Fasting plasma glucose related genetic variants show ethnic variation in Hungarian populations: Szilvia Fiatal

Background

Elevated fasting glucose (FG) level is a risk factor for diabetes. Meta-analyses of European populations’ data identified many loci associated with glucose metabolism, but knowledge on other ethnicities is limited. Consequently, the generalizability of results to non-Europeans remains to be clarified. We addressed the question whether the individual effect of established FG related genetic loci are applicable to the Roma considered to have a South Asian origin.

Methods

Genotyping of 21 SNPs in 19 genes (among them C2CD4B, CDC123, CDKAL1, CDKN2A/B, FADS1, G6PC2, GCKR, GLIS3, SLC2A2, TCF7L2) was performed in Hungarian populations (General: N = 1411; Roma: N = 576). Single SNP associations (direction and magnitude of the effect on FG level) were estimated by regression models using covariates (age, sex). Data obtained on study populations were compared to formerly published European data and to each other. Individuals on antidiabetic treatment were excluded.

Results

The beta coefficients for SNPs on the sample of general population were found to be almost (except rs5219, rs174550) identical both in direction and magnitude, indicating that the effect is mostly concordant with the results obtained in large scale studies on Europeans. However, among Roma in case of 5 SNPs (rs10946398, rs340874, rs7034200, rs1111875 and rs174550) the effect direction differed in comparison with European data. The rs174550 variant showed difference in effect direction in both populations compared to data on Europeans. Between study populations in case of 5 SNPs (rs10946398, rs340874, rs7034200, rs1111875, rs5219) the association differed in direction, in addition one out of these variants (rs1111875) differed also in magnitude, pointing out the notable ethnic variation among Hungarians.

Conclusions

Differences observed in effect direction in case of 5 SNPs indicate that data obtained on European populations can be utilised in genetic risk estimation models for Roma with cautions.

Key messages:

• The vast majority of polymorphisms found to be related to FG level in populations of European descent are relevant for the Hungarian general but less relevant for the Hungarian Roma population.

• Differences in effect size measures found in case of some SNPs indicate that data obtained on European populations can be utilised in genetic risk estimation models for Roma only with cautions.