Aberrant Methylation of the 16q22.1 Tumor Suppressor Gene CDH11 Promotes Tumorigenesis and Progression of Renal Cell Carcinoma through NF-kB Pathway

Background: To discover new epigenetic biomarkers for early cancer detection. The CDH11 gene has been reported as a critical tumor suppressor gene (TSG) for multiple tumors, although it has not yet been reported in renal cell carcinoma (RCC). We explored its epigenetic alteration in RCC and analyzed the possible biological function and mechanism in tumorigenesis and progression of RCC. Materials and Methods: We examined CDH11 gene expression and methylation using semiquantitative reverse transcriptase PCR (RT-PCR) and methylation-specic polymerase chain reaction (MSP) in RCC cell lines before and after treatment with 5-aza-2’-deoxycytidine (5-Aza). MSP was further applied to 93 RCC primary tumors, and the relationship between CDH11 gene methylation and clinicopathological features was discussed. A selection of the cell lines and specimens was subsequently examined using bisulte genomic sequencing (BGS) and real-time PCR. Meanwhile, assays of cell viability, colony formation, migration and invasion, wound healing, and western blot were performed to conrm the tumor-suppressive function and mechanism of CDH11 gene. Results: CDH11 gene methylation was detected in 4 of 5 RCC cell lines with silenced expressions. Treatment with 5-Aza reversed methylation and restored CDH11 gene expression. Aberrant methylation was further detected in 41 of the 93 (44.1%) primary tumors. Furthermore, CDH11 gene methylation was signicantly associated with tumor stage and nuclear grade in patients with RCC (p<0.05). Due to the phenomenon of aberrant methylation, ectopic low-level expression of CDH11 gene could result in promotion of tumorigenesis and progression in RCC cell lines, which might be mediated through NF-kB pathway. Conclusions:


Background
RCC is a malignant tumor with a relatively poor prognosis and high rates of metastasis to distal organs, accounting for 2% of all adult tumors [1] . The detailed molecular mechanisms that underlie RCC development remain poorly understood. However, an increasing number of studies have determined that the inactivation of TSG is frequently involved in the tumorigenesis of RCC as a result of epigenetic abnormalities in DNA methylation [1][2][3][4][5][6] .
Currently, many TSGs in a wide range of cancers have been found to be inactivated by the methylation of promoter regions of DNA. This methylation frequently occurs at 5'-CpG islands, which are regions with a high density of CG dinucleotides found in approximately 70% of the coding genes in mammals [7] . It has been proposed that the aberrant promoter methylation status of speci c TSGs could be a potentially sensitive marker for use in RCC diagnosis and prognosis prediction. However, the rate of aberrant promoter methylation of most TSGs is relatively lower in RCC than in other solid tumors, correspondingly to only approximately 10% to 30%. Therefore, identifying a critical TSG with a higher methylation rate in RCC is critical. Our team has always been devoted to this research and a series of highly aberrant methylation TSGs of RCC had been explored [8][9][10][11][12] .
The CDH11 gene, located in the 16q22.1 region and belonging to the E-cadherin family, is often involved in an important group of cell-cell adhesion molecules that mediate intercellular adhesion by Ca 2+dependent hemophilic interactions [13] . Because cadherins have been widely documented in cancer development [14] , the promoter CpG island hypermethylation-associated silencing of the CDH11 gene, a component of this superfamily, is consistent with the hypothesis that epigenetic inactivation of this gene could be involved in the tumorigenesis and progression of carcinomas. L Li et al [15] found that the CDH11 gene is frequently methylated in a variety of tumor tissues including esophageal squamous cell carcinoma (13/14, 93%), nasopharyngeal carcinoma (17/18, 94%), hepatocellular carcinoma (28/42, 66%), breast carcinoma (11/12, 91%), gastric carcinoma (13/13, 100%), colon carcinoma (10/11, 90%) and other carcinomas.
To our knowledge, we are the rst throughout the world to explore the CDH11 gene methylation status in RCC cell lines and primary tumors. Besides, its roles in RCC suppression remain unclear up to present. Based on the current situation above, this investigation will aid in elucidating whether the CDH11 gene is a potential novel biomarker and therapeutic target for early detection and prognosis of RCC.

Patient and tissue samples
All human primary RCCs (93 cases) and adjacent normal tissues (20 cases) were obtained from the Urology Department, Peking University First Hospital, Beijing, from September 2012 to May 2013. The specimens were collected yia radical nephrectomy, and the RCC diagnoses were con rmed according to the pathological ndings. The adjacent normal tissues were resected at least 2cm from the tumors.
Additionally, 3 similarly aged normal renal parenchyma specimens were collected during non-cancerrelated kidney surgeries that were performed during the same time period for use as normal controls. All of the resected tissues were snap frozen in liquid nitrogen and stored at -80℃. Patients with localized lymph node or distant metastases detected by a preoperative computed tomography scan were excluded from this study. The participants gave informed written consent for participation in this research study and all of the research procedures were approved by the Institutional Review Board of Peking University First Hospital. The study was also strictly conducted according to the principles de ned in the Declaration of Helsinki.
The tumor set was composed of samples taken from 64 males and 29 females ranging from 28-78 years of age, with a median age of 58.3 years at diagnosis. Of the samples, 54 primary tumors were located on the left side, and 39 on the right. The average preoperative diameter of the primary tumors was 5.5cm (range 2.1-12.5cm). The classi cation of the tumors was based on the staging system of the 2018 American Joint Committee on Cancer (AJCC). Additionally, the nuclear grade of the tumors was determined during postoperative pathological analysis.

Cell line preparation and treatment of RCC cell lines with 5-Aza
Five RCC cell lines (A498, Caki-2, Ketr-3, Osr, and 786-O) obtained from the American Type Culture Collection (ATCC, Manassas, VA, USA) were prepared to validate the CDH11 gene methylation status. The HEK293 human normal embryonic kidney cell line and HK-2 human kidney proximal tubular epithelial cell line were both routinely cultured and served as "normal" controls. All cell lines were maintained in DMEM supplemented with 2mM glutamine and 10% fetal bovine serum at 37℃ with 5% CO 2. The demethylating agent 5-Aza (Sigma®) was freshly prepared in ddH 2 O and lter sterilized. Afterwards, the RCC cell lines were treated with 10μM 5-Aza for 3 days. The media was changed every 24hrs. After treatment, the cells were pelleted and washed with PBS, and the DNA/RNA was extracted.
2.3 DNA extraction from the RCC tissue samples DNA was extracted from the RCC primary tumors and adjacent normal tissues. Genomic DNA was isolated from the tissues using the TIANamp Genomic DNA Kit (TIANGEN®) protocol. The quality of the isolated DNA was subsequently assessed by electrophoresis.

Bisul te modi cation of specimen DNA and MSP
Bisul te modi cation of the specimen DNA was carried out as described previously [16] . To examine the presence of epigenetic alterations in the CDH11 gene of the RCC primary tumors, bisul te-modi ed DNA (50μg) was used in MSP with speci cally methylated and unmethylated primers. All of the primers were provided by Sangon Biotech® Co. Ltd. (Shanghai) and were previously shown to not amplify nonbisul ted DNA. The 12.5μl MSP mixture contained 0.5μl of bisulfate-treated genomic DNA, 1μl of dNTPs (2.5mM), 0.75μl of 5'-primers (10μM), 0.75μl of 3'-primers (10μM), 2μl of 5xFlexi buffer, 1μl of MgCl 2 (25mM), 0.1μl of AmpliTaq Gold® (5U/μl ) and 6.4μl of ddH 2 O. The MSP products were checked on a 2% agarose gel stained with ethidium bromide.

Immunohistochemistry (IHC) assays
IHC assays were performed with an UltraSensitive SP Kit (Maixin-Bio, Fujian, China) following the manufacturer's instructions. Tissue sections were dewaxed, rehydrated, processed for antigen retrieval, and blocked. The sections were incubated with a primary antibody against CDH11 overnight at 4°C. Following treatment with secondary antibodies, 3,3′-diaminobenzidine, and hematoxylin staining, the staining was evaluated with Image-Pro Plus, version 6.0 (Image Pro, Silver Spring, MD).

Quantitative real-time PCR
Total RNA from 10 pairs of patient tumors and adjacent normal tissue samples was isolated using the TRIZOL reagent (Invitrogen), and 2μg of total RNA was reverse-transcribed using a reverse transcription system. Real-time PCR was performed using an ABI Prism 7500 TM instrument (Applied Biosystems) system with SYBR green PCR mix purchased from Sangon Biotech® Co. Ltd. (Shanghai). GAPDH was used as an internal control. The relative expression levels of the CDH11 gene were calculated according to the 2 -△△CT method [17] .

BGS of the CDH11 gene in the cell lines and specimens
BGS was then performed to analyze the detailed methylation status of the CDH11 gene in a selection of the cell lines and specimens, and this method was performed according to the previous report [18] . The ampli ed BGS products were TA cloned, and 6 colonies were randomly chosen and sequenced.

Cell viability assay
Cells were replated into 96-well plates after being transfected with CDH11 expression vector and empty vector for 48h, and further measured using the Cell Counting Kit-8 (CCK-8, Beyotime, Shanghai, China) as described previously [19][20] .

Colony formation assay
Monolayer culture was used for colony formation assay. Cells were selected for ~14 days with presence of G418 after transfection for 48h. Surviving colonies (≥50 cells/colony) were counted and stained with gentian violet. All experiments were performed in triplicate wells with three times. The detailed procedures could be traced by HY Xu et al [21] .

Migration and invasion assays
Cell motility and invasive abilities were assessed by Transwell® and Matrigel™ invasion assays (Corning Life Sciences, Bedford, MA) as described previously [22][23] . Cells were migrated and invaded to the lower side of themembrane, and further stained with 0.1% crystal violet. Cells were then counted in ve microscopic elds, and the mean values were counted.

Wound healing assay
Wound healing assay was performed as described previously [19][20] . A mechanical wound was created after scratching with a pipette tip, and images were taken at different time points. The distance between the wound edges was measured and quanti ed.

Statistical analysis
Clinicopathological data of the patients were obtained from the institutional database of Peking University First Hospital. The CDH11 gene methylation status and the clinicopathological features of the RCC patients were analyzed. Continuous data were shown as means ± standard deviation (SD). All of the statistical analyses were performed using Student' s t test, Fisher's exact test or the chi-square test with the SPSS17.0 software (StatSoft Inc., Tulsa, OK, USA). A p-value <0.05 was considered statistically signi cant. Figure 1A listed the primer sequences and cycling parameters for the RT-PCR and MSP. CDH11 gene expression was silenced in all of the 5 RCC cell lines as shown in Figure 1B. As expected, it was aberrantly methylated in 4 of the 5 RCC cell lines (80.0%), including 786-O, A498, Osr and Ketr-3.

Down-regulatedCDH11 gene expression is aberrantly methylated in RCC cell lines
After 5-Aza treatment, CDH11 gene expression levels were restored in 4 of the 5 (80%) RCC cell lines. Meanwhile, MSP also showed that the CDH11 gene was partially or totally demethylated following pharmacological demethylation. These changes con rmed that CDH11 gene expression silencing was directly mediated by aberrant promoter methylation.
3.2 Down-regulatedCDH11 geneexpression is aberrantly methylated in RCC primary tumors, which is signi cantly associated with tumor stage and nuclear grade We subsequently used this validated system to analyze the CDH11 gene methylation status in a series of 93 RCC samples and 20 adjacent normal tissues. Aberrant promoter methylation was detected in 41 of the 93 (44.1%) RCC primary tumors and in 3 of the 20 (15.0%) adjacent normal tissues (p<0.05), and representative results were shown in Figure 2A.
To con rm that the methylation of the CDH11 gene indeed correlated with the down-regulation of its gene expression, we performed real-time PCR to detect the mRNA expression levels of the CDH11 gene in 10 primary tumors paired with their adjacent normal tissues (10T/N, 11T/N, … , 18T/N, 19T/N), as demonstrated in Figure 2B. As expected, in 8 out of the 10 (80%) RCC primary tumors (10T, 11T, 12T, 14T, 15T, 16T, 18T, and 19T), the expression levels of the CDH11 gene were signi cantly lower than that of the adjacent normal tissues, which was consistent with our MSP results. Because both 13T and 13N were methylated, there was no signi cant difference in the expression level of 13T/N. However, MSP-negative results were identi ed for the 11T/N, 16T/N and 18T/N tumors, which also had signi cantly different expression from their normal tissue pairs. By the IHC assays, the discrepancy of the protein expression levels in the CDH11 gene could be also directly re ected in Figure 2C. Figure 2D listed the patients' clinicopathological features. The CDH11 gene methylation status was not signi cantly associated with gender, age, tumor location or tumor diameter. However, there existed a signi cant association between AJCC pathological stage and nuclear grade in the patients with different CDH11 gene methylation status (p<0.05). The percentage of methylated RCC tumors increased dramatically with more advanced stages or grades.

3.3
The methylation status of the CDH11 gene is con rmed by BGS analysis BGS was performed to con rm the methylation status of the CDH11 gene in HEK293 and A498/Ketr-3 cell lines (before and after 5-Aza treatment) and three methylated tumors along with their adjacent normal tissues. The BGS primers were listed in Figure 1A. The results were consistent with the MSP results, as a high density of methylated CpG sites were detected in the A498/Ketr-3 cell lines and in the methylated tumor samples (Figure 3A-3B). The BGS results for the HEK293 and A498/Ketr-3 cell lines after 5-Aza treatment and for the matched adjacent normal tissues ( Figure 3A-3B) were consistent with CDH11 gene down-regulation, which was observed in RCC cell lines and primary tumors.

Aberrant methylation of the CDH11 gene can promote tumorigenesis and progression ofRCC, which might be mediated through NF-kB pathway
We further assessed tumor-suppressive functions of CDH11 gene. Initially, real-time PCR con rmed CDH11 gene overexpression in CDH11-transfected RCC cell lines (Figure4A). Colony formation and CCK8 assays showed that overexpression of CDH11 gene had signi cantly detectable effect on the proliferation of RCC cell lines (p<0.05) ( Figure 4B-4C). Scratch wound healing assays also revealed that CDH11transfected cells showed slower closure of the scratched wound in RCC cell lines (p<0.05) ( Figure 4D). In addition, we measured cell migration and invasion using Transwell® assays (with or without Matrigel) in CDH11-transfected and the respective control cells. Results showed that CDH11 overexpression signi cantly decreased cell migration and invasion in RCC cell lines (p<0.05) ( Figure 4E).
To further explore molecular mechanisms responsible for CDH11-mediated tumor-suppressive functions, we analyzed the effects on NF-kB signaling pathways. As expected, CDH11-transfected A498 and 786-O cell lines both displayed lower levels of pp65 and MMP2 than the control cells ( Figure 4F).
These ndings suggested that CDH11 may act as an antagonist of NF-kB signaling pathway in RCC.

Discussion
RCC is often distinguished by a set of genetic and epigenetic abnormalities. It is well known that inactivation of TSGs frequently occurs in RCC via DNA methylation and deletions, but only a few genes with lower rates of methylation in RCC have been reported, which may not be biologically relevant. Thereafter, the identi cation of genes with higher rates of methylation in RCC is critical. In this study, we validated the methylation status of a critical 16q22.1 TSG, the CDH11 gene, in RCC cell lines and primary tumors. More importantly, we explored its possible biological function and mechanism in tumorigenesis and progression of RCC. Similar to other classical cadherin family members (CDH1, CDH3, CDH5, CDH8 and CDH13), the CDH11 gene protein can cluster through a zipper-like mechanism while their intracellular domain is anchored to the actin cytoskeleton through α/β-catenin [24] . These interactions have crucial roles in maintaining tissue architecture and cell polarity and in limiting cell movement and proliferation, thus resulting in tumor inhibition [25] . Recent studies have shown that the CDH11 gene meets several of the criteria of classic TSGs: (1) it shares the same locus of 16q22 as CDH1 and CDH13, which has been identi ed as involved in the inhibition of cell proliferation and invasiveness and the promotion of apoptosis; and (2) it is capable of inhibiting the growth of various types of tumor cells and colony formation, followed by frequent deletion and frequent methylation in multiple cancers.
The CDH11 gene was rst identi ed as a TSG candidate in a murine retinoblastoma model by Marchong et al in 2010 [26] . Our investigation is the rst study to report the prevalence of CDH11 gene methylation in a large set of RCC cell lines and tumor samples, although in the year of 2019 Chunsheng Li et al [27] has suggested a hypothesis that CDH11 gene may be a potential core gene in metastatic RCC using bioinformatics analysis without objective experimental veri cation. Using MSP, we discovered that the CDH11 gene is aberrantly methylated both in RCC cell lines and primary tumors, indicating that the CDH11 gene, a novel functional TSG, is frequently epigenetically inactivated in RCC. MSP revealed that the CDH11 gene promoter is methylated in 80% of RCC cell lines and in 44.1% of RCC primary tumors. The higher methylation rate of RCC cell lines compared to RCC primary tumors indicates that some RCC cell lines may have acquired the CDH11 gene methylation property during the establishment or maintenance process. Certainly, there were unmethylated alleles in the Caki-2 RCC cell line, and no expression was detected, which suggests that other transcriptional regulatory mechanisms, such as histone modi cation or transcriptional repressors, may also contribute to the silencing of this gene.
The AJCC pathological tumor stage and nuclear grade are commonly used for prognostic diagnoses in patients with RCC. Our statistical analysis of clinicopathological features in patients with RCC showed a positive correlation between the CDH11 methylation status and the tumor stage or nuclear grade, which was consistent with the conclusions obtained from gastric cancer and non-muscle invasive bladder cancer [28][29][30][31] . Our data in the subsequently functional trails also showed that ectopic expression of CDH11 gene inhibited the proliferation, migration and invasion in RCC cell lines. Moreover, this study appears to be also the rst to reveal its underlying mechanism. We found that the inhibition of tumor migration and invasion by CDH11 might be associated with deregulation of NF-kB signaling pathway. Above all, these results all suggest that CDH11 gene functions a tumor suppressor through suppressing proliferation, migration and invasion through NF-kB pathway in RCC.
Surprisingly, CDH11 gene methylation was also detected in some adjacent normal tissues. Methylation of the CDH11 gene promoter was not detected in normal renal parenchyma specimens, although our investigation revealed that 15.0% of the adjacent normal tissues were methylated. More importantly, the tumor samples of these 3 adjacent normal tissues were also found to be aberrantly methylated, indicating that aberrant promoter methylation might be caused as part of a premalignant eld effect [32] .
Unfortunately, the limited sample size made it di cult to draw an accurate conclusion concerning the signi cance of CDH11 gene methylation, and the in-depth upstream regulatory mechanisms underlying the tumor-suppressive biological functions still remain fuzzy. These parameters will be improved in our subsequent investigation.

Conclusions
We report for the rst time that the CDH11 gene is often down-regulated by aberrant promoter methylation in RCC cell lines and primary tumors, indicating that it plays a critical role as a TSG in RCC. Additionally, we conclude that CDH11 gene methylation may be involved in the tumorigenesis and progression of RCC through NF-kB pathway, thus which could potentially serve as a novel biomarker and

6.Authors' contributions
Ben Xu and Haifeng Song carried out the design of this research, analysis and interpretation of data, and drafted the manuscript. Cheng Luo and Lei Liang participated in the collection of data and data analysis. Qian Zhang assisted in the design of this research and project development. Ben Xu conceived the study, reviewed all of the statistical analysis of the data, and revised the manuscript. All authors read and approved the nal manuscript. 7