Potent USP10/13 antagonist spautin‐1 suppresses melanoma growth via ROS‐mediated DNA damage and exhibits synergy with cisplatin

Abstract Malignant melanoma is one of the most invasive tumours. However, effective therapeutic strategies are limited, and overall survival rates remain low. By utilizing transcriptomic profiling, tissue array and molecular biology, we revealed that two key ubiquitin‐specific proteases (USPs), ubiquitin‐specific peptidase10 (USP10) and ubiquitin‐specific peptidase10 (USP13), were significantly elevated in melanoma at the mRNA and protein levels. Spautin‐1 has been reported as a USP10 and USP13 antagonist, and we demonstrated that spautin‐1 has potent anti‐tumour effects as reflected by MTS and the colony formation assays in various melanoma cell lines without cytotoxic effects in HaCaT and JB6 cell lines. Mechanistically, we identified apoptosis and ROS‐mediated DNA damage as critical mechanisms underlying the spautin‐1‐mediated anti‐tumour effect by utilizing transcriptomics, qRT‐PCR validation, flow cytometry, Western blotting and immunofluorescence staining. Importantly, by screening spautin‐1 with targeted or chemotherapeutic drugs, we showed that spautin‐1 exhibited synergy with cisplatin in the treatment of melanoma. Pre‐clinically, we demonstrated that spautin‐1 significantly attenuated tumour growth in a cell line‐derived xenograft mouse model, and its anti‐tumour effect was further enhanced by cotreatment with cisplatin. Taken together, our study revealed a novel molecular mechanism of spautin‐1 effecting in melanoma and identified a potential therapeutic strategy in treatment of melanoma patients.


| INTRODUC TI ON
Malignant melanoma, a solid tumour produced by the malignant transformation of melanocytes from the skin and other organs, has an increasing incidence and a poor prognosis. 1 Compared with other tumours originating from the epidermis and accessory organs, the cutaneous form of the disease is common in the Western world and is late onset with no obvious early symptoms, which causes the three-quarters of the deaths related to skin cancer. 2 The early stage of melanoma has a good prognosis after sufficient surgical treatment. However, the advanced stage of melanoma is still a therapeutic challenge despite the development of multiple new therapeutic methods. Although chemotherapeutics or targeted therapeutics alter tumour growth, in most cases, the effect is not long-lasting resulting in drug resistance that causes treatment failure and impacts the survival of patients. 3,4 New immunotherapies have been recently approved to induce long-lasting responses in patients with metastatic cancers, even so, resistance and poor response rates still persist. 5 As melanoma eventually becomes resistant to these novel therapies, the development of more effective approaches is required.
Ubiquitination is a post-translational protein modification that is regulated by a series of ubiquitination-associated enzymes, including the ubiquitin-activating enzyme (E1), the ubiquitin-conjugating enzyme (E2), the ubiquitin ligase (E3) and deubiquitinases (DUBs). 6 Ubiquitination spans wide-spectrum functions in the cell that include cell death, DNA damage repair and protein degradation. 7 Thus, dysregulation of ubiquitination leads to an imbalance between promoting and suppressing pathways of tumours. Recently, accumulative evidence has established the critical role of ubiquitination in cancer pathogenesis and has therefore revealed a potential therapeutic role in various cancers. In particular, the advent of some new inhibitors revealed suppressive effect in melanoma. 8,9 Ubiquitin-specific pep-tidase10 (USP10) and ubiquitin-specific peptidase 13 (USP13) both belong to USP family, the largest subfamily of DUB. 10 USP10 and USP13 are involved in a number of biological processes and impact the development of various tumours by stabilizing several proteins.
USP13 and USP10 usually function as functional partners to regulate the stability of several proteins, such as p53. 11,12 Whether targeting these two molecules can inhibit the melanoma or not is still unclear.
In this study, we employed a potent USP10/13 deubiquitinating activity antagonist, spautin-1, to investigate the effect of targeting USP10/USP13 as an anti-melanoma treatment and found that melanoma cell proliferation was significantly suppressed by spautin-1-induced ROS-mediated DNA damage. Besides, spautin-1 treatment could combine with cisplatin, a well-known chemotherapeutic drug, which was shown to enhance the anticancer effect of cisplatin both in vivo and in vitro.

| NCBI GEO data analysis
To identify the expression of USP10 and USP13 in malignant melanoma, data were obtained from the GEO database. The raw data set GSE3189 was downloaded, which provided the USP10 and USP13 miRNA profiles including 7 normal skin samples, 18 naevus samples and 45 melanoma samples. These data were analysed based on the GPL96 platform.

| Cell lines and culture
The human malignant melanoma cell lines (A375 and SK-Mel-28), the human keratinocyte HaCaT and the mouse epidermis-derived JB6 Cl 41-5a cells were purchased from the American Type Culture Collection (ATCC). All cells were grown in Dulbecco's modified Eagle's medium (BI) supplemented with 10% foetal bovine serum (BI) at 37°C in a 5% CO 2 humidified incubator. a novel molecular mechanism of spautin-1 effecting in melanoma and identified a potential therapeutic strategy in treatment of melanoma patients.

K E Y W O R D S
combination therapy, DNA damage, malignant melanoma, ROS, spautin-1 F I G U R E 1 USP10 and USP13 mRNA and protein levels were elevated in tumour tissue of melanoma patients and in melanoma human cell lines. A and B, USP10 (A) and USP13 (B) expression in normal skin (n = 7), naevus (n = 18) and malignant melanoma (n = 45) in the GSE3189 data sets. A, USP10 mRNA expression in the GSE3189 data sets. Significant differences were evaluated using Student's t test. *P < .05 vs malignant melanoma. B, USP13 mRNA expression in the GSE3189 data sets. Significant differences were evaluated using Student's t test. *P < .05 vs malignant melanoma. C and D, Immunohistochemical staining of USP10 (C) and USP13 (D) in normal naevi tissues and malignant melanoma. Right panel: quantification of USP10 and USP13 in naevi and melanoma tissues. Significant differences were evaluated using Student's t test. *P < .05 vs naevus. E and F, The protein level of USP10 (E) and USP13 (F) in the human skin keratinocytes cell line (HaCat) and different melanoma cell lines were measured by Western blot (upper panel). Semi-quantitative analysis of USP10 and USP13 proteins expression compared with GAPDH. (Lower panel) (mean values ± SEM, n = 3) Significant differences were evaluated using Student's t test. *P < .05 vs HaCat | 4327 GUO et al.
After 48 hours, the drug-containing medium was replaced with complete growth medium for two weeks until the visible colony formation. During the process, the medium was refreshed every three days. Finally, the cells were washed with PBS, fixed with 4% paraformaldehyde (Servicebio) and stained with 0.5% crystal violet (DingGuo). At the end-point, the images were captured.

| Cell transfection
Negative control small interfering siRNA (si-NC), USP10 small interfering RNA (si-USP10) and USP13 small interfering RNA (si-USP13) were constructed by Gene Pharma. For transfection experiments, A375 and Sk-Mel-28 cells (2 × 10 5 per well) were seeded in a 6-well plate to allow attachment and culture for 24 hours, grown to 70%-80% confluence the day of transfection. The cells were cotransfected with different oligonucleotides using TurboFect (Thermo Scientific) following the manufacturer's protocol. Cells were used to subsequent experiments after transfection for 48 hours.

| Cell apoptosis and cell cycle assay
Apoptosis and cell cycle distribution were detected by flow cytometer. For the cell cycle assays, cells treated with various concentrations of spautin-1 (0, 2.5, 10 μmol/L) were trypsinized and washed with cold PBS. Then, the collected cells were suspended in 75% ethanol at 4°C overnight. The next day, the cells were stained with 0.5 mL of propidium iodide (PI) staining (Beyotime Biotechnology) according to the manufacturer's instructions. The data were analysed using ModFit software. For the cell apoptosis, the treated cells were digested by trypsin solution without EDTA (Beyotime Biotechnology), washed with PBS, and incubated with 5 µL annexin V staining for minutes at room temperature, and then stained with 10 µL of propidium iodide (BD Biosciences) before being detected. The samples were run on a DxP cytometer (Cytek), and the data were analysed by FlowJo 10 software.

| Western blotting
Cells were harvested in RIPA Lysis Buffer (DingGuo) with a protease inhibitor and phosphatase inhibitors (Selleck), and protein concentration was determined using a BCA assay kit (Beyotime). Protein samples (30 μg) were subjected to 8%-12% SDS-polyacrylamide gel electrophoresis (SDS-PAGE) and then transferred onto polyvinylidene fluoride membranes (Millipore). The membranes were blocked with 5% non-fat milk or 5% bovine serum albumin (BSA) for one hour at room temperature and then incubated with following primary and second-

| Immunofluorescence assay
Cells (2 × 10 5 per well) were seeded on coverslips in a 6-well plate to allow attachment and then exposed to spautin-1 (10 μmol/L) for 0, 24 or 48 hours. Cells were fixed in 4% paraformaldehyde for 20 minutes and permeabilized with 0.5% Triton X-100 for 1 hour.

| Mitochondrial membrane potential detection
Cells (2 × 10 5 per well) were seeded on coverslips in a 6-well plate to allow attachment and then exposed to spautin-1 (10 μmol/L) for 0, 6, 12 or 24 hours. Then, the cells were incubated with 1 mL of JC-1 working solution for 30 minutes according to the manufacturer's instruction (Beyotime Biotechnology). After being washed with JC-1 staining buffer twice, the cells were observed and captured by fluorescence microscopy (Ts2R, Nikon).

| Measurement of ROS generation
The effect of spautin-1 on intracellular ROS generation was measured using a DCFH-DA reactive oxygen species ROS fluorescence probe (Solarbio). Cells (3 × 10 5 per well) were seeded in a 6-well plate to allow attachment and then exposed to spautin-1 for 0, 3, 6 or 12 hours. Then, the cells were incubated with DCFH-DA for 30 minutes according to the manufacturer's instruction. The intracellular ROS generation was measured by a DxP Athena cytometer (Cytek), and the data were analysed by the FlowJo10 software.

| Statistical analysis of drug interaction
The intensity and properties of the interaction of spautin-1 and cisplatin could be quantitatively determined by the combination index (CI) of the Chou-Talalay method using the Calcusyn version 2.0. A CI value >1 was antagonistic, equal to 1 was considered to be additive and <1 was synergistic. were calculated by the formula (length × width ×height × π/6). At the end of the experiment, the tumour-bearing mice were sacrificed, and the xenografts were removed for histology and further analysis.

| Immunohistochemistry
The tumour tissue was made into sections. After a series of dewaxing and gradient alcohol dehydration, antigen repair was performed in a microwave. The slices were cooled to room temperature and washed 3 times with PBS. An immunohistochemical pen was used to draw a circle, and peroxidase was blocked for 10 minutes. The primary antibodies against Ki67 (1:400, Abcam) and γ-H2AX (1:200, CST) were applied to the slides which were incubated in a humidified chamber at 4℃ overnight. The next day, the slides were incubated with the secondary antibody for 30 minutes, and then the reaction was visualized by DAB staining. After that, the slides were counterstained with haematoxylin. Finally, after a series of dehydrations, the sections were sealed with cover glass and neutral resin.

| RNA-sequencing analysis
RNA-seq was completed on the BGISEQ-500 platform (BGI). After the quality control was adjusted, the quantitative analysis of the gene was carried out based on the analysis of the gene expression level (principal component, correlation, differential gene screening, etc). The differentially expressed genes between samples were analysed by Gene Ontology (GO), KEGG pathway, enrichment cluster, protein interaction network, and other in-depth analyses according to BGI's instructions.

| Quantitative real-time PCR
Total RNA was extracted from cells using TRIzol reagent (Bioteke  Table S1. F I G U R E 3 Transcriptomics coupled with qRT-PCR identified apoptosis and DNA damage as potential mechanisms underlying spautin-1-mediated anti-tumour effect. A, Sk-Mel-28 cells were treated with 10 µmol/L Spautin-1 for 48 h. Cluster analysis of the genes expressed in each comparison group, displaying with a heat map. B, The pathways correlate with the differential expression genes was analysed by enriched KEGG pathway. The enriched bubble chart shows the relative top 20 enriched pathways. C, Gene Ontology analysis of the relative differential expression genes. Only top 20 enriched pathways are shown in the bubble chart. D and E, A375 (D) and Sk-Mel-28 (E) were treated with various dosages of spautin-1 for 48 h. The key differential expression genes related to cell cycle and DNA damage were validated by qRT-PCR. (Mean values ± SEM, n = 3) Significant differences were evaluated using a one-way ANOVA. *P < .05 vs control

| Statistical analysis methods
Data are presented as the mean ± SEM. Statistical analysis was performed using GraphPad Prism software (version 6.01). Differences between the means of 2 groups were compared by Student's t test, and the differences of multiple groups were compared by one-way ANOVA, followed by a Tukey multiple-comparisons test. P < .05 was considered significant.

| RE SULTS
3.1 | USP10 and USP13 mRNA and protein levels were elevated in melanoma patient samples and various human melanoma cell lines USP10 and USP13 expression was significantly up-regulated in the melanoma tissues compared to naevus or normal skin tissue according to the GSE3189 data sets ( Figure 1A,B). To assess the role of USP10 and USP13 in melanoma progression, we applied melanoma tissue microarray and immunohistochemistry to detect their expression. As shown in Figure 1C,D, USP10 and USP13 were highly expressed in malignant melanoma specimens compared to the naevus tissue. We have checked the expression of USP10 and USP13 in different melanoma cell lines and human skin keratinocytes cell line (HaCaT). We found that there were higher USP10 and USP13 expression in melanoma cell lines compared with control cells. In particular, their expression was relatively higher in A375 and SK-Mel-28 melanoma cell lines compared to Sk-Mel-05 melanoma cell line ( Figure 1E,F). All the results suggested that USP10 and USP13 might play detrimental roles in melanoma.

| Spautin-1, a dual USP10 and USP13 antagonist, inhibits melanoma cell growth in vitro
Spautin-1, as a potential USP10/13 antagonist, was reported to play an anti-tumour effect in various cancers, such as chronic myeloid leukaemia, ovarian cancer and lung cancer. 13 In this study, we inves-

| Transcriptomics coupled with qRT-PCR validation identified a potential mechanism of spautin-1-mediated anti-tumour growth effects
To identify the molecular mechanism of the spautin-1-mediated inhibitory effect on melanoma cell growth, we analysed the global transcriptomic alteration of melanoma cells after treatment with spautin-1. Bioinformatic analysis identified that 699 genes were up-regulated, and 356 genes were down-regulated after spautin-1 treatment for 48 hours ( Figure 3A). Furthermore, the relative differ-  Figure 3B,C. It is suggested that pathways including cell F I G U R E 4 Flow cytometry and Western blot validated that spautin-1 induce G2/M cell cycle arrest and increased cell apoptosis in melanoma cell lines. A, Representative data (Left) and quantitative analysis (Right) of the cell cycle distribution of spautin-1-treated cells. A375 and Sk-Mel-28 cells were treated with spautin-1 at the indicated concentrations for 48 h. Cell cycle profile was detected using flow cytometry, and the data were analysed by ModFit software. (Mean values ± SEM, n = 3) Significant differences were evaluated using a one-way ANOVA. *P < .05 vs control. B, (Left) Representative data of apoptosis of spautin-1-treated cells. A375 and Sk-Mel-28 cells were disposed with different dosages of spautin-1 for 48 h and then cells were stained with annexin V and PI to measure the percentage of apoptotic cells. (Right) The bar graphs showed the percentage of apoptotic cells, early apoptosis (annexin V positive, PI negative) and late apoptosis (annexin V positive, PI positive). (Mean values ± SEM, n = 3) Significant differences were evaluated using a one-way ANOVA. *P < .05 vs control group, # P < .05 vs 2.5 µmol/L. C, Cells were treated with control (DMSO), spautin-1 (2.5 or 10 μmol/L) for 48 h. Western blot was used to analysis the expression of G2/M phase arrest-associated proteins, including CDC2 and Cyclin B1. GAPDH was used as an internal control (upper panel). Semi-quantitative analysis of the proteins expression compared with GAPDH. (Lower panel) (Mean values ± SD, n = 3) Significant differences were evaluated using a one-way ANOVA. *P < .05 vs control group, # P < .05 vs 2.5 µmol/L. D, A375 and Sk-Mel-28 cells were treated with control (DMSO), spautin-1 (2.5 or 10 μmol/L) for 48 h. Western blot assay was then implied to detect the indicated protein level. GAPDH was used as an internal control (upper panel). Semi-quantitative analysis of cleaved-PARP proteins expression compared with PARP, BAX and BCL2 proteins expression compared with GAPDH. (Lower panel) (Mean values ± SEM, n = 3) Significant differences were evaluated using a one-way ANOVA. *P < .05 vs control group, # P < .05 vs 2.5 µmol/L cycle, apoptosis, PI3K-Akt and DNA damage-related pathway were ranked among the top pathways and that those pathways may play an important role in the mechanism of spautin-1 inhibition of the growth of melanoma. Then, qRT-PCR was utilized to further validate the transcriptomic results by quantifying several key genes related

| Spautin-1 induced G2/M cell cycle arrest and cell apoptosis via up-regulation of ROS-mediated DNA damage in melanoma cell lines in vitro
To test whether spautin-1-mediated anti-proliferation in melanoma

| Spautin-1 had a synergistic effect with cisplatin in the treatment of melanoma cells by inducing DNA damage in vitro
To test the clinical relevance and clinical significance of spautin-1 in the treatment of melanoma, we assessed the synergistic effect of spautin-1 and several first-line clinical drugs. Our results showed that there was no synergistic effect of combining spautin-1 with several first-line targeted drugs, such as vemurafenib and trametinib ( Figure S4C-F), respectively. However, analysis of the cell viability revealed that a combination of spautin-1 and cisplatin caused markedly more cell inhibition than spautin-1 or cisplatin alone within F I G U R E 5 Spautin-1 caused ROS-mediated DNA damage in melanoma cells. A, Effects of spautin-1 on ROS generation in A375 and Sk-Mel-28 cells. The cells were treated with 10 μmol/L spautin-1 for 0, 3, 6 or 12 h. Cells were stained with DCF fluorescence probe and the generation of ROS was measured by flow cytometry. The bar graphs showed the relative ROS levels. (Mean values ± SEM, n = 3) *P < .05. B, A375 and Sk-Mel-28 were pre-treated with or without antioxidant NAC (2 mmol/L) for 1 h and then, exposed to spautin-1 (10 μmol/L) or control medium for another 6 h. Cells were stained with DCF fluorescence probe and the generation of ROS was measured by flow cytometry. The bar graphs showed the relative ROS levels. (Mean values ± SEM, n = 3) *P < .05. C, Cells were treated with spautin-1 (control, 2.5 or 10 μmol/L) for 48 h. Effects of spautin-1 on the expression levels of ATM, p-ATM, ATR, p-ATR and γ-H2AX in A375 and Sk-Mel-28 cells were examined by Western blot assay. D, A375 and Sk-Mel-28 were pre-treated with or without antioxidant NAC (2 mmol/L) for 1 h and then exposed to spautin-1 (10 μmol/L) or control medium for another 6 h. Western blot assay was then implied to detect the indicated protein level. E, (Left panel) Representative images of immunofluorescence staining of γ-H2AX in A375 and Sk-Mel-28 cells treated with 10 μmol/L spautin-1 for 0-48 h. (Right panel) Quantitative analysis of γ-H2AX nucleus foci. (Mean values ± SEM, n = 3) Significant differences were evaluated using a one-way ANOVA. *P < .05 vs control a wide range of drug concentration in A375 and SK-Mel-28 cells ( Figure S4A,B). A widely accepted quantitative analysis method for drug combination is the Chou-Talalay method. The properties of drug combination can be quantitatively defined by the value of the combination index (CI), a theorem of Chou-Talalay. It defines CI = 1 as an additive effect, CI < 1 as a synergistic effect and CI > 1 as an antagonistic effect. 18 Our data showed that the combination of spautin-1 and cisplatin at a non-constant ratio achieved CI < 1 in different melanoma cell lines ( Figure 6A), and the particular CI values are shown in Figure 6B. A large number of studies reported that cisplatin attenuated cancer cell growth through inducing DNA damage and apoptosis. [19][20][21] Previous data suggested that spautin-1 might

| Spautin-1 treatment attenuated tumour growth in a cell-derived xenograft mouse model, and its anti-tumour effect was further enhanced by cotreatment with cisplatin
To further investigate the effect of spautin-1 on malignant melanoma in vivo, a cell-derived xenograft melanoma model was utilized. The A375 cell line has been adopted to establish subcutaneous xenograft tumour models. 22 Consistent with our in vitro results, spautin-1 significantly attenuated tumour growth in vivo and further enhanced the sensitivity of tumours to cisplatin compared to the control group, reflected by smaller tumour sizes and tumour weight ( Figure 7A-C). Meanwhile, the bodyweight of tumourbearing mice in all treatment groups had no significant change during this progress, ruling out the toxicity of spautin-1 in vivo ( Figure S4B). Immunohistochemical staining of the tumour tissues showed decreased expression of Ki-67, a well-accepted proliferation marker, in the spautin-1-treated group compared with the control group, with further decreased expression in the combination group compared with the single-drug group ( Figure 7D). In addition, histopathological findings of the tumours showed that the expression of γ-H2AX in the combined drug group was significantly higher than in the single-drug group ( Figure 7D), which was consistent with previous results. Therefore, these results suggested that spautin-1 alone or combined with cisplatin could significantly inhibit the proliferation of malignant melanoma in vivo. Spautin-1, a USP10/USP13 inhibitor, was also reported to be an autophagy inhibitor by enhancing the degradation of beclin-1, a protein required for the initiation of autophagy. 12 It was reported that loss of beclin-1 could lead to DNA damage. 25 Previous study showed that spautin-1 inhibits cell growth and angiogenesis through inactivating the PI3K/AKT pathway in chronic myeloid leukaemia. 26 In particular, it can suppress the growth of ovarian cancer and lung cancer by regulating MCL1 stability. 24 However, its role in malignant melanoma has not been appreciated thus far. In this study, we revealed that spautin-1 is a novel and a safe approach to inhibiting malignant melanoma proliferation.

| D ISCUSS I ON
Mechanistically, by utilizing RNA-sequencing coupled with qRT-PCR, we identified enriched pathways and genes involved in the potential molecular mechanisms of the anti-proliferative role of spautin-1 in melanoma. As such GADD45A and GADD45B were significantly increased after spautin-1 treatment. These two proteins were targets of DNA damage, and they played critical roles in apoptosis and G2/M checkpoint in response to DNA damage. [27][28][29] In addition, BTG2, a cell cycle control gene, initiated the G2/M phase arrest and inhibited cell growth and was induced in response to DNA damage mediated by p53-dependent mechanisms. 30,31 Reactive oxygen species (ROS), such as H 2 O 2 , superoxide anion (O − 2 ) and hypochlorous acid (HOCl), 32 were recognized as a mediator of DNA damage. 33 F I G U R E 6 Spautin-1 had synergistic effect with cisplatin in the treatment of melanoma cells by induction of DNA damage in vitro. A and B, The synergistic effect of spautin-1 in combination with cisplatin on the growth of A375 and SK-Mel-28 cells. Combination index (CI) values were calculated at the drug concentration of spautin-1 (5 μmol/L) plus cisplatin (2 μmol/L), spautin-1 (5 μmol/L) plus cisplatin (4 μmol/L), spautin-1 (5 μmol/L) plus cisplatin (6 μmol/L), spautin-1 (5 μmol/L) plus cisplatin (8 μmol/L), spautin-1 (10 μmol/L) plus cisplatin (2 μmol/L), spautin-1 (2 μmol/L) plus cisplatin (2 μmol/L), spautin-1 (10 μmol/L) plus cisplatin (4 μmol/L), spautin-1 (10 μmol/L) plus cisplatin (6 μmol/L) and spautin-1 (10 μmol/L) plus cisplatin (8 μmol/L) using the Chou-Talalay method. C, A375 and SK-Mel-28 cell lines were treated with DMSO, spautin-1 (5 μmol/L) and/or cisplatin (4 μmol/L) for 48 h. Western blot analysis of γ-H2AX protein expression was performed (Left panel). Semi-quantitative analysis of γ-H2AX protein expression compared with GAPDH (right panel). (Mean values ± SEM, n = 3) Significant differences were evaluated using a one-way ANOVA. *P < .05 Sputin-1 vs control group, # P < .05 cisplatin vs control group, & P < .05 spautin-1 vs spautin-1 + cisplatin, @ P < .05 cisplatin vs spautin-1 + cisplatin The accumulation of high levels of ROS may cause further DNA damage and then induce cell death. [34][35][36] In this study, we found that spautin-1 significantly induced rapid ROS generation. Furthermore, we revealed that spautin-1 caused DNA damage. Previous study reported that the overload of mitochondria ROS resulted in the reduction of mitochondrial membrane potential (MMP) and subsequently caused the cellular apoptosis. [37][38][39] We assessed the changes of the MMP in the spautin-1-treated cells with a JC-1 probe. 40 The cells showed a sharp MMP decline after being treated with spautin-1 ( Figure S5). The cytometry results showed that spautin-1 treatment significantly induced melanoma cell apoptosis ( Figure 4B). We further utilized Western blotting to reveal the molecular mechanism of spautin-1-mediated proapoptotic effect in melanoma. PARP, a well-known DNA strand break-binding enzyme, is cleaved and activated and then causes apoptosis. 41  Cisplatin, as a well-known chemotherapeutic drug, plays an anti-tumour role through a variety of mechanisms, the most important of which is to activate the DNA damage response and induce apoptosis. 43,44 The cisplatin benefits the overall survival (OS) and relapse-free survival (RFS) of advanced melanoma and metastatic melanoma patients. 45 However, single-agent chemotherapy regimens have produced low response rates, and the median response duration is only 4-5 months. 46,47 According to reported studies, cisplatin is used for advanced melanoma in combination with other chemotherapeutic drugs or after targeted drugs. 20,48 Our results showed that spautin-1 inhibited melanoma cell proliferation by ROS-induced DNA damage, and spautin-1 had a synergistic effect with cisplatin, but not vemurafenib or trametinib. Vemurafenib (selective BRAF inhibitors) and trametinib (Inhibitors of the downstream MAP kinase MEK) were targeted inhibitors utilized for the treatment of melanoma. Previous studies reported that the molecular mechanism of vemurafenib-and trametinib-mediated anti-tumour effects was inhibition of MAPK pathway. 49,50 It was reported that elevation of ROS activated MAPK-related pathway. [51][52][53] Therefore we speculated that spautin-1 had no synergistic effect with vemurafenib or trametinib may be due to the opposite effect in terms of regulating MAPK pathways in the treatment of melanoma. Taking together, spautin-1 may be not only a potential therapeutic approach for melanoma cells but also a sensitizer for cells that are resistant to conventional chemotherapeutic agents, especially cisplatin.
Taken together, our study revealed that spautin-1 induced G2/M cell cycle arrest and cell apoptosis via up-regulation of ROSmediated DNA damage and had a synergistic effect with cisplatin ( Figure 7E), which indicated that spautin-1 mediated beneficial effects in melanoma and identified a potential therapeutic strategy in the treatment of melanoma patients.

CO N FLI C T O F I NTE R E S T
The authors declare that they have no competing interests.

E TH I C S A PPROVA L A N D CO N S E NT TO PA RTI CI PATE
The animal protocol was approved by the Ethics Committee of Xiangya Hospital (Central South University, China).

DATA AVA I L A B I L I T Y S TAT E M E N T
The data sets used and/or analysed during the current study are available from the corresponding author on reasonable request. F I G U R E 7 Combination treatment of cisplatin significantly enhances the anti-tumour effect of spautin-1 in the treatment of melanoma xenograft tumour models. A-C, A375 melanoma cells (5 × 10 6 cells/0.1 mL) were xenografted into nude mice. The mice were randomized for intraperitoneal injection of spautin-1 (40 mg/kg, once every other day) and/or gavage administration of cisplatin (3 mg/kg, once a week) for 14 days. The bodyweight (C) and tumour growth (B) were measured around twice per week. All the tumours were removed, and the weight was measured for each group at the last day (mean values ± SEM, n = 7). Significant differences were evaluated using a one-way ANOVA. *P < .05 vs control group, # P < .05 vs spautin-1 group, & P < .05 vs cisplatin group. D, Immunohistochemistry assay assessed Ki67 (upper) and γ-H2AX (lower). The representative images are shown. E, Proposed working model of spautin-1 on melanoma cells