Original Article Variants in ARHGEF11, a Candidate Gene for the Linkage to Type 2 Diabetes on Chromosome 1q, Are Nominally Associated With Insulin Resistance and Type 2 Diabetes in Pima Indians Lijun Ma, 1 Robert L. Hanson, 1 Lorem N. Que, 1 Anna M.G. Cali, 1 Mao Fu, 2 Janel L. Mack, 1 Aniello M. Infante, 1 Sayuko Kobes, 1 the International Type 2 Diabetes 1q Consortium, Clifton Bogardus, 1 Alan R. Shuldiner, 2 and Leslie J. Baier 1 A prior genome-wide linkage scan in Pima Indians indi- cated a young-onset (aged <45 years) type 2 diabetes susceptibility locus on chromosome 1q21-q23. ARHGEF11, which encodes the Rho guanine nucleotide exchange factor 11, was analyzed as a positional candidate gene for this linkage because this protein may stimulate Rho-dependent signals, such as the insulin signaling cascade. The ARH- GEF11 gene, and two adjacent genes NTRK1 and INSRR, were sequenced in 24 Pima Indians who were not first- degree relatives. Sequencing of the coding regions, 5� and 3� untranslated regions and putative promoter regions of these genes, identified 28 variants in ARHGEF11, 11 vari- ants in NTRK1, and 8 variants in INSSR. These 47 variants, as well as 84 additional public database variants within/ between these genes, were genotyped for association anal- ysis in the same group of Pima Indians who had participated in the linkage study (n � 1,228). An R1467H in ARHGEF11, and several additional noncoding variants that were in high linkage disequilibrium with this variant, were nominally associated with young-onset type 2 diabe- tes (P � 0.01; odds ratio 3.39) after adjusting for sex, family membership, and Pima heritage. The risk allele H had a frequency of 0.10. In a subgroup of 262 nondiabetic, full-heritage Pima Indians who had undergone detailed metabolic testing, the risk allele H also was associated with a lower mean insulin-mediated glucose disposal rate and a lower mean nonoxidative glucose storage rate after adjust- ing for age, sex, nuclear family membership, and percent- age of body fat (P < 0.01). These findings suggest that variation within ARHGEF11 nominally increases risk of type 2 diabetes, possibly as a result of increased insulin resistance. Diabetes 56:1454–1459, 2007 T he Pima Indians of Arizona have an extremely high prevalence of type 2 diabetes (1). Their diabetes is characterized by obesity, dysfunction of insulin secretion, insulin resistance (de- creased insulin-mediated glucose disposal), and increased rates of postabsorptive endogenous glucose output (2). Studies (3–5) have shown that type 2 diabetes, insulin resistance, the acute insulin response, and obesity are highly heritable in this population. A prior genome-wide linkage scan in Pima Indians indicated a susceptibility locus for young-onset (onset age �45 years) type 2 diabetes on chromosome 1q21-q23 at D1S1677 (6). Linkage to type 2 diabetes on chromosome 1q21–23 has subsequently been observed in seven other populations who have formed the International Type 2 Diabetes 1q Consortium (7–14). Within this region of linkage, there is a cluster of three genes (ARHGEF11, NTRK1, and INSRR) located between 153 and 154 Mb, which encode proteins that have putative roles in the insulin signaling system. ARHGEF11 encodes the Rho guanine nucleotide exchange factor ARHGEF11 (also called PDZ-RhoGEF and KIAA0380). ARHGEF11 interacts with small GTPases (G-protein, guanine nucleotide-bind- ing proteins), such as Rho, that function as molecular switches in signaling pathways that include the insulin signaling cascade (15,16). NTRK1 encodes the neurotro- phic tyrosine kinase receptor type 1, which recruits insulin receptor substrate (IRS)-1 and IRS-2 (17), and INSRR encodes the insulin receptor–related receptor, which po- tentially phosphorylates IRS-1 and IRS-2 (available at http://genecards.bcgsc.bc.ca/cgi-bin/carddisp?insrr). In this study, ARHGEF11 and the two adjacent genes, NTRK1 and INSRR, were analyzed as positional candidate genes for type 2 diabetes in the Pima Indians. RESEARCH DESIGN AND METHODS The subjects who were sequenced (n � 24) and genotyped (n � 1,228) were participants in our prior genome-wide linkage study for diabetes susceptibility loci in Pima Indians (6) and are part of our ongoing longitudinal study of the etiology of type 2 diabetes among the Gila River Indian Community in central Arizona (1). All individuals in the longitudinal study are invited to participate in a standardized health examination biannually. To determine diabetes From the 1Diabetes Molecular Genetics Section, Phoenix Epidemiology and Clinical Research Branch, National Institute of Diabetes and Digestive and Kidney Diseases, National Institutes of Health, Department of Health and Human Services, Phoenix, Arizona; and the 2Division of Endocrinology, Diabetes and Nutrition, University of Maryland School of Medicine, Baltimore, Maryland. Address correspondence and reprint requests to Leslie J. Baier, PhD, Diabetes Molecular Genetics Section, Phoenix Epidemiology and Clinical Research Branch, National Institute of Diabetes and Digestive and Kidney Diseases, National Institutes of Health, 445 N. Fifth St., Suite 210, Phoenix, AZ 85004. E-mail: lbaier@phx.niddk.nih.gov. Received for publication 9 May 2006 and accepted in revised form 22 January 2007. Published ahead of print at http://diabetes.diabetesjournals.org on 29 Jan- uary 2007. DOI: 10.2337/db06-0640. Additional information for this article can be found in an online appendix at http://dx.doi.org/10.2337/db06-0640. IRS, insulin receptor substrate; LD, linkage disequilibrium; SNP, single nucleotide polymorphism. © 2007 by the American Diabetes Association. The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked “advertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact. 1454 DIABETES, VOL. 56, MAY 2007 status, a 75-g orally administered glucose tolerance test is given and the results are interpreted according to the criteria of the World Health Organi- zation (18). All studies were approved by the tribal council of the Gila River Indian Council and the institutional review board of the National Institutes of Diabetes and Digestive and Kidney Diseases. Among 1,228 subjects that were genotyped, 262 of the nondiabetic subjects had additionally been studied for measures of pre-diabetic phenotypes in our clinical research center. Only individuals (aged 18 – 45 years) who are con- firmed to be healthy by medical history, physical examination, and routine laboratory tests and are not taking medications are studied. Oral glucose tolerance is measured after 2–3 days on a weight-maintaining diet of mixed composition. Blood for plasma glucose and insulin measurements is drawn before ingesting 75 g of glucose and at 30, 60, 120, and 180 min thereafter. Subjects also receive a 25-g intravenous injection of glucose over 3 min to measure the acute insulin response. Blood samples were collected before infusion and at 3, 4, 5, 6, 8, and 10 min after infusion for determination of plasma glucose and insulin concentrations. The acute insulin response was calculated as the mean increment in plasma insulin concentrations from 3 to 5 min (19). The hyperinsulinemic-euglycemic clamp technique was used to determine basal glucose appearance and insulin-stimulated glucose disappear- ance (uptake) rates (19). Briefly, insulin was infused to achieve physiologic plasma insulin concentrations (137 � 3 �U/ml) for 100 min. Plasma glucose concentrations were held constant at 100 mg/dl by a variable 20% glucose infusion. Tritiated glucose was infused for 2 h before the insulin infusion to calculate glucose disappearance rates during the insulin infusion. During the last 40 min of the insulin infusion, the rate of insulin-stimulated glucose disposal was calculated, adjusted for steady-state plasma glucose and insulin concentration, and normalized to estimated metabolic body size (fat-free mass plus 17.7 kg) as described (19,20). Ventilated-hood indirect calorimetry was used to estimate rates of glucose and lipid oxidation before and during the insulin infusions (20). Glucose and lipid oxidation rates were normalized to estimated metabolic body size (fat-free mass plus 17.7) as described (19,20). Body composition was estimated by underwater weighing until January 1996 and currently is measured by dual-energy X-ray absorptiometry (DPX-1; Lunar Radiation, Madison, WI). A conversion equation derived from comparative analyses is used to make estimates of body composition comparable between methods (21). Single nucleotide polymorphism identification and genotyping. To iden- tify sequence variants, all exons, exon-intron boundaries, 5� and 3� untrans- lated regions, and 2 kb of the putative promoter regions of ARHGEF11, NTRK1, and INSRR were PCR amplified and sequenced in DNA samples from 24 non–first-degree–related Pima Indians. Among 24 subjects, 12 developed diabetes before the age of 25 years and 12 were nondiabetic and were at least 45 years of age. Sequencing was performed using the Big Dye Terminator (Applied Biosystems) on an automated DNA capillary sequencer (model 3730xl; Applied Biosystems). Single nucleotide polymorphisms (SNPs) iden- tified by sequencing were genotyped in DNA from 1,228 subjects using the TaqMan Allelic Discrimination Assay (Applied Biosystems), direct sequencing (as described above; Applied Biosystems), or SNPplex (Applied Biosystems) following the manufacture’s protocol. Sequence information for all oligonu- cleotide primers and probes is available from the authors upon request. Statistical analysis. For the association analysis with young-onset type 2 diabetes, only siblings who were confirmed to have developed type 2 diabetes before the age of 45 years (n � 525) or confirmed to be nondiabetic and at least 45 years of age (n � 88) were analyzed. Statistical analyses were performed using the software of the SAS Institute (Cary, NC). Numeric variables are expressed as means � SE. The association of genotypes with diabetes was assessed by logistic regression analysis. For continuous vari- ables, linear regression analysis was used. Both linear and logistic models were fit with the general estimating equation procedure because some subjects were siblings. This procedure accounts for family membership by modeling the phenotypic variance/covariance matrix among siblings. For the present analyses, an exchangeable correlation matrix was assumed and the “empirical” SE, which is robust to the assumed correlation structure, was used. Plasma insulin concentrations and glucose disposal rate during the physi- ological plasma concentration were log transformed before analyses to approximate a normal distribution. For analysis under an additive model, homozygotes for the major allele (1/1) and heterozygotes (1/2) and homozy- gotes for the minor allele (2/2) were coded to a continuous numeric variable for genotype (as 0, 1, and 2). The recessive model was defined as contrasting genotypic groups 1/1 vs. 1/2 � 2/2, where allele 1 is defined as the major allele. Although the general estimating equation approach accounts for the correla- tion among family members, it does not provide a specific within-family test of association and, thus, still is potentially confounded by population strati- fication. Therefore, data also were analyzed using a modification of the method of Abecasis et al. (22) to test for within-family association. In brief, this method partitions the association into between- and within-family com- ponents, represented, respectively, by the sibship mean of the continuous numeric variable for genotype and each individual’s deviation from this mean. The test of the significance of the within-family components is a test of cotransmission among siblings, which is robust to population stratification (although it is less powerful than the more general test). Analyses of all traits were adjusted for potentially confounding variables (e.g., age, sex, percentage of body fat); all adjustments were specified a priori based on previous studies of the determinants of these traits. To examine pairwise linkage disequilib- rium (LD), haplotype frequencies were estimated with the estimating haplo- type (EH) program (Xie and Ott) (available at http://linkage.rockefeller.edu/ ott/), and these haplotype frequencies were used to calculate D� and � (2). P values of �0.05 were considered to be of nominal statistical significance. RESULTS Sequencing and genotyping of ARHGEF11. Sequenc- ing of the 41 exons, exon-intron boundary regions, and 2 kb of the 5� (putative promoter) region of ARHGEF11 in 24 non–first-degree–related Pima Indians identified 28 variants, 3 of which were nonsynonymous amino acid substitutions (R293Q, S1456G, and R1467H). These 28 variants, and 66 additional database SNPs positioned outside of the regions that were sequenced, were geno- typed for association analysis in the same Pima Indian subjects who were used for our prior type 2 diabetes linkage study (online appendix Table 1 [available at http:// dx.doi.org/10.2337/db06-0640]). Three coding variants in ARHGEF11 were nominally associated with young-onset type 2 diabetes. The H allele (frequency � 0.10) of R1467H was associated with young-onset type 2 diabetes (P � 0.01; odds ratio 3.4 [95% CI 1.29 – 8.93], recessive model) after adjusting for sex, family membership, and Pima heritage (Fig. 1; Table 1). The G allele (frequency � 0.26) of S1456G also was associated with young-onset type 2 diabetes (P � 0.005, 1.85 [1.20 –2.85], additive model; P � 0.012, 1.93 [1.16 –3.22], recessive model) after adjusting for sex, fam- ily membership, and Pima heritage (Fig. 1; Table 1). The Q allele of R293Q was extremely rare (frequency � 0.002), and although the association data for this variant are included in Table 1, these data may not be reliable due to the limited sample power. Thirteen noncoding SNPs also were nominally associated with young-onset type 2 diabe- tes (listed in Table 1 legend). All of these noncoding SNPs, with the exception of rs3753213 (shown in Table 1), were in high LD (D� � 0.99, r2 � 0.99) with either R1467H or S1456G (online appendix Fig. 3). We further assessed whether the R1467H and S1456G were associated with metabolic traits predictive of type 2 diabetes in Pima Indians who had not yet developed diabetes. In a subgroup of 262 nondiabetic, full-heritage Pima Indians who have undergone detailed metabolic testing, including measurements of body composition, oral glucose tolerance, insulin secretory function, insulin-stim- ulated glucose uptake, and indirect calorimetry, the risk allele (H) for R1467H was nominally associated with a lower mean glucose disposal rate (P � 0.005) and a lower mean nonoxidative glucose storage rate (P � 0.01) during a euglycemic-hyperinsulinemic clamp under physiologic concentrations of plasma insulin infusion, after adjusting for age, sex, nuclear family membership, and percentage of body fat (Table 2; Fig. 2). The R1467H was not associ- ated with insulin secretion as assessed by the acute insulin response to a 25-g intravenous bolus injection of glucose or the insulin level at 30 min during an oral glucose tolerance test (Table 2). No association was found be- tween S1456G and metabolic predictors of type 2 diabetes L. MA AND ASSOCIATES DIABETES, VOL. 56, MAY 2007 1455 in nondiabetic, full-heritage Pima subjects (data not shown). Sequencing and genotyping of NTRK1 and INSRR. Direct sequencing of all exons, exon-intron boundary regions of NTRK1 and INSRR, and 2.5 kb of the common 5� flanking regulatory region between these two genes in 24 non–first-degree–related Pima Indians identified 19 variants (11 in NTRK1 and 8 in INSRR; online appendix Table 1). Four of these variants predicted nonsynonymous amino acid changes (Phe316Leu, Arg869Leu, Pro928Leu, and Arg999STOP) located in INSRR. None of these coding variants were detected when INSRR previously was screened in Pima Indians using the technique of denatur- ing high-performance liquid chromatography on pooled DNA samples (23). The 19 variants, and 18 additional database SNPs positioned outside of the regions that were sequenced, were genotyped for association analysis in the same Pima Indian subjects who were used for the type 2 diabetes linkage study. None of the amino acid changes were associated with young-onset type 2 diabetes or metabolic predictors of type 2 diabetes in Pima Indians. Only SNP rs3753213, located in the shared common 5� flanking regulatory region (2.5 kb) between INSRR and NTRK1, showed a nominal association with young-onset FIG. 1. Association plot of the �log P values (additive model) for SNPs with young-onset type 2 diabetes versus position along a 300-kb region of chromosome 1. The relative positions of exons for INSRR, NTRK1, and ARHGEF11 within this region are shown beneath the plot. TABLE 1 Representative* SNPs associated with young-onset type 2 diabetes in Pima Indians SNP SNP (1/2) Gene locus Minor allele frequency Type 2 diabetes by genotype (n)† Nondiabetic by genotype (n)‡ Association in general (upper) and within-family (lower, italics) models 11 12 � 22 11 12 � 22 P value Odds ratio rs3753213 C/T INSRR/NTRK1 0.34 233 273 28 59 0.04 0.52 (0.27–0.96) Intergenic 0.18 0.55 (0.23–1.32) R1467H G/A ARHGEF11 0.10 422 102 82 6 0.01§ 3.39 (1.29–8.93) Exon 39 0.03§ 5.09 (1.20–21.55) S1456G A/G ARHGEF11 0.26 256 253 59 29 0.01 1.93 (1.16–3.22) Exon 39 0.24 1.54 (0.76–3.18) R293Q G/A ARHGEF11 0.002 523 2 86 2 0.04 0.18 (0.03–0.93) EXON 11 NA NA SNP alleles are given as 1 � major and 2 � minor. The risk allele is given in bold. Due to the low frequencies of these SNPs, only the recessive analytical model is given, which is defined as comparing genotype groups 1/1 vs. 1/2 � 2/2. P values/odds ratios were adjusted for sex, family membership, and Pima heritage. Odds ratios were calculated by defining allele 2 as the defaulted risk factor. *R1467H is in nearly perfect LD (D� � 0.99, r2 � 0.99) with rs865239, rs822430, rs1336149, rs1572407, rs6664744, rs822577, rs822578, rs822579, hcv27852229, and rs4661014. S1456G is in nearly perfect LD with rs1336150 and rs4661077. †Subjects with type 2 diabetes diagnosed before the age of 45 years. ‡Subjects determined to be nondiabetic by oral glucose tolerance test after the age of 45 years. §The empirical P values obtained with 5,000 permutations were 0.011 for the general association test and P � 0.038 for the within-family test, respectively, which were very similar to the nominal values. ARHGEF11, TYPE 2 DIABETES, AND PIMA INDIANS 1456 DIABETES, VOL. 56, MAY 2007 type 2 diabetes (P � 0.04; Table 1; Fig. 1), but this variant was not associated with any metabolic predictors of type 2 diabetes in nondiabetic, full-heritage Pima Indians (data not shown). The variants detected in INSRR and NTRK1 are in a distinct LD block compared with variants in ARHGEF11 in Pima Indians (online appendix Fig. 3). Haplotype analyses for ARHGEF11 and INSRR. To detect an effect of multiple, potentially functional SNPs on the development of type 2 diabetes, haplotype analyses were done by combining the Arg1467His and Ser1456Gly in ARHGEF11 and the Arg999STOP and Pro928Leu in INSRR. However, neither haplotype analysis provided additional information to the single SNP analyses (Table 3). The Arg193Gln in ARGEF11 and the Phe316Leu and Arg869Leu in INSRR were not included in the haplotypes due to their extremely rare allele frequencies. DISCUSSION Our data suggest that the H allele at R1467H in ARHGEF11 (or a variant that is carried on the H allele in Pima Indians) increases risk of type 2 diabetes by reducing insulin- stimulated glucose uptake. Among the nondiabetic, full- heritage Pima Indians, subjects with an H allele (homozygotes and heterozygotes) had a lower mean glu- cose disposal rate during a hyperinsulinemic-euglycemic clamp after adjusting for age, sex, nuclear family member- ship, and percentage of body fat. This lower glucose disposal rate mainly was attributed to reduced nonoxida- tive glucose storage because the other component of glucose disposal rate, glucose oxidation, was similar be- tween subjects with an H allele and subjects who were homozygous for the R allele. These findings indicate that ARHGEF11 may alter insulin-mediated glucose storage pathways, such as glycogen synthesis in the muscle and liver. ARHGEF11 is ubiquitously expressed and is present in both liver and muscle, which are important organs for FIG. 2. Comparison of glucose disposal rates in nondiabetic, full- heritage Pima Indians grouped by R1467H genotypes. Insulin-mediated glucose disposal rate is measured during a euglycemic-hyperinsuline- mic clamp during infusion of physiological concentrations of plasma insulin and is composed of nonoxidative glucose storage and glucose oxidation. f, glucose oxidation; u, nonoxidative glucose storage. TABLE 2 Mean values for insulin action and insulin secretion among nondiabetic, full-heritage Pima Indians grouped by ARHGEF11 R1467H genotypes R1467R R1467H P value* n (male/female) 208 (121/87) 54 (35/19) 0.44 Age (years) 27 � 0.4 26 � 0.8 — Percentage of body fat 33 � 0.6 31 � 1.3 0.08 Glucose disposal rate (mg � min�1 � kg EMBS�1) 3.61 � 0.08 3.47 � 0.14 0.005 Nonoxidative glucose storage (mg � min�1 � kg EMBS�1) 1.60 � 0.07 1.47 � 0.11 0.01 Glucose oxidation (mg � min�1 � kg EMBS�1) 2.02 � 0.04 2.01 � 0.07 0.42 Oral glucose tolerance test Fasting glucose (mg/dl) 91 � 0.6 92 � 1.3 0.62 2-h glucose (mg/dl) 128 � 2.1 125 � 4.2 0.95 log10 fasting insulin (mIU/ml) 1.56 � 0.02 1.55 � 0.03 0.20 log10 2-h insulin (mIU/ml) 2.21 � 0.02 2.20 � 0.05 0.40 log10 acute insulin response (mIu/ml) (n � 138 normal glucose tolerance) 2.32 � 0.02 2.35 � 0.04 0.69 log10 insulin 30� during oral glucose tolerance test (mIU/ml) (n � 138 normal glucose tolerance) 2.33 � 0.02 2.33 � 0.05 0.99 Data are means � SE. Among this subgroup there were no subjects homozygous for H1467H. Glucose disposal rate is measured as insulin-stimulated glucose uptake during a hyperinsulinemic-euglycemic clamp at the physiologic plasma insulin concentrations. Estimated metabolic body size (EMBS) equals fat-free mass plus 17.7 kg. *P value for sex distribution between two groups was calculated by �2. P value for percentage of body fat was assessed using the general estimating equation procedure after adjusting for age, sex, and family membership. P values for glucose disposal rate, oral glucose tolerance glucose, and insulin levels were obtained using the general estimating equation procedure after adjusting for age, sex, family membership, and percentage of body fat. P values for log-transformed acute insulin response was assessed in 138 normal glucose-tolerant full-heritage Pima subjects after adjusting for age, sex, percentage of body fat, family membership, and insulin action. P values for log-transformed insulin level at 30 min during an oral glucose tolerance test were obtained after adjusting for age, sex, percentage of body fat, family membership, insulin action, and glucose level at 30 min during oral glucose tolerance tests. L. MA AND ASSOCIATES DIABETES, VOL. 56, MAY 2007 1457 insulin-mediated glucose metabolism (24). ARHGEF11 is a member of a subfamily of RhoGEFs that contain regulator of G-protein signaling domains. LARG (Leukemia-associ- ated Rho guanine nucleotide exchange factor; also called ARHGEF12) also is a member of this regulator of G- protein signaling subfamily. We previously studied LARG as a candidate gene for type 2 diabetes and identified a functional Tyr1306Cys variant that also was associated with reduced insulin-stimulated glucose uptake among nondiabetic Pima Indians (16). Although nominally significant associations were ob- served in our analysis of ARHGEF11, it should be cau- tioned that multiple comparisons inherent to association studies can lead to the discovery of false-positives, and for any given variant the prior probability of a true association is probably low. For these two reasons, particularly strin- gent P value thresholds for declaring a significant associ- ation have been proposed (25,26). The P values given in this study are unadjusted, and, if corrected for the 131 variants studied (which could be argued to be an overcor- rection due to high LD among many of the variants), the P values of �0.01 certainly would not be significant (P � �0.5). Yet, the performance of a correction for multiple comparison proceeds on the assumption that the “global” null hypothesis (no association with any of the variants) is of interest, and it is not clear that this is appropriate for a physiologic candidate gene in a region with replicated evidence for linkage. In this context, its rank relative to other SNPs genotyped as part of our fine-mapping efforts across this region of linkage, along with potential func- tional consequences of the variant, may provide more relevant information about the importance of this gene. The P value of 0.01 for R1467H puts this variant within the top 2.5% of the 4,500 variants genotyped to date in Pima Indians across a 37-Mb region on chromosome 1q. The R1467H also is located in the COOH-terminal region of ARHGEF11 protein, which has been shown to regulate ARHGEF11 protein activity (27). Replication of associations in other populations also can provide evidence that associations are not spurious. The region spanning ARHGEF11, NTRK1, and INSRR also has been sequenced and densely genotyped in the Amish population, where variants within ARHGEF11 also were found to be associated with type 2 diabetes and oral glucose levels in response to an oral glucose tolerance test. However, the specific variants in ARHGEF11 demon- strating the best associations differed between the Amish and the Pima populations (28). The chromosomal region spanning ARHGEF11 also has been genotyped at a �5-kb density as part of the fine-mapping efforts of the Interna- tional Type 2 Diabetes 1q Consortium (29), but among the 75 database variants (which did not include R1467H) genotyped in the consortium’s multiethnic subjects (n � 3,700) there was only weak evidence for an association with type 2 diabetes (lowest P value among 75 SNPs � 0.03). Therefore, we do not propose that the R1567H in ARHGEF11 explains the linkage to type 2 diabetes ob- served in multiple populations on chromosome 1q21–23. However, the R1567H, or another variant carried on the H allele of ARHGEF11, may have played a subtle role in our ability to detect linkage in this region in Pima Indians. Adjustment of the original linkage peak, centered in D1S1677, for the effect of R1467H results in a 12% reduc- tion in the evidence for linkage (logarithm of odds drops from 2.88 to 2.54), which is among the top 5% of the linkage reductions observed among 3,900 SNPs that have been analyzed, to date, within this region in Pima Indians. In summary, we propose that ARHGEF11 may have a minor, or modifying, role in influencing susceptibility to type 2 diabetes via its effect upon insulin action in Pima Indians. However, due to the relatively low frequency of risk allele H versus the high prevalence of type 2 diabetes in Pima Indians and the fact that this gene shows little evidence for association in other populations with linkage to type 2 diabetes on chromosome 1q21–23, except for the Amish, we speculate that there are additional type 2 diabetes susceptibility genes, with larger and more univer- sal effects, on chromosome 1q21-q23. ACKNOWLEDGMENTS This study was supported by the intramural research program of the National Institute of Diabetes and Diges- tive and Kidney Diseaes, National Institutes of Health (NIH). Funding was also provided by NIH Grant R01 DK073490. L.M. was supported by a mentor grant from the American Diabetes Association. REFERENCES 1. Knowler WC, Bennett PH, Hamman RF, Miller M: Diabetes incidence and prevalence in Pima Indians: a 19-fold greater incidence than in Rochester, Minnesota. Am J Epidemiol 108:497–505, 1978 2. Bogardus C: Metabolic abnormalities in the development of non-insulin dependent diabetes mellitus. In Diabetes Mellitus: A Fundamental and Clinical Text. LeRoith D, Taylor SI, Olefsky JM, Eds. Philadelphia, Lippincott-Raven, 1996, p. 459 – 467 3. Baier LJ, Hanson RL: Genetic studies of the etiology of type 2 diabetes in Pima Indians: hunting for pieces to a complicated puzzle. Diabetes 53:1181–1186, 2004 4. Ravussin E, Bogardus C: Energy expenditure in the obese: is there a thrifty gene? Infusionstherapie 17:108 –112, 1990 5. Sakul H, Pratley R, Cardon L, Ravussin E, Mott D, Bogardus C: Familiality of physical and metabolic characteristics that predict the development of TABLE 3 Haplotype analyses for AHRGEF11 (Arg1467His and Ser1456Gly) and INSRR (Arg999STOP and Pro928Leu) Haplotype Frequency (type 2 diabetes)* Frequency (non–type 2 diabetes)† Odds ratio (95% CI) (additive model) P value (additive model) Corrected P value 1467R-1456S 0.71 0.83 0.52 (0.32–0.86) 0.0106 0.0211 1467R-1456G 0.19 0.14 1.31 (0.80–2.16) 0.2870 0.4916 1467H-1456G 0.10 0.03 3.58 (1.28–10.03) 0.0151 0.0300 999R-928P 0.75 0.75 1.02 (0.66–1.59) 0.9295 0.9950 999R-928L 0.23 0.23 1.03 (0.66–1.62) 0.8975 0.9895 999STOP-928P 0.02 0.03 0.66 (0.24–1.95) 0.4815 0.7312 Corrected P is the P value corrected for multiple comparisons within haplotype combinations. The haplotype 1467H-1456S and 999STOP-928L were extremely rare and therefore were not included. *Subjects with type 2 diabetes diagnosed before the age of 45 years. †Subjects determined to be nondiabetic by oral glucose tolerance test after the age of 45 years. ARHGEF11, TYPE 2 DIABETES, AND PIMA INDIANS 1458 DIABETES, VOL. 56, MAY 2007 non-insulin-dependent diabetes mellitus in Pima Indians. Am J Hum Genet 60:651– 656, 1997 6. Hanson RL, Ehm MG, Pettitt DJ, Prochazka M, Thompson DB, Timberlake D, Foroud T, Kobes S, Baier L, Burns DK, Almasy L, Blangero J, Garvey WT, Bennett PH, Knowler WC: An autosomal genomic scan for loci linked to type II diabetes mellitus and body-mass index in Pima Indians. Am J Hum Genet 63:1130 –1138, 1998 7. Elbein SC, Hoffman MD, Teng K, Leppert MF, Hasstedt SJ: A genome-wide search for type 2 diabetes susceptibility genes in Utah Caucasians. Diabetes 48:1175–1182, 1999 8. Hsueh WC, St. Jean PL, Mitchell BD, Pollin TI, Knowler WC, Ehm MG, Bell CJ, Sakul H, Wagner MJ, Burns DK, Shuldiner AR: Genome-wide and fine-mapping linkage studies of type 2 diabetes and glucose traits in the Old Order Amish: evidence for a new diabetes locus on chromosome 14q11 and confirmation of a locus on chromosome 1q21– q24. Diabetes 52:550 – 557, 2003 9. Wiltshire S, Hattersley AT, Hitman GA, Walker M, Levy JC, Sampson M, O’Rahilly S, Frayling TM, Bell JI, Lathrop GM, Bennett A, Dhillon R, Fletcher C, Groves CJ, Jones E, Prestwich P, Simecek N, Rao PV, Wishart M, Bottazzo GF, Foxon R, Howell S, Smedley D, Cardon LR, Menzel S, McCarthy MI: A genome-wide scan for loci predisposing to type 2 diabetes in a U.K. population (the Diabetes UK Warren 2 Repository): analysis of 573 pedigrees provides independent replication of a susceptibility locus on chromosome 1q. Am J Hum Genet 69:553–569, 2001 10. Vionnet N, Hani El-H, Dupont S, Gallina S, Francke S, Dotte S, De Matos F, Durand E, Lepretre F, Lecoeur C, Gallina P, Zekiri L, Dina C, Froguel P: Genome-wide search for type 2 diabetes-susceptibility genes in French whites: evidence for a novel susceptibility locus for early-onset diabetes on chromosome 3q27-qter and independent replication of a type 2-diabetes locus on chromosome 1q21– q24. Am J Hum Genet 67:1470 –1480, 2000 11. Meigs JB, Panhuysen CI, Myers RH, Wilson PW, Cupples LA: A genome- wide scan for loci linked to plasma levels of glucose and HbA(1c) in a community-based sample of Caucasian pedigrees: the Framingham Off- spring Study. Diabetes 51:833–1840, 2002 12. Xiang K, Wang Y, Zheng T, Jia W, Li J, Chen L, Shen K, Wu S, Lin X, Zhang G, Wang C, Wang S, Lu H, Fang Q, Shi Y, Zhang R, Xu J, Weng Q: Genome-wide search for type 2 diabetes/impaired glucose homeostasis susceptibility genes in the Chinese: significant linkage to chromosome 6q21– q23 and chromosome 1q21– q24. Diabetes 53:228 –234, 2004 13. Ng MC, So WY, Lam VK, Cockram CS, Bell GI, Cox NJ, Chan JC: Genome-wide scan for metabolic syndrome and related quantitative traits in Hong Kong Chinese and confirmation of a susceptibility locus on chromosome 1q21– q25. Diabetes 53:2676 –2683, 2004 14. Ehm MG, Karnoub MC, Sakul H, Gottschalk K, Holt DC, Weber JL, Vaske D, Briley D, Briley L, Kopf J, McMillen P, Nguyen Q, Reisman M, Lai EH, Joslyn G, Shepherd NS, Bell C, Wagner MJ, Burns DK, the American Diabetes Association GENNID Study Group: Genetics of NIDDM: genome- wide search for type 2 diabetes susceptibility genes in four American populations. Am J Hum Genet 66:1871–1881, 2000 15. Ishizuka T, Cooper DR, Hernandez H, Buckley D, Standaert M, Farese RV: Effects of insulin on diacylglycerol-protein kinase C signaling in rat diaphragm and soleus muscles and relationship to glucose transport. Diabetes 39:181–190, 1990 16. Kovacs P, Stumvoll S, Bogardus C, Hanson RL, Baier LJ: A functional Tyr1306Cys variant in LARG is associated with increased insulin action in vivo. Diabetes 55:1497–1503, 2006 17. Miranda C, Greco A, Miele C, Pierotti MA, Van Obberghen E: IRS-1 and IRS-2 are recruited by TrkA receptor and oncogenic TRK-T1. J Cell Physiol 186:35– 46, 2001 18. World Health Organization: Diabetes Mellitus: Report of a WHO Study Group. Geneva, World Health Org., 1985 (Tech. rep. ser. no. 727). 19. Lillioja S, Mott DM, Spraul M, Ferraro R, Foley JE, Ravussin E, Knowler WC, Bennett PH, Bogardus C: Insulin resistance and insulin secretory dysfunction as precursors of non-insulin-dependent diabetes mellitus: prospective studies of Pima Indians. N Engl J Med 329:1988 –1992, 1993 20. Lillioja S, Mott DM, Howard BV, Bennett PH, Yki-Jarvinen H, Freymond D, Nyomba BL, Zurlo F, Swinburn B, Bogardus C: Impaired glucose tolerance as a disorder of insulin action: longitudinal and cross-sectional studies in Pima Indians. N Engl J Med 318:1217–1225,1988 21. Norman RA, Tataranni PA, Pratley R, Thompson DB, Hanson RL, Prochazka M, Baier L, Ehm MG, Sakul H, Foroud T, Garvey WT, Burns D, Knowler WC, Bennett PH, Bogardus C, Ravussin E: Autosomal genomic scan for loci linked to obesity and energy metabolism in Pima Indians. Am J Hum Genet 62:659 – 668, 1998 22. Abecasis GR, Cardon LR, Cookson WOC: A General test of association for quantitative traits in nuclear families. Am J Hum Genet 66:279 –292, 2000 23. Wolford JK, Thameem F, Bogardus C, Prochazka M: Polymorphism screening of the insulin receptor-related receptor gene (INSRR) on 1q in Pima Indians. Mol Cell Probes 15:223–227, 2001 24. Hirotani M, Ohoka Y, Yamamoto T, Nirasawa H, Furuyama T, Kogo M, Matsuya T, Inagaki S: Interaction of plexin-B1 with PDZ domain-containing Rho guanine nucleotide exchange factors. Biochem Biophys Res Commun 297:32–37, 2002 25. Colhoun HM, McKeigue PM, Smith GD: Problems of reporting genetic associations with complex outcomes. Lancet 361:865– 872, 2003 26. Risch N, Merikangas K: The future of genetic studies of complex human diseases. Science 273:1516 –1517, 1996 27. Chikumi H, Barac A, Behbahani B, Gao Y, Teramoto H, Zheng Y, Gutkind JS: Homo- and hetero-oligomerization of PDZ-RhoGEF, LARG and p115RhoGEF by their C-terminal region regulates their in vivo Rho GEF activity and transforming potential. Oncogene 23:233–240,2004 28. Fu M, Sabra MM, Damcott C, Pollin TI, Ma L, Ott S, Shelton JC, Shi X, Reinhart L, O’Connell J, Mitchell BD, Baier LJ, Shuldiner AR: Evidence that Rho guanine nucleotide exchange factor 11 (ARHGEF11) on 1q21 is a type 2 diabetes susceptibility gene in the Old Order Amish. Diabetes 56: 1363–1368, 2007 29. Zeggini E, Damcott CM, HansonRL, Karim MA, Rayner NW, Groves CJ, Baier LJ, Hale TC, Hattersley AT, Hitman GA, Hunt SE, Knowler WC, Mitchell BD, Ng MCY, O’Connell JR, Pollin TI, Vaxillaire M, Walker M, Wang X, Whittaker P, Xiang K, Jia W, Chan JCN, Froguel P, Deloukas P, Shuldiner AR, Elbein SC, McCarthy MI, the International Type 2 Diabetes 1q Consortium: Variation within the gene encoding the upstream stimula- tory factor 1 does not influence susceptibility to type 2 diabetes in samples from populations with replicated evidence of linkage to chromosome 1q. Diabetes 55:2541–2548, 2006 L. MA AND ASSOCIATES DIABETES, VOL. 56, MAY 2007 1459