BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS ARTICLE NO.

234, 764– 768 (1997)

RC976695

Identification of Two Novel Insulin Receptor Mutations, Asp59Gly and Leu62Pro, in Type A Syndrome of Extreme Insulin Resistance Mathias Rouard,* Franc¸oise Macari,* Olivier Bouix,*,† Corinne Lautier,* Jean Fre´de´ric Brun,† Patrick Lefebvre,‡ Eric Renard,‡ Jacques Bringer,‡ Claude Jaffiol,‡ and Florin Grigorescu*,1 *Laboratoire d’Endocrinologie Mole´culaire, Institut Universitaire de Recherche Clinique and Centre de Recherche de Biochimie Macromole´culaire, ERS 155 CNRS, 34093 Montpellier, France; and †Service d’Explorations des Hormones et Me´tabolismes and ‡Service d’Endocrinologie, Hoˆpital Lapeyronie, 34295 Montpellier, France

Received April 22, 1997

To elucidate genetic determinants of insulin resistance, we investigated insulin receptor (IR) and insulin receptor substrate-1 (IRS-1) genes, in vitro IR function and in vivo insulin sensitivity in a family with Type A syndrome. Two missense IR mutations (Asp59Gly and Leu62Pro) found in the proband, resulted in reduction by 90% of insulin binding to erythrocytes, decreased receptor autophosphorylation and a dramatic reduction of insulin sensitivity. The proband and mother were heterozygote for Gly972Arg IRS-1 variant. Asp59Gly mutation, also carried by proband’s brother with no consequence on insulin sensitivity, was inherited from the mother who is diabetic and insulin resistant and Leu62Pro was from the father. We conclude that severity of insulin resistance in the proband may be explained by the genetic condition of compound heterozygote for IR mutations while severe insulin resistance in the mother raises the possibility that other genetic factors, like IRS-1 polymorphisms, may contribute to the phenotypic expression of IR mutations. q 1997 Academic Press

Insulin signaling at the cellular level is initiated by activation of the insulin receptor tyrosine kinase and subsequent tyrosine phosphorylation of intracellular substrates. Among these, IRS-1, the major insulin receptor substrate, undergoes rapid phosphorylation on multiple tyrosine residues and thus bind Src Homology 1 To whom correspondence should be addressed at the Laboratoire d’Endocrinologie Mole´culaire, I.U.R.C. 75, rue de la Cardonille, 34093 Montpellier, France. Fax: 33 467 542 731. E-mail: florin@ spectrum.iurc.inserm.montp.fr. Abbreviations used: NIDDM, non-insulin-dependant diabetes mellitus; IVGTT, intravenous glucose tolerance test; PCR, polymerase chain reaction; IGF-1, insulin-like growth factor-1.

domains (SH2) of signaling proteins (1). Defects at various steps in the insulin transduction pathway may be involved in the pathogenesis of insulin resistance (2). The role of insulin receptor defects has been demonstrated by transgenic experiments and identification of gene mutations in genetic syndromes of extreme insulin resistance, including Type A syndrome (2-5). By contrast, the role of IRS-1 remains more elusive (6, 7). Data on IRS-1 gene defects in syndrome of extreme insulin resistance are not yet available, however the Gly972Arg variant could enhance the risk of NIDDM in obese subjects (8, 9). Observations on a novel polygenic model of NIDDM showed that heterozygote mice for both insulin receptor and IRS-1 null alleles become overtly diabetic (10). These data give new insight on how combination of two defects in insulin signaling may induce insulin resistance and diabetic phenotype. In contrast to similar syndromes such as RabsonMendenhall or leprechaunism, where mutations in the insulin receptor are constantly present, the prevalence for receptor mutations was estimed to be much lower among patients with Type A syndrome (3, 5). To determine whether this underestimated prevalence is due to still undetected mutations in the insulin receptor gene or coexistence of other post-receptor genetic defects, we have screened the insulin receptor gene for mutations and extended the search towards IRS-1 gene. We describe, for the first time in Type A syndrome, two novel insulin receptor mutations, Asp59Gly and Leu62Pro as compound heterozygote. Additionally, we report IRS-1 gene polymorphisms and correlate with insulin sensitivity in vivo. MATERIALS AND METHODS Subjects. The proband, designated A-Mont-3, was a white female who was found to have glycosuria at age 11 and investigated for

764

0006-291X/97 $25.00 Copyright q 1997 by Academic Press All rights of reproduction in any form reserved.

AID BBRC 6695

/

692d$$$861

05-09-97 12:10:00

bbrcg

AP: BBRC

Vol. 234, No. 3, 1997

BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS TABLE 1

Clinical and Laboratory Features of the Proband with Type A Syndrome of Insulin Resistance and First-Degree Relatives

BMI (kg/m2) NIDDM Fasting insulin (pmol/l) IVGTT (minimal model) K g (1002 min01) Si (1003 pmol/l min01) Sg (1002 min01)

Proband A-Mont-3

Mother

Father

Brother

19.6 / 633

38.6 / 283

34.1 0 206

28.1 0 120

1.04 0.25 1.36

0.92 0.17 1.51

2.03 5.66 3.02

2.27 17.3 3.24

Controlsa (n Å 38) 22.9 { 2.3 50.1 { 2.6 2.3 { 1.1 18.9 { 8.3 3.1 { 1.3

Note. Kg, index of glucose tolerance; Si , insulin sensitivity index; Sg , glucose effectiveness. a Values are given in mean { SD.

hirsutism, acanthosis nigricans and primary amenohrea at 14. Laboratory examination revealed hyperandrogenism and celioscopic examination confirmed polycystic ovarian disease. The patient’s diabetes and insulin resistance were further documented by fasting hyperinsulinemia, oscillating between 633 to 1867 pmol/l (occasionally peaking at 3028 pmol/l), presence of acanthosis nigricans and fasting blood glucose levels oscillating between 3.5 to 12.3 mmol/l, peaking at 21.7 mmol/l after 75 g of oral glucose load. Glycated hemoglobin was comprised between 8 and 10.5%. The proband’s mother was 55, obese and known diabetic since age 43. The father was 54, obese with normal glucose tolerance whilst the proband’s brother was 29, only moderately overweight and with normal glucose tolerance. The degree of insulin resistance in all family members was assessed by

the minimal model approach of Bergman from data of IVGTT as previously described (11, 12) and illustrated in Table 1. Screening for mutations in insulin receptor and IRS-1 genes. Genomic DNA was isolated from peripheral blood cells using Puregene DNA Isolation Kit (Gentra Systems, Minneapolis, USA). The 22 exons and flanking introns of the insulin receptor gene were amplified using PCR and specific primers adapted from Ref. 13. PCR conditions were denaturation at 947C for 5 min followed by 35 cycles of denaturation (947C for 1 min), annealing (567C for 50 sec) and extension (727C for 1 min), and 10 min of post-extension at 727C. The amplification of exon 1 was carried out in the presence of 5% v/v dimethylsulfoxide. The PCR products were purified using the Pharmacia EasyPrep PCR

FIG. 1. Partial nucleotide sequence at mutated sites of IR from proband with Type A syndrome and first-degree relatives. Direct sequencing of PCR products was performed with ABI PRISM Dye Terminator Cycle Sequencing Ready Reaction Kit. Panels correspond to electrophoregrams for IR obtained from ABI Sequencing Analysis Software. Mutations are indicated by arrows and corresponding residues are indicated in the aminoacid sequence on the top of each panel. Symbols Ω, ø and j indicate hyperinsulinemia, NIDDM and Type A syndrome, respectively. 765

AID BBRC 6695

/

692d$$$861

05-09-97 12:10:00

bbrcg

AP: BBRC

Vol. 234, No. 3, 1997

BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS

FIG. 2. 125I-insulin and -IGF-1 binding to erythrocytes from the proband (A-Mont-3) with Type A syndrome and first-degree relatives. Hormone binding to intact erythrocytes was performed in cell suspension of 4.4 110 9 cells/ml at 157C and is expressed as percent of total radioactivity. Panel A, competition-inhibition curves of insulin binding. Panel B, competition-inhibition curves of IGF-1 binding. Symbols l, s and h indicate the proband, mother and father respectively. Shaded area represents values of control subjects expressed as mean { 1SD.

ciated mutations, although previously described DNA polymorphisms were noted: silent polymorphisms homozygous for the variant allele in exon 1 within the signal peptide codon 020 GGG (Gly)rGGA (Gly) and in exon 3 codon 276 CAA(Gln)rCAG (Gln), heterozygote polymorphisms in exons 8 and 9 codon 519 GAT (Asp)/ GAC (Asp) and codon 642 TTC (Phe)/TTT (Phe). Intronic polymorphisms were also found in intron 6, position /31 heterozygous C/T. The proband and her mother were heterozygote for IRS-1 972 variant, GGG(Gly)rAGG(Arg), whereas her father and brother carried normal variants. Sequencing the entire IRS-1 gene revealed no other known polymorphisms in family members, except silent heterozygote polymorphism Gly235 (GGGrGGA) in the proband and her mother. A new missense mutation was found in the father at codon 834, CAT(His)rTAT(Tyr), not inherited by either sibs.

product Prep Kit (Pharmacia Biotech, Uppsala, Sweden) and then sequenced directly from both ends using the same primers as used for PCR and a AmpliTaq FS ABI PRISM Dye Terminator cycle sequencing kit (ABI, Foster City, CA) on an ABI 373A DNA sequencer according to the manufacturer specifications. The exon containing the coding region of the IRS-1 gene was amplified as series of overlapping fragments of 242 to 519bp2 which were sequenced directly as described above. Mutations in the insulin receptor gene were confirmed by cloning the exon 2 PCR products in pGEM-T vector (Promega, Madison, USA) and sequencing clones corresponding to both alleles. IRS-1 Gly972Arg variant was confirmed by restriction endonuclease digestion (SmaI, New England Biolabs, Beverly, USA) of the corresponding PCR product.

Biochemical defects of insulin receptor. 125I-insulin binding to circulating erythrocytes was decreased by 90% in the proband and 60% in both parents compared to controls. By contrast 125I-IGF-1 binding to proband’s cells was in the higher normal range. Analysis of insulin binding data revealed combined defects in affinity constants and number of available sites (Fig. 2). Autophosphorylation of erythrocyte purified receptors from the proband was stimulated in the presence of insulin by 2.1-fold relative to basal level, representing 30% decrease compared to simultaneous control. Similar decrease was detected in phosphoprotein content in the 95 kDa protein under stimulation with IGF-1 (Fig. 3).

Biochemical studies. 125I-insulin and -IGF-1 binding to intact erythrocytes was performed as previously described (14) using 125I [TyrA14] insulin and [Tyr3] recombinant IGF-1 with 2000 Ci/mmol specific activity (Amersham, Buckinghamshire, U.K). Receptors were extracted from inside-out erythrocytes vesicles and purified by affinity chromatography on Ricinus communis lectin (RCA60) coupled to agarose. Autophosphorylation was performed with 10 mg purified protein as described (14) and phosphoproteins were precipitated with anti-phosphotyrosine antibodies (kindly supplied by Dr. M. White) and protein A (Pharmacia, Uppsala, Sweden).

DISCUSSION We report in this paper a unique genetic condition associating two novel insulin receptor heterozygote compound mutations in a patient with Type A syndrome of extreme insulin resistance. In addition we

RESULTS Genetic features. Each exon and flanking intron region of the insulin receptor gene of the proband was amplified using PCR and directly sequenced. There were two nucleotide substitutions in exon 2 that resulted in amino acid replacements: codon 59, GAT (Asp)rGGT (Gly) and codon 62, CTG (Leu)rCCG (Pro). Sequencing of exon 2 of the proband’s parents indicated that the Gly59 allele was inherited from the mother and Pro62 from the father (Fig. 1). Thus, the proband was a compound heterozygote. The sequences of exons 1 and 3-22 were normal with no potential disease-asso2 Primer sequences are presented in a manuscript in preparation: F. Macari et al.

FIG. 3. Autophosphorylation of erythrocyte IR from the proband (A-Mont-3) with Type A syndrome and simultaneous control. Erythrocyte insulin receptors were solubilized with Triton X-100 and further purified by Ricin (RCA60) affinity chromathography and in vitro stimulated with maximal concentrations of insulin or IGF-1 (10 mg/ ml). Proteins were phosphorylated in the presence of g-32P [ATP], precipitated with anti-phosphotyrosine antibodies, resolved in 7.5% polyacrylamide gels and autoradiographed.

766

AID BBRC 6695

/

692d$$$861

05-09-97 12:10:00

bbrcg

AP: BBRC

Vol. 234, No. 3, 1997

BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS

have investigated the degree of insulin resistance in vivo and identified IRS-1 gene polymorphisms in family members. The insulin receptor mutations were not previously reported, they are located in exon 2 of the gene coding for a region that has been implicated in insulin binding (15). This is in concordance with the dramatic reduction in the insulin receptor binding to erythrocytes from the proband, similarly to some other mutations in exon 2 (5, 16). Decrease of insulin binding by 60% in parents, carrying one of each mutation, suggests that both mutations would affect binding function, although, hyperinsulinemia per se found in both obese parents, could contribute by down regulation to a reduction in number of available sites. The reduced insulin binding on cells from these patients may be explained either by alteration of the insulin receptor processing or direct effect of the replaced residues involved in the putative insulin binding site. Indeed, previous studies have shown that residues 168 of the a-subunit confer ligand specificity (17). Moreover, these two mutations occur in a particular location involved in the architectural organization of the insulin receptor a-subunit. Indeed, the tertiary structure of insulin receptor extracellular model contains loosely conserved motifs which are, schematically, hydrophobic aminoacids centered generally by a Gly residue (18). In the insulin receptor these motifs are repeated four times in two non-contiguous domains, L1 (aa 1-119) and L2 (aa 311-428) encompassing the cystein rich region. The mutation Asp59Gly affects the central amino acid of the second hydrophobic motif of domain L1. The mutation Leu62Pro eliminates a conserved hydrophobic residue in the same motif. Previous studies using alanine scanning mutagenesis have shown that insulin binding to Asp59Ala mutated protein resulted in a reduction in Kd by 1.5 fold, while Leu62Ala mutant is not processed (19). Alteration of biochemical events downstream from insulin binding was indicated by decreased receptor autophosphorylation, although, in quantitative terms, the reduction by only 30% was unexpectedly modest when compared to insulin binding defect. It is likely that transphosphorylation between insulin and IGF-1 receptors would contribute to concurrent potentiation of IR and reduced IGF-1 receptor phosphorylation (20). Phosphorylation of IRS-1 could not be investigated in freshly isolated cells from this patient. However, preliminary experiments using monocytes from the proband, have shown that phosphatidylinositol (PtdIns) 3* kinase activity was poorly stimulated by insulin, 1.4 versus 5.2 { 1.5 (arbitrary units) in controls. The significance of genetic defects may be partially approached from in vivo insulin sensitivity in these patients, although considerable fasting hyperinsulinemia and hyperglycemia as well as effects of age and body mass preclude the Bergman’s model. Neverthe-

less, insulin sensitivity was dramatically impaired in the proband who carried two missense mutations in the insulin receptor gene (Table 1). Abnormalities found in vivo in both parents, raise the question of what would be the relative contribution of each receptor mutation on the overall insulin resistance. The father, carrying the heterozygote receptor Leu62Pro mutation, showed Si decreased by 66% from control population, but comparable to that of non-diabetic obese subjects (SiÅ7.18{1.75 10 03 pmol/l.min01, nÅ21). This value of Si do not point out considerable effect of paternal gene defects. Although, specific functional consequences are difficult to consider since they potentially overlap those related to His834Tyr IRS-1 mutation and obesity. The proband’s brother, expressing heterozygote receptor mutation Asp59Gly was phenotypically normal except overweight and showed Si values in the range of control non-diabetic lean subjects. The mother carrying the same heterozygote receptor mutation appeared to be severely insulin resistant with overt diabetes, raising the possibility that other genetic factors contribute to the phenotypic expression of Asp59Gly mutation. Our finding of heterozygote Gly972Arg IRS-1 variant in the mother and the proband, but not in the other sib, suggest that IRS-1 may be one of these contributing factors. This hypothesis is also supported by the likely diabetogenic role of this IRS-1 variant in a context of obesity (10), impaired PtdIns 3’ kinase activation by insulin in transfected cells expressing this gene defect (10, 11) and more recent polygenic models of NIDDM involving both insulin receptor and IRS-1 genes (9). Our data suggest that screening for mutations through the insulin receptor gene and exploration for a combined IRS-1 gene defects seems necessary before excluding the involvement of these transduction proteins in many cases of Type A syndrome. In conclusion, the genetic defects uncovered in this family seems to reasonably contribute to the generation of the insulin resistant state in the proband and explain discrepancies of diabetes phenotype expression among first degree relatives. ACKNOWLEDGMENTS We thank Dr. G. I. Bell for his critical reading of the manuscript. This work was supported by following grants: ER87 from Ministe`re de l’Enseignement Supe´rieur et de la Recherche (MESR) to C.J., Grant 701016 from CNAMTS to F.G., Fondation pour la Recherche Me´dicale to F.G. M.R. has been supported by Novo Nordisk, France, Ligue Re´gionale contre le Cancer, Languedoc Roussillon and Association pour la Recherche sur le Cancer.

REFERENCES 1. Cantley, L. C., Auger, K. R., Carpenter, C., Duckworth, B., Graziani, A., and Kapeller, R. (1991) Cell 64, 281 –302. 2. Kahn, C. R. (1994) Diabetes 43, 1066 –1084.

767

AID BBRC 6695

/

692d$$$861

05-09-97 12:10:00

bbrcg

AP: BBRC

Vol. 234, No. 3, 1997

BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS

3. Moller, D. E., and O’Rahilly, S. (1993) in Insulin Resistance (Moller, D. E., Ed.), pp. 49 –81, Wiley, Chichester, UK. 4. Accili, D., Drago., J, Lee, E. J., Johnson, M. D., Cool, M. H., Salvatore, P., Asico, L. D., Jose´, P. A., Taylor, S. I., and Westphal, H. (1996) Nature Genetics 12, 106–109. 5. Taylor, S. I., Kadowaki, T., Kadowaki, H., Accili, D., Cama, A., and McKeon, C. (1990) Diabetes-Care 13, 257– 279. 6. White, M. F., Maron, R., and Kahn, C. R. (1985) Nature 318, 183–186. 7. Sun, X. J., Rothenberg, P., Kahn, C. R., Backer, J. M., Araki, E., Wilden, P. A., Cahill, D. A., Goldstein, B. J., and White, M. F. (1991) Nature 352, 73– 77. 8. Almind, K., Bjorbaek, C., Vestergaard, H., Hansen, T., Echwald, S., and Pedersen, O. (1993) Lancet 342, 828–832. 9. Almind, K., Inoue, G., Pedersen, O., and Kahn, C. R. (1996) J. Clin. Invest. 97, 2569 – 2575. 10. Bru¨ ning, J. C., Winnay, J., Bonner-Weir, S., Taylor, S. I., Accili, A., and Kahn, C. R. (1997) Cell 88, 561– 572. 11. Bergman, R. N. (1989) Diabetes 38, 1512 –1527.

12. Brun, J. F., Bouix, O., Monnier, J. F., Blachon, C., Jourdan, N., Baccara, M. T., Fe´dou, C., and Orsetti, A. (1996) Acta Diabetol. 33, 1– 6. 13. Seino, S., Seino, M., and Bell, G. I. (1990) Diabetes 39, 123–128. 14. Grigorescu, F., Flier, J., and Kahn, C. R. (1986) Diabetes 35, 127–138. 15. De Meyts, P., Gu, J. L., Shymko, R. M., Kaplan, B., Bell, G. I., and Whittaker, J. (1990) Mol. Endocrinol. 4, 409–416. 16. Moritz, W., Boni-Schnetzler, M., Stevens, W., Froesch, E. R., and Levy, J. R. (1996) Eur. J. Endocrinol. 135, 357– 363. 17. Kjeldsen, T., Andersen, A. S., Wiberg, F. C., Rasmussen, J. S., Scha¨ffer, L., Balschmidt, P., Moller, K. B., and Moller, N. P. (1991) Proc. Natl. Acad. Sci. USA 88, 4404 – 4408. 18. Bajaj, M., Waterfield, M. D., Schlessinger, J., Taylor, W. R., and Bludel, T. (1987) Biochim. Biophys. Acta 916, 220–226. 19. Williams, P. F., Mynarcik, D. C., Yu, G. Q., and Whittaker, J. (1995) J. Biol. Chem. 270, 3012 –3016. 20. Moxham, C. P., and Jacobs, S. (1992) J. Cell. Biochem. 48, 136– 140.

768

AID BBRC 6695

/

692d$$$861

05-09-97 12:10:00

bbrcg

AP: BBRC