The effect of transforming growth factor beta on human neuroendocrine tumor BON cell proliferation and differentiation is mediated through somatostatin signaling.

The dual effect of the ubiquitous inflammatory cytokine transforming growth factor beta1 (TGF beta) on cellular proliferation and tumor metastasis is intriguing but complex. In epithelial cell- and neural cell-derived tumors, TGF beta serves as a growth inhibitor at the beginning of tumor development but later becomes a growth accelerator for transformed tumors. The somatostatin (SST) signaling pathway is a well-established antiproliferation signal, and in this report, we explore the interplay between the SST and TGF beta signaling pathways in the human neuroendocrine tumor cell line BON. We defined the SST signaling pathway as a determinant for neuroendocrine tumor BON cells in responding to TGF beta as a growth inhibitor. We also determined that TGF beta induces the production of SST and potentially activates the negative growth autocrine loop of SST, which leads to the downstream induction of multiple growth inhibitory effectors: protein tyrosine phosphatases (i.e., SHPTP1 and SHPTP2), p21(Waf1/Cip1), and p27(Kip1). Concurrently, TGF beta down-regulates the growth accelerator c-Myc protein and, collectively, they establish a firm antiproliferation effect on BON cells. Additionally, any disruption in the activation of either the TGF beta or SST signaling pathway in BON leads to "reversible" neuroendocrine-mesenchymal transition, which is characterized by the loss of neuroendocrine markers (i.e., chromogranin A and PGP 9.5), as well as the altered expression of mesenchymal proteins (i.e., elevated vimentin and Twist and decreased E-cadherin), which has previously been associated with elevated metastatic potential. In summary, TGF beta-dependent growth inhibition and differentiation is mediated by the SST signaling pathway. Therefore, any disruption of this TGF beta-SST connection allows BON cells to respond to TGF beta as a growth accelerator instead of a growth suppressor. This model can potentially apply to other cell types that exhibit a similar interaction of these pathways.