Knockdown by MSH2 and EPCAM siRNA suppress Wnt/β-Catenin Pathway in HCT116 Cell Line

Purpose: Small interfering RNA (siRNA) has the potential as a therapeutic approach against selective pathways in colorectal cancer. EPCAM, a transmembrane glycoprotein mediating cell adhesion, was known to be involved in suppressing Wnt/β-catenin pathway, an important pathway for tumour progression in colon cancer cells. EPCAM deletions caused a transcriptional read-through that may silence its neighbouring gene, MSH2. This study aimed to investigate the effect of co-siRNA targeted genes, MSH2 and EPCAM, in colon cancer cell line, HCT116, and their effect in regulating the Wnt/β-catenin pathway. Methods: Pre-designed siRNA of MSH2 and EPCAM were transfected into HCT116 cells. The cells were divided into six group of treatments: untreated cell group, cells treated with negative control siRNA, MSH2-siRNA treated cells, EPCAM-siRNA treated cells, cells treated with both EPCAM and MSH2-siRNA, and cells treated with transfection reagent (mock control). The mRNA and protein expression following the individual and combined siRNA treatments were assessed by quantitative polymerase chain reaction and Western blot. Results: The mRNA and protein expression levels of MSH2, EPCAM and β-catenin were reduced in the individual MSH2 and EPCAM-siRNA treated samples as compared to the untreated sample. Further reduction of mRNA and protein expressions for MSH2, EPCAM and β-catenin were identied in combined siRNA treatments. Conclusion: Reduction of β-catenin expression by simultaneous silencing of MSH2 and EPCAM suggested that these genes may play a role in supressing the Wnt/β-catenin pathway in cancer cells.


Introduction
Colorectal cancer (CRC) is the third most common cancer worldwide and the fourth most common cause of death [1,2]. The most common hereditary colon cancer is Lynch syndrome (LS) which is caused by germline mutations in any of the four mismatch repair (MMR) genes namely MLH1, MSH2, MSH6 and PMS2 [3] and to date, EPCAM gene was also reported to be associated with LS [4]. EPCAM functions in intracellular signalling, differentiation and proliferation of normal and cancer tissues [4]. This was suggested by the EPCAM signalling that is activated by intra-membrane proteolysis in which the extracellular domain is owed off and the intracellular domain (EpICD) is then released into the cytoplasm. In the cytoplasm, EPCAM formed a nuclear complex with transcriptional regulators β-catenin and Lef, both are parts of Wnt signalling pathway [5]. Germline deletions in the last exon of the non-MMR gene, EPCAM, may silence its neighbouring gene, MSH2, through promoter hypermethylation [6]. Previous study has shown that in EPCAM-silenced colon cancer cells, the expression level of β-catenin was decreased, and the result showed that silencing of EPCAM could inactivate the Wnt/β-catenin pathway in tumour cells [7]. Hence, EPCAM was identi ed as a principal target for the treatment of tumours, through the mechanism of suppressing the Wnt/β-catenin pathway. EPCAM deletions cause a transcriptional read-through and silenced the MSH2 gene [8]. The mechanism of promoter hypermethylation happens when a number of methyl groups bound to the MSH2 promoter region, and consequently decrease the expression of MSH2 protein products in the epithelial cells. Considering the major role of MSH2 protein in DNA repair mechanism, the loss of this protein may subsequently affect the process of DNA repair, and mistakes accumulated upon cell division [4].
Wnt signalling pathway has become one of the major pathway in CRC and LS predisposition with about 65% overactivation in colorectal cancers associated to LS and activating β-catenin mutations were identi ed in approximately 40% of these tumours [9]. Although CRC was commonly triggered by the alterations in the signalling components of the Wnt signalling pathway [10], the relationship between MSH2 and EPCAM genes towards the effect of associated pathway is yet to be discovered. Therefore, this study aimed to investigate the effect of MSH2 and EPCAM siRNA mediated gene knockdown in regulating the Wnt/β-catenin pathway.

Materials And Methods
Culture of HCT116 cell line HCT116 cell line was purchased from RIKEN BioResource Centre, Japan, and was established according to manufacturer's protocol. The cell suspension was centrifuged in complete culture medium (DMEM supplemented with 1.0 g/l of glucose, L-glutamine, sodium pyruvate and 10% FBS) for centrifugation at 1000 rpm for 3 minutes at room temperature. Followed by centrifugation and resuspension, the cell suspension was seeded into two 25 cm 2 culture asks and incubated at 37°C in humidi ed incubators with 5% CO 2 . The growth of cells was maintained daily and fresh culture medium was changed for every two days or when necessary. Sub-culturing was carried out when the cells reached 80% con uence. When the cells reached 80% con uence or 80% of the ask surface area was covered by the cells, sub-culturing was carried to ensure continuous growth of cells during their exponential growth. The rst few cultures were sub-cultured with split ratio of 1:2 to 1:3 depending on the growth of cells. Before sub-culturing, the culture medium was discarded and the adherent cells were washed with 1 ml of PBS. After two to three washing steps, the cells were digested with 0.25% trypsin-EDTA for 5 minutes at 37°C in humidi ed incubators with 5% CO 2 . Upon con rmation of cell detachment, 1 ml of complete culture medium was added and gently dispersed by pipetting over the surface of the cell layer to deactivate the trypsin activity.
The trypsin and cell suspension were mixed thoroughly and transferred into 15 ml centrifuge tube prior to centrifugation at 1200 rpm for 5 minutes at room temperature. Supernatant was then discarded and the cell pellet was re-suspended in 1 ml of complete culture medium. The cell suspension was sub-divided according to appropriate split ratio into the new culture asks containing fresh complete culture medium. The cells were then maintained at 37°C in humidi ed incubators with 5% CO 2 .

Sirna Preparation And Transfection
Hiperfect® Transfection Reagent (Qiagen, Denmark) was used for siRNA transfection of MSH2 and EPCAM genes into the HCT116 cells. SiRNA was prepared in an RNAse-free environment. The target sequence of four pre-designed siRNAs (FlexiTube GeneSolutions, Qiagen, Denmark) for each MSH2, EPCAM and negative control siRNA were listed in Table 1. RNase-free water in a total volume of 100 µl was added to 1 nmol of lyophilized siRNA for MSH2 and EPCAM genes to obtain a nal stock concentration of 10 µM. For the negative control siRNA, a total volume of 250 µl of RNase-free water was added for a nal stock concentration of 20 µM. For the preparation of 1 µM working solution, 2 µl of siRNA was added to the stock solution for a nal volume of 20 µl. All siRNAs were stored at -20°C until further use. Approximately 60 000 cells were seeded in each well of 24-well plate containing 500 µl of complete culture medium. The cells were then incubated overnight under normal growth condition at 37°C with 5%

Rna Extraction And Cdna Synthesis
RNA extraction was performed after 48 hours of transfection by using a commercial RNA extraction kit, RNeasy Mini Kit (QIAGEN, Germany). The protocols of RNA extraction were carried out according to the manufacturer's protocols. The quality of the RNA was assessed by gel electrophoresis. Synthesis of cDNA was performed by using a commercial kit, SensiFAST™ cDNA Synthesis Kit (Bioline, USA) according to manufacturer's protocols. A total of 1 µg RNA served as an initial concentration for cDNA synthesis followed by the following conditions; 25°C for 10 minutes (primer annealing) followed by reverse transcription at 42°C for 15 minutes and the inactivation step at 85°C for 5 minutes with a nal hold at 4°C. The nal volume of 20 µl cDNA in RNase-free was stored in -20°C until further use.
Gene Expression Using Quantitative Real-time Pcr The relative quanti cation of gene expression for β-actin and the targeted genes, MSH2, EPCAM and βcatenin was performed using Stratagene Mx3000P qPCR System (Agilent Technologies, USA). cDNA was diluted in 10-fold dilution of 5 different concentrations with the initial concentration of 100 ng/µl to further determine the PCR e ciency of each targeted gene. Real-time PCR (qPCR) ampli cation was carried out using commercial kit, Quantinova™ SYBR® PCR Kit (QIAGEN, Germany). The kit was used in combination of real-time predesigned primers of Quantitect Primer Assay (QIAGEN, Germany). The reaction mastermix was prepared in 96 well plate according to the proposed reaction setup according to the manufacturer's protocol. β-actin was used as a reference gene for the relative quanti cation of MSH2, EPCAM and β-catenin genes. The reaction mixture for β-actin, MSH2, EPCAM and β-catenin was prepared in a total volume of 20 µl containing 10 ng of cDNA, 10 µl of 1X SYBR Green PCR buffer, 2 µl of 1X QN ROX reference dye, 1X primer assay and RNAse free water. The samples were run in triplicates for each set of the primers. The conditions for the ampli cation were 95°C for 2 min for the initial heat inactivation, denaturation at 95°C for 5 sec and 45 cycles of combined annealing and extension at 60°C for 10 sec followed by dissociation stage at 95°C for 1 min, 55°C for 30 sec and 95°C for 30 sec.

Relative Quanti cation Analysis And Statistical Analysis
Relative quanti cation analysis was performed to determine the expression of the targeted genes (MSH2, EPCAM and β-catenin) in all samples by using comparative double ΔCt method.
Statistical analysis was performed by SPSS® statistical package, version 24.0 (SPSS Inc., Chicago, IL, USA). All data were presented as mean ± SD. The statistical comparison of more than two groups in this experiment was tested using one-way ANOVA. P < 0.05 was considered signi cant.

Western Blot Analysis
Protein extraction was performed by using RIPA Buffer (Nacalai Tesque, Japan) according to the manufacturer's protocols. The cells were washed twice with cold PBS and 1X RIPA buffer was added to the culture depending on the number of cells in the culture ask. The protein concentration in a sample was measured using Bradford protein assay. Each sample replicates (n = 3) for each treatment were pooled separately with concentration of 20 µg. The protein samples and the standard marker subsequently subjected to 10% SDS-PAGE electrophoresis at 100 V for 80 minutes. The samples were transferred to PVDF membrane and was run at 25 V for 2 hours prior to incubation in 5% blocking buffer for 1 hour at room temperature. The membrane was then incubated with 5% blocking buffer mixed with an appropriate dilution of each primary antibody; MSH2 (1:2000), EPCAM (1:2000) and β-catenin (1:1800) and β-actin (1:7500), for overnight at 4°C with gentle agitation. After an overnight incubation, the primary antibody was discarded. The membrane was washed twice at room temperature with TBS-T buffer followed by one time washing with TBS buffer. The membrane was then incubated with Goat Anti-Mouse IgG secondary antibody conjugated to horseradish peroxidase (HRP) in 5% blocking buffer at 1:20 000 dilutions for 1 hour with gentle agitation. The samples were visualized using chemiluminescent detection kit (Nacalai Tesque, Japan).

Densitometry Analysis
The intensity of the protein bands was semi-quantitatively analysed by Image J tool (NIH, Bethesda, MD, USA) (http://rsb.info.nih.gov/ij/index.html) to compare the intensity of each protein band across different samples. A pro le plot for each band that represented the density of the band was then created by using the Image J tool. The relative protein level of each sample was then calculated based on the following equation [11].

Effect of siRNA transfection
The effect of MSH2 and EPCAM on the Wnt/β-catenin was investigated by transfecting the HCT116 cell lines with siRNA against the MSH2 and EPCAM gene. The comparison of cell growth and morphology in each group of treated samples were observed after 48 hours of treatment (Fig. 1). The number of attached cells in the MSH2 and EPCAM treated cells or MSH2 and EPCAM knockdown cells as well as cells which were simultaneously treated by both genes were found to decrease notably. From the microscopic observation, the cells in Group 1, Group 2 and Group 6 was appeared as epithelial in shape as compared to the group of cells treated with the targeted genes. The morphology of cells in Group 3, Group 4 and Group 5 were observed to change into rounded bodies with a few smaller cells in epithelial shape. In the treated samples, the detached cells were observed to be aggregated with a few cells remained attached to the surface of the ask. As compared to Group 3 and Group 4 samples, the cells in Group 5 displayed the highest cell aggregation with a few numbers of cells attached on the surface of the ask.

Gene expression analysis of MSH2 and EPCAM knockdown
The gene expression level of MSH2, EPCAM and β-catenin in six different groups were determined by comparing the 2 −ΔΔCt values of the target genes normalized to the housekeeping genes, β-actin with respective to the control group. The expression level of individual gene knockdown, MSH2-siRNA treated group and EPCAM-siRNA treated group and simultaneous double gene knockdown, MSH2 + EPCAM-siRNA treated group were shown to be statistically signi cant with p-value < 0.05 as compared to the control (Fig. 2). MSH2-siRNA signi cantly inhibited the MSH2 mRNA expression at the level of 0.17 ± 0.04 (p-value = 0.005) as compared to the untreated one (Fig. 2a). Meanwhile, a slight decrease which was not much altered were observed for mRNA expression level in negative control siRNA with no signi cant difference compared to the untreated control. A notable knockdown of MSH2 expression was also observed by EPCAM-siRNA at the level of 0.25 ± 0.05 (p-value = 0.012) (Fig. 2a). Similarly, EPCAM expression level was inhibited by both MSH2-siRNA and EPCAM-siRNA with the level of 0.25 ± 0.03 (pvalue = 0.007) and 0.11 ± 0.04 (p-value = 0.010), respectively (Fig. 2b). A signi cant knockdown was also observed in β-catenin in which both MSH2-siRNA and EPCAM-siRNA has reduced β-catenin expression at the level of 0.34 ± 0.07 (p-value = 0.044) and 0.32 ± 0.04 (p-value = 0.016), respectively (Fig. 2c). The effect of simultaneous gene expression was determined in these targeted genes. The mRNA expression level in these three targeted genes were observed to have signi cantly reduced by MSH2 + EPCAM-siRNA (Fig. 2).

Protein expression analysis of MSH2 and EPCAM knockdown
Protein expression was observed after 48 hours of transfection, including the untreated control. Reduced band intensity was observed for MSH2-siRNA and EPCAM-siRNA treated samples that indicated less protein was expressed for these three targeted genes as compared to untreated control. Reduced band intensity of MSH2 + EPCAM siRNA treated samples were further observed that may indicate the possible effect of simultaneous gene knockdown resulted in low expression of the two targeted proteins (Fig. 3). The percentage of MSH2 protein level transfected by the individual MSH2-siRNA and EPCAM-siRNA were reduced at 33% and 32% respectively (Fig. 4a). Increased reduction of protein level was observed in the simultaneous knockdown of MSH2 and EPCAM siRNA with the percentage protein level of 14% (Fig. 4a). For EPCAM gene, the protein level was also identi ed to decrease in both targeted siRNA by 44% in MSH2-siRNA transfected sample and 43% of protein level in the EPCAM-siRNA transfected sample (Fig. 4.2b). Reduced protein level was identi ed in both genes transfected sample with the remaining protein level of 10% (Fig. 4.2b). In addition, for β-catenin gene, the MSH2 and EPCAM-siRNA transfected sample was decreased to 38% and 33% respectively (Fig. 4c). Notable reduction of β-catenin protein level was observed in the simultaneous gene knockdown with the percentage remaining protein level of 5% (Fig. 4c).

Discussions
Germline deletion that cause the MSH2 inactivation was considered a novel ndings in the predisposition of LS [4,5] including a novel large duplication of MSH2-EPCAM previously reported in a LS patient [12]. However, the association of these two genes in the pathway associated to CRC and LS have not yet being elucidated. Previous study showed that EPCAM was suggested as an important target gene that can disrupt the mechanism of Wnt/β-catenin pathway in HCT116 cells [7] but tumour cells commonly involved multiple genetic and epigenetic alterations and single inhibition of one tumour associated gene as a therapeutic strategy may inadequate to inhibit the development of tumour [13].
The present study was carried out in colon cancer cell line, HCT116 to evaluate the role of MSH2 and EPCAM gene in modulating the Wnt/β-catenin pathway using siRNA target gene. The resulted gene silencing after siRNA transfection was evaluated by analysis of gene expression and protein expression by Western blot analysis which further veri ed the e cient gene knockdown in the mRNA and protein level. The expression of individual gene knockdown has reduced the mRNA and protein expression of MSH2, EPCAM and β-catenin in EPCAM-siRNA and MSH2-siRNA transfected cells. This nding was concordant with previous study that reported a signi cant reduction of proliferative activity and reduced mRNA and protein expression in EPCAM-siRNA transfected in HCT116 cells [7]. The expression level of βcatenin was also reduced in the EPCAM-siRNA transfected cells [7]. In addition, MSH2 knockdown was previously carried out in other colon cancer cell lines, SW480 using shRNA-mediated gene silencing which resulted in the reduction of cell proliferative activity and decreased mRNA expression [14].
The simultaneous knockdown of MSH2 and EPCAM may also be associated with the epigenetic modi cations that usually occurs at the promoter region or enhancers of tumour-suppressor genes that often cause tumorigenesis [15]. MSH2 methylation was reported as disease speci c due to the absence of MSH2 methylation in normal tissues as well as in sporadic CRC cases [16]. In addition, promoter region of MSH2 was suggested as a target of aberrant methylation in LS due to the presence of high level of promoter methylation [16]. Previous study demonstrated the role of 3'-end EPCAM deletion that may cause MSH2 methylation in patients with no LS germline mutations in MSH2 gene [8]. Due to the downstream position of MSH2 gene to EPCAM gene, gene silencing by transcriptional read-through of a neighbouring gene could represent a general mutational mechanism and also caused by a second somatic hit that inactivates MSH2 in tumours with EPCAM deletion [8].
A signi cant reduction of β-catenin expression was observed in the simultaneous gene knockdown of MSH2 and EPCAM which suggested that the silencing of these two genes may interrupt the activation of Wnt/β-catenin pathway in cancer cells. It has been well known that the degradation of Tcf/β-catenin complex formation decreased EPCAM gene expression in normal human hepatocytes culture and HCC cell lines [17] and human colon cancer line [7]. The key transcription regulator of Wnt signalling pathway is CTNNB1 gene which was identi ed to encode for β-catenin [18]. The mechanism took place when the phosphorylation of β-catenin by CK1α followed by GSK3β mediated phosphorylation of the destruction complex and targeting β-catenin for ubiquitination and subsequent proteolysis [19]. However, the subsequent degradation of β-catenin will be avoided by the point mutations at these amino acids by hindering the β-catenin from being phosphorylated [18]. Mutations in CTNNB1 and AXIN2 in CRC were mostly arise in tumours with MMR genes; MLH1, MSH2 or PMS2 inactivation [20]. Although previous invivo study has showed that the de ciency of MSH2 led to an enhance of β-catenin activity and cellular hyperproliferation in colon epithelial cells, however, this activity was dependent on gut microbes [21].
In conclusion, Wnt pathway played an important role as molecular signalling in various cancers and due to frequent abnormality of Wnt activation in colorectal cancer, it has been proposed as one of the key pathway in CRC predisposition [22,23]. The activation of Wnt signalling commonly occurs by the presence of genetic alterations in APC, β-catenin gene, AXIN1 and AXIN2 that further caused β-catenin to be accumulated in the cytoplasm [20]. Based on the ndings in the in-vitro study, the profound effect of the simultaneous gene knockdown of MSH2 and EPCAM gene to the reduction of β-catenin expression may indicate the ability of these two genes to become co-target genes in regulating the Wnt/β-catenin pathway. This combined siRNA approach has also suggested to be a new therapeutic approach in the treatment of CRC as well as LS through suppression of Wnt/β-catenin pathway.