Sox4 Regulates the Sensitivity of Canine Mammary Gland Tumor Cells to Cisplatin via the Wnt/β-catenin Signaling Pathway

Background: The development of cisplatin resistance is one of the major causes of breast cancer treatment failure, and is associated with changes in Sox4 gene expression. In this study, a cisplatin-resistant cell line, CHMp CIS , was constructed from the cell line CHMp, which was isolated from the primary lesion of a malignant canine mammary gland tumor (CMGT). Sox4 expression was evaluated to assess its roles in cisplatin sensitivity, proliferation and apoptosis, epithelial-mesenchymal transition (EMT), cancer stem cell (CSC) features, and activation of the Wnt/β-catenin signaling pathway in CMGT cells. Results: CHMp CIS Cells exhibited changes in morphology, slower proliferation, and greater anti-apoptotic ability, EMT and CSC features, and the Wnt/β-catenin pathway was activated in CHMp CIS cells. In CMGT tissues, Sox4 expression was elevated. In CHMp CIS cells, silencing Sox4 inhibited cisplatin resistance, EMT and CSC features, and Wnt/β-catenin signaling activation. Then activating the Wnt/β-catenin signaling pathway increased Sox4 expression levels. Conclusions: Silencing Sox4 inhibited the above-mentioned cancer cell characteristics in CHMp CIS cells compared with CHMp cells. In addition, activating the Wnt/β-catenin signaling pathway increased Sox4 expression levels, as part of a positive feedback loop. These ndings may provide new targets and therapeutic strategies for the clinical treatment of CMGT as well as a reference for human mammary gland tumor (HMGT) research. form cell Relative expression levels of Wnt/β-catenin pathway-related in CHMp and CHMpCIS cells

The Wnt/β-catenin pathway regulates the process of cell development, and affects cell morphology, proliferation, differentiation, apoptosis, EMT and cancer stem cell (CSC) features in various tissue environments. Disorders of this pathway may also lead to breast tumor formation [8][9][10][11]. Sox4 (sexdetermining region Y-box4) is a member of the Sox transcription factor family and a so-called "cancer characteristic genes" [12]. High Sox4 expression is associated with poor prognosis and may promote the development of cancer through a variety of mechanisms, including regulation of cellular proliferation, EMT, self-renewal, apoptosis, and signaling pathways such as Wnt, PI3K/Akt, Notch, and Hedgehog [13]. Sox4 is particularly important for Wnt signaling and can directly interact with β-catenin to activate gene expression [14,15]. The present study explores the effects of Sox4 on CMGT progression and cisplatin resistance and the interaction between Sox4 and the Wnt/β-catenin pathway. ow cytometry analysis showed that in the presence of cisplatin, the apoptotic rates of CHMp CIS and CHMp cells were 6.0% and 49.3%, respectively (Figure 1 F). The expression of the apoptotic protein Bax was remarkably reduced in CHMp CIS cells, whereas the expression of the anti-apoptotic protein Bcl-2 and the multidrug resistance proteins MDR1 and MRP1 was remarkably increased (Figure 1 G). These results indicated that CHMp CIS cells have enhanced anti-apoptotic ability in addition to drug resistance.

EMT and CSC Features Are Enhanced in CHMp CIS Cells
Compared with CHMp cells, CHMp CIS cells exhibited similar migration ability but distinctly stronger invasion ability in the absence of cisplatin. The addition of cisplatin obviously inhibited the migration and invasion ability of CHMp cells but had little effect on CHMp CIS cells (Figure 2 A, B). In addition, the expression of epithelial cadherin (E-cadherin) was lower in CHMp CIS cells than in CHMp cells, whereas the expression of N-cadherin, a protein that provides greater connection exibility, was increased in CHMp CIS cells (Figure 2 C). These observations indicate enhanced EMT in CHMp CIS cells. In the mammosphere experiment, CHMp CIS cells formed more and larger II MS than CHMp cells; in addition, the majority of the mammospheres formed by CHMp CIS cells, up to 70%, were large spheres (Figure 2 D). As shown in Figure   2 E, CHMp CIS cells formed more clonal cell clusters than CHMp cells in the presence or absence of cisplatin; in fact, in the presence of cisplatin, CHMp cells were unable to form clonal cell clusters. Moreover, the expression of three key proteins related to cell stemness, Nanog, Oct4, and Sox2, was notably higher in CHMp CIS cells than in CHMp cells (Figure 2  clusters, albeit smaller in size, in the presence of FH535 as in its absence. In CHMp CIS cells, exposure to FH535 increased the expression of the key Wnt/β-catenin pathway proteins β-catenin and Wnt3a but decreased the expression of Gsk3β, a protein involved in the degradation of β-catenin. Accordingly, the expression of P-Gsk3β and P-β-catenin increased and decreased, respectively. The above results indicate that the Wnt/β-catenin pathway is active in CHMp CIS cells.

Sox4 Expression Is Elevated in CMGT Tissues
The analysis of Sox4 protein and mRNA expression showed that Sox4 expression was signi cantly higher in CMGT tissues (CMGTT) than in adjacent tissues (CAMGTT) (Figure 4

Discussion
Cisplatin resistance remains an important issue in the clinical treatment of breast cancer. Here, we demonstrated that MDR1 and MPR1 expression, anti-apoptotic ability, EMT, CSC features, and activation of the Wnt/β-catenin signaling pathway were enhanced in cisplatin-resistant CMGT cells compared with parent CMGT cells. Moreover, the expression of Sox4 was elevated in CMGT tissues compared with adjacent tissues. Cisplatin resistance, apoptosis-and drug resistance-related protein expression, EMT and CSC features were all related to the expression of Sox4 in CHMp CIS cells. Sox4 silencing signi cantly reversed cisplatin resistance, related biological characteristics and Wnt/β-catenin pathway activity in CHMp CIS cells. Finally, activating the Wnt/β-catenin signaling pathway increased Sox4 expression. Taken together, these ndings indicate that Sox4 plays a key role in the development of cisplatin resistance by increasing autocrine Wnt signaling, thereby enabling escape from apoptosis and enhancing EMT and CSC features.
In ovarian cancer cells, cisplatin resistance is associated with upregulated expression of the antiapoptotic protein Bcl-2, activation of Fas, blockade of Caspase-3 and Caspase-8 activation, reduced levels of the pro-apoptotic protein Bax, and increased cell proliferation due to elevated Wnt/β-catenin expression [17,18]. There is evidence that the multi-drug resistance gene MDR1 is also a target of TCF/βcatenin [19]. MDR1 and MRP1 are cisplatin-pumping proteins, and their overexpression is considered one of the main mechanisms of cisplatin resistance. The expression of β-catenin is positively correlated with upregulation of MDR1 gene expression [20,21]. Consistent with our ndings, the Wnt/β-catenin pathway enhances the EMT and CSC features of human tongue squamous cell carcinoma and breast cancer cells [22,23].
Sox4 is an important developmental transcription factor that regulates stemness, differentiation, progenitor cell development and multiple developmental pathways, including PI3K, Wnt and TGFβ signal transduction [13]. Sox4 also inhibits terminal differentiation. These functions are strongly associated with the development of malignant tumors, and Sox4 gene upregulation has been observed in a variety of cancers [24,25]. Increased Sox4 activity promotes cancer cell survival, proliferation [26], migration [15], and self-renewal [27]. High Sox4 expression is associated with poor tumor prognosis in patients, and thus Sox4 is a pan-cancer prognostic biomarker [28]. However, in dogs, the prognostic value of Sox4 expression has not been examined. In the present study, silencing Sox4 reduced β-catenin activation, resulting in a weakened invasive phenotype. Similarly, silencing Sox4 has been shown to prevent cancer progression before prostate intraepithelial neoplasia becomes cancerous [29].
The induction of β-catenin/TCF activity by Sox4 depends on the stability of the β-catenin protein and not the induction of β-catenin transcription [30]. We showed that Sox4 knockdown suppressed the Wnt/βcatenin signaling pathway, thereby restraining the proliferation, migration, invasion, and self-renewal of CMGT cells. Sox4 gene deletion also reduces active β-catenin levels in prostate cancer [31]. The mechanism by which Sox4 deletion inhibits β-catenin activation remains to be determined, but there are several possibilities. First, Sox4 may directly interact with TCF and β-catenin to stabilize the βcatenin/TCF complex [14]. Second, Sox4 may stimulate β-catenin activity indirectly by maintaining active AKT. Crosstalk is known to occur between the PI3K-AKT and Wnt/β-catenin pathways, as AKT phosphorylation and inhibition of GSK3β increase the activity of Wnt pathway mediators and β-catenin [32]. Third, it has been proposed that Sox4 stabilizes β-catenin by inducing the transcription of CK2, thereby protecting it from degradation. Finally, in endometrial cancer, Sox4 has been found to promote Wnt/β-catenin signal transduction through direct transcription to activate the expression of TCF4, leading to the activation of Wnt target genes [28].
Activation of Wnt/β-catenin signaling pathway further regulates Sox4 gene expression through a positive feedback loop, thereby promoting tumor progression. Regulation of Sox4 transcriptional activity and expression occurs not only at the mRNA level but also via post-translational modi cations (PTMs) and protein-protein interactions. In prostate cancer, C-myc and CUL4B act directly on Sox4 to promote progression [33,34], forming a positive feedback loop. The transcriptional activity and target gene speci city of Sox4 are also controlled by synergistic interactions with different transcription factors and cofactors. Syntenin-1 regulates Sox4 protein stability and transcriptional activity by interacting with syndecans [35], ephrin [36], Frizzled [37], etc. Other mutations in cancer cells, including in receptor signaling pathways and DNA damage pathways, may also affect Sox4 transcriptional activity [28].
Taken together, these observations emphasize the relevance of the interaction of Sox4 with the Wnt/βcatenin signaling pathway in CMGT and cisplatin-resistance. The development of cisplatin resistance increases the expression of Sox4, leading to Wnt signal activation, enhanced cancer cell migration and invasion capabilities, and cancer stem cell population enrichment. Targeting Sox4 may facilitate the differentiation of stem cells and increase the sensitivity of cells to cisplatin. Thus, treatment methods that reduce Sox4 expression may prevent cancer recurrence in CMGT and provide a theoretical basis for the treatment of HMGT. CHMp cells with 0.25% EDTA-trypsin (Beyotime, China, C0201-100ml). The cells were then diluted with complete culture medium containing 20% FBS and inoculated into 96-well plates. An inverted microscope was used to identify wells containing single cells, and then the plates were incubated at 37°C and viewed once every 24h. The surviving cells were expanded and exposed to a drug concentration gradient plus drug maintenance method ( nal cisplatin concentration of 20 μmol/L) for 10 months to obtain the CHMp CIS cell line.

Western Blot Analysis
Cell proteins were extracted with the BCA Protein Assay Kit (Beyotime Biotechnology, P0012S) and separated by SDS-polyacrylamide gel electrophoresis. The separated proteins were transferred to an NC membrane, which was blocked and then incubated sequentially with primary and secondary antibodies. Finally, ECL Luminescent Solution (Beyotime Biotechnology, P0018FS) was added for imaging. Quanti cation was performed using Image J software.

Cell Counting Kit-8 (CCK8) Assay
Logarithmic-phase CHMp and CHMp CIS cells were adjusted to 3×10 4 cells/mL and transferred to 96-well plates at 100μL/well for culture at 37℃. After 24 h, the culture medium was replaced with medium containing a gradient of cisplatin concentrations, and the cells were again cultured at 37°C for 24 h. The culture medium was then replaced with CCK8 detection solution (Abcam, UK, ab228554), and the cells were incubated at 37°C for 30 minutes. Finally, OD values were recorded using a microplate reader. The half maximal inhibitory concentration (IC50) of cisplatin was calculated according to relative cell survival.
In separate assays, logarithmic-phase CHMp and CHMp CIS cells were adjusted to 1×10 4 cells/mL and transferred to 96-well plates at 100 μL/well for culture at 37℃. On days 1, 2, 3, 4, or 5, the culture medium was replaced with CCK8 detection solution, the OD value was detected, and the value-increasing curves were plotted.

Flow Cytometry
Cells adjusted to 1×10 5 cells/mL were seeded in a 6-well plate and cultured in a 37°C incubator. After 48 h, the cell culture solution was collected, and the cells were harvested by digestion with trypsin and centrifugation. Apoptosis was detected by ow cytometry with Annexin V-FITC/PI (Bioss, China, BA00101).

Transwell Assay
Cells adjusted to 5×10 4 cells/mL in serum-free culture medium were transferred to the upper chamber of a Transwell apparatus at 100 μL/well (Corning, USA, 3422). During the migration experiment, Matrigel was not added to the upper chamber, whereas for the invasion experiment, the upper chamber was coated with Matrigel in advance. Complete medium containing 20% fetal bovine serum was added to the lower chamber at 600 μL per chamber and the entire apparatus was placed in a 37°C incubator. After migration for 24 h/invasion for 48 h, the cells on the lower surface of the upper chamber were xed with paraformaldehyde, stained with crystal violet, washed with distilled water, and dried in a fume hood. Cells on transparent plates were photographed under an inverted microscope (×200 magni cation). For counting, 5 elds were randomly selected from each group of cell samples.

Clone Formation Assay
Cells were seeded in a 6-well culture plate at 500 cells/well. The medium was replaced with fresh medium on day 5. On day 10, the cells were xed in 4% paraformaldehyde and stained with crystal violet, and the number of colonies was recorded.

Cell Transfection
Transfection kits, siR-NC and siR-SOX4 were purchased from Ribobio. Transfection was performed according to the manufacturer's instructions.

Reverse-transcription Quantitative PCR (QRT-PCR)
Total RNA was extracted using TRIzol reagent (Invitrogen, USA, 15596026). The total RNA concentration was measured with an ultra-micro ultraviolet spectrophotometer (Thermo, USA, ND-ONE-W), and the RNA was reverse-transcribed using a cDNA rst-strand synthesis kit (Tiangen, China, KR118) with TB Green II dye (TaKaRa, Japan, RR820A). Finally, a Light Cycler R 480 System (Roche, Basel, Switzerland) was used for quantitative analysis. Each gene was analyzed at least 3 times in each sample, and the average value was calculated. The data were analyzed by the 2 -δδCT method. See Supplementary Table 1 for primer sequences.

Statistical Analysis
Data are presented as the mean ± SD. Statistical comparisons between treated and control groups were performed using Student's t-test in GraphPad Prism 5, and multiple groups were compared by two-way ANOVA. Differences that are not signi cant (P>0.05) are not marked; signi cant differences (0.01 P 0.05) are marked with "*"/"#", and extremely signi cant differences (P<0.01) are marked with "**"/"##". The animal study was reviewed and approved by Northeast Agricultural University, Harbin, Heilongjiang Province. Written informed consent was obtained from the owners for the participation of their animals in this study.

Consent for publication
Not applicable Availability of data and materials All of the data generated or analysed during this study are available from the rst author on reasonable request.

Competing interests
The authors declare that they have no competing interests.