Introduction

Melanoma, a malignant tumor derived from melanocytes, causes the majority of deaths attributed to skin cancer. The incidence of melanoma, the most fatal form of skin cancer, has increased significantly over the past 20 years [1]. Melanoma has been treated by various ways. Surgical excision is a representative treatment in early stages of the disease [2], in addition, RAF and MEK kinase-targeted chemotherapy or immunotherapy using immune checkpoint inhibitors are approved for patients with advanced stage [3]. However, melanomas have diverse genetic alterations in key molecules on growth/proliferation signaling pathways and cell cycle control, such as BRAF, NRAS, KIT, PTEN and CDKN2A [4,5,6], and the poor prognosis-associated with metastatic melanoma has made it necessary for a novel pharmacological treatment options or more efficient chemotherapeutic intervention.

Apoptosis, known as programmed cell death, is triggered by extrinsic and intrinsic ligands. These two different apoptotic pathways are regulated by diverse intracellular signaling pathways and coordinated by a network of genes [7]. During the apoptotic process, activation of caspase cascade leads to the cleavage of essential proteins that is involved in anti-apoptotic pathways, such as Bcl-2 family proteins [8]. Apoptosis is a critical process for cellular homeostasis, therefore, apoptosis became a key target for the development of new anti-cancer therapies [9].

Reactive oxygen species (ROS) are generated by the mitochondria in most mammalian cells [10]. ROS have been related to diverse physiological and pathologic condition. Elevated ROS also act as signaling molecules in the maintenance of physiological functions [11], however, excessive ROS above the physiological concentration can lead to carcinogenesis or cell death [12].

Oxidative stress is due to the imbalance between scavenging activities of intracellular anti-oxidant and production of ROS [13], and also elevates intracellular levels of ROS that cause damage to various biological molecules, such as DNA, proteins and lipids [14]. In addition, oxidative stress has been reported to induce apoptosis via diverse pathways, such as endoplasmic reticulum (ER) stress and mitochondrial cascade [15]. Therefore, manipulating intracellular ROS levels is a way to selectively kill various cancer cells, which has been involved in the anti-cancer effects of several therapeutic agents, including paclitaxel, cisplatin, and TNF-related apoptosis-inducing ligand (TRAIL/Apo-2L) [16,17,18].

It has been demonstrated that signal transducer and activator of transcription 3 (STAT3), a member of the STAT family, plays critical roles in signaling of various cytokines, hormones, and growth factors on diverse human malignancies [19, 20]. STAT3 is constitutively activated in most human solid tumors, and constitutive activation of STAT3 signalling pathway has been demonstrated to mediate critical processes in tumor cells, such as survival, proliferation, angiogenesis, and metastasis [21,22,23]. In previous studies, many researchers indicated that STAT3 might be a novel target for anti-cancer therapy [24,25,26].

Rosmarinus officinalis, generally known as rosemary, is an aromatic evergreen herb which is in use of traditional medicine, fragrance, and spice. Rosmarinus officinalis contains rosmarinic acid, ursolic acid, betulinic acid, carnosic acid and carnosol as major bioactive polyphenolic components [27]. Among these components, carnosol (Fig. 1a) has a wide range of biological activities, including anti-cancer, anti-oxidant, and anti-inflammatory effects [28,29,30].

In this study, we aimed to investigate the underlying molecular mechanisms of anti-cancer effects of carnosol in malignant human melanoma G361 cells. Our results suggest that carnosol treatment induced apoptosis in G361 cells through oxidative stress and the interference of STAT3-mediated signaling pathway.

Materials and methods

Chemicals and reagents

Carnosol (purity 99%), N-acetyl cysteine (NAC), 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT) and anti-β-actin antibody were purchased from Sigma-Aldrich (St. Louis, MO, USA). Antibodies against cleaved caspase-3, -7, -9, poly(ADP-ribose) polymerase (PARP), Bcl-2, Bcl-xl, Bax, cytochrome c, STAT3, p-STAT3 (Y705), Src, p-Src, cyclin D1, D2, D3 and survivin were bought from Cell Signaling Technology Inc. (Beverly, MA, USA). Antibodies against p53, murine double minute-2 (Mdm2), and horse-radish peroxidase-conjugated secondary antibodies were purchased from Santa Cruz Biotechnology (Paso Robles, CA, USA). The 2′,7′-dichlorofluorescein diacetate (DCF-DA) was procured from Invitrogen (Carlsbad, CA, USA). Hank’s balanced salt solution (HBSS) was purchased from the Meditech (Herndon, VA, USA).

Cell lines and cell culture

G361 cells were obtained from American Type Culture Collection (ATCC) and maintained in Dulbecco’s modified Eagle’s medium (DMEM) supplemented with 10% fetal bovine serum and antibiotics (100 U/ml penicillin G and 100 mg/ml streptomycin) at 37 °C in a humidified incubator with 5% CO2-containing atmosphere. In all experiments, cells were seeded at 50–60% confluency. All chemicals were dissolved in ethanol and the final concentration of ethanol was less than 0.1% in culture media.

Cell viability assay

The effect of carnosol on cell viability was measured by MTT assay. Cells were plated at 2 × 103 cells/well onto a 96-well plate and incubated with 100 µl of carnosol-containing media at 37 °C for indicated time. Thereafter, 10 μl of MTT stock solution (5 mg/ml) was added to each well and incubated for 4 h. Medium was removed, insoluble formazan was dissolved in DMSO and absorbance at 550 nm was measured by using microplate reader (Tecan Trading AG, Switzerland). Cell viability was described as the relative percentage of control, and calculated based on technical triplicate repeats.

Annexin V-propidium iodide (PI) staining

Annexin V-PI staining was performed using fluorescein isothiocyanate (FITC)-Annexin V staining kit (BD Biosciences, San Jose, CA, USA) following the manufacturer’s instructions. Briefly, carnosol-treated cells were washed with PBS and resuspended in binding buffer containing Annexin V and PI. The fluorescence intensity of cells was analyzed by FACSVerse™ (BD Biosciences).

Immunoblot analysis

Cells were harvested and lysed with radioimmunoprecipitation assay (RIPA) buffer, and then cell lysates were quantified by using a BCA protein assay kit (Pierce Biotechnology, Rockford, IL, USA). Equal amount of cell lysates was denatured by boiling at 100 °C for 5 min in SDA-PAGE sample buffer. The cell lysates were separated on 8–15% sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDS-PAGE) and transferred onto polyvinylidene difluoride (PVDF) membrane. The blotto was blocked with 5% skim milk in Tris-buffered saline containing 0.1% of Tween 20 (TBS-T) for 1 h at room temperature. The blot membrane was washed for 3 times with TBS-T after blocking and hybridized with corresponding antibodies at 4 °C for overnight. After then, the blotto was washed for 3 times and incubated for 1 h with horseradish peroxidase-conjugated secondary antibodies at room temperature. The membranes were washed 3 times for 10 min each with TBS-T buffer. Immunoblot s were developed using enhanced chemiluminescence (ECL) reagent (GE Healthcare, NJ, USA) and visualized with Imagequant™ LAS 4000 (Fujifilm Life Science, Japan).

Electrophoretic mobility gel shift assay (EMSA)

The EMSA for STAT3 DNA binding was performed using a DNA–protein binding detection kit according to the manufacturer’s protocol (GIBCO BRL, Grand Island, NY). Nuclear extracts were prepared from cells cultured with or without TQ. The STAT3 oligonucleotide probe 5′-AGC TTC ATT TCC CGT AAA TCC CTA-3′ (Bionics, South Korea) was labeled with [γ-32P] ATP and EMSA was performed according to the protocol previously described [31].

Measurement of ROS accumulation

Cells were treated with carnosol in the presence or absence of NAC for 12 h and then loaded with 25 μM of DCF-DA. After incubation for 30 min at 37 °C in 5% CO2-containing atmosphere, cells were washed twice with HBSS solution, and examined under a fluorescence microscope to detect the intracellular ROS. The positive control cells were treated with 500 μM of hydrogen peroxide (H2O2). For flow cytometry, cells were trypsinized and re-suspended in growth media, then cells were applied to FACSVerse™ for further analysis.

Statistical analysis

The data were expressed as mean ± SD based on at least triplicate repeats, and statistical analysis between each single experimental group was performed using the Student’s t-test, and p value less than 0.05 was considered statistically significant.

Results

Effect of carnosol on cell viability in human melanoma G361 cells

To examine the effect of carnosol on the viability of human melanoma G361 cells, we performed MTT assay after treatment of carnosol. As shown in Fig. 1b, carnosol significantly reduced cell viability with a time- and concentration-dependent manner in comparison to the untreated cells. Carnosol treatment also induced the apoptosis-like morphological changes as the dose increases in G361 cells (Fig. 1c). We further analyzed these morphological changes after carnosol treatment by Annexin V-PI staining and calculated the number of Annexin V(+)/PI(−) cells. At 100 μM of carnosol-treated group, the cells were revealed approximately 80% of the apoptotic cell population as compared to untreated cells after 24 h of treatment (Fig. 1d, e). These results showed that the treatment of carnosol to G361 cells dramatically increased apoptotic cell death as compared with untreated control cells.

Fig. 1
figure 1

Concentration- and time-dependent effect of carnosol on cell viability of G361 cells. a Chemical structure of carnosol. b Concentration- and time-dependent decrease of viability of human melanoma G361 cells after carnosol treatment. c The morphological changes of G361 cells after dose-dependent carnosol treatment after 24 h. d The FACS analysis for carnosol-treated G361 cells (20, 50, 100 μM) using Annexin V-PI staining. The data were presented by dot plots. Annexin V-positive cells were calculated and plotted as the mean ± S.D e *p < 0.05, **p < 0.01 compared to control group

Full size image

Effect of carnosol on activation of caspases, levels of anti-apoptotic Bcl-2 family protein, and p53

The caspase cascade is one of the key components in induction of apoptosis, and is closely linked to mitochondrial proteins that control the mitochondrial membrane permeability [32]. To elucidate the involvement of caspases and Bcl-2 family proteins in carnosol-induced apoptosis, we assessed the cellular levels of cleaved caspases, Bcl-2, Bcl-xL, and Bax by immunoblot. After 24 h of carnosol treatment with indicated concentration, caspase-9, -7, and -3, which play a critical role in the execution-phase of apoptosis, were sequentially activated by proteolytic cleavage. In addition, cleaved PARP, a well-known substrate of active caspase-3, was detected in a dose-dependent manner (Fig. 2a). In addition, immunoblot analysis against anti-apoptotic Bcl-2 family proteins revealed that the expression levels of Bcl-2 and Bcl-xL were concentration-dependently decreased, while pro-apoptotic Bax protein was elevated its expression by carnosol treatment (Fig. 2b). As shown in Fig. 2c, carnosol increased expression level of p53, however, diminished protein expression level of Mdm2, a regulator of p53 through proteosomal degradation, in a concentration-dependent manner. These data suggest that carnosol induces apoptosis through the decrease of Bcl-2 and Bcl-xL and accumulation of p53 by elimination of Mdm2 in G361 cells.

Fig. 2
figure 2

Effect of carnosol on proteolytic cleavage of caspases and protein levels of anti- and pro-apoptotic Bcl-2 family and p53 in apoptosis of G361 cells. a, b The cellular levels of apoptosis marker proteins (cleaved caspase-9, -7, -3, and PARP) and anti- and pro-apoptotic proteins (Bax, Bcl-2, and Bcl-xL) in G361 cells after carnosol treatment (0, 20, 50, 100 μM) for 24 h. β-actin was as a loading control. c The expression levels of p53 and mdm2 in G361 cells with concentration-dependent carnosol treatment

Full size image

Inhibition of STAT3 activation and STAT3-mediated gene expression by carnosol

STAT3 plays a key role in cell growth and apoptosis through transcriptional activation of diverse genes in response to cell stimuli [33]. To further analyze the underlying mechanism in carnosol-induced apoptosis, the cellular levels of p-STAT3 and the activity of STAT3-stimulating protein kinases, Janus kinases (JAK) and Src, were examined by immunoblot analysis. Treatment of carnosol in G361 cells inhibited phosphorylation of STAT3 on tyrosine 705 residue (Fig. 3a), while p-JAK and JAK were not detected under the same condition (data not shown). In addition, we observed that carnosol inhibited DNA-binding activity of STAT3, which is assessed by EMSA (Fig. 3b). Then, we assessed the expression levels of survivin and D-series cyclins, the representative STAT3-regulated cell proliferative genes. As shown in Fig. 3c, the expression levels of Cyclin D1, D2, and D3 were gradually decreased by carnosol treatment in a concentration-dependent manner, and the cellular level of survivin was also decreased from 20 μM of carnosol treatment. These results suggest that carnosol treatment suppresses activity of Src kinase to activate STAT3 and this inhibition down-regulates STAT3-mediated gene expression in G361 cells.

Fig. 3
figure 3

Carnosol-induced inhibition of STAT3-mediated signaling and gene expression in G361 cells. a G361 cells were treated with indicated concentrations of carnosol. The cellular levels of p-STAT3 (Y705), STAT3, p-Src, and Src were determined by immunoblot. b The DNA-bound STAT3 was analyzed after carnosol treatment. c Immunoblot analysis was assessed to elucidate the expression levels of cyclin D1, D2, D3, and survivin after 24 h treatment with indicated concentrations of carnosol

Full size image

Involvement of ROS in carnosol-induced apoptosis

Since it is well known that the accumulation of intracellular ROS could induce apoptosis in various cancer cells, we examined the involvement of ROS in carnosol-induced apoptosis. The cells were treated with carnosol (20, 50, and 100 μM) for 12 h, and then we stained the cells with DCF-DA to measure intracellular total ROS (Fig. 4a, b). The carnosol-induced ROS generation was analyzed by fluorescence microscopy and flow cytometry. The treatment of carnosol in G361 cells led to the production of ROS in a concentration-dependent manner, especially, 100 μM of carnosol-treated G361 cells generated more ROS than H2O2-treated cells. However, carnosol-induced ROS generation was attenuated by pre-treatment of NAC (Fig. 5).

Fig. 4
figure 4

Increase in total ROS level after carnosol treatment in G361 cells. a Cells were treated with carnosol (20, 50 or 100 μM) for 12 h and cells were stained with 25 μM of DCF-DA. Then, intracellular accumulation of ROS was examined by fluorescence microscopy (×200) or analyzed by flow cytometry (b, left panel). H2O2 (500 μM)-treated cells were used as positive control. The increase in the intensity of DCF-DA fluorescence was compared to blank group, and plotted as a graph using geomean value (c, right panel). **p < 0.01 compared to control group

Full size image
Fig. 5
figure 5

NAC-dependent elimination of carnosol-induced ROS generation. Cells were treated with NAC (5 mM) for 12 h with or without carnosol treatment (100 μM), then ROS levels were measured by fluorescence microscopy (a) or by flow cytometry (b) after DCF-DA staining. c The percent intensity of ROS generation compared to non-stained blank was calculated using geomean value and plotted as a graph. The statistical significance was determined by Student’s T-test against non-treated control. *p < 0.01 compared to control group

Full size image

Prevention of carnosol-induced apoptosis by scavenging of ROS

To explore whether carnosol-induced apoptosis regulated by ROS, we assessed the amelioration of carnosol-induced apoptosis including cell viability, STAT3-mediated signal pathway, and proteolytic activation of caspases in NAC-treated G361 cells. Pre-treatment of NAC mitigated carnosol-induced changes on the phosphorylation of STAT3 and Src activity (Fig. 6a), DNA-binding activity of STAT3 (Fig. 6b), and level of cleaved caspases (Fig. 6c). The pre-treatment of NAC reduced expression level of p53 and cleaved PARP. These findings suggest that ROS play critical roles in carnosol-induced apoptosis in G361 cells.

Fig. 6
figure 6

The amelioration of carnosol-induced apoptosis by elimination of ROS. a The NAC-induced restoration in phosphorylation levels of STAT3 and Src (a), in DNA-bound STAT3 (b). c The diminution of cleaved apoptosis marker proteins and decrease of p53 level after NAC treatment in carnosol-treated G361 cells. d, e The decrease of apoptotic cell population by NAC treatment. The percent of apoptotic cell of each group was calculated and plotted. **p < 0.01 compared to control group

Full size image

Discussion

Carnosol, a phenolic diterpene in rosemary, has been reported to exert anti-cancer activities on various cancer cells [34, 35]. However, the biochemical basis of carnosol-induced anti-cancer mechanism remains elusive. In this study, we discovered that carnosol could induce apoptosis in human melanoma G361 cells in a concentration- and time-dependent manner. This apoptosis-inducing effect corresponds to the apoptosis by carnosol in other cancer cells [34, 36, 37]. In this study, we determined the apoptosis-inducing effect of carnosol on G361 cells and demonstrated that carnosol treatment would induce apoptosis through ROS-dependent inhibition of STAT3 signaling pathway.

The proteolytic activation of caspase-9 following the release of cytochrome c from mitochondria and activation of apaf-1 by intrinsic apoptotic signals activates downstream executioner caspases, such as caspase-3 and -7, then, initiates apoptosis [38]. During mitochondria-mediated apoptosis, the transition of mitochondrial membrane potential is an irreversible step and this process is regulated by anti- or pro-apoptotic BCL-2 family proteins [39]. According to current reports, the inhibition of Bcl-2 and Bcl-xL, anti-apoptotic BCL-2 family proteins, could facilitate apoptosis and be an overcoming strategy for drug resistance in human melanoma [40, 41]. From our results, the carnosol treatment revealed the proteolytic activation of caspase-3 and the cleavage of PARP. In addition, carnosol showed inhibition of Bcl-2 and Bcl-xL expression, while the expression level of Bax was increased. These results suggest that carnosol might cause apoptosis through the mitochondria-mediated intrinsic pathway.

The cellular level of tumor suppressor p53 is regulated by its cytosolic repressor, Mdm2, through proteasomal degradation [42]. Negative regulation of Mdm2 by chemicals stabilizes p53, and induces growth arrest and apoptosis in human melanoma cells [43, 44]. In the present study, the level of p53 was increased following the decrease of Mdm2 after carnosol treatment. These results suggest that accumulation of p53 through negative regulation of Mdm2 might contribute to carnosol-induced apoptosis.

The mitochondrion is a major subcellular organelle producing ROS as a by-product of the oxidative phosphorylation process. The numerous researchers have been examined chemicals from natural products for their anti-cancer activity [45, 46], and their large part of anti-cancer activity relies on positive- or negative-regulation of ROS level [47]. In our study, carnosol treatment generated ROS from the lowest dose, 20 μM, and this increase in cellular ROS level was analogous to 500 μM of H2O2-treated positive control group. Moreover, carnosol treatment-induced ROS generation in G361 cells was decreased by pre-treatment of anti-oxidant, NAC, and this pre-treatment of the cells with NAC attenuated carnosol-induced cleavage of caspase-3 and PARP. These evidences showed that ROS plays pivotal roles in carnosol-induced apoptosis.

Separately, several reports support that the expression of Bcl-2 and Bcl-xL is mediated by STAT3 in human melanoma models [48], and abnormal activation of STAT3 exerts effects on proliferation, invasion and migration of melanoma cells [49]. Moreover, the inhibiting STAT3 signal exhibits cytotoxicity and down-regulated expression of Bcl-xL and metalloprotease-2 and -9 in melanoma cells [50]. We assessed the carnosol-induced changes on STAT3 signal pathway in G361 cells. STAT3 is activated by phosphorylation on tyrosine 705 residue by upstream protein kinases, such as JAK2 [51], and Src kinase [52], followed by homodimer formation. Our results revealed that carnosol treatment decreased phosphorylation on STAT3 by Src kinase, and this inhibition on STAT3 phosphorylation reduced STAT3-regulated genes including Cyclin D1, D2, D3, and survivin. However, we could not detect phosphorylation of JAK2 in this model (data not shown). These findings suggest that decreased expression of Cyclin D1, D2, D3, and survivin by carnosol treatment is connected to carnosol-induced apoptosis.

Taken together, we demonstrated that carnosol treatment-induced accumulation of p53, inactivation of STAT3-mediated signal, generation of ROS, and caspase cascade cause apoptosis in human melanoma G361 cells. Moreover, we described our proposed action mechanism of carnosol in Fig. 7. Our findings would provide a molecular basis of carnosol-induced apoptosis in human melanoma cells, and might contribute to development of anti-melanoma drugs using carnosol.

Fig. 7
figure 7

Schematic diagram of possible carnosol-induced apoptosis pathway in human melanoma G361 cells

Full size image