USP47-mediated deubiquitination and stabilization of YAP contributes to the progression of colorectal cancer

The Hippo tumor suppressor pathway plays an important role in development and tumorigenesis. Yes-associated protein (YAP) is a major effector of the Hippo pathway regulating the expression of genes involved in cell proliferation, cell death, and cell differentiation (Pan, 2010; Johnson andHalder, 2014; Piccolo et al., 2014; Yu et al., 2015). In intestine, YAP ismainly expressed in leucine rich repeat containing G protein-coupled receptor 5 (LGR5) marked stem cells, and is required for regeneration following tissue damage and the development of colon cancer inmice (Cai et al., 2010; Barry et al., 2013;Wang et al., 2017).Moreover, in colorectal cancer (CRC) specimens, elevated YAP expression and activity is frequently observed (Steinhardt et al., 2008). It has been shown that high YAP expression is associated with cancer metastasis, poor prognosis, resistance to chemotherapy, and greater possibility of relapse (Johnson andHalder, 2014; Yu et al., 2015). However, the molecular mechanism underlying elevated YAP protein expression in CRC and other cancers remains poorly defined. The ubiquitin-proteasome system (UPS) plays a critical role in tumorigenesis (Popovic et al., 2014). Dysregulated expression of E3 ligases or deubiquitinating enzymes (DUB) is frequently observed in human cancers, including CRC. The protein stability of YAP is mainly regulated by phosphorylation and ubiquitination, and the latter is carried out by beta-transducin repeat containing E3 ubiquitin protein ligase (β-TRCP) (Zhao et al., 2010). However, the DUB responsible for YAP deubiquitination and stabilization in CRC is currently unknown. It has been shown previously that ubiquitin specific peptidase 7 (USP7, a DUB) interacts with β-TRCP (Peschiaroli et al., 2010). We speculated that β-TRCP and USP47 may work together to fine-tune the ubiquitination and protein stability of YAP. To verify this hypothesis, we first tested the interaction between USP47 and YAP. In co-immunoprecipitation (Co-IP) assays, recombinant USP47 could pull-down YAP and vice versa, suggesting that USP47 and YAP physically interact with each other (Fig. 1A). The intermolecular interaction was mediated by the peptidase domain (C19C) and coiled coil (CC) domains in USP47 and N-terminal domain in YAP (Fig. S1). In an in vitro deubiquitination assay, addition of USP47 led to deubiquitination of YAP in a dose dependent manner (Fig. S2). Consistently, overexpression of USP47 resulted in a decrease of ubiquitination of ectopic or endogenous YAP in HEK293T cells (Fig. 1B and 1C). Moreover, knockdown of USP47 in HEK293T cells increased YAP ubiquitination (Fig. 1D and 1E). Taken together, these results indicate that YAP physically interacts with USP47, and USP47 serves as a DUB for YAP. Ubiquitination often is coupled with protein destabilization via proteasomal degradation. In the presence of cycloheximide (CHX, an inhibitor of protein translation), overexpression of USP47 notably delayed the protein turnover of YAP in HEK293T cells (Fig. S3). On the other hand, in HEK293T, HCT116, and HT29 cells, knockdown of USP47 induced YAP degradation, and this effect was rescued by inhibition of proteasome (by MG132) or ectopic expression of shRNA resistant USP47 (Figs. 1F and S4). Thus, USP47 is critical in regulating protein stability of YAP. We retrieved gene expression omnibus (GEO) dataset GDS2609 which contains mRNA expression data of colon mucosae from early onset CRC patients and healthy controls, and analyzed the mRNA expression patterns of USP47, YAP, and YAP target genes. YAP mRNA levels in CRC samples and healthy controls are similar, however, the mRNA levels of USP47 and representative YAP target genes, such as CYR61, TAGLN, and CTGF, are increased in CRC patients’ mucosae (Figs. 1G and S5). Hence, USP47 mRNA level is elevated at the early stage of CRC, and may promote the expression of YAP target genes. Consistently, in HCT116 cells, knockdown of USP47 has no effect on YAP mRNA level but dramatically decreased YAP protein levels and the expression of YAP target genes (Fig. 1H and 1I). Collectively, these data suggest that USP47 promotes YAP stability and transcriptional activity through a post-transcriptional mechanism. We also tested the effect of OTU domain-containing ubiquitin aldehyde-binding protein 2 (OTUB2) and OTU deubiquitinase 1 (OTUD1), two recently reported DUBs for YAP in mammary cells (Yao et al., 2018; Zhang et al., 2019).


Stable Cells
The ORF of YAP was cloned into lentiviral vector pLVX-IRES-neo for stable cells establishment.

Colony Formation Assay and Growth Curve
The proliferation potential of cells was assessed by plating 2000/4000/8000 cells separately in 6-well plates. After 10 days of growth, cells were fixed with methanol, then stained with crystal violet (Sigma) for 30 mins at room temperature and the number of colonies was counted. For the growth curve assay, 2000 cells were plated in 96-well plates with four replicates. CellTiter 96® AQueous One solution cell proliferation assay Kit (Promega) was used to measure the cell growth. The original medium was removed and 100 μl empty medium mixed with 20 μl agent was added in per well. After 1 hr of incubation, the absorbance was measured at OD490. The absorbance of the initial time point was measured after 6 hr of seeding cells to normalize the following results. Then the absorbance was measured every 24 hours.

Statistical analysis
All statistical analyses were performed using SPSS program (SPSS Inc, Chicago, IL).
Experimental results were evaluated statistically using the Student t test, the Student t' test, the Pearson chi-square test, and the ANOVA test. P < 0.05 was considered statistically significant. Datasets GSE23232 and GSE17537 were downloaded from GEO database and relevant information was obtained by GEO2R.

RNA isolation and Real-time PCR
Total RNA was prepared with Trizol reagent (Invitrogen) and the cDNA was synthesized using M-MLV Reverse Transcriptase (TaKaRa). Quantitative real-time PCR (qRT-PCR) was performed using a 7500 detection system (Applied Biosystems) and Go Taq qPCR Master Mix (Promega). The relative mRNA levels of target samples to that of control samples were calculated according to 2 -∆∆Ct method, in which the difference in Ct values (∆ Ct) between the target gene and the reference gene internal control (GAPDH) was calculated for normalization and the ∆Ct of the different samples was compared directly (∆∆Ct).

Inhibitors
The human CRC cell lines (HCT116 and HT29) were incubated with USP47/USP7 dual inhibitor P5091 (SelleckChem) at the indicated concentrations and time. To determine the mechanism in which USP47 regulates the protein level of YAP, cells were treated with 20μM MG132 (Sigma) or 100μg/ml CHX (INALCO) to inhibit proteasome or protein synthesis separately. Vehicle-treated cells were used as controls.

Immunohistochemistry
Tissue microarrays containing 90 colon tumors and the corresponding normal colonic mucosa were from by Shanghai Outdo Biotech Co., Ltd (Cat No. FM-S4006-1 Vision1.0/Lot No. HCol-Ade180CS-01). Tissue microarrays or adjacent tissues were incubated with USP47 antibody (Santa Cruz), or YAP antibody (Cell Signaling Technology). Immunohistochemistry was performed using EnVision Kit from DAKO.
The staining intensity was scored using ImageJ software.

siRNA and Oligos
The oligonucleotides were synthesized chemically by Sangon Biotech (Shanghai, (A) Protein domains of USP47 and YAP, related deletion mutants were used in co-immunoprecipitations (Co-IP) in B and C; (B) C19C and CC domains of USP47 is required for interaction with YAP. HA-tagged USP47 deletion mutants were co-transfected with Flag-tagged YAP into HEK293T cells, and cell lysates were used for Co-IP with HA antibody; (C) N-terminal (NT) of YAP is required for interaction with USP47. Fla-tagged YAP were co-transfected with HA-tagged USP47 into HEK293T cells, and cell lysates were used for Co-IP with Flag antibody.

Figure S2. USP47 deubiquitinates YAP in vitro.
Flag-YAP and HA-Ub were overexpressed in HEK293T cells. Flag-YAP was immunoprecipitated with α-Flag beads and then incubated with recombinant USP47 protein. The ubiquitination of YAP was analyzed by immunoblotting with anti-YAP or anti-HA antibody.     HEK293T cells were transfected with Flag-tagged YAP or HA-tagged Ub, and treated with or without P5091 for indicated time. Flag-YAP was immunoprecipitated and probed with HA antibody. The HA-Ub signaling was increased in cells treated with P5091 for 24 or 36 hrs. Figure S10. Knockdown of USP7 has no effect on YAP protein expression. USP7, YAP and β-actin protein expressions were determined in control or USP7 knockdown HCT116 and HT29 cells.  TABLE   Table S1. The expression levels of USP47 or YAP in CRCs.