WISP1 and miR-29c-3p are novel prognostic biomarkers and therapeutic targets for bladder cancer

Background: Aberrant expression of WISP1 is associated with carcinogenesis; however, the expression and prognostic values of WISP1 in bladder cancer (BC) remain elusive. Therefore, the present study aimed to investigate the WISP1 expression in BC and explore the possible mechanisms and clinical value of WISP1 in BC. Methods: The in silico analysis based on oncomine and Kaplan-Meier Plotter databases were to reveal WISP1 expression and prognosis in BC. Besides, qRT-PCR, immunohistochemistry (IHC) and western blot assays were used to detect WISP1 expression both in BC tissues and BC cell lines. TargetScan database was used to predict miRNAs that putatively regulate the expression of WISP1. COX regression and the Kaplan-Meier method were applied to evaluate the prognostic value of WISP1 and miRNAs in BC. Results: WISP1 was up-regulated in BC and associated with poor survival in patients with BC through in silico analysis. Besides, WISP1 expression was identi�ed to be up-regulated both in BC tissues and cell lines. We identi�ed 13 miRNAs that putatively regulate the expression of WISP1. Of these miRNAs, only miR-29c-3p was found to be signi�cantly negatively correlated with WISP1 in BC tissues. The correlation of in situ expressions of WISP1 and miR-29c-3p by immunohistochemistry (IHC) and clinical characteristics revealed that WISP1 was signi�cantly associated with tumor size and hsa-miR-23b-3p expression, and miR-29c-3p was associated with tumor size, M stage, and WISP1 expression. Multivariate Cox regression analysis indicated that TNM stage and WISP1 expression were predictors of unfavorable prognosis, while hsa-miR-29c-3p was a predictor of favorable prognosis in patients with BC. Conclusions: Collectively, the �ndings indicated that WISP1 and miR-29c-3p might serve as novel prognostic biomarkers and potential therapeutic targets for BC.


Introduction
Globally, bladder cancer(BC) ranks the tenth most frequently diagnosed urological malignancy, with approximately 550,000 new cases (nearly 425,000 in males and 125,000 in females) diagnosed in 2018 worldwide.It is among the leading causes of cancer-associated mortality, with approximately 200,000 deaths reported in 2018.Histologically, urothelial carcinoma (transitional cell) represents the predominant type of BC, accounting for 90 percent of all BC 1 .Clinically, two main phenotypes, including the muscleinvasive and non-muscle-invasive, with different pathogenesis, molecular characteristics, and clinical outcomes, have been identi ed 3 .Recent advances in high-throughput sequencing technologies have facilitated the rapid molecular characterization and enhanced our understanding of the pathogenesis of BC, leading to the identi cation of actionable therapeutic targets for BC 4 .However, the molecular mechanisms underlying the occurrence and development of bladder carcinogenesis remain to be completely elucidated.
WNT1-inducible signaling pathway protein 1 (WISP1), a secreted matricellular protein, is a member of the connective tissue growth factor/CCN protein family, which is found in the extracellular matrix (ECM).
WISP1 is predominantly involved in various biological processes, including cell adhesion, proliferation, differentiation, survival, and carcinogenesis.The elevated expression of WISP1 has been detected in several cancers, including melanoma, glioblastoma, and hepatocellular carcinoma [5][6][7] .WISP1 expression has also been associated with tumor purity, an in amed tumor microenvironment, advanced disease, EMT, and macrophage M2 polarization in multiple solid tumors [8][9][10][11] .However, its expression and physiological function in BC remain elusive.MicroRNAs (miRNAs) are small 18-24 nucleotide long, single-stranded non-coding RNAs involved in post-transcriptional gene regulation by pairing to the mRNAs of its target protein-coding gene), thereby causing translational repression or mRNA degradation of the target gene 12 .Recently, several miRNAs, such as miRNA-217 and miRNA-616 13,14 , have been identi ed as prognostic biomarkers in BC 15,16 .However, there is a paucity of studies on WISP1-related miRNAs in BC.Therefore, in the present study, we investigated the expression levels of WISP1 in BC using quantitative real-time PCR (qRT-PCR), Western blot assay, and bioinformatics analyses.
Furthermore, we used TargetScan to predict the potential WISP1-related miRNAs.Univariate and multivariate COX regression and the Kaplan-Meier method were applied to evaluate the prognostic value of WISP1 and miRNAs in BC.

Bioinformatics analysis
Oncomine (www.oncomine.org),an online microarray database, was used to analyze the mRNA levels of WISP1 in BC and normal bladder tissue samples 17,18 .Kaplan-Meier Plotter (kmplot.com/analysis/) was used to analyze the prognostic value of WISP1 in BC 19 .TargetScan (7.2 version, http://www.targetscan.org/)was used to identify the potential miRNAs targetingWISP1 20 .

BLAD Patient Samples
One hundred and thirty-two BC tissues and twenty paired BC tissues and adjacent normal tissues were obtained from patients who underwent surgical treatment for either diagnostic biopsy or surgical resection for BC at Wuxi People's Hospital A liated to Nanjing Medical University.The study protocols were approved by the Ethics Committee of the Institutional Review Board of Wuxi People's Hospital, Nanjing Medical University.Written informed consent was obtained from each patient.All protocols involving human patients were performed in accordance with the relevant guidelines and regulations of Wuxi People's Hospital, Nanjing Medical University.The tissue samples were immediately stored at −80℃ until use.The histopathological examination was performed to con rm the diagnosis.

Quantitative reverse transcription PCR (RT-qPCR)
Total RNA was extracted from fresh frozen tissue samples using Trizol reagent (Invitrogen, Carlsbad, CA, United States) according to the manufacturer's instructions.cDNA was synthesized using a reverse transcription kit (PrimeScript RT-PCR kit; Takara, Japan) following the manufacturer's protocol.Total RNA was isolated from the cells using the PureLink™ RNA Mini Kit (Life Technologies, Carlsbad, CA, USA) according to the manufacturer's instructions.Reverse transcription reactions were performed using iScript cDNA synthesis kit (Bio-Rad) for mRNA, or TaqMan miR Reverse Transcription kit (Applied Biosystems) for mature miR as described earlier.qRT-PCR was performed using SYBR Green Master Mix (Takara, Japan)on an ABI 7500 RealTime PCR System (Applied Biosystems, United States).β-actin and U6 were used as the reference genes.Primers used in the study were listed in Table1.All reactions were performed in triplicate.The relative expression levels of miRNA and mRNA were evaluated using the 2 -∆∆Ct method.

Table1 Primers of genes and miRNAs
Names Forward Western Blot assay Total cell proteins were extracted from BC, and paracancerous tissues with RIPA lysis buffer (Pierce, Thermo Scienti c, Cramlington, United Kingdom) supplemented with protease inhibitor cocktail.Protein concentrations were determined by BCA protein concentration reagent kit (Beyotime, China).Proteins were separated using sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDA-PAGE) and transferred to PVDF membrane (Bio-Rad, USA).The membranes were blocked with 5% non-fat dry milk for 1 h at room temperature (RT).Subsequently, the membranes were incubated with the following primary antibodies: anti-WISP1 (Abcam, ab178547) and anti-β-actin (Abcam, ab8226) at 4°C for overnight.Then, membranes were washed and incubated with goat anti-rabbit IgG secondary antibody conjugated with horseradish peroxidase (1:1000).The target bands were visualized using an electrochemiluminescence (ECL) detection system.β-actin was used as an endogenous control.

Immunohistochemistry (IHC)
The immunohistochemical assay was performed on cancer tissues and adjacent normal tissues from BC patients.Tissue specimens were xed with 4% paraformaldehyde, dehydrated, and embedded in para n.
The tissue section was cut into a thickness of 5 μm and then mounted on a glass slide.IHC staining was performed according to the manufacturer's protocol.In brief, slides were depara nized and then rehydrated in successively graded ethanol.Antigen retrieval was performed by heating the sections at 100°C for 30 minutes in the microwave.Subsequently, the slides were incubated with the primary antibody against WISP1 (diluted 1:200, ab178547, Abcam, USA) at 4°C overnight.The sections were washed 3 times in TBST for 5 minutes each and then incubated with the rabbit secondary antibody for 1 hour at room temperature.The sections were then stained with diaminobenzidine (DAB) and counterstained with hematoxylin.Three pathologists independently performed blinded analysis of the WISP1 immunostaining intensity under a light microscope.By analyzing the Based on the percentage of positively staining cells, the sections were grade as 0 = 5% or none of the cells were stained, 1 = 6-25% of the cells stained positive, 2 = 26-50% of the cells stained positive; 3 = 51-75% of the cells stained positive, and 4 = more than 76% of the cells stained positive.Similarly, the staining intensity was graded as 0 = no staining, 1 = weak staining, 2 = moderate staining, and 3 = strong staining.The two scores were multiplied, and the immune-reactive score (values from 0-12) for each case was determined 21 .

Statistical Analysis
All experiments were performed in triplicate.Data from three or more independent experiments were presented as the mean ± standard deviation (SD).A paired Student's t-test was used to analyze the nal score of cancer tissues and adjacent normal tissues.A Chi-square test was performed to assess the relationship between WISP1 and miRNAs expression and clinicopathological characteristics.The Kaplan-Meier method was used to calculate survival functions, and differences were compared using the logrank test.Univariate and multivariate Cox proportional hazards analyses were used to identify the independent prognostic factors for overall survival in BC.Statistical analyses were performed with the R version 3.6.1 (R Core Team, 2019) with RStudio and GraphPad Prism 8.0 (GraphPad Software, San Diego, CA, United States).A P-value of < 0.05 (two-tailed) was considered statistically signi cant.

WISP1 was overexpressed in BC
From the oncomine database, as illustrated in Figure 1a, we determined that WISP1 was up-regulated in BC.From the Kaplan-Meier plotter database, as presented in Figure 1b, the high expression of WISP1 was signi cantly associated with poor overall survival.These results indicated that the aberrant expression of WISP1 in BC was associated with unfavorable survival.qRT-PCR and Western blot analyses of cancer tissues and adjacent normal tissues from BC patients revealed that WISP1 was noticeably up-regulated in BC tissues, as shown in Figure 1c, d/e.Consistently, a signi cant over-expression of WISP1 was observed in BC cells (UMUC3, RT4, T24, UC9, and 5637) compared to normal cell lines (SV-HUC-1) as represented in Figure 1f/g.Collectively, these results suggested that WISP1 was signi cantly up-regulated in BC.

WISP1 was associated with tumor size and hsa-miR-23b-3p expression
To further verify WISP1 expression in clinical specimens of BC, IHC and qRT-PCR was performed in 132 BC tissue specimens.The protein expression of WISP1 was localized both in the nucleus and cytoplasm of cells (Figure3b WISP1 and hsa-miR-23b-3p were independent prognostic factors for BC Using Kaplan-Meier survival analysis, BLAD patients high WISP1 exhibited a signi cantly low OS (P = 0.011, Figure 4a); BLAD patients with high hsa-miR-23b-3p expression exhibited a noticeably good OS (P = 0.015, Figure 4b).The results of the univariate analysis indicated that age (>60 vs. <=60), M stage (M1 vs. M0), TNM stage (III-IV vs. I-II), hsa-miR-29c-3p (High vs. Low), and WISP1 (High vs. Low) were critical factors affecting the overall survival in patients with BC (P < 0.05, Table 2).
The results of multivariate Cox regression analysis showed that TNM stage (III-IV vs. I-II) and WISP1 (High vs. Low) were independent predictors of unfavorable prognosis, while hsa-miR-29c-3p (High vs. Low) represented a predictor of favorable prognosis in patients with BC (P < 0.05, Table 3).Together, these results revealed that WISP1 and hsa-miR-23b-3p were independent prognostic factors of BC.

Discussion
WISP1, a cysteine-rich protein, belongs to the family of matricellular proteins involved in developmental functions and carcinogenesis.Furthermore, WISP1 polymorphisms have been recognized as a biomarker or a therapeutic target in urothelial cell carcinoma 22 .From the oncomine database and Kaplan-Meier plotter database, we found that WISP1 was up-regulated in BC and associated with signi cantly poor overall survival, suggesting a potential role of WISP1 in BC.The expression of WISP1, both at mRNA and protein levels, was evaluated in BC cell lines and tissues, and the results indicated that WISP1 was indeed overexpressed in BC. miRNAs, critical regulators of gene expression, are often dysregulated in cancer. 23- 26.Using TargetScan, we predicted 13 miRNAs, which potentially target the WISP1.However, correlation analysis revealed that only hsa-miR-23b-3p was signi cantly negatively correlated with WISP1 (P=0.006,r=-0.633) in BC.Notably, miR-23b-3p has also been reported as a robust normalizer for urine microRNA studies in BC 27 .Accumulating studies have indicated that miR-23b-3p acts as a tumor suppressor in multiple cancer, including laryngeal squamous cell carcinoma 28 , ovarian cancer 29 , esophageal Carcinoma 30 , and BC 31 ; however, in abeta-treated neuroblastoma cells, miR-23b-3p functions as an oncogenic factor 32 .
Furthermore, tumor size and hsa-miR-23b-3p expression were signi cantly associated with WISP1, and tumor size, M stage and WISP1 were signi cantly correlated wth hsa-miR-23b-3p expression.All these ndings indicated that WISP1 and hsa-miR-23b-3p might play crucial roles in BC.The results of univariate Cox analysis revealed that age, M stage, TNM stage, hsa-miR-29c-3p, and WISP1 were critical factors affecting the survival time in patients with BC.The multivariate Cox survival analysis revealed that TNM stage and WISP1were predictors of unfavorable prognosis, while hsa-miR-29c-3p represented a factor of favorable prognosis in patients with BC.Furthermore, BC patients with high WISP1 expression exhibited a signi cantly lower OS, while patients with high hsa-miR-23b-3p expression exhibited an evidently higher OS.Taken together, these results suggested that WISP1 and hsa-miR-23b-3p were independent prognostic factors of BC.
Several potential limitations of the present study should be noted.Firstly, although validated in databases, our bioinformatics analyses with the current databases may have certain limitations and aws.However, the evidence supporting the involvement of WISP1 and hsa-miR-23b-3p in BC remains very limited.Thus, further studies are warranted to validate with mechanistic evidence the designation of WISP1 as a potential target of hsa-miR-23b-3p for BC.Secondly, the functional mechanism underlying WISP1 overexpression in BC was not investigated; therefore, further studies are required to identify the mechanisms of WISP1 overexpression in bladder carcinogenesis.

Conclusions
In conclusion, the present study demostrated that WISP1 is upregulated and associated with poor prognosis in BC.The miR-29c-3p negatively relativated with WISP1 is an independent prognostic factor in BC.The results of the present study indicated that WISP1 and miR-29c-3p may serve as novel prognostic biomarkers and potential therapeutic targets for BC.

Figure 2 The
Figure 2

Figure 3 The
Figure 3

Table 3 .
Univariate and multivariate COX regression analyses of the prognostic factors in bladder cancer