Background:
TGF-β signal transduction initiates with receptor-mediated C-terminal phosphorylation of Smad2/3, subsequent complex formation with Smad4, and nuclear translocation for target gene regulation. Regulatory phosphorylation steps at serine/threonine residues in the Smad linker (L) domain create phosphorylated forms with distinct cellular functions, including transcriptional activity and proteasomal degradation. TGF-β signaling is impaired in several human tumors, however, the contribution of pSmad patterns to the development and progression of cancer is poorly understood. Here, we present evidence that Smad3 linker phosphorylation is a marker of cholangiocellular carcinoma (CCA).
Methods:
We examined healthy and tumorigenic liver tissues from CCA patients for the phosphorylation status of Smad3 at C-tail- and linker residues (S204, S208, and S213), using immunoblotting and immunohistochemistry with Smad phosphorylation site-specific antibodies. A kinase screen was performed in selected CCA cell lines using several inhibitors as well as a novel small molecule (designated as E738), targeting various kinases among others, GSK3 and MEK/ERK1/2. Readouts were Smad2/3 proteasomal degradation, Smad phosphorylation, cytotoxicity and apoptosis. We also included the kinase-inactive analogue of E738 (XC47) to further confirm the significance of kinase inhibition in the observed effects.
Results:
CCA patients and cell lines display variant Smad3 linker and C-terminal phosphorylation patterns. Tumorigenic areas from intrahepatic CCA patients present decreased pSmad3C and increased pSmad3L (S213 and 204) levels. In the CCSW1 cell line, we identified a counter-regulation of C-terminal versus linker phosphorylation, and GSK3, MEK/ERK1/2, and p38 MAPK as regulators of pSmad3L signalling. E738 was found to inhibit pSmad3L by Smad2/3 degradation, most likely through its kinase inhibitory function. Using a combination of inhibitors of the above kinases and E738, we observed synergistic effects on inhibition of pSmad3L protein levels, which coincided with reduced cellular viability and increased apoptotic cell death in CCA cell lines.
Conclusion:
We suggest that pSmad3L (S204) and pSmad3L (S213) are predominant in CCA and interfere with the cytostatic pSmad3C (S423/425) signaling, inducing a malignant switch in TGF-β's responses. We show that this could be translated into a therapeutic strategy through the inhibition of kinases involved in Smad3 linker phosphorylation.
