretedrelati v e to insulin. For example, perfused pancreasfrom rats treated with dexamethasone or intraperitone-ally administered glucose (hyperglycemic) show signifi-cantly increased IAPP:insulin ratio in the perfusatewhen stimulated by 16.7 mM glucose, while pancreasesfrom fasted rats showed a significantly reduced IAP-P:insulin ratio ( OBrien et al., 1991 ). A recent study hasalso shown that when human islets in culture areexposed to 24.4 mM glucose there is an increase in thecellular IAPP:insulin ratio and a similar increase in themRNA ratios of these hormones. Under these condi-

tions the human islets also showed an increase in theamount of IAPP secreted through the constituati v epathway ( Gasa et al., 2001 ). In a mouse model of insulin resistance, a high fat diet was associated withmaintenance of IAPP mRNA le v els but decreasedexpression of insulin mRNA compared to mice on thelow fat control diet ( Mulder et al., 2000 ). These findingstaken together show that conditions present in thediabetic state (hyperglycemia, corticosteroid excess)and prediabetic state (insulin resistance, hypertriglycer-idemia) may increase the absolute amount of IAPPsynthesized and secreted. Furthermore, these conditionsmay increase the IAPP:insulin ratio. Experiments show-ing that insulin inhibits IAPP-amyloidogenesis suggest apossible role for increased IAPP:insulin expression inthe de v elopment of IA ( Charge et al., 1995; Westermarket al., 1996; Kud v a et al., 1998 ).

The associations discussed between IA formation and

the de v elopment of type 2 DM, and IA formation andthe loss of b-cells ha v e, in the past, been consideredintriguing but did not re v eal whether the association wasmerely coincidental (i.e. an epiphenomenon), or if therewas a direct cause and effect relationship. Now howe v er,sev eral additional lines of e v idence strongly support therole of IA in the de v elopment of FDM and T2DM.These include e v idence from: (1) se v eral lines of humanIAPP (hIAPP) transgenic mice; (2) in v itro experimentsshowing cytotoxic effects of fibrillar forms of hIAPP;and (3) e v idence concerning the serine to glycinemutation at position 20 of hIAPP (S20G) which isassociated with early onset T2DM.

Transgenic mouse models pro v ide strong e v idence of the importance of hIAPP fibrillogenesis in the destruc-tion of b-cells and the de v elopment of diabetes mellitus(de Koning et al., 1994; Couce et al., 1996a; Jansen etal., 1996; Verchere et al., 1996; Soeller et al., 1998;Ho ppener et al., 2000 ). For example, male, heterozy-gous, human-IAPP-transgenic mice which were madeinsulin resistant by treatment with dexamethasone andgrowth hormone de v eloped diabetes mellitus and isletamyloidosis within 6 weeks while non-transgenic micewere unaffected ( Couce et al., 1996a ). b-cells in thesehuman-IAPP-transgenic mice frequently contained ab-

normal IAPP-immunoreacti v e deposits within the cyto-plasm and in many instances had classical amyloiddeposits within their cytoplasm ( Couce et al., 1996a ). Itwas also noted that affected b-cells often showedultrastructural degenerati v e changes and alterationsconsistent with apoptosis. Furthermore, male homozy-gous hIAPP-transgenic mice de v eloped spontaneousdiabetes mellitus with selecti v e and extensi v e b-cell lossby 8 weeks of age without any induction of insulinresistance ( Jansen et al., 1996 ). Female hIAPP-trans-genic mice of this strain also de v eloped diabetes, isletamyloidosis, and b-cells loss at a more ad v anced age. Inanother unrelated strain of mice, transgenic for hIAPP,

T.D. OBrien / Molecular and Cellular Endocrinology 197 (2002) 213 / 219216

8/12/2019 O'Brien, 2002

5/7

a close association between the de v elopment of hyper-glycemia and islet amyloidosis was found ( Verchere etal., 1996 ). Recently, it was shown that insulin-resistantmale agouti mice, which were heterozygously transgenicfor hIAPP, spontaneously de v eloped diabetes and isletamyloidosis ( Soeller et al., 1998 ). In each of thesetransgenic models, the production of high le v els of hIAPP by the b-cells was associated with the formationof abnormal IAPP aggregates, selecti v e b-cell loss, andthe de v elopment of diabetes mellitus. Thus, micetransgenic for hIAPP de v elop lesions that closelyresemble the islet lesions in humans with T2DM, whilemice transgenic for non-amyloidogenic rodent IAPP donot de v elop either IA or loss of b -cells.

Data supporting the concept that IAPP-deri v edamyloid fibrils are cytotoxic and associated with apop-totic cell death and/or necrosis are becoming increas-ingly abundant. Studies using hIAPP transfected COS-1

cells which expressed hIAPP at high le v els demonstratedthe formation of intracellular IAPP-deri v ed amyloidfibrils which was associated with cellular degenerationand death by 96 h ( OBrien et al., 1995 ). In contrast,COS-1 cells transfected with non-amyloidogenic ratIAPP also showed high expression of IAPP but showedno ad v erse effects. Subsequent experiments with thissystem ha v e demonstrated that there is a significantincrease in apoptosis of COS-1 cells expressing hIAPPbut no increase in apoptosis in cells expressing rodentIAPP ( Hiddinga and Eberhardt, 1999 ). Human IAPP-deri v ed fibrils ha v e also been shown to be toxic toisolated islets in culture ( Lorenzo et al., 1994 ). Themechanism of cell death in this study in v ol v ed synthesisof RNA and protein, plasma membrane blebbing,chromatin condensation and DNA fragmentation, con-sistent with apoptotic cell death. Lastly, there ha v e beensev eral studies demonstrating toxic effects of hIAPP-deri v ed fibrils on neurons and PC12 cells (rat pheochro-mocytoma cell line) ( May et al., 1993; Mattson andGoodman, 1995; Dore et al., 1997 ). In these systemshIAPP demonstrates cytotoxic effects that are identicalto those of A b (the fibrillogenic protein that formsamyloid deposits in Alzheimers disease) whereas, ro-dent IAPP is non-toxic ( May et al., 1993 ). For both

IAPP and A b, it is the fibrillar forms that are cytotoxicwhile monomers and related non-amyloidogenic poly-peptides are not cytotoxic ( Howlett et al., 1995; Lorenzoand Yankner, 1994, 1996; Schubert et al., 1995 ). Recentdata indicates that it is fibrillar assemblies in the rangeof 50000 /200 000 Da that are cytotoxic and disruptcellular membranes ( Janson et al., 1999 ). Interestingly,the toxicity of both hIAPP and A b fibrils can becounteracted in cultured neurons by IGF-1 ( Dore etal., 1997 ) and Congo red ( Burge v in et al., 1994 ), whilerifampicin and its analogues inhibit the toxicity of fibrillar IAPP and A b on PC12 cells ( Mattson andGoodman, 1995 ). These similarities in cytotoxicity by

IAPP and A b fibrils support the concept that similarmechanisms are in v ol v ed in neuronal cell death inAlzheimers disease and b-cell death in T2DM.

Gi v en the similarities of IAPP and A b fibril cytotoxi-city, the mechanisms of A b toxicity in neurons are of particular interest, in that they may illuminate cytotoxicmechanisms in v ol v ed in b-cell damage by IAPP fibrils.Mechanisms in v ol v ed in A b cytotoxicity include oxida-ti v e damage by reacti v e oxygen species, lipid peroxida-tion, reduced mitochondrial transmembrane potential,and destabilization of intracellular calcium homeostasis(Hensley et al., 1994; Mark et al., 1997a,b ). Membranelipid peroxidation initiated by A b induces apoptosis inPC12 cells and cultured hippocampal neurons. Thisprocess is mediated by 4-hydroxynonenal and is pre-v ented by Bcl-2 and antioxidants ( Kruman et al., 1997 ).The neuroprotecti v e actions of cycloheximide againstAb induction of apoptosis is mediated by increased

expression of the bcl-2 gene product ( Furukawa et al.,1997 ). Ab toxicity on hippocampal neurons is alsopre v ented by EUK-8 a synthetic catalytic free radicalsca v enger ( Bruce et al., 1996 ). Of special interest is thefinding that membrane lipid peroxidation initiated byAb is also associated with impaired glucose transportinto cultured rat hippocampal neurons ( Mark et al.,1997b ). If similar alterations are induced in b-cells byIAPP fibrillogenesis, this may be of critical importancein impairing normal b-cell function.

The recent disco v ery of the S20G hIAPP has pro v idedyet another line of e v idence supporting a role of IA in

the de v elopment and progression of T2DM ( Sakaga-shira et al., 1996 ). Patients identified with this mutationhad relati v ely early onset of diabetes ( 5 /35 years of age),relati v ely sev ere diabetes, and a strong family history of T2DM. A role for IAPP amyloidogenesis in T2DMwould therefore be supported if the S20G mutant ismore amyloidogenic than the more common allele of hIAPP. Indeed, COS-1 cells transfected with the S20GhIAPP showed significantly more apoptosis 96 h aftertransfection v ersus the common hIAPP allele ( Sakaga-shira et al., 2000 ). It was shown in an in v itrofibrillogenesis assay that the S20G mutant formed

approximately twofold more amyloid and at a rateapproximately 3-fold higher than the common hIAPPallele. Thus, these experiments pro v ide e v idence for alinkage between increased amyloidogenicity of hIAPPand early onset of T2DM.

In summary, the e v idence for an important role forIAPP-deri v ed IA in the pathogenesis of FDM andT2DM is increasing. Howe v er, a great deal of work stillneeds to done to elucidate the mechanisms underlyingthe initiation and toxicity of IAPP fibrillogenesis, andthe apoptotic pathways in v ol v ed in b-cell death.Furthermore, these mechanisms need to be unequi v o-cally demonstrated to be operational in the feline and

T.D. OBrien / Molecular and Cellular Endocrinology 197 (2002) 213 / 219 217

8/12/2019 O'Brien, 2002

6/7

8/12/2019 O'Brien, 2002

7/7

Lorenzo, A., Yankner, B.A., 1994. Beta-amyloid neurotoxicity re-quires bril formation and is inhibited by Congo red. Proc. Natl.Acad. Sci. USA 91, 12243 /12247.

Lorenzo, A., Yankner, B.A., 1996. Amyloid bril toxicity in Alzhei-mers disease and diabetes. Ann. N. Y. Acad. Sci. 777, 89 /95.

Lorenzo, A., Razzaboni, R., Weir, G.C., Yankner, B.A., 1994.Pancreatic islet cell toxicity of amylin associated with type-2

diabetes mellitus. Nature 368, 756 /760.Lukinius, A., Wilander, E., Westermark, G.T., Engstrom, U., Wester-

mark, P., 1989. Co-localization of islet amyloid polypeptide andinsulin in the B cell secretory granules of the human pancreaticislets. Diabetologia 32, 240 /244.

Ma, Z., Westermark, G.T., Johnson, K.H., OBrien, T.D., Wester-mark, P., 1998. Quantitati v e immunohistochemical analysis of isletamyloid polypeptide (IAPP) in normal, impaired glucose tolerant,and diabetic cats. Amyloid: Int. J. Exp. Clin. In v est. 5, 255 /261.

Mark, R.J., Keller, J.N., Kruman, I., Mattson, M.P., 1997. Basic FGFattenuates amyloid beta-peptide-induced oxidati v e stress, mito-chondrial dysfunction, and impairment of Na //K /-ATPaseacti v ity in hippocampal neurons. Brain Res. 756, 205 /214.

Mark, R.J., Pang, Z., Geddes, J.W., Uchida, K., Mattson, M.P., 1997.Amyloid beta-peptide impairs glucose transport in hippocampaland cortical neurons: in v ol v ement of membrane lipid peroxidation.J. Neurosci. 17, 1046 /1054.

Mattson, M.P., Goodman, Y., 1995. Different amyloidogenic peptidesshare a similar mechanism of neurotoxicity in v ol v ing reacti v eoxygen species and calcium. Brain Res. 676, 219 /224.

May, P.C., Boggs, L.N., Fuson, K.S., 1993. Neurotoxicity of humanamylin in rat primary hippocampal cultures: similarity to Alzhei-mers disease amyloid- b neurotoxicity. J. Neurochem. 61, 2330 /2333.

Mulder, H., Martensson, H., Sundler, F., Ahren, B., 2000. Differentialchanges in islet amyloid polypeptide (amylin) and insulin mRNAexpression after high-fat diet-induced insulin resistance in C57BL/6J mice. Metab. Clin. Exp. 49, 1518 /1522.

Nakazato, M., Asai, J., Miyazato, M., Matsukura, S., Kangawa, K.,

Matsuo, H., 1990. Isolation and identication of islet amyloidpolypeptide in normal human pancreas. Regul. Peptides 31, 179 /186.

OBrien, T.D., Hayden, D.W., Johnson, K.H., Ste v ens, J.B., 1985.High dose intra v enous glucose tolerance test and serum insulin andglucagon le v els in diabetic and non-diabetic cats: relationships toinsular amyloidosis. Vet. Pathol. 22, 250 /261.

OBrien, T.D., Hayden, D.W., Johnson, K.H., Fletcher, T.F., 1986.Immunohistochemical morphometry of pancreatic endocrine cellsin diabetic, normoglycaemic glucose-intolerant and normal cats. J.Comp. Pathol. 96, 357 /359.

OBrien, T.D., Westermark, P., Johnson, K.H., 1991. Islet amyloidpolypeptide (IAPP) and insulin secretion from isolated perfusedpancreas of fed, fasted, glucose-treated and dexamethasone-treatedrats. Diabetes 40, 1701 /1706.

OBrien, T.D., Butler, P.C., Westermark, P., Johnson, K.H., 1993.Islet amyloid polypeptide: a re v iew of its biology and potentialroles in the pathogenesis of diabetes mellitus. Vet. Pathol. 30, 317 /332.

OBrien, T.D., Butler, P.C., Kreutter, D.K., Kane, L.A., Eberhardt,N.L., 1995. Intracellular amyloid associated with cytotoxicity in

COS-1 cells expressing human islet amyloid polypeptide. Am. J.Pathol. 147, 609 /616.

OBrien, T.D., Wagner, J.D., Litwak, K.N., Carlson, C.S., Cefalu,W.T., Jordan, K., Johnson, K.H., Butler, P.C., 1996. Islet amyloidand islet amyloid polypeptide in Cynomolgus macaques ( Macaca fascicularis ): an animal model of human non-insulin-dependentdiabetes mellitus. Vet. Pathol. 33, 479 /485.

Panciera, D.L., Thomas, C.B., Eicker, S.W., Atkins, C.E., 1990.Epizootiologic patterns of diabetes mellitus in cats: 333 cases(1980 /1986). J. Am. Vet. Med. Assoc. 197, 1504 /1508.

Rahier, J., Goebbels, R.M., Henquin, J.C., 1983. Cellular compositionof the human diabetic pancreas. Diabetologia 24, 366 /371.

Saito, K., Yaginuma, N., Takahashi, T., 1979. Differential v olumetryof A, B, and D cells in the pancreatic islets of diabetic and non-diabetic subjects. Tohoku J. Exp. Med. 129, 273 /283.

Sakagashira, S., Sanke, T., Hanabusa, T., Shimomura, H., Ohagi, S.,Kumagaye, K.Y., Nakajima, K., Nanjo, K., 1996. Missensemutation of amylin gene (S20G) in Japanese NIDDM patients.Diabetes 45, 1279 /1281.

Sakagashira, S., Hiddinga, H.J., Tateishi, K., Sanke, T., Hanabusa, T.,Nanjo, K., Eberhardt, N.L., 2000. S20G mutant amylin exhibitsincreased in v itro amyloidogenicity and increased intracellular

cytotoxicity compared to wild-type amylin. Am. J. Pathol. 157,2101 /2109.

Sanke, T., Bell, G.I., Sample, C., Rubenstein, A.H., Steiner, D.F.,1988. An islet amyloid peptide is deri v ed from and 89-amino acidprecursor by proteolytic processing. J. Biol. Chem. 263, 17243 /17246.

Schubert, D., Behl, C., Lesley, R., Brack, A., Dargusch, R., Sagara,Y., Kimura, H., 1995. Amyloid peptides are toxic v ia a commonoxidati v e mechanism. Proc. Natl. Acad. Sci. USA 92, 1989 /1993.

Soeller, W.C., Janson, J., Hart, S.E., Parker, J.C., Carty, M.D.,Ste v enson, R.W., Kreutter, D.K., Butler, P.C., 1998. Islet amyloid-associated diabetes in obese A v y /a mice expressing human isletamyloid polypeptide. Diabetes 47, 743 /750.

Verchere, C.B., DAlessio, D.A., Palmiter, R.D., Weir, G.C., Bonner-Weir, S., Baskin, D.G., Kahn, S.E., 1996. Islet amyloid formation

associated with hyperglycemia in transgenic mice with pancreaticbeta cell expression of human islet amyloid polypeptide. Proc. Natl.Acad. Sci. USA 93, 3492 /3496.

Westermark, P., 1972. Quantitati v e studies of amyloid in the islets of Langerhans. Ups. J. Med. Sci. 77, 91 /94.

Westermark, P., Wernstedt, C., OBrien, T.D., Hayden, D.W.,Johnson, K.H., 1987. Islet amyloid in type 2 human diabetesmellitus and adult diabetic cats is composed of a no v el putati v epolypeptide hormone. Am. J. Pathol. 127, 414 /417.

Westermark, P., Wernstedt, C., Wilander, E., Hayden, D.W., OBrien,T.D., Johnson, K.H., 1987. Amyloid brils in human insulinomaand islets of Langerhans of the diabetic cat are deri v ed from aneuropeptide-like protein also present in normal islet cells. Proc.Natl. Acad. Sci. USA 84, 3881 /3885.

Westermark, P., Engstro m, U., Johnson, K.H., Westermark, G.T.,

Betsholtz, C., 1990. Islet amyloid polypeptide: pinpointing aminoacid residues linked to amyloid bril formation. Proc. Natl. Acad.Sci. USA 87, 5036 /5040.

Westermark, P., Li, Z.C., Westermark, G.T., Leckstrom, A., Steiner,D.F., 1996. Effects of b-cell granule components on human isletamyloid polypeptide bril formation. FEBS Lett. 379, 203 /206.

T.D. OBrien / Molecular and Cellular Endocrinology 197 (2002) 213 / 219 219