Limits...
Distinct routes to metastasis: plasticity-dependent and plasticity-independent pathways

View Article: PubMed Central - PubMed

ABSTRACT

The cascade that culminates in macrometastases is thought to be mediated by phenotypic plasticity, including epithelial-mesenchymal and mesenchymal-epithelial transitions (EMT and MET). While there is substantial support for the role of EMT in driving cancer cell invasion and dissemination, much less is known about the importance of MET in the later steps of metastatic colonization. We created novel reporters, which integrate transcriptional and post-transcriptional regulation, to test whether MET is required for metastasis in multiple in vivo cancer models. In a model of carcinosarcoma, metastasis occurred via an MET-dependent pathway; however, in two prostate carcinoma models, metastatic colonization was MET-independent. Indeed, the MET-independent pathway was supported by reanalysis of pre-clinical and clinical data. These results provide evidence for both MET-dependent and MET-independent metastatic pathways.

No MeSH data available.


Related in: MedlinePlus

CS-99 human carcinosarcoma cells undergo MET in vitroA. Cultured CS-99 cells possess cell-cell contacts, interspersed with spindle-shaped cells. CS-99 cells transfected with Rint express DsRed while cells transfected with RIIIcI2 express only background levels of DsRed. B. CS-99 cells express FGFR2-IIIc mRNA. Epithelial DT and mesenchymal AT3 cells are included as controls for the IIIb and IIIc isoforms, respectively. For each cell type, RT-PCR products are mock digested (M), digested with the FGFR2 IIIb-specific enzyme, AvaI (A), or digested with the FGFR2 IIIc-specific enzyme, EcoRV (E). C. CS-99 cells lack E-cadherin and express vimentin. Epithelial LNCaP and mesenchymal PC3 cells are included as controls. D. Reverse transfection of miR200a, miR200b, and miR200c into CS-99 cells leads to a morphological change consistent with MET; cells change shape from spindle-like, single cells to clusters of rounded cells. E. Expression of miR200s in CS-99s induces upregulation of E-cadherin and loss of vimentin. F. The miR200 targets, ZEB1 and Slug, are downregulated and upregulated, respectively. G. Induction of MET in CS-99 cells has no effect luciferase expression from a control reporter, FFint, but a significant increase in luciferase expression is observed from the E-cadFFIIIcI2 MET reporter.
© Copyright Policy
Related In: Results  -  Collection


getmorefigures.php?uid=PMC4940344&req=5

Figure 5: CS-99 human carcinosarcoma cells undergo MET in vitroA. Cultured CS-99 cells possess cell-cell contacts, interspersed with spindle-shaped cells. CS-99 cells transfected with Rint express DsRed while cells transfected with RIIIcI2 express only background levels of DsRed. B. CS-99 cells express FGFR2-IIIc mRNA. Epithelial DT and mesenchymal AT3 cells are included as controls for the IIIb and IIIc isoforms, respectively. For each cell type, RT-PCR products are mock digested (M), digested with the FGFR2 IIIb-specific enzyme, AvaI (A), or digested with the FGFR2 IIIc-specific enzyme, EcoRV (E). C. CS-99 cells lack E-cadherin and express vimentin. Epithelial LNCaP and mesenchymal PC3 cells are included as controls. D. Reverse transfection of miR200a, miR200b, and miR200c into CS-99 cells leads to a morphological change consistent with MET; cells change shape from spindle-like, single cells to clusters of rounded cells. E. Expression of miR200s in CS-99s induces upregulation of E-cadherin and loss of vimentin. F. The miR200 targets, ZEB1 and Slug, are downregulated and upregulated, respectively. G. Induction of MET in CS-99 cells has no effect luciferase expression from a control reporter, FFint, but a significant increase in luciferase expression is observed from the E-cadFFIIIcI2 MET reporter.

Mentions: We noted a number of similarities between the AT3 tumors and prostate carcinosarcomas, including their undifferentiated histology, mesenchymal-like gene expression profiles (17), aggressive nature, and preference for metastasizing to the lungs and lymph nodes (24). Indeed, AT3 cells resemble the more poorly differentiated component of human carcinosarcomas in their histopathology (S6A), their likely epithelial origin and weak to negative co-expression of epithelial biomarkers, such as cytokeratin, and expression of vimentin (S6B). To understand whether MET-independent metastasis is a property of human carcinosarcomas, we tested our suicide reporters in the human uterine carcinosarcoma cell line, CS-99 (25). Morphologically, CS-99 cells exhibit a spindle-like morphology, but will form cell-cell attachments at higher confluence (Figure 5A). CS-99s stably transfected with Rint, a plasmid encoding DsRed interrupted by a constitutively spliced intron, express DsRed; however, CS-99s transfected with RIIIcI2 exhibit only low levels of DsRed expression (Figure 5A), which is consistent with FGFR2 exon IIIc inclusion and a mesenchymal phenotype. We also characterized the expression of endogenous FGFR2 isoforms in CS-99s by RT-PCR (Figure 5B). Similar to the results from the RIIIcI2 reporter, CS-99 cells exclusively express FGFR2-IIIc mRNA (Figure 5B). CS-99s lack E-cadherin and express vimentin (Figure 5C). Together, these data verify the mesenchymal-like phenotype and biomarker expression of the CS-99 cell line.


Distinct routes to metastasis: plasticity-dependent and plasticity-independent pathways
CS-99 human carcinosarcoma cells undergo MET in vitroA. Cultured CS-99 cells possess cell-cell contacts, interspersed with spindle-shaped cells. CS-99 cells transfected with Rint express DsRed while cells transfected with RIIIcI2 express only background levels of DsRed. B. CS-99 cells express FGFR2-IIIc mRNA. Epithelial DT and mesenchymal AT3 cells are included as controls for the IIIb and IIIc isoforms, respectively. For each cell type, RT-PCR products are mock digested (M), digested with the FGFR2 IIIb-specific enzyme, AvaI (A), or digested with the FGFR2 IIIc-specific enzyme, EcoRV (E). C. CS-99 cells lack E-cadherin and express vimentin. Epithelial LNCaP and mesenchymal PC3 cells are included as controls. D. Reverse transfection of miR200a, miR200b, and miR200c into CS-99 cells leads to a morphological change consistent with MET; cells change shape from spindle-like, single cells to clusters of rounded cells. E. Expression of miR200s in CS-99s induces upregulation of E-cadherin and loss of vimentin. F. The miR200 targets, ZEB1 and Slug, are downregulated and upregulated, respectively. G. Induction of MET in CS-99 cells has no effect luciferase expression from a control reporter, FFint, but a significant increase in luciferase expression is observed from the E-cadFFIIIcI2 MET reporter.
© Copyright Policy
Related In: Results  -  Collection

Show All Figures
getmorefigures.php?uid=PMC4940344&req=5

Figure 5: CS-99 human carcinosarcoma cells undergo MET in vitroA. Cultured CS-99 cells possess cell-cell contacts, interspersed with spindle-shaped cells. CS-99 cells transfected with Rint express DsRed while cells transfected with RIIIcI2 express only background levels of DsRed. B. CS-99 cells express FGFR2-IIIc mRNA. Epithelial DT and mesenchymal AT3 cells are included as controls for the IIIb and IIIc isoforms, respectively. For each cell type, RT-PCR products are mock digested (M), digested with the FGFR2 IIIb-specific enzyme, AvaI (A), or digested with the FGFR2 IIIc-specific enzyme, EcoRV (E). C. CS-99 cells lack E-cadherin and express vimentin. Epithelial LNCaP and mesenchymal PC3 cells are included as controls. D. Reverse transfection of miR200a, miR200b, and miR200c into CS-99 cells leads to a morphological change consistent with MET; cells change shape from spindle-like, single cells to clusters of rounded cells. E. Expression of miR200s in CS-99s induces upregulation of E-cadherin and loss of vimentin. F. The miR200 targets, ZEB1 and Slug, are downregulated and upregulated, respectively. G. Induction of MET in CS-99 cells has no effect luciferase expression from a control reporter, FFint, but a significant increase in luciferase expression is observed from the E-cadFFIIIcI2 MET reporter.
Mentions: We noted a number of similarities between the AT3 tumors and prostate carcinosarcomas, including their undifferentiated histology, mesenchymal-like gene expression profiles (17), aggressive nature, and preference for metastasizing to the lungs and lymph nodes (24). Indeed, AT3 cells resemble the more poorly differentiated component of human carcinosarcomas in their histopathology (S6A), their likely epithelial origin and weak to negative co-expression of epithelial biomarkers, such as cytokeratin, and expression of vimentin (S6B). To understand whether MET-independent metastasis is a property of human carcinosarcomas, we tested our suicide reporters in the human uterine carcinosarcoma cell line, CS-99 (25). Morphologically, CS-99 cells exhibit a spindle-like morphology, but will form cell-cell attachments at higher confluence (Figure 5A). CS-99s stably transfected with Rint, a plasmid encoding DsRed interrupted by a constitutively spliced intron, express DsRed; however, CS-99s transfected with RIIIcI2 exhibit only low levels of DsRed expression (Figure 5A), which is consistent with FGFR2 exon IIIc inclusion and a mesenchymal phenotype. We also characterized the expression of endogenous FGFR2 isoforms in CS-99s by RT-PCR (Figure 5B). Similar to the results from the RIIIcI2 reporter, CS-99 cells exclusively express FGFR2-IIIc mRNA (Figure 5B). CS-99s lack E-cadherin and express vimentin (Figure 5C). Together, these data verify the mesenchymal-like phenotype and biomarker expression of the CS-99 cell line.

View Article: PubMed Central - PubMed

ABSTRACT

The cascade that culminates in macrometastases is thought to be mediated by phenotypic plasticity, including epithelial-mesenchymal and mesenchymal-epithelial transitions (EMT and MET). While there is substantial support for the role of EMT in driving cancer cell invasion and dissemination, much less is known about the importance of MET in the later steps of metastatic colonization. We created novel reporters, which integrate transcriptional and post-transcriptional regulation, to test whether MET is required for metastasis in multiple in vivo cancer models. In a model of carcinosarcoma, metastasis occurred via an MET-dependent pathway; however, in two prostate carcinoma models, metastatic colonization was MET-independent. Indeed, the MET-independent pathway was supported by reanalysis of pre-clinical and clinical data. These results provide evidence for both MET-dependent and MET-independent metastatic pathways.

No MeSH data available.


Related in: MedlinePlus