Pancreatic cancer still remains one of the most aggressive and lethal of human cancers, with a lack of effective treatments contributing to its poor prognosis and high mortality. Perturbations in tyrosine kinase signaling are characteristic of many human cancers. In order to detect aberrant protein tyrosine phosphorylation events in pancreatic cancer, and to subclassify pancreatic cancer cell lines based on their tyrosine phosphorylation patterns, tryptic digests from a wide panel of pancreatic cancer cell lines were subjected to anti-phosphotyrosine immunoprecipitation, then analysed by liquid chromatography coupled with tandem mass spectrometry (MS/MS). Proteins were ranked according to differences in median phosphorylation levels across the cell line panel, and those that exhibited significantly variable phosphorylation levels were selected as candidates for further evaluation as biomarkers and therapeutic targets.
One candidate identified by this approach was SgK223, a 175 kDa atypical kinase with substitutions in the DFG motif of its kinase domain. Since this is one of the conserved motifs critical for full catalytic activity, SgK223 may be catalytically inactive. Western blotting revealed that SgK223 protein expression varied across the panel, with a subset of pancreatic cancer lines exhibiting overexpression relative to normal human pancreatic duct epithelial (HPDE) cells. To further interrogate the role of SgK223 in pancreatic cancer, SgK223 was overexpressed in HPDE cells. This resulted in an elongated cell morphology and increased cellular migration and invasion in transwell assays. MS/MS analysis determined that SgK223 was phosphorylated on tyrosine residues 159 and 413. Since the former is a candidate Stat3 binding site, we characterised Stat3 signalling in this model. SgK223 overexpression led to enhanced Stat3 tyrosine phosphorylation and transcriptional activity. These data identify a role for SgK223 as a mediator of cell invasion in PC, and support the use of MS-based phosphoproteomic profiling as an effective discovery tool for cancer-associated signaling proteins.