Novel chloroquine derivative suppresses melanoma cell growth by DNA damage through increasing ROS levels

Abstract Melanoma is a fatal cancer with a significant feature of resistance to traditional chemotherapeutic drugs and radiotherapy. A mutation in the kinase BRAF is observed in more than 66% of metastatic melanoma cases. Therefore, there is an urgent need to develop new BRAF‐mutant melanoma inhibitors. High‐dose chloroquine has been reported to have antitumour effects, but it often induces dose‐limiting toxicity. In this study, a series of chloroquine derivatives were synthesized, and lj‐2‐66 had the best activity and was selected for further investigation. Furthermore, the anti‐BRAF‐mutant melanoma effect and mechanism of this compound were explored. CCK‐8 and colony formation assays indicated that lj‐2‐66 significantly inhibited the proliferation of BRAF‐mutant melanoma cells. Flow cytometry revealed that lj‐2‐66 induced G2/M arrest in melanoma cells and promoted apoptosis. Furthermore, lj‐2‐66 increased the level of ROS in melanoma cells and induced DNA damage. Interestingly, lj‐2‐66 also played a similar role in BRAF inhibitor‐resistant melanoma cells. In summary, we found a novel chloroquine derivative, lj‐2‐66, that increased the level of ROS in melanoma cells and induced DNA damage, thus leading to G2/M arrest and apoptosis. These findings indicated that lj‐2‐66 may become a potential therapeutic drug for melanoma harbouring BRAF mutations.


| INTRODUC TI ON
Melanoma is a solid tumour resulting from malignant transformation of melanocytes in the skin and other organs. Although melanoma is rare (accounting for only 4% of skin cancer cases), it is a very deadly disease that accounts for 75% of skin cancer deaths. 1 BRAF mutation occurs in more than 66% of cases of metastatic melanoma. 2 The discovery of BRAF inhibitors such as vemurafenib and dabrafenib has improved the prognosis of melanoma patients. 3 However, a large proportion of patients rapidly develop secondary drug resistance. 4 Therefore, it is an urgent task to develop new drugs for melanomas with BRAF mutations.
Chloroquine has been found to exert inhibitory effects against melanoma via a variety of mechanisms. For example, it promotes the apoptosis of melanoma cells by inhibiting the degradation of the P53 upregulated modulator of apoptosis (PUMA) protein, suppresses melanoma cell invasion and metastasis by normalizing the tumour vasculature, and increases the sensitivity of GNAQ/11-mutant melanoma to MEK1/2 inhibition. [5][6][7] However, although chloroquine has considerable antimelanoma effects, a very high dose is often required to achieve these antitumour effects due to its weak antitumour activity. Systemic application of high-dose chloroquine may cause extensive and serious side effects, among which the most serious complications are retinopathy, cardiomyopathy, neuromuscular disease and myopathy. 8,9 Chloroquine derivatives containing a 7-chloroquinoline moiety were reported to have high antitumour activity. [10][11][12][13][14] Moreover, 2-methylquinoline and methylcarbazole were found to be important fragments with antitumour activity. [15][16][17] In our previous report, we found that the 2-methylquinoline and carbazole hybrid compounds 5I and 8g have certain antimelanoma activity. 18 Based on the above information, we speculated that combining 2-methyl-7-chloroquinoline and methylcarbazole hybrid fragments may be a promising direction for the design of inhibitors targeting BRAF-mutant melanoma. Thus, we designed and synthesized a variety of chloroquine derivatives-2-methyl-7chloroquinoline and methylcarbazole hybrid compounds ( Table 1 and Figure S1). Among these compounds, lj-2-66, containing a 2-methyl-7-chloroquinoline fragment and a methylcarbazole fragment connected by a carbon atom linker, had the highest anti-BRAF-mutant melanoma activity.
Through further study, we found that lj-2-66 exerts antitumour effects in melanoma harboring BRAF mutations both in vivo and in vitro and that the effective concentration was much lower than that of chloroquine. Further research showed that the compound causes DNA damage by increasing reactive oxygen species (ROS) levels, leading to G2/M arrest and apoptosis. In addition, lj-2-66 suppressed the proliferation of BRAF inhibitor-resistant (BRAFiresistant) melanoma cells independent of their resistance status, suggesting that lj-2-66 may be a potential drug for BRAF-mutant melanoma treatment.

| Chemicals
The method for synthesizing chloroquine derivatives is outlined in Figure S1. Cross-dehydrogenative coupling of 1 with aldehyde 2 using hypervalent iodine and a TMSN 3 system afforded Compound 3, which was then condensed with 4 to produce the product lj-2-64.
By changing aldehydes 2 to 4 using the same strategy described for the synthesis of Compound 3, we easily obtained lj-2-65. With the important intermediate lj-2-65 in hand, we could easily prepare lj-2-66, lj-2-67 and lj-2-68 by reduction with NaBH4, substitution with N,N-carbonyldiimidazole (CDI), and Wittig olefination with methyltriphenylphosphonium bromide (Compound 5) respectively. Then, lj-2-114 was prepared by modified click chemistry by condensation of 8 and 7 using a CuSO 4 /sodium L-ascorbate/Et 3 N system, and 7 was obtained by substitution of 6 with NaN 3 .

| Cell culture
The human melanoma cell lines Sk-Mel-5, Sk-Mel-28 and A375 (maintained in our laboratory) and the human melanocyte cell line PIG1 (a gift from Department of Dermatology, Xiangya Third Hospital) were used in this study. Sk-Mel-5, Sk-Mel-28 and A375 cells were cultured in Dulbecco's modified Eagle's medium (BI, Israel) supplemented with 10% foetal bovine serum (FBS) (BI, Israel) at 37 °C in 5% CO 2 . PIG1 was cultured in Medium 254 (Gibco) supplemented with 10% FBS (BI, Israel) at 37°C in 5% CO 2 . A375 cells were continually treated with 2 µM vemurafenib for more than 3 months to obtain a vemurafenib-resistant cell line (labelled RA), and the drug was removed one week before use.

| Colony formation assay
Cells were seeded into 6-well plates (1.5 × 10 3 cells in each well), cultured in complete growth medium, and incubated overnight at 37°C in 5% CO 2 . The next day, different concentrations of lj-2-66 or DMSO (control group) were added to the wells. After 48 h, the drugcontaining medium was replaced with medium containing 10% FBS.
Culture was terminated after colonies were macroscopically visible (approximately 2-3 weeks). Then, the colonies were washed twice with phosphate buffered saline (PBS), fixed with 4% paraformaldehyde for 15 min and stained with 0.5% crystal violet staining solution for 20 min. After counting the colonies, GraphPad software was used for statistical analysis of the data.

| Apoptosis assay
Cells were seeded into six-well plates at a density of 3 × Apoptosis was detected by flow cytometry, and data were analysed with FlowJo software. Each sample was analysed in triplicate.

| Cell cycle assay
Cells were seeded into six-well plates at a density of 3 × 10 5 cells per well, incubated overnight in complete growth medium at 37°C and treated with different concentrations of lj-2-66 or DMSO (control group). After 48 h, the cells were washed twice with precooled PBS and digested with 0.25% trypsin digestion solution. The collected cells were fixed with pre-cooled 70% ethanol and incubated overnight at 4°C. Then, the collected cells were incubated with propidium iodide stain (Biyuntian, C1052, China) according to the instruction manual. The cell cycle assay was conducted by flow cytometry, and data were analysed with ModFit software. Each sample was analysed in triplicate.

| Immunoblotting
Cells were lysed with RIPA lysis buffer containing protease and phosphatase inhibitors (Selleck, USA), and protein quantification was performed with BCA according to the instruction manual.

| RNA sequencing (RNA-seq)
After treating Sk-Mel-28 cells with 100nM lj-2-66 or an equal volume of DMSO (control group) for 24 or 48 h, the cells were collected and sent to Wuhan Huada Sequencing Company for RNA-seq.

| Xenograft tumour model
The

| Immunohistochemistry
Mouse tumours were fixed with formalin and embedded in paraffin. The slices were placed in a 65°C oven for 1.5 h. After the tissue was dewaxed with turpentine, an alcohol gradient was used to rehydrate the tissue. The slices were placed in preboiled sodium citrate and incubated in a pressure cooker for 6 min for antigen retrieval.
After the sections returned to room temperature, they were blocked with peroxidase for 10 min and with goat serum for 1 h. Then, the slides were incubated with primary antibodies specific for Ki67 (1:300, Abcam, ab16667), P21 (1:50; CST; 2947S) and P53 (1:50; Santa Cruz; 47698) in the dark at 4 °C overnight. The next day, the slides were incubated with the secondary antibody and stained with 3,3′-diaminobenzidine (DAB).

| Statistical analysis
The statistical results are presented as the mean ± SEM values and were analysed by Student's t test and one-or two-way ANOVA with GraphPad Prism software (version 6.01). A p value <.05 was considered to be significant.

| Chloroquine derivatives had antimelanoma activity
We designed several 2-methyl-7-chloroquinoline and methylcarbazole hybrid chloroquine derivatives and tested their effect on the viability of Sk-Mel-5 cells. As shown in Table 1 (Figures S3b and 2f). This result indicated that lj-2-66 significantly inhibited the growth of BRAFmutant melanoma in vivo.

| RNA-seq analysis identified the effect of lj-2-66 on the transcriptome and key pathways
To further explore the mechanism of lj-2-66 in melanoma, we per-  ±SD of each group, and an asterisk (*) indicates a significant difference compared with control group using one-way ANOVA (p < .05). (e) SK-Mel-5, SK-Mel-28 and A375 were treated with NAC for 1 hour and subsequently treated with100 nM lj-2-66 or equal volume DMSO (control group) for 6 h. Then, the cells were incubated with DCFH-DA as described in the methods. The data from multiple experiments are expressed as the mean (n = 3) ± SD. Significant differences were evaluated using two-way ANOVA, and an asterisk (*) indicates a significant difference (p < .05) Moreover, γH2AX accumulated in the nucleus after treatment with lj-2-66 for 24 h (Figure 4b,c). ROS are well recognized as mediators of DNA damage. 19 Therefore, we tested the effect of lj-2-66 on the level of ROS and found that lj-2-66 increased the production of ROS in numerous BRAF-mutant melanoma cell lines (Figure 4d). NAC is recognized as an inhibitor of ROS. We found that the promotive effect of lj-2-66 on ROS production was partially suppressed by NAC ( Figure 4e). These results suggested that lj-2-66 was likely to induce DNA damage by increasing the production of ROS, leading to G2/M arrest and apoptosis. when the dose was raised to 100 nM, the rate increased to 8.13%

| lj-2-66 inhibited the growth of RA
( Figure 5e). Therefore, we detected the expression of apoptosisrelated proteins in RA cells treated with lj-2-66. As expected, lj-2-66 treatment increased the expression of BAX and caused PARP cleavage (Figure 5f). We also performed RT-PCR to verify the expression of CDKN1A, GADD45A, BBC3, CDK1, MCM3, MCM4, MCM5 and MCM6. As expected, the changes in the expression of these genes were consistent with those identified by RNA-seq (Figure 5g). In addition, lj-2-66 treatment increased the ROS level, and this increase was partially suppressed by NAC (Figure 6a,b). We also tested the expression of DNA damage-related proteins and found that lj-2-66 increased the expression of P53, p-P53, P21 and γH2AX in a dosedependent manner (Figure 6c). Moreover, we observed accumulation of γH2AX in the nucleus (Figure 6d).

| DISCUSS ION
Chloroquine is regarded as a promising drug for tumour treatment, but it usually requires high doses, which may cause extensive and serious side effects. Therefore, the development of chloroquine derivatives with higher anticancer activity has become a hot topic in cancer treatment. [10][11][12][13][14] In this study, we designed several chloroquine derivatives-2-methyl-7-chloroquinoline and methylcarbazole hybrid compounds that have certain BRAF-mutant melanoma inhibitory activities-and identified the most active compound, lj-2-66 ( Table 1).
The IC 50 values of lj-2-66 in Sk-Mel-5, Sk-Mel-28 and A375 cells were 130, 80 and 100 nM respectively (Figure 1b). In addition, it had similar suppressive effects on RA cells and non-BRAFi-resistant melanoma cells, independent of the BRAFi resistance status (Figure 5a,b). Furthermore, we found that the antimelanoma activity of lj-2-66 was achieved via the induction of G2/M arrest and apoptosis (Figure 2a,b). Then, RNA-seq was performed to explore the mechanism of lj-2-66's effect on BRAF-mutant melanoma cells.
The results showed that cell cycle-and tumour-related pathways were significantly altered after lj-2-66 treatment ( Figure S4). Then, DNA damage and promote apoptosis by increasing ROS levels. 33 The CDK1 protein is a key cyclin that accelerates the G2/M transition and decreases the entire cell cycle time. 34 The expression of MCM protein in several malignant tissues (such as breast, gastrointestinal, lung and F I G U R E 6 lj-2-66 induced DNA damage by increasing the production of ROS in RA. (a) RA was treated with 100 nM lj-2-66 or equal volume DMSO (control group) for 3 or 6 h, then incubated with DCFH-DA as described in the methods. The results represent the mean (n = 3) ±SD of each group, and an asterisk (*) indicates a significant difference using one-way ANOVA (p < .05). (b) RA was treated with NAC for 1 h and subsequently treated with 100 nM lj-2-66 or equal volume DMSO (control group) for 6 h, then incubated with DCFH-DA as described in the methods. The results represent the mean (n = 3) ±SD of each group, and an asterisk (*) indicates a significant difference using two-way ANOVA (p < .05). (c) RA was treated with various dosages of lj-2-66 (20, 50 and 100 nM) for 48 h. Then, Western blotting was performed for the indicated antibodies. The histograms indicated relative protein expression, as means ± SD, and an asterisk (*) indicates a significant difference using two-way ANOVA (p < .05). (d) RA was seeded on coverslips (2 × 10 4 /well) and exposed to various dosages of lj-2-66 (50 and 100 nM) for 24 h, and then, immunofluorescence was performed forγ-H2AX and the photographs were taken by confocal microscope. The results represent the mean (n = 4) ± SD of each group, and an asterisk (*) indicates a significant difference using one-way ANOVA (p < .05) Therefore, we evaluated the expression levels of DNA damagerelated proteins and found that lj-2-66 increased the expression levels of P-53, p-P53, P21 and γH2AX in a dose-dependent manner ( Figure 4a). In addition, we found that γH2AX accumulated in the nucleus after treatment with lj-2-66 (Figure 4b,c). p53 is a widely studied tumour suppressor that affects the response of various cells to DNA damage. 39,40 In addition, P53 can be a target of ROS. 41 Histone H2AX is a chromatin factor that has been widely studied in the DDR. 42 These results indicate that lj-2-66 induces DNA damage. repair processes. Therefore, synthetic glucocorticoids may be coadministered with lj-2-66 to boost the potential therapeutic efficacy of the latter for melanoma harboring BRAF mutations. 45,46 Although we found that lj-2-66 has a much higher IC 50 value in immortalized nontumorigenic cells than in melanoma cells, there are still limitations of this study because we were not able to calculate the therapeutic index of the compound. In future studies, to provide more evidence supporting final clinical administration, the safety of the compound in vivo needs to be tested by approaches such as drug toxicology, metabolism and other related studies.

| CON CLUS IONS
In summary, we synthesized the chloroquine derivative lj-2-66, which can induce DNA damage by increasing the level of ROS, leading to G2/M arrest and apoptosis in melanoma cells and subsequently inhibiting the growth of melanoma both in vivo and in vitro. These results provided evidence that the compound lj-2-66 has an anti-BRAF-mutant melanoma effect (Figure 7).

CO N FLI C T O F I NTE R E S T
The authors confirm that there are no conflicts of interest.

DATA AVA I L A B I L I T Y S TAT E M E N T
The datasets generated and/or analysed during the current study are available from the corresponding author on request.