Identifying key regulatory genes in drug resistance acquisition: Modeling pseudotime trajectories of single-cell transcriptome

Abstract

Single-cell RNA-sequencing (scRNA-seq) technology has provided significant insights into cancer drug resistance at the single-cell level. However, understanding dynamic cell transitions at the molecular systems level remains limited, requiring a systems biology approach. We present an approach that combines mathematical modeling with pseudotime analysis using time-series scRNA-seq data obtained from the breast cancer cell line MCF-7 treated with tamoxifen. Our single-cell analysis identified five distinct subpopulations, including tamoxifen-sensitive and -resistant groups. Using a single-gene mathematical model, we discovered approximately 560-680 genes out of 6,000 exhibiting multistable expression states in each subpopulation, including key estrogen receptor-positive breast cancer cell survival genes, such as RPS6KB1. Bifurcation analysis elucidated their regulatory mechanisms, and we mapped these genes into a molecular network associated with cell survival and metastasis-related pathways. Our modeling approach comprehensively identifies key regulatory genes for drug resistance acquisition, enhancing our understanding of potential drug targets in breast cancer. SummaryDespite recent advancements in experimental technology for genome-wide molecular profiling, our understanding of the dynamic mechanism underlying cancer drug resistance remains limited. In this study, we present an approach that combines mathematical modeling with the pseudotime analysis of single-cell time-series transcriptome data of drug-treated breast cancer cells. Our method identifies approximately 600 genes out of 6,000 exhibiting multistable expression states, including RPS6KB1, a predictor of poor prognosis, cell survival, and growth in estrogen receptor-positive breast cancers. The bifurcation analysis elucidates the regulatory mechanisms of the key regulatory genes, which can also be mapped into a molecular network based on cell survival and metastasis-related pathways, providing a comprehensive understanding of the interplay between signaling pathways and regulatory genes. Our method serves as a powerful tool for deciphering the complexities of drug resistance mechanisms in human diseases.