Abstract
Intratumoral genetic heterogeneity has been characterized across cancers, including chronic lymphocytic leukemia (CLL), and presents challenges to therapy as presence of phenotypically diverse cell populations commonly fuels relapse and resistance to treatment. While insights have been previously gained from bulk analysis, further characterization on the single-cell level is needed to more accurately dissect the pathway and regulatory features associated with distinct genetic subclones. In order to more accurately identify subclones, define phylogenetic relationships, and probe genotype-phenotype relationships, we developed methods for targeted mutation detection in DNA and RNA isolated from thousands of single cells from five CLL samples.
To precisely establish the subclonal architecture of the five CLL samples, we first applied targeted single-cell DNA sequencing for detection of mutation and copy number variation using CLL samples previously characterized by bulk whole exome sequencing (WES). Our results identified multiple genetically distinct subclones in four out of five CLL samples. Next, to understand the subclonal architecture and their associated gene expression pattern, we performed single-cell whole transcriptome sequencing (scRNA-seq) and pathway and geneset overdispersion analysis. Although scRNA-seq identified clear transcriptional heterogeneity, these data could neither be used to confidently resolve the genetic subclone structure nor assess the correspondence of genetic structure with the observed transcriptional heterogeneity due to sparseness of coverage at mutation sites of interest. Thus, we developed a targeted RNA approach to assess transcript expression profiles and mutational status in the same single cells. Our approach reliably detects subclonal mutations and enables recapitulation of single-cell DNA information, including phylogenetic structure. Integrative analysis to correlate genotype and phenotype reveals phenotypic convergence between distinct subclones and prompts the idea that convergent evolution generates phenotypically similar cells in distinct genetic branches, thus creating a cohesive expression profile in each CLL sample despite the presence of genetic heterogeneity. Specifically, by clearly resolving phylogenic relationships and simultaneously assessing DNA and RNA-level information from the same single cells, we uncovered mutated LCP1 and WNK1 as novel CLL drivers, supported by functional evidence demonstrating their impact on CLL pathways.
Overall, we demonstrate the ability to robustly integrate DNA- and RNA-level information in order to dissect the impact of somatic mutations on cellular phenotype. Our study highlights the potential for single-cell RNA-based targeted analysis to sensitively determine transcriptional and mutational profiles of individual cancer cells leading to an increased understanding of driving events in malignancy.
Citation Format: Jean Fan, Lili Wang, Joshua M. Francis, George Georghiou, Sarah Hergert, Shuqiang Li, Rutendo Gambe, Chensheng W. Zhou, Chunxiao Yang, Sheng Xiao, Paola Dal Chin, Michaela Bowden, Dylan Kotliar, Sachet A. Shukla, Jennifer R. Brown, Donna Neuberg, Dario R. Alessi, Cheng-Zhong Zhang, Peter V. Kharchenko, Kenneth J. Livak, Catherine J. Wu. Integrated analysis of targeted single-cell genetic and transcriptional heterogeneity suggests novel drivers of chronic lymphocytic leukemia [abstract]. In: Proceedings of the Second AACR Conference on Hematologic Malignancies: Translating Discoveries to Novel Therapies; May 6-9, 2017; Boston, MA. Philadelphia (PA): AACR; Clin Cancer Res 2017;23(24_Suppl):Abstract nr 27.