TGFβ Signaling in Prostate Stroma Supports Carcinoma Growth by Upregulating Stromal Genes Related to Tissue Remodeling

Increasing evidence points to an active stromal involvement in cancer initiation and progression. Cytokines derived from tumor cells are believed to modulate stromal cells to produce growth and angiogenic factors which in turn provide the tumor with the necessary microenvironment for expansion and invasion. Transforming growth factor beta (TGFβ) has been implicated as a candidate cytokine to mediate this communication. However, how its signaling in stromal cells regulates tumorigenesis and tumor progression remains unresolved. We show that normal, pre-senescent fibroblasts or prostate stromal cells cotransplanted with prostate carcinoma cells subcutaneously into nude mice reduced tumor latency and accelerated tumor growth. When their TGFβ signaling was blocked, the fibroblasts and stromal cells still stimulated tumor initiation, but no longer supported tumor growth as control cells did. The loss of the tumor growth-promoting activity of the stromal cells with attenuated TGFβ signaling was not associated with altered cellular senescence or tumor angiogenicity. TGFβ and the medium conditioned by the prostate carcinoma cells stimulated myofibroblast differentiation of the intact stromal cells, but not the stromal cells with attenuated TGFβ signaling. Gene microarray and quantitative RTPCR analyses showed that TGFβ upregulated a host of genes in stromal cells that are involved in tissue remodeling and wound healing. Thus, our study provides evidence for TGFβ as a supporting agent in tumor progression through the induction of a perpetual wound healing process in the tumor microenvironment. Keywords: Human prostate stromal cells (PrSC) Total RNA was extracted from confluent PrSC cells and DNRII transfected PrSC cells with and without 24-hour treatment with 5.0 ng/mL TGFb3. The RNA was purified using a RiboPure™Kit from Ambion, Inc., following the manufacturer’s suggested protocol. Triplicate sample hybridizations were done for each comparison. TGFbeta treated PrSC cell versus non treated PrSC cells and TGFbeta treated DNRII PrSC cell versus non treated PrSC cells. For each comparison, one dye swap experiment was performed. The raw data for the dye swap experiment was reversed so that the data match the other two replicates.