TRPV4 Regulates Breast Cancer Cell Extravasation, Stiffness and Actin Cortex

Abstract

Metastasis is a significant health issue. The standard mode of care is combination of chemotherapy and targeted therapeutics but the 5-year survival rate remains low. New/better drug targets that can improve outcomes of patients with metastatic disease are needed. Metastasis is a complex process, with each step conferred by a set of genetic aberrations. Mapping the molecular changes associated with metastasis improves our understanding of the etiology of this disease and contributes to the pipeline of targeted therapeutics. Here, phosphoproteomics of a xenograft-derived in vitro model comprising 4 isogenic cell lines with increasing metastatic potential implicated Transient Receptor Potential Vanilloid subtype 4 in breast cancer metastasis. TRPV4 mRNA levels in breast, gastric and ovarian cancers correlated with poor clinical outcomes, suggesting a wide role of TRPV4 in human epithelial cancers. TRPV4 was shown to be required for breast cancer cell invasion and transendothelial migration but not growth/proliferation. Knockdown of Trpv4 significantly reduced the number of metastatic nodules in mouse xenografts leaving the size unaffected. Overexpression of TRPV4 promoted breast cancer cell softness, blebbing, and actin reorganization. The findings provide new insights into the role of TRPV4 in cancer extravasation putatively by reducing cell rigidity through controlling the cytoskeleton at the cell cortex.

Figure 1. Discovery and validation of TRPV4 as an aberrantly expressed phosphoprotein in metastatic breast cancer cells.

(a) Pervanadate-induced tyrosine phosphorylation profiles of the cell lines in Breast Cancer Metastasis (BCM) model. (b) Schematic diagram showing the workflow of iTRAQ-based experiments to identify phosphotyrosine containing proteins in the Breast Cancer Metastasis (BCM) model. (c) Ingenuity Pathway Analysis of the phosphoproteomics gene set. (d) Immunoprecipitation shows that TRPV4 is both tyrosine phosphorylated and overexpressed in 4T07 and 4T1 breast cancer cells. Bar chart showing the average values (n = 3, ±s.d.) and p values from unpaired Student’s t-test are also shown (e) Q-PCR showing increased TRPV4 transcript levels across the BCM model. Bar chart was plotted from 3 technical replicates and p-values obtained by unpaired Student’s t-test.

Figure 2. Clinical significance of TRPV4 in human epithelial cancers.

(a) Top: Gene Set Analyses of 1881-sample breast tumors and 51-sample breast cancer cell lines using GOBO. Bottom: Correlation of TRPV4 expression with DMFS using KMplotter on a total of 361 breast cancers with all breast cancer tumors, luminal B subtype, or HER2 subtype. (b) Top left: Correlation of EMT score with TRPV4 expression in breast tumors. Kaplan Meier Analyses correlating the expression of TRPV4 with overall and/or disease free survival in ovarian (top right) and gastric tumors (bottom panels).

Figure 3. In vitro cell-based assays following Trpv4 knock down.

(a) Top left panel shows effective silencing of TRPV4 expression. Bottom left and right panels show that Trpv4 siRNA inhibited the migration of 4T07 cells. (b) Trpv4 siRNAs inhibited the chemotaxis, invasion, transendothelial migration but not proliferation of 4T07 cells. For all, except proliferation, knock down of Trpv4 produced statistically significant effect on the cellular processes studied, p < 0.05. (c) Shows the inhibitory effects of RR and RES on cell migration and (d) invasion of 4T07 cells but not 67NR cells. Average values (n = 3, ± s.d.) from three biological repeats and p values from unpaired Student’s t-test are shown.

Figure 4. Effects of exogenous TRPV4 on human breast cancer cellular processes.

(a) Expression of wild-type human TRPV4 in stably retroviral-transduced breast cancer cell lines. Upper panel: Immunoblotting of exogenous TRPV4 and endogenous Actin (loading control) in whole-cell lysates. Lower panel: Immunoblotting of TRPV4 following Tunicamycin treatment (5 ug/mL), PNGaseF and Endo-H digestion (500 U/reaction). (b) Calcium imaging of TRPV4 stable transductants following treatment with 10 μM of 4α-PDD. (c) FACS analysis of MB468-TRPV4 stable transductants. Left panel: after gating on the live cells, single cells were gated using width and area parameters. Right panel: Histogram showing the percentages of cells in G1, S, and G2M phases. Average values (n = 3, ±s.d.) from two biological repeats and P values from unpaired Student’s t-test are shown. (d) Physical examination of TRPV4 stable transductants. Upper panel: Phase-contrast micrographs of the cell morphology of MB468 and MCF7 transduced with pBABE or pBABE-TRPV4 were obtained at sub-confluent density grown in standard medium containing 10% serum. Bottom left panel: relative cell surface area of control and TRPV4 stable transductants; bottom right panel: size distribution of control and TRVP4 stable transductants based on FSC-A data from FACS analysis and P values from unpaired Student’s t-test are shown (e) Effects of TRPV4 on the matrigel invasion and chemotaxis of MB468 cells transduced with pBABE or pBABE-TRPV4 retroviral particles using the Boyden chamber assays. Cells that migrated or invaded through the barrier were stained with hematoxylin/eosin and counted (for invasion assay) or subject to colormetric measurements (chemotaxis assay) in three independent experiments. Data are expressed as migrated/invaded cells per field (n = 3, means ± s.d.), P values from unpaired Student’s t-test are shown.

Figure 5. (a)

Effect of Trpv4 knock down on metastasis of 4T1 cells. Top panel: H&E staining of lung tissue sections from mice injected with 4T1 cells transfected with control or Trpv4-specific siRNA. ‘T’ indicates tumor nodules and ‘N’ indicates normal lung tissues. Three magnifications (40x, 200x and 600x) of the representative H&E images (n = 10 for each groups). Bottom panel: Table showing the number of mice in the control or Trpv4knocked-down condition that possessed the categorical numbers of nodules. (b) Upper panel: Representative IHC images showing the expression of TRPV4 in the nodules of lungs of mice from control and Trpv4-knocked down conditions. Lower Panel: Box plots showing the expression of TRPV4 in the metastatic nodules in the lungs from mice in the control and Trpv4-knocked down conditions. (c) Top panel: The size of all the nodules in lungs of every mouse in control and Trpv4-knocked down conditions were measured and the numbers of nodules within 3 size groups were plotted. Bottom left panel: Box plots showing the distribution of nodules size in the mice within the control and Trpv4-silenced conditions. ‘n’ refers to the sample size of the nodules in each condition. Bottom right panel: Box plots of the total number of nodules in each mouse within the control and Trpv4-silenced condition. ‘n’ refers to sample size of mice.

Figure 6. Micropipette aspiration experiment to measure cancer cell rigidity following Trpv4 knock down or overexpression in 4T07 and MB468 cells, respectively.

(a) The percentage of 4T07 cells that formed blebs before and after Trpv4 silencing. (b) Arrows show that bleb formed in control cells but not cells transfected with S1- and S3- Trpv4-specific siRNAs. The minimum pressure at which bleb developed in (c) 4T07 cells and (d) MB468 cells were plotted. (e) The shear modulus (Pa) in the MB468 cells was determined from using the linear elastic solid model. P values from unpaired Student’s t-test are shown.

Figure 7. TRPV4 overexpression affected the cytoskeleton network compared to the control.

(a) Lysates from control or stable TRPV4 MB468 transductants untreated or treated with deionized water for 3 min were immunoblotted with the indicated antibodies. Data shown are mean ± s.d. (n = 3) (b) Insoluble (contains F-actin) and soluble fractions (contains G-actin) isolated from the total lysates were probed for Actin. Data shown are mean ± s.d. (n = 4). (c) Lysates from control or stable TRPV4 MB468 transductants were immunoblotted with the phospho-Cofilin or total Cofilin antibodies. (d) Lysates from MB468-Vector and MB468-TRPV4 cells not treated or treated with deionized water were probed for total VASP and phospho-VASP antibodies. Protein expression levels were normalized and quantified as shown by the bar chart. GAPDH were used as a loading control. Average values from three biological repeats (n = 3, ±s.d.) and p values from unpaired Student’s t-test are shown. (e) MB468-Vector and MB468-TRPV4 cells were not pre-treated or treated with RR for an hour before the replacement of the media with deionized water or fresh growth medium. Densitometry was performed. Protein expression levels were quantified and presented as relative change in E-cadherin expression levels in the various conditions compared to their respective untreated samples. β-Tubulin were used as a loading control. Average values from three biological repeats (n = 5, ±s.d.) and p values from unpaired Student’s t-test are shown.