The rising healthcare costs of obesity and obesity-associated health problems are sufficient to justify prevention as daily work of the pediatric clinicians. By recognizing genetic variants that predispose to obesity, pediatricians could classify obese children into subgroups that might respond to specific diets or physical activity regimes, drugs or surgeries. In this study, we selected the most extensively studied SNP in APOA5 gene, tag SNP rs662799 in the promoter region and another SNP rs651821, located 3 bp upstream from the predicted start codon of APOA5, which is in linkage disequilibrium with the much studied rs662799. The two tightly linked SNPs both show high minor allele frequencies in Chinese population, our results may be generalizable to other ethnic child population which have similar minor allele frequencies.
Our available data supported the hypothesis that the TG-raising genetic variants in the APOA5 gene may also have high risks for obesity in Chinese children and adolescents. In consistent with our results, Horvatovich et al. revealed that APOA5*2 haplotype (containing the minor alleles of rs662799, rs2072560 and rs2266788) confers susceptibility to development of obesity in pediatric patients . Additionally, in Aberle’s study, at baseline, TT homozygotes had a lower BMI compared to carriers of the C allele, although without a statistical significance . Until now much is learned about the biologic function of APOA5, however, little is known about the associations between APOA5 variants and body fat composition. Thus, mechanisms underlying our findings still remain hypothetic, requiring further work to prove. One speculated mechanism is that the effect of APOA5 gene variants on the risk of obesity might be attributed to this locus being in interaction with other obesity-risk genes, such as APOA1, APOC3 and APOA4 genes [15, 16]. Since the APOA1/C3/A4/A5 cluster occupies a restricted chromosomal region, phenotypic effects that appear to result from mutations in one member of the cluster might actually derive from allelic associations with functional variants in another . Other potential mechanisms could be linked to lipoprotein lipase (LPL) activation by the variable effects from multiple novel APOA5 variants, resulting in varying degrees of impaired lipolysis . As obesity is a complex, multifactorial condition for which heritability estimates vary widely. It might be the case in the association of obesity with APOA5 gene variants, as a result, further studies are needed to clarify this issue.
Although the association between APOA5 genetic variants and hypertriglyceridemia has been extensively studied [19–21] and also replicated in our study, there have been few studies, particularly in children and adolescents, on the association between APOA5 genetic variants and non-HDL-C levels. Furthermore, the relationships for TG or non-HDL-C were not eliminated by adjustment for age, sex and BMI, suggesting that this genetic factors act independently of age, sex or BMI. In consistent with our results, Hubacek et al. also suggested that the rs662799 variant in the APOA5 gene could be a significant genetic determinant for plasma non-HDL-C levels . So far as we know, there are scarce studies to provide some mechanism by which APOA5 gene variants could affect non-HDL-C levels. It is known that APOA5 protein accelerates plasma hydrolysis of TG-rich lipoproteins by activation of plasma LPL . We speculate that abnormalities in LPL activity could directly lead to delayed hydrolysis of TG-rich lipoproteins and elevated levels of intermediate metabolites in plasma, such as intermediate density lipoprotein (IDL), VLDL and chylomicrons, which resulted in an elevation of plasma non-HDL-C levels. Certainly, the detailed description of this hypothetic mechanismis needed for more intensive investigation.
The current study has several limitations that should be noted. First, the interactions between gene-gene, gene-environment and even different polymorphic loci of the same gene may influence the biological effects of the polymorphisms of the genes. Second, for all the polymorphisms, the number of the available studies in some subgroups, such as the control group was small. Third, due to the cross-sectional nature of the study design, the causal relationships between gene variants and clinical disorders can’t be drawn. Thus, we hope to perform full longitudinal association analyses in larger samples and more candidate genes in the future prospective studies. Furthermore, studies in the other populations will be useful to confirm our results.