Breast cancer is still the most common type of cancer in women; an important role in carcinogenesis is actually attributed to cancer-associated fibroblasts. In this study, we investigated whether it is possible to assess the functional state of cancer-associated fibroblasts through tumor tissue proteome profiling. Tissue proteomics was performed on tumor-central, tumor-near, and tumor-distant biopsy sections from breast adenocarcinoma patients, which allowed us to identify 2074 proteins. Data were interpreted referring to reference proteome profiles generated from primary human mammary fibroblasts comprising 4095 proteins. These cells were analyzed in quiescent cell state as well as after in vitro treatment with TGFβ or IL-1β, stimulating wound healing or inflammatory processes, respectively. Representative for cancer cells, we investigated the mammary carcinoma cell line ZR-75-1, identifying 5212 proteins. All mass analysis data have been made fully accessible via ProteomeXchange, DOI PXD001311 and PXD001323-8. Comparison of tissue proteomics data with all of those reference profiles revealed predominance of cancer cell-derived proteins within the tumor and fibroblast-derived proteins in the tumor-distant tissue sections. Remarkably, proteins characteristic for acute inflammation were hardly identified in the tissue samples. In contrast, several proteins found by us to be induced by TGFβ in mammary fibroblasts, including fibulin-5, SLC2A1, and MUC18, were positively identified in all tissue samples, with relatively higher abundance in tumor neighboring tissue sections. These findings indicate a predominance of cancer-associated fibroblasts with wound healing activities localized around tumors.
Keywords: breast cancer; cancer-associated fibroblasts; comparative proteome profiling; label-free quantification; primary human cells; tissue proteomics.