We examined the genetic associations of previously identified sequence variants with type 2 diabetes mellitus (T2DM) and its potentially genetic heterogeneity by gender in a large-scale cohort. A total of 613 T2DM patients and 8221 control subjects from the Korea Association REsource (KARE) cohort were included in the analysis of genetic association of T2DM with 33 nucleotide polymorphic markers identified by previous studies. The association analysis was further conducted with data partitioned by gender. The association analysis resulted in five nucleotide sequence variants associated with the susceptibility of T2DM after Bonferonni correction (Po0.0015). One was located near the gene of hematopoietically expressed homeobox (HHEX), and the others were all in the gene of cyclin-dependent kinase 5 regulatory subunit-associated protein 1-like 1 (CDKAL1). Further analysis revealed that the sequence variant (rs5015480) near HHEX and two SNPs (rs7756992 and rs9465871) in CDKAL1 were associated with the susceptibility of T2DM in females (Po0.005), but not in males (P40.005). We suggested heterogeneous genetic associations of the T2DM susceptibility with the CDKAL1 and HHEX genes by gender.

Recently, the use of a mixed model methodology in genome-wide association studies (GWAS) has been considered effective for controlling population stratification and explaining the polygenic effects of complex traits. However, estimating polygenic variance components and heritability was biased when the mixed model was used. This bias results from a diluted genetic relationship covariance structure, particularly with a limited number of underlying causal variants. We simulated disease and quantitative phenotypes with a variety of heritabilities (0.1, 0.2, 0.3, 0.4, and 0.5), prevalence rates (0.1, 0.2, 0.3, and 0.5), and causal variant numbers (10, 30, 50, and 100). Heritabilities from the simulated data using restricted maximum likelihood were underestimated in many populations (Po0.05). The underestimation increased with a large heritability, a small prevalence, and a small number of causal variants. The underestimation was larger in analyzing disease traits compared with quantitative traits. This study suggests an underestimated heritability in GWAS upon using the mixed model methodology with an excessively larger number of variants versus causal variants.

The current study examined the promoter activity of an association signal in a 5'-upstream region of the gene encoding CCAAT/enhancer binding protein epsilon (CEBPE) identified from a recent genome-wide association study (GWAS) for complex acute lymphoblastic leukemia (ALL). This follow-up study first compared the activity of reporter constructs with three haplotypes estimated with the rs2239633 and its proximity nucleotide variants in strong linkage. The most frequent haplotype was CTTTTGT (H1), and the second most frequent haplotype consisted of entirely opposite alleles to H1 (TCGCACC, H2). Their luciferase activity revealed the strongest expression with H2 and the weakest with H1. Subsequent analysis revealed that different luciferase activity was found by the single-nucleotide substitution at rs2239632 and rs2239633 (P<0.05). Especially, the difference in luciferase activity between two alleles of rs2239632 corresponded to the difference between H1 and H2. We concluded that rs2239632 could regulate the expression of the CEBPE gene. We suggest a hypothesis that its risk allele (G) might increase the gene product and lead to leukemogenesis. As a result, a person with the allele or the corresponding haplotype might have increased susceptibility to ALL. Further research is warranted to investigate this hypothesis and the underlying mechanisms.

Supplemental Digital Content is available in the text

We examined the promoter activity of an association signal in an upstream region of the gene encoding fucosyltransferae 6 (FUT6) identified from a recent genomewide association study for the N-glycan level. The luciferase assay using reporter constructs with T and C alleles at rs3760776 revealed differential promoter activity. The amount of luciferin expressed with the C allele corresponded to that without the reporter construct (P > 0.05). On the other hand, the expression was dramatically reduced with the T allele (P < 0.05). The difference in transcriptional activity between the two alleles was confirmed by an electrophoretic mobility shift assay. It demonstrated that the promoter with a T allele had a stronger binding affinity to nuclear factors than that with the C allele. We concluded that the T allele of rs3760776 might repress the transcription of the FUT6 gene. Further studies are warranted to understand its underlying mechanism and its influence on susceptibility to potential diseases.