Genome-wide association study and meta-analysis of phytosterols identifies a novel locus for serum levels of campesterol

Abstract

Sitosterolemia is a rare inherited disorder caused by mutations in the ABCG5/ABCG8 genes. These genes encode proteins involved in the transport of plant sterols. Mutations in these genes lead to decreased excretion of phytosterols, which can accumulate in the body and lead to a variety of health problems, including premature coronary artery disease. We conducted the first genome-wide association study (GWAS) in the Middle East/North Africa population to identify genetic determinants of plant sterol levels in Qatari people. GWAS was performed on serum levels of β-sitosterol and campesterol using the Metabolon platform from Qatar Biobank (QBB) and genome sequence data provided by Qatar Genome Program. A trans-ancestry meta-analysis of data from our Qatari cohort with summary statistics from a previously published large cohort (9758 subjects) of European ancestry was conducted. Using conditional analysis, we identified two independent single nucleotide polymorphisms associated with β-sitosterol (rs145164937 and rs4299376), and two others with campesterol (rs7598542 and rs75901165) in the Qatari population in addition to previously reported variants. All of them map to the ABCG5/8 locus except rs75901165 which is located within the Intraflagellar Transport 43 (IFT43) gene. The meta-analysis replicated most of the reported variants, and our study provided significant support for the association of variants in SCARB1 and ABO with sitosterolemia. Evaluation of a polygenic risk score devised from European GWAS data showed moderate performance when applied to QBB (adjusted-R² = 0.082). These findings provide new insights into the genetic architecture of phytosterol metabolism while showing the importance including under-represented populations in future GWAS studies.

Keywords: Genetic predispositions; Genome-wide association study; Meta-analysis; Middle Eastern; Polygenic risk score; Sitosterolemia.

Fig. 1

Correlation analysis between β-sitosterol, Campesterol, and Total Cholesterol. A Correlation plot for campesterol versus $\beta$-sitosterol, B campesterol versus total cholesterol, C β-sitosterol versus total cholesterol. Pearson’s correlation analyses of β-sitosterol, Campesterol, and Total Cholesterol were presented for quality control passed by participants of the QBB (n = 1652; 1,451; 1652), respectively. Significant P values are revealed by P value less than 0.05. Abbreviations: R, Pearson’s rank coefficient; X, predictor (independent); Y, outcome (dependent)

Fig. 2

Genome-wide association results of metabolomic β-sitosterol and Campesterol levels were presented as Manhattan and Q–Q plots. The Manhattan plot illustrates the chromosomal positions of the genetic variants (N = 8,218,470) in relation to the –log10 P value. The analysis was performed using linear mixed models that were adjusted for age, sex, total cholesterol, the first four principal components. The horizontal red line in the plot represents the genome-wide significance threshold (P < 5.0 × 10^–8). The Q–Q plot displays the observed –log10 P values versus the expected –log10 P values

Fig. 3

Single-tissue violin plots for the independent SNPs. Gene expression of the associated variants (expression quantitative trait loci (eQTL)) using the Genotype-Tissue Expression (GTEx) portal. The data Source: GTEx Analysis Release V8 (dbGaP Accession phs000424.v8. p2). A rs6544713 (chr2_43846742_T_C). B rs4245791 (chr2_43847292_C_T). C rs4299376 (chr2_43845437_G_T). D rs7598542 (chr2_43866137_G_C)

Fig. 4

The performance of polygenic scores calculated from the European studies on the Qatari population. The relationship between A inverse-normalized baseline levels of β-sitosterol and weighted polygenic risk scores (PRS) derived from the Scholz et al. study, and B inverse-normalized baseline levels of campesterol and weighted polygenic risk scores (PRS) derived from the GWAS catalog. Regression analysis was conducted to assess this correlation, with the blue line representing the most suitable fit