Abstract LB-49: Defining hierarchical organization of breast cancer cells and cancer activated fibroblasts using single-cell gene expression profiling: switch-like and transient transitions between novel stem cell and progenitor pools

Abstract

Abstract The cancer stem cell (CSC) model holds that cancers are organized in a hierarchical structure and derived from cellular populations with stem cell properties. CSC characteristics include self-renewal, tumorigenicity, multi-lineage differentiation as well as increased resistance to radio- and chemotherapy-induced cell death. In order to better understand the complex hierarchical structure of breast cancer cells in vivo as well as phenotypes of stromal cells as cancer associated fibroblasts, tools need to be developed, enabling analyses of key regulators and biomarkers on a single-cell level. We therefore established single-cell reverse transcription quantitative real-time PCR and gene expression profiling, including more than 100 key regulators associated with the cell cycle, differentiation, EMT, stemness and fibroblast activation and performed extensive single-cell PCR analyses, using cell lines and primary breast cancer samples. Three strategies were used to enrich for cancer stem cells and progenitor populations corresponding to 1) cells surviving anchorage-independent growth conditions, 2) cells in mammospheres showing few cell divisions, defined by a cell division tracker and 3) cells grown under hypoxic conditions. All methods have earlier been shown to enrich for cells that have an increased cancer initiating ability in mouse model systems. Interestingly, all strategies increased two separate tumor subpopulations defined by lower proliferative features and various EPCAM, ERBB2, CDH1, ID1, ABCG2, POU5F1, NANOG and SOX2 levels. Besides, a clear switch-like transition between the various populations, a transition state linking the cancer stem cell and progenitor populations with the high proliferative main cluster was observed. When specifically analyzing the few cells that had only undergone one asymmetrical cell division in 5 days mammosphere cultures, it was clear that this potential stem cell pool had a distinctive expression profile based on the included key regulators. We next analyzed immortalized primary fibroblasts and tumor activated counterparts using single-cell PCR analyses. The two cell types clearly separated from each other and showed various activation states defined by key regulators as COL1A1, ALDH1A3, TGFB, PDGFA, ACTA2, CXCL12 and cell cycle regulators. The data support a continuing transition towards myofibroblast differentiation with gradual loss of proliferation, whereas the activation markers peak in an intermediate state before entering a final low proliferative condition. When analyzing primary tumors we observed distinct clusters of epithelial cells and fibroblasts supporting the validity of the defined assay. A detailed description of primary breast cancer samples and important microenvironmental components combined with functional assays on a single cell level are now ongoing. Within this comprehensive program we have identified diverse previously unknown CSC/progenitor pools or distinct cellular states and we are now able to model how cells actually transfer between various differentiation and stem cells states, thereby better understanding the hierarchical composition of breast cancer on a single cell level. Citation Format: Anders Ståhlberg, Nina Akrap, Daniel Andersson, Pernilla Gregersson, Susann Busch, Göran P. Landberg. Defining hierarchical organization of breast cancer cells and cancer activated fibroblasts using single-cell gene expression profiling: switch-like and transient transitions between novel stem cell and progenitor pools. [abstract]. In: Proceedings of the 105th Annual Meeting of the American Association for Cancer Research; 2014 Apr 5-9; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2014;74(19 Suppl):Abstract nr LB-49. doi:10.1158/1538-7445.AM2014-LB-49