Resequencing of LPL in African Blacks and associations with lipoprotein–lipid levels

Abstract

Genome-wide association studies have identified several loci associated with plasma lipid levels but those common variants together account only for a small proportion of the genetic variance of lipid traits. It has been hypothesized that the remaining heritability may partly be explained by rare variants with strong effect sizes. Here, we have comprehensively investigated the associations of both common and uncommon/rare variants in the lipoprotein lipase (LPL) gene in relation to plasma lipoprotein–lipid levels in African Blacks (ABs). For variant discovery purposes, the entire LPL gene and flanking regions were resequenced in 95 ABs with extreme high-density lipoprotein cholesterol (HDL-C) levels. A total of 308 variants were identified, of which 64 were novel. Selected common tagSNPs and uncommon/rare variants were genotyped in the entire sample (n=788), and 126 QC-passed variants were evaluated for their associations with lipoprotein–lipid levels by using single-site, haplotype and rare variant (SKAT-O) association analyses. We found eight not highly correlated (r2<0.40) signals (rs1801177:G>A, rs8176337:G>C, rs74304285:G>A, rs252:delA, rs316:C>A, rs329:A>G, rs12679834:T>C, and rs4921684:C>T) nominally (P<0.05) associated with lipid traits (HDL-C, LDL-C, ApoA1 or ApoB levels) in our sample. The most significant SNP, rs252:delA, represented a novel association observed with LDL-C (P=0.002) and ApoB (P=0.012). For TG and LDL-C, the haplotype analysis was more informative than the single-site analysis. The SKAT-O analysis revealed that the bin (group) containing 22 rare variants with MAF≤0.01 exhibited nominal association with TG (P=0.039) and LDL-C (P=0.027). Our study indicates that both common and uncommon/rare LPL variants/haplotypes may affect plasma lipoprotein–lipid levels in general African population.

A novel LPL intronic variant: g.18704C>A identified by re-sequencing Kuwaiti Arab samples is associated with high-density lipoprotein, very low-density lipoprotein and triglyceride lipid levels

“…In addition, sequence analysis of the LPL gene has revealed over one hundred single nucleotide polymorphisms (SNPs) within both the coding and noncoding region [ 8 , 14 – 19 ]. Most variants that have been identified appear to have been based almost entirely on populations of European and/or African descent explaining <10% of the heritability factor on plasma lipid levels in the studied population [ 3 , 5 , 17 , 18 ].…”

Exautomate: A user-friendly tool for region-based rare variant association analysis (RVAA)

“…Particularly with SKAT and SKAT-O, even slight parameter adjustments at the outset can produce significant differences in output. In a substantial search of published articles reporting the use of SKAT or SKAT-O for different phenotypes-including cardiovascular disease, body-mass index, height, amyotrophic lateral sclerosis, red blood cell traits, Alzheimer's disease, Parkinson's disease, lipid traits, and blood pressure-there are virtually no statements or reporting of the specific parameters used, other than perhaps the variant frequency threshold [19][20][21][22][48][49][50][51][52][53][54] . Unfortunately, this trend of minimal methodological information neither instills confidence in the research, nor does it facilitate replications of results.…”

“…One method frequently used for region-based RVAA is the sequence kernel association test (SKAT) [15][16][17][18] . While the method has been useful in revealing rare variants and genomic loci of interest in complex traits and diseases-such as cardiovascular disease, body-mass index, height, and neurodegenerative diseases [19][20][21][22] -reproducing these results can be difficult. SKAT is challenging to implement for exome-and genome-scale analyses as it requires significant data preprocessing involving additional software dependencies and variables that can complicate reproducibility 23 .…”

Ultrassequenciamento exômico dos principais genes relacionados com a hipercolesterolemia familial

“…As variantes APOA5 rs2075291 (c.553G>A; p.Gly185Ser), rs3135506 (c.56C>G; p.Ser19Trp) e rs34282181 (c.111C>A; p.Asp37Glu); APOC1 rs12721054 (c.-100A>G); LPL rs268 (c.953A>G; p.Asn318Ser) e rs5934 (c.1279G>A; p.Ala427Thr) foram previamente associadas com aumento de TG (tabela 9) (ELBERS et al, 2012;PIRIM et al, 2015;RABACCHI et al, 2015;VAN LEEUWEN et al, 2016;XIAO et al, 2017). Mas apenas a variante APOA5 rs3135506 (c.56C>G; p.Ser19Trp) também foi associada com altas concentrações de TG nas análises estatísticas deste estudo (quadro 6 anexo H).…”

“…A associação com LDL-c é controversa entre os estudos, no entanto, como previamente discutido os mecanismos relacionados a este aumento são plausíveis (QIAN et al, 2018;YOU et al, 2018). LUIS et al, 2018;HASCHKE-BECHER et al, 2010;HEIDARI-BENI et al, 2015;PIRIM et al, 2015;SCHARNAGL et al, 2014;SMITH et al, 2010;ZAMBRANO et al, 2015). Outra variante previamente associada com aumento das concentrações de apoB-100 e CT foi a APOA4 rs5110 (c.1140G>T; p.Gln380His) (tabela 9).…”