Surveying the serologic proteome in a tissue-specific kras(G12D) knockin mouse model of pancreatic cancer.

ABSTRACT

We have applied a serologic proteomic workflow involving three complementary MS approaches to a tissue-specific Kras(G12D) -knockin mouse model of pancreatic cancer that consistently forms precancerous lesions by 4 months of age. The three proteomics applications were highly complementary and allowed us to survey the entire range of low to high molecular weight serologic proteins. Combined, we identified 121 (49↓, 72↑) unique and statistically relevant serologic biomarkers with 88% previously reported to be associated with cancer and 38% specifically correlated with pancreatic cancer. Four markers, lysozyme C2, cytokeratin 19, Serpina1A and Pcf11, were further verified by Western blotting. When applying systems analysis, the top-associated gene ontology functions were tied to wound healing, RXR signaling, growth, differentiation and innate immune activation through the JAK/STAT pathway. Upon further investigation of the apparent immune response using a multiplex cytokine screen, we found that IFN-γ, VEGF and GM-CSF were significantly increased in serum from the Kras(G12D) animals compared to littermate controls. By combining three complementary MS applications, we were able to survey the native intact peptidome and the global proteome in parallel, unveiling pathways that may be biologically relevant to promotion of pancreatic cancer progression and serologic markers of noninvasive early-stage neoplasia.