Abstract
Genotyped human B-lymphoblastoid cell lines (LCLs) are widely used models in mapping quantitative trait loci for chromatin features, gene expression, and drug response. The extent of genotype-independent functional genomic variability of the LCL model, although largely overlooked, may inform association study design. In this study, we use flow cytometry, chromatin immunoprecipitation sequencing and mRNA sequencing to study surface marker patterns, quantify genome-wide chromatin changes (H3K27ac) and transcriptome variability, respectively, among five isogenic LCLs derived from the same individual. Most of the studied LCLs were non-monoclonal and had mature B cell phenotypes. Strikingly, nearly one-fourth of active gene regulatory regions showed significantly variable H3K27ac levels, especially enhancers, among which several were classified as clustered enhancers. Large, contiguous genomic regions showed signs of coordinated activity change. Regulatory differences were mirrored by mRNA expression changes, preferentially affecting hundreds of genes involved in specialized cellular processes including immune and drug response pathways. Differential expression of DPYD, an enzyme involved in 5-fluorouracil (5-FU) catabolism, was associated with variable LCL growth inhibition mediated by 5-FU. The extent of genotype-independent functional genomic variability might highlight the need to revisit study design strategies for LCLs in pharmacogenomics.
Highlights
In the past few years, the biomedical research community empowered by high-throughput technologies have gained a broader appreciation for the importance of intra- and inter-cell line variability and dynamics
Using cell lines prepared from the same individual and excluding cell culturing-related confounders, sGT lymphoblastoid cell lines (LCLs) represent a suitable model for exploring molecular phenotype variability among LCLs
Pairwise comparison of sGT LCLs, indicating the number of differentially expressed genes (DEGs) per sGT LCL pair. (c) The number of genes expressed in a certain number of sGT LCLs (CPM >5 in both replicates)
Summary
In the past few years, the biomedical research community empowered by high-throughput technologies have gained a broader appreciation for the importance of intra- and inter-cell line variability and dynamics. The main advantages of this cell line model include easy culturing and manipulation, stable karyotype, and publicly available genotype information for a large number of LCLs for molecular association and pharmacogenomics. Based on the above observations, many studies assume that LCLs from different individuals offer a good model for how genotypes affect functional genomic patterns and the response to environmental cues, including various pharmaceuticals. At least half of the catalogued genes are expressed in LCLs, rendering it an appropriate model for finding associations between genetic and transcriptomic signatures versus drug response for lymphoproliferative and non-B cell-related diseases[42]. We used a model of five LCLs derived from the same individual (isogenic, bearing identical genomic material), prepared in the same laboratory using the same EBV strains to characterise inter-cell line variability at the epigenetic, transcriptomic, and protein surface marker levels. Our results have implications for pharmacogenomics research, supporting a more widespread adoption of the triangle model, which considers baseline RNA levels in LCL-based study design in pharmacogenomics
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
