USP13 Regulates the Breast Tumor Proliferation and Migration through the Stabilization of PDCD4 Protein

Background: Programmed cell death protein 4 (PDCD4), which serves as a tumor suppressor protein, plays a important role in cell proliferation,apoptosis, cell migration and DNA-damage response.However, the exact mechanism for the deubiquitination of PDCD4 remain unclear. Methods: Western blotting was used to detect the expression of PDCD4 in the breast cancer tissues and BC cell lines. We identied the potential PDCD4 associated deubiquitinase by RNAi screening. GST-Pull down and domain-mapping analysis were used to reveal that USP13 and PDCD4 directly interact with each other.Flow cytometry was used to detect the changes of G1 to S phase. Soft agar assay was used to measure the changes of the cell proliferation eciency. Results: The expression of PDCD4 was decreased in the breast cancer tissues and BC cell lines. USP13 as a potential PDCD4 associated deubiquitinase. USP13 physically interacted with PDCD4 and greatly increased the steady state of PDCD4 through the ubiquitin-proteasome pathway.Importantly, silencing of the USP13 facilitated cell cycle from G1 to Sphase, promoted breast tumor cells proliferation and migration through downregulation of PDCD4. Conclusions: Together, these results suggest that USP13 plays an important role in the breast tumor proliferation and migration through modulating PDCD4 stability.


Introduction
Breast cancer, as the primary malignant tumor of female cancers, which resulted in million deaths yearly [1]. Early stage breast cancer could be treated by surgery [2] and radical radiotherapy [3], which can reduce local breast tumor recurrence. However, effective therapies for both radiotherapy-resistant and advanced BC remain lacking in the clinic. Emerging evidence has reported that abnormal cell proliferation [4] and migration [5,6] are involved in BC tumorigenesis.Therefore, it is essential to explore the molecular mechanisms, which will provide new targeted treatment strategies to improve the long-term survival of BC patients.
Deubiquitinating enzymes (DUBs) are a group of proteases that reverse the functional effects of ubiquitination by cleaving ubiquitin from ubiquitylated proteins [7,8]. DUBs, composed of 100 deubiquitinating enzymes,are classi ed into six subfamilies based on its domains [9]. It has long been known that DUBs regulate a series of essential proteins,which implicated in Anti-cancer or oncogenesis [10][11][12]. Ubiquitin-speci c peptidases (USP) have been reported to be the largest subfamilies of DUBs.
In the study, we observed that PDCD4 was decreased at protein level in the breast cancer tissues and cell lines. Here, we newly identi ed the deubiquitinating enzyme, USP13, may as an PDCD4 associated deubiquitinase by screening the USPs-shRNA library.USP13 stabilized PDCD4 protein via direct binding and deubiquitination of PDCD4. We con rmed USP13 as the deubiquitylase that stabilizes the PDCD4 protein by protecting PDCD4 from proteasome-mediated degradation. Silencing of USP13 enhanced PDCD4 ubiquitination, thus accelerating the transformation of cell cycle G1/S phase and promoting BC proliferation.Meantime, loss of USP13 promoted the protein expression involved in the cell migration through downregulation of PDCD4 in breast cancer cells. Conversely, overexpression of USP13 suppressed migration in wild-type but not its catalytically mutant. In the present study, we clarify that USP13 plays an important role in the regulation of BC proliferation and migration through modulating the stability of PDCD4.

Reagents,tissue specimens and cell culture
The primary antibodies were purchased from Abcam(Cambridge, UK) and Santa Cruz Biotechnology(Santa Cruz, CA, USA).All experiments were carried out in accordance with the Guidelines of the Jiaozuo Women's and Children's Hospital (Jiaozuo, China), and experiments were performed according to the ethics committee at the Jiaozuo Women's and Children's Hospital. Informed consents were acquired from all participants of this study. 30 breast patients' and adjacent normal tissues were respectively obtained from the Pathology Department, the Jiaozuo Women's and Children's Hospital. (Cat: 11415-064, Gibco) supplemented with 10% FBS.T47D and ZR-75-30 cells were cultured in RPMI 1640 (Gibco,catalog no. 12633012) supplemented with 10% FBS (Cat#1500-500, Seradigm). Cells were incubated at 37°C in 5% CO2/95% mixed ambient air.

Western blot analysis and Coimmunoprecipitation (co-IP) assay
The Western blot was performed as we describe below. Brie y,the cells were collected and incubated in RIPA buffer (Thermo Fisher Scienti c, USA) and then the centrifuged protein lysates were separated by the SDS-PAGE gel. The protein bands were incubated with primary antibodies (anti-USP13, anti-PDCD4, anti-Flag, anti-Myc, Anti-HA and anti-GAPDH) at 4°C for overnight, followed by incubation in secondary antibody at room temperature for 1 h. The bands were visualized using the chemidoc XRS imaging system (Bio-Rad) using an enhanced ECL detection reagents (Thermo).
Alternatively,the whole-cell protein lysates were incubated with primary antibody pre-absorbed 5 μL protein G sepharose beads at 4°C for 6 h. Then, the incubation solution were washed with the RIPA lysis buffer, boiled in 2×loading buffer for 10 min, and analyzed by Western blot.

Immuno uorescent staining
Cells were xed with 4% paraformaldehyde followed by permeabilization using Triton X-100 and closed with 10% goat serum and 1% BSA for 1 h. Cells were incubated with PDCD antibody at 4 ℃ for overnight, then were stained with DyLight 594-conjugated secondary antibody together with DAPI for 1 h. Cells were examined and analysed in two independent experiments in triplicate by a uorescence microscope.

GST pull down Assays
According to the manufacturer's instructions, BL21 cells were transfected with pGEX-4T-1-USP13 construct and induced by 0.1m ispropyl b-D-1-thiogalactopyranoside (IPTG) at 20°C overnight.Bacterially expressed GST-USP13 proteins were puri ed with glutathione-sepharose beads (Beyotime cat:P2226) and incubated with biosynthesis His-PDCD4 for overnight at 20 °C. The processed incubation were immunoblotted with USP13 antibodies overnight at 4°C . Finally,the incubation were eluted at 100 °C for 10min, and subjected to western blotting.

Construction of the recombinant vector and cell transfection
The full-length USP13 open reading frame was ampli ed from MCF-7 by RT-PCR, and cloned into the

Protein Half-Life Assays
Cells were incubated with cycloheximide(50 μg·mL −1 ) to block protein synthesis. The protein was cellected for various periods of time and western blot analysis was conducted as described above.

In vivo Ubiquitylation Assay.
MCF-7 and ZR-75-30 cells transfected HA-Ub were transfected with the indicated lentiviruses or plasmid, followed by treated with 20 μM MG132 for 6 h. Then, Cell lysate and dilution buffer (10 mM Tris·HCl, pH8.0,150 mM NaCl, 2 mM EDTA, 1% Triton) were incubated at 4 °C for 30 min with rotation. Protein fraction were obtained by centrifugation and incubated with the indicated anti-PDCD4 antibodies overnight. Next, protein-antibody complexes and protein A/G agarose beads were incubated for an additional 1 h at room temperature, followed by centrifugation. The beads were washed three times with wash buffer (1M NaCl,10 mM Tris·HCl, 1mM EDTA, 1% Nonidet P-40) and boiled in SDS/PAGE sample buffer. The ubiquitination levels were analyzed using immunoblotting with anti-Ub antibodies.

PDCD4 was decreased in Breast cancer tissues and cell lines
Previous studies have demonstrated that PDCD4 was downregulated in breast cancer samples [31]. To gain a closer look at the correlations between PDCD4 expression and BC clinicopathologic features, we again examined the protein expression of PDCD4 in paired non-tumor tissue and BC obtained from the Pathology Department using Western blot. The results con rmed that PDCD4 protein levels were decreased in BC tissues compared with adjacent normal tissues (Fig. 1A). We further analyzed the expression of PDCD4 in BC cell lines and found that endogenous PDCD4 expression was decreased in BC cell lines, compared with the human breast epithelial cells (MCF-10A) (Fig. 1B)

Screening of DUBs for PDCD4
Multiple E3 ubiquitin ligase-SKP2, DTL, and CRL3IBTK-have been reported to induce PDCD4 ubiquitylation and degradation. Importantly, ubiquitination and deubiquitination are the two main types of protein post-translational modi cation that can regulate protein homeostasis [32]. Therefore, it is likely that a deubiquitinase related with PDCD4 to prevent its ubiquitin proteasome-mediated degradation. To identify PDCD4 speci c DUBs, we employed short hairpin RNAs (shRNAs) to individually inhibit the expression of 30 DUBs. MCF-7 cells transfected with lentivirus carrying these shRNAs were analysed for PDCD4 using immuno uorescence.Interestingly,we found that loss of USP13 induced downregulation of PDCD4 (Fig. 2).

USP13 interacts with PDCD4
As depletion of USP13 affects PDCD4 expression, we questioned whether USP13 interacts directly with PDCD4. Coimmunoprecipitation analysis con rmed that endogenous PDCD4 was pulled down by Flag-USP13 (Fig.3A), and that endogenous USP13 was pulled down by Myc-PDCD4 (Fig.3B).The interaction between endogenous USP13 and endogenous PDCD4 was also detected by coimmunoprecipitations (Fig.3C). Based on the above conclusions, we next determined which domain of USP13 is responsible for PDCD4 interaction. USP13 consists of an UBP (ubiquitin-speci c processing protease) domain, an USP and two UBA domains [33]. (Fig. 3D). We tested the interaction between Myc-PDCD4 and Flag -UBP domain, Flag -amino terminuses (an USP and two UBA domains) of USP13 or Flag-USP13 employing coimmunoprecipitation assay. Domain-mapping analysis revealed that PDCD4 was pulled down by amino terminuses of USP13 and full length USP13, suggesting that USP13 bound PDCD4 through its amino terminuses (Fig. 3E). To determine whether USP13 and PDCD4 directly interact with each other, we puri ed recombinant USP13 and PDCD4 in vitro. GST-USP13, but not GST-control , was able to pull down His-PDCD4 (Fig. 3F). Together, these data support that USP13 physically interacts with PDCD4.

USP13 is an PDCD4 deubiquitinase
Based on the above results that silencing of USP13 decreases PDCD4 expression and the convinced interaction of USP13 with PDCD4, we investigated whether USP13 affects the protein levels of PDCD4. To verify the function of USP13 as an PDCD4 deubiquitinase, we rst transfected siRNAs-USP13 into MCF-7. Intriguingly, silencing of USP13 resulted in a signi cant decrease of PDCD4 protein levels (Fig.4A).
In contrast, USP13 overexpression resulted in a signi cant increase of PDCD4 levels by performing the overexpression experiment in the MDA-MB-231 cells. However, overexpression of a catalytically inactive USP13 mutant (C345A) or employing spautin-1 that speci cally inhibit USP13 activity had no effect on PDCD4 levels, suggesting that USP13-mediated PDCD4 upregulation may depend on the USP13 deubiquitinase activity (Fig.4B).Furthermore, USP13 knockdown alone abrogated PDCD4 levels and decreased protein expression of PDCD4 was rescued by proteasome inhibitor MG132 (Fig.4C), suggesting that USP13 stabilizes PDCD4 by abrogating its proteasomal degradation. Because USP13 regulates the levels of PDCD4, we speculated USP13 is able to stabilize PDCD4. To con rm this hypothesis, the de novo protein synthesis in MCF-7 was restrained by cycloheximide(CHX) in the USP13 knockdown. We found that USP13 knockdown shortened the half-life of PDCD4 (Fig.4D). Considering that USP13 stabilizes protein by promoting their deubiquitination, we further measured the levels of PDCD4 polyubiquitylation. The results showed that USP13 knockdown in MCF-7 cells enhanced ubiquitination of endogenous PDCD4. Conversely, overexpression of USP13 but not the enzymatic inactive mutant USP13 reversed PDCD4 ubiquitination levels (Fig.4E). These data clarify that USP13 is an PDCD4 deubiquitinase.

USP13 alters cell cycle G1/S phases and regulates breast cancer cell proliferation in a PDCD4-Dependent Manner
The above research showed that USP13 positively regulated PDCD4 expression by promoting PDCD4 deubiquitination, we questioned whether USP13 inhibits tumorigenesis by stabilizing PDCD4. To test this hypothesis, the percentage of cells with S phase DNA content from MCF-7 cells was determined. This is manifested by increased accumulation of cells in S phase with USP13 knockdown in MCF-7 cells compared to controls.However, silencing of USP13 transfected with exogenous PDCD4 could reverse the cell G1/S transition process (Fig.5A), suggesting that USP13-mediated G1/S transition is dependent on PDCD4. Then, We probed whether USP13 inhibited cell proliferation via PDCD4 suppression. The cell proliferation e ciency was determined by colony-forming assay in MCF-7 cells. USP13 knockdown in MCF-7 cells displayed higher cell proliferation compared with control cells, and USP13-depleted cells transfected with exogenous PDCD4 fully reversed the cell proliferation (Fig.5B). These results demonstrate that USP13 regulates breast cancer cells proliferation by promoting PDCD4 expression.
3.6 USP13 may in uence cancer cell motility abilities by regulating PDCD4 It has long been known that PDCD4 is a essential regulator of tumor invasion and migration. Nieves et al found that overexpression of PDCD4 attenuates invasion in breast cancer and colon cells [34,35]. Consistent with the above ndings,PDCD4 knockdown promotes invasion [36,37]. Meantime, silencing of PDCD4 inhibits E-cadherin expression and activates β-catenin [38]. E-cadherin and vimentin have been reported to be markers of epithelial cells and mesenchymal cells,respectively [39]. Given that USP13 can mediate PDCD4 stability,it is likely that USP13 is participate in the expression of protein involved in the regulation of cell migration. To determine the effect of USP13 on E-cadherin and vimentin,we transfected a control-shRNA and USP13-shRNA into MCF-7 cells. The results indicated that USP13 alone knockdown inhibited E-cadherin protein expression and improved vimentin expression, but exogenous PDCD4 fully reversed E-cadherin and vimentin expression (Fig.6A).Furthermore, overexpression of wild USP13,but not a catalytically inactive USP13 mutant(C345A) improved E-cadherin expression and inhibited vimentin expression. In contrast, PDCD4-depleted cells transfected with exogenous USP13 had no effect on the expression of E-cadherin and vimentin. (Fig.6B). Therefore, we con rmed that USP13 plays a positive role in regulating the expression of protein involved in the regulation of cell migration by targeting PDCD4.

Discussion
Previous studies show that PDCD4 expression decreases in many types of cancers and PDCD4 inhibits the formation of protein initiation complexes, thereby inhibiting tumorigenesis [40].We found that expression of PDCD4 was decreased in the breast cancer tissues and BC cell lines. These results indicated that loss of PDCD4 may contribute to tumorigenesis and PDCD4 function as a tumor suppressor in breast cancer cells.
PDCD4 is a tumor suppressor with a relatively short half-life. PDCD4 is regulated mainly by posttranslational modi cations,such as phosphorylation, deubiquitylation and ubiquitylation.In the present study, we identify USP13 as a potential PDCD4 deubiquitylase.Our results show that USP13 and PDCD4 interact with each other. Silencing of USP13 decreases PDCD4 levels by increasing its ubiquitylation, whereas overexpression of USP13 stabilizes PDCD4 by removing ubiquitin chain. PDCD4 can be mediated by USP13 that directly deubiquitylation.
It has been reported that PDCD4 plays a functional role in tumor cells by regulating proliferation, migration, and invasion.Our results show that USP13 alters G1/S phases in a PDCD4-dependent manner, as USP13 knockdown alone promotes G1/S transition, whereas silencing of USP13 cells transfected with exogenous PDCD4 reverse G1/S transition. We also con rm that USP13 acts as tumor suppressor in breast tumor cells, as cells with USP13 depletion exhibit increased cell proliferation. This is accordance with previous reports that USP13 functions as a tumor suppressor [41][42][43]. In addition, USP13-depleted cells transfected with exogenous PDCD4 restored the cell proliferation and migration,indicating that USP13 exerts its function via PDCD4.

Conclusions
we identi ed Ubiquitin Speci c Peptidase 13 (USP13) as an PDCD4 associated deubiquitinase.Our nding that there is an interaction between USP13 and PDCD4. We con rmed USP13 as the deubiquitylase that stabilizes the PDCD4 protein by protecting PDCD4 from proteasome-mediated degradation. Silencing of USP13 enhanced PDCD4 ubiquitination, thus accelerating the transformation of cell cycle G1/S phase and promoting BC proliferation and migration.As far as we know, this is perhaps the rst report demonstrating that USP13 is an PDCD4 deubiquitinase.   Screening of DUBs for PDCD4 MCF-7 cells carrying out the screen using shRNAs to inhibit 30 USPs expression were analysed the expression of PDCD4.    A. USP13 knockdown alone inhibited E-cadherin expression and improved vimentin expression, however, exogenous PDCD4 reversed the trend. B. USP13,but not USP13 mutant,improved E-cadherin expression