Search tips
Search criteria 


Logo of plosonePLoS OneView this ArticleSubmit to PLoSGet E-mail AlertsContact UsPublic Library of Science (PLoS)
PLoS One. 2010; 5(7): e11428.
Published online 2010 July 8. doi:  10.1371/journal.pone.0011428
PMCID: PMC2900202

Common Polymorphisms in MTNR1B, G6PC2 and GCK Are Associated with Increased Fasting Plasma Glucose and Impaired Beta-Cell Function in Chinese Subjects

Kathrin Maedler, Editor



Previous studies identified melatonin receptor 1B (MTNR1B), islet-specific glucose 6 phosphatase catalytic subunit-related protein (G6PC2), glucokinase (GCK) and glucokinase regulatory protein (GCKR) as candidate genes for type 2 diabetes (T2D) acting through elevated fasting plasma glucose (FPG). We examined the associations of the reported common variants of these genes with T2D and glucose homeostasis in three independent Chinese cohorts.

Methodology/Principal Findings

Five single nucleotide polymorphisms (SNPs), MTNR1B rs10830963, G6PC2 rs16856187 and rs478333, GCK rs1799884 and GCKR rs780094, were genotyped in 1644 controls (583 adults and 1061 adolescents) and 1342 T2D patients. The G-allele of MTNR1B rs10830963 and the C-alleles of both G6PC2 rs16856187 and rs478333 were associated with higher FPG (0.0034<P<6.6×10−5) in healthy controls. In addition to our previous report for association with FPG, the A-allele of GCK rs1799884 was also associated with reduced homeostasis model assessment of beta-cell function (HOMA-B) (P = 0.0015). Together with GCKR rs780094, the risk alleles of these SNPs exhibited dosage effect in their associations with increased FPG (P = 2.9×10−9) and reduced HOMA-B (P = 1.1×10−3). Meta-analyses strongly supported additive effects of MTNR1B rs10830963 and G6PC2 rs16856187 on FPG.


Common variants of MTNR1B, G6PC2 and GCK are associated with elevated FPG and impaired insulin secretion, both individually and jointly, suggesting that these risk alleles may precipitate or perpetuate hyperglycemia in predisposed individuals.


Elevated fasting plasma glucose (FPG) level is an important risk factor contributing to cardiometabolic diseases. Impaired fasting glucose, defined as FPG from 5.6 to 6.9 mmol/l (ADA criteria) [1] or 6.1 to 6.9 mmol/l (WHO criteria) [2], is associated with increased risk of diabetes and cardiovascular disease [3], [4]. Adequate insulin secretion and sensitivity is critical in the maintenance of euglycemia [5]. FPG level is shown to be moderately heritable in twin and family studies (heritability estimate h2 = 0.20–0.28 [6], [7], [8]. Recent genome-wide association studies (GWAS) have uncovered a few loci associated with FPG, including genes encoded for melatonin receptor 1B (MTNR1B), glucose-6-phospate catalytic subunit 2 (G6PC2), glucokinase (GCK) and glucokinase regulatory protein (GCKR) [9]. Surprisingly, these genes only demonstrated modest or weak association with type 2 diabetes (T2D) [9], [10], , despite a close interplay between T2D and high FPG level. Although most T2D genes are implicated in beta-cell function [12], high FPG level from other causes can worsen insulin secretion or sensitivity by setting up a vicious cycle via glucotoxicity [13].

MTNR1B, GCK and G6PC2 proteins are expressed in the human beta cells [14], [15], [16]. MTNR1B encodes a high affinity receptor for melatonin, a hormone primarily secreted by the pineal gland to regulate circadian rhythm and sleep cycles [17]. Plasma melatonin follows an opposite circadian rhythm to plasma insulin and glucose, rising by night and falling by day, which suggests that melatonin may affect insulin release and glucose level via its islet-specific receptor [18]. Large scale association studies [9], [10], [14] suggested that two common MTNR1B SNPs rs1387153 and rs10830963 (r2 = 0.7 in Europeans) may affect insulin secretion and glucose homeostasis.

Glucokinase (encoded by GCK) is a glucose-sensing enzyme that determines the threshold for glucose-stimulated-insulin-secretion (GSIS) in islets, and controls gluconeogenesis and glycogen synthesis in hepatocytes [15]. The hepatic activity of GCK is inhibited by its regulatory protein GCKR in a dose-dependent manner, which competes with glucose for binding site [19]. Two GCKR variants rs780094 and rs1260326 (r2 = 0.93 in Europeans) are associated with HOMA-IR, FPG and triglyceride levels [20], [21]. A functional study showed that this effect was mediated through reduced repression on GCK [22].

The GCK variant rs1799884 is associated with higher FPG in Caucasian and Chinese cohorts [23], [24], [25], and rare GCK mutations confer a form of maturity-onset diabetes of the young (MODY 2) characterized by hyperglycemia [26]. Glucose-6-phospate catalytic subunit 2 (encoded by G6PC2) is proposed to counteract pancreatic GCK activity by competing for glucose usage [16]. G6PC2-null mice demonstrated a ~15% reduction in FPG level [27], and two SNPs rs560887 and rs563694 were repeatedly associated with hyperglycemia in Europeans [10], [28], [29], [30]. However, these two variants are nearly monomorphic in Asians, so Hu et al. adopted a tagSNP approach and reported a FPG-associated SNP rs16856187 in Chinese [31].

Based on their common effects on FPG levels and co-presence in the beta-cell and liver, variants in MTNR1B, GCK, G6PC2 and GCKR are hypothesized to have interactions or joint effects. Bouatia-Naji et al. have reported their additive effects on FPG [9], [32], and we have previously shown that a variant in GCKR modifies the association between GCK and FPG [24]. In this study, we aimed to validate the reported dosage effects on FPG in a Chinese cohort, and examined their associations with insulin secretion and sensitivity as estimated by HOMA-IR and HOMA-B indices.

Materials and Methods


We have previously described the study design, ascertainment, inclusion criteria and phenotyping procedures of subjects included in this study [33]. All subjects were of southern Han Chinese ancestry residing in Hong Kong. The control cohort consists of 1644 subjects with FPG<6.1 mmol/l ascertained from a) 583 hospital staff and volunteers from a community-based health screening program (mean age 41.4±10.5 years, 45% male) and b) 1061 adolescents from a community-based school survey (mean age 15.4±1.9 years, 45% male). A subgroup of 420 adult controls also underwent a 75g oral glucose tolerance test (OGTT). The case cohort consists of 1342 unrelated T2D patients (mean age 50.5±13.7 years, 41% male, mean duration of T2D 6.1±6.5 years) selected from the Hong Kong Diabetes Registry (HKDR). T2D was diagnosed according to the 1998 World Health Organization (WHO) criteria. Patients with classic type 1 diabetes with acute ketotic presentation or continuous requirement of insulin within 1 year of diagnosis were excluded. The clinical characteristics of subjects in the three cohorts are summarized in Table 1. Written informed consent was obtained from all adult subjects and parents of the adolescents while the adolescents gave verbal consent. This study was approved by the Clinical Research Ethics Committee of the Chinese University of Hong Kong.

Table 1
Clinical and metabolic characteristics of 1644 healthy Chinese adults and adolescents and 1342 T2D patients.

Clinical studies

All study subjects were examined in the morning after an overnight fast. Anthropometric measurements including body weight and height were documented. Fasting blood samples were collected for DNA extraction and measurements of FPG and fasting plasma insulin (FPI). Homeostasis model assessment of insulin resistance (HOMA-IR) was calculated as (FPI×FPG)An external file that holds a picture, illustration, etc.
Object name is pone.0011428.e001.jpg22.5, and homeostasis model assessment of beta-cell function (HOMA-B) was calculated as FPI×20An external file that holds a picture, illustration, etc.
Object name is pone.0011428.e002.jpg(FPG - 3.5) [34]. Insulinogenic index was calculated as (PI during OGTT for 30 min - 0 min)An external file that holds a picture, illustration, etc.
Object name is pone.0011428.e003.jpg(PG during OGTT for 30 min - 0 min) [35]. Data were discarded if PI or PG level at 0 min were higher than that of 30 min. Insulin sensitivity index (ISI) was estimated using the formula proposed by Matsuda and DeFronzo [36]: 10,000An external file that holds a picture, illustration, etc.
Object name is pone.0011428.e004.jpg[FPG×FPI×(mean PG during OGTT)×(mean PI during OGTT)]An external file that holds a picture, illustration, etc.
Object name is pone.0011428.e005.jpg. Insulin disposition index (IDI) was calculated as ISI×insulinogenic indexAn external file that holds a picture, illustration, etc.
Object name is pone.0011428.e006.jpg100 [36].


We genotyped five SNPs in four genes including MTNR1B rs10830963, G6PC2 rs16856187 and rs478333, GCK rs1799884 and GCKR rs780094 in all study subjects due to their reported associations with FPG, beta-cell function and T2D [29], [31], [32], except for G6PC2 rs478333. Although G6PC2 rs560887 [28], [30] and G6PC2 rs563694 [29] showed association with FPG in Caucasian populations, both of their corresponding minor allele frequencies (MAF) are rare in Chinese population (0.006 and 0.012 in HapMap CHB for rs560887 and rs563694, respectively). Additionally, Shanghai study [31] found that rs16856187 showed the strongest signal for both T2D and FPG. To clarify these inter-ethnic differences, we genotyped rs16856187 and another nearby SNP rs478333 located in the 3′ flanking region, which has common allele frequencies in both Chinese (0.29 in HapMap CHB) and Caucasian (0.49 in HapMap CEU) populations. We did not test for associations for all tagging SNPs of the respective genes. Genotyping on genomic DNA was performed either at deCODE Genetics using the Centaurus (Nanogen) platform or at the McGill University and Genome Quebec Innovation Centre using the Sequenom MassARRAY platform (San Diego, CA, USA). The concordance rate for part of the samples genotyped on both platforms is >99%. All SNPs were in Hardy-Weinberg equilibrium (P>0.05) in control cohorts using the exact test implemented in PLINK [37]. The overall genotype call rates were >96% and the minor allele frequencies (MAF) in normal controls (MAF of MTNR1B rs10830963 = 0.44 for both adult and adolescent controls; G6PC2 rs16856187 and rs478333 = 0.30 and 0.35, respectively for both adult and adolescent controls; GCK rs1799884 = 0.16 and 0.19 in adult and adolescent controls, respectively) were comparable with the HapMap CHB data (0.48 for MTNR1B rs10830963; 0.28 and 0.29 for G6PC2 rs16856187 and rs478333, respectively; 0.20 for GCK rs1799884), except the one of GCKR rs780094. MAFs of GCKR rs780094 in our data (0.46 for both adult and adolescent cohorts) were lower than seen in the HapMap CHB (0.60), but they were similar to the frequency reported in a group of Han Chinese (0.44) study [21].

Systematic Review

A systematic literature search was performed according to the MOOSE guidelines [38] for the meta-analysis of observational studies. The description of studies and details of the literature search process are outlined in the table (Table S1) and the flow chart (Figure S1), respectively. We searched the PubMed database from inception to January 2010 for association studies between fasting glucose and MTNR1B or G6PC2 genes. The keywords used were MTNR1B, G6PC2 and fasting glucose. We restricted our analysis to human studies, and placed no language restriction. We included studies if they (a) reported the association results for subjects in case-control or population-based studies; (b) genotyped MTNR1B rs10830963, G6PC2 rs560887 or rs16856187 and measured fasting glucose levels for the studied subjects; (c) presented results as mean ± SD with sample size stratified by genotypes with or without adjustment for covariate. We excluded studies if they (a) were reviews or abstract; (b) were duplicate reports on previously published studies; (c) did not provide sufficient information for computation of a quantitative effect estimate of the relationship between FPG and genetic variants.

Statistical analysis

All data are presented as percentage, mean ± SD or median (interquartile range), as appropriate. Insulinogenic index, FPI, HOMA-IR, HOMA-B, ISI and IDI were logarithmically transformed due to skewed distributions. Each trait was winsorized separately in adult and adolescent cohorts by replacing extreme values with 4 standard deviations from the mean. Less than 0.2% of data were replaced.

Within each control cohort, associations between genotypes and phenotypic traits were tested by multivariate linear regression adjusted for sex, age, and BMI (where appropriate) under the additive genetic model. In the combined analysis, an additional dummy variable “study cohort” coded as 0 for adult controls and 1 for adolescent controls was included in the regression model. Multiple testing of phenotypic traits and SNPs were corrected by controlling the false discovery rate (FDR) using the Benjamini-Hochberg approach [39]. An alternative method for controlling multiple testing is developed by Conneely and Boehnke [40], which accounts for correlation both among SNPs and among phenotypes and is less conservative than FDR. To assess gene-gene interaction effects on phenotypic traits, linear regression analyses including the main and pairwise interaction effects of SNPs under an additive genetic model were applied. The joint effects of the SNPs was assessed by calculating the estimated marginal mean with 95% confidence intervals (CIs) in a general linear model (including sex, age, BMI and study cohort as covariates), categorized by the number of risk alleles assuming an additive genetic model. Risk alleles were defined as alleles that increased fasting plasma glucose described either in literature or in the present study. The significance of the trend was tested by linear regression using the number of risk alleles carried as an independent variable.

Frequencies of genotypes and number of risk alleles between T2D cases and healthy controls were compared using logistic regression adjusted for age, sex and BMI. Odd ratios (ORs) with 95% CIs were presented.

Meta-analyses for the association of FPG were calculated based on the Hedges g statistic which was used to calculate the standardized mean difference (SMD) across studies under the fixed effects model. To account for heterogeneity of SMDs across studies (Cochran's Q statistic P<0.1), the overall effect size (SMD) under the random effects model was reported, in which both random variations within and between different studies were incorporated [41].

We estimated study power using genetic power calculator [42]. Assuming an additive model with the frequencies of 0.41 [43] for the G-allele of MTNR1B rs10830963, 0.30 [31] for the C-alleles of G6PC2 rs16856187, 0.17 [24] for the A-allele of GCK rs1799884 and 0.54 [24] for the C-allele GCKR rs780094 in a Chinese population, our sample size has >90% power to detect a T2D risk under the prevalence of 0.1 with an odds ratio of 1.16 [43], 1.19 [31], 1.22 [25] and 1.18 [21] respectively, and a per-allele effect of increasing FPG by >0.068 (total QTL variance = 0.012) and >0.067 (total QTL variance = 0.012) mmol/l [31], [43] for rs10830963 and rs16856187 respectively, at the α level of 0.05.

All statistical analyses were performed using SAS v.9.1 (SAS Institute, Cary, NC, USA) or SPSS for Windows v.15 (SPSS, Chicago, IL, USA) unless specified otherwise. Two-tailed P values <0.05 were considered statistically significant.


Associations with FPG, beta cell function and T2D

We have previously shown that the minor A-allele of GCK rs1799884 was associated with higher FPG, nevertheless, no association was detected for GCKR rs780049. In this study, we further observed consistent and significant association of the minor G-allele of MTNR1B rs10830963 (Beta±S.E. = 0.037±0.012 and P = 0.0034 in combined analysis) as well as the C-alleles of both G6PC2 rs16856187 (Beta±S.E. = 0.059±0.015 and P = 6.6×10−5 in combined analysis) and rs478333 (Beta±S.E. = 0.050±0.013 and P = 0.0002 in combined analysis) with increased FPG after adjustment for age, gender, BMI and/or study cohorts (Table 2 and and33).

Table 2
Associations of MTNR1B rs10830963 with type 2 diabetes related traits in Chinese control subjects.
Table 3
Associations of G6PC2 rs16856187 and rs478333 with type 2 diabetes related traits in Chinese control subjects.

In addition, association with reduced beta-cell function as assessed by HOMA-B was also observed for the A-allele of GCK rs1799884 (Beta±S.E. = −0.081±0.026 and P = 0.0015), as well as trend for associations with the G-allele of MTNR1B rs10830963 (Beta±S.E. = −0.037±0.019 and P = 0.0532) and the C-allele of G6PC2 rs478333 (Beta±S.E. = −0.039±0.021 and P = 0.0623) in the combined control samples (Tables 2, ,3,3, ,44).

Table 4
Associations of GCK rs1799884 and GCKR rs780094 with type 2 diabetes related traits in Chinese control subjects.

The associations of FPG with MTNR1B rs10830963 (P = 0.0162), G6PC2 rs16856187 (P = 0.0013) and rs478333 (P = 0.0019), as well as GCK rs1799884 with HOMA-B (P = 0.0095) in the combined control samples remained significant after controlling for FDR. None of the SNPs revealed association with T2D (Table S2), BMI, FPI or insulin sensitivity, as measured by HOMA-IR (Tables 2, ,3,3, ,4)4) or OGTT-based traits (Table S3).

Interaction and joint effect of genes on FPG, beta-cell function and T2D

We did not detect any novel pairwise interaction between genes on FPG level or beta-cell function in the combined healthy controls (data not shown), apart from the one previously reported between GCK rs1799884 and GCKR rs780094. We then examined the joint effects of selected SNPs from GCK, GCKR, MTNR1B, G6PC2 genes on FPG and beta-cell function, assuming that all risk alleles have similar effect sizes. Due to the relatively high linkage disequilibrium between G6PC2 rs16856187 and G6PC2 rs478333 (r2 = 0.62 in both HapMap CHB data and the present study), only the most significant SNP was included for the joint analysis of FPG or beta-cell function, respectively. We tested for the independence between loci by using all four corresponding loci in the regression models. All loci were independent (P<0.05) except for GCKR rs780094. Subjects with increasing number of risk alleles showed higher FPG concentration (P = 2.9×10−9) and lower value of HOMA-B (P = 1.1×10−3) (Figure 1A–B) in a dose-dependent manner. In addition, we did not find any association between T2D and combined gene variants (Table S2).

Figure 1
Additive effects on fasting plasma glucose and HOMA beta-cell function.

Meta-analyses of MTNR1B rs10830963 and G6PC2 rs16856187 with FPG

In our previous study, we have confirmed the association of GCK rs1799884 with FPG in a meta-analysis including both Europeans and Chinese populations. In the present study, meta-analysis of association of MTNR1B (rs10830963) in Europeans showed increases of 0.15 (0.11–0.19) and 0.29 (0.25–0.34) in (SMD) of FPG for CG and GG genotypes, respectively, when compared to the CC reference genotype (Figure S2). However, weaker and dominance effect was found in Chinese, with increases of 0.20 (0.11–0.28) and 0.23 (0.12–0.34) in SMD of FPG for CG and GG genotypes, respectively (Figure S2). In the combined meta-analysis for all European and Chinese cohorts, we confirmed the additive effect of the G-allele in MTNR1B rs10830963 with FPG. Furthermore, an additive trend of 0.15 (0.03–0.27) and 0.31 (0.19–0.44) increases in SMD of FPG for AC and CC genotypes, respectively, when compared to the AA reference genotype, was also observed for G6PC2 rs16856187 in Chinese populations (Figure S3). Due to significant heterogeneity amongst the study cohorts (P<0.1), the combined SMDs were calculated based on the random effect models, only using the fixed effect models for the association of MTNR1B rs10830963 in all Chinese cohorts (P>0.1) (Figure S3).


Here we reported the association of rs10830963 in MTNR1B, rs16856187 and rs478333 in G6PC2, and rs1799884 in GCK with higher FPG and lower HOMA-B levels. Further analyses of the risk alleles (G-allele of rs10830963, C-alleles of both rs16856187 and rs478333, A-allele of rs1799884 and C-allele of GCKR rs780094) confirmed their joint effects on FPG and beta-cell function. These consistent findings from three independent cohorts strongly support the risk effects of these variants on GSIS to predispose hyperglycemia in Chinese.

Consistent with previous studies [9], [10], [28], [30], [32], [44], [45], [46], [47], we observed the individual and joint effects of risk alleles in MTNR1B, G6PC2, and GCK on FPG. Compared to the carriers with 0 or 1 alleles, each additional allele increases mean FPG level by 0.048 (0.032–0.064) mmol/l (Figure 1). This effect size is similar to those observed in Dutch (0.05 (0.04–0.07)) [32], French (0.07 (0.06–0.08) [9], and Japanese (0.055 (0.045–0.065)) [45] populations, despite the use of different risk variants. While GCK rs1799884 and MTNR1B rs10830963 are strongly associated to FPG in both Caucasian and Asian cohorts [21], [43], [45], [48], we and others have observed an Asian specific G6PC2 risk variant (rs16856187 in Hu et al [31] and the current study, and rs3755157 in Takeuchi et al [45], r2≥0.9) in addition to the Caucasian reported rs563694 (Figure S4). This suggests that while these loci are reproducibly associated to FPG across populations, as shown in our meta-analyses (Figures S2, S3), replication in multiple ethnicities could help to identify population specific risk variants and filter for causal variants. Since G6PC2 rs478333 is common in both Asian (0.29 in Hapmap CHB) and Caucasian (0.49 in HapMap CEU), its replication could provide new insight. Together, the Asian-specific rs16856187 and rs478333 and the Caucasian specific rs563694 (r2 = 0.02 and 0.03 to rs16856187 and rs47833, respectively) helped refined a 8.8 kb region of interest (Figure S4), which would be useful for future functional studies.

We further observed that these variants exerted individual and combined effects on beta-cell function, as estimated by the homeostasis model (HOMA-B). To our knowledge, this is the first study showing that the same risk alleles can jointly impair insulin secretion and elevate FPG level. Each allele decreases mean HOMA-B level by 0.04%, which may partially explain the concomitant increase of 0.048 mmol/l in mean FPG level (Figure 1). A recent GWAS meta-analysis involving ~100,000 Europeans demonstrated that the same risk alleles in MTNR1B, G6PC2, and GCK were associated to FPG and HOMA-B at genome-wide significant levels [48]. Most of the 17 FPG-associated loci were also consistently associated to HOMA-B [48]. Our result and other literature support that impaired beta-cell function and hyperglycemia likely share the same underlying pathogenic mechanism.

Although we had sufficient power (>90%) to detect T2D risks with odds ratio ranging from 1.16 to 1.22 [21], [25], [31], [43], we did not detect T2D-associations for these four loci (P = 0.47–0.88). Several large-scale studies failed to find T2D-associations as well [28], [29], and even if found, their effects on T2D are shown to be much weaker than their effects on FPG [9], [10], [11], [21], [31], [32], [43]. Dupuis et al. estimated that MTNR1B, GCK and GCKR were associated to T2D with modest effect sizes of 1.06–1.09 [48], so our result may be explained by a lack of power due to the small sample size.

In pancreatic islets, glucose is phosphorylated by GCK into glucose-6-phosphate (G6P), committing it for glycolysis and the subsequent glucose-stimulated insulin secretion (GSIS). G6PC2 is hypothesized to counteract this process, which removes the phosphate group and releases glucose from the beta-cell [16]. MTNR1B may down regulate GCK expression and GSIS by lowering intracellular cAMP level [17], [49]. Indeed, GCK, G6PC2 and MTNR1B knock out mice demonstrated significantly lower FPG levels [18], [27], [50]. Thus, the coexistences of risk alleles in MTNR1B, G6PC2, and GCK may confer high melatonin level, low intra-islet glucose oxidation, and low GCK activity in carriers, causing decreased insulin secretion and increased FPG level as observed in this study. The low number of overlapping loci between FPG and T2D GWAS studies [48] suggested that genetic variants may disturb beta-cell function and affect physiological fasting glucose levels beneath the pathological thresholds of T2D.

In conclusion, we showed that risk alleles in GCK, GCKR, G6PC2, and MTNR1B exert joint effects on FPG and HOMA-B. The Asian-specific risk variants in G6PC2 may help to fine map the causal region within the gene (or possibly the G6PC2-ABCB11 region, given the proximity and strong linkage disequilibrium between the two genes). Concordant with previous studies, we assumed each allele contributes equal dosage, despite minor differences among their effect sizes. Some limitations of this study include using adolescents as controls, which may reduce our power as they may develop T2D later in life. This concern is partially alleviated by obtaining similar results compared to the Chinese and Caucasian adult cohorts (Figures S2, S3). Other studies using adolescent cohorts also reported consistent results [44]. Our study could be improved by using directly measured insulin data rather than surrogate measures such as HOMA-B and HOMA-IR, and increasing the sample size of our OGTT-based associations. Our results could be further improved by using the PACT method instead of FDR to correct for multiple comparisons, which considered the correlation structure among both SNPs and phenotypes and is less conservative [40]. Future studies in other populations would substantiate our finding.

Supporting Information

Table S1

Descriptions of studies.

(0.13 MB DOC)

Table S2

Associations of SNPs and risk allele scores with type 2 diabetes.

(0.07 MB DOC)

Table S3

Associations of MTNR1B rs10830963, GCK rs1799884 as well as G6PC2 rs16856187 and rs478333 with OGTT-based traits in combined Chinese control subjects (adults and adolescents).

(0.07 MB DOC)

Figure S1

Flow chart of literature search for studies on the association of fasting glucose with a) MTNR1B rs10830963 and b) G6PC2 rs560887/rs16856187.

(0.16 MB DOC)

Figure S2

Meta-analysis of associations of MTNR1B rs10830963 with fasting plasma glucose in European and Chinese populations.

(0.06 MB DOC)

Figure S3

Meta-analysis of associations of G6PC2 rs560887 and rs16856187 with fasting plasma glucose in European and Chinese populations, respectively.

(0.06 MB DOC)

Figure S4

Linkage disequilibrium for SNPs within the region near G6PC2 and ABCB11 at chromosome 2 between 169.47 Mb and 169.49 Mb.

(0.12 MB DOC)


We are grateful to all study subjects who took part in these studies. We thank Ms. Cherry Chiu and Dr. Gary Wong for recruitment of study subjects, and Mr. Alex Ng and Ms. Patty Tse for technical assistance. We thank deCODE Genetics, Iceland, for help with genotyping of samples, and the CUHK Centre for Clinical Trials and Information Technology Services Centre for support of computing resources. Special thanks are extended to all nursing and medical staff at the PWH Diabetes and Endocrine Centre for their dedication and professionalism.


Competing Interests: This study received technical support from Roche Pharmaceuticals (Research collaboration agreement TH053001). Dr. Soren Germer and Dr. Mitchell Martin from the Roche Pharmaceuticals played a role in the study design, review and approval of the manuscript. No author has any other conflict of interest to declare.

Funding: This study was funded by the Hong Kong Government Research Grant Committee Central Allocation Scheme (CUHK 1/04C), Research Grant Committee (CUHK472407) and the Chinese University of Hong Kong Direct Grant (2006.1.041).


1. Diagnosis and classification of diabetes mellitus. Diabetes Care. 2006;29(Suppl 1):S43–48. [PubMed]
2. Alberti KG, Zimmet PZ. Definition, diagnosis and classification of diabetes mellitus and its complications. Part 1: diagnosis and classification of diabetes mellitus provisional report of a WHO consultation. Diabet Med. 1998;15:539–553. [PubMed]
3. Shaw JE, Zimmet PZ, Hodge AM, de Courten M, Dowse GK, et al. Impaired fasting glucose: how low should it go? Diabetes Care. 2000;23:34–39. [PubMed]
4. Unwin N, Shaw J, Zimmet P, Alberti KG. Impaired glucose tolerance and impaired fasting glycaemia: the current status on definition and intervention. Diabet Med. 2002;19:708–723. [PubMed]
5. Piche ME, Arcand-Bosse JF, Despres JP, Perusse L, Lemieux S, et al. What is a normal glucose value? Differences in indexes of plasma glucose homeostasis in subjects with normal fasting glucose. Diabetes Care. 2004;27:2470–2477. [PubMed]
6. Freeman MS, Mansfield MW, Barrett JH, Grant PJ. Heritability of features of the insulin resistance syndrome in a community-based study of healthy families. Diabet Med. 2002;19:994–999. [PubMed]
7. Li JK, Ng MC, So WY, Chiu CK, Ozaki R, et al. Phenotypic and genetic clustering of diabetes and metabolic syndrome in Chinese families with type 2 diabetes mellitus. Diabetes Metab Res Rev. 2006;22:46–52. [PubMed]
8. Zhang S, Liu X, Yu Y, Hong X, Christoffel KK, et al. Genetic and Environmental Contributions to Phenotypic Components of Metabolic Syndrome: A Population-based Twin Study. Obesity (Silver Spring) 2009 [PubMed]
9. Bouatia-Naji N, Bonnefond A, Cavalcanti-Proenca C, Sparso T, Holmkvist J, et al. A variant near MTNR1B is associated with increased fasting plasma glucose levels and type 2 diabetes risk. Nat Genet. 2009;41:89–94. [PubMed]
10. Prokopenko I, Langenberg C, Florez JC, Saxena R, Soranzo N, et al. Variants in MTNR1B influence fasting glucose levels. Nat Genet. 2009;41:77–81. [PMC free article] [PubMed]
11. Zeggini E, Scott LJ, Saxena R, Voight BF, Marchini JL, et al. Meta-analysis of genome-wide association data and large-scale replication identifies additional susceptibility loci for type 2 diabetes. Nat Genet. 2008;40:638–645. [PMC free article] [PubMed]
12. Kahn SE, Hull RL, Utzschneider KM. Mechanisms linking obesity to insulin resistance and type 2 diabetes. Nature. 2006;444:840–846. [PubMed]
13. Issad T, Kuo M. O-GlcNAc modification of transcription factors, glucose sensing and glucotoxicity. Trends Endocrinol Metab. 2008;19:380–389. [PubMed]
14. Lyssenko V, Nagorny CL, Erdos MR, Wierup N, Jonsson A, et al. Common variant in MTNR1B associated with increased risk of type 2 diabetes and impaired early insulin secretion. Nat Genet. 2009;41:82–88. [PubMed]
15. Matschinsky F, Liang Y, Kesavan P, Wang L, Froguel P, et al. Glucokinase as pancreatic beta cell glucose sensor and diabetes gene. J Clin Invest. 1993;92:2092–2098. [PMC free article] [PubMed]
16. Newgard CB, Lu D, Jensen MV, Schissler J, Boucher A, et al. Stimulus/secretion coupling factors in glucose-stimulated insulin secretion: insights gained from a multidisciplinary approach. Diabetes. 2002;51(Suppl 3):S389–393. [PubMed]
17. Peschke E. Melatonin, endocrine pancreas and diabetes. J Pineal Res. 2008;44:26–40. [PubMed]
18. Muhlbauer E, Gross E, Labucay K, Wolgast S, Peschke E. Loss of melatonin signalling and its impact on circadian rhythms in mouse organs regulating blood glucose. Eur J Pharmacol. 2009;606:61–71. [PubMed]
19. Van Schaftingen E. Short-term regulation of glucokinase. Diabetologia. 1994;37(Suppl 2):S43–47. [PubMed]
20. Saxena R, Voight BF, Lyssenko V, Burtt NP, de Bakker PI, et al. Genome-wide association analysis identifies loci for type 2 diabetes and triglyceride levels. Science. 2007;316:1331–1336. [PubMed]
21. Qi Q, Wu Y, Li H, Loos RJ, Hu FB, et al. Association of GCKR rs780094, alone or in combination with GCK rs1799884, with type 2 diabetes and related traits in a Han Chinese population. Diabetologia. 2009;52:834–843. [PubMed]
22. Beer NL, Tribble ND, McCulloch LJ, Roos C, Johnson PR, et al. The P446L variant in GCKR associated with fasting plasma glucose and triglyceride levels exerts its effect through increased glucokinase activity in liver. Hum Mol Genet 2009 [PMC free article] [PubMed]
23. Rose CS, Ek J, Urhammer SA, Glumer C, Borch-Johnsen K, et al. A -30G>A polymorphism of the beta-cell-specific glucokinase promoter associates with hyperglycemia in the general population of whites. Diabetes. 2005;54:3026–3031. [PubMed]
24. Tam CH, Ma RC, So WY, Wang Y, Lam VK, et al. Interaction effect of genetic polymorphisms in glucokinase (GCK) and glucokinase regulatory protein (GCKR) on metabolic traits in healthy Chinese adults and adolescents. Diabetes. 2009;58:765–769. [PMC free article] [PubMed]
25. Vaxillaire M, Veslot J, Dina C, Proenca C, Cauchi S, et al. Impact of common type 2 diabetes risk polymorphisms in the DESIR prospective study. Diabetes. 2008;57:244–254. [PubMed]
26. Froguel P, Zouali H, Vionnet N, Velho G, Vaxillaire M, et al. Familial hyperglycemia due to mutations in glucokinase. Definition of a subtype of diabetes mellitus. N Engl J Med. 1993;328:697–702. [PubMed]
27. Wang Y, Flemming BP, Martin CC, Allen SR, Walters J, et al. Long-range enhancers are required to maintain expression of the autoantigen islet-specific glucose-6-phosphatase catalytic subunit-related protein in adult mouse islets in vivo. Diabetes. 2008;57:133–141. [PubMed]
28. Bouatia-Naji N, Rocheleau G, Van Lommel L, Lemaire K, Schuit F, et al. A polymorphism within the G6PC2 gene is associated with fasting plasma glucose levels. Science. 2008;320:1085–1088. [PubMed]
29. Chen WM, Erdos MR, Jackson AU, Saxena R, Sanna S, et al. Variations in the G6PC2/ABCB11 genomic region are associated with fasting glucose levels. J Clin Invest. 2008;118:2620–2628. [PMC free article] [PubMed]
30. Rose CS, Grarup N, Krarup NT, Poulsen P, Wegner L, et al. A variant in the G6PC2/ABCB11 locus is associated with increased fasting plasma glucose, increased basal hepatic glucose production and increased insulin release after oral and intravenous glucose loads. Diabetologia. 2009;52:2122–2129. [PubMed]
31. Hu C, Zhang R, Wang C, Ma X, Fang Q, et al. A genetic variant of G6PC2 is associated with type 2 diabetes and fasting plasma glucose level in the Chinese population. Diabetologia. 2009;52:451–456. [PubMed]
32. Reiling E, van 't Riet E, Groenewoud MJ, Welschen LM, van Hove EC, et al. Combined effects of single-nucleotide polymorphisms in GCK, GCKR, G6PC2 and MTNR1B on fasting plasma glucose and type 2 diabetes risk. Diabetologia 2009 [PMC free article] [PubMed]
33. Ng MC, Park KS, Oh B, Tam CH, Cho YM, et al. Implication of genetic variants near TCF7L2, SLC30A8, HHEX, CDKAL1, CDKN2A/B, IGF2BP2, and FTO in type 2 diabetes and obesity in 6,719 Asians. Diabetes. 2008;57:2226–2233. [PMC free article] [PubMed]
34. Matthews DR, Hosker JP, Rudenski AS, Naylor BA, Treacher DF, et al. Homeostasis model assessment: insulin resistance and beta-cell function from fasting plasma glucose and insulin concentrations in man. Diabetologia. 1985;28:412–419. [PubMed]
35. Phillips DI, Clark PM, Hales CN, Osmond C. Understanding oral glucose tolerance: comparison of glucose or insulin measurements during the oral glucose tolerance test with specific measurements of insulin resistance and insulin secretion. Diabet Med. 1994;11:286–292. [PubMed]
36. Matsuda M, DeFronzo RA. Insulin sensitivity indices obtained from oral glucose tolerance testing: comparison with the euglycemic insulin clamp. Diabetes Care. 1999;22:1462–1470. [PubMed]
37. Purcell S, Neale B, Todd-Brown K, Thomas L, Ferreira MA, et al. PLINK: a tool set for whole-genome association and population-based linkage analyses. Am J Hum Genet. 2007;81:559–575. [PubMed]
38. Stroup DF, Berlin JA, Morton SC, Olkin I, Williamson GD, et al. Meta-analysis of observational studies in epidemiology: a proposal for reporting. Meta-analysis Of Observational Studies in Epidemiology (MOOSE) group. JAMA. 2000;283:2008–2012. [PubMed]
39. Benjamini Y, Drai D, Elmer G, Kafkafi N, Golani I. Controlling the false discovery rate in behavior genetics research. Behav Brain Res. 2001;125:279–284. [PubMed]
40. Conneely KN, Boehnke M. So Many Correlated Tests, So Little Time! Rapid Adjustment of P Values for Multiple Correlated Tests. Am J Hum Genet. 2007;81 [PubMed]
41. DerSimonian R, Laird N. Meta-analysis in clinical trials. Control Clin Trials. 1986;7:177–188. [PubMed]
42. Purcell S, Cherny SS, Sham PC. Genetic Power Calculator: design of linkage and association genetic mapping studies of complex traits. Bioinformatics. 2003;19:149–150. [PubMed]
43. Ronn T, Wen J, Yang Z, Lu B, Du Y, et al. A common variant in MTNR1B, encoding melatonin receptor 1B, is associated with type 2 diabetes and fasting plasma glucose in Han Chinese individuals. Diabetologia. 2009;52:830–833. [PubMed]
44. Kelliny C, Ekelund U, Andersen LB, Brage S, Loos RJ, et al. Common genetic determinants of glucose homeostasis in healthy children: the European Youth Heart Study. Diabetes. 2009;58:2939–2945. [PMC free article] [PubMed]
45. Takeuchi F, Katsuya T, Chakrewarthy S, Yamamoto K, Fujioka A, et al. Common variants at the GCK, GCKR, G6PC2-ABCB11 and MTNR1B loci are associated with fasting glucose in two Asian populations. Diabetologia. 53:299–308. [PubMed]
46. Sparso T, Bonnefond A, Andersson E, Bouatia-Naji N, Holmkvist J, et al. G-allele of intronic rs10830963 in MTNR1B confers increased risk of impaired fasting glycemia and type 2 diabetes through an impaired glucose-stimulated insulin release: studies involving 19,605 Europeans. Diabetes. 2009;58:1450–1456. [PMC free article] [PubMed]
47. Langenberg C, Pascoe L, Mari A, Tura A, Laakso M, et al. Common genetic variation in the melatonin receptor 1B gene (MTNR1B) is associated with decreased early-phase insulin response. Diabetologia. 2009;52:1537–1542. [PMC free article] [PubMed]
48. Dupuis J, Langenberg C, Prokopenko I, Saxena R, Soranzo N, et al. New genetic loci implicated in fasting glucose homeostasis and their impact on type 2 diabetes risk. Nat Genet. 2010;42:105–116. [PMC free article] [PubMed]
49. Fernandez-Mejia C, Vega-Allende J, Rojas-Ochoa A, Rodriguez-Dorantes M, Romero-Navarro G, et al. Cyclic adenosine 3′,5′-monophosphate increases pancreatic glucokinase activity and gene expression. Endocrinology. 2001;142:1448–1452. [PubMed]
50. Grupe A, Hultgren B, Ryan A, Ma YH, Bauer M, et al. Transgenic knockouts reveal a critical requirement for pancreatic beta cell glucokinase in maintaining glucose homeostasis. Cell. 1995;83:69–78. [PubMed]

Articles from PLoS ONE are provided here courtesy of Public Library of Science