| Genetic contexts play an important role in disease pathogenesis in both human and animal models. Ventral prostate (VP)-specific transgenic AKT1 mice develop prostate intraepithelial neoplasia (PIN), a precursor of invasive prostate cancer. To understand why constitutive AKT1 activation is not sufficient to drive murine prostate carcinogenesis, we hypothesized that modifier genes may modulate the phenotypic outcome of Tg-AKT1 in FVB mice. We show that Tg-F1 offspring between FVB-AKT1 and C57B/6 mice have a marked increase in the number of ventral prostate BrdU positive cells. We demonstrate that the VP mRNA expression states are unique for each of the six inbred strains---AKR, SWR, 129X1, FVB, BALB/c, and C57BL/6, and VP proliferation rates vary across Tg-F1 progeny generated from intercrossing these strains with FVB-AKT1 mice. Candidate modifier genes have been identified by correlation analysis of these two data. Using the gene set enrichment analysis (GSEA), we identify modifiers that function at the pathway level. One particular candidate modifier pathway, the glycolysis pathway, has been validated using human prostate cancer expression datasets and B6- AKT1 mice. This strategy is named parental strain expression mapping (PSEM).; To uncover quantitative trait loci (QTL) regulating AKT1-induced proliferation, we measured BrdU incorporation in 144 VPs of Tg-F2 animals obtained from intercrossing FVB-AKT1 and B6 mice. Genome scan was carried out with 173 SNP markers using Sequenom technology. One suggestive locus was identified using interval mapping on chromosome 5. Two additive QTLs on chromosome 4 were identified using two-dimensional analysis. Homologous mapping linked the QTL on chromosome 5 to a known human prostate cancer susceptibility locus on 4821. With mRNA expression data, 12 candidate genes were located within the 1-lod interval of the QTL on chromosome 5. One of the candidates, Cxc113, stimulated cell proliferation, consistent with its expression being upregulated in B6, the high proliferating strain.; We demonstrated previously that expression microarray analysis could be used to identify novel biomarkers that are indicative of mTOR activity in FVB-AKT1 mice. Here, in order to differentiate pharmacodynamic and response biomarkers associated with targeted inhibition of mTOR by RAD001, we extended the analysis to longer time points (d6, d9 and d15). Using a step-wise approach, 383 mTOR-dependent and 107 PIN-dependent biomarkers were identified. Potential serum biomarkers were discovered through GO analysis, mass spectrometry analysis, and immunoblotting. The human relevance of these murine biomarkers was assessed using human expression array datasets. |
