BackgroundIntra-tumoral genetic and functional heterogeneity correlates with cancer clinical prognoses. However, the mechanisms by which intra-tumoral heterogeneity impacts therapeutic outcome remain poorly understood. RNA sequencing (RNA-seq) of single tumor cells can provide comprehensive information about gene expression and single-nucleotide variations in individual tumor cells, which may allow for the translation of heterogeneous tumor cell functional responses into customized anti-cancer treatments.ResultsWe isolated 34 patient-derived xenograft (PDX) tumor cells from a lung adenocarcinoma patient tumor xenograft. Individual tumor cells were subjected to single cell RNA-seq for gene expression profiling and expressed mutation profiling. Fifty tumor-specific single-nucleotide variations, including KRASG12D, were observed to be heterogeneous in individual PDX cells. Semi-supervised clustering, based on KRASG12D mutant expression and a risk score representing expression of 69 lung adenocarcinoma-prognostic genes, classified PDX cells into four groups. PDX cells that survived in vitro anti-cancer drug treatment displayed transcriptome signatures consistent with the group characterized by KRASG12D and low risk score.ConclusionsSingle-cell RNA-seq on viable PDX cells identified a candidate tumor cell subgroup associated with anti-cancer drug resistance. Thus, single-cell RNA-seq is a powerful approach for identifying unique tumor cell-specific gene expression profiles which could facilitate the development of optimized clinical anti-cancer strategies.Electronic supplementary materialThe online version of this article (doi:10.1186/s13059-015-0692-3) contains supplementary material, which is available to authorized users.

Genetic and epigenetic intra-tumoral heterogeneity cooperate to shape the evolutionary course of cancer 1 . Chronic lymphocytic leukemia (CLL) is a highly informative model for cancer evolution as it undergoes substantial genetic diversification and evolution with therapy 2 , 3 . The CLL epigenome is also an important disease-defining feature 4 , 5 , and growing CLL populations diversify through stochastic DNA methylation (DNAme) changes – epimutations 6 . However, previous studies based on bulk DNAme sequencing could not answer whether epimutations affect CLL populations homogenously. To measure epimutation rate at single-cell resolution, we applied multiplexed single-cell reduced representation bisulfite sequencing (MscRRBS) to healthy donors B cell and CLL patient samples. We observed that the common clonal CLL origin results in consistently elevated epimutation rate, with low cell-to-cell epimutation rate variability. In contrast, variable epimutation rates across normal B cells reflect diverse evolutionary ages across the B cell differentiation trajectory, consistent with epimutations serving as a molecular clock. Heritable epimutation information allowed high-resolution lineage reconstruction with single-cell data, applicable directly to patient samples. CLL lineage tree shape revealed earlier branching and longer branch lengths than normal B cells, reflecting rapid drift after the initial malignant transformation and a greater proliferative history. MscRRBS integrated with single-cell transcriptomes and genotyping confirmed that genetic subclones map to distinct clades inferred solely based on epimutation information. Lastly, to examine potential lineage biases during therapy, we profiled serial samples during ibrutinib-associated lymphocytosis, and identified clades of cells preferentially expelled from the lymph node with therapy, marked by distinct transcriptional profiles. The single-cell integration of genetic, epigenetic and transcriptional information thus charts CLL’s lineage history and its evolution with therapy.

Mutations in genes involved in DNA methylation (DNAme; e.g., TET2, DNMT3A) , are frequently observed in hematological malignancies 1 – 3 and clonal hematopoiesis 4 , 5 . Applying single-cell sequencing to murine hematopoietic stem and progenitor cells, we observed that these mutations disrupt hematopoietic differentiation, causing opposite shifts in the frequencies of erythroid vs. myelo-monocytic progenitors upon Tet2 or Dnmt3a loss. Notably, these shifts trace back to transcriptional priming skews in uncommitted hematopoietic stem cells (HSCs). To reconcile genome-wide DNAme changes with specific erythroid vs. myelo-monocytic skews, we provide evidence in support of differential sensitivity of transcription factors due to biases in CpG enrichment in their binding motif. Single-cell transcriptomes with targeted genotyping showed similar skews in transcriptional priming of DNMT3A -mutated human clonal hematopoiesis bone marrow progenitors. These data show that DNAme shapes the hematopoietic differentiation topography, and support a model in which genome-wide methylation changes are transduced to differentiation skews through biases in transcription factor binding-motif CpG enrichment.