SOX8 Knockdown Overcomes Enzalutamide Resistance in Castration-Resistant Prostate Cancer by Inhibiting the Notch Signaling Pathway

Castration-resistant prostate cancer (CRPC) is still challenging to treat. Dissatisfaction with androgen signal-targeted therapy forces people to look for other treatment strategies. Therefore, this study is aimed at exploring the role of SOX8/Notch signaling in CRPC. The upregulation of SOX8, Notch4, and Hes5 indicated a poor progression-free survival (PFS) in CRPC patients. The expression of these proteins was also upregulated in enzalutamide-resistant LNCaP cells (Enza-R). Moreover, knocking down SOX8 inhibited malignant biological behaviors and decreased the activation of Notch signaling in Enza-R cells. Importantly, knocking down SOX8 obviously reversed the enzalutamide resistance in Enza-R cells, while RO0429097 (a γ secretase inhibitor inactivates Notch signaling) exerted similar effects. At last, we found that both SOX8 knockdown and/or RO0429097 suppressed tumor growth and bone metastasis in vivo. Altogether, our study indicated that the SOX8/Notch signaling is involved in CRPC and that these enzymes are possible targets to develop novel treatment for CRPC.


Introduction
Prostate cancer (PCa) is the most common cancer among men and represents one of the leading causes of cancer-related deaths in developed countries [1]. For advanced PCa, androgen deprivation therapy (ADT) is the mainstream therapy. However, most patients with advanced PCa develop a castration-resistant prostate cancer (CRPC) within 18-24 months after ADT [2,3]. Unfortunately, current treatment strategies, including endocrine therapy, such as abiraterone and enzalutamide, and chemotherapy, result in poor longterm survival for CRPC patients [4][5][6][7][8]. Therefore, other molecular mechanisms that lead to CRPC have been investigated as an attempt to uncover novel therapeutic targets.
The Notch signaling pathway is highly conserved in mammalian cells as it determines the fate and differentiation of cells. At the same time, it participates in the development of many organs, including the prostate [9,10]. Interestingly, there is a controversy about whether the Notch signaling pathway acts as a tumor suppressor or as an oncogene [11][12][13][14][15][16]. Excessive activation of the Notch signaling pathway has been reported in PCa, including in patients with CRPC [17][18][19]. Overexpression of Notch signaling molecules has been associated with PCa, while downregulation of Notch receptors inhibited malignant biological behaviors of PCa cells [20][21][22][23][24]. More importantly, Notch inhibitors (γ-secretase inhibitors), such as PF-3084014 and GSI-IX, enhance the efficacy of ADT in PCa [25][26][27]. In addition, our recent study has indicated that PF-3084014 partly reverses enzalutamide resistance in CRPC cells by inhibiting the Notch1 receptor [28]. Chemoresistance in PCa has also been associated with the dysregulation of Notch2 receptors [29,30]. Despite these studies, underlying mechanisms of Notch receptors involved in ADT resistance and CRPC are still unclear.
The SOX (SRY-related HMG-box) family of genes includes approximately 30 different subtypes, termed from A to H. These genes are found in multiple types of progenitor cells and play a key role in the regulation of cell development [31]. SOX8 belongs to SOX group E, and it was shown to be overexpressed in various cancer types including triplenegative breast cancer (TNBC), ovarian cancer, and tongue squamous cell carcinoma [32][33][34], while promoting the tumorigenesis and progression of tumors. Importantly, SOX8 was shown to be responsible for the chemotherapy resistance of multiple cancers [32,34]. However, the role of SOX8 in PCa, especially in CRPC, is still unknown.
In this study, we reported that SOX8, Notch4, and Hes5 were significantly elevated in CRPC samples when compared with those of PCa samples. Increased levels of SOX8, Notch4, and Hes5 represented a worse prognosis for CRPC patients. We also found that these enzymes were upregulated in CRPC cells (named as Enza-R cells), when compared to their parental cells, LNCaP. Moreover, downregulating SOX8 significantly inhibited malignant behaviors of both CRPC and DU145 cells and reversed the resistance to enzalutamide by decreasing activities of Notch signaling. Importantly, a γ-secretase inhibitor (Notch signaling inhibitor) RO0429097 obviously restored the sensitivity of Enza-R cells to enzalutamide. Finally, SOX8 knockdown or RO0429097 was able to block the growth and bone metastasis of Enza-R cells in vivo. Taken together, our results indicate that the SOX8/Notch signaling axis may be a promising therapeutic strategy for CRPC.

Patients and Tissue Samples.
A total of 45 PCa samples were collected from the Affiliated Hospital of North Sichuan Medical College, Nanchong, China, between April 2018 and April 2020. The inclusion criteria for CRPC were (a) to abide by the EAU guidelines on CRPC and (b) that patients had available CRPC specimens and complete clinical data. Therefore, 35 CRPC tissues, including frozen and paraffinembedded tissues, were obtained from the Affiliated Hospital of North Sichuan Medical College (10 cases); the First Affiliated Hospital of Chongqing Medical University, Chongqing, China (18 cases); and the Fuling Central Hospital, Chongqing, China (7 cases). These cases were collected between May 2008 and April 2020. All PCa and CRPC tissues were confirmed by a trained pathologist. The study protocol was approved by the Human Ethics Review Committee of the First Affiliated Hospital of Chongqing Medical University and Ethics Committee of the Affiliated Hospital of North Sichuan Medical College. This study conforms to the provisions of the Declaration of Helsinki. Informed consent was obtained from the patients or their family members prior to inclusion in the study.
2.5. Protein Expression by Western Blotting. Total protein was extracted from cell lines using RIPA buffer containing phosphatase inhibitors (Beyotime Institute of Biotechnology, Beijing, China) according to the manufacturer's procedures. Protein samples (50 μg) were transferred to PVDF membranes (EMD Millipore, MA, USA). After blocking with 5% nonfat milk for 2 hours at room temperature, membranes were treated with the following various primary antibodies overnight at 4°C (Table 1). GAPDH was used as a loading control. All Western blot experiments were repeated at least 3 times.

Colony Formation Assay.
A total of 400 cells/well were plated into 6-well plates and were cultured for two weeks until the number of each clone reached 50 cells. Then, the cells were fixed with 4% paraformaldehyde for 15 minutes and stained with 0.05% crystal violet for 20 minutes at room temperature. Each group was replicated in three wells.
2.9. Transwell and Wound Healing Assay. For the Transwell assay, 1:0 × 10 4 cells were seeded in the upper chamber of the insert with Matrigel (BD Biosciences, USA). After incubation with serum-free medium for 48 h, cells were stained with 0.1% crystal violet and 4% formaldehyde. The number of cells fixed on the bottom membrane of the inserts was counted under an optical microscope. For the wound healing assay, 5 × 10 4 cells/well were plated into a 6-well plate. After incubating for 24 hours, cells were wounded with a yellow pipette tip. Then, the cells were cultured for 24 hours, and the wound healing was observed under an optical microscope at indicated time-points. 2.11. Statistical Analysis. Statistical analyses were performed using the SPSS 19.0 software. Numerical data are shown as mean ± SD. Studentʼs t-test, χ 2 test, Mann-Whitney test, Pearsonʼs analysis, one-way ANOVA, two-way ANOVA, and Kaplan-Meier survival analysis were performed as appropriate. P < 0:05 was used to infer statistical differences.
Next, a Kaplan-Meier survival analysis was used to evaluate the relationship between the progression-free survival (PFS) of CRPC patients and the expression of SOX8, Notch4, and Hes5. This analysis revealed that the median PFS was 24 months in CRPC patients that were positive for SOX8 and 43 months in SOX8-negative patients ( Figure 1(i), P = 0:0365). Moreover, the PFS of Notch4positive CRPC patients was obviously shorter than that of Notch4-negative patients (24 months vs. 50.5 months, P = 0:0101) (Figure 1(j)). Finally, the PFS of Hes5-positive CRPC patients was shorter than that of Hes5-negative patients (22 months vs. 38 months, P = 0:0235) (Figure 1(k)).

SOX8, Notch4, and Hes5 Are Upregulated in CRPC Cells.
To determine a possible role of SOX8 and Notch signaling in CRPC, we constructed Enza-R cells by continuously treating LNCaP with enzalutamide for at least 6 months. As shown in Figure 2(a), the resistance to enzalutamide in Enza-R cells increased nearly 100-fold compared to their parental cells. Next, the expression of SOX8, Notch4, and Hes5 was detected by RT-qPCR, Western blot, and immunofluorescence. As expected, both mRNA and protein expressions of SOX8, Notch4, and Hes5 were upregulated in DU145 and Enza-R cells. However, these proteins were not detected in   (Figure 3(j)).

SOX8 Knockdown Inhibits Malignant Biological
Behaviors of Enza-R Cells through Regulating the Notch Signaling Pathway. As mentioned above, the expression of SOX8 was positively correlated with Notch signaling in CRPC tissues (Figures 1(g) and 1(h)). We hypothesized that knocking down SOX8 would inhibit malignant biological behaviors of resistant cells through the downregulation of To determine the role of Notch signaling in Enza-R cells, we treated cells with 5 μM RO04929097 (a γ-secretase inhibitor) for 48 h, which inactivates Notch signaling. As shown in Figure 4(d), a combination of SOX8 knockdown and RO04929097 led to a more potent inhibition of the expression of multiple oncogenic pathways, such as p21 and c-myc. In addition, to investigate the correlation between SOX8 and Cyclin family members in Enza-R cells, the expression of Cyclin E1, Cyclin D1, and Cyclin D3 was detected after knocking down SOX8. As shown in Figure 4(e), downregulation of SOX8 decreased the activity of Cyclin family members, suggesting that the dysregulation of SOX8 promotes the proliferation of Enza-R cells. When SOX8 knockdown cells were treated with RO04929097, there was a more obvious decrease of Cyclin family members, indicating that both SOX8 and Notch signaling are involved in regulating the mitosis of Enza-R cells. Moreover, a synergistic effect between SOX8 downregulation and RO04929097 was observed on the apoptosis of Enza-R cells, as evidenced by the upregulation of BAX and BAK and the downregulation of Bcl-2 and Bcl-xl (Figures 4(e) and 4(f)).
Next, we investigated if SOX8-mediated effects on CRPC cells were due to the regulation of the Notch signaling pathway. Notch4 receptor was knocked down by adenoviruses in LNCaP cells that overexpressed SOX8. The CCK8 assay showed that overexpression of SOX8 promoted the   BioMed Research International proliferation of LNCaP; however, when Notch4 receptor was concomitantly knocked down, the growth of LNCaP was obviously inhibited (Figures 5(a) and 5(b)). Moreover, knocking down Notch4 impaired the invasion ability caused by the overexpression of SOX8 in LNCaP cells ( Figure 5(b)). As shown in Figure 5(c), Notch4 knockdown led to a downregulation of the expression of Cyclin E1, Bcl-2, and Ncadherin and an upregulation of the expression of BAK1 and E-cadherin. Our data suggest that knocking down Notch4 could rescue the proliferation and invasion caused by the overexpression of SOX8 in CRPC cells. More importantly, the overexpression of SOX8 increased enzalutamide resistance by 2-fold. However, when knocking down Notch4, such drug resistance was reversed in LNCaP cells ( Figure 5(d)). Taken together, data herein presented support that Notch4 downregulation can rescue the proliferation, invasion, metastasis, and drug resistance caused by the over-expression of SOX8 in CRPC cells, thus inhibiting malignant biological behaviors of CRPC cells. Xie et al. reported that SOX8 confers chemoresistance and stemness properties and mediates EMT in tongue squamous cell carcinoma via bounding to the promoter region of Frizzled-7 (FZD7) and inducing the FZD7-mediated activation of the Wnt/β-catenin pathway [32]. Moreover, previous studies revealed that the induction of the Notch ligand/ receptor was regulated by β-catenin hyperactivation in intestinal tumorigenesis [35][36][37]. Therefore, we hypothesized that SOX8 regulates Notch signaling through β-catenin in CRPC.
As shown in Figure 5(e), SOX8 downregulation decreased the expression of β-catenin, p-β-catenin, Notch1, and Notch4. However, when Enza-R cells were treated with AZD2858 (a Wnt/β-catenin activator), the downregulation of Notch1 and Notch4 was rescued. Similarly, the combination of knocking down SOX8 and treatment with PNU74654   Enza-R cells was examined by Western blot. Cells were transfected with lentiviruses containing LV-NC or LV-shSOX8. GAPDH served as a loading control. (f, g) A Transwell assay was performed to examine the invasive ability of DU145 and Enza-R cells following SOX8 knockdown (magnification, ×400). (h, i). The migratory capacity of Enza-R cells was evaluated after the cells were wounded with a yellow pipette tip for 0 h, 24 h, and 48 h. (j) Enza-R cells were exposed to increasing concentrations of enzalutamide (0, 1, 5, 25, 50, 100, 200, 300, and 400 μM) for 48 h, and the half maximal inhibitory concentration (IC 50 ) was determined by a CCK8 assay. * P < 0:05 and * * P < 0:01. Enza-R: enzalutamide-resistant LNCaP cells. 7 BioMed Research International (a Wnt/β-catenin inhibitor) led to a stronger inhibition of Notch signaling molecules when compared to that of either treatment alone. Consistent with our hypothesis, the activation of Notch signal mediated by SOX8 is achieved through the regulation of β-catenin in Enza-R cells.

Inhibition of the Notch Signaling Pathway by Both
Knocking Down SOX8 and γ-Secretase Inhibitor (RO04929097) Significantly Reversed the Enzalutamide Resistance. As mentioned above, downregulation of SOX8 inhibited the proliferation, invasion, and migration of Enza-R cells. Thus, we hypothesized that the overexpression of SOX8 was responsible for enzalutamide resistance in Enza-R cells. As expected, the CCK-8 assay showed that the downregulation of SOX8 increased the sensitivity of the Enza-R cells to enzalutamide by 5.2-fold ( Figure 6(a)). Similar results were found in DU145 cells, in which there was a 2.26-fold in reversing resistance (Figure 6(c)), suggesting that dysregulation of SOX8 is responsible for enzalutamide resistance. Importantly, a γ-secretase inhibitor named  Figure 4: Knocking down SOX8 and treatment with RO04929097 inhibit Notch signaling and its downstream. (a-c) Notch signaling activity in Enza-R cells after knocking down SOX8 was detected by RT-qPCR, Western blot, and immunofluorescence. Cell nuclei were stained with DAPI (magnification, ×200). In order for the figure to be more concise, panels for DAPI staining alone are not shown. (d-f ) Downstream genes of Notch signaling were detected in DU145 and Enza-R cells after knocking down SOX8 and/or treating with RO04929097 using Western blot and immunofluorescence. * * * P < 0:001; GAPDH served as a loading control. Enza-R: enzalutamideresistant LNCaP cells. 8 BioMed (e) Enza-R cell was subjected to SOX8 knockdown and/or treatment with AZD2858 (5 nM 24 hours) or PNU74654 (5 nM 24 hours). The expression of SOX8, β-catenin, p-β-catenin, Notch1, and Notch4 was detected by Western blot assay; , GAPDH served as a loading control, * * * P < 0:001. 9 BioMed Research International RO04929097 was also able to reversed enzalutamide resistance by 5.8-fold in Enza-R cells and by 1.9-fold in DU145 cells (Figures 5(b) and 5(d)), indicating that pharmacological intervention of Notch signaling by RO04929097 may represent a promising therapeutic strategy for CRPC.
Next, to determine the antitumor effect of RO04929097 on the CRPC cell model, the CCK-8 assay was performed to evaluate the proliferation of Enza-R and DU145 cells following treatment with various concentrations of RO04929097 (5,10,20,30,40,60,80, and 100 μM). As shown in Figures 6(e) and 6(f), RO04929097 exerted a dose-dependent and powerful antitumor effect on both Enza-R and DU145 cells, suggesting that inhibition of γsecretase decreased Notch signaling and may become a novel and potent therapy for CRPC.

Combination of Enzalutamide with RO04929097
Displays a Synergic Effect in Blocking the Growth and Bone Metastasis of Enza-R Cells In Vivo. As mentioned above, SOX8 and Notch inhibitor RO04929097 effectively suppressed malignant biological behaviors of Enza-R cells in vitro. Next, we evaluated the therapeutic potential of these strategies in vivo. Xenograft tumor models were constructed by treating castrated nude mice with a combination of

11
BioMed Research International enzalutamide and RO04929097. Compared to the control group, both LV-shSOX8 and RO04929097 groups had significantly decreased volume and weight of their xenograft tumor (Figures 7(a)-7(c) and 7(f)). Of note, when LV-shSOX8 and RO04929097 were combined, the inhibitory effect became stronger (Figures 7(a)-7(c) and 7(f)). Next, we subcutaneously injected Enza-R cells with LV-NC or LV-shSOX8 into the right tibia to construct a bone metastasis model. Mice were treated with enzalutamide and/or RO04929097 injected into the tail vein. X-ray, H&E histology, and IHC were performed to evaluate bone destruction. As expected, SOX8 knockdown as well as RO04929097 obviously reduced the bone metastasis, compared to the control group (Figures 7(d)-7(f)). Moreover, a synergistic effect was detected in preventing bone destruction when SOX8 knockdown and RO04929097 were combined (Figures 7(d)-7(f)).

Discussion
In healthy organisms, SOX genes regulate cell differentiation, organogenesis, and many other developmental processes [38][39][40][41]. However, SOX gene members are frequently dysregulated in various tumors [42,43]. SOX2 is weakly detected in benign prostate tissues; however, it is highly expressed in PCa tissues, including in CRPC ones. More importantly, SOX2 promotes tumor tumorigenesis and progression. Reduced SOX2 levels were shown to attenuate the proliferation and invasion while increasing the redifferentiation of PCa cells [44,45]. Meanwhile, SOX4 is also overexpressed in PCa tissues and cell lines, and its upregulation is correlated with a higher Gleason score.
Moreover, decreased SOX4 induces death of PCa cells, indicating that SOX4 might be a therapeutic target for PCa [46,47]. Interestingly, SOX11 was recently reported to act as a tumor suppressor in PCa, since its overexpression suppressed the migration and invasion of PCa cells [48,49]. In addition, SOX9, also known as a soxE member, was overactivated in PCa cells and its downregulation inhibited tumorsphere formation in androgen-deficient hosts [50]. In our study, we found that SOX8, another soxE member, was highly expressed in both CRPC tissues and Enza-R cells and that SOX8 was associated with a worse prognosis of CRPC patients. Reducing the expression of this enzyme significantly inhibited malignant biological behaviors of Enza-R cells and reversed enzalutamide resistance, suggesting that SOX8 may be a potential target for CRPC therapy.
The Notch receptor members are recognized as an oncogene in various tumors, including PCa. Notch1 was found to be overactivated in PCa, while its inhibition by a γ-secretase inhibitor restored enzalutamide function [24,26,27,51]. Also, inhibition of Notch2 activation by GSI-1, another γsecretase inhibitor, decreased the cell survival of prostate cells and promoted their apoptosis [24,52]. A recent study reported that Notch3 is responsible for PCa-induced bone lesion by activating MMP-3 signaling [53]. Another study revealed that hypoxia triggers the activation of Notch3, which, in turn, sustains the survival and proliferation of PCa cells in vivo [54]. Notch4 is involved in the progression of PCa given that Notch4 ablation inhibits PCa growth and EMT via the NF-κB pathway [55]. Here, we discovered that Notch4 is highly expressed in CRPC tissues and associated with a poorer prognosis of CRPC patients. Our previous studies indicate that Notch receptors 1, 2, 3, and 4 have no  statistically significant relationship with bone metastasis [20]. However, the present study found that Notch4 instead of Notch1, 2, and 3 was significantly correlated with bone metastasis (data not show). It can be explained that we have added research center and obtained some new samples from it. Moreover, Notch4 reduction, achieved by knocking down SOX8 and/or treatment with RO04929097, significantly inhibited the survival and growth of Enza-R cells and restored enzalutamide sensitivity in Enza-R cells, suggesting that interfering with the Notch/SOX8 axis may be a potential target for the treatment of CRPC.

Conclusion
Our data indicated that overactivated SOX8, Notch4, and Hes5 predict more susceptibility to bone metastases and shorter PFS in CRPC tissues. Furthermore, the SOX8/ Notch4 signaling axis is responsible for enzalutamide resistance, and knocking down SOX8 may be a novel strategy for the treatment of CRPC. Importantly, the pharmacological inhibition of Notch signaling by RO04929097 may be a promising therapeutic strategy for CRPC.

Data Availability
All datasets of this article are included within the article. More supporting data is available under reasonable request.

Ethical Approval
The study protocol was approved by the Human Ethics Review Committee of the First Affiliated Hospital of Chongqing Medical University and Ethics Committee of the Affiliated Hospital of North Sichuan Medical College. This study conforms to the provisions of the Declaration of Helsinki. Animal studies were performed according to the Institutional Animal Care and Use Committee of Chongqing Medical University.