OBJECTIVES: Carcinoma-associated fibroblasts are reported to communicate microenvironment-derived signals through chemokine/chemokine receptor interaction, influencing carcinogenesis. We sought to characterize roles of CXCL12/CXCR4 in crosstalk between non-small cell lung cancer epithelial cell and carcinoma-associated fibroblasts and in tumor growth. METHODS: Non-small cell lung cancer tumor samples obtained at surgery and from tumor arrays, as well as primary carcinoma-associated fibroblast and epithelial cell lines generated from fresh tumors, were assessed for CXCL12/CXCR4 expression, tissue localization, and production. Colony assays, extracellular signal-regulated kinase signaling, and chemokine production were measured to assess cancer cell responsiveness to CXCL12 stimulation with or without CXCR4 antagonists. RESULTS: CXCL12 and CXCR4 were detected in all major subtypes of non-small cell lung cancer. CXCL12-expressing carcinoma-associated fibroblasts were mostly located near CXCL12-negative tumor cells, whereas CXCL12-positive tumor cells were mostly surrounded by CXCL12-negative stroma. Intratumoral CXCL12 levels were significantly higher than serum levels. CXCL12 expression correlated with advanced disease stage. In vitro, tumor cell lines produced variable amounts of CXCL12 and expressed high levels of CXCR4. Carcinoma-associated fibroblasts cell lines produced high amounts of CXCL12 and expressed variable levels of CXCR4. Stimulation of non-small cell lung cancer neoplastic cells with CXCL12 increased colony-forming capacity, induced extracellular signal-regulated kinase phosphorylation, and production of the proinflammatory chemokine CCL20. CXCR4 antagonists attenuated these effects. CONCLUSIONS: Interaction between carcinoma-associated fibroblasts and tumor epithelial cells through the CXCL12/CXCR4 axis plays a role in non-small cell lung cancer tumor proliferation, marking this axis as a target for immune intervention.
OBJECTIVES:Carcinoma-associated fibroblasts are reported to communicate microenvironment-derived signals through chemokine/chemokine receptor interaction, influencing carcinogenesis. We sought to characterize roles of CXCL12/CXCR4 in crosstalk between non-small cell lung cancer epithelial cell and carcinoma-associated fibroblasts and in tumor growth. METHODS:Non-small cell lung cancer tumor samples obtained at surgery and from tumor arrays, as well as primary carcinoma-associated fibroblast and epithelial cell lines generated from fresh tumors, were assessed for CXCL12/CXCR4 expression, tissue localization, and production. Colony assays, extracellular signal-regulated kinase signaling, and chemokine production were measured to assess cancer cell responsiveness to CXCL12 stimulation with or without CXCR4 antagonists. RESULTS:CXCL12 and CXCR4 were detected in all major subtypes of non-small cell lung cancer. CXCL12-expressing carcinoma-associated fibroblasts were mostly located near CXCL12-negative tumor cells, whereas CXCL12-positive tumor cells were mostly surrounded by CXCL12-negative stroma. Intratumoral CXCL12 levels were significantly higher than serum levels. CXCL12 expression correlated with advanced disease stage. In vitro, tumor cell lines produced variable amounts of CXCL12 and expressed high levels of CXCR4. Carcinoma-associated fibroblasts cell lines produced high amounts of CXCL12 and expressed variable levels of CXCR4. Stimulation of non-small cell lung cancer neoplastic cells with CXCL12 increased colony-forming capacity, induced extracellular signal-regulated kinase phosphorylation, and production of the proinflammatory chemokine CCL20. CXCR4 antagonists attenuated these effects. CONCLUSIONS: Interaction between carcinoma-associated fibroblasts and tumor epithelial cells through the CXCL12/CXCR4 axis plays a role in non-small cell lung cancer tumor proliferation, marking this axis as a target for immune intervention.
Authors: Sophie Kirshberg; Uzi Izhar; Gail Amir; Jonathan Demma; Fiona Vernea; Katia Beider; Zippora Shlomai; Hanna Wald; Gideon Zamir; Oz M Shapira; Amnon Peled; Ori Wald Journal: PLoS One Date: 2011-09-15 Impact factor: 3.240