Transforming growth factor (TGF-) signaling induces proliferating endothelial cells to undergo an epithelial-to-mesenchymal transition, which generates a pool of activated fibroblasts in solid tumors.
The accumulation of activated fibroblasts within solid tumours is widely accepted, but it remains unclear where these fibroblasts originate from. Raghu Kalluri and colleagues have evidence that some originate from endothelial cells that have undergone endothelial-to-mesenchymal transition (EndMT).
Fibroblasts arising as a result of EndMT have been documented before, both in normal heart development and in its fibrosis. To see whether EndMT also occurs in carcinomas, Kalluri and colleagues initially stained B16F10 mouse melanoma tumours for the endothelial marker CD31 (also known as PECAM1) and the fibroblast markers fibroblast-specific protein 1 (FSP1) and -smooth muscle actin (SMA, also known as ACTA2). Approximately 11% of SMA-expressing cells and 40% of FSP1+ cells were positive for CD31. However, this staining could not confirm that these cells originated from endothelial cells, so the authors used Tie2–Cre;R26Rosa–lox–stop–lox–LacZ transgenic mice in which Tie2–cre endothelial cells continuously express LacZ. Examination of subcutaneous B16F10 melanomas in these mice indicated that LacZ-positive cells were evident in blood vessels as expected, but were also scattered within the tumour tissue. Double-labelling experiments for FSP1 or SMA along with LacZ indicated that 30% of the LacZ cells stained with FSP1 and 12% with SMA. These results indicate that LacZ-positive endothelial cells can acquire mesenchymal characteristics.
So what is the trigger for EndMT? Previous studies in the heart have implicated transforming growth factor (TGF), a known inducer of epithelial-to-mesenchymal transition. In vitro experiments with mouse lung endothelial cells indicated that TGF could induce EndMT. Furthermore, the authors showed, again using double-labelling experiments, that FSP1 and CD31 double-positive cells were present in pancreatic tumours that develop in the Rip1–Tag2 mouse cancer model, and that expression of TGF is also evident.
Further investigations are needed to show whether EndMT also occurs in other tumour types, in both man and mouse, and how this contributes to tumour progression.
Nicola McCarthy
References
Zeisberg, E. M., Potenta, S., Xie, L., Zeisberg, M. & Kalluri, R. Discovery of endothelial to mesenchymal transition as a source for carcinoma-associated fibroblasts. Cancer Res.67, 10123–10128 (2007) | Article | PubMed |
Kalluri, R. & Zeisberg, M. Fibroblasts in cancer. Nature Rev. Cancer6, 392–401 (2006) | Article | PubMed |
(TGF-
-smooth muscle actin (
40% of FSP1+ cells were positive for CD31. However, this staining could not confirm that these cells originated from endothelial cells, so the authors used Tie2–Cre;R26Rosa–lox–stop–lox–LacZ transgenic mice in which Tie2–cre endothelial cells continuously express LacZ. Examination of subcutaneous B16F10 melanomas in these mice indicated that LacZ-positive cells were evident in blood vessels as expected, but were also scattered within the tumour tissue. Double-labelling experiments for FSP1 or 