*3.4. ABCA1, LDL-C and Triglyceride (TG) Serum Levels*

While the effect of ABCA1 on HDL-C plasma levels is clear, the effect of ABCA1 loss of function on other lipid traits is less evident. Several authors report that TD patients have significantly elevated plasma TG levels and reduced LDL-C concentrations than normal controls, although plasma TG levels vary in these patients [57–60]. Clee et al. reported overall higher TG levels in subjects heterozygous for *ABCA1* mutations than in controls, although TG levels were variable and not elevated in all mutation carriers [61]. Using an extreme phenotype approach, Frikke-Schmidt et al. described nine patients heterozygous for *ABCA1* mutations with very low HDL-C levels, six of which had elevated TG levels (>2.2 mmol/L) [12]. In contrast, heterozygous carriers of *ABCA1* mutations have no significant change in LDL-C levels [62–64].

Most GWAS have not reported *ABCA1* as a locus associated with TG and LDL-C levels [65–68]. However, recent multiethnic GWAS, including hundreds of thousands of cases and controls, have identified different *ABCA1* variants associated with TG (rs2575876, rs1799777, rs1883025 and rs1800978) and LDL-C levels (rs7873387, rs2575876, rs2740488, rs11789603, rs2066714) with genome-wide significance, although with small effect sizes [69–72]. In one of these studies, associations with TG levels were observed in current drinkers and/or regular drinkers [69]. In addition, several candidate gene studies also reported *ABCA1* SNPs associated with LDL-C and TG levels, with inconsistent results. While various studies failed to find associations of *ABCA1* gene variation with TG and/or LDL-C levels [40,73–75], *ABCA1* polymorphisms were associated with TG levels but not with LDL-C levels in Brazilians [76], Chinese [77], Turkish [78], Iranians [79] and Mexican school-aged children [80]. The functional *ABCA1*/R230C variant was associated with lower triglyceride levels only in Pimas and Mayans, but not in Mexican mestizos [41]. Moreover, *ABCA1* gene variants have been associated with LDL-C but not TG levels in a cohort of Greek nurses [81], and in male individuals with hypercholesterolemia [82]. In a large multiethnic cohort studying over 150 common variants, *ABCA1* was associated with both TG and LDL-C levels [83].

Postprandial hypertriglyceridemia is an important factor in developing atherosclerotic plaque and is closely related to the occurrence of cardiovascular events [84,85]. Although TG levels are usually estimated in a fasting state, several epidemiological studies have demonstrated that non-fasting hyperlipidemia is more harmful [86–88]. The high interindividual variability of TG levels observed in TD may be due to the inherent heterogeneity in individual triglyceride levels in different postprandial dietary lipid absorption states [89,90]. While several studies have documented single candidate SNPs associated with postprandial TG metabolism modulation [91,92], studies analyzing the effect of *ABCA1* gene variants on postprandial lipid metabolism are scarce. A recent study identified that most of the interindividual variability in the postprandial chylomicron TG response to dietary fat in healthy male adults could be explained by a combination of 42 SNPs in 23 genes, including *ABCA1* [91]. Moreover, Delgado-Lista et al. showed that major allele homozygotes for rs2575875 and rs4149272 had lower postprandial increases in TG and large-triglyceride rich lipoproteins, suggesting these variants may regulate the clearance of postprandial triglycerides [93].

It is evident that altered ABCA1 function and gene variation do not always affect TG levels. This may have to do with ethnicity, sex-specific effects and with interactions with other gene variants and environmental factors. In this regard, a small number of studies have reported interactions between *ABCA1* gene variants and dietary macronutrient proportions affecting plasma TG levels. In the Mexican population, premenopausal women carrying the *ABCA1*/R230C risk allele and consuming higher carbohydrate/lower fat diets

showed an unfavorable metabolic pattern including higher TG levels, with a statistically significant interaction [43]. An independent study in the Inuit population also reported an interaction between the *ABCA1*/R219K variant with saturated fat intake affecting plasma TG levels [94]. These facts indicate that gene-diet interactions may help better predict inter-individual variations in plasma lipid levels and may provide information useful to design diet intervention studies.
