0021-972X/01/$03.00/0 The Journal of Clinical Endocrinology & Metabolism Copyright © 2001 by The Endocrine Society
Vol. 86, No. 6 Printed in U.S.A.
Naturally Occurring Mutations in the Melanocortin Receptor 3 Gene Are Not Associated with Type 2 Diabetes Mellitus in French Caucasians* EL HABIB HANI, SOPHIE DUPONT, EMMANUELLE DURAND, CHRISTIAN DINA, SOPHIE GALLINA, IRA GANTZ, AND PHILIPPE FROGUEL Lille Institute of Biology-Centre National de la Recherche Scientifique 8090 and Lille Pasteur Institute (E.H.H., S.D., E.D., C.D., S.G., P.F.), 59000 Lille, France; and University of Michigan Medical School (I.G.), Ann Arbor, Michigan 48109 ABSTRACT Familial genetic studies of type 2 diabetes (T2DM) of different human populations, including the French Caucasians, suggested evidence for linkage of T2DM and human chromosome 20q13, a region where maps the melanocortin 3 receptor gene (MC3R). Likewise, its homologous MC4R in human obesity, MC3R gene is also a good candidate for genetic susceptibility to glucose intolerance and T2DM. We therefore undertook a molecular study to assess the role of genetic variations of this gene in a large cohort of French families with T2DM. In these patients, we identified two missense mutations in the MC3R gene: Val81Ile and Lys6Thr. These two variants, which were in com-
T
YPE 2 DIABETES mellitus (T2DM), formerly called noninsulin-dependent diabetes mellitus), is a very common metabolic disorder of glucose homeostasis, with an obvious but complex genetic predisposition (1). We originally reported significant evidence for linkage between T2DM and the human chromosome 20q13, near the phosphoenolpyruvate carboxykinase locus (PCK1), in French Caucasian families with T2DM (2, 3). Other groups have provided further support for a susceptibility locus for T2DM on this chromosomal region in other populations (4 – 6). In addition, a quantitative trait locus (QTL) for plasma insulin levels as well as for obesity-associated traits has been mapped on this same chromosomal region in Quebec families of French ancestry (7). These different indications of linkage between T2DM (or related quantitative traits) and this genomic region on 20q13 collectively suggest the presence of a putative gene with variants increasing the susceptibility to glucose intolerance and diabetes/obesity and possibly affecting the -pancreatic insulin secretory function. Many candidate genes may be considered in this diabeteslinked chromosomal region. Among these, the melanocortin 3 receptor gene (MC3R), a single exon-encoding gene with an open reading frame of 360 amino acids (1083 nucleotides, GenBank EMBL no. L06155) maps to this human chromosome 20q13 region (8, 9). The MC3R belongs to a family of Received October 4, 2000. Revision received February 26, 2001. Accepted March 2, 2001. Address all correspondence and requests for reprints to: Prof. Philippe Froguel, Institut Pasteur de Lille, 1 rue du Pr Calmette 59000 Lille, France. E-mail:
[email protected]. * This work was supported by grants from the Centre National de la Recherche Scientifique and the Re´gion Nord-Pas de Calais.
plete linkage disequilibrium, were also present in nondiabetic controls. Based on association and familial linkage disequilibrium tests results, we found that these MC3R gene-coding variants were not associated with diabetes or obesity. These variants were found, however, marginally associated with insulin and glucose levels during oral glucose tolerance testing in normoglycemic subjects. Overall, the present study provides no evidence for a major role of the MC3R coding mutations underlying the genetic linkages of T2DM and the MC3R gene region on chromosome 20q13 in T2DM families from France and other geographical origins. (J Clin Endocrinol Metab 86: 2895–2898, 2001)
seven transmembrane G protein-coupled receptors. The MC3R is expressed in brain, including hypothalamic centers involved in the regulation of feeding behavior (10), and in gastrointestinal tissues, including the pancreas (8). Like the melanocortin 4 receptor (MC4R) and other melanocortin receptors (MCRs), the MC3R recognizes the core melanocortin heptapeptide sequence (His-Phe-Arg-Trp), but is distinguished pharmacologically from other MCRs by its greater affinity for ␥MSH (8, 9). Two endogenous melanocortin receptor agonists have been identified, agouti and agoutirelated protein (AGRP) (11, 12). Agouti mutant mice that ectopically express agouti (13) or transgenic mice ubiquitously expressing AGRP (14) are phenotypically characterized by obesity and defects in carbohydrate metabolism via the competitive antagonism of the ␣MSH action. One major site of the action of agouti and AGRP in these transgenic rodent models is at the MC4R in hypothalamic feeding centers. Accordingly, Huszar et al. found that inactivation of the murine ortholog of MC4R by gene targeting in mice resulted in an obesity syndrome associated with marked carbohydrate metabolic defects (15). Similarly, defective mutations of the MC4R have been shown to cause rare extreme obesity in humans (16, 17). The MC3R has a notable structure similarity with the MC4R, and recently, like the MC4R, it has been demonstrated that agouti is also an antagonist of MC3R (18). Based on this structure/function similarity and on the MC4R mutation-induced metabolic phenotypes in animals and humans, it may be speculated that MC3R mutations can predispose to T2DM in humans or to energy metabolism-related defects. This idea is strongly suggested by a recent report describing an obesity-related phenotype associated with in-
2895
2896
JCE & M • 2001 Vol. 86 • No. 6
HANI ET AL.
creased fat mass along with defects in energy homeostasis in MC3R gene-deficient mice (19). Collectively, the chromosomal location of the MC3R locus (20q13) as well as its biology make it a plausible candidate gene for the indications of linkage observed between T2DM and the human chromosome 20q13. This candidacy is further prompted by the expression pattern of MC3R (pancreas and brain) (8) and the coincidence of a diabetogenic as well as an insulin-QTL on the chromosome 20q13 near the MC3R locus in humans from French and other geographical origins (2–7). Hence, the aim of the present study was to search for the presence of MC3R mutations in French type 2 diabetic families and to evaluate their association with T2DM or T2DMrelated traits. We therefore screened the MC3R gene in a large group of French families presenting with T2DM using PCR and direct DNA sequencing and assessed their relationship to the T2DM affection status and its related traits through association-based tests. Subjects and Methods Experimental subjects and cohorts The probands and families studied here are part of a large collection of French multiplex T2DM families recruited all over France through a multimedia-based campaign. Forty probands were initially selected (1 proband from each family) from the collection of families that showed evidence for linkage of T2DM with MC3R gene chromosomal region (3). The MC3R gene-encoding sequences were screened for mutations by PCR and DNA direct sequencing on both MC3R DNA strands in these probands. Identified variants were subsequently tested for association with diabetes or quantitatively related traits by comparing their frequencies in groups of 308 diabetic probands (each representing a single family) and in 218 nondiabetic nonobese healthy controls. All of the T2DM families studied here had no known etiology for diabetes. The clinical characteristics of the proband group, subjected to the primary direct gene screening, and the association test groups are summarized in Table 1.
using specific oligonucleotides: MC3R-For1 (5⬘-GAATAAACTCTAAGTCCAGATT-3⬘) and MC3R-Rev1 (5⬘-ATCTGGGTCTGCTGCGGCGTC-3⬘) for the first half of the coding sequence of the MC3R gene (fragment I) and MC3R-For2 (5⬘-CGTCTGTGGCGTGGTGTTC-3⬘) and MC3R-Rev2 (5⬘-TGAAAACTGAAAGGTAGGTGC-3⬘) for the second overlapping half of the gene (fragment II). PCR was carried out on 100 ng genomic DNA in a 50-L volume with 0.25 mol/L of each primer and 1.5 U AmpliTaq Gold (Perkin-Elmer Corp., Foster City, CA). For both fragments, PCR cycling consisted of an initial denaturation at 94 C for 12 min, followed by 30 cycles of a denaturation step at 94 C for 30 s, an annealing step at 57 C for 30 s, and an extension step at 72 C for 45 s, with a final extension step at 72 C for 10 min. The amplified fragments were screened for mutations on both strands using direct DNA sequencing optimized protocols as previously described (20). Sequence comparison analyses were independently scored by 2 readers to ensure maximal accuracy for mutation detection.
Association studies The Lys6Thr and Val81Ile variations identified in the 40 T2DM initially screened subjects were found in complete linkage disequilibrium, and therefore only 1 of these (the Val81Ile) variation was genotyped to test for association in larger cohorts (308 diabetics vs. 218 controls). The presence of the Val81Ile variant results in a gain of MnlI endonuclease restriction site polymorphism, and therefore, all subjects were screened for this substitution by PCR-restriction fragment length polymorphism. The DNA fragments were amplified using the same oligonucleotides as those described above, the amplified product was incubated and digested with 5 U MnlI according to the manufacturer’s instructions (New England Biolabs, Inc., Beverley, MA), and the restriction products were analyzed on 2% agarose gels and scored with ethidium bromide staining.
Statistics Genotype frequencies in the T2DM probands group (n ⫽ 308) and control groups (n ⫽ 218) were compared using 2 ratios calculated under dominant, codominant, and recessive models. We used ANOVA to test the association between the Val81Ile substitution and diabetic status, body mass index, and age at diagnosis of diabetes. ANOVA was also performed, in the control group exclusively, to test the influence of the Val81Ile variation (and of the Lys6Thr variant as well, due to complete linkage disequilibrium to the Val81Ile variation) on glucose and insulin levels during a standard oral glucose tolerance testing.
Mutational screening of MC3R In the initially selected group of 40 T2DM probands, the entire exon of the MC3R gene was amplified by PCR in 2 overlapping segments
TABLE 1. Clinical characteristics of the studied patients’ groups Characteristics
No. of probands Sex ratio (% men) Age at diagnosis (mean ⫾ SD) BMI (kg/m2; mean ⫾ SD) Fasting plasma glucose (mmol/L; mean ⫾ SD) Fasting plasma insulin (mU/L; mean ⫾ SD) Actual diabetes medication (%) Oral hypoglycemic agents (OHA) Insulin OHA ⫹ insulin Diet None
Primary mutational screening probands’ group
40 62.5 33.6 ⫾ 7.7
Type II diabetics group/control group
308/218 51.6/44.9 45 ⫾ 11.5/⫺
27.5 ⫾ 4.5 9.8 ⫾ 3.2
27.6 ⫾ 4.8/23.3 ⫾ 2.9 9.88 ⫾ 3.8/5.1 ⫾ 0.5
9.2 ⫾ 9
9.3 ⫾ 8.6/7.7 ⫾ 5.3
60
69.4
22.5 15 0 2.5
19.1 6.7 1.7 2.1
Results and Discussion
We identified two genetic variations in the MC3R gene upon screening of 40 French T2DM probands belonging to a set of families that have previously been reported to provide evidence for a genetic linkage between T2DM and the MC3R gene region on chromosome 20q13. These 2 genetic variants, C-to-A and G-to-A variations, result in a lysine to threonine (Lys6Thr) and a valine to isoleucine (Val81Ile) amino acids changes at codons 6 and 81 of the MC3R, respectively. MC3R protein is a seven-transmembrane receptor, and codons 6 and 81 are located in the N-terminal extracellular part of the protein and in the first transmembrane helix, respectively. We therefore speculated that these variations may have an effect on MC3R function and thus be associated with alterations in carbohydrate metabolism resulting in T2DM in the context of the genetic linkages in the vicinity of the MC3R locus on chromosome 20q. In the initially screened group of 40 probands, the 2 variations were found in complete linkage disequilibrium; consequently, we performed association and family-based linkage tests only with the Val81Ile substitution. The G/A mutation at codon 81 and consequently, due to complete
MUTATIONS OF MC3R GENE IN TYPE 2 DIABETES TABLE 2. Glucose tolerance profiles comparisons between carriers and noncarriers of the MC3R Lys6Thr and Val81Ile variants among nondiabetic control subjects Clinical parameters (mean ⫾ SD)
Plasma glucose 0 min 30 min 60 min 90 min 120 min Plasma insulin 0 min 30 min 60 min 90 min 120 min
Allele variant carriers (n ⫽ 39)
Wild-type allele carries (n ⫽ 179)
5.26 ⫾ 0.5 8.33 ⫾ 1.6 7.34 ⫾ 1.8 6.21 ⫾ 0.9 5.99 ⫾ 0.9
5.07 ⫾ 0.5 8.49 ⫾ 1.6 7.62 ⫾ 1.9 6.07 ⫾ 1.6 5.29 ⫾ 1.3
10.37 ⫾ 6.6 69.10 ⫾ 43.7 76.15 ⫾ 45.0 60.92 ⫾ 43.0 37.30 ⫾ 30.6
7.24 ⫾ 4.9 31.44 ⫾ 15.8 42.2 ⫾ 26.4 30.82 ⫾ 21.2 36.32 ⫾ 36.6
P value
0.04 0.79 0.69 0.80 0.11 0.003 0.0002 0.009 0.008 0.90
P values are provided based on ANOVA comparisons between wild-type genotype carriers and the variant genotype carriers.
linkage disequilibrium, the C/A mutation at codon 6 as well were also found in normoglycemic control subjects. The allelic frequencies of the Val81Ile and the Lys6Thr variants were not significantly different when comparing the T2DM group (n ⫽ 308) and the nondiabetics (n ⫽ 218). These were 0.08 and 0.10 in the diabetic and control groups, respectively (genotype frequencies, 12.66% vs. 16.51% of heterozygotes and 1.62% vs. 1.83% of mutated homozygotes were found in the diabetic and control groups, respectively). No significant correlation was found between the Val81Ile polymorphism with diabetes neither body mass index or age at onset of the disease (data not shown). Moreover, sib-TDT analysis results (21) gave no evidence for linkage disequilibrium of these variants with the affected status (P ⫽ 0.35; data not shown). This is not surprising, as the number of informative families for such analysis was as low as 19, which does not guarantee satisfactory power in the context of a susceptibility gene with reduced penetrance and possible high phenocopy rate. Collectively, these results indicate that both variations are unlikely to be the diabetogenic mutations underlying the observed genetic linkage between diabetes and chromosome 20q13 near the MC3R locus (2–7) and exclude MC3R as a T2DM susceptibility gene in this population. However, it has been reported in the Quebec-based population that a QTL affecting insulin levels maps to the chromosomal region of the MC3R locus (7). Thus, it may be possible that these variations affect some subtle phenotypes related to diabetes, such as plasma insulin and glucose levels during a glucose challenge. When we assessed the effects of these variations in the control nondiabetic group with respect to glycemia-related metabolic profiles, we found that the carriers of the rare A allele (i.e. carriers of the Lys6Thr and Val81Ile variations at codons 6 and 81, respectively) had slightly higher fasting plasma glucose (5.26 vs. 5.07 mmol/L; P ⫽ 0.036) and insulin levels (10.4 vs. 7.2 mU/L; P ⫽ 0.003) and a significantly higher fasting and 30-min oral glucose tolerance test insulin levels (69.1 vs. 31.4 mU/L; P ⫽ 0.0002) compared with noncarriers (Table 2). This result, suggesting a difference between carriers and noncarriers of the investigated MC3R gene variants among nondiabetic subjects in terms of insulin secretion function, should be cautiously
2897
considered, because the number of variant allele carriers is quite small (n ⫽ 39) and can only correspond to a spurious association. Association tests in larger samples are needed to formally conclude in favor of this association. In this regard, the Val81Ile has recently been reported in U.S. Caucasians by Li and colleagues, who found no association of this polymorphism with obesity in cohorts of extremely obese Caucasian and Afro-American women (22). It is noteworthy that, unlike the MC4R gene, no frameshift mutation has been identified in MC3R in humans to date. In conclusion, it appears unlikely that the MC3R gene variants described here have functional effects that would significantly contribute to the metabolic defects underlying the development of T2DM in humans. This study suggests that coding mutations MC3R variants are not major contributors to the genetic linkages to T2DM previously observed near the MC3R locus in T2DM families, although this does not fully exclude the involvement of the MC3R gene in the pathogenesis of T2DM or obesity due to other defects, for instance in the regulatory regions of the gene, or to other mutations in populations from other ethnic origins. Acknowledgments We thank type 2 diabetic patient volunteers and their families for their participation in the study. We are indebted to P. Gallina for his valuable efforts in the recruitment of diabetic families and their members for the studies, and to L. Zekiri and M. Caudrelier for the secretarial assistance.
References 1. DeFronzo RA. 1997 Pathogenesis of type 2 diabetes: metabolic and molecular implications for identifying diabetes genes. Diabetes Rev. 5:177–269. 2. Hani EH, Zouali H, Philippi A, et al. 1996 Indication for genetic linkage of the phosphoenolpyruvate carboxykinase (PCK1) gene region on chromosome 20q to non insulin-dependent diabetes mellitus. Diabetes Metab. 22:451– 454. 3. Zouali H, Hani EH, Philippi A, et al. 1997 A susceptibility locus for early-onset non insulin-dependent (type 2) diabetes mellitus maps to chromosome 20q, proximal to the phosphoenolpyruvate carboxykinase gene. Hum Mol Genet. 9:1401–1408. 4. Bowden DW, Sale M, Howard TD, et al. 1997 Linkage of genetic markers on human chromosomes 20 and 12 to NIDDM in Caucasian sib pairs with a history of diabetic nephropathy. Diabetes. 46:882– 886. 5. Ji L, Malecki M, Warram JH, Yang Y, Rich SS, Krolewski AS. 1997 New susceptibility locus for NIDDM is localized to human chromosome 20q. Diabetes. 46:876 – 881. 6. Ghosh S, Watanabe RM, Hauser ER, et al. 1999 Type 2 diabetes: Evidence for linkage on chromosome 20 in 716 Finnish affected sib pairs. Proc Natl Acad Sci USA. 96:2198 –2203. 7. Lembertas AV, Perusse L, Chagnon YC, et al. 1997 Identification of an obesity quantitative trait locus on mouse chromosome 2 and evidence of linkage to body fat and insulin on the human homologous region 20q. J Clin Invest. 100:1240 –1247. 8. Gantz I, Konda Y, Tashiro T, et al. 1993 Molecular cloning of a novel melanocortin receptor. J Biol Chem. 268:8246 – 8250. 9. Gantz I, Tashiro T, Barcroft C, et al. 1993 Localization of the genes encoding the melanocortin-2 (adrenocorticotropic hormone) and melanocortin-3 receptors to chromosomes 18p11.2 and 20q13.2-q13.3 by fluorescence in situ hybridization. Genomics. 18:166 –167. 10. Roselli-Rehfuss L, Mountjoy KG, Robbins LS, et al. 1993 Identification of a receptor for ␥melanotropin and other proopiomelanocortin peptides in the hypothalamus and limbic system. Proc Natl Acad Sci USA. 90:8856 – 8860. 11. Lu D, Willard D, Patel IR, et al. 1994 Agouti protein is an antagonist of the melanocyte-stimulating-hormone receptor. Nature. 371:799 – 802. 12. Ollmann MM, Wilson BD, Yang YK, et al. 1997 Antagonism of central melanocortin receptors in vitro and in vivo by agouti-related protein. Science. 278:135–138. 13. Yen TT, Gill AM, Frigeri LG, Barsh GS, Wolff GL. 1994 Obesity, diabetes, and neoplasia in yellow A(vy)/⫺ mice: ectopic expression of the agouti gene. FASEB J. 8:479 – 488. 14. Graham M, Shutter JR, Sarmiento U, Sarosi I, Stark KL. 1997 Overexpression of Agrt leads to obesity in transgenic mice. Nat Genet. 17:273–274.
2898
HANI ET AL.
15. Huszar D, Lynch CA, Fairchild-Huntress V, et al. 1997 Targeted disruption of the melanocortin-4 receptor results in obesity in mice. Cell. 88:131–141. 16. Vaisse C, Clement K, Guy-Grand B, Froguel P. 1998 A frameshift mutation in human MC4R is associated with a dominant form of obesity. Nat Genet. 20:113–114. 17. Yeo GS, Farooqi IS, Aminian S, Halsall DJ, Stanhope RG, O’Rahilly S. 1998 A frameshift mutation in MC4R associated with dominantly inherited human obesity. Nat Genet. 20:111–112. 18. Yang YK, Thompson DA, Dickinson CJ, et al. 1999 Characterization of agoutirelated protein binding to melanocortin receptors. Mol Endocrinol. 13:148 –155.
JCE & M • 2001 Vol. 86 • No. 6
19. Chen AS, Marsh DJ, Trumbauer ME, et al. 2000 Inactivation of the mouse melanocortin-3 receptor results in increased fat mass and reduced lean body mass. Nat Genet. 26:97–102. 20. Boutin P, Wahl C, Samson C, et al. 2000 Big Dye terminator cycle sequencing chemistry: accuracy of the dilution process and application for screening mutations in the TCF1 and GCK genes. Hum Mutat. 15:201–203. 21. Horvath S, Laird NM. 1998 A discordant-sibship test for disequilibrium and linkage: no need for parental data. Am J Hum Genet. 63:1886 –1897. 22. Li WD, Joo EJ, Furlong EB, et al. 2000 Melanocortin 3 receptor (MC3R) gene variants in extremely obese women. Int J Obes Relat Metab Disord. 24:206 –210.
