Inhibition of cathepsin K sensitizes oxaliplatin-induced apoptotic cell death by Bax upregulation through OTUB1-mediated p53 stabilization in vitro and in vivo

Cathepsin K is highly expressed in various types of cancers. However, the effect of cathepsin K inhibition in cancer cells is not well characterized. Here, cathepsin K inhibitor (odanacatib; ODN) and knockdown of cathepsin K (siRNA) enhanced oxaliplatin-induced apoptosis in multiple cancer cells through Bax upregulation. Bax knockdown significantly inhibited the combined ODN and oxaliplatin treatment-induced apoptotic cell death. Stabilization of p53 by ODN played a critical role in upregulating Bax expression at the transcriptional level. Casein kinase 2 (CK2)-dependent phosphorylation of OTUB1 at Ser16 played a critical role in ODN- and cathepsin K siRNA-mediated p53 stabilization. Interestingly, ODN-induced p53 and Bax upregulation were modulated by the production of mitochondrial reactive oxygen species (ROS). Mitochondrial ROS scavengers prevented OTUB1-mediated p53 stabilization and Bax upregulation by ODN. These in vitro results were confirmed by in mouse xenograft model, combined treatment with ODN and oxaliplatin significantly reduced tumor size and induced Bax upregulation. Furthermore, human renal clear carcinoma (RCC) tissues revealed a strong correlation between phosphorylation of OTUB1(Ser16) and p53/Bax expression. Our results demonstrate that cathepsin K inhibition enhances oxaliplatin-induced apoptosis by increasing OTUB1 phosphorylation via CK2 activation, thereby promoting p53 stabilization, and hence upregulating Bax.


INTRODUCTION
Cathepsin K is one of the major proteases in the lysosomal cysteine protease family. Cathepsin K has functional roles in multiple physiological processes, such as MHC-II-mediated antigen presentation, bone resorption, and keratinocyte differentiation [1,2]. Cathepsin K expression is elevated in cancer cells [3]. Cathepsin K acts in cancer progression and invasion by indirectly or directly degrading extracellular matrix proteins [3]. Cathepsin K inhibitors have been proposed for the treatment and prevention of bone cancer and bone metastases [4]. Odanacatib (ODN) is a small-molecule selective cathepsin K inhibitor that prevents binding to its substrates. Combined treatment with SM934 (a novel water-soluble artemisinin analog) and testosterone inhibited proliferation and metastasis of cancers by inhibiting cathepsin K expression, which in turn inhibited Bcl-xL [5]. Recently, ODN has been found to enhance TNF-related apoptosis-inducing ligand (TRAIL) sensitivity via ubiquitin-specific peptidase 27x (USP27x)mediated Bim upregulation [6]. ODN induces proteasomedependent degradation of regulatory associated protein of mammalian target of rapamycin (Raptor), followed by production of mitochondrial ROS, which has a critical role in USP27x-mediated Bim stabilization [6]. Furthermore, cathepsin K overexpression promotes cell proliferation, migration, and invasion in non-small cell lung cancer [7], and cathepsin K inhibition prevents the establishment and progression of prostate cancer in bone [8].
Oxaliplatin, a platinum compound, is one of the most heavily utilized chemotherapeutic agents for colorectal cancer [9,10]. Oxaliplatin exerts its cytotoxic effect chiefly by inducing DNA strand breakage and inhibiting DNA synthesis and repair, ultimately leading to apoptosis. However, the anticancer effect of oxaliplatin is limited by intrinsic or acquired drug resistance. Oxaliplatin has also been administered in combination therapy as an effective strategy for reducing adverse side effects and resistance in various types of cancers. Further study is required to understand oxaliplatin's mechanisms of cytotoxicity and to identify chemical reagents that improve its anticancer effects. This will lead to more effective anticancer drug treatment strategies.
Therefore, our objectives were to understand how ODN affects oxaliplatin-induced apoptosis and to identify the molecular mechanisms by which combined oxaliplatin and ODN treatment induce apoptosis in human renal carcinoma cells.
Loss of mitochondrial membrane potential (MMP) plays a critical role in apoptosis, by releasing cytochrome c into the cytoplasm [11]. Therefore, we examined the effect of the combined treatment on MMP. The combined treatment induced a reduction in MMP (Fig. 1G) and an increase in cytosolic cytochrome c levels (Fig. 1H). To verify the involvement of cathepsin K inhibition in sensitization to oxaliplatin-induced apoptosis, we examined apoptosis using cathepsin K siRNA. Oxaliplatin alone markedly increased the sub-G1 population and induced PARP cleavage in cathepsin K siRNA transfected all p53 wild-type (WT) cancer cells, including Caki-1, ACHN, U87MG, and MCF7 cells (Fig. 1I). Therefore, our data indicate that cathepsin K inhibition enhanced oxaliplatin-mediated apoptosis in multiple cancer cells.
Upregulation of Bax expression plays a critical role in combined treatment with a cathepsin K inhibitor and in oxaliplatin-induced apoptosis To elucidate the molecular mechanisms leading to apoptosis in ODN-treated cells, we analyzed the regulation of apoptosis-related protein expression. The expression of the apoptosis-related proteins that we studied was not altered by ODN treatment in Caki-1 cells ( Fig. 2A). However, ODN significantly upregulated Bax, a proapoptotic protein. Further, ODN-induced Bax upregulation in ACHN, U87MG, and MCF7 cells, but not normal MC and normal TCMK-1 cells (Fig. 2B, C). ODN increased Bax mRNA expression and its promoter activity in cancer cells (Fig. 2D). Since Bax activation plays a critical role in the release of cytochrome c from the mitochondria [26], we examined Bax activation using conformation-specific anti-Bax antibodies (6A7) in ODN-plusoxaliplatin-treated cells. Combined treatment markedly induced Bax activation and Bax oligomerization (Fig. 2E, F). We subsequently investigated the functional importance of Bax. Bax knockdown via siRNA markedly inhibited ODN-plus-oxaliplatininduced apoptosis and PARP cleavage (Fig. 2G). Therefore, our findings show that upregulation of Bax expression is associated with ODN-plus-oxaliplatin-induced apoptosis.

P53 upregulation plays a critical role in ODN-induced Bax expression
To identify the molecular mechanism of Bax upregulation, we investigated the expression of p53, a key transcriptional factor of Fig. 1 Cathepsin K inhibition sensitizes oxaliplatin-mediated apoptosis. A-E Cancer or normal cell lines were treated with a 25 μM oxaliplatin or/and 2 μM odanacatib (ODN) for 24 h. Quantification of DNA fragments was determined using a DNA fragmentation assay kit (D). Detection of caspase activity was measured using a DEVDase colorimetric assay kit (E). F Caki-1 cells were treated with a combination of 2 μM ODN and 25 μM oxaliplatin in the presence or absence of a pan-caspase inhibitor, 20 μM z-VAD-fmk (z-VAD), for 24 h. G, H Caki-1 cells were treated with a combination of 2 μM ODN and 25 μM oxaliplatin for the indicated times. Flow cytometry was used to detect fluorescence intensity to measure MMP, using rhodamine 123 fluorescent dye (G). Cytochrome c release was analyzed by cytoplasmic fraction. MnSOD was used as a mitochondrial fraction marker (H). I The cancer cell lines were transfected with control siRNA or cathepsin K siRNA and were treated with 25 μM oxaliplatin for 24 h. Apoptosis and protein expression were measured by flow cytometry (A-C, F, and I) and western blotting (A, B, F, H and I). Cell morphology was assessed using a microscope; scale bar: 50 µm (C). The values in the graphs A-G, and I represent the mean ± SD of three independent experiments. *P < 0.01 compared to the control. # P < 0.01 compared to the ODN-plus-oxaliplatin combination. **P < 0.01 compared to the cathepsin K siRNA-transfected cells treated with oxaliplatin.
Bax [12]. ODN increased p53 protein expression within 3 h, in all of the p53 WT cancer cell lines that we tested (Fig. 3A). Cathepsin K knockdown via siRNA also upregulated p53 and Bax expression in p53 WT cancer cells (Fig. 3B). Furthermore, ODN did not increase Bax promoter activity and Bax expression in HCT116 p53-null cells (Fig. 3C, D). However, cotransfection with the Bax-promoter construct and p53-expression plasmid increased Bax promoter activity and expression in p53-null cell lines (Fig. 3E). Transfection of p53 markedly increased combined ODN and oxaliplatininduced apoptosis, PARP cleavage, and Bax upregulation in p53null NCI-H1299 and SaOS-2 cells, but not in vector-transfected cells (Fig. 3F). In addition, siRNA-mediated p53 knockdown completely blocked ODN-induced Bax expression (Fig. 3G) and abolished ODN-plus-oxaliplatin-induced apoptosis and PARP cleavage in p53 WT Caki-1 cells (Fig. 3H). Taken together, these results indicate that p53 upregulation induced by ODN is involved in Bax upregulation, thereby contributing to the sensitization to oxaliplatin-mediated apoptosis in p53 WT cancer cells.
ODN induces p53 upregulation via OTUB1 phosphorylation Next, we investigated the mechanism whereby ODN induces p53 expression. p53 WT Caki-1 cells were treated with or without ODN in the presence of cycloheximide (CHX). Including ODN significantly enhanced p53 stabilization compared with using CHX alone (Fig. S1A). Because p53 stability is mainly regulated by the MDM2 E3 ligase [13], we investigated the effect of ODN on MDM2 expression. ODN did not alter MDM2 expression (Fig. S1B). Furthermore, Nutlin-3 (a small-molecule MDM2 antagonist) and MDM2 siRNA did not affect ODN-mediated p53 and Bax expression (Fig. S1C, D). Therefore, these findings suggest that ODN-induced p53 stabilization is not associated with MDM2.

Cotreatment with ODN and oxaliplatin suppresses tumor growth in vivo
To verify the synergistic effect of ODN and oxaliplatin in an in vivo p53 WT HCT116 xenograft model, single (ODN or oxaliplatin) or combined treatment (ODN plus oxaliplatin) in tumor-bearing mice were applied. Single treatment showed a weak inhibitory effect on tumor volume, whereas combined treatment markedly reduced tumor volume (Fig. 8A). However, body weight did not change in combined treatment (Fig. 8B). Furthermore, TUNEL-positive signals were detected in combination treatment (Fig. 8C). We checked the related protein levels to ODN or/and oxaliplatin using in vivo samples. Similar to our in vitro results, ODN or ODN plus oxaliplatin increased Ser16 phosphorylation of OTUB1, p53, and Bax expression, whereas decreased caspase-3 expression (Fig. 8D). These data implied that combined treatment ODN and the anticancer drug has synergistic effects in the in vivo.

Phosphorylation of OTUB1 correlates with p53 and Bax in human RCC patient tissues
We collected 38 specimens of human renal clear carcinoma (RCC) tissues and analyzed related proteins expression. The results revealed that OTUB1 Ser16 phosphorylation, p53, and Bax were upregulated in RCC tissues (Fig. 9A, B). The expression levels of three proteins were quantified in all samples, 84.2% (32/38) of OTUB1 Ser16 phosphorylation, 86.8% (33/38) of p53, and 81.6% (31/38) of Bax were significantly higher in tumor tissues compared to adjacent normal tissue, respectively (Fig. 9B). Moreover, Ser16 phosphorylation of OTUB1 has positive relationship with p53 and Bax (Fig. 9C). Collectively, our findings indicate that the mitochondrial ROS/CK2/OTUB1 phosphorylation/p53/Bax axis signaling pathway plays a critical role in ODN-plus-oxaliplatininduced apoptosis.

DISCUSSION
In this study, we have demonstrated that cathepsin K inhibition enhanced oxaliplatin-induced apoptosis in p53 WT cancer cells. Both pharmacological inhibitions via ODN and genetic ablation of cathepsin K upregulated Bax, a proapoptotic protein.
We found that both specific inhibition and knockdown of cathepsin K increased Bax expression in multiple cancer cells lines (Figs. 2B and 3B). ODN-mediated Bax upregulation was modulated at the transcriptional level by p53 (Fig. 3C, D), but was not detected in p53-null cancer cell lines (Fig. 3C-F). Several studies have reported that Bax expression is regulated by p53-dependent transcriptional activation, occurring at two p53 half-sites plus an adjacent six base pairs (5′-GGGCGT-3′) [12,17]. We also investigated whether ODN upregulates other p53-target proteins, such as p21, PUMA, Noxa, and DR5. However, ODN did not affect the expression of these proteins (Fig. S2). Based on a comparative analysis of p53 targets genes, it is known that Bax gene transcription is differentially regulated by p53 [18][19][20]. In our study, ODN stabilized p53 to induce Bax expression. Further, MDM2, a major E3 ligase of p53 [13,21], was not involved in p53 stabilization (Fig. S1B-D). Therefore, we focused on the role of deubiquitinases in regulating p53 activity. OTUB1 had a critical role in p53 stabilization (Fig. 4A). OTUB1 performs both canonical and non-canonical deubiquitination [22]. Sun et al. have shown that OTUB1 directly suppresses MDM2-mediated p53 ubiquitination. Ectopic expression of OTUB1 stabilizes and activates p53, leading to apoptosis and inhibition of cell growth in a p53dependent manner [23]. Wiener et al. reported that deleting the first 45 OTUB1 residues makes it unable to non-canonically inhibit E2 enzymes [24,25]. Consistent with this observation, the N-terminal deletion mutant (Δ1-45) abolished ODN-induced p53 stabilization (Fig. 4D). Interestingly, ODN markedly increased phosphorylation of OTUB1 at Ser16 but did not increase OTUB1 expression (Fig. 4B). CK2 is known to regulate the function of OTUB1 via OTUB1 phosphorylation at Ser16 [15]. In our study, CK2 inhibition and knockdown also induced inhibition of OTUB1 phosphorylation and increased p53 stabilization (Fig. 5A, B). Therefore, CK2-induced OTUB1 phosphorylation is critical for p53 stabilization and nuclear localization in ODN-induced Bax upregulation.
ROS generation is tightly controlled, because an imbalance in redox reactions can impair cellular function. Cathepsin deficiency induces ROS production and mitochondrial dysfunction. Proteome analysis of cathepsin L-deficient myocardium revealed a significant reduction in the levels of respiratory chain components, compared with levels in the WT [26]. The use of cathepsin E-deficient macrophages revealed augmented ROS production and upregulation of oxidized peroxiredoxin-6, but reduced levels of the antioxidant glutathione [27]. Cathepsin S inhibition increases intracellular ROS levels by inducing mitochondrial dysfunction [28]. Recently, we reported that ODN induces mitochondrial fusion and mitochondrial ROS generation via raptor downregulation [6]. ODN-induced mitochondrial ROS are also tested by using pHyPer-dMito vector in this study (Fig. 6B). Also,  ODN reduced OXPHOS complex I and II expression (Fig. 6D). Mitochondrial ROS blockers (Mito-TEMPO and MnTMPyp) inhibited ODN-induced phosphorylation of OTUB1 at Ser16 and upregulation of p53 and Bax (Fig. 7C). Mitochondrial ROS may be involved in regulating CK2 activation. Neither pharmacological inhibition nor genetic ablation of CK2 affected ODN-induced mitochondrial ROS production (Fig. 7E, F). Our findings suggest that mitochondrial ROS function as a signaling trigger upstream of CK2. However, further study is required to elucidate how mitochondrial ROS modulate CK2 activity.
In conclusion, our findings suggest that cathepsin K inhibition enhances oxaliplatin-induced apoptosis by upregulating Bax, a proapoptotic protein. Mitochondrial ROS production induced phosphorylation of OTUB1 at Ser16 via CK2 activation; phosphorylated OTUB1, in turn, increased p53 stabilization, resulting in Bax upregulation at the transcriptional level. Therefore, the combination of cathepsin K inhibition and anticancer drugs may provide a novel and effective strategy for cancer therapy.

Flow cytometry analysis
Harvested cells were resuspended in 100 µl phosphate-buffered saline (PBS) and fixed using 200 µl of 95% ethanol at 4°C. After 1 h, the cells were washed with PBS, resuspended in 1.12% sodium citrate buffer (pH 8.4) containing 12.5 µg RNase, and incubated at 37°C for 30 min. Then, 250 µl of propidium iodide solution (50 µg/ml) was added to the cells, followed by incubation at 37°C for 30 min. The number of apoptotic cells was measured using a BD Accuri TM C6 flow cytometer (BD Biosciences, San Jose, CA).

Western blot analysis
Cells were lysed using RIPA buffer containing a protease inhibitor [29]. Cell lysates were collected by centrifugation at 13,000 × g for 15 min at 4°C. Protein samples were separated using SDS-PAGE and transferred onto nitrocellulose membranes (GE Healthcare Life Science, Pittsburgh, PA). Protein bands were detected using an enhanced chemiluminescence kit (EMD Millipore, Darmstadt, Germany).

DNA fragmentation and Asp-Glu-Val-Asp-ase (DEVDase) activity assay
The amount of fragmented DNA was detected using the Cell Death Detection ELISA Plus kit (Roche, Basel, Switzerland) [6]. For the analysis of caspase-3 activation, 20 µg of cell lysates were incubated with an acetyl-Asp-Glu-Val-Asp p-nitroanilide (Ac-DEVD-pNA) for 2 h at 37°C and measured at 405 nm absorbance using a spectrophotometer [6]. The values in the graphs D-F represent the mean ± SD of three independent experiments. *P < 0.01 compared to the ODN-plus-oxaliplatin treatment.
Detection of mitochondrial membrane potential and cytochrome c release Mitochondrial membrane potential (MMP) was analyzed using a 5 µM rhodamine 123 dye (Molecular Probes Inc., Eugene, OR), and fluorescence was measured by BD Accuri TM C6 flow cytometer (BD Biosciences). For the analysis of cytochrome c release [30], cell lysates were centrifuged at 13,000 × g for 15 min at 4°C. The supernatants (cytosolic extract) and the pellets (mitochondrial extract) were collected.  The amount of cytochrome c released into the cytoplasm was analyzed using western blotting.

Reverse transcription PCR
Reverse transcription PCR (RT-PCR) and quantitative PCR were performed as previously described [31]. The primer sequences used are described in Supplementary Table 1.

Promoter activity measurement
Cells were transfected with the Bax/−600 luciferase promoter, using Lipofector p-MAX (AptaBio, Korea), and were treated with ODN for 12 h. The lysates were then incubated with the luciferase substrate, luciferin (Promega, Madison, WI).

Bax-activation analysis
Cells were harvested and fixed by adding 4 % paraformaldehyde for 30 min. Cells were incubated with the Bax (6A7) antibody in PBS / 1% FCS / 0.1% saponin for 1 h at 4°C. The cells were then washed using PBS / 1% FCS, and incubated with secondary antibody in PBS / 1% FCS / 0.1% saponin for 1 h at 4°C. Bax activation and oligomerization were the same as described previously [30].

Deubiquitination assay
This assay was performed using a tagged-ubiquitin plasmid and pretreatment with MG132, as previously reported [32]. Briefly, the cells were cotransfected with HA-tagged ubiquitin (HA-Ub) and Flag-OTUB1 wild-type (WT), Flag-OTUB1 Δ1-45, and Flag-OTUB1 S16A plasmids, and then treated with MG132 for 12 h. Immunoprecipitation was performed using an anti-p53 antibody. Ubiquitination of endogenous p53 was identified using HRP-conjugated anti-Ub under denaturation conditions.

ROS production assay
After treatment, the cells were incubated with H 2 DCF-DA or MitoSOX Red (Thermo Fisher Scientific, Waltham, MA) for 10 min at 37°C, then harvested and resuspended in PBS. To examine source of ROS, we transfected using pHyPer-cyto, pHyPer-nuc, or pHyPer-dMito vector (Evrogen, Moscow, Russia) in the cells. ROS production through fluorescence was measured using BD Accuri™ C6 flow cytometer (BD Biosciences) or fluorescence microscope.

Xenograft model
Male BALB (bagg and albino)/c-nude mice were purchased from the JA BIO, Inc. (Suwon, Korea), and mice were maintained in a pathogen-free conditioned room. The IRB Keimyung University Ethics Committee (KM-2020-03R2) approved our study protocol. HCT116 cells were subcutaneously injected on each flank of mice. After 3 weeks, mice were randomly divided into four groups and injected intraperitoneal (i.p.) of 5 mg/kg ODN (in 2% DMSO/PBS) or 5 mg/kg oxaliplatin (PBS) three times a week. The tumor size (length × width 2 )/2 was measured every time using a Vernier's caliper (Mitutoyo Co., Tokyo, Japan). Apoptosis was detected ApopTag Fluorescein in situ Apoptosis Detection Kit (Millipore).

Patient specimens
A total of 38 patients diagnosed with RCC were included in this retrospective study. RCC tissues were collected from patients undergoing surgery in Keimyung University Dongsan Medical Center (Daegu, Korea). Tissue samples were immediately frozen in liquid nitrogen and stored at −196°C until Western blotting analysis. Tissue samples were provided by the Biobank of Keimyung University Dongsan Hospital Biobank (IRB-2019-11-040).

Statistical analysis
Statistical analyses were performed using Statistical Package for Social Science (SPSS, Version 26.0; IBM SPSS, Armonk, NY, USA). All experiments were repeated three times or more. The data are presented as means. The data were analyzed using a one-way ANOVA and post-hoc comparisons (Student-Newman-Keuls). We determined the sample size based on the smallest effect we wished to measure. The association between the relative expression levels of p-OTUB1 (S16), p53, and Bax has assessed the Pearson's correlation coefficients for continuous variables. R is Pearson's correlation coefficient value. P < 0.05 was considered statistically significant.