Project description:We found that RANKL, expressed by cancer cells or derived from exogenous sources, consistently induced human prostate, breast, kidney, lung and liver cancer cells to colonize or metastasize to bone in an animal model of cancer bone metastasis. RANK-mediated signaling established a premetastatic niche through a forward feedback loop by inducing RANKL and c-Met expression and downstream signaling via upregulation of master regulator transcription factors regulating EMT (Twist1, Slug, Zeb1, Zeb2), stem cells (Sox2, Myc, Oct3/4 and Nanog), neuroendocrine cells (Sox 9, HIF-1α and FoxA2) and osteomimicry (c-Myc/Max, Sox2, Sox9, HIF1α and Runx2). Abrogating RANK or its downstream signaling network, c-Myc/Max or c-Met, abolished PCa skeletal metastasis in mice. We observed that a small number of RANKL-expressing PCa cells can initiate bone and soft tissue metastases by recruiting non-tumorigenic or bystander PCa or host cells from the circulation or at metastatic sites to co-colonize bone. The recruited bystander PCa cells assume the phenotypes of RANKL-expressing PCa cells by expressing increased c-Met, phosphorylated c-Met and RANKL. RANKL expression at a single cell level in primary PCa tissues predicted disease-specific survival, reflecting the significant role of RANKL-RANK signaling in the development of lethal bone metastasis.
Project description:We found that RANKL, expressed by cancer cells or derived from exogenous sources, consistently induced human prostate, breast, kidney, lung and liver cancer cells to colonize or metastasize to bone in an animal model of cancer bone metastasis. RANK-mediated signaling established a premetastatic niche through a forward feedback loop by inducing RANKL and c-Met expression and downstream signaling via upregulation of master regulator transcription factors regulating EMT (Twist1, Slug, Zeb1, Zeb2), stem cells (Sox2, Myc, Oct3/4 and Nanog), neuroendocrine cells (Sox 9, HIF-1α and FoxA2) and osteomimicry (c-Myc/Max, Sox2, Sox9, HIF1α and Runx2). Abrogating RANK or its downstream signaling network, c-Myc/Max or c-Met, abolished PCa skeletal metastasis in mice. We observed that a small number of RANKL-expressing PCa cells can initiate bone and soft tissue metastases by recruiting non-tumorigenic or bystander PCa or host cells from the circulation or at metastatic sites to co-colonize bone. The recruited bystander PCa cells assume the phenotypes of RANKL-expressing PCa cells by expressing increased c-Met, phosphorylated c-Met and RANKL. RANKL expression at a single cell level in primary PCa tissues predicted disease-specific survival, reflecting the significant role of RANKL-RANK signaling in the development of lethal bone metastasis. Global gene expression analysis perturbed by RANKL in LNRANKL compared to LNNeo cells.
Project description:We used microarrays to probe the global programme of gene expression under treatment with c-Met inhibitor and identified distinct classes of up- and down-regulated genes during this process We treated MET addicted EBC-1 and MKN-45 cells with selective c-Met kinase inhibitor SGX-523 at 1uM for 24 and 48 hours, using HCC827 cells with activated c-Met but lacking MET dependency as a negative control
Project description:Somatic hotspot mutations and structural amplifications and fusions affecting fibroblast growth factor receptor 2 (FGFR2) occur in multiple cancer types. However, clinical responses to FGFR inhibitors (FGFRi) have remained variable, emphasizing a need to better understand which FGFR2 alterations are oncogenic and targetable. Here we applied transposon-based screening and tumor modelling in mice to uncover truncation of exon (E) 18 of Fgfr2 as a potent driver mutation. Human oncogenomic datasets revealed a diverse set of FGFR2 alterations, including rearrangements (REs), E1-E17 partial amplifications, and E18 nonsense and frameshift mutations, each causing transcription of E18-truncated FGFR2 (FGFR2deltaE18). Somatic modelling in mice and human tumor cell lines using a compendium of FGFR2deltaE18 and full-length variants identified FGFR2deltaE18-truncation as potent single-driver alteration in cancer. Here we show the phosphoproteomic landscape of FGFR2 variants in murine epithelial cell (MEC) lines and mouse tumors. Global (STY) phosphoproteomics using IMAC and phosphotyrosine phosphoproteomics using pTyr IP’s are combined with DIA protein expression data to uncover oncogenic signaling of clinically-relevant FGFR2 variants.
Project description:Genome-wide association studies have identified a locus within the second intron of the FGFR2 gene that is consistently the most strongly associated with estrogen receptor-poisive breast cancer risk. However, we know little about the mechanisms by which the FGFR2 locus mediates risk or the pathways in which multiple risk loci may combine to cause disease. Previously, a systems biology approach was adopted to elucidate the regulatory networks operating in MCF-7 breast cancer cells in order to examine the role of FGFR2 in mediating risk. Here, the same approach has been employed using MCF-7 cells that have been treated with siRNA directed against FGFR2, in order to knock-down FGFR2 expression, to confirm that the differential gene expression that we see when FGF10 signalling is perturbed, on a background of estrogen signalling, is mediated via FGFR2 stimulation.
Project description:MET amplification is present in 20% of gastric cancers and has been confirmed as a therapeutic target in clinical trials. The molecular mechanisms of response and resistance to MET inhibitors are not well understood. We investigated the determinants of MET dependency in human gastric cancer. MET inhibition inhibited proliferation and induced cell death only in MET-amplified gastric cancer cell lines. The effects on growth arrest were stronger than the effects on cell death. To identify possible resistance mechanisms, we performed whole-genome mRNA expression profiling. Molecular changes related to autophagy were among the top alterations observed. Consistent with these findings, autophagy levels increased in a concentration-dependent manner when MET-amplified cells were exposed to crizotinib. Autophagy inhibition caused a dramatic decrease in apoptosis in one of the MET-amplified cell lines (MKN45) but not in the other (SNU-5). Because autophagy may provide energy in cells subjected to growth factor deprivation, we explored the effects of MET or autophagy inhibition on cellular ATP levels. This revealed that autophagy-dependent ATP production was selectively required for apoptosis in the MKN45 cells and that chemical ATP depletion mimicked the effects of autophagy inhibition to block cell death. Overall, the data reveal a novel relationship between ATP depletion and resistance to MET inhibitor-induced cell death. Our observations suggest that autophagy inhibitors could have unintended consequences when they are combined with growth factor receptor inhibitors in tumors that require autophagy-dependent ATP production for apoptosis. 12 samples triplicate samples of SNU-5 and MKN45 +/- criztonib for 24 hours
Project description:MET amplification has been clinically credentialed as a therapeutic target in gastric cancer, but the molecular mechanisms underlying sensitivity and resistance to MET inhibitors are still not well understood. Using whole-genome mRNA expression profiling, we identified autophagy as a top molecular pathway that was activated by the MET inhibitor crizotinib in drug-sensitive human gastric cancer cells, and functional studies confirmed that crizotinib increased autophagy levels in the drug-sensitive cells in a concentration-dependent manner. We then used chemical and molecular approaches to inhibit autophagy in order to define its role in cell death. The clinically available inhibitor of autophagy, chloroquine, or RNAi-mediated knockdown of two obligate components of the autophagy pathway (ATG5 and ATG7) blocked cell death induced by crizotinib or RNAi-mediated knockdown of MET, and mechanistic studies localized the effects of autophagy to cytochrome c release from the mitochondria. Overall; the data reveal a novel relationship between autophagy and apoptosis in gastric cancer cells exposed to MET inhibitors. The observations suggest that autophagy inhibitors should not be used to enhance the effects of MET inhibitors in gastric cancer patients.
Project description:We used microarrays to probe the global programme of gene expression under treatment with c-Met inhibitor and identified distinct classes of up- and down-regulated genes during this process
Project description:Genome-wide association studies for breast cancer have identified over 80 different risk regions in the genome, with the FGFR2 locus consistently identified as the most strongly associated locus. However, we know little about the mechanisms by which the FGFR2 locus mediates risk or the pathways in which multiple risk loci may combine to cause disease. Here we use a systems biology approach to elucidate the regulatory networks operating in breast cancer and examine the role of FGFR2 in mediating risk. Using model systems we identify FGFR2-regulated genes and, combining variant set enrichment and eQTL analysis, show that these are preferentially linked to breast cancer risk loci. Our results support the concept that cancer-risk associated genes cluster in pathways
Project description:Genome-wide association studies for breast cancer have identified over 80 different risk regions in the genome, with the FGFR2 locus consistently identified as the most strongly associated locus. However, we know little about the mechanisms by which the FGFR2 locus mediates risk or the pathways in which multiple risk loci may combine to cause disease. Here we use a systems biology approach to elucidate the regulatory networks operating in breast cancer and examine the role of FGFR2 in mediating risk. Using model systems we identify FGFR2-regulated genes and, combining variant set enrichment and eQTL analysis, show that these are preferentially linked to breast cancer risk loci. Our results support the concept that cancer-risk associated genes cluster in pathways The data consists of 71 microarray samples from MCF-7 cells treated under different conditions, at 3 time points (0, 6 and 24 h) in order to perturb FGFR2 signalling using the iF2 construct system. The data have been pre-processed in R using the beadarray package, and are presented in the form of log2 expression values. The experiment was carried out on 6 Humanv4 BeadChips using 12 samples per BeadChip. The original arrays contain 48324 features, with a mean of 22 beads per feature (Standard Deviation of 5)