DIABETES/METABOIJSMRESEARCH AND REVIEWS Diabetu Metab ResRev 2OOL; 17:363-373. DOI: 10.1002/ dmn.

225

REVIEW ARTICLE

Molecular mechanism resistance of insulin in mellitus:roleof the insulin type 2 diabetes receptor variant forms
Giorgio Sestil* Massimo Federici2 Davide Lauro2 Paolo Sbraccia2 Renato Lauro2
7llniv er sity of Catanzar o-Magnq Graecia" Italy 2Laboratory of Moleatlar Medicine, Departtnent of Internal Medicine, University of Rome'Tor Vergata', Rome, Italy *Correspondenceto: Giorgio Sesti, c/o Diparrimento di Medicina lnterna, Universit di Roma Tor Vergata', Via Tor Vergata 135, 00133 Rome, Italy. E-mail : sesti@uniroma2.it

Summary
Type 2 diabetes is a heterogeneousand polygenic disorder resulting from interaction of genetic factors with environmental influences. Numerous candidate genes for insulin signaling proteins have been screened,but no single major zusceptibility genefor type 2 diabeteshas been identified. Due to its pivotal role in insulin action, the insulin receptor was considered a plausiblecandidate gene.The insulin receptor existsin two isoforms differing (8x11 -) or presence (Exl1*) of a 12 amino acid sequence by the absence in the COOH-terminusof the c-zubunit, as a consequence of alternative splicing of exon 11. The Exll- binds insulin with two-fold higher affinity than the Ex1l* . This difference is paralleled by a decreasedsensitivity for metabolic actions of insulin. Some,but not all, studieshave reported that expressionof the low-affiniry Exlf * is increased in target tissues from t5pe 2 diabetic patients, thus suggesting that alterations in abundance of the nn'o isoforms might contribute to insulin resistance.Insulin and type 1 IGF receptors have been shown to form hybrid receptorsin tissuesco-expressing both molecules. Hybrid receptorsbind IGF-I,but not insulin, with high affiniry and behaveas IGF-I holoreceptors, rather than insulin receptors, in terms of receptor autophosphorylation, ild horrnone lnlsmalization. It has been shown that the abundanceof hybrid receptorsis increasedin skeletalmuscle and adipose tissue from tlpe 2 diabetic patients, and is negatively correlated with in vrvo insulin sensitivity. Mutations in the insulin receptor genehave been identified in snrdieswhidr examinedan appropriately sized population of patients with tynr- 2 diabetes. The prevalenceof mutations in the insulin receptor gene ranged from O.4o/r7.8o/o.This review will focus on the structural and functional heterogeneity of the insulin receptor, and will discuss the pathogenetic role of insulin receptor variant forms and polymorphisms in the development of the corrmon form of We 2 diabetes.Copyright CI 2001 John Wiley & Sons, Ltd. Keywords We2 diabetes;inzulin receptor isoforms; IGF-I receptor; hrid receptor; insulin resistance;geneticsof tSpe 2 diabetes

lntroduction
T1pe 2 diabetes is a complex disease characterized by a combination of impaired insulin action, increased hepatic glucose production, and insulin secretorydefects [1]. Longirudinat studies suggestthat abnormalities in both insulin action and insulin secretion occur early in the pathogenesis of type 2 diabetes[2]. Genetic factorscontribute to ttre pathogenesis of type 2 diabetes, given the high concordance rate between monozygotic twins, the high prevalencein certain ethnic groups,and the increasedprevalencein offspring

Receive 17 January 2OO7 Rcvise 13 Apnl2001 Arrepte 23 April 2001

C-opynght O 2001 John Wiley & Sons, Ltd.

. of affected patients [2]. At a molecular level, multiple defectsin insulin signaling have beenshown to contribute to peripheral insulin resistance, including decreasesin insulin receptor number or affinity, altered insulin receptor kinase activity, decreasesin phosphorylation of intracellular substrates, and abnormalities in glucose transporter translocation and activation [1]. In contrast, the regulation,of insulin secretionfrom pancreaticp-cells has yet to be completely clarified, as a consequence of its extreme complexity, ild the molecular mechanism(s) underlying the relative dysfunction of p-cells in tlpe 2 diabetes is still undefined. Understanding the geneticsof type 2 diabetes is complicated by the fact that multiple genes are likely to be involved in phenotypically different subgroups of patients with rype 2 diabetes,ild also in a glven patient with the disease. Accordingly, considerable effort has been devoted to identifying genes that contribute to this genetic predisposition.In one approach to this question, several groups of investigators have screenedfor the presenceof mutations in candidategenes whose products have important functions in the biochemical pathways regulating insutin secretion or insulin action. Indeed, molecular scanning of candidate genes has resulted in the elucidation of the genetic basisof rare monogenic forms of diabetes with defined modes of inheritance. For instance, mutations in the insulin receptor gene have been identified in rare syndromesof exfieme insulin resistance such as kprectraunism and Rabson-Mendenhall [3]. Primary ddects in pancreatic p-cellshave been demonstratedin maruriry onsetdiabetes of the young (MODY), a rare forrr of diabetes causedby mutations in genes encoding glucokinase [4J, hepatocyte nuclear factor (HNF) l-alpha [5], and HNF zfalpha [6], HNF l-beta l7l, or IPFI t8l. In spite of intense investigation, the genetic basis of the common form of type 2 diabetes has not yet been identified. lnzuIin exerts its multiple effecs on cell growth and metabolism by binding to its cell surface transmembrane receptor that belongs to the large family of growth factor receptors with intrinsic tyrosine kinase activity. Insulin binding to the insulin receptor extracellular c-zubunit stimulates autophosphorylation on multiple tyrosine residuesof the cytoplasmicportion of the transmembrane p-subunits. This results in activation of its intrinsic tyrosine kinase, which catalyzes the phosphorylation of a variety of intracellular docking proteins induding insulin receptor substrate (RS) proteins [9,10]. Once phosphorylated, IRS proteins bind and activate severalSrc homolory 2 (SH2) domain proteins including the p85 regulatory subunit of phosphatidylinositol 3-kinase (PI 3-kinase), the ryrosine kinasesFyn and Csh the tyrosine protein phosphataseSFIP-2,and several smaller adapter molecules such as the growth factor receptor-binding protein Grb-2, Crk, and Nck [9,t01. The activation of ttrese SH2 domain proteins initiates signaling cascades, leading to the activation of multiple downstream effectors whidr ultimately rransmit the insulin signal to a branching series of intracellular pathways that regulate cell metabolism, sunrival, growth, and differentiation. Recent
Copyright @) 2001 John Wiley & Sons, Ltd.

G.Sestief oL studies have shown that some componentsof the insulin signaling pathway play a role in the growth and secretory function of pancreatic p-cells [11-13]. Mice with p-cellspecific disruption of the insulin receptor gene show impaired insulin secretion in responseto glucose, and a progressive glucose intolerance [13]. Ttrese data have raised the intriguing possibility ttrat molecular defeas in insulin signaling may contribute to both peripheral insulin resistance and impaired insulin secretion, the t! /o key components in the pathogenesis of tSpe 2 diabetes.In the light of its key role, the insulin receptor has been considereda plausible candidategene for type 2 diabetes. This review will focus on the strucnrral and functional heterogeneity of the insulin receptor, and will then discuss the pathogenetic role of insulin receptor variant forms and polymorphisms in the development of the common form of type 2 diabetes.

Insulinreceptor structure
The insulin receptor gene is located on the short arm of chromosome 19 and contains 22 exons and 21 introns (Figure 1) [14-16]. The mature insulin receptor is a heterotetrameric glycoprotein composed of two usubunits and two membrane-spanningp-subunits linked by disulfide bonds. The insulin receptor is synthesizedas a single high molecular weight proreceptor (Mr - 180 000) that s proteolytically cleaved at a tetrabasic amino acid sequence(Arg-Lys-Arg-Arg)located at the junction of the a- and p-subunitsto yield a-p monomers.The a-p linkage is ensured by disulfide bonds connecting Cys-647 in the a-subunit with Cys-872 in the p-subunit. Two a-p monomers are connected together by an intersubunit disulfide bond bennreen a Cys-524 residue and the corresponding residue in the adjacent a-subunit. In addition, one of the cysteines 682, 683 or 685 was suggested to form an a-a, disulfide bridge resulting in the mature heterotetrameric a2p2 configuration. The r-subunit is entirely extracellular and is responsible for high affinity insulin binding. The N-terminal (exons 1-2) and the cysteine-rich domains (exons 3-5) are required for high affiniry binding in combination with a COOHterminal domain (residues 70+719) [17]. Furthennore, the presence of a 12 amino acid sequenceencoded by exon 11 contributes to the modulation of insulin binding affinity, as discussed below. The unoccupied a-subunit inhibits the intrinsic tyrosine kinase activity of the p-subunit and may be viewed as a regulatory subunit of the catalytic intracellular subunit. Removal of the a-subunits by proteolytic deavage or by deletion mutagenesis releasesthis inhibitory effect and activates the intrnsic kinase.Thus, in the absenceof ligand, the insulin receptor a-subunit maintains a structural constraint on the consttutively active p-subunit kinase. The p-subunit of the insulin receptor is composedof a short extracellular domain, a transmembrane domain, and a cytoplasmic domain, which possesses the intrinsic tyrosine kinase activity. Distinct functional regions have been defined in
DiabetesMetab ResRev 2001; 77:363-373.

lR in Type z Diabetes t2 13 L4 ls 16 17 lE 19 20 21 22

365

-s-s-

Insulin Binding Domain Cysteine-Rich

- s - s-

Alternatelv SplicedExonll

oomainl Transmembrane

+
\

Juxtamembrane Domain ATP-Binding Site

-)> /'

B
Figure 1. Schematic illustration

\ \

IR-Ex 11-

IR-Ex 11+

of the human insulin receptor gene and the rwo insulin recePtor protein isoforrrs

the intracellular portion of the p-subunit including the ATP binding domain (Gly-X-Gly-X-X-Gly) and three autophosphorylation sites located in the intracellular juxtamembrane region (Tyr-960), in the middle of the intracellular domain (Tyr-1158-X-X-X-Tyr-1162-Tyr1163), and at ttre COOH-terminus(Tyr-1328 and Tyrat site 960 in the juxtamembrane 1334). Phosphorylation region creates an NPXpX-recognitionmotif for the PTB domain of the IRS proteins, and is also important for receptor internalization. Autophosphorylation of the r.wosineresiduesin the l2OOfYYof the kinase regulatory domain is essential for insulin-stimulated kinase activity and biological actions. In addition, the activated kinase regulatory loop binding (KRLB) domain is required for interaction between a unique region of IRS-2 comprising amino acids 591-786 and the insulin receptor [18].

isoforms Insulinreceptor
Cloning of the insulin receptor cDNA revealed the exisence of nrro receptor isoforms which differ by the (HIR-Bor Exll*) (HIR-Aor Ex11-) or Presence absence of the a-subunit of 12 anino acidsat the COOH-terminus (Figure l) [4,15]. Analysisof the exon-intron organizauon of the HIR gene revealed that this 12 amino acid seSmentis encodedby a discrete36 basepair exon (exon are I l) F6l. Thus,two insulin receptormRNAtranscripts of alternativesplicingof exon generatd as a consequence involvedin the regulation I l. The molecularmechanisms process are not understood.It the alternative splicing of has been recently reported that the intron 10 contains ttlro
':-ryr:ght .Q 2001 Joha Wiley & Sons, Ltd.

that modulatealternativesplicingof exon 11: a sequences 48 nudeotide purine-rich sequenceat the S'-end that functions as a splicing enhancerleading to an increasein exon 11 indusion, and an inhibitory 43 nudeotide sequenceat the 3'-end upstream from the branch point that favorsskipping of exon I 1 [ 19] . In addition, sequence to haveboth positiveand negative the exon 11 itself appears regulatory effectson alternative splicing t191. There is evidencesuggestingthat the two HIR isoforms are immunologicallyand functionallydistinct l2o-27l.It has been suggestedthat the COOH-terminal domain of to residuesT05-73I the Exll+ a-subunitcorresponding is a major autoantigenic site [20]. Consistent with these observations, it has been shown that an IgG fraction, isolated from a patient affected by autoimmune hypoglycemia due to anti-insulin receptor autoantibodies, inhibited insulin binding to cells transfected with and expressing the Ex1l- from without affecting insulin binding to cells transfected with and expressing the Exl1+ form l2:.l. An antipeptide antibody directed against the NH2-terminal region of the sequenceencoded by exon 11 was able to inhibit insulin binding to the Exl1+ form without affecting the binding to the Ex1lform [22]. Using transfectedmammalian cells expressing separatelythe nvo receptor isoforms, it has been shown that the binding affinity for insulin of ttre b(11- form is approximately mro-fold higher than that of the Exll+ form 123,25,27f.This differencein ligand binding affinity is paralleled by changesin sensitivity for bottr metabolic and mitogenic actions of insulin [26]. Furthermore,the Ex11- form appearsto have higher rates of internaliz.asome LT7l,but not all [26], tion and reryding 124,251.In
Diabeta Metab ResRev 2001.; 17:363-373-

,66 transfected cell lines expressing separately the two receptor forms, studieshave detectedpreferential insulinform. More induced down-regulation of the Exllrecently, it has been demonstrated that the Exll- form is a physiological receptor for the insulin-like growth factor II (IGF-II) [28]. Using fibroblastsderived from mice with targeted disruption of the tlpe 1 insulin-like growth factor I (IGF-D receptor and expressingseparatelythe nryo receptor isoforms, it has been shown that IGF-II interacts with the Exl1- form, but not with the Ex1l+ form, with relatively high affinity (4o/oof ttrat of insulin). The atrinity of IGF-II for the Exll- form was similar to that of IGF-II binding to the type I IGF receptor. ln addition, IGF-II predominantly elicits mitogenic,rather than metabolic effects, which are paralleled by differences in the activation of IRS-I and Shc [28]. Relative expressionof the wvo insulin recePtor nRNA transcripts is regulated in a tissue- and developmentspecific manner [28-31]. Hematopoietic and neuronal form, tissues such as cells express only the Exl l placenta, kidney, adipose tissue, and skeletal muscle express both isoforms whereas liver predominandy expressesthe Ex1l+ form 29-311. Furthermore, the Exll- form is predominantly expressedin human fetal tissues induding kidney, skeletal musde, liver, and fibroblasts as compared with adult tissues [28]. HIR isoform expression is also modulated by differentiation. In colon cancer, breast carcinomas, and cultured breast cancercell lines, the Exll- is the predominantreceptor form expressed, thus raising the possibility that IGF-II may modulate cancer cell growth in an autocrine/ paracrine manner f28,321. In vivo t33l and in vtro [34] studies have shown that hormona-l and metabolic factors can regulate the alternative splicing of the insulin receptor mRNA. Thus, in skeletal muscle, increased expression of the low-affinity Exl1+ form has been positively correlated with both high glucose levels and high fasting insulin concentrations [35]. Taken together, encoded these data suggestthat the amino acid sequence by exon 11 plays a role in modulating the high affiniry binding of insulin to the receptor by acting in concertwith other receptor regions to form the three-dimensional structure of the insulin binding site. Furthermore, it is possible that the presence of exon 11 may have a regulatory role for other receptor functions.

G.Sestief ol. would be expected to respond only to its own ligand under some under physiological conditions. Nevertheless, insulin can function as a growth factor and circumstances IGF-I can have metabolic effects similar to those of insulin. Thus, the specificity of action of IGF-I and insulin must depend on the binding affinity and tissueexpression of their correspondingreceptors. Sincethe two receptors are closely homologous, and the position of many rysteine residues thought to be involved in a-a intersubunit disulfide bonds are also conserved,the possibility *45 leised that insuliVIGF-I hybrid receptors composed of an insulin receptor cp-heterodimer and an IGF-I cp-heterodimer could be formed in tissuesco-expressing both molecules(Figure2). Hybrid receptors were first identified by studies demonsuating high affinity IGF-I binding immunoreactive with insulin receptor-specificantibodies [37]. The shown by formation of hybrid receptorswas subsequently using several approachesinduding immunoprecipitation with rype I IGF receptor-specificantibody and microsequencing [38] or expression of recombinant human insulin or type 1 IGF receptorsin transfectedrodent cells and characterization of their reactivity with species-and receptor-spcific antibodies [39]. Assembly of insuliV IGF-I hybrid receptors was also demonstrated in vrno under defined ligand incubation conditions [4O]. It has beenshownthat hybridreceptorsbind IGF-Iwithan affinity similar to that seen with t)"e 1 IGF receptors,but bind insulin with lower affiniry than dassic insulin receptors f4I,421. Furthermore, hybrid receptorsbehaveas an IGF-I receptor rather than a t)"ical insulin receptor or some intermediate version of the two receptors in terms of receptor autophosphorylation, ild hormone inlsmalization and degradation [43]. The distribution of hybrid receptors varies among human tissues. Analysis of the distribution of insuliVIGF-I hybrid receptors has revealed that hybrid receptorsrepresenta high proportion of to 550/o in total type 1 IGF receptors varying from 360/o human tissues zuch as placenta, skeletal muscle, fat, erythrocytes, leukocytes, and fibroblasts [44]. Accordingly, a siguificant fraction of total insulin receptor was assembledas hybrids in the human tissues,varying from 37o/orn placenta to 42o/oin skeletal musde, with the

Insulin/lcF-l hybridreceptors
The insulin and tlpe 1 insulin-like growth factor I (IGF-I) receptors share considerablehomology of structure and function [36]. Although the trro receptorsappearsimilar in signaling potential, it is well recognized that insulin functions primarily in the acute regulation of glucoseand lipid metabolismsin muscle, liver, and adipose tissue, whereasIGF-I functions in long-term control of growth in srudieshave shown that the most tissues.Cross-reaction the insulin receptor have low IGF receptor and rype 1 affinity for the homologus hormone so that eachreceptor
copyright o 2001 John wiley & sons,Ltd.

INSULIN RECEPTOR

IIYBRID RECEPTOR

TYPE T IGF RECEPTOR

Figure 2. Schematic illustration IGF receptor, and insulinzlGF-I

of the insulin receptor, tyf) hrid receptor

Diabeta Metab Res Rev 2OOl; a7: 363-373.

lRin Typez Diabetes exception of fat tissue where the fraction of insulin receptorsforming hybrids was 77o/o 1441. Factors regulating hyrid receptors formation in vivo are not completely defined. Studieswith transfectedcells both insulin and type 1 IGF receptors [43] co-expressing and with mammalian tissues [45] have suggestedt]rat hybrid receptors are formed by a process of random assembly in which hybrids and dassical receptors are formed with equal efficiency and in proportions determined by the molar ratio of the insulin and t5pe 1 IGF receptor content. Since both insulin and IGF-I regulates the expression of their own receptors, it is conceivable that changes in insulin receptor and,/or type 1 tGF receptor content induced by their respective hormones would result in modifications in hybrid receptor abundance. It has been demonstrated that in skeletal musde from subjects with insulinoma, a state of primary nongeneticallydetermined hlperinsulinemia, hybrid recePtor abundancewas increased, and was inversely correlated wittr insulin receptor number and positively correlated with increased plasma insulin levels [46]. In another study, an increased abundance of hybrid recePtor was found in skeletal musde of obese individuals who were characterized by increased fasting insulin levels and plasma IGF-I concentrations [47]. The Propordecreased in tion of hybrid receptorswas correlatedwith a decrease insulin receptor number and an increase in tyPe 1 IGF receptor number. These observations strongly suggest that downregulation of insulin receptorsor upregulation of changesin of type 1 IGF receptors, as a consequence may lead to the levels, and IGF-I circulating insulin formation of a higher proportion of hybrid receptors. A more recent study in skeletal musde from partially pancreatectomizeddiabetic rats has shown that hlperglycemia causesan increasein hybrid receptor formation through a posttranslational medranism mediated by the glucosamine pathway [48]. It is possible t]rat increased activation of glucosanine biosynthetic pathway induced by the high concentrations of glucose may alter the glycosylation pattern of insulin and type 1 IGF receptors. It is known that insulin and tyPe 1 IGF receptor glycosylation precedesthe formation of intra- and intersubunit disulfide bonds in the endoplasmic reticulum (ER). Since it is likely that hybrids are assembledduring bioqgrthesis in the ER where insulin and tlpe 1 IGF receptor precursors become disulfide linked to one another rather than to a homologous precursor, it is possible that hyperglycemia, via the glucosamine pathway, may facilitate hybrid receptor formation by affecting the spatial or temporal separation during the synthesis of the two receptor homodimers. Overall, these srudiesindicate that both hormonal and metabolic factors can regulate insuliVlGF-l hybrid receptor formation
tn l'I]'o.

367 by IGF-I rather than insulin under physiological conditions. This hlpothesis was tested by analyzing expression and function of hybrid receptors in 3T3-L1 preadipocyte cell model t491. During differentiation into adipocytes, expression of the insulin receptor increases whereas expression of type 1 IGF receptor remains relatively constant, thus leading to formation of a high proportion of hybrid receptors. In fully differentiated 3T3-L1 adipocytes, it was shown ttrat virnrally all type 1 IGF receptors were assembledas hybrids, and, more importantly, IGF-I was as potent as insulin in stimulating glucose uptake, thus indicating that hybrid receptorscan mediate IGF-I biological actions [49]. Overall, these observations raise the possibility that tissue-specific differences in expressionof hybrid receptorsmay contribute to regulate tissue sensitiviry to insulin and IGF-I.

variantformsand lnsulin receptor mutationsin type z diabetes

Insulin receptor isoforrns in tyrpe 2 diabetes
The finding that the two insulin receptor isoforms have distinct functional characteristicshave led to the hypothesis ttrat tissue-specific alterations in the relative abundanceof the two isoforms may play a role in causing insutin resistance,a typical feature of obesity and type 2 diabetes. Studies on the relative expression of the two mRNA transcripts have led to divergent observations [50-55]. Thus, three studies have reported that expression of ttre two mRNA speciesmeasured by polymerase chain reaction (PCR) is altered in skeletal musde of both type 2 diabetic and pre-diabetic insulin-resistantsubjects as comparedto normal insulin-sensitivecontrols [50-52]. These observations, however, were not confirmed by three other studies[53-55]. Discordantresultshavebeen aslo reported when the relative abundance of the nnro receptor protein isoformswas analyzedby immunological In two studies,the relative distribution of the nro assays. receptor protein isoforms was measured by using an c autoantiserum that differentially inhibits isoform-specifi insulin binding to the two receptor protein isoforms of [56,57]. In thesestudies,ttrey showedthat expression the trnroisoforms is altered in isolated adipocytesand in skeletal muscle membranes from subjects with type 2 diabetes 156,571.In a different study, a specific dual peptide-based radioimmunoassay (RLA), which quntitated the two human insulin receptor isoforms independently of their binding properties, ws used to measure abundanceof the two isoforms [35]. This study demonstrated that expression of the low-affinity Ex11+ form was significantly increasedin fat and skeletalmusclefrom obeseand qrpe 2 diabetic subjectsas comparedwith nonobese controls [35]. In addition, increasedexpressionof the Exll+ form was significantly correlated with both body mass index (BMI) and fasting glucose levels, and
DiabetesMetab Ra Rev 2001; 17:363-373.

An imponant issue that remains to be addressed is rshether mitogenic and metabolic effects of IGF-I can be equdll' signaled via h1rid receptors. Because hybrid receptors bind IGF-I, but not insulin, with high affinity, it s ukell'that h1rid receptors may be primarily activated
C--r.:g-: C 2S: j, -i<r & Sons. Ild

*t tended to be correlatedwith high fasting insulin levels.In another snrdy, relative abundance of the two isoforms was determined by using an immunoprecipitation assay based on the relative ability of an exon 11* specific monoclonal antibody to immunoprecipitate receptors labeled *i,6 rzsl-insutin [53]. This study failed to show a significant difference in the relative proportion of the two receptor protein isoforms in the skeletal muscle of normal, obese non-diabetic, and obese type 2 diabetic subjects, alttrough a slightly higher mean percentageof the low-affinity Exl1+ receptor was observed in obese and type 2 diabetic subjecs compared with normal controls. It is possible that the lack of statistical significancein the relative abundanceof the two receptor isoforms observedin this latter study might be due to the small number of subjectsexamined. In addition, because l2sl-insulin was usedas tracer in the immunoprecipitation assay, it is possible that this srudy underestimated the acnral abundance of the low-affinity Exll* isoform as a consequence of the different insulin binding affiniry betweenEx11- and Exll* isoforms(Exll- >Exll+). Other possible reasons for the discrepancies among studies include differencesin PCR protocols, contamination of musde specimensby connectivetiszueand blood, site of tissue biopsies,conuol groups, or, perhapsmore importantly, differences in metabolic stants of examined that subjects.In this respect, it has been demonsrrated expressionof the low-affinity Exl1+ form was significantly increased in skeletal muscle of patients with insulinoma when compared with control zubjects. Furthermore, this increase was positively correlated with plasma insulin concentration and negatively correlated with in vivo insulin sensitiviry [58]. The resuls of this study may provide a key to the interpretation of the discordantdata reportedby the different studies[5G55]. In these latter reports, the metabolic state of the zubjects studied ranged from glucoseintolerance to overt diabetes and since this progression is usually paralleled by the dedine in plasma insulin levels whi accomp:rny increases in glycemia, the differences in relative abundance of the two insulin receptor isoforms bennreen normoglycemic and diabetic subjects may have been minimal or absent due to similar plasma insulin concentrauons.

G. Sestief ot

correlated with a decrease in both insulin binding affiniry and rn vrvo insulin sensitivity measured by an insulin tolerance test. Similar results have been reported in adipose tissue from patients with ttTpe2 diabetes [60]. Four studies have addressedthe question of whether the obserwedalterations in hybrid abundance are primary defects specifically associatedwith type 2 diabetes or represent abnormalities associatedwith other cormon states of insulin resistance. In one study, it was demonstrated ttrat abundance of insuliVIGF-I hytrid receptors was increased in ttre placenta of insulinresistant women with gestational hypertension [61]. In a secondstudy, hybrid receptor abundancewas found to be increasedin the skeletalmuscle of obesesubjects,and was correlated to increased BMI and decreasedrn vivo insulin sensitivity t471. In another study, abundance of hybrid receptors was quantitated in the skeletal muscle of patients with insulinoma t461. It was found that hybrid receptor abundancewas significantly increasedin patients with insulinoma as compared with control subjects, and was correlated to both increased plasma insulin levels and reduced n vivo insulin-mediated glucose uptake measured by the euglycemic hyperinsulinemic clamp technique t461. Fina[y, in a more recent study, the abundance of hybrid receptors was measured in skeletal muscle from lean, glucose-tolerant,non-obese subjecs with a different degreeof insulin sensitivity, ild from overweight non-diabetic subjects 621. It was observed that hybrid receptor abundance did not differ benveenthe more sensitiveand the lesssensitivesubjects, although the hybrid receptor,/insulin receptor ratio was slighdy increasedin the latter group of subjects [62]. Overall, these findings are supporrive of the hypothesis that dterations in the proportion of insuliVIGF-I hybrid receptors might contribute to impair insulin action in insulin resistantsubjecs by sequestering insulin receptors in a less responsive form, and raise the possibility that increasedexpressionof hybrid receptorsmight represent a generalized defect associatedwith different states of insulin resistance. Finally, because expression of both the low-affinity Exll* form and insuliVIGF-I h)rid receptorshas been found to be increased in the same conditions of insulin one might speculate that the Ex1l+ form has resistance, a higher ability to form hybrids as compared to the Insulin/IGF-I hybrid receptors in t5pe 2 Ex11- form. Although a study with transfected cells overexpressingthe nryo receptor isoforms separatelyhas diabetes shown no difference in their ability to form hybrids with As describedabove,insulin/IGF-I hybrid receptorsbehave endogenous rodent type 1 IGF receptors f27f , further as rype 1 IGF receptors rather than typical insulin studies are needed to elucidate this issue. receptors with respect to various receptor functionsThus, an increasedproportion of hybrid receptorswould Insulin receptor polymorphisms and be expected to reduce insulin sensitivity in target tissues mutations of insulin action contributing to cellular insulin resistance.Quantitation of hybrid receptorsusing a microwell- The human insulin receptor has been cloned indepenbased immunoassay revealed that the proportion of dently by nuo groups of researchers[14,15]. Besides hybrids was significantly higher in skeletal muscle from silent polymorphisms that do not alter amino acid patients with rype 2 diabetes as compared with normal sequence,other polymorphisms have been described subjects [59]. The proportion of hybrid receptorswas that alter the predicted amino acid sequence(Table 1).
Copyright O 2001 John Wiley & Sons, Ltd. DiabetesMetab Ru Rev 2001: 77:363-373.

lR in Type z Diabetes

369

polymorphisms of the humaninsulinreceP' with insulin-resistant Pima lndians, a racial group with Table1. Unconfirmed tor that alter predictedamino acid sequence'
Codon Ullrichet aL (1985) Tvr (tAC) lle (ATC) Gln (CAG)

Ebina et a/.(1985)
His (CAC) Thr (ACC) Lys (MG) val (GTC) Ser (TCC) Lvs (MG)

1M 421 465 873 874 125',1

tup (GAc) Thr(ACC) tun (AAC)

studies in addition underlined have been identified inother acids Theamino eta/.[15]. of Ullrich etal.l14l and Ebina reports to theoriginal Sequencingof the insulin receptor cDNAsfrom two Pima Indians [63] or analysis by allele-specific amplification of genomic DNA from an ltalian population including have revealed that three codons 66 subjects tg) repofted by Ullrich et aI. correspondedto the sequence to the sequence reportedby and three conesponded [14], Ebina et al. [15]. These data indicate that these polymorphisms represent a cloning artifact or a sequencing error. Another insulin receptor variant causing a Val-' Met change at codon 985 has been described in both glucose-tolerant zubjects and type 2 diabetic patients t651. Transfection snrdies in mammalian cells have shown that the expressedMet%s receptor bound insulin with normal affinity, internalizecl norrrally when bound to insulin, had normal insulin-stimulated kinase activiry, activated normally the IRS-1,rPI3-kinase pathway, and elicited normal insulin-stimulated metablic and mitogenic effects t661. Thus, it is likely that the Mets insulin receptor is a relatively common poymorphism that does not causecellular insulin resistance. A few studies have addressedthe quetion of whether mutations in the insulin receptor gene may conm-buteto the etiology of the common form of rlpe 2 diabetes (Figure 3) t3l. The first investigationswere carried out
INST]LIN RECEPTOR DOMAINS Insulin Binding Domain $eineRftlt Domain Alternately Spliced Eionll \

the highest documented prevalence of type 2 diabetes in the world t2l. In one srudy, CDNA from an insulinresistant Pima lndian was doned, and both alleles of the insulin receptor gene were found to encode a normal protein sequencel6n. h the second study, cDNA from nro Pima tndians with rype 2 diabetes was amplified by PCR and sequenced [63]. The nudeotide sequencein bottr subjectswas identical to the nonnal sequence[63]. In another study, the sequenceof exons 1G22 encoding the tyrosine kinase domain of the receptor was analyzed from the DNA of six patients with type 2 diabetes using PCR amplification and direct sequencing t681. None of the diabetic subjects showed mutations in the exons examined. In other studies, molecular scanning techniques have been applied for screening population of rype 2 diabetic patients for mutations in the insulin receptor gene. ln one study, analysis of single-stranded conformational polyrrorphisms (SSCP) was utilized to screenexons 77-27 of the insulin receptor in 30 unrelated Caucasian patiens \rith t)" 2 diabetes t651. Two patients were found to have variant arrino acid sequences in the tyrosine kinase domain. One of these was a heterozygousmissensemutation causing a Lys--+Glulo6a changeat cdon 108. This non-conservative substitution is located within the ryrosine kinasedomain of the insulin receptor thus it is likely to have functional consequences, although these have not been investigatd. The second parient was heterozygousfor the Metessinsulin receptor variant. ln the same snrdy, the Metess variant was also detected n 1/73 control zubjecs. In a snrdy of a Welsh population, the Metessinsulin receptor variant was found n a8O control subjects but in O/L6O type 2 diabetic zubjects,while the Glu168 variant was not found either in control or diabetic subjects [69]. Anot]rer snrdy carried LIST OF MUTATIONS

- s-s
f[1)[l3arr

+J

Transmembrarc Domain Juxtamembrane Domain ATP-Binding Ste

aF

ValMeP8s

+1-

Lys)Gluroee Arg)Glnue+
TyrClsBs

FfJe tlcF

!. f.t-ic tE C :lr: .5

ibrin

of

human insulin

neoeptor with map of strucnrral

domains

and mutations

in the insulin

-ft3

rrlr

tc.

i..d

Diafutes Metoh ResRev 2001; 17: 363-373.

t70

G.Sestiet o[

out in a Danish population failed to show any significant target tissues, and ttrat alterations in ttre relative association between the Metess variant and t'?e 2 abundance of the two isofonns, especially in skeletal diabetes [70]. In contrast, in a Dutch population recruited muscle and fat, might contribute to the development of in Rotterdam, the Metess variant was found to be insulin resistance in type 2 diabetes. The pathophysioas compared logicat role of hybrid receptors is still undear. Because increasedin type 2 diabetic subjects(5.60/o) (1.3olo) In another snrdy hybrid receptors display functional properties similar to with control subjects [71]. carried out in the Netherlands whidr induded a popula- those of tSpe 1 IGF receptors,it is a plausible hlpothesis tion recruited in Hoorn, the prevalence of the Metess that in insulin-resistant states such as type 2 diabetes, an variant was similar in tlpe 2 diabetic subjecs (3.7o/o)arrd increased proportion of hybrid receptors in target tissues control subjects (2.7o/o) 721. However, in a combined of insulin action might affect insulin binding and thereby analysisof the Hoorn and the Rotterdam data, the Metess insulin sensitivity by sequesteringinsulin receptors in a variant was found at frequencies of 4.4o/oin rlpe 2 less insulin-responsive form. Furthermore, it is possible in control subjects,suggesting that in human tissues hybrid receptors may be primarily diabetic subjectsand L.8olo an association between the Metess variaat and tlpe 2 responsive to mitogenic effects of IGF-I rather than diabetes l72l at least in the Netherlands.Another study metabolic effecs of insulin. Thus, the upregulation of used denaturing gradient gel electrophoresis(DGGE) to hybrid receptors in tissues such as vascular smooth screen an ltalian population of 103 type 2 diabetic musde cells may increasethe sensitivity to the mitogenic subjecs for mutations in exons 17-27 of the insulin effects of IGF-I, thus contributing to diabetic vascular receptor gene.One patient was identified asheterozygous complications. Alternatively, the upregulaton of hybrid changeat receptorsinduced byhormonal and metabolic factorsmay mutation causingaArg-+6h11r64 for a missense codon 1164 1731. Functional characterization of the be adaptive, perhaps protecting tissues from excessive Ghr1164variant revealed that this receptor exhibits a insutin-induced glucose enury. In this respect, it is an constitutive increasein basal tyrosine kinaseactiviry but a intriguing hlpothesis that insuliVIGF-I hybrid receptors 'lizes futsulin reduced or absent response to insulin 174). This may provide a medranism whereby IGF-I u constitutive activation of the recePtor kinase activity signaling pathways for eliciting metabolic effects. Since leads to a maximal increase in basal glucose transPort elegant studies of kinase-deficient insulin receptor mutants have shown that rraru-phosphorylation is the with no further stimulation in responseto insutin 74l.In a study caried out in a Japanesepopulation of 51 qpe 2 major mechanismof intramolecular autophosphoryladon diabetic subjects,all22 exonsof the insulin recePtorgene 142), it is possible to speculate that IGF-I binding to were amplified and sequenced[75]. Three patients were hybrid receptors may activate the adjacent insulin mutation causing receptor p-subunit or cellular substrates specifically identified with a heterorygous missense a Thr-'Ala83l change in exon 13, and one patient with a associated with the insulin receptor p-zubunit tts missensemutation causinga Tyr-tgysr33a exerting more pronounced metabolic effects in addition heterozygous substitution in exon 22. Transfection snrdies in marnma- to typical growth-promoting effects. Recombinant human lian cells showed that the expressed41"831and Cys133a IGF-I (rhIGF-I) has been proposed as a potentid crapy variants function normally [75]. However, the pedigree for type 2 diabetes. Preliminary studies have Soryn that and linkage analysissuggestedthat the Ala$r variant may treatrnent with rhIGF-I increases insulin spncitivity and improves glycemic control in type 2 diabctic patients be responsiblefor the onset of type 2 diabetes. 176,771.Ttre mechanismsby whidr rhIGF'l elers these effects nvvo are undear as IGF-I cin trt outtt type I IGF receptors,insulin receptors,or botlu Thc oeertrations of increased abundance of hybrids in *dcal muscle of Understanding the molecular basis of impaired insulin type 2 diabetic subjects have rabcd c inriguing EansPort action in type 2 diabetesis a challenglngproblem. Recent possibility ttrat IGF-I might stimrrlla Sluc musch ES instiVIGF-I glycogen in synthesis yielded and considerable studies have biochemical cell biolory cryression of insights into the molecular mechanismsof insulin action hybrid receptors. Therefore, iDorcd to an increase in and their role in the pathophysiolory of diabetes. The hybrid receptors could contribc patiens The dinical cloning of the two functionally distinct insulin receptor rhIGF-I sensitivity in qpe 2 dit isoforms and the identification of hybrid receptors relevance of differences in bo fu lw-affinity Exll+ rcccptors is at present and a form and insuliVlGF-l lDd composed of an insulin receptor aB-hemirecePtor by Z)-Xr| of insulin binding is provided Reduction unclear. type 1 IGF receptor ap-hemireceptor have insulin resistance observed the mertd to cause unlikely by which significant information on the mechanism qpe by itself 2 abetes because of the patients with in binding can modulate receptor in insulin heterogeneiry metabolic and mitogenic signaling in target tissues in presence of spare rctprs. Nevertheless,it is possible both physiological and pathological states. Thus, func- that defects in iri$lb linding, even if minor, may be tional srudies on the role of the two insulin receptor additive to genctic c acquired defects in intracellular isoforms have led to the suggestionthat insulin receptor signaling thus cocuting to the resistance to insulin mRNA alternative splicing represents a regulatory action. Ty?e 2 diabcrcr b the leading cause of renal failure, mechanism for modulation of insulin sensitivity in

Conclusions

Copyright O 2001 John Wiley & Sons, Ltd.

Diaberzs Metab Ru Rcv 2OO7; 17: 363-373.

lR in Typez Diabetes blindness, and lower limb amputations in adult individuals and is a major risk factor for cardiovascular diseases. Genetic predisposition is a primary determinant for development of type 2 diabetes in the urbanized Western world. Therefore, identification of the genetic defects resulting in the susceptibiliry to acquire We 2 diabetes is currently a major challenge for dabetes research. Several mutations in the insulin receptor gene have been described in genetic syndromes of severe insulin resistance. However, these are clinically and pathogenetically distinct from rype 2 diabetes. Mutations in the insulin receptor gene have been identified in every study that examined an appropriatelv sized patient population of r,vpe 2 diabetic subjects. In these srudies the insulin receptor gene the prevalence of mutations to 7.8o/o.Hon'ever. in most studies, ranged from O.4o/o of the insulin of the protein coding sequence only 20-3 Oo/o receptor gene \r'as investigated. Funhermore. the molecular scanning strategiesn'ere des:gned in a n'av that rvould onlv identifv certain n?es of nutations. but n'ould n r u : a i i o : s : ; t h e p r o m o t e ro r i n fajl to identifl' deletions. domains of the insu.in receptor gene. In other regirlatoryrot tnvestigated, the as much as 70-80c'oof thc-ge;e \\.a-c true prevalence of mutations l;r :hc. :nsu:n receptor rn could be as high as 5o/o conunon forms of +'pe 2 cirabeic-s

377

subsuate (IRS) transducrion system: distinct and overlapping signaling potential. DiabetesMetab ResRev 2000; 16: 434 44I. 11. Porzio O, Federici M, Hribal ML, et ol. The Glye72--fu'g amino acid polymorphism in IRS-1 impairs insulin secretion in pancreatic beta-cells.J CIin Invest 1999; lO4:357-364. 12. Federici M, Hribal ML, Ranalli M, et aL The common Arg972 polymorphism in insulin receptor substrate-1causesapoptosisof human pancreatic islets.IASEB J ZOOL;15:22-24. 13. Kulkarni RN, Bruning JC, Winnay JN, Postic C, Magnuson MA, Kahn CR. Tissue-specificknockout of the insulin receptor in pancreaticbeta cells createsan insulin secretorydefect similar to that in type 2 diabetes. CeII 7999; 96: 329-339. 14. Ullrich A" Bell JR, Chen E{, et aI. Human insulin receptor and its relationship to the tyrosine kinase family of oncogenes-Nature 1985; 313: 756-767. 15. Ebina Y, Ellis L, Jarnagin K, et aI. The suuctural basis for hormone-activated transmembrane signalling. CeIl 1985; 4O: 747-758. 16. Seino S, Seino M, Nishi S, Bell GI. Structure of the human insulin receptor gene and characterization of its promoter. Proc Nqtl Acad Sci U S A 7989; 86: 114-118. 17. Iftistensen C, Wiberg FC, Schaffer L, Andersen AS. Expression and characterization of a 70-kDa fragment of the insulin receptor that binds insulin. J BioI Chem 1998; 273: 7778077786. 18. Sawka-VerhelleD, Tartare-Deckert S, White MF, Van Obberghen E. Insulin receptor substrate-2 binds to the insu[n receptor through its phosphotyrosine-binding domain and through a newly identified domain comprising amino acid 591-786. J Biol Chem 1996; 271: 5980-5983. 19. Kosaki A, Nelson J, Webster NJG. Identification of inron and exon sequences involved in alternative splicing of insulin receptor pre-mRNA.J Biol Chem 7998;273: 10331-10337. 20. Sakata S, Kobayashi M, Miura K, Atassi MZ. Moiecular [3]. recognition of human insulin receptor by autoantibodies in a human serum. Immunol Invest 1988; 77:237-242. 21. Sesti G, Marini MA, Montemurro Pu et o.l. Evidence that Acknowledgements rwo naturally occurring human insulin receptor alphasubunit variants are immunologically distinct. Diabetes 1992; European This study was srppore: : ir:': :.' 41:6-11. Ricerca 22. Sesri G, Tullio AN, Marini MA, et al. Role of the exon 11 of the Community No. QLGl-CT-1999 ,-'-'r insulin receptor gene on insulin binding identified bv antiCOFIN iiralizzata RF9B from \fi;::st::c, :'r..a pepride antibodies. Mol CeII Endocinol 1994; 7O7: 127-127. ^1 9 9 f r o m M i n i s t e r o d e l l ] * : : : ' ' ' e : s : : a 23. Mosthaf L, Grako D, Dull TJ, CoussensL, Ullrich A, Mcclain D.{. ..c:rologica (GS and RL). Functionally distinct insulin receptors generated by nssuespecific alternative splicing. EMBO J I99O;9: 2409-2413. 24. Yogt B, Carrascosa JM, Ermel B, Ullrich A, Haring HU. The rwo isorypes of the human insulin receptor (HIR-A and HIR-B) follow different internalization kinetics. Biochem Biophys Ru Commun I99I; 177 : 1013-1018. "5. Yamaguchi Y, Flier JS, Yokota A, BeneckeH, Backer JM, Moller of ri? 2 diabetes:rr.t':atr,lrcand RA. Pathogenesis DE. Functional properries of n"ro naturally occurring isoforms of implications for idendhrng Jlabetes genes D::iei.:i the human insulin receptor in Chinese hamster ovary cells. 5: 177-269. Endocinologr 1991; 129: 2058-2066. 3ogardusC, Mon D\1. PratlevF.E.Tne naturai h;sion 26. \f cclain DA. Different ligand affinities of the rwo human insu[n :. secretorydysfunction and insuhn resistancein the recepror splice variants are reflected in parallel changes in : ' - : i o f r y p e2 d i a b e t e s J C I i n/ n y e s t1 9 9 9 : 1 0 4 : mcllitus. sensitiriw for insulin action. MoI Endocrinol I99I; 5: 734-739. :::a A. Accili D. et al. Ilutation,s t h e i n s u L n 27. \'amaguchiY, Flier JS, BeneckeH, RansilBJ, Moller DE. Ligandbinding propenies of the two isoforms of the human insulin i.^.docineRev 1992; 13: 56G-595. receptor. Endocrinologt 1993; 732: 1 132-1 138. ' ..:,:re l i n k a g eo f g l u c o k i n a s e M, SunF, et aI.Close i--,rne 7p to early-onsetnon-insulin-dependent 28. FrascaF. Pandini G, SciaccaL, et al.Insulin receptor isoform Ao a '.'::ure 7992: 356: I62-lG. nertlv recognized, high-affiniry insulin-like growth factor II receptor rn fetal and cancer cells. Mol CeII BioI 7999; 79:. \. Kaisaki PJ, et aI. \!utations in the 3278-3288. actor-1 alpha Sene in marurin'-onset 29. Seino S. Bell GI. Alternative splicing of human insulin receptor . - ' \ ' . r r u r e1 9 9 6 ; 3 8 4 : 4 5 H 5 8 . messenger RNA. Biochem Biophys Res Commun 1989; 159: : tla N, et al- Mutations in the hepatoqre ''r:e in maturiry-onsetdiabetesof the 3 12 - 3 1 6 . expression 3t{ 30. Moler DE. Yokota A, Caro JF, Flier JE. Tissue-specific of nvo alternarive spliced insulin receptor mRNA in man. Mol \1. et aI. Mutation in hepatocl-ie Endocnnol1989; 3: 1263-1269. --:2) associated r'r'ith \ODY. \'ct 31. SesriG, Tullio AN. D'Alfonso R, et al. Tissue-specific expression . .': V, Clarke \4L. Habener JF. of nso alternatively spliccd isoforms of the human insuLin rccL-prorprotein. Acta Diobetologico1994; 31: 59-65. a single nucleodde der'don 3'. Sciacca L. Costantino A, Pandini G, et ai. Insulin receptor . ience. Nat Genet 1997: I 5: acti'''etion bv IGF-lI inbreast carcer: evidence for a new au:ocnne'paracrine mechanism. Oncogene1999; 18: 2471and the IRS protein.

References

Insubn receptor

33 iorir"n

S. Zierath J, Wedell A, Waleberg-HenrikssonH,

DiabetesMetab ResRev 2001: 17:363-373.

,T2 Llthman H. Regulation of human insulin receptor RIrIA splicing in vivo. Proc NatI Acad Sci U S A 1994;91: 1465-7469. Norgren S, Sheng Li I Luthman H. Regulation of human insu[n receptor RNA splicing in HepG2 cells: effecs of glucocorticoid and low glucose concentration. Bbchem Bioplrys Rs Commun 7994; 199: 277-284. Sesti G, D'Alfonso R, Vargas Punti MD, et aI. Peptide-based radioirnmunoassay for the nro isoforms of the human insulin receptor. Diabetologis 1.995; 38: 4H53. Lftidl,\ Gray A, Tam AW, et cl. Insulin-like growth fuctor I receptor primary stmcture: comparision with insulin receptor suggests structural determinants ttrat define functional specific:ry. EMBO J 1986; S:250T2512. Soos lvf,qb Siddle K Immunological relationships between receptors for insulin and insulin-like growth factor-l. Biodtem J L989;263: 553-563. Iksuya J, Paz B, Maddux BA Goldfine ID, Hefta Sd FujitaYamagudri Y. Characterization of human placental insulin-fke growth factor-Vinsulin hybrid receptors by protein microsequencing and purification. Biochemistry 1993; 32: 13531L3536. Soos M,\ Whittaker J, l,ammers R, Ulldch,\ Siddle K Receptor for insulin and insulin-like growth factor-I can form hybrid dimers. Biochem J 199O; 27O: 383-390. Treadway L, Morrison B, Goldfine ID, Pessin JE. Assembb of growth factor-I hybrid recPtors in vino. insuliVinsulin-like J Biol Chem 1989;264:27450_21453. Soos lvrAoField CE, Siddle K Purified hybrid insuliVirrsulin-like growth factor-I receptors bind insulin-like growth factor-I, but not insulin, with high affinity. Bicrjr.emJ 7993;2%): 4L9426. Langlois WJ, Sasaoka T, Yip CC, Oleftky JM. Functional draracterization of hybrid receptors composed of a truncated insulin receptor and wild type insulin-like growth factor I or insulin receptor. Endocrinolog 1995; 136: 197V7986. Seely BL, Reichart DR, Takata Y, Yip CC, Olefrky JM. A growth factor-I functional assessment of inzuliVinsulin-like hybrid receptors. Eniloainologr 1995; 136: 163!1641. Federici M, Porzio O, Zucaro l- et aI- Distribution of insuliV insulin-like growttr factor-I hybrid receptors in human tissues. Mol CelI Endocrinol L997; 129: 727-126. Bailyes EM, Nave BT, Soos I\,fA,,Orr SR, Hayward AC, Siddle K Insulin receptor/IGF-I receptor hybrids are widely distributed in mammalian tissues: quantification of individual receptor species by selective immunoprecipitation and immunobloning. Biochem J 1997;327:2O9-2L5. Federici M, Lauro D, D'Adamo M, et aI. brpression of insuliV IGF-I hybrid reaeptors is increased in skeletal musde of patiens with clnonic primary hyperinsulinemia. Dicbeta L998; 47: 87-92. Federici M, Porzio O, lauro D, et ol.lncreased abundance of growth factor-I hybrid receptors in skeletal insuliry'insulin-like musde of obese subjects is correlated with in vtvo insulin sensitivity. J Clin Endoqinol Metab 1998; 83: 2917-291,5. Federici M, Giaccari,q, Hribal MI+ et cl. Evidences for glucose/ hexosamine in vivo regulation of insulinlGF-I hrid recePtors assembly. Diabetes 1999 ; 48: 2277 -2285. Modan-Moses D, Janicot M, Mdenithan JC, Iane MD, Casella SJ. growth factor I E:rpression and function of insuliVinzulin-like hybrid receptors during differentiation of 3T3-L1 preadipocytes. BiochemJ 1998;333: 82$S31. Mosthaf L, Vogt B, Haring HU, Ullrich,A- Altered expression of insulin receptor tlpes A and B in the skeletal musde of noninsulin-dependent diabetes mellitus. hoc NatI Acad Sci USA 1991;88: 472U73O. Mosthaf I Eriksson d Haring HU, Groop L, Widen E, Ullrich AInsulin receptor isotypes expression correlates with risk of noninsulin-dependent diabetes mellitus. Proc NatI Acad. Sci U S A 1993; 9O: 2633-2635. Norgren S, Zierath J, Galuska D, Walleberg-Henrilsson H, Luthman H. Differences in the ratio of RNA encoding nao isoforms of the insulin receptor between control and NIDDM patients. Diabetes 1993; 42: 675487. Benecke H, Flier JS, Moller DE. Alternatively spliced variants of the human insulin receptor protein. J CIin lnvest 1992; 89:206-2070. Anderson CM, Henry R& IGrudson PE, Olesky JM, Webster NJG. Relative expression of insulin receptor isoforms does not differ

G. Sesff ct ot
diabetes mellitus in lean, obese, and noninsulin-dependent subjects. J Clin Eniloainol Metab L993; 72: 1380-f 383Hansen I Bjorbaek C, Vestergaard H, Gronskov K Bak JF, Pedersen O. Expression of insulin receptor spliced rrarians and their funaional correlates in musde from patients with noninsulin-dependent diabetes mellitus. J CIin Enilwinol Metob 1993; 77: 1500-L505. 56. Sesti G, Marini lvf{ Tullio Al.I, et cL Altered eryression of the two naturally occuring human insulin receptor rrarians in diabetes mellitus isolated adipocytes of non-insulindependent patients. BiodtemBiophys Fcs Commtn 1991; 181: 1419-142457. Kellerer M, Sesti G, Seffer E, et aL. Altered panern of insulin receptor isotlpes in skeletal muscle membranes of type II (noninsulin-dependent) diabetic subjects. Diabetologia 1993; 36: 628H'32. 58. Sbraccia P, D'Adamo M, Ironetti F, et aL Chronic primary hyperinsulinemia is associated with altered insulin recptor nRNA sp[cing in musde of patients with insulinoma. Diabtobgis. 1993; 39: 22O-225. 59. Federici M, Zucaro L, Porzio O, et al.Increased eryression of growth factor-I hybrid receptors in skdetal insuliVinsulin-like diabetes mellitus subjects. musde of non-insulin-dependent J Clin Invut 1996;98:2887-2893. 60. Federici M, Porzio O, Zucaro In et ol.Increased abundance of hybrid reaeptors in adipose tissue from NIDDM insuliMcF-I patients. Mol CeLlEndocrinol l99T; 135: 4147. 6 1 . Valensise H, Uu YY, Federici M, et al.Increased expression of low-affinity insulin receptor isoform and insuliVinsulin-like growth factor-I hybrid receptors in term placenta from insulin resistant women with gestational hyperterrsion. Diafutologin 7996; 39: 952-960. Spampinato D, Pandini G, Iuppa ,t Trischina V, Vigneri R, growth factor-I h),brid recepors Friaitra L Insulin/insulin-like overexpression is not an early defect in insulin-resistant subjectr J Clin EndoqinolMetob 2000; 85: 42194223. 63. Moller DE, Yokota d Flier JS. Normal insulin reaeptor cDl{A sequence in Pima lndians with NIDDM. Diabetes 1989; 3t. 149G1s00. 64. D'Alfonso R, Fusco d Napolitano ML, et aI. UncorH polymorphisms in the human insulin receptor gene. fry Cb Endoqinol 1993; l0t (Suppl. 2):52-53. 65. OTahitly J, Choi WH, Patel P, Turner RC, Flier JS, Mob f' Detection of mutations in insulin-receptor gene ia liDfl pl7. patients by analysis of single-suanded conformetid morphisms. Diohetu 799L; 4O: 777:782. tlr. A 66. Strack V, Bossenmaier B, Stoyanov B, Mushacl J, l{rf i$tin 973 valine to methionine mutation of drc b d Shc receptor: interaction with insulin-receptor subs-l in HEK 293 cells. Diabetobgia 1997;4O: f f3!lla d ia$lin 67. Cama d Patterson ,\ thdowaki T, et aL Oq J CIin breceptor cDl.IA from an insulin resistan Pb Endoqinol Metab l99O; 7O: 1155-1161. Aoelysb of 68. Kusari J, Verma US, Buse JB, Henry RR' Olq.IiEb-scnsitive the gene sequences of the inzulin reccPao3 d fl ornrnon-r)?e glucose transporter (GLU"I-4) in prtil 1991; 88: non-insulin-dependent diabetes rrpllinr t H

34.

35.

36.

37.

38.

39.

40.

41.

42.

43.

44.

45.

46.

47.

48.

49.

r323-r.330. R, R..t A, tcr JS,Moller DE. J, IGookd Morgan 69. O'Rahilly

Eansporter Insulin receptor and insulin-r6ptb@ (GLUT-4) mutations and pofulh e Welstr rvp. 2 (non-insulindependent) fudogia 1992; diaba. ffa. 35:486489. Hansen I. Hansen T, Claus IL r d- Tb n1985 met insulinpopulation: lack of ht receptor variant in thc D diabetes mellitus or associations y,'i so6-irprh-rrt. lgyh 0: 1532-1535. inzulin resistanae- Attr J t- DE, Van Der Does FE, f[ Glce t Hart LM, Stolk RP, ll* variant Met-985 inlin-rcceptor Maassen J,{- Associedo ddiabetes melliwith hypergbrc@ir ad eanindcpendent study. Am J Hum r lryhinased tus in the Nethcrl Genet 1996; 59: I I llll2S'tHanLM, Stoll RP. t5cl ol- Prevalence of variants in nelligus in the Netherlands: b ts2dirbetes c,ndidate gm drc Hoorn study. J Clin Endocrinol d tlre Ronerdan qt Metah 199*, Ut: Ef@6. G, et al. MDDM associated with C-ocgzzaS, PqcI Risrdi finase domain of insulin receptor genemutarioo r! DDinfu.a f9: r|L ftl-526.

50.

70.

51.

71. ' 72.

52.

53.

73 -

54.

Copynght @) 2001 John Wiley & Sons, Ltd.

Diatetzs Maab Res Rev 2OOI;77:. 363-373-

lR in Type z Diabetes 74. Pea'vzziello d Formisano P,'Miele C, et al. Defective insulin action in fibroblasts from noninsulin-dependent diabetes mellitus patiens with Gln 1L52 insulin receptor mutation. J CIin Endotinol Metah L993;77: 409412. 75. Kan M, Kanai F, Iida M, et o1. Frequency of mutations of insulin receptor gene in Japanese patients with NIDDM. Dinbetes l99S; 44: 108L-1086. 76. Moses AC, Young SCJ, Morrow L,t Otsrien M, Clemmons DR

373 Recombinant human insulin-like-growttr factor I increases insulin sensitivity and improves glycemic control in rype II diabetes. Dioheta 1996; 45: 9f-100. 77. Cusi K DeFronzo R- Ecombinant human insulin-like growth factor I treatment for 1 week improrres metabotc control in type 2 diabetes by ameliorating hepatic and muscle insulin resisrance. J Clin EndoainolMetab 2000; 85: 3O77-3OU-

C6I

G If

.IbrLtr

J-.

Ld.

Diabtes Metab ResRev 2fi)1: 17:363-373.