Abstract
Statins are the most commonly prescribed cardiovascular disease drug, but their inter-individual efficacy varies considerably. Genetic factors uncovered to date have only explained a small proportion of variation in low-density lipoprotein cholesterol (LDLC) lowering. To identify novel markers and determinants of statin response, we used whole transcriptome sequence data collected from simvastatin and control incubated lymphoblastoid cell lines (LCLs) established from participants of the Cholesterol and Pharmacogenetics (CAP) simvastatin clinical trial. We looked for genes whose statin-induced expression changes were most different between LCLs derived from individuals with high versus low plasma LDLC statin response during the CAP trial. We created a classification model of 82 “signature” gene expression changes that distinguished high versus low LDLC statin response. One of the most differentially changing genes was zinc finger protein 542 pseudogene (ZNF542P), the signature gene with changes most correlated with statin-induced change in cellular cholesterol ester, an in vitro marker of statin response. ZNF542P knock-down in a human hepatoma cell line increased intracellular cholesterol ester levels upon simvastatin treatment. Together, these findings imply a role for ZNF542P in LDLC response to simvastatin and, importantly, highlight the potential significance of noncoding RNAs as a contributing factor to variation in drug response.
Highlights
Statins, the most widely prescribed drugs for reducing cardiovascular disease risk, exhibit wide inter-patient variability in their efficacy with regard to the magnitude of plasma low-density lipoprotein cholesterol (LDLC) reduction[1,2]
We first identified signature genes from lymphoblastoid cell lines (LCLs) established from the tails of the European American LDLC response distribution, and we tested for replication in the extremes of the African American LDLC response distribution
Using the expression changes of the identified European American signature genes that were identified as predictors in radial-basis support vector machine (SVM) classification models, we trained and predicted their ability to distinguish 12 high and 14 low African American responders (Supplementary Fig. S1b) whose LDLC statin response was as extreme as the European American subset
Summary
The most widely prescribed drugs for reducing cardiovascular disease risk, exhibit wide inter-patient variability in their efficacy with regard to the magnitude of plasma low-density lipoprotein cholesterol (LDLC) reduction[1,2]. Genome-wide association studies (GWAS) have identified a modest number of SNPs associated with LDLC statin response in or near genes such as APOE, LPA, ABCG2, SORT1, and SLCO1B13–6. These variants together account for only about 5% of the variance in LDLC statin response[6]. Given the success of RNA expression profiling of patient-derived cell lines for the identification of novel drug response genes, here we sought to extend our findings from the CAP LCL expression array studies by using whole transcriptome sequencing (RNA-seq) data from in vitro simvastatin versus control incubated cell lines established from 104 European American and 53 African American ancestry CAP participants. Zinc finger protein 542 pseudogene (ZNF542P) emerged as a novel candidate gene implicated in LDL simvastatin response
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
