The Combination of Thymoquinone and Chloroquine Dose-Dependently Regulates Autophagy and Potentiates Metastatic Melanoma Cell Death via Autophagy-Dependent and -Independent Mechanisms

TL;DR

Combining thymoquinone and chloroquine increases cancer cell death in melanoma by affecting autophagy, a process cells use to survive stress.

Contribution

This study reveals how thymoquinone and chloroquine interact to enhance melanoma cell death through both autophagy-dependent and -independent mechanisms.

Findings

The combination of thymoquinone and chloroquine showed additive or sub-additive cytotoxic effects in melanoma cell lines.

Thymoquinone increased autophagosome accumulation in WM9 cells while reducing chloroquine's anti-autophagic effect.

The drug combination altered cell morphology and autophagy dynamics in metastatic melanoma cells.

Abstract

Although combination therapies with mitogen activated protein kinase inhibitors remain among the most effective treatments for malignant melanoma, they are not universally applicable to all subtypes of this cancer, and their efficacy decreases in the presence of distant metastases. Drug resistance, often associated with elevated autophagy in tumor cells, and adverse effects of the treatment also reduce the survival time of melanoma patients. Therefore, the aim of our research was to assess the cytotoxicity of the combination of a late-stage autophagy blocker chloroquine with thymoquinone, a natural substance with anticancer potential and low toxicity towards healthy cells, in metastatic melanoma cell lines. Using the WST-1 assay, we examined the cytotoxicity of the combination of chloroquine and thymoquinone in melanoma WM9 and WM852 cell lines and assessed the type of their interactions. Additionally, using time-lapse bright field and fluorescent microscopy, we assessed changes in cell morphology during 48 h incubation with the tested compounds and their combinations, as well as their effect on autophagy. We identified an additive and sub-additive cytotoxic effect of thymoquinone/chloroquine combinations against WM9 and WM852 cells. Moreover, we found that thymoquinone combined with chloroquine caused an increase in autophagosome accumulation in WM9 cells, while attenuating the chloroquine’s anti-autophagic effect. A thorough understanding of the mechanism of drug interactions with natural substances is crucial for the development of new effective anticancer therapies.

Figures

Full text

1. Introduction

Melanoma is a malignant tumor resulting from a mutation in melanocytes [1]. Although it is not the most frequently diagnosed skin cancer, it is responsible for the largest number of deaths caused by skin cancers [2], which constitutes 0.6% of cancer-related deaths worldwide [3]. Melanoma occurs mainly in the skin, but can also appear in other tissues where melanocytes are present, such as the eyes, digestive and genitourinary systems, and the nasopharynx [1]. Genetic mutations associated with melanoma affect signaling pathways that regulate key cellular processes such as growth, differentiation, migration, division, and apoptosis. Dysfunctions in the RAS-RAF-MEK-ERK mitogen activated protein kinase (MAPK) cell signaling pathway, often associated with a mutation in the BRAF gene, lead to uncontrolled cell growth. Activating BRAF V600E mutations (a valine-to-glutamic acid substitution) are characteristic of younger people and sun-undamaged skin, and occur in nearly 50% of melanoma cases [4]. Mutations in the gene coding for NRAS protein represent the second most common genetic alteration in melanoma, present in up to 20% of cases, particularly in individuals with periodically sun-exposed skin [5]. Current treatments for melanoma include surgical resection of the tumor, targeted therapy, and immunotherapy. Surgical removal of the lesion is still the mainstay of treatment, but adjuvant therapies may be necessary in cases of metastasis. Three immune checkpoint inhibitors, ipilimumab (anti-CTLA-4, cytotoxic T-lymphocyte associated protein 4), nivolumab, and pembrolizumab (anti-PD-1, programmed cell death 1), have been approved for the treatment of melanoma, although not all patients respond to them, and some experience side effects of the therapy [2,6]. Combination therapy with MAPK inhibitors, such as BRAF and MEK inhibitors, has shown efficacy in treating melanoma with a BRAF V600 mutation, improving response rate and progression-free survival [7]. Three combinations of these inhibitors (dabrafenib and trametinib, vemurafenib and cobimetinib, encorafenib and binimetinib) have been approved by the Food and Drug Administration (FDA) for the treatment of unresectable disease [8,9,10]. Furthermore, the combination of dabrafenib and trametinib has been approved as adjuvant therapy after resection of stage III/IV melanoma. This underscores the clinical importance of testing patients for the presence of BRAF and NRAS mutations, as such molecular profiling enables the selection of the most appropriate strategy for each patient, identifies individuals who are less likely to benefit from specific treatment regimens, and informs the inclusion of difficult-to-treat subgroups in clinical trials evaluating emerging therapeutic agents. The latter is especially important for patients with NRAS mutation, for whom the treatment prognosis is unfavorable, due to high tumor aggressiveness, limited availability of effective targeted therapies, and the rapid emergence of therapeutic resistance [11]. Additionally, most patients who initially respond well to treatment with BRAF or MEK kinase inhibitors develop acquired resistance, as a consequence of reactivation of the MAPK pathway [12], although the combination treatment delays the appearance of this phenomenon [13], and a small percentage of them do not respond to the treatment due to intrinsic mechanisms of resistance [2]. This raises the continuing need to develop anti-melanoma therapies that overcome resistance mechanisms.

Although many factors may contribute to treatment resistance, autophagy has recently been implicated among them [14]. In healthy cells, this catabolic process plays a protective role, eliminating damaged organelles or recycling macromolecular components. For cancer cells, however, it is believed to be a protective shield that allows them to survive in an unfavorable environment of lack of nutrients, oxygen, or growth factors [15]. It may also increase their survival in the presence of chemotherapeutics or drugs used in targeted therapies [14]. The antimalarial chloroquine (CQ), used as an anti-inflammatory agent in dermatology and rheumatology [16,17], is, next to hydroxychloroquine, the only drug approved by the FDA and used in clinical trials to treat cancer by inhibiting the autophagy process [18]. It has been proven that chloroquine can lead to inhibition of melanoma development by various mechanisms. It inhibits tumor metastasis and promotes apoptosis by inhibiting the degradation of the p53 protein, and activation of the p53 upregulated modulator of apoptosis (PUMA) [19]. In the case of melanoma with the GNAQ/11 mutation, sensitivity to MEK1/2 kinases inhibition increases [20]. Since high doses of chloroquine are needed to achieve the desired anticancer effect, its possible adverse reactions must be taken into account [21]. The use of chloroquine in combination with a natural substance with confirmed anticancer activity and minimal toxicity towards healthy cells would allow for a reduction of its dose and thus also of the side effects of the therapy.

Thymoquinone (TQ) is a benzoquinone belonging to the monoterpenoid class with well-documented therapeutic effects, including anticancer properties. In cancer cells, TQ acts by interfering with oncogenic pathways, preventing inflammation and oxidative stress, inhibiting metastasis and angiogenesis, and inducing apoptosis [22,23,24]. At certain concentrations, TQ causes DNA damage and promotes the formation of reactive oxygen species, inducing apoptosis in cancer cells. The anti-tumor activity of TQ has been attributed to the inhibition of Janus kinase 2 (Jak 2) signaling pathways [23,24,25]. Studies confirm the efficacy of thymoquinone in combating various types of cancers, including malignant melanoma [23,26,27,28,29,30], although to date, only one study has examined human metastatic melanoma [22].

There are numerous reports on the synergy of TQ with radiotherapy and chemotherapy [28,31,32,33,34,35,36,37,38,39]. TQ has been shown to act synergistically with cisplatin in both non-small cell and small cell lung cancer, without increased toxicity [32]. It has also been found to have a more negligible effect on healthy tissues compared to traditionally used therapeutic agents, such as cisplatin [40]. In the case of colon cancer, TQ has been shown to act synergistically with 5-fluorouracil [33]. In turn, studies on breast cancer have shown synergy of TQ with radiation and substances such as resveratrol, cyclophosphamide, cabazitaxel, docetaxel, gemcitabine, and ferulic acid [35,36,37,38,39,40]. In modern medicine, TQ has significant potential to be incorporated into pre-existing cancer treatment regimens.

Taking the above into account, our study has aimed to investigate the potential cytotoxicity of combinations of different concentrations of thymoquinone with the late autophagy blocker chloroquine towards human metastatic melanoma cells of the BRAF V600E-mutated WM9 line and NRAS-mutated WM852 cells and to compare the antiproliferative effect of the tested combinations with the effect of these substances used separately. Additionally, we have attempted to assess the interactions (synergism, additivity, antagonism) of the tested combinations of thymoquinone and chloroquine. Furthermore, we have assessed morphological changes in the two melanoma cell lines exposed to the TQCQ combinations and the effect of the mixtures on the autophagy process. The presented results indicate differences in the biological activity of the TQCQ combination in relation to TQ and CQ used separately and suggest a possible mechanism of their action.

2. Results

2.1. Thymoquinone/Chloroquine Combinations Enhance Cytotoxic Responses in BRAF- and NRAS-Mutated Metastatic Melanoma Cells

To evaluate the response of WM9 and WM852 melanoma cell lines to TQ, CQ, and their combinations, cells were treated with 5, 10, 20, and 40 μM CQ or TQ for 48 h. The subsequent cytotoxicity analysis (WST-1 assay) revealed a very strong negative correlation between CQ concentration and the viability of both lines (r = −0.89 for WM9, r = −0.88 for WM852) (Figure 1A,D). Similarly, a strong (WM9: r = −0.66, Figure 1B) and very strong (WM852: r = −0.98, Figure 1E) negative dose–response relationship was observed for TQ. Compared to the control, the viability of WM9 was significantly lower after treatment with 10, 20, and 40 μM CQ and 20 and 40 μM TQ (Figure 1A and Figure 1B, respectively), whereas all doses of TQ and CQ were toxic to WM852 cells (Figure 1D and Figure 1E, respectively). Next, the cells were incubated with the following TQCQ combinations: 5/10, 10/10, and 20/10 μM. Mixtures containing CQ at a concentration of 10 μM were selected to study compound interactions because it proved to be the lowest toxic concentration of CQ alone towards WM9 cells, and the one that resulted in a minimum of 25% reduction in cell viability in the case of both cell lines examined. In WM9, the dose–response correlation for TQCQ10 mixtures was moderate (r = −0.41, Figure 1B), while in WM852, the toxic effect of the TQCQ10 combinations was very strongly correlated with the TQ dose (r = −0.96, Figure 1E). In parallel, significantly lower percentage of viable, metabolically active cells was noted in populations treated with 10/10 and 20/10 μM (both cell lines, Figure 1B,E), as well as with 5/10 μM (WM852, Figure 1E), compared to the control. The TQ40CQ10 combination was not selected for cytotoxicity testing because TQ40 alone reduced cell viability by an average of about 64% (in the case of BRAF^mut^ cells) and 99% (in the case of NRAS^mut^ cell line), and our goal was to find a combination of TQ and CQ that would prove to be significantly toxic at the lowest possible concentration of the substances tested. A significantly lower number of viable WM9 cells was detected after their treatment with TQ10CQ10 and TQ20CQ10 mixes than in the cells incubated with the same concentrations of TQ alone. However, substantially lower cell viability was revealed only for the TQ20CQ10-treated populations, when compared to CQ10-treated cells (Figure 1C). Unlike WM9 cell line, NRAS^mut^ cells showed reduced viability when treated with TQ10CQ10 and TQ20CQ10 combination, compared to chloroquine, but only the first mixture was also proved to be more cytotoxic, relative to TQ in monotherapy. When used alone, both TQ and CQ were shown to exert a lower viability-reducing effect in WM852 cells, compared to TQ5CQ10 combination (Figure 1F). To sum up, TQ exhibited a stronger cytotoxic effect, both as monotherapy and in combination therapy with CQ, against NRAS-mutated melanoma cell line than against BRAF-mutated cells.

As TQ10CQ10 and TQ20CQ10 combinations proved to induce a greater cytotoxic effect than TQ or TQ and CQ used alone in WM9 cells, respectively, cell growth inhibition rate was computed for these combinations, based on the cell viability results obtained. For WM852 cells, the same calculations were performed for the combinations described above and, additionally, for the TQ5CQ10 mixture, as they exhibited a significantly higher cytotoxic effect than either agent used individually (TQ5CQ10 and TQ10CQ10) or CQ alone (TQ20CQ10). Cell growth inhibition rate, together with subsequently calculated Q value indicated an additive effect for the combination of 10 μM TQ and 10 μM CQ, and sub-additive for TQ20CQ10 mixture in WM9 cell line. A sub-additive effect was also demonstrated for all three combinations in the WM852 cells. The results are presented in Table 1.

2.2. The TQ20CQ10 or TQ5CQ10 Combinations Increase Vacuolation and Vesicle Release Prior to Cell Death in WM9 and WM852 Cells, Respectively

To pre-assess the mechanism of TQCQ combination toxicity, and to compare it with the toxicity mechanisms of TQ and CQ, WM9 and WM852 cells were first treated with TQCQ mixes, as well as with the two substances alone, and imaged for 48 h in a bright field mode. Lower concentrations of TQ (5 and 10 μM) and their respective combinations with CQ were applied to WM852 cells, compared to the WM9 cells (10 and 20 μM), due to the WM852 cell line’s increased sensitivity to the tested compound, as determined by the cytotoxicity test. Enhanced plasma membrane budding, vacuolation and abrupt release of vesicles upon plasma membrane rupture have been noticed for the CQ10-treated WM9 cell line (Figure 2A, Video S1). In contrast, TQ10-treated BRAF-mutated cells continue to grow and divide and reach confluence at the end of the 48 h incubation period, although more cells with plasma membrane budding were seen in this population (Figure 2B, Video S2), compared to control (Figure 2F, Video S6). Conversely, WM9 cells exposed to TQ20 alone exhibited cell swelling accompanied by their severe vacuolation, although numerous cells with vesicle budding were also observed in this experimental variant (Figure 2C, Video S3). As for the TQCQ-treated WM9 populations, cells with clearly visible budding plasma membranes were seen in the WM9 cells incubated with the mixture containing a lower TQ concentration (Figure 2D, Video S4). Additionally, some cells were characterized by the presence of large vacuoles in the cytoplasm, whereas others showed swelling or slight release of vesicles. Interestingly, cells treated with the TQ20CQ10 combination increased their volume and showed very strong vacuolation, and subsequent massive vesicle release, prior to cell death (Figure 2E, Video S5), which resembled the behavior of cells incubated with CQ alone (Video S1). However, the vacuoles were much larger in the mixture-treated cells than in those incubated only with CQ. When the cells treated with two TQCQ mixtures are compared, larger vacuoles and larger amount of vesicles released were observed in the population incubated with the combination containing a higher concentration of TQ.

In NRAS^mut^ cells, incubation with chloroquine induced similar morphological changes as in WM9 cells, i.e., vacuolization and vesicle release, along with plasma membrane budding (Figure 3A, Video S7). However, these effects were less pronounced than in WM9 cell line. In addition, vesicle budding was observed in WM852 cells following treatment with both TQ5 and TQ10 (Figure 3B, Video S8, and Figure 3C, Video S9, respectively), similar to those seen in WM9 cells. However, in contrast to the BRAF^mut^ cell line, WM852 cells failed to reach confluence after TQ10 treatment, and a substantial amount of dead cells were observed within the field of vision. Treatment of NRAS^mut^ cells with the TQ5CQ10 combination induced morphological changes similar to those observed with chloroquine alone; however, the cytotoxic effects became apparent more rapidly with the combination (Figure 3D, Video S10). In contrast to WM9 cell line, WM852 cells treated with the TQ10CQ10 combination did not exhibit increased cytoplasmic vacuolation but instead showed plasma membrane budding with sporadic vesicle release (Figure 3E, Video S11). Consistent with the other compounds and their combinations tested, cytotoxic effects were observed earlier in the NRAS^mut^ cell line than in WM9 cells. A solvent control for WM852 cells with no visible signs of increased cell death, is presented in Figure 3F and in Video S12 (Figure 3F, Video S12). The most characteristic changes in the morphology of WM9 and WM852 cells incubated with the compounds tested are summarized in Table 2.

2.3. The Autophagy-Regulating Effect of TQ10CQ Mixture Does Not Outweigh the Effect of CQ in Monotherapy

In order to investigate whether the chloroquine-caused late-stage autophagy inhibition is enhanced or weakened by the addition of TQ, and whether this potential change of the interference of TQCQ mixtures in the process of autophagy can be the cause of decreased cell viability, a time-lapse fluorescence microscopy analysis following transfection of the cells with the microtubule-associated protein 1 light chain 3B-green fluorescent protein (LC3B-GFP) construct and their treatment with TQCQ mixtures, as well as TQ and CQ alone was performed. The subsequent calculation of the number of LC3B-positive spots, corresponding to the number of autophagosomes, per number of cells, revealed the highest number of LC3B-positive spots relative to the number of cells in the 3rd, 4th and 5th hour of incubation, in the CQ10-treated population of WM9 cells (Figure 4A, red curve). Similarly, cells treated with both TQCQ combinations showed the highest number of autophagosomes/nuclei in the 4th hour of the experiment (Figure 4A, black and pink curves). No distinctive peak corresponding to the amount of autophagosomes/cell has been detected for TQ-treated cells (Figure 4A, green and cyan curves), as well as for the control (Figure 4A, blue curve). In contrast to WM9 cells, the WM852 cell line showed only a slight increase in the number of autophagosomes/cell, observed exclusively for the TQ5CQ10 (Figure 4D, orange curve) and TQ10CQ10 (Figure 4D, black curve) mixtures at 1st, 2nd, and 3rd hour of incubation, with the most pronounced peak occurring at the 2nd hour. For further analysis, data collected in the 3rd, 4th, and 5th hour, for WM9 cells, and in the 1st, 2nd, and 3rd hour of the live-cell experiment, in the case of the WM852 line, have been averaged and assessed. No significant correlation between the concentration of TQ used alone and the amount of autophagosomes/cell has been detected, either for BRAF^mut^, or for NRAS^mut^ cells (Figure 4B and Figure 4E, respectively). Additionally, none of the TQ concentrations increased the number of autophagosomes significantly, compared to the control. Conversely, the number of LC3B-positive spots/cell in the WM9 cell line, but not in the WM852 cell populations, incubated with both TQCQ mixtures, as well as in the cells treated with CQ10, was statistically higher than in the solvent control (Figure 4C and Figure 4F, respectively).

When comparing the LC3B+ spot/cell ratio in the 4th hour (WM9) or in the 2nd hour (WM852) in the cells treated with TQCQ combinations with this parameter in populations incubated with the tested compounds separately, a significantly higher amount of the analyzed autophagy marker has been observed in the WM9 cells exposed to TQ10CQ10 mixture (Figure 5A,C, Video S13), compared to the cells treated with TQ10 alone (Figure 5A,D, Video S14). The treatment with this TQCQ combination, however, did not result in a significant increase or decrease in autophagosomes/cell in comparison to CQ-treated cells (Figure 4C and Figure 5A,G, Video S15). Conversely, TQ20CQ10-treated populations (Figure 5A,E, Video S16) expressed significantly fewer LC3B-positive spots/cell than cells incubated with CQ only (Figure 4C and Figure 5A,G, Video S15), although no significant increase/decrease in the number of autophagosomes/cell was seen for this TQCQ mixture, compared to TQ20-treated cells (Figure 5A,F, Video S17). Almost no green-fluorescing autophagosomes were visible in the control (Figure 5H, Video S18). No significant differences in the number of autophagosomes/cell between the groups treated with TQCQ combinations and those incubated with TQ and CQ in monotherapy were observed in WM852 cells (Figure 5B and Figure S1). Overall, both CQ alone and TQCQ combinations increased the accumulation of LC3B-positive puncta in WM9 cells, while no effect on autophagosome number was observed in the WM852 cell line.

3. Discussion

The use of a combination of substances acting on different defense mechanisms of cancer cells often results in an enhanced therapeutic effect by eliminating the phenomenon of drug resistance and consequently complete recovery or extending the patients’ survival time [41]. Combining registered anticancer drugs, both chemotherapeutics and biological drugs, apart from greater treatment effectiveness, when compared to therapy with single substances, unfortunately, also shows high toxicity towards non-cancerous cells, causing side effects [42,43,44]. Natural substances with observed anticancer effects, but lower toxicity towards healthy cells, therefore, seem to be a good alternative to pharmaceuticals [22,41]. In this study, an attempt was made to investigate the effect of a combination of thymoquinone, a natural substance with anticancer properties, and chloroquine, a late-stage autophagy blocker, on the survival of BRAF^mut^ WM9 cells, originally derived from a malignant melanoma metastasis to a lymph node, as well as on BRAF-wild type (WT) NRAS^mut^ WM852 metastatic melanoma cells. Given that CQ exhibits toxicity toward both cancerous [19,20,45] and noncancerous cells [21,46], our goal was to combine CQ with thymoquinone to reduce CQ’s concentration to a level at which it would demonstrate satisfactory antitumor activity with minimal toxicity to healthy cells. As expected, our study showed very strong negative dose–effect relationship for chloroquine both in WM9 and WM852 melanoma cells, when their viability is concerned, with the lowest toxic CQ concentration of 10 μM for BRAF mutants and 5 μM in the case of NRAS^mut^ cells. Analogously, thymoquinone proved to exert a strong proliferation-reducing effect in WM9 melanoma cell line, the stronger the higher the concentration, which confirmed our already published data, although a wider range of concentrations was tested at that time [22]. The lowest cytotoxic TQ concentration, however, proved to be twice as low as the previously confirmed lowest cytotoxic TQ dose for WM9 cells (20 μM vs. 40 μM). This could be attributed to the fact that cancer cells in the metastatic phase are characterized by high heterogeneity and significant aggressiveness. These features are intended to ensure cells survive and multiply as quickly as possible [47]. Therefore, even small differences in experimental settings can affect cell behavior and, consequently, cause discrepancies in results. Similarly to chloroquine, TQ was found to be more toxic to BRAF WT NRAS-altered cells than to the WM9 cell line, with the lowest toxic dose being 5 μM and a stronger concentration-response correlation in the case of the former. With respect to the antimalarial agent studied, a few literature reports showing its toxicity towards melanoma cells mention a noticeable decrease in their survival at a CQ concentration of 50 μM, within 24 h [48]. The same concentration was the lowest one to induce apoptosis in Mel624 melanoma cell line [19] and chloroquine proved to be toxic to A375, DM6, and SK-Mel-2 melanoma cell lines in both normoxic and hypoxic conditions [48]. Importantly, the effective concentrations of CQ towards WM9 and WM852 cell lines (10 and 5 μM, respectively), as shown in our study, turned out to be substantially lower than the CC50 values (>19 μM) previously determined for non-tumorigenic cell lines [49]. Many scientific articles focus on the effects of combining CQ with anticancer drugs on melanoma cell lines [50,51]. The enhanced cytotoxic and proapoptotic effects of CQ in combination with chemotherapeutic agents were also observed in studies involving non-melanoma cell lines in combination with carboplatin [52], or doxorubicin [53]. However, to our knowledge, the toxicity of the TQCQ combination has not been tested before, and the results presented in this paper make a new contribution to the research on the use of combinations of autophagy inhibitors and natural substances in anticancer therapies. Conversely to TQ and CQ alone, TQ in TQCQ mixtures showed a moderate viability-reducing dose–effect relationship in WM9 cells. However, for both combinations tested, the effect was stronger than that of TQ used separately, with TQ20CQ10 mixture significantly reducing cell viability also compared to CQ alone. In the case of the WM852 cell line, both monotherapy and combination treatment showed a greater cell viability-reducing response, with lower doses of TQ becoming toxic when administered together with chloroquine, compared to WM9 cells. When the interaction between the two drugs is discussed, an additive cytotoxic effect of TQ and CQ was observed for TQ10CQ10 mixture in the WM9 cells. For all other TQCQ combinations both in WM9 and WM852 cell lines, the observed combination effect was higher than the effect of each compound alone but lower than the sum of their individual effects. This indicates a positive, sub-additive interaction, meaning the combination improves efficacy over single compounds but does not reach full additivity or classical synergy [54]. Although the applied reference model suggested additivity at high concentrations, this discrepancy likely reflects ceiling effects and limitations of the model under conditions of high toxicity, rather than true absence of interaction. Such results should be described as a partial additive or positively interacting combination, avoiding overinterpretation as full synergy [55,56]. Although the most desired effect when testing combination therapies is a synergistic effect, an additive effect also underlies many therapies used to treat a variety of conditions, e.g., hypertension, asthma, chronic obstructive pulmonary disease, or cancer [57,58,59]. It allows the use of drugs that are part of combination therapy in their optimal concentrations, which minimizes side effects and drug toxicity. In this context, our result showing an additive, or partial additive anticancer effect of the TQCQ combinations on metastatic melanoma cells is a positive news and creates an opportunity to undertake further research on the use of these combinations in melanoma therapy. Moreover, it gives hope for the development of more effective treatment for BRAF WT NRAS-altered melanoma, which is characterized by higher aggressiveness and worse patient outcomes [60]. However, to confirm the universality of this TQCQ interaction in advanced melanoma, additional studies using other metastatic melanoma cell lines are needed.

To elucidate the mechanism of TQCQ combination-caused cytotoxicity, we evaluated the changes in morphology in WM9 and WM852 cells treated with the TQCQ mixtures, which showed a more substantial proliferation-reducing effect than TQ and CQ in monotherapy. The observed cell budding, which might be indicative of apoptosis, was not a surprise, providing that TQ was previously identified as an apoptosis-inducing substance [21,23,25,27], also in WM9 melanoma cell line [22]. This type of cell death was dominant in WM852 cells, regardless of the TQ concentration alone or in the combinations tested. However, to confirm the participation of apoptosis in TQCQ-evoked toxicity, additional analyses are required. Interestingly, the apparent strong vacuolation in WM9 cells incubated with TQ20CQ10 combination and less noticeable in TQ5CQ10- and TQ10CQ10-treated WM852 cell line suggested the participation of more than just one cell death mechanism in cell toxicity. Although autophagy as one of them seemed likely, given that chloroquine is a known late autophagy inhibitor [18], the presence of large vacuoles and subsequent cell swelling and rupture of plasma membrane in WM9 cells could also indicate oncosis, paraptosis, necroptosis, or methuosis [61,62,63,64]. So far, no direct information indicating the involvement of TQ in any of the above-mentioned cell death mechanisms in melanoma cells is available. Nevertheless, the ability of certain quinones, including thymoquinone, to induce reactive oxygen species generation and subsequent mitochondrial membrane depolarization could result in mitochondrial damage and reduced ATP production. This in turn could lead to apoptosis, oncosis, paraptosis, and necroptosis [62]. Further studies including the detection of markers characteristic of these types of cell death, and the use of inhibitors specific for them, are necessary.

Chloroquine can affect cancer cells via various mechanisms leading to cell death, [19,48,51,65]. However, in experiments aimed at developing new anticancer therapies, it is often used as an inhibitor of late-stage autophagy. Interfering with the process of cellular ’self-eating’, chloroquine increases the pH of lysosomes and inhibits the fusion of autophagosomes with lysosomes, which disrupts protein degradation [18]. Reports on the effect of thymoquinone on autophagy provide conflicting information. On one hand, TQ increases the recruitment and accumulation of autophagosomal protein LC3II and p62 and induces lysosomal permeabilization in glioblastoma cells, suggesting inhibition of the late stage of autophagy [66]. On the other hand, it can induce protective autophagy in gastric cancer cells by suppressing the PI3K/Akt/mTOR pathway [67]. Additionally, at low doses (<10 μM), TQ triggered protective autophagy in cardiomyocytes, which increased their survival after subsequent incubation with doxorubicin [68]. Considering the above, we attempted to explain the additive and sub-additive cytotoxic effect of TQ and CQ combinations in WM9 and WM852 cells by analyzing their influence on the late stage of autophagy, using modern quantitative live cell fluorescent microscopy. We found that TQ in combination with 10 μM chloroquine increased the accumulation of autophagosomes per cell in both concentrations studied (10 and 20 μM), in WM9 cells, and that the effect of each combination was significantly greater compared to the control and—in case of TQ10CQ10 combination—also compared to thymoquinone alone. At the same time, however, this effect was not stronger than (in case of TQ10CQ10), or even weaker (in case of TQ20CQ10) than that of chloroquine used separately, which could imply that TQ at certain concentrations partially attenuates the strong autophagy-inhibiting effect of CQ. This, in turn, could suggest an action of TQ opposite to that of chloroquine, i.e., stimulation of autophagy. Autophagosome accumulation, in isolation, is, however, insufficient to conclusively demonstrate inhibition of the “self-eating” process, and more studies are needed to elucidate the mechanism of autophagy-regulating effect of TQ. However, based on the results obtained, it can be stated that TQ at a dose of 20 μM weakens the anti-autophagic action of chloroquine in the WM9 cell line. Taking into account the data obtained by our group during the study of cytotoxicity of both TQCQ combinations, it can be concluded that the action of TQ20 opposite to that of chloroquine ultimately does not cause weakening of the cytotoxic properties of the latter and the reduced survival of WM9 cells exposed to this combination is not dependent solely on the autophagy-regulating effect of the tested substances and may be associated with other mechanisms leading to cell death. The results obtained for the WM852 cell line, showing no significant dependence of cell death on blocking autophagy and a simultaneous significant enhancement of the toxic effect of TQ in combination with CQ, seem to confirm this hypothesis. This could be in line with recent reports suggesting that chloroquine-evoked cytotoxicity is largely independent of autophagy, but it is rather associated with its action on lysosomes [69]. Treatment with chloroquine can also result in enhanced anti-tumor immunity, activation of the integrated stress response, and PUMA or p53 stabilization-induced apoptosis [19,70]. Recent reports indicating the efficacy of combining autophagy inhibition with the MEK inhibitor that elicits protective autophagy in RAS-mutated cancer cells [71], as well as suggesting the potential effectiveness of this strategy in NRAS-mutated melanoma, underscore the need for further investigation of combination therapies incorporating chloroquine/hydroxychloroquine in these types of cancers [72]. Interestingly, in the case of melanoma, the role of autophagy in its development, progression and resistance to treatment has not been clearly defined yet. Some reports indicate that the use of BRAF kinase inhibitors in BRAF-mutated cells causes activation of autophagy and that blocking of the “self-eating” process is associated with tumor progression, metastasis and resistance to therapy [73,74]. In this context, the use of a combination of chloroquine with a natural substance, which could reduce the CQ-evoked inhibition of late-stage autophagy, in melanoma therapy seems justified. Further studies are, however, necessary to elucidate the exact mechanism of action of thymoquinone and chloroquine in metastatic melanoma cells.

When interpreting the findings of this study in the context of the therapeutic utility of TQCQ combinations in melanoma treatment, the limited scope of the investigations conducted should be taken into account. Although the findings presented in this study suggest the potential for developing a novel therapeutic approach for metastatic BRAFV600E-and NRAS-mutated melanoma that may be less susceptible to the development of drug resistance, further investigations are required to confirm its applicability to skin cancers harboring other mutations. In addition, the effects of the TQCQ combination on non-malignant cells that are susceptible to drug-induced toxicity should be evaluated to exclude the possibility of greater toxicity toward healthy cells. Furthermore, in light of the potential immunosuppressive properties of TQ, the evaluation of the effects of the TQCQ mixtures on the tumor microenvironment is required. Moreover, it is necessary to further explore whether the tested drug combinations may influence melanoma cells through mechanisms beyond autophagy regulation, including the induction of other forms of programmed cell death, such as apoptosis, paraptosis, oncosis, or pyroptosis. Finally, the effects and properties of TQ alone must be taken into consideration when developing a novel anti-melanoma therapy, namely its dose-dependent antioxidant or prooxidant action, and poor water solubility. In vivo and clinical studies will be required to define the therapeutic window and fully assess the potential dose-dependent toxicity of TQ and its nanocarrier formulations, both alone and in combination with CQ, which will be the focus of our future investigations.

4. Materials and Methods

4.1. Cell Culture

WM9 and WM852 melanoma cells (Rockland Immunochemicals, Pottstown, PA, USA) were grown in a high-glucose Dulbecco’s modified Eagle’s medium (Sigma-Aldrich Merck Group, St. Louis, MO, USA) with the addition of 2 mM L-glutamine (Serana, Pessin, Germany), 1 mM sodium pyruvate (Sigma, Gillingham, UK), antibiotics (penicillin and streptomycin; Serana, Pessin, Germany), and 5% fetal bovine serum (Serana, Pessin, Germany). The incubation conditions were standard for human cells, i.e., 37 °C, 5% CO_2_, and a humidified atmosphere.

4.2. Cell Toxicity Assessment/Drug Interaction Evaluation

A 10 mM stock solution of TQ in dimethylsulfoxide (DMSO) (Sigma-Aldrich, St. Louis, MO, USA), was prepared immediately prior to each experiment and subsequently and gradually diluted with culture medium to the desired concentrations before administration to the cells. Chloroquine diphosphate (30 mM aqueous solution) was stored at temperatures below 25 °C for no longer than six months. Both TQ and CQ were obtained from Cayman Chemicals (Ann Arbor, MI, USA). WM9 cells (2.5 × 10^3^ cells/well) and WM852 cells (5 × 10^3^ cells/well) were cultivated in 96-well plates for 24 h. After this time, the culture medium was replaced with a fresh one containing TQ, CQ, or a mixture of the two substances, for the next 48 h. The final concentrations of the drugs were 5, 10, 20, and 40 μM for both of the chemicals, and 5/10, 10/10 and 20/10 μM for TQCQ mixtures. Cells treated with 0.5% DMSO were used as a negative control and the substances in medium, but without the cells as blanks. To assess the cytotoxic effect of the substances tested, or the effect of their combinations, the Cell Proliferation Reagent WST-1 assay (Roche Diagnostics, Mannheim, Germany) was used. Briefly, the WST-1 reagent was added to the cells and after a 30-min incubation the absorbance was measured at 450 nm (with 620 nm background correction) with the use of a microplate reader (Infinite 200 Pro, Tecan, Männedorf, Switzerland). To calculate the viability of the cells, the following formula was used: [(A_test_ − A_blank_)/(A_control_ − A_blank_)] × 100%. The readings were taken from at least eight independent experiments, each conducted in quadruplicate. The inhibition rate was calculated using the following equation: (1 − average OD of experimental group/average OD of control group) × 100, and the cytotoxic effect of the combinations of the drugs was assessed with the help of the following formula: Q = E(A + B)/[(EA + EB) − (EA × EB)], where EA and EB correspond to the inhibition rate for drug A and B, respectively, and E(A + B) represents the inhibition rate for the combined therapy. A Q-value < 0.85 indicates an antagonistic effect, 0.85 < Q < 1.15—an additive effect, whereas a synergistic effect is represented by Q > 1.15 [75].

4.3. Bright Field Live-Cell Imaging

WM9 cells (2.5 × 10^3^ per/well) and WM852 cells (5 × 10^3^ cells/well) were seeded in a 96-well flat bottom plate and cultivated for 24 h. Then, the following substances were added to the culture: TQ (5, 10 and 20 μM), or CQ (10 μM), or the combinations of TQCQ (5/10, 10/10, 20/10 μM). The concentrations were chosen based on the cytotoxicity results. Cells treated with 0.025% DMSO were used as a negative control. Upon treatment, the cells were immediately placed in the incubation chamber of Agilent BioTek Lionheart FX automated microscope (BioTek Instruments, Inc., Winooski, VT, USA), with CO_2_ concentration set to 5% and temperature to 37 °C. The images were taken every 10 min, in a bright field mode, using the 20× objective and laser autofocus, for the next 48 h. The videos were created using Gen5™ Control and Data Analysis Software, ver. 3.12 (BioTek Instruments, Inc., Winooski, VT, USA). Images and videos were selected in a condition-blinded manner.

4.4. Time-Lapse Imaging of Autophagosome Formation and Quantitative Analysis of Autophagic Flux

WM9 and WM852 cells were transfected using Premo Autophagy Sensor LC3B-GFP (BacMam 2.0) kit (Life Technologies Corporation, Eugene, OR, USA), according to the manufacturer’s instruction. Briefly, WM9 cells were trypsinized, washed, and 40 × 10^3^ cells were mixed with 12 μL of LC3B-GFP baculovirus-based construct. The cells (2.5 × 10^3^ (WM9) or 5 × 10^3^ (WM852)) were subsequently cultured in 96-well black, clear bottom plate for 24 h. After this time, the cell nuclei of WM9 cells were stained with Hoechst 33342 nuclear stain (1:10,000, filtered) (Pierce Biotechnology, Rockford, IL, USA), for 10 min, 37 °C. In the case of WM852 cell line, Hoechst stain was replaced with a SiR-DNA far red probe (Spirochrome AG, Stein am Rhein, Switzerland) (1 μM, 25 min, 37 °C), due to the higher rate of cell death caused by the UV light. The cells were then washed briefly with culture medium, and they were treated with the tested substances as above (see section Bright field live cell imaging). Autophagosome formation was captured using Agilent BioTek Lionheart FX automated microscope (BioTek Instruments, Inc., Winooski, VT, USA), with DAPI or Cy5 (nuclei) and GFP (LC3B) light cubes. Images were collected every 1 h, for 48 h. Throughout the entire time of image collection, cells were incubated at 37 °C, 5% CO_2_. The images were processed and analyzed with the use of Gen5™ Control and Data Analysis Software, ver. 3.12 (BioTek Instruments, Inc., Winooski, VT, USA). The autophagic flux was expressed as the spot/nuclei ratio, corresponding to autophagosomes/cell, and calculated by using the following formula: DS1/DS2*1, where DS1 represents the average amount of LC3B-positive spots/image, and DS2—the average amount of cells/image. The spot/nuclei ratio was calculated from at least two fields of vision (FOV) per image, from a minimum of seven independent experiments.

4.5. Statistical Analysis

The data has been presented as individual samples, or mean ± standard deviation. The differences between the cells treated with TQ, CQ, or mixtures, and control groups were assessed using Mann–Whitney U test, with a p-value < 0.05 considered as statistically significant. To evaluate the relationship between the concentration of the drugs used and their effects on cells, Spearman’s rank correlation coefficient (r) was used. Statistical analysis was conducted with the use of Statistica 13.1 (Statsoft, Round Rock, TX, USA).

5. Conclusions

Thymoquinone at doses of 10 and 20 μM or 5 and 10 μM in combination with 10 μM chloroquine significantly reduces the survival of the WM9 and WM852 metastatic malignant melanoma cell lines, respectively, compared to the control and to thymoquinone used separately. The TQ10CQ10 combination shows an additive cytotoxic effect of TQ and CQ in the WM9 cell line, whereas TQ10CQ10 (in WM9 cells), and TQ5CQ10, TQ10CQ10, and TQ20CQ10 (in WM852 cell line) decrease cell proliferation sub-additively. TQ10CQ10 and TQ20CQ10 combinations increase the accumulation of autophagosomes in BRAF^mut^ cell line, compared to control, but simultaneously TQ in the TQ20CQ10 mixture reduces the anti-autophagic effect of chloroquine. The regulating effect of thymoquinone in combination with CQ on the autophagy process does not weaken the cytotoxic effect of the latter. The combinations of TQ and CQ used do not significantly modulate autophagy in WM852 cells, while showing significantly higher cytotoxicity compared to monotherapy. Knowledge of the mechanisms of action of natural medicinal substances and their potential interactions in combination with drugs is a prerequisite for the development of new, effective and less toxic anticancer therapies.