The oncogenic role of MUC12 in RCC progression depends on c‐Jun/TGF‐β signalling

Abstract Renal cell carcinoma (RCC) is a common kidney cancer worldwide. Even though current treatments show promising therapeutic effectiveness, metastatic RCC still has limited therapeutic options so that novel treatments were urgently needed. Here, we identified that MUC12 was overexpressed in RCC patients and served as poor prognostic factor for RCC progression. Overexpression of MUC12 increased RCC cell growth and cell invasion while deficiency of MUC12 exerted opposite effects on RCC cells. Mechanistic dissection demonstrated that MUC12‐mediated RCC cell growth and cell invasion were dependent of TGF‐β1 signalling because they could be blocked in the presence of TGF‐β1 inhibitor. Moreover, the regulation of TGF‐β1 by MUC12 relied on the transactivation of c‐Jun. MUC12 promoted the recruitment of c‐Jun on the promoter of TGF‐β1, leading to its transcription. Importantly, knockdown of c‐Jun also attenuated MUC12‐mediated TGF‐β1 induction and RCC cell invasion. In summary, our study defines the role of MUC12 in RCC progression and provides rational to develop novel targeted therapy to battle against RCC.

hypoxia-inducible factors (HIF-1α and HIF-2α) and its inactivation increases HIF levels in RCC tumours. 8,9 The well-known target of HIF signalling is VEGF family, which plays central role in the development of neo-angiogenesis and tumour metastasis. Therefore, targeting the angiogenetic signalling pathway becomes an ideal therapeutic strategy for RCC patients. As a first-line medicine for metastatic RCC, sunitinib is designed to suppress angiogenesis by inhibiting receptor tyrosine kinases (RTKs) in both RCC cells and endothelial cells. 10,11 Mucin family belongs to glycoprotein and mainly locates in cell membrane. [12][13][14] The dysregulation of mucin proteins has been frequently observed in malignant cancers. 14 The participation of mucin proteins into cancer progression relies on its capacity to transduce intracellular signallings. Previous studies have demonstrated that MUC1 is overexpressed in several cancers including breast cancer and pancreatic cancer. 14,15 MUC1 not only serves as a cancer biomarker but also functions to regulate several biological events of cancer including proliferation, invasion, immune-therapy and drug resistance. 15 However, as one of mucin family member, MUC12 is rarely investigated, especially in RCC.
The transforming growth factor-β (TGF-β) signalling is involved in considerably biological events including cell survival, cell differentiation, immune response and cancer development. 16 Upon the TGF-β binding, TGF-β receptor is phosphorylated and activated, which in turn promotes the phosphorylation of SMAD2/SMAD3. The phosphorylated SMAD2/ SMAD3 is then complexed with SMAD4 and translocation to the nucleus to regulate a plethora of genes. 17,18 Initially, TGF-β signalling is considered to play tumour suppressing role by inhibiting cell proliferation and inducing apoptosis. 19 However, as tumour cells progress, they develop mechanisms to switch TGF-β signalling as a driving force for cancer progression. Indeed, mounting evidence suggests that TGF-β signalling plays central role in the process of epithelial-mesenchymal transition (EMT), an early process facilitates tumour cells to metastasize to distant organs. 20 In this study, we found that MUC12 was overexpressed in RCC patients compared with normal kidney tissues. Knockdown of MUC12 suppressed RCC cell growth and cell invasion while induction of MUC12 in RCC cells revealed the opposite phenotypes. Mechanistically, MUC12 triggered TGF-β1 activation via increasing its reliable transcription factor c-Jun. Inhibition of TGF-β1 or c-Jun could attenuate MUC12induced RCC cell growth and cell invasion. Overall, our data strengthen the oncogenic role of MUC12 in RCC progression and provide strong rational to develop novel therapy to better suppress advanced RCC.

| Human clinical samples
All tumour tissues and adjacent normal tissues were obtained from Department of Urology, The Affiliated Changzhou No. 2 People's Hospital of Nanjing Medical University, from 2016 to 2018. We performed IHC on 4% formaldehyde-fixed tissues. Twenty-four pairs of fresh tumour tissues and the corresponding peritumoural tissues were used for qRT-PCR and Western blot (WB) assays. HK2, SW839, 786-O, OSRC A498 and Caki-1 were obtained from   the American Type Culture collection (ATCC, Rockville, MD, USA).

| Cell culture
Cells were cultured in DMEM media containing 1% penicillin and streptomycin, supplemented with 10% foetal bovine serum (FBS) and maintained in a 5% (v/v) CO2 humidified incubator at 37°C.

| Quantitative real-time PCR
A 1 μg of total RNA was subjected to reverse transcription using Superscript III Transcriptase (Invitrogen, Grand Island, NY). Quantitative real-time PCR (qRT-PCR) was performed in a Bio-Rad CFX96 machine with SYBR green to determine the interested mRNAs. GAPDH mRNA was used as internal control. The primers used in this study were listed in Table S3.
Then, medium was removed and OD570 value was measured.

| Colony formation assay
RCC cells with gene manipulation were seeded into 6-well plate at the density of 500. 2 weeks later, cells were fixed by cold methanol and stained with 0.1% crystal violet.

| Invasion assay
A 8 µm transwells were pre-coated with diluted Matrigel (1:10 with serum-free medium) and dried for 2 hours in the incubator. Infected RCC cells were seeded into the upper transwell chamber at the density of 1 × 10 5 . 20% serum medium was added in the lower chamber to attract the cells. One day later, the invaded cells were fixed by cold methanol and stained with 0.1% crystal violet. Images were captured by microscopy.

| Western blot analysis
Renal cell carcinoma human samples or cells were lysed in RIPA buffer. 40 μg proteins were loaded for separation on 8%-12% SDS/ PAGE gel. Then, proteins were transferred onto PVDF membranes (Millipore, Billerica, MA, USA). After being blocked with 10% milk for 1 hour, the membranes were blotted with specific primary antibodies at 4°C overnight. Membranes were washed with 0.1% Tween-20 TBS and incubated with HRP-conjugated secondary antibodies for another one hour. Blots were visualized using ECL system (Thermo

| Analysis of TCGA data set
KIRC raw data were downloaded from TCGA database for the analy-

| Chromatin immunoprecipitation (ChIP)
ChIP assay was performed as described in previous report. 21 Protein-DNA complexes were cross-linked by formaldehyde for 10 minutes and then quenched by 125 mmol/L glycine for 10 minutes. Genomic DNAs were sonicated to an average of 500 bp size. After centrifugation, the DNA-protein complexes were precipitated with c-Jun antibody overnight at 4 ºC. Next, pre-cleared A/G beads were added for the purification of c-Jun binding DNAs.

| Luciferase assay
RCC cells were cotransfected with pGL3-TGFβ-WT or pGL-TGFβ-MUT (200 ng/well), c-Jun siRNAs (10 mmol/L) or negative siRNAs (10 mmol/L) and pRL-TK (5 ng/well) using Lipofectamine 3000 according to the manufacturer's instructions. 48 hours later, cells were lysed and the luciferase activity was detected by the dual luciferase assay using pRL-TK as the internal control. Each reading was performed in triplicate.

| Animal studies
The animal protocol was approved by the Institutional Animal Use and Care Committee of Nanjing Medical University. As for xenografts group, mice were inoculated subcutaneously with 1 × 10 6 /100 μL 786-O cells stably transfected with shMUC12. The implanted tumour volume of each mouse was monitored every 7 days, and tumour weight was calculated after 5 weeks when mice were sacrificed. Differences in different groups were analysed by Student's t test or one-way ANOVA. Correlations between different parameters were analysed using a Spear man rank test. Statistical tests were performed using GraphPad Prism 8.0. P < 0.05 was considered statistically significant. The detailed results were *P < 0.05, **P < 0.01 and ***P < 0.001.

| Analysis of TCGA data set showed that MUC12 levels were positively correlated with RCC progression
To investigate the potential role of MUC12 in RCC carcinogenesis and development, we first analysed its expression levels based on RNA-seq from TCGA database. Analysis showed that expression levels of MUC12 were significantly up-regulated in RCC patients (n = 523) compared with normal kidney tissues (n = 100) ( Figure 1A).
In addition, Kaplan-Meier overall survival (OS) and disease-free survival (DFS) analyses all suggested that MUC12 served as a poor prognostic factor for RCC patients ( Figure 1B-C). Importantly, MUC12 expression levels were also strongly correlated with clinical stage, pathological grade, tumour recurrence and tumour metastasis ( Figure 1D), showing that higher MUC12 levels were observed in more advanced RCC patients. Besides, the ROC curve revealed that MUC12 could be as a worse prognostic indicator (95% CI: 0.592-0.680, P < 0.001) for RCC patients ( Figure 1E). Overall, MUC12 was an independent risk factor for worse overall survival, tumour recurrence, grade and metastasis according to univariate or multivariate analysis ( Figure 1F-G and Table S1,S2). Together, all these data indicate that MUC12 may serve as tumour-promoting factor in RCC patients.

| Experimental examination of MUC12 in RCC patients
To verify the above RNA-seq data, we collected RCC patients as well as the corresponding normal tissues and examined MUC12 expression levels. Consistent with online analyses, data displayed that  Figure 2C). Taking together, we speculate that MUC12 plays an oncogenic role in RCC progression.

| MUC12 promoted RCC cell growth
To test whether MUC12 is a causal factor for RCC progression, we Besides, the linear curve also recorded that knocking down MUC12 dramatically suppressed the average weight ( Figure 3K) of tumours in nude mice. Together, these results indicate that MUC12 acts as a tumour supporting factor to promote RCC cell growth.

| MUC12 enhanced RCC cell invasion via promoting epithelial-mesenchymal transition (EMT) process
The ability of cell invasion is another essential hallmark for cancers. To  Figure 4E). In contrast, an increase of N-cadherin, Snail-1and Vimentin while reduction of E-cadherin and ZO-1 was observed in MUC12 expressing A498 cells ( Figure 4F). Western blotting analyses also confirmed the above findings ( Figure 4G). In addition, ZO-1 detection by immunofluorescent staining also verified that ZO-1 signals were reversely correlated with MUC12 levels ( Figure 4H). All these findings demonstrate that MUC12 promotes EMT of RCC cells to increase their invasive capacity.
F I G U R E 1 Analysis of TCGA data set showed that MUC12 levels were positively correlated with RCC progression. A, MUC12 was up-regulated in RCC patients compared to normal kidney tissues. B-C, MUC12 was poor prognostic factor for RCC overall survival (B) and disease-free survival (C). D, MUC12 was up-expression in high clinical stage, pathological grade, tumour recurrence and tumour metastasis RCC. E, ROC curve revealed that MUC12 could be as a worse prognostic indicator (95% CI: 0.592-0.680, P < 0.001) for RCC patients. F-G, Univariate (F) and multivariate (G) analysis showed that MUC12 was an independent risk factor for overall survival, tumour recurrence, tumour grade and cancer metastasis. *P < 0.05

| Inhibition of TGF-β1 attenuated MUC12mediated RCC cell growth and cell invasion
To find the underlying mechanism by which MUC12 promotes RCC progression, we performed single-gene set enrichment analysis (GSEA) using the MUC12 median level as a cut-off. Result exhibited that the activation of TGF-β1 signalling, a central signalling involved in cancer growth and migration, was observed in MUC12 high group ( Figure 5A). In addition, TCGA data set also showed that MUC12 was positively correlated with TGF-β1 at mRNA level ( Figure 5B, P < 0.001, r = 0.17). Therefore, we sought to examine whether MUC12 could directly affect TGF-β1 signalling. Expectedly, we found that MUC12

| MUC12-mediated RCC cell growth was dependent of c-Jun
c-Jun as a transcription factor has been reported to regulate TGF-β1 at the transcriptional level. Next, we sought to examine whether c-Jun was involved in MUC12 induced RCC cell growth and cell invasion. Importantly, knockdown of c-Jun by two independent siRNAs, whose efficiency was confirmed in Figure 6C by Western blotting, could reverse MUC12-mediated induction of TGF-β1 levels ( Figure 6D) as well as MUC12-mediated increase of RCC cell invasion ( Figure 6E). Taken together, these data suggest that c-Jun was required for MUC12 to activate TGF-β1 signalling and to promote RCC development.

| Activation of TGF-β1 by MUC12 relied on the transactivation of c-Jun
Next, we want to investigate whether MUC12 relies on c-Jun's transactivating ability to regulate TGF-β1. We first confirmed that c-Jun depletion by siRNAs could reduce TGF-β1 at both protein and mRNA level ( Figure 6G,H). By analysing the promoter region of TGF-β1 using c-Jun as a bait, we identified there were four potential DNA responsive elements of c-Jun ( Figure 7A,B). To end this, we designed two pairs of primers (c-JUNE1 and c-JUNE2) as it is difficult to separately analyse them, to examine whether c-Jun directly binds to TGF-β1's promoter. ChIP assay result indicated that c-Jun could bind to c-JUNE2 but not c-JUNE1 promoter region of TGF-β1 ( Figure 7C). Next, we constructed 2 kb wild-type promoter region of TGF-β1 (WT) as well as the c-JUNE2 mutated form (MUT) into pGL3 to investigate whether MUC12 and c-Jun could regulate its activity ( Figure 7D). As expected, depletion of c-Jun could suppress the WT activity but fail to affect it when c-JUNE2 was mutated ( Figure 7E).
Importantly, MUC12 could promote the recruitment of c-Jun to the promoter region of TGF-β1 ( Figure 7F). Similarly, MUC12 could enhance the promoter activity of WT TGF-β1 but lost this ability when the WT was replaced by MUT ( Figure 7G). All these results demonstrate that MUC12 relies on the transactivation of c-Jun to activate TGF-β1, leading to RCC progression.

| D ISCUSS I ON
RCC remains the most common kidney cancer. 3 Although current therapeutic treatments benefit RCC patients a lot, the survival rate of metastatic RCC is still very low. Thus, there is an urgent need of developing novel therapies for better treatment of RCC. In this study, we found that MUC12 was overexpressed in RCC patients compared to normal kidney tissues and its levels were gradually increased as this type of cancer progressed to later stage according to TCGA data set. In vitro experimental results also confirmed that MUC12 is membrane glycoprotein and its biological contributions to cells may rely on signal transduction cassette. Here, our data exhibited that MUC12 bore the ability to increase c-Jun protein levels, which in turn transcriptionally regulated TGF-β1.
Given the fact that c-Jun is a transcription factor working in the nucleus, the process by which MUC12 regulates c-Jun has a spatial distance. We hypothesize that MUC12 may serve as scaffold protein to activate certain kinase cassette, which phosphorylates and activates c-Jun. Previous reports have demonstrated that the phosphorylation of c-Jun in its N-terminal by ERK or JNK could increase its stability and enhance its DNA binding ability. 22,23 Interestingly, both ERK and JNK have change to be activated by F I G U R E 5 Inhibition of TGF-β1 attenuated MUC12-mediated RCC cell growth and cell invasion. A, GSEA analysis of TCGA data set revealed that TGF-β1 signalling was activated in MUC12 higher group. B, A positive correlation between MUC12 and TGF-β1 was observed in human RCC samples based on TCGA dataset. P < 0.001, r = 0.17. C, Knockdown of MUC12 decreased TGF-β1 levels as well as p-Smad3 in 786-O cells while overexpression of MUC12 in A498 cells boosted TGF-β1 levels as well as p-Smad3. GAPDH served as internal control. D, Inhibition of TGF-β1 signalling by its specific inhibitor could reverse MUC12-mediated induction of TGF-β1 levels and p-Smad3 in A498 cells. GAPDH was loading control. E,G, Inhibition of TGF-β1 signalling by its specific inhibitor blocked MUC12-mediated RCC cell growth. E, representative images of colonies; G, a statistical analysis of E. F,H, Inhibition of TGF-β1 signalling by its specific inhibitor attenuated MUC12-mediated RCC cell invasion. F, representative images of invaded cells; H, a statistical analysis of G. *P < 0.05; **P < 0.01; ***P < 0.001 TGF-β1 mRNA levels were normalized to GAPDH mRNA levels. *P < 0.05; **P < 0.01 F I G U R E 7 Activation of TGF-β1 by MUC12 relied on the transactivation of c-Jun. A,B, There are four potential c-Jun binding sites in the promoter region of TGF-β1 according to online prediction. C, ChIP assay showed that c-Jun could directly bind to c-JUNE2 but not c-JUNE1. D, Cartoon was drawn to compare the wild-type and mutant form of TGF-β1's promoter. E, Knockdown of c-Jun suppressed the activity of WT TGF-β1's promoter but failed to affect MUT TGF-β1's promoter. F, MUC12 promoted the recruitment of c-Jun to TGF-β1's promoter. G, MUC12 increased the activity of WT TGF-β1's promoter but failed to affect MUT TGF-β1's promoter. *P < 0.05; **P < 0.01 membrane proteins such as G-protein receptor, which suggests that MUC12 also has chance to activate them. Nevertheless, the detailed mechanisms of how MUC12 activates c-Jun deserve future intensive investigations.
As a membrane protein, MUC12 has a promising therapeutic potential. Recently, chimeric antigen receptor (CAR) T cell therapy has been merged as one alternative therapy for multiple cancers. 24 CAR-T cells with gene modification can express chimeric receptor and specifically bind to tumour antigen, leading to specific cytotoxicity of T cells towards cancer cells. However, CAR-T cell therapy in advanced RCC is still on its infancy. The property of MUC12 in metastatic RCC makes it a promising target for CAR-T cells. In fact, MUC family member has been viewed as ideal target for CAR-T cells in various cancers: CAR-T cell therapy specific for MUC1 has been clinically tested in HCC, breast and glioma 25 ; CAR-T cells specific for MUC16 have also being investigated in ovarian cancer. 26 Thus, it is also reasonable to construct specific CAR-T cells to kill MUC12positive advanced RCC.
TGF-β signalling is widely known to promote cancer development. In one hand, TGF-β signal can promote the epithelial to mesenchymal transition (EMT) of cancerous cells, which bestows cancer cells metastatic potential and allows them to move to distant organs. 20 In another hand, TGF-β signalling also can educate tumour microenvironment and make it suitable for cancer survival. 27 For instance, the transformation of fibroblasts into myofibroblasts (socalled cancer-associated fibroblasts or CAFs) is accelerated in the presence of TGF-β. 27 These CAFs play critical role in tumour survival and cancer metastasis. Due to the importance of TGF-β in cancer progression, now several inhibitors are being tested in clinical trials including metelimumab, fresolimumab and AVID200. 28 Our study also pointed out that TGF-β1 was involved in MUC12-mediated cell invasion, strengthening the role of TGF-β1 signalling in the development of advanced RCC.
In summary, our data identify MUC12 as a tumour promoting factor in advanced RCC by activating TGF-β1 signalling and provide compelling rational to develop MUC12-targeted therapies for RCC patients.

CO N FLI C T O F I NTE R E S T
This study has no potential conflict of interest to declare.

DATA AVA I L A B I L I T Y S TAT E M E N T
The data sets used and/or analysed during the current study are available from the corresponding author on reasonable request.