Pyruvic acid/ethyl pyruvate inhibits melanogenesis in B16F10 melanoma cells through PI3K/AKT, GSK3β, and ROS‐ERK signaling pathways

Abstract

Melanin is the main product of human melanocytes and functions to protect skin from ultraviolet (UV) radiation while conferring color to skin and hair. Tyrosinase is the rate‐limiting enzyme for melanin synthesis along with tyrosinase‐related protein (TRP)‐1 and TRP‐2. Microphthalmia‐associated transcription factor regulates tyrosinase gene expression and is in turn regulated by extracellular signal‐regulated kinase (ERK), phosphoinositide 3‐kinase (PI3K)/AKT, and glycogen synthase kinase (GSK)3β signaling pathways. Pyruvic acid (PA) is an energy source for ATP synthesis in the tricarboxylic acid cycle that also acts as a reactive oxygen species (ROS) scavenger. As UV irradiation induces melanin synthesis and ROS generation, we speculated that PA or ethyl pyruvate (EP), a stable form of pyruvate, regulates melanogenesis. B16F10 melanoma cells served as a melanin synthesis model. Treatment with PA or EP suppressed melanin synthesis while increasing intracellular ROS levels, which was accompanied by increased ERK phosphorylation in the case of EP treatment. PA and EP induced GSK3β phosphorylation and activated PI3K/AKT signaling, leading to decreased melanin synthesis. These results indicate that PA and EP inhibit melanogenesis via PI3K/AKT and GSK3β signaling and targeting the ERK and GSK3β pathways, respectively. Thus, PA and EP can potentially be used for treatment of hyperpigmentation disorders.

1 INTRODUCTION

Melanin is the molecule responsible for skin color (Seo, Sharma, & Sharma, 2003) that protects skin from ultraviolet (UV) radiation‐induced damage (Chen et al, 2015). The pigment is synthesized in melanocytes in an organelle known as the melanosome, a process referred to as melanogenesis (Videira, Moura, & Magina, 2013). Tyrosinase is a critical enzyme for melanogenesis and along with tyrosinase‐related protein (TRP)‐1/2 catalyzes the rate‐limiting steps in melanin synthesis (Eves, MacNeil, & Haycock, 2006; Jiménez‐Cervantes et al., 2001). Microphthalmia‐associated transcription factor (MITF) is the transcription factor regulating tyrosinase expression and is thus a key regulator of melanogenesis (Chang, 2012). MITF expression and activity are controlled by various upstream factors. Alpha melanocyte‐stimulating hormone (α‐MSH) released by keratinocytes in response to UV irradiation (Thody & Graham, 1998) binds to melanocortin‐1 receptor and stimulates cAMP (Bertolotto et al., 1998), which activates protein kinase (PKA) and results in the phosphorylation of cAMP response element‐binding protein (CREB). Phosphorylated CREB directly induces the transcription of MITF to promote melanogenesis (Flaherty, Hodi, & Fisher, 2012).

In addition to the PKA/CREB signaling pathway, MITF expression and activity are regulated by extracellular signal‐regulated kinase (ERK), a mitogen‐activated protein kinase (MAPK) family member (Peng, Lin, Wang, Shih, & Chou, 2014) that phosphorylates MITF at S73 and targets the protein for degradation by the proteasome (Lee, Lee, Chang, & Lee, 2015; Wu, Lin, Yang, Weng, & Tsai, 2011). Phosphoinositide 3‐kinase (PI3K)/AKT (Chae et al., 2017; Hwang et al., 2016) and glycogen synthase kinase 3 beta (GSK3β; Shin et al., 2015) also negatively regulate MITF activity and thereby suppress melanogenesis.

UV radiation induces melanogenesis, which is accompanied by an increase in reactive oxygen species (ROS) production (Dong et al., 2010; Mastore, Kohler, & Nappi, 2005). Pyruvic acid (PA) is a simple three‐carbon α‐keto monocarboxylic acid that plays a critical role in glycol metabolism and is a component of the tricarboxylic acid cycle (Fink, 2004). Pyruvate is also an effective ROS scavenger and has beneficial effects in diseases caused by increased ROS generation (Salahudeen, Clark, & Nath, 1991; Varma, Devamanoharan, & Ali, 1998). Aqueous solutions of pyruvate have low stability; the ethyl ester form, ethyl pyruvate (EP), is more stable, and has anti‐inflammatory and anticoagulant properties in addition to its scavenging function (Fink, 2007).

In this study, we tested the hypothesis that PA or EP modulates melanogenesis using B16F10 melanoma cells.

2 RESULTS

2.1 PA and EP inhibit melanogenesis

To determine the safe concentrations of PA and EP, the cytotoxicity of the two compounds was evaluated in B16F10 melanoma cells with the 3‐(4,5‐dimethylthiazol‐2‐yl)‐2,5‐diphenyltetrazolium bromide (MTT) assay. Cells were treated with PA and EP concentrations of 0.1–15 mM. A dose‐dependent cytotoxicity was observed at concentrations higher than 10 mM (Figure 1a,b). Although 5 mM of PA and EP significantly decreased cell viability, the cell growth curve showed that there was no notable cytotoxicity (Figure 1c). Therefore, 5 mM of PA and EP was used in the following experiments.

Figure 1

PA and EP dose‐dependently decreased melanogenesis. B16F10 melanoma cells were treated with different concentrations of PA and EP, with or without MSH and forskolin (Fsk). Cells were collected after 48 hr and used in subsequent assays. Ct represents the control group without any treatment. (A and B) Effect of PA or EP on B16F10 melanoma cell viability. (C) B16F10 melanoma cell growth curve by 5 mM PA and EP treatment. (D and E), PA and EP dose‐dependently decreased melanin synthesis in B16F10 melanoma cells. Data were normalized to total protein content. (F and G) Melanogenesis promoters α‐MSH and Fsk were used as positive controls. Treatment with α‐MSH and Fsk induced melanin synthesis, which was suppressed by 5 mM PA or EP treatment. Data were normalized to total protein content. Data represent mean ± SD (n ≥ 3). *p < 0.05, **p < 0.01, ***p < 0.005 compared with Ct. ^#p < 0.05, ^##p < 0.01, ^###p < 0.005 compared with positive control

Next, we examined the effect of PA and EP on melanin synthesis. Here, we normalized the intracellular melanin content with cellular total protein to avoid the effect of cell number decrease and found that intracellular melanin levels were decreased in a dose‐dependent manner in the presence of either compound (Figure 1d,e).

Furthermore, cells treated with the melanogenesis‐promoting molecules α‐MSH and forskolin (Lehraiki et al., 2014) produced more melanin (Figure 1f,g). Howeve, this increase in melanogenesis was down‐regulated by PA and EP. Therefore, PA and EP effectively inhibited melanogenesis.

2.2 PA and EP inhibit tyrosinase activity and expression

Since tyrosinase is the main enzyme catalyzing melanin synthesis, we measured tyrosinase activity in B16F10 cells based on the catalysis of the tyrosinase substrate 3,4‐dihydroxy‐l‐phenylalanine (l‐DOPA) to melanin. Cells were treated with 5 mM PA or EP for 48 hr, and the melanin production at 30 min was considered to represent tyrosinase activity. Both PA (Figure 2a) and EP (Figure 2b) down‐regulated tyrosinase activity, as evidenced by the decrease in melanin levels.

Figure 2

PA and EP decreased tyrosinase activity and gene expression. Cells were treated with PA and EP for 24 or 48 hr. After treatment, cells were collected for subsequent assays. Ct represents the control group without any treatment. (A and B) Tyrosinase activity was evaluated with 50 μg of cell lysate. The absorbance increase at 30 min based on the production of melanin was measured and calculated to represent tyrosinase activity. PA and EP dose‐dependently decreased the tyrosinase activity. (C) Cells treated for 48 hr were collected for western blot. Tyrosinase, TRP‐1, and TRP‐2 protein levels were down‐regulated by 5 mM PA or EP treatment. (D and E) Tyrosinase mRNA was suppressed by PA and EP after 24 and 48 hr of treatment. These mRNA quantities were normalized by GAPDH quantity. Data represent mean ± SD (n ≥ 3). ***p < 0.005

Given our observation that PA and EP treatment decreased tyrosinase activity, we examined tyrosinase expression by western blotting. Cells treated with PA and EP were collected at 24 and 48 hr. We observed the reducing effect of PA and EP on tyrosinase expression at 48 hr (Figure 2c), while the difference at 24 hr was not clear (data not shown). TRP‐1 and TRP‐2 (also known as dopachrome tautomerase, DCT), along with tyrosinase, regulate melanin synthesis; therefore, we examined the expression of TRP‐1 and TRP‐2 by western blotting. The expression of both proteins was decreased by PA and EP treatment.

To confirm that the melanogenesis suppression by PA and EP was caused by down‐regulation of tyrosinase gene expression, the mRNA of tyrosinase was measured. The relative quantity of tyrosinase mRNA was decreased in the presence of PA or EP for 24 and 48 hr. Taken together, the results indicated that PA and EP inhibit melanogenesis by negatively regulating the expression of melanin synthesis‐related genes.

2.3 PA and EP regulate MITF gene expression

MITF regulates tyrosinase and TRP‐1 and TRP‐2 expression, indicating that it plays a significant role in melanogenesis. Therefore, we examined MITF expression by western blot. Cells treated with PA and EP were collected after 48 hr. The tyrosinase expression was the same as in Figure 2c, and MITF was significantly decreased by the treatment (Figure 3a).

Figure 3

PA or EP regulated MITF time‐dependent gene expression. B16F10 melanoma cells were treated with PA and EP for 48 hr, and cells were collected at 0, 1, 3, 6, 9, 12, 24, and 48 hr. (A) Although tyrosinase was decreased, MITF was insignificantly down‐regulated by PA and EP treatment at 48 hr. (B and C) After collection of cells, mRNA quantity was measured by RT‐PCR. mitf mRNA in the nontreatment group (Ct) was increased at 3 hr by refeeding fresh medium and returned to a normal level after 6 hr without MSH or forskolin treatment. Both PA and EP reduced the increase in mitf mRNA, especially at 3 hr, but not at 9 hr. All data were normalized by GAPDH mRNA quantity. Data represent mean ± SD (n ≥ 3). *p < 0.05, **p < 0.01, ***p < 0.005 compared with Ct at the same time point

To investigate whether MITF was involved in the suppression of melanogenesis by PA or EP, we further analyzed mitf mRNA quantity. After refreshing the medium with PA or EP, which did not contain any promoting agents like MSH or forskolin, the cells were collected after 0, 1, 3, 6, 9, 12, 24, and 48 hr. The mRNA quantity at 0 hr without any treatment was considered 1, and the relative value of that at other time points was calculated. The mitf mRNA in nontreated cells was increased at 3 hr and reduced to a normal level after 6 hr, a phenomenon that has never been mentioned previously (Figure 3b,c). Consistent with our expectations, PA and EP reduced the mitf mRNA quantity at almost all time points, except at 9 hr. These results suggested that MITF was involved in PA‐ and EP‐induced melanogenesis down‐regulation.

2.4 ROS‐ERK signaling mediates the effects of PA and EP on melanogenesis

ERK phosphorylation has been shown to induce MITF degradation (Song et al., 2015) and lead to the inhibition of melanogenesis. Both PA and EP induced ERK phosphorylation without altering ERK gene expression after 24 hr (Figure 4a). Treatment of cells with the ERK inhibitor U0126 increased melanogenesis (Figure 4b,c). When U0126 was coadministered with PA or EP, melanin synthesis was more reduced than with inhibitor treatment alone. On the other hand, PA alone reduced the melanin content to the same level as PA and U0126, although there was a significant difference between levels following EP treatment and EP/U0126 cotreatment.

Figure 4

PA and EP regulated melanogenesis through ROS‐ERK signaling. (A) ERK phosphorylation was increased by PA and EP treatment, whereas ERK gene expression was unaffected. (B and C) B16F10 melanoma cells were treated with PA, EP, and U0126 for 48 hr; melanin content was increased by U0126, but this effect was reversed by PA and EP treatment. (D) After treatment with PA and EP for 48 hr, cells were treated with DCFDA for 1 hr, and fluorescence was measured (excitation/emission =495/529 nm). PA and EP increased ROS generation in B16F10 cells. Data represent mean ± SD (n ≥ 3). *p < 0.05, **p < 0.01, ***p < 0.005

PA and EP are both effective intracellular ROS scavengers (Salahudeen et al., 1991; Varma et al., 1998), and ROS is an activator of the MAPK signaling pathway (Kim et al., 2014). Therefore, we examined intracellular ROS levels in cells treated with PA and EP. Both compounds increased ROS production (Figure 4d), which was consistent with the observed increase in ERK phosphorylation.

These results suggested that the inhibition of melanogenesis by PA was independent of the ERK pathway although PA increased ERK phosphorylation and ROS generation. In contrast, EP appeared to partially target the ERK pathway to inhibit melanin synthesis.

2.5 GSK3β is involved in the regulation of melanogenesis by PA and EP

PI3K activates AKT, which then inhibits GSK3β through phosphorylation at Ser9 (Cross, Alessi, Cohen, Andjelkovich, & Hemmings, 1995). Recent studies have shown that GSK3β regulates MITF activity (Shin et al., 2015) but the detailed mechanisms remain unclear. To determine the contribution of PI3K/AKT and GSK3β signaling to melanogenesis, we treated B16F10 cells with the PI3K inhibitor LY294002 and GSK3β inhibitor (2Z,3E)‐6‐bromoindirubin‐3'‐oxime (BIO). Treatment with LY294002 and BIO increased melanin content in a dose‐dependent manner (Figure 5a,b). In addition, tyrosinase and MITF gene expression increased with melanin synthesis (Figure 5c,d). PI3K inhibitor treatment decreased phosphorylated (p‐)AKT level, whereas GSK3β inhibitor reduced GSK3β and p‐GSK3β levels. Thus, activated AKT induces GSK3β phosphorylation, which has an inhibitory effect on melanogenesis. We then cotreated cells with PA or EP and the inhibitors and found that PA decreased melanin content to a level similar to that of the control and PI3K inhibitor groups but not the GSK3β inhibitor group; the same trend was observed in cells treated with EP (Figure 5e,f). PA and EP also increased p‐GSK3β and weakly increased p‐AKT^Ser473 (Figure 5g). These results suggested that PI3K/AKT and GSK3β suppress melanin synthesis and that GSK3β is an important Mediator of the inhibitory effects of PA and EP on melanogenesis.

Figure 5

GSK3β plays a key role in the regulation of melanogenesis by PA and EP. Cells were treated with different concentrations of LY294002 and BIO (1, 5, and 10 μM) for 48 hr. (A and B) Inhibition of PI3K and GSK3β (with LY294002 and BIO, respectively) dose‐dependently increased melanin synthesis. (C) Effect of BIO on GSK3β phosphorylation and melanogenesis‐related gene expression. (D) Effect of LY294002 on AKT phosphorylation and melanogenesis‐related gene expression. (E and F) Cells treated with CA were cotreated with BIO or LY294002 for 48 hr; the absorbance of the lysates at 405 nm was measured to determine melanin content. Untreated cells served as the control group (Ct). (G) Cells were treated with PA and EP for 6 and 12 hr; lysates were then analyzed by western blotting. GSK3β phosphorylation was increased by PA and EP treatment, whereas AKT phosphorylation at Ser473 was increased by PA treatment. Data represent mean ± SD (n ≥ 3). *p < 0.05, **p < 0.01, ***p < 0.005

3 DISCUSSION

The results of this study demonstrate for the first time that PA and EP inhibit melanogenesis. In cell metabolism, pyruvate is an important intermediate product of glycolysis that is used for ATP synthesis (Fink, 2004). We found that PA, EP, and even sodium pyruvate increased ATP production (Supporting Information Figure S1a), whereas only PA and EP suppressed melanogenesis (Figure 1d,e and Supporting Information Figure S2a), indicating that the latter effect was independent of ATP production. PA and SP (sodium pyruvate) hydrolyze to pyruvate in aqueous solution, indicating that pyruvate alone does not regulate melanogenesis. We also used NaOH to neutralize PA and obtained the same results as for PA (Supporting Information Figure S2b), demonstrating that the negative regulation of melanin synthesis by PA is not through acidification. Therefore, the effect of PA on down‐regulating melanogenesis requires pyruvic acid itself.

Figure 6

Signaling pathway of PA/EP‐induced melanogenesis down‐regulation in B16F10 melanoma cells

Tyrosinase is the main enzyme for melanin synthesis, and its transcription is regulated by MITF (Chang, 2012; Eves et al., 2006; Jiménez‐Cervantes et al., 2001). In this study, we found that PA and EP treatment suppressed MITF and consequently tyrosinase gene expression, leading to a decrease in melanin synthesis.

MITF gene expression is regulated by many factors, including CREB, PI3K/AKT (Chae et al., 2017; Hwang et al., 2016), GSK3β (Shin et al., 2015), and MAPK (ERK, c‐Jun N‐terminal kinase, and p38; Peng et al., 2014). In our study, ERK phosphorylation was increased by PA and EP treatment, which corresponded to a decrease in MITF protein level and an increase in ROS generation. The latter result contradicts the reported role of PA and EP as ROS scavengers (Salahudeen et al., 1991; Varma et al., 1998). ERK inhibitor treatment increased melanin synthesis, indicating the role of the ERK pathway in inhibiting melanogenesis. Although the coadministration of PA or EP prevented melanin synthesis, there was a significant increase in melanin synthesis following EP/U0126 cotreatment. It appears that ERK was not involved in suppression of melanogenesis by PA and EP, but the increase in p‐ERK and ROS corresponded with melanin synthesis. Thus, we suspect that ROS‐ERK signaling was not the main regulatory pathway and only partially involved in the down‐regulation of melanogenesis by PA and EP.

Inhibition of PI3K increased melanin synthesis, indicating that PI3K/AKT signaling negatively regulates melanogenesis. On the other hand, GSK3β inhibition also increased melanogenesis. PI3K/AKT is located upstream of GSK3β; thus, our results indicated that the PI3K/AKT‐GSK3β axis mediates the effects of PA and EP on melanogenesis. The melanin content following cotreatment with PI3K inhibitor/PA or EP was not only significantly higher than that following PA or EP treatment alone but also lower than that following inhibitor treatment alone. On the other hand, the results of cotreatment with GSK3β inhibitor/PA or EP were not significantly different from those of inhibitor treatment alone. Western blot also revealed that PA and EP induced GSK3β phosphorylation, indicating that GSK3β phosphorylation is critical in PA‐ and EP‐induced down‐regulation of melanogenesis.

In recent studies, AKT also directly induced MITF phosphorylation at Ser510; p‐MITF is degraded by protease (Wang et al., 2016). This could explain why the PI3K inhibitor induced more melanin synthesis than the GSK3β inhibitor.

In conclusion, the results of this study demonstrate that PA and EP negatively regulate melanin production through ERK, PI3K/AKT, and GSK3β signaling (Figure 6) and have the potential to be used to treat hyperpigmentation disorders.

4 EXPERIMENTAL PROCEDURES

4.1 Materials

B16F10 melanoma cells were obtained from RIKEN Institute of Physical and Chemical Research Cell Bank (Tsukuba, Japan). PA and EP were purchased from Wako Pure Chemical Industries (Osaka, Japan). Roswell Park Memorial Institute (RPMI)‐1640 medium, MTT, and l‐DOPA were purchased from Sigma‐Aldrich (St. Louis, MO, USA). Anti‐tyrosinase and anti‐TRP‐1 and ‐2 antibodies were purchased from Santa Cruz Biotechnology (Santa Cruz, CA, USA). Antibodies against MITF, total GSK3β, p‐GSK3β^Ser9, AKT, p‐AKT^Thr308, p‐AKT^Ser473, and total β‐actin were obtained from Cell Signaling Technology (Tokyo, Japan).

4.2 Cell culture

B16F10 cells were cultured in RPMI‐1640 medium supplemented with 10% fetal bovine serum at 37°C and 5% CO₂. After culturing in 10‐cm dishes from frozen stocks, cells were seeded at a density of 3.4 × 10⁴ cells/well in a 6‐well plate, at 3.0 × 10³ cells/well in a 24‐well plate, or at 1.0 × 10³ cells/well in a 96‐well plate. Cells were allowed to attach to the bottom of the plate for 24 hr before experiments.

4.3 MTT assay

B16F10 cells were seeded in a 96‐well plate. After treatment for 48 hr with PA or EP, the medium was replaced with fresh medium containing 500 μg/ml MTT reagent, and the cells were incubated at 37°C for 6 hr. A 10% SDS solution was added, and plates were maintained at room temperature overnight. The absorbance at 570 nm was measured with a microplate reader (Tecan, Kawasaki, Japan).

4.4 Measurement of melanin content

Melanin content was measured as previously described (Lehraiki et al., 2014). B16F10 cells were seeded in a 6‐well plate. After treatment for 48 hr with PA or EP, cells were collected by trypsinization and resuspended in 100 μl of 1 N NaOH. After heating at 80°C for 1 hr, the absorbance at 405 nm was measured on a microplate reader. Melanin content was normalized to total protein content.

4.5 Intracellular tyrosinase activity

Tyrosinase activity was estimated by measuring the rate of l‐DOPA oxidation (Lee et al., 2013). B16F10 cells were seeded in a 6‐well plate and treated with PA and EP for 72 hr. The cells were collected by trypsinization, resuspended in phosphate buffer containing 10% Triton X‐100, sonicated, and centrifuged at 18000 g for 20 min. The supernatant was collected and assayed for total protein, and 50 μg was mixed with 2 μl of 10% (m/v) l‐DOPA in phosphate buffer followed by incubation at 37°C. The absorbance was measured at 475 nm at 0 and 30 min. Melanin (Sigma‐Aldrich) was used for standard concentration. Tyrosinase activity was calculated with the following formula: quantity of melanin at 30 min – quantity of melanin at 0 min.

4.6 Detection of intracellular ROS

B16F10 cells were treated with PA and EP for 48 hr, then washed twice with PBS, and treated with 2',7'‐dichlorodihydrofluorescein diacetate (DCFDA) for 1 hr. The cells were washed with PBS to remove the DCFDA, and fluorescence (excitation/emission = 495/529 nm) was measured on a microplate reader.

4.7 Western blotting

B16F10 cells (2.5 × 10⁵) were seeded in a 6‐cm dish. After treatment with samples for 48 hr, cells were collected in radioimmunoprecipitation assay buffer, sonicated, and centrifuged at 10,000 g for 10 min. The supernatant was collected, and protein assay was performed with the Pierce^TM BCA Protein Assay Kit (Thermo Fisher Scientific, Waltham, MA, USA). Next, 20 μg protein from each sample was separated by sodium dodecyl sulfate‐polyacrylamide gel electrophoresis (SDS‐PAGE) and transferred to a polyvinylidene difluoride membrane that was blocked for 1 hr in 2% bovine serum albumin (BSA) and then incubated overnight at 4°C with primary antibodies against MITF, ERK, p‐ERK, AKT, p‐AKT^Thr308, p‐AKT^Ser473, GSK3β, and p‐GSK3β^Ser9, tyrosinase, and TRP‐1 and TRP‐2. This was followed by incubation for 60 min at room temperature with appropriate secondary antibodies. The membrane was stained with LuminoGLO (Cell Signaling Technology), and protein bands were visualized with an AE‐9300H EZ‐Capture MG imager (ATTO Corporation, Tokyo, Japan).

4.8 Real‐time PCR

B16F10 cells were seeded in a 6‐well plate, and after PA or EP treatment for 0, 1, 3, 6, 9, 12, 24, or 48 hr, RNA was extracted from the cells using RNAiso Plus (Takara Bio, Otsu, Japan). Next, the RNA was reverse transcribed to cDNA, which was used as a template for real‐time PCR amplification of mitf (forward: GTGAGATCCAGAGTTGTCGT; reverse: AGTACA GGAGCTGGAGATG) and tyrosinase (forward: TGACTCTTGGAGGTAGCTGT; reverse: AACAATGTCCCAAGTACAGG) genes. GAPDH (forward: TGCCGTTGAATTTGC CGTGAGT; reverse: TGGTGAAGGTCGGTGTGAACGG) was used as an internal reference for normalization.

4.9 Statistical analysis

Results are reported as the mean ± SD. Group means were compared with Student's t test, and differences were considered significant at p < 0.05.

ACKNOWLEDGMENTS

This work was supported in part by Grants‐in‐Aid for Scientific Research and Education from the University of Tsukuba, Japan.