Endothelial-mesenchymal transition of brain endothelial cells: possible role during metastatic extravasation - PubMed (original) (raw)

Endothelial-mesenchymal transition of brain endothelial cells: possible role during metastatic extravasation

István A Krizbai et al. PLoS One. 2015.

Erratum in

Abstract

Cancer progression towards metastasis follows a defined sequence of events described as the metastatic cascade. For extravasation and transendothelial migration metastatic cells interact first with endothelial cells. Yet the role of endothelial cells during the process of metastasis formation and extravasation is still unclear, and the interaction between metastatic and endothelial cells during transendothelial migration is poorly understood. Since tumor cells are well known to express TGF-β, and the compact endothelial layer undergoes a series of changes during metastatic extravasation (cell contact disruption, cytoskeletal reorganization, enhanced contractility), we hypothesized that an EndMT may be necessary for metastatic extravasation. We demonstrate that primary cultured rat brain endothelial cells (BEC) undergo EndMT upon TGF-β1 treatment, characterized by the loss of tight and adherens junction proteins, expression of fibronectin, β1-integrin, calponin and α-smooth muscle actin (SMA). B16/F10 cell line conditioned and activated medium (ACM) had similar effects: claudin-5 down-regulation, fibronectin and SMA expression. Inhibition of TGF-β signaling during B16/F10 ACM stimulation using SB-431542 maintained claudin-5 levels and mitigated fibronectin and SMA expression. B16/F10 ACM stimulation of BECs led to phosphorylation of Smad2 and Smad3. SB-431542 prevented SMA up-regulation upon stimulation of BECs with A2058, MCF-7 and MDA-MB231 ACM as well. Moreover, B16/F10 ACM caused a reduction in transendothelial electrical resistance, enhanced the number of melanoma cells adhering to and transmigrating through the endothelial layer, in a TGF-β-dependent manner. These effects were not confined to BECs: HUVECs showed TGF-β-dependent SMA expression when stimulated with breast cancer cell line ACM. Our results indicate that an EndMT may be necessary for metastatic transendothelial migration, and this transition may be one of the potential mechanisms occurring during the complex phenomenon known as metastatic extravasation.

PubMed Disclaimer

Conflict of interest statement

Competing Interests: The authors have declared that no competing interests exist.

Figures

Fig 1

Fig 1. TGF-β1 induces EndMT in primary rat BECs.

(a) BECs were subjected to 48 hrs TGF-β1 treatment, then fixed and stained for claudin-5 and VE-cadherin. Control cells showed claudin-5 and VE-cadherin at their intercellular borders, which was depleted in TGF-β1 treated cells (400x magnification). BECs were treated with TGF-β1 for 48 hrs, and were analyzed for endothelial, mesenchymal and myogenic marker expression by Western blot. TGF-β1 treatment led to the (b) down-regulation of tight and adherens junction protein expression (claudin-5, occludin, VE-cadherin), it induced β1-integrin expression and (c) and led to a marked increase in N-cadherin expression (2.18±0.48 fold based on three independent experiments, where N-cadherin expression increased 2.11, 1.52 and 2.89 fold respectively, as compared to controls). TGF-β1 treated cells showed a robust expression of fibronectin (d), SMA and calponin (e). SMA expression was observed by immunofluorescence microscopy (400x magnification) under similar conditions as well (f).

Fig 2

Fig 2. TGF-β1 induced SMA expression in BECs requires TGFβR and ROCK. TGF-β1 treatment leads to SRF nuclear translocation.

(a) Cells were preincubated with 10 μM SB-431542 for 60 min and then treated with TGF-β1 for 48 hrs in the presence of SB-431542, and analyzed by Western blot. TGF-β1 induced SMA and calponin expression was mitigated in SB-431542 treated cells. (b) Cells were pretreated with 10 μM Y-27632 for 60 min and then treated with TGF-β1 for 48 hrs in the presence of the inhibitor, and analyzed by Western blot. TGF-β1 induced SMA expression was inhibited in the presence of Y-27632. (c) BECs were subjected to 24 hrs of TGF-β1 treatment, then were fixed and stained for SRF. Specimens were analyzed by immunofluorescence microscopy. Control cells expressed SRF mainly in their cytoplasm, whereas TGF-β1 treatment led to its nuclear translocation (400x magnification).

Fig 3

Fig 3. Activated B16/F10 conditioned medium induces EndMT in BECs through TGF-β signaling.

PDS-free BEC medium was conditioned with B16/F10 cells for 24 hrs, heat activated and then supplemented with PDS. Conditioned medium (CM) or heat-activated conditioned medium (ACM) was then used to stimulate BECs for 48 hrs, samples being analyzed by Western blot. Similarly to the parallel TGF-β1 treatments, ACM induced EndMT of BECs as evidenced by the reduction of claudin-5 protein levels (a), as well as by the expression of fibronectin (b) and SMA (c). (d) BECs were stimulated with TGF-β1 or B16/F10 ACM for 30 minutes. Both stimuli induced phosphorylation of Smad2 and Smad3. Cells were pre-treated with 10 μM SB-431542 for 60 min and then treated with conditioned medium (CM) or heat-activated conditioned medium (ACM), in the presence of the inhibitor. Samples were analyzed by Western blot. SB-431542 treatment of RBECs preserved claudin-5 levels (e). The presence of the inhibitor prevented fibronectin (f) and SMA (g) expression of RBECs when stimulated with B16/F10 conditioned/activated medium.

Fig 4

Fig 4. Activated cancer cell conditioned medium induces SMA expression through TGF-β signaling in RBECs.

PDS-free BEC medium was conditioned with A2058 human melanoma, MCF-7 human breast cancer or MDA-MB231 human breast cancer cells for 24 hrs. A half of the conditioned media were subjected heat activation, and then both conditioned (CM) and activated conditioned media (ACM) were supplemented with PDS. Cells were pre-treated with 10 μM SB-431542 for 60 min and then treated with CM or ACM, in the presence of SB-431542. Samples were analyzed by Western blot. SB-431542 inhibited A2058 ACM induced SMA expression and VE-cadherin down-regulation (a), as well as MCF-7 ACM induced SMA and fibronectin expression (b). SB-431542 prevented MDA-MB231 ACM induced SMA expression and VE-cadherin down-regulation (c).

Fig 5

Fig 5. TGF-β1 and activated B16/F10 conditioned medium modulate TEER, melanoma-endothelial cell adhesion and transendothelial migration of melanoma cells.

(a) TGF-β1 and activated B16/F10 conditioned medium reduce TEER of RBECs. Both TGF-β1 and activated B16/F10 conditioned medium caused a significant, time dependent decline of the transendothelial electric resistance (TEER) of the RBEC monolayer. SB-431542 (60 min) pretreatment and treatment of RBECs inhibited or reduced the decline of TEER upon stimulation with either stimulus. The graph summarizes the results of 3 independent measurements for each treatment, average and SE values are presented. Comparing the TEER between TGF and control (*), ACM and control (*), TGF+SB and TGF (**), ACM+SB and ACM (***) we obtained p<0.05 in all three time points, except for ACM vs. control at 2 hrs where the significance was lower (p<0.01, ****), as assessed by ANOVA and Bonferroni’s post hoc test. (b) TGF-β-dependent adhesion of B16/F10 melanoma cells to RBECs. Confluent RBECs were pretreated with TGF-β1, B16/F10 conditioned media (B16 CM) or B16/F10 activated conditioned media (B16 ACM) in the presence or absence of SB-431542 (SB) for 5 hrs. Fluorescently labeled B16/F10 melanoma cells (5x104/well) were plated onto confluent RBECs and left for 70 min. After washing of non-adherent cells, attached melanoma cells were counted. TGF-β1 pretreatment led to a marked increase of melanoma cells attached to the endothelial monolayer. Similarly to TGF-β1, B16/F10 ACM pre-treatment also enhanced the attachment of B16/F10 melanoma cells onto the endothelial monolayer, in a TGF-β- dependent manner, since this effect was mitigated in the presence of the TGF-β inhibitor SB-431542 (p<0.05 in case of TGF vs. control (*), B16 ACM vs. control (*) and B16 ACM+SB vs. B16 ACM (**), as assessed by ANOVA and Bonferroni’s post hoc test). (c) Enhanced TGF- β-dependent transendothelial migration of melanoma cells. RBECs were cultured until confluence in 12 well plates and treated with TGF-β1 or B16/F10 ACM for 5 hrs. Fluorescently labelled B16/F10 melanoma cells (2 x 104/well) were plated onto the monolayer. Cells were monitored for 10 hrs and then transmigrating melanoma cells were counted. Both stimuli enhanced the number of transmigrating melanoma cells. Preincubation of RBECs with 10 μM SB-431542 for 60 min inhibited ACM induced transendothelial migration of melanoma cells (*- p<0.01 for TGF vs. control and B16 ACM vs control, **- p<0.01 for B16 ACM+ SB vs. B16 ACM, ANOVA and Bonferroni’s post hoc test).

Fig 6

Fig 6. Activated breast cancer cell conditioned medium induces SMA expression through TGF-β signaling in HUVECs.

FBS-free M199 medium was conditioned with MDA-MB231 (a) or SK-BR3 (b) human breast cancer cells for 24 hrs. A half of the conditioned media were subjected heat activation, and then both conditioned (CM) and activated conditioned media (ACM) were supplemented with FBS. Cells were pre-treated with 10 μM SB-431542 for 60 min and then treated with CM or ACM for 72 hrs, in the presence of the inhibitor. Samples were analyzed by Western blot. SB-431542 inhibited MDA-MB231 or SK-BR3 ACM induced SMA expression. (c) Co-culture with melanoma cells leads to expressional changes characteristic to EndMT in HUVECs. Gene expression profiles of HUVECs and HUVECs co-cultured with 1205Lu human metastatic melanoma cells were analyzed and compared. In HUVECs co-cultured with melanoma cells there was a marked decrease in expression levels of several endothelial markers (KRT7, KRT18, TJP2), as well as a severe decrease in FST expression. In parallel, co-cultured HUVECs exhibited elevated expression levels of EndMT markers (FN1, COL3A1, S100A4, MMP2, COL1A2) and transcriptional regulators (ZEB1, Wnt5a, TWIST1, Snai2).

References

    1. Eisenberg LM, Markwald RR. Molecular regulation of atrioventricular valvuloseptal morphogenesis. Circ Res. 1995;77: 1–6. - PubMed
    1. Zeisberg EM, Potenta S, Xie L, Zeisberg M, Kalluri R. Discovery of Endothelial to Mesenchymal Transition as a Source for Carcinoma-Associated Fibroblasts. Cancer Res. 2007;67: 10123–10128. - PubMed
    1. Zeisberg EM, Tarnavski O, Zeisberg M, Dorfman AL, McMullen JR, Gustafsson E, et al. Endothelial-to-mesenchymal transition contributes to cardiac fibrosis. Nat Med. 2007;13: 952–961. - PubMed
    1. Zeisberg EM, Potenta SE, Sujimoto H, Zeisberg M, Kalluri R. Fibroblasts in kidney fibrosis emerge via endothelial to mesenchymal transition. J Am Soc Nephrol. 2008;19: 2282–2287. 10.1681/ASN.2008050513 - DOI - PMC - PubMed
    1. Maddaluno L, Rudini N, Cuttano R, Bravi L, Giampietro C, Corada M, et al. EndMT contributes to the onset and progression of cerebral cavernous malformations. Nature. 2013;498: 492–496. 10.1038/nature12207 - DOI - PubMed

Publication types

MeSH terms

Substances

LinkOut - more resources