Fibroblast-led cancer cell invasion is activated by epithelial-mesenchymal transition through platelet-derived growth factor BB secretion of lung adenocarcinoma.

Cancer-associated fibroblast (CAF)-dependent local invasion is the process by which cancer cells invade the extracellular matrix using tracks that have been physically remodeled by CAFs. In the present study, we investigated the process by which the epithelial-mesenchymal transition (EMT) of cancer cells affect CAF-dependent local invasion. Using an in vitro collagen invasion assay, we showed cancer cells undergoing EMT to promote the matrix-remodeling ability of CAFs and thereby enhance CAF-dependent local cancer cell invasion. Platelet-derived growth factor (PDGF)-BB secretion was significantly elevated in cancer cells undergoing EMT, and this induced an increase in the invasion ability of both CAFs and cancer cells. Conversely, knockdown of PDGF-B expression in cancer cells undergoing EMT, or treatment with a PDGF-receptor inhibitor, decreased the invasion ability of both CAFs and cancer cells. By analyzing the gene expression profiles of 442 patients with lung adenocarcinomas, we established that high expression of PDGF-B and presentation of mesenchymal-like tumors were significantly associated with a high rate of disease recurrence and poor patient prognosis. Thus, cancer cells undergoing EMT may accelerate their own ability to invade local tissues via PDGF-BB secretion to promote CAF matrix remodeling. Therefore, targeting PDGF signaling between cancer cells undergoing EMT and CAFs is a promising therapeutic target to inhibit cancer progression and improve patient prognosis.

Cancer cell invasion driven by extracellular matrix remodeling is dependent on the properties of cancer-associated fibroblasts.

PURPOSE: As one form of tumor invasion, cancer cells can invade the extracellular matrix (ECM) through tracks that have been physically remodeled by cancer-associated fibroblasts (CAFs). However, CAFs are a heterogeneous population with diverse matrix-remodeling capacities. The purpose of this study was to investigate how CAFs with various matrix-remodeling capacities influence cancer cell invasion. METHODS: We established single-cell-derived clones from three primary cultures of CAFs from lung adenocarcinoma patients (Case 1, 5 clones; Case 2, 5 clones; and Case 3, 7 clones). Using a co-culture model, we evaluated the correlations between the number of invaded cancer cells and the remodeling areas generated by CAF clones in each case. RESULTS: When A549 lung adenocarcinoma cells and CAF clones were co-cultured, both the numbers of invaded cancer cells and the remodeling areas generated by the CAF clones varied greatly. The number of invaded cancer cells was moderately and strongly correlated with the remodeling areas generated by each CAF clone originating from Cases 1 and 2 (R(2) value = 0.53 and 0.68, respectively), suggesting that the remodeling areas in the ECM may determine the number of invaded cancer cells. In contrast, the number of invaded cancer cells was not correlated with the remodeling areas generated by CAF clones originating from Case 3, suggesting that factors other than the remodeling areas might determine the number of invading cancer cells. CONCLUSIONS: These findings showed two types of fibroblast-dependent cancer cell invasion that are dependent on and independent of the remodeling areas generated by CAFs.

Teachable, high-content analytics for live-cell, phase contrast movies.

CL-Quant is a new solution platform for broad, high-content, live-cell image analysis. Powered by novel machine learning technologies and teach-by-example interfaces, CL-Quant provides a platform for the rapid development and application of scalable, high-performance, and fully automated analytics for a broad range of live-cell microscopy imaging applications, including label-free phase contrast imaging. The authors used CL-Quant to teach off-the-shelf universal analytics, called standard recipes, for cell proliferation, wound healing, cell counting, and cell motility assays using phase contrast movies collected on the BioStation CT and BioStation IM platforms. Similar to application modules, standard recipes are intended to work robustly across a wide range of imaging conditions without requiring customization by the end user. The authors validated the performance of the standard recipes by comparing their performance with truth created manually, or by custom analytics optimized for each individual movie (and therefore yielding the best possible result for the image), and validated by independent review. The validation data show that the standard recipes' performance is comparable with the validated truth with low variation. The data validate that the CL-Quant standard recipes can provide robust results without customization for live-cell assays in broad cell types and laboratory settings.