TAZ is involved in breast cancer cell migration via regulating actin dynamics

Background Cancer metastasis is dependent on cell migration. Several mechanisms, including epithelial-to-mesenchymal transition (EMT) and actin fiber formation, could be involved in cancer cell migration. As a downstream effector of the Hippo signaling pathway, transcriptional coactivator with PDZ-binding motif (TAZ) is recognized as a key mediator of the metastatic ability of breast cancer cells. We aimed to examine whether TAZ affects the migration of breast cancer cells through the regulation of EMT or actin cytoskeleton. Methods MCF-7 and MDA-MB-231 cells were treated with siRNA to attenuate TAZ abundance. Transwell migration assay and scratch wound healing assay were performed to study the effects of TAZ knockdown on cancer cell migration. Fluorescence microscopy was conducted to examine the vinculin and phalloidin. Semiquantitative immunoblotting and quantitative real-time PCR were performed to study the expression of small GTPases and kinases. Changes in the expression of genes associated with cell migration were examined through next-generation sequencing. Results TAZ-siRNA treatment reduced TAZ abundance in MCF-7 and MDA-MB-231 breast cancer cells, which was associated with a significant decrease in cell migration. TAZ knockdown increased the expression of fibronectin, but it did not exhibit the typical pattern of EMT progression. TGF-β treatment in MDA-MB-231 cells resulted in a reduction in TAZ and an increase in fibronectin levels. However, it paradoxically promoted cell migration, suggesting that EMT is unlikely to be involved in the decreased migration of breast cancer cells in response to TAZ suppression. RhoA, a small Rho GTPase protein, was significantly reduced in response to TAZ knockdown. This caused a decrease in the expression of the Rho-dependent downstream pathway, i.e., LIM kinase 1 (LIMK1), phosphorylated LIMK1/2, and phosphorylated cofilin, leading to actin depolymerization. Furthermore, myosin light chain kinase (MLCK) and phosphorylated MLC2 were significantly decreased in MDA-MB-231 cells with TAZ knockdown, inhibiting the assembly of stress fibers and focal adhesions. Conclusion TAZ knockdown inhibits the migration of breast cancer cells by regulating the intracellular actin cytoskeletal organization. This is achieved, in part, by reducing the abundance of RhoA and Rho-dependent downstream kinase proteins, which results in actin depolymerization and the disassembly of stress fibers and focal adhesions.


Introduction
Breast cancer is one of the most common types of cancer and the second leading cause of death among women (1).Despite advancements in the diagnosis and treatment of early-stage disease, approximately 6-10% of breast cancer patients are diagnosed with metastasis.Furthermore, it has been estimated that around 30% of patients who are initially diagnosed with early-stage breast cancer will ultimately experience recurrent advanced or metastatic disease (2).As opposed to the primary tumor, the cause of mortality in breast cancer patients is predominantly the metastatic dissemination to multiple organs, including bone, lung, and liver (3,4).Thus, the heterogeneous nature of breast cancer metastasis makes it challenging to identify risk factors and underlying mechanisms for disease progression and define appropriate treatments for each breast cancer patient (5).
Tumor metastasis involves a complex set of sequential events.For instance, local invasion into the surrounding tissues, intravasation by infiltrating the blood or lymphatic vessels, and survival as circulating tumor cells (CTCs) in the circulation are all required.In addition, the release of CTCs from the circulatory system, extravasation, adaptation to the microenvironments as disseminated tumor cells, and metamorphosis into metastasisinitiating cells are necessary for the formation of a metastatic lesion (6)(7)(8)(9).In particular, cell migration, a fundamental process that is implicated in numerous biological phenomena, is critical for the invasion and metastasis of cancer (10)(11)(12).
Several mechanisms could be implicated in cancer cell migration, including epithelial-to-mesenchymal transition (EMT) and actin fiber formation (13).In addition, we previously demonstrated that aquaporin-5 (AQP5) expression is associated with breast cancer cell migration, activated Rac1, and cell detachment and dissemination from migrating cell sheets (14)(15)(16)).The EMT is known to be engaged in the initiation of cancer cell migration, as it involves the loss of epithelial markers, trans-differentiation into mesenchymal-like cells, and the acquisition of motile and invasive capacities (17,18).For instance, Snail1, a crucial transcription factor of EMT, is associated with the tumor grade, a high rate of recurrence, and distant or lymph node metastasis of invasive ductal carcinoma (19,20), suggesting a plausible target for therapeutic intervention.Furthermore, cancerassociated fibroblasts (CAF) assemble a fibronectin-rich extracellular matrix that facilitates CAF-cancer cell interaction and directs cell migration (21).
Importantly, the Hippo signaling pathway, which regulates cell proliferation, differentiation, and apoptosis (22,23), has emerged as a cancer signaling network in a range of malignancies, including breast cancer (24).Dysregulation of the Hippo pathway is associated with breast cancer metastasis (25)(26)(27).Particularly, Bartucci et al. (28) demonstrated that the loss of TAZ (transcriptional co-activator with PDZ-binding motif, also known as WWTR1) in breast cancer stem cells significantly hindered metastasis colonization formation and reduced chemoresistance.Based on these studies, we hypothesized that Hippo signaling might change the ability of cancer cells to metastasize (i.e., cell migration) by controlling EMT or actin fiber formation, in addition to regulating the expression of genes linked to tumor metastasis (29).
This study aimed to investigate the role of TAZ in breast cancer cell migration and its underlying mechanisms.We have focused on studying the alterations in EMT and actin fiber formation in response to TAZ knockdown in breast cancer cells, as these changes may be linked to cancer cell migration.Furthermore, we used next-generation sequencing (NGS) to study alterations in genes associated with TAZ knockdown in breast cancer cells.

Quantitative real-time PCR
MDA-MB-231 cells were seeded in 6-well plates and transfected with 25 nM of TAZ-siRNA for 24 h.Total RNA purification was performed by Direct-zol RNA MiniPrep, and total RNA (1 mg) was used to synthesize cDNA with the Takara cDNA synthesis kit (6110A, Takara, Otsu, Shiga, Japan).The relative mRNA expression of TAZ, RhoA, RhoC, ROCK1, ROCK2, Cdc42, Rac1, and Myosin Light Chain 2 (MLC2) was analyzed using a QuantiTect SYBR Green PCR Kit (204143, QIAGEN, Hilden, Germany).b-actin mRNA was used as an internal control.qRT-PCR was run on Rotor-Gene-A (QIAGEN, Hilden, Germany) and threshold was set by 0.02 to determine the threshold cycle (Ct) value.The relative mRNA expression was calculated, as we described previously (31).Each sample was tested in duplicate, and the primer sequences used for qRT-PCR are depicted in Table 1.

Transwell cell migration assay
Breast cancer cell lines (MCF-7 and MDA-MB-231) were transfected with 25 nM of control siRNA or TAZ-siRNA using DharmaFECT 4 reagent.After treatment with siRNA for 24 h and changing to fresh growth media, the cells were grown for another 24 h and detached using trypsin/EDTA.The detached cells were resuspended in low-or high-glucose DMEM without FBS and were seeded on the permeable filter of the transwell system (8 mm pore size, transwell) at a density of 2.0 x10 4 cells per well.Cell migration was stimulated with DMEM media containing 10% FBS for 6 h (MDA-MB-231 cells) or for 12 h (MCF-7 cells).The upper chamber contained the medium without FBS, and the lower chamber was filled with the medium containing 10% FBS.Migrated cells on the lower face were fixed with 4% paraformaldehyde for 10 min, then stained with 0.03% crystal violet for 10 min.The numbers of the migrated cells in five randomly selected fields were counted under a light microscope at 100 X magnification.

Scratch wound healing assay for cell migration
In 96-well IncuCyte® ImageLock Plates (Essen BioScience, Ann Arbor, MI), MDA-MB-231 and MCF7 cells were grown to confluency.Negative control-siRNA or TAZ-targeting siRNA were transfected into cells in low or high glucose DMEM when they reached 50 -60% confluence.MDA-MB-231 and MCF-7 cells were transfected with 25 nM of either control siRNA or TAZ-siRNA.After treatment with siRNA for 24 h and changing to a fresh growth medium, cells were grown for an additional 24 h and then preincubated with mitomycin C (10 mg/mL, Sigma, St. Louis, MO) for 4 h before scratching to prevent cell proliferation.Mitomycin C treatment was maintained throughout the wound closure assay.A 96-pin WoundMaker ™ (IncuCyte ZOOM® Live-Cell Imaging System; Essen BioScience, Ann Arbor, MI) was used to scratch the cell monolayer.After scratching, cell migration was induced for 24 h by 10% FBS-containing media.Every 2 h, images were automatically captured, saved, and registered by the IncuCyte ™ software system.The data was analyzed using an integrated metric.The values were expressed as relative wound density.

Next-generation sequencing
Total RNA was isolated from control-siRNA or TAZ-siRNAtreated MDA-MB-231 cells.The number of cell preparations that were included in each group was three.For transcriptome sequencing, mRNA libraries were generated using TruSeq Stranded mRNA library prep kit according to the manufacturer's protocol (TruSeq Stranded mRNA reference guide # 1000000040498v00).
Libraries were pooled and sequenced to obtain 100-bp paired-end reads on the Illumina NovaSeq platform to a depth of more than 40 million reads per sample (>10 million reads per cell).Genomic reference (GRCh38) was used to map cDNA fragments obtained through RNA-seq data.The quality of sequencing data was checked using FastQC.Trimmed reads were mapped to known reference genomes using the HISAT2 program after preprocessing.Genes/ transcripts were assembled using the reference gene model through the StringTie program.After assembly, the abundance amount of the corresponding transcript was calculated using read count and normalized value; FPKM (Fragments Per Kilobase of transcript per Million mapped reads) and TPM (Transcripts Per Kilobase Million) were estimated.Gene ontology was analyzed using g:Profiler (https:// biit.cs.ut.ee/gprofiler/orth) for the list of significant differentially expressed genes (DEGs).Gene set enrichment by functional classification: biological process (BP), molecular function (MF), and cellular component (CC) analyses were carried out.

Statistical analysis
Values are presented as means ± SEM.Comparisons between the two groups were made by unpaired t-test.A comparison of multiple groups was made by one-way ANOVA followed by a posthoc Bonferroni's multiple comparison test.Multiple comparison tests were only applied when a significant difference was determined by ANOVA (P < 0.05).P values < 0.05 were considered statistically significant.

Results
Reduced cell migration in breast cancer cells (MCF-7 and MDA-MB-231 cells) treated with TAZ-siRNA MCF-7 and MDA-MB-231 breast cancer cell lines were treated for 24 h with control-siRNA or TAZ-siRNA.Immunoblot analysis demonstrated that TAZ knockdown was induced by TAZ-siRNA treatment in both MCF-7 and MDA-MB-231 cells (Figures 1A-D).We then conducted a transwell cell migration assay and a scratch wound healing assay to determine whether TAZ knockdown affects the migration of breast cancer cells.On the upper side of the transwell, serum-free DMEM media was added with cells treated with siRNA.In order to stimulate cell migration, 10% FBS was added to the DMEM media in the lower chamber.The migration of the MCF-7 cells was observed for 12 h, while MDA-MB-231 cells were observed for 6 h.TAZ knockdown significantly decreased cell migration of MCF-7 cells (49 ± 1% of control-siRNA, P < 0.05, Figures 1E, F) and MDA-MB-231 cells (73 ± 2% of control-siRNA, P < 0.05, Figures 1G, H), as determined by the transwell cell migration assay counting the numbers of migrated cells on the lower face.Furthermore, to perform the scratch wound healing assay, cells were scratched and subsequently starved for 24 h without FBS.The results demonstrated that TAZ knockdown significantly reduced cell migration in both MCF-7 and MDA-MB-231 cells (Figures 1I-L).Compared to control-siRNA, TAZ knockdown reduced cell migration in both MCF-7 cells (from 12 to 22 h, P < 0.05, Figures 1I, J) and MDA-MB-231 cells (from 4 to 24 h, P < 0.05, Figures 1K, L).

Epithelial-to-mesenchymal transition in MDA-MB-231 cells with siRNA-mediated TAZ knockdown
To determine if TAZ-knockdown per se is associated with EMT in MDA-MB-231 cells, the changes in the expression of EMT markers in breast cancer cells were examined using semiquantitative immunoblotting.As epithelial markers, E-cadherin and occludin were used, while fibronectin, N-cadherin, and vimentin were employed as mesenchymal markers.The protein abundance of Ecadherin (188 ± 23% of control-siRNA, P < 0.05, Figures 2A, C) and fibronectin (172 ± 18% of control-siRNA, P < 0.05, Figures 2A, D) was significantly increased in response to TAZ-knockdown (Figure 2B), whereas N-cadherin was decreased (77 ± 9% of control-siRNA, P < 0.05, Figures 2A, F).The protein levels of occludin and vimentin were unaltered (Figures 2A, E, G).
Next, NGS data were categorized using the Pathcards database (https://pathcards.genecards.org),an integrated database consolidating 1,626 SuperPath entries from 11 sources and clustering human pathways based on gene content similarity.Each PathCard offers information about a SuperPath, representing one or more human pathways (32).The superpathway regarding the "cytoskeleton remodeling regulation of actin cytoskeleton by Rho GTPases" had the highest relevance score among the actin-related pathways (https:// pathcards.genecards.org/Search/Results?query=actin).The superpathway includes 5 subpathways.A total of 187 genes were included in the superpathways database, and 68 genes from our NGS data were found in the superpathways (Table 3).The superpathway regarding the "actin nucleation by ARP-WASP complex" had the second highest relevance score among the actin-related pathways.The superpathway consists of 4 subpathways, including a total of 344 genes, and 109 genes in our NGS data were found in the superpathways (Table 4).
Next, g:Profiler (https://biit.cs.ut.ee/gprofiler/orth) was used to examine gene ontology to compile a list of important DEGs.Gene set enrichment using functional classification, e.g., the studies of cellular components (CC), molecular functions (MF), and biological processes (BP) were performed.Supplementary Table 1 displays the biological process subcategories in which gene expression was substantially changed in response to TAZ knockdown.Supplementary Table 1 categorizes the genes with significantly increased or decreased expression by TAZ knockdown that are involved in biological processes.Furthermore, we classified genes with significantly increased or decreased expression by TAZ knockdown into molecular functions (Supplementary Table 2).Previous studies revealed that the Rho family GTPases, RhoA, Cdc42, and Rac1, are associated with cytoskeleton rearrangements and cell motility (33,34).Our NGS data revealed that several genes which are involved in regulation of actin dynamics and Rho GTPases were downregulated in MDA-MB-231 cells with TAZ knockdown (Tables 3, 4).The alterations in the abundance of Rho GTPase proteins and mRNAs were further examined in MDA-MB-231 cells with siRNA-mediated TAZ knockdown.siRNA-mediated TAZ knockdown significantly decreased the protein abundance of TAZ (5 ± 2% of control-siRNA, Figures 4A, B), RhoA (50 ± 5% of control-siRNA, P < 0.05, Figures 4A, D).In contrast, YAP, ROCK, RhoC, Cdc42, and Rac1 were unchanged (Figures 4A, C, E-H).In addition, quantitative real-time PCR demonstrated that TAZ-siRNA mediated TAZ knockdown significantly decreased TAZ mRNA expression (10 ± 6% of control-siRNA, P < 0.05, Figure 5A).In contrast, RAC1 mRNA levels were increased (130 ± 7% of control-siRNA, P < 0.05, Figure 5H) in TAZ-knockdown condition, whereas YAP, RHOA, RHOC, ROCK1, ROCK2 and CDC42 mRNA levels were unchanged (Figures 5B-G).
Altered stress fiber formation and expression of myosin light chain kinase in breast cancer cells with siRNA-mediated TAZ knockdown We examined the changes in the stress fiber formation in MDA-MB-231 cells in response to TAZ knockdown.Stress fibers were stained using phalloidin, demonstrated in red (Figures 8, 9), and vinculin was immunolabeled in green as a marker of focal adhesion (arrows in Figures 8C, D).In control-siRNA-treated MDA-MB-231 cells, stress fibers were observed to be widely distributed intracellularly in both the central and peripheral regions of the cells (Figures 8A, E, 9A, C, D, F), and immunolabeling of vinculin was observed at the ends of the stress fibers (arrows in Figure 8C).Moreover, stress fibers were more abundantly expressed on the rear side of the cells (arrows in Figure 9D), and lamellipodia were predominantly detected on the front side (arrowheads in Figure 9D).On the contrary, TAZ-siRNA-treated MDA-MB-231 cells exhibited a decrease in the labeling of stress fibers, observed mainly in the central regions of the cells (asterisks in Figures 8B, 9H,  K) and the immunolabeling of vinculin (focal adhesion) was mainly observed at the periphery of the cells (arrows in Figure 8D).Furthermore, the disassembly of stress fibers was observed in MDA-MB-231 cells treated with TAZ-siRNA (Figures 8B, F, 9G,  I, J, L).We quantified the intensity of phalloidin (stress fiber) staining using ImageJ.The five regions of interest per cell were randomly selected and measured.Fluorescence intensity was measured in 10 cells per group, and a significant decrease in the phalloidin intensity was demonstrated in TAZ-siRNA treated MDA-MB-231 cells (68 ± 10% of control-siRNA, P < 0.05, Figure 8G).

Discussion
The characteristics of metastatic breast cancer cells are diverse, including EMT and increased AQP5 expression (14,(36)(37)(38).In particular, the metastatic breast cancer cells in human patients display an increased level of TAZ expression (39,40).In the present study, we demonstrated that TAZ knockdown inhibited the migration of highly invasive triple-negative breast cancer cells (MDA-MB-231) by reducing the protein abundance of RhoA and Rho-dependent kinases, particularly LIMK1 and MLCK.The downregulation of RhoA and Rho-dependent kinases in response to TAZ knockdown is likely to play a role in the reduction of stress fibers in the central region of the cells, disassembly of stress fibers, and diffuse focal adhesions throughout the cell periphery, resulting in a decrease in cell migration.Although the findings were obtained purely from invasive triple-negative breast cancer cells, the results may suggest that TAZ could be investigated as a therapeutic target for the inhibition of the migration of breast cancer cells and metastasis.characteristics.Even though there was less TAZ in MCD-7 cells than in MDA-MB-231 cells, siRNA effectively reduced TAZ levels in both cell lines.Moreover, TAZ knockdown decreased cell migration in both MCF-7 and MDA-MB-231 cells.However, we conducted subsequent experiments exclusively with MDA-MB-231 cells due to the limited cell proliferation and migration observed in MCF-7 cells.Furthermore, previous studies demonstrated an association between higher levels of the TAZ protein and aggressive and invasive metastatic breast cancer cells (39,41).

TAZ knockdown did not induce EMT in MDA-MB-231 cells
Several previous studies have demonstrated that TAZ induces EMT in various types of cancer (42)(43)(44).We examined whether TAZ knockdown affects the changes in EMT makers in MDA-MB-231 cells.As shown in Figure 2, an increase in fibronectin abundance may indicate an increase in the mesenchymal characteristics of cancer cells.On the other hand, the changes in E-cadherin, occludin, N-cadherin, and vimentin did not exhibit the typical pattern of EMT progression.
Then, we examined the effects of TGF-b treatment on cell migration under TAZ knockdown.An experiment was conducted to determine whether inducing TAZ knockdown in MDA-MB-231 cells, which already exhibit mesenchymal characteristics, could inhibit the morphological changes and cell migration induced by TGF-b treatment.Interestingly, TGF-b treatment resulted in a more pronounced reduction in the protein abundance Decrease in RhoA, phosphorylated LIMK1, and phosphorylated cofilin in response to TAZ knockdown in MDA-MB-231 cells The NGS analysis of control-siRNA-treated and TAZ-siRNAtreated MDA-MB-231 cells showed substantial changes in the expression levels of genes encoding small GTPases, Rho-GAPs, and Rho-GEFs, all of which are involved in the organization of the actin cytoskeleton.The small GTPases RhoA and RhoC exhibited the highest levels of expression within the Rho family, as demonstrated by NGS analysis.RhoA tended to decrease with TAZ knockdown, while RhoC showed no difference.Semiquantitative immunoblotting in Figure 4 revealed a significant decrease in RhoA protein levels in TAZ-siRNA-treated MDA-MB-231 cells.We tested other small GTPases (Cdc42 and Rac1), but there were no differences in mRNA or protein levels.Rho-dependent kinase (ROCK) is a downstream effector of RhoA and RhoC, which regulates actomyosin assembly, actomyosin contractility, and reorganization of the actin cytoskeleton by regulating the phosphorylation of downstream kinases or phosphatases (45).The binding of active RhoA to ROCK initiates a phosphorylation cascade that controls the dynamics of cytoskeletal actin (46,47).ROCK mRNA and protein levels stayed the same in MDA-MB-231 cells treated with TAZ-siRNA.However, the lower expression of RhoA may have an effect on the kinases that follow in the RhoA-dependent pathway.In particular, LIM kinase (LIMK), activated by the small GTPase Rho and its downstream protein kinase ROCK, phosphorylates cofilin, an actindepolymerizing factor, to regulate actin cytoskeletal reorganization.(35).Notably, we demonstrated that RhoA and LIMK1 were significantly decreased by TAZ knockdown in MDA-MB-231 cells.As shown in Figure 7, the TAZ knockdown resulted in a significant decrease in the abundance of phosphorylated LIMK (LIMK1 at threonine 508 and LIMK-2 at threonine 505) at ~72 kDa, which represents an activated form of LIMK.Furthermore, in MDA-MB-231 cells treated with TAZ-siRNA, phosphorylated cofilin at serine 3 was significantly reduced, thereby enhancing cofilin-induced actin depolymerization (48).The findings indicate that TAZ knockdown reduces RhoA protein expression, which subsequently causes the downregulation of the RhoA/ROCK/ LIMK1/cofilin pathway, resulting in a change in actin dynamics.In accordance with this observation, the migration of MDA-MB-231 cells was found to be diminished upon RhoA knockdown.On the other hand, the results of the immunoprecipitation assay did not provide any evidence of direct binding between TAZ and RhoA.Therefore, it is necessary for future research to investigate the underlying mechanisms by which TAZ controls the expression of RhoA and downstream kinases.
We also examined the changes in actin dynamics by phalloidin staining and immunofluorescence labeling.We observed the changes in the expression of vinculin, a highly expressed anchor protein involved in focal adhesion that interacts with F-actin and the membrane.Vinculin (Figure 8) was mainly observed at the end of stress fibers distributed in both the periphery and central region of the control-siRNA-treated MDA-MB-231 cells.On the other hand, in response to TAZ knockdown, the cells showed exclusive vinculin localization at the periphery, with no apparent stress fiber staining in the central region.These findings are consistent with actin depolymerization that occurred due to decreased phosphorylation of cofilin.

Decreased MLCK and phosphorylated MLC in response to TAZ knockdown in MDA-MB-231 cells
Rho-kinase and MLCK are known to regulate the organization and contraction of the stress fibers in the central region of the cells and the organization of focal adhesion (49)(50)(51).siRNA-mediated TAZ knockdown in MDA-MB-231 cells banished the stress fibers and focal adhesion in the central region of the cells.We looked at the Rho/ROCK/MLCK pathway, which controls the formation and contractility of central stress fibers, to see if TAZ knockdown changed the phosphorylation of MLCK and MLC.Consistently, in TAZ-siRNA-treated MDA-MB-231 cells, the protein abundance of MLCK and phosphorylated MLC2 at serine 19 was significantly decreased, but the total abundance of MLC2 was unchanged.
In summary, the findings observed in response to TAZ knockdown demonstrated that a decrease in LIMK1 and phosphorylated cofilin caused actin depolymerization, while a reduction in MLCK and phosphorylated MLC2 led to the disassembly of stress fibers and focal adhesions, ultimately resulting in a decrease in breast cancer cell migration.

1
FIGURE 1 Semiquantitative immunoblotting of TAZ and cell migration assay in breast cancer cells (MCF-7 and MDA-MB-231) treated with control-siRNA or TAZ-siRNA.(A-D) Changes in protein abundance of TAZ in breast cancer cells treated with control-siRNA or TAZ-siRNA for 24 hours.*P < 0.05, when compared with control-siRNA.The immunoblots were reacted with antibodies against TAZ (~49 kDa) and b-actin (~42 kDa).n indicates the number of cell preparation in each group.(E-H) Transwell cell migration assay of breast cancer cells treated with control-siRNA or TAZ-siRNA.The numbers of migrated cells were counted in the randomly selected fields (x100) per well.n indicates the number of randomly selected fields in each group.*P < 0.05, when compared with control-siRNA.(I-L) Scratch wound healing assay.*P < 0.05, when compared with control-siRNA.n indicates the number of wells containing cell treated with control-siRNA or TAZ-siRNA.

5
FIGURE 5 Quantitative real-time PCR for Rho family GTPases.(A-H) mRNA levels of TAZ, YAP, RhoA, RhoC, ROCK1, ROCK2, Cdc42, and Rac1 in MDA-MB-231 cells treated with control-siRNA or TAZ-siRNA.*P < 0.05, when compared with control-siRNA.n indicates the number of cell preparation in each group.

6
FIGURE 6 Semiquantitative immunoblotting of RhoA-siRNA-treated MDA-MB-231 cells and cell migration assay in MDA-MB-231 cells treated with control-siRNA, TAZ-siRNA, or RhoA-siRNA, respectively.(A-D) Changes in protein abundance of RhoA and TAZ in MDA-MB-231 cell treated with control-siRNA or RhoA-siRNA for 24 hours.*P < 0.05, when compared with control-siRNA.The immunoblots were reacted with antibodies against RhoA (~22kDa), TAZ (~49 kDa), and b-actin (~42 kDa).n indicates the number of cell preparation in each group.(D, E) Transwell cell migration assay of MDA-MB-231 cells treated with control-siRNA, TAZ-siRNA, or RhoA-siRNA treatment.The numbers of migrated cells were counted in the randomly selected fields (x100) per well.n indicates the number of randomly selected fields in each group.*P < 0.05.

FIGURE 8 Fluorescence
FIGURE 8 Fluorescence microscopic examination of stress fibers and immunolabeling of vinculin in MDA-MB-231 cells treated with control-siRNA (A, C, E) or TAZ-siRNA (B, D, F).Nuclei were labeled with 4',6-Diamidino-2-Phenylindole (DAPI).Scale bars, 20 mm.(G) The quantification of phalloidin intensity.Relative phalloidin intensity was measured by averaging the fluorescence intensity of 5 randomly selected regions of interest in MDA-MB-231 cells.n indicates the number of cells in each group.*P < 0.05.

TABLE 1
Primer sequences for quantitative real time PCR.

TABLE 2
Significantly changed gene expressions.

TABLE 3
Cytoskeleton remodeling regulation of actin cytoskeleton by Rho GTPases.

TABLE 3 Continued
*Fold change: siRNA-mediated TAZ knockdown vs. control-siRNA treatment in MDA-MB-231 cells.Five pathways included in the Cytoskeleton remodeling Regulation of actin cytoskeleton by Rho GTPases. 1. Cell adhesion Integrin-mediated cell adhesion and migration.2. Cytoskeleton remodeling Integrin outside-in signaling.3. Cytoskeleton remodeling Regulation of actin cytoskeleton by Rho GTPases.4. Cell adhesion Tight junctions.5. Development MAG-dependent inhibition of neurite outgrowth.o indicates the genes involved in each pathway.x indicates the genes not involved in each pathway.

TABLE 4
Actin nucleation by ARP-WASP complex.