In this study we used the novel approach of searching for common polymorphisms located in the 5’UTR region of the LDLR that might be involved in the pathogenesis of primary hypercholesterolemia because of their possible influence in the transcription of the LDLR gene. The major conclusions drawn from this study are that, the LDLR g.3131C > T polymorphism in the 5’ region of the LDLR is responsible for a major effect on LDL-C concentration because of the increase in the transcriptional activity of the LDLR gene, and also indicates that the LDLR variation is associated with PH.
It has been shown that 58.9% of the variants involved in regulation changes are located within the first 500 bp upstream of the transcription start site, but 12.8% are more than 1 kb upstream and show transcription factor switching, with each allele having a higher affinity for a different transcription factor [15, 28]. As the LDLR promoter is well-conserved in most of its positions among chimps and humans, the sequencing of a 3103 pb fragment, from -625 to +2478 of the ENSG00000130164 reference, showed only 6 variants, all located within the first and second introns of the LDLR gene and with ≤ 8.5% minor allele frequency.
Simultaneously to the sequencing of the LDLR promoter we analyzed four polymorphisms that appeared to be relevant in the regulation of LDLR expression according to the bioinformatics analysis. We found that only the variant g.3131C > T had a different distribution of genotypes between a control group of 525 individuals carefully selected among a working population of 1137 subjects and a group of 477 hypercholesterolemic individuals without known FH-causing mutations (p < 0.00001). A lower frequency of the minor T allele of this SNP was observed in the total hypercholesterolemic population and also in each of the three subgroups taken separately. Surprisingly, the T-allele is also the ancestral one, as it has been observed by sequence comparison between 20 eutherian mammals, and, according to our results, this allele would have a protective effect against hypercholesterolemia.
The g.3131C > T is placed in a large linkage disequilibrium (LD) block. In this work we had sequenced this LD region in 125 subjects, including the rs6511720 SNP, and we have calculated the LD between these SNPs (r2 = 0.211 and D’ = 0.487) not observing an association in this group.
A limitation to the study would be the diagnosis with the LIPOchip® in the PH subjects as non-carriers of FH causal mutations. It would be remotely possible to have missed a variant located in a region not sequenced with this platform, although coefficients for specificity and sensitivity are 99.7% and 99.9%, respectively . Another limitation to the study is the different gender distribution in the PH and control groups. However, regression analyses in men and in the overall population showed a similar independent inverse association of the LDLR g.3131C > T variant with LDL-C that, together with the lipoprotein(a) concentration and age, explained ≈ 10% of LDL-C variability. The increase of the ancestral (and minor) allele frequency in populations with a lower concentration of LDL-C comparing to PH group and in replication cohorts suggests that T-allele of the g.3131C > T variant in the LDLR gene protects against hypercholesterolemia. This observation might be explained by the lower affinity of the g.3131 T variant for some repressors of LDLR gene transcription, and as a consequence, it would increase the transcriptional activity of the LDLR gene comparing with the wild-type allele (Figure 1).
Taking together the genetic analyses and the functional assays, our results suggest that the ancestral and minor allele T at g.3131C > T is associated with LDL-C levels and explains part of the LDL-C variability. This variant is located in a regulatory element, and the shift of nucleotide C to T produce a change in the affinity for transcription factors as well as an increase of 2.5 times in the transcriptional activity of the LDLR gene explaining the protective effect of the T allele. Further studies aimed to expand knowledge on the regulatory elements in the distal region of the LDLR promoter are warranted. Future studies should also consider the effect of this variant on the LDL-C response to lipid-lowering drugs or lifestyle changes.
The association of the rs17248720 polymorphism with hypercholesterolemia confirms previous observations from GWAS suggesting that non-coding variants located at the LDLR locus are associated with blood lipid levels, including LDL-C [8, 9, 12, 13]. While studies attempt to find new loci implicated in the hereditability of lipid levels, the strongest locus associated with LDL-C continues to be that containing the LDLR gene .