Screening of an Epigenetic Drug Library Identifies 4-((hydroxyamino)carbonyl)-N-(2-hydroxyethyl)-N-Phenyl-Benzeneacetamide that Reduces Melanin Synthesis by Inhibiting Tyrosinase Activity Independently of Epigenetic Mechanisms

The aim of this study was to identify novel antimelanogenic drugs from an epigenetic screening library containing various modulators targeting DNA methyltransferases, histone deacetylases, and other related enzymes/proteins. Of 141 drugs tested, K8 (4-((hydroxyamino)carbonyl)-N-(2-hydroxyethyl)-N-phenyl-benzeneacetamide; HPOB) was found to effectively inhibit the α-melanocyte-stimulating hormone (α-MSH)-induced melanin synthesis in B16-F10 murine melanoma cells without accompanying cytotoxicity. Additional experiments showed that K8 did not significantly reduce the mRNA and protein level of tyrosinase (TYR) or microphthalmia-associated transcription factor (MITF) in cells, but it potently inhibited the catalytic activity TYR in vitro (IC50, 1.1–1.5 µM) as compared to β-arbutin (IC50, 500–700 µM) or kojic acid (IC50, 63 µM). K8 showed copper chelating activity similar to kojic acid. Therefore, these data suggest that K8 inhibits cellular melanin synthesis not by downregulation of TYR protein expression through an epigenetic mechanism, but by direct inhibition of TYR catalytic activity through copper chelation. Metal chelating activity of K8 is not surprising because it is known to inhibit histone deacetylase (HDAC) 6 through zinc chelation. This study identified K8 as a potent inhibitor of cellular melanin synthesis, which may be useful for the treatment of hyperpigmentation disorders


Introduction
Human skin color is controlled by the composition and distribution of various chromophores such as melanin, hemoglobin, and carotenoids [1]. Melanin is a dark polymer pigment synthesized by melanocytes [2]. Melanosomes containing melanin migrate from melanocytes to neighboring keratinocytes via dendrites, resulting in the distribution of melanin throughout the epidermis [3]. Melanin plays an important role in protection against carcinogenic ultraviolet radiation [4]; therefore, melanin metabolism is critical in epidermal homeostasis [5,6].
Dysfunctions associated with melanin synthesis cause clinically relevant pigmentation disorders that can be congenital or acquired, cutaneous or systemic, temporary or permanent, and related to hypo- In the preliminary screening assay, each drug was tested for its effects on melanin synthesis in murine melanoma-derived B16-F10 cells cultured in vitro. After treating the cells with a drug solution (final concentration, 1.0 µM) or vehicle, melanin synthesis was stimulated with α-MSH (final concentration, 100 nM). As shown in Figure 1a, α-MSH increased cellular melanin synthesis, as monitored by optical density (OD) at 400 nm, and the induction was modulated by the various drugs. The effects of the drugs on cell viability were also determined as shown in Figure 1b. Comparing activity and cytotoxicity, drugs H4 and K8 were selected for additional study because they markedly inhibited cellular melanin synthesis without reducing cell viability at the tested concentration. monitored by optical density (OD) at 400 nm, and the induction was modulated by the various drugs. The effects of the drugs on cell viability were also determined as shown in Figure 1b. Comparing activity and cytotoxicity, drugs H4 and K8 were selected for additional study because they markedly inhibited cellular melanin synthesis without reducing cell viability at the tested concentration. Figure 1. Effects of drugs on melanin synthesis and viability in B16-F10 cells. (a) Cells were treated with each drug at 1.0 µM and stimulated with 100 nM α-melanocyte-stimulating hormone (α-MSH) for 72 h for relative melanin synthesis assay (mean ± SD, n = 3). ** p < 0.01, * p < 0.05 versus α-MSH control. (b) Cells were treated with each drug at 1.0 µM for 48 h for relative viability assay (mean ± SD, n = 3). ** p < 0.01, * p < 0.05 versus control. Open and closed histograms represent the absence and (a) Cells were treated with each drug at 1.0 µM and stimulated with 100 nM α-melanocyte-stimulating hormone (α-MSH) for 72 h for relative melanin synthesis assay (mean ± SD, n = 3). ** p < 0.01, * p < 0.05 versus α-MSH control. (b) Cells were treated with each drug at 1.0 µM for 48 h for relative viability assay (mean ± SD, n = 3). ** p < 0.01, * p < 0.05 versus control. Open and closed histograms represent the absence and presence of α-MSH addition, respectively. Drugs that inhibited melanin synthesis without cytotoxicity are highlighted in red.

Activity Versus Cytotoxicity for H4 and K8
These two drugs, H4 (CBHA) and K8 (HPOB) were compared in the next experiment for their antimelanogenic activity and cytotoxicity. B16-F10 cells were treated with drug at varied concentrations and stimulated with 100 nM α-MSH for 72 h. As shown in Figure 2a,b, the increase in OD 400 due to melanin synthesis was attenuated by both drugs in a dose-dependent manner. However, drug H4 showed significant cytotoxicity above 3 µM, whereas K8 did not show toxic effects up to 10 µM (Figure 2c,d). Thus, K8 was selected for further mechanistic study. presence of α-MSH addition, respectively. Drugs that inhibited melanin synthesis without cytotoxicity are highlighted in red.

Activity Versus Cytotoxicity for H4 and K8
These two drugs, H4 (CBHA) and K8 (HPOB) were compared in the next experiment for their antimelanogenic activity and cytotoxicity. B16-F10 cells were treated with drug at varied concentrations and stimulated with 100 nM α-MSH for 72 h. As shown in Figure 2a,b, the increase in OD 400 due to melanin synthesis was attenuated by both drugs in a dose-dependent manner. Effects of H4 and K8 on the melanin synthesis and viability of B16-F10 cells. Cells were treated with each drug at various concentrations and stimulated with 100 nM α-melanocytestimulating hormone (α-MSH) for 72 h for melanin synthesis assay (a,b) or for 48 h for relative viability assay (c, d). Data are presented as percentages relative to the control value (mean ± SD, n = 4 for a and b; n = 3 for (c,d)). ** p < 0.01. Chemical structures of H4 (N-hydroxy-3-[3-(hydroxyamino)-3oxo-1-propen-1-yl-benzamide; CBHA) and K8 (4-((hydroxyamino)carbonyl)-N-(2-hydroxyethyl)-Nphenyl-benzeneacetamide; HPOB) are shown. Open and closed histograms represent the absence and presence of α-MSH addition, respectively.

Effects of K8 on Melanin Content and TYR Activity in Cells
Additional assays were undertaken to separately quantify extracellular and intracellular melanin levels of cells treated with K8. As shown in Figure 3a,b, both the extracellular and intracellular melanin content, normalized for total cellular protein content, were increased by α-MSH treatment, which was suppressed by K8 in a dose-dependent manner (1-10 µM).
Changes in melanin content of cells may be closely related to TYR activity in cells. In order to directly examine this possibility, cellular TYR activity was assessed after treating different concentrations of drugs and a fixed amount of α-MSH. As a result, α-MSH increased cellular TYR activity, and the change was attenuated by K8 ( Figure 3c). The less expected observation was that the TYR activity of the cells was only partially lowered by 10 µM K8 that suppressed almost all melanin synthesis as observed above. Effects of H4 and K8 on the melanin synthesis and viability of B16-F10 cells. Cells were treated with each drug at various concentrations and stimulated with 100 nM α-melanocyte-stimulating hormone (α-MSH) for 72 h for melanin synthesis assay (a,b) or for 48 h for relative viability assay (c, d). Data are presented as percentages relative to the control value (mean ± SD, n = 4 for a and b; n = 3 for (c,d)). ** p < 0.01. Chemical structures of H4 (N-hydroxy-3-[3-(hydroxyamino)-3-oxo-1-propen-1-yl-benzamide; CBHA) and K8 (4-((hydroxyamino)carbonyl)-N-(2-hydroxyethyl)-N-phenyl-benzeneacetamide; HPOB) are shown.
Open and closed histograms represent the absence and presence of α-MSH addition, respectively.

Effects of K8 on Melanin Content and TYR Activity in Cells
Additional assays were undertaken to separately quantify extracellular and intracellular melanin levels of cells treated with K8. As shown in Figure 3a,b, both the extracellular and intracellular melanin content, normalized for total cellular protein content, were increased by α-MSH treatment, which was suppressed by K8 in a dose-dependent manner (1-10 µM).
Changes in melanin content of cells may be closely related to TYR activity in cells. In order to directly examine this possibility, cellular TYR activity was assessed after treating different concentrations of drugs and a fixed amount of α-MSH. As a result, α-MSH increased cellular TYR activity, and the change was attenuated by K8 ( Figure 3c). The less expected observation was that the TYR activity of the cells was only partially lowered by 10 µM K8 that suppressed almost all melanin synthesis as observed above. , cells were stimulated with 100 nM α-MSH for 60 h in the presence or absence of K8 at the indicated concentrations. Extra-and intracellular melanin levels were determined using the conditioned medium and cell extracts, respectively, and the values were normalized to total cellular protein content (n = 3). In (c), cells were stimulated with 100 nM α-MSH in the presence or absence of K8 at various concentrations for 24 h. Cell lysates were used for the enzyme activity assay (n = 4). Data are presented as percentages relative to control value (mean ± SD). ** p < 0.01. Open and closed histograms represent the absence and presence of α-MSH addition, respectively.

Effects of K8 on the mRNA and the Protein Expression Levels of TYR, TRP1, and DCT
We further examined the effects of K8 on the gene expression of TYR and other related enzymes involved in melanogenesis. Expression levels of the mRNAs for TYR, tyrosinase-related protein 1 (TYRP1), and dopachrome tautomerase (DCT) were assessed in B16-F10 cells by quantitative reverse transcription polymerase chain reaction (qRT-PCR) using to glyceraldehyde 3-phosphate dehydrogenase (GAPDH) as a control (Figure 4a-c). α-MSH significantly increased the mRNA levels of TYR and TRP1, whereas the change in DCT was insignificant. mRNA levels of TYR, TYRP1, and DCT were not significantly affected by K8.
Western blotting showed that α-MSH also increased the protein levels of TYR, TYRP1, and DCT in B16-F10 cells (Figure 4d-f). K8 did not significantly attenuate the increase in the protein levels of TYR, TRP1, and DCT caused by α-MSH stimulation. In (a,b), cells were stimulated with 100 nM α-MSH for 60 h in the presence or absence of K8 at the indicated concentrations. Extra-and intracellular melanin levels were determined using the conditioned medium and cell extracts, respectively, and the values were normalized to total cellular protein content (n = 3). In (c), cells were stimulated with 100 nM α-MSH in the presence or absence of K8 at various concentrations for 24 h. Cell lysates were used for the enzyme activity assay (n = 4). Data are presented as percentages relative to control value (mean ± SD). ** p < 0.01. Open and closed histograms represent the absence and presence of α-MSH addition, respectively.

Effects of K8 on the mRNA and the Protein Expression Levels of TYR, TRP1, and DCT
We further examined the effects of K8 on the gene expression of TYR and other related enzymes involved in melanogenesis. Expression levels of the mRNAs for TYR, tyrosinase-related protein 1 (TYRP1), and dopachrome tautomerase (DCT) were assessed in B16-F10 cells by quantitative reverse transcription polymerase chain reaction (qRT-PCR) using to glyceraldehyde 3-phosphate dehydrogenase (GAPDH) as a control (Figure 4a-c). α-MSH significantly increased the mRNA levels of TYR and TRP1, whereas the change in DCT was insignificant. mRNA levels of TYR, TYRP1, and DCT were not significantly affected by K8.
Western blotting showed that α-MSH also increased the protein levels of TYR, TYRP1, and DCT in B16-F10 cells (Figure 4d-f). K8 did not significantly attenuate the increase in the protein levels of TYR, TRP1, and DCT caused by α-MSH stimulation.

Figure 4.
Effects of K8 on the mRNA and protein levels of tyrosinase (TYR), tyrosinase-related protein 1 (TYRP1) and dopachrome tautomerase (DCT) in B16-F10 cells stimulated with alpha-melanocytestimulating hormone (α-MSH). Cells were stimulated with 100 nM α-MSH in the presence or absence of K8 at different concentrations for 12 h for mRNA analysis or for 24 h for protein analysis. The mRNA levels of TYR, TYRP1, and DCT were determined by quantitative reverse transcription polymerase chain reaction (qRT-PCR) and normalized to glyceraldehyde 3-phosphate dehydrogenase (GAPDH) (a-c). Western blotting of cell lysates was performed for TYR, TYRP1, and DCT using βactin as a loading control (d-f). Typical blot images are shown. Data are presented as percentages relative to the control value (mean ± SD, n = 3 for a, b, and c; n = 4 for d, e, and f). ** p < 0.01. Open and closed histograms represent the absence and presence of α-MSH addition, respectively.

Effects of K8 on the mRNA and the Protein Expression Levels of MITF
MITF is a key transcription factor that regulates the gene expression of TYR and other melanogenic enzymes [2,8]. As shown in Figure 5a,b, α-MSH induced MITF mRNA and protein expression but these changes were not significantly affected by K8. Figure 4. Effects of K8 on the mRNA and protein levels of tyrosinase (TYR), tyrosinase-related protein 1 (TYRP1) and dopachrome tautomerase (DCT) in B16-F10 cells stimulated with alpha-melanocyte-stimulating hormone (α-MSH). Cells were stimulated with 100 nM α-MSH in the presence or absence of K8 at different concentrations for 12 h for mRNA analysis or for 24 h for protein analysis. The mRNA levels of TYR, TYRP1, and DCT were determined by quantitative reverse transcription polymerase chain reaction (qRT-PCR) and normalized to glyceraldehyde 3-phosphate dehydrogenase (GAPDH) (a-c). Western blotting of cell lysates was performed for TYR, TYRP1, and DCT using β-actin as a loading control (d-f). Typical blot images are shown. Data are presented as percentages relative to the control value (mean ± SD, n = 3 for a, b, and c; n = 4 for d, e, and f). ** p < 0.01. Open and closed histograms represent the absence and presence of α-MSH addition, respectively.

Effects of K8 on the mRNA and the Protein Expression Levels of MITF
MITF is a key transcription factor that regulates the gene expression of TYR and other melanogenic enzymes [2,8]. As shown in Figure 5a,b, α-MSH induced MITF mRNA and protein expression but these changes were not significantly affected by K8. Figure 5. Effects of K8 on the mRNA and protein levels of microphthalmia-associated transcription factor (MITF) in B16-F10 cells stimulated with alpha-melanocyte-stimulating hormone (α-MSH). Cells were stimulated with 100 nM α-MSH in the presence or absence of K8 at various concentrations for 12 h for the mRNA analysis or for 24 h for the protein analysis. The mRNA level of MITF was determined by quantitative reverse transcription polymerase chain reaction (qRT-PCR) and normalized to glyceraldehyde 3-phosphate dehydrogenase (GAPDH) (a). The protein level of MITF was determined by Western blotting using β-actin as a loading control (b). Typical blot images are shown. Data are presented as percentages relative to the control value (mean ± SD, n = 3 for a; n = 4 for b). ** p < 0.01, * p < 0.05. Open and closed histograms represent the absence and presence of α-MSH addition, respectively.

Effects of K8 on the Catalytic Activity of Murine and Human TYRs
Despite the potent inhibitory effects of K8 on cellular melanin synthesis, the mRNA and protein levels of TYR and other related enzymes were not significantly inhibited. This suggests that mechanisms other than gene expression regulation exist. As one possibility, we examined whether the drug directly inhibits the catalysis of TYR. Two different TYR sources were used for this purpose. One is the protein extract of B16F10 cells stimulated by α-MSH and thus expresses a high amount of TYR. The other is the protein extract of HEK293-TYR, a stable cell line overexpressing human TYR. In these experiments, the effect of K8 on the catalytic activity of TYR was compared with that of βarbutin or kojic acid [23]. As shown in Figure 6a, K8 appeared to inhibit murine TYR (IC50, 1.5 µM) much more potently than β-arbutin (IC50, 700 µM). As shown in Figure 6b, K8 (IC50, 1.1 µM) inhibited and TYR much more potently than β-arbutin (IC50, 500 µM) and kojic acid (IC50, 63 µM).

Figure 5.
Effects of K8 on the mRNA and protein levels of microphthalmia-associated transcription factor (MITF) in B16-F10 cells stimulated with alpha-melanocyte-stimulating hormone (α-MSH). Cells were stimulated with 100 nM α-MSH in the presence or absence of K8 at various concentrations for 12 h for the mRNA analysis or for 24 h for the protein analysis. The mRNA level of MITF was determined by quantitative reverse transcription polymerase chain reaction (qRT-PCR) and normalized to glyceraldehyde 3-phosphate dehydrogenase (GAPDH) (a). The protein level of MITF was determined by Western blotting using β-actin as a loading control (b). Typical blot images are shown. Data are presented as percentages relative to the control value (mean ± SD, n = 3 for a; n = 4 for b). ** p < 0.01, * p < 0.05. Open and closed histograms represent the absence and presence of α-MSH addition, respectively.

Effects of K8 on the Catalytic Activity of Murine and Human TYRs
Despite the potent inhibitory effects of K8 on cellular melanin synthesis, the mRNA and protein levels of TYR and other related enzymes were not significantly inhibited. This suggests that mechanisms other than gene expression regulation exist. As one possibility, we examined whether the drug directly inhibits the catalysis of TYR. Two different TYR sources were used for this purpose. One is the protein extract of B16F10 cells stimulated by α-MSH and thus expresses a high amount of TYR. The other is the protein extract of HEK293-TYR, a stable cell line overexpressing human TYR. In these experiments, the effect of K8 on the catalytic activity of TYR was compared with that of β-arbutin or kojic acid [23]. As shown in Figure 6a, K8 appeared to inhibit murine TYR (IC 50 , 1.5 µM) much more potently than β-arbutin (IC 50 , 700 µM). As shown in Figure 6b, K8 (IC 50 , 1.1 µM) inhibited and TYR much more potently than β-arbutin (IC 50 , 500 µM) and kojic acid (IC 50 , 63 µM).

Enzyme Kinetic Study to Determine Inhibition Types of K8 and β-arbutin.
As shown in Figure 7a, the lines of Lineweaver-Burk plots for TYR activity obtained from the uninhibited enzyme and different concentrations of K8 virtually converged to the same intercept on the 1/activity axis, indicating a competitive inhibition. As shown in Figure 7b, the study with varying concentrations of β-arbutin also resulted in a group of lines with the same intercept on the 1/activity axis, indicating a competitive inhibition.

Enzyme Kinetic Study to Determine Inhibition Types of K8 and β-arbutin.
As shown in Figure 7a, the lines of Lineweaver-Burk plots for TYR activity obtained from the uninhibited enzyme and different concentrations of K8 virtually converged to the same intercept on the 1/activity axis, indicating a competitive inhibition. As shown in Figure 7b, the study with varying concentrations of β-arbutin also resulted in a group of lines with the same intercept on the 1/activity axis, indicating a competitive inhibition.

Enzyme Kinetic Study to Determine Inhibition Types of K8 and β-arbutin.
As shown in Figure 7a, the lines of Lineweaver-Burk plots for TYR activity obtained from the uninhibited enzyme and different concentrations of K8 virtually converged to the same intercept on the 1/activity axis, indicating a competitive inhibition. As shown in Figure 7b, the study with varying concentrations of β-arbutin also resulted in a group of lines with the same intercept on the 1/activity axis, indicating a competitive inhibition.

Copper Chelating Activity of K8 and Kojic Acid
Pyrocatechol violet was used to compare the copper chelating activities of K8, kojic acid, and β-arbutin. As shown in Figure 8a, the absorption spectrum of pyrocatechol changed in the presence of copper ions. The increase of absorbance of pyrocatechol violet at 632 nm due to copper chelation was attenuated by K8 and kojic acid, but not by β-arbutin (Figure 8b). Equimolar concentrations (200 µM) of K8 and kojic acid were estimated to chelate 27% and 36% of 200 µM copper ions, respectively, whereas β-arbutin did not show copper chelating activity. Assays were also performed in the presence of K8 (a) or β-arbutin (b) at the indicated concentrations (µM). Lineweaver-Burk plots are drawn using the mean values of three different measurements in order to determine the inhibition types of K8 and β-arbutin. Both K8 and β-arbutin show competitive inhibition modes against TYR activity.

Copper Chelating Activity of K8 and Kojic Acid
Pyrocatechol violet was used to compare the copper chelating activities of K8, kojic acid, and βarbutin. As shown in Figure 8a, the absorption spectrum of pyrocatechol changed in the presence of copper ions. The increase of absorbance of pyrocatechol violet at 632 nm due to copper chelation was attenuated by K8 and kojic acid, but not by β-arbutin (Figure 8b). Equimolar concentrations (200 µM) of K8 and kojic acid were estimated to chelate 27% and 36% of 200 µM copper ions, respectively, whereas β-arbutin did not show copper chelating activity.

Screening of a Drug Library for TYR Inhibitory Activity
The next experiment examined the library for other TYR inhibitors. As a result of evaluating the

Screening of a Drug Library for TYR Inhibitory Activity
The next experiment examined the library for other TYR inhibitors.

Discussion
Melanin synthesis is an important research topic in the context of skin pigmentation disorders [6,26,27]. Skin pigmentation disorders can cause severe mental stress that reduces productivity and quality of life [28] and are thus considered a socially important disease. Various approaches are available for the treatment of hyperpigmentation, such as chemical peeling, laser surgery, pharmacotherapy, and cosmetic camouflage [29][30][31], but these remain unsatisfactory. Hydroquinone is a commonly used treatment for skin pigmentation disorders that can be administered alone or in combination with other drugs, but it is associated with side effects such as irritation, allergy, and cancer [32]. Although various inhibitors of melanin synthesis are used in the cosmetic industry to control skin pigmentation, consumer satisfaction is low, and more effective skin lightening treatments are in high demand [33][34][35].
Melanin is synthesized in cells through a multistep reaction that begins with the oxidation of Ltyrosine and/or L-DOPA to L-dopaquinone catalyzed by the TYR enzyme [36]. Thus, TYR is a useful target for the control of unwanted skin pigmentation, and several strategies have been proposed to target TYR for the control melanin synthesis [21,37]. Gene expression of melanogenic enzymes, such as TYR, TYRP1, and DCT, is induced by MITF in melanocytes [2,8]. Melanocortin 1 receptor (MC1R) is stimulated by α-MSH and other proopiomelanocortin-derived peptide hormones, which activates adenylate cyclase [8,38]. The increase in cyclic AMP subsequently activates protein kinase A and Figure 9. Effects of drugs on human TYR activity in vitro. The activity of human TYR was determined in the absence or presence of each drug (1.0 µM). Data are presented as percentages of uninhibited activity (mean ± SD, n = 3). ** p < 0.01 vs. vehicle control. Drugs that inhibited the catalytic activity of TYR in vitro are highlighted in red.

Discussion
Melanin synthesis is an important research topic in the context of skin pigmentation disorders [6,26,27]. Skin pigmentation disorders can cause severe mental stress that reduces productivity and quality of life [28] and are thus considered a socially important disease. Various approaches are available for the treatment of hyperpigmentation, such as chemical peeling, laser surgery, pharmacotherapy, and cosmetic camouflage [29][30][31], but these remain unsatisfactory. Hydroquinone is a commonly used treatment for skin pigmentation disorders that can be administered alone or in combination with other drugs, but it is associated with side effects such as irritation, allergy, and cancer [32]. Although various inhibitors of melanin synthesis are used in the cosmetic industry to control skin pigmentation, consumer satisfaction is low, and more effective skin lightening treatments are in high demand [33][34][35].
Melanin is synthesized in cells through a multistep reaction that begins with the oxidation of L-tyrosine and/or L-DOPA to L-dopaquinone catalyzed by the TYR enzyme [36]. Thus, TYR is a useful target for the control of unwanted skin pigmentation, and several strategies have been proposed to target TYR for the control melanin synthesis [21,37]. Gene expression of melanogenic enzymes, such as TYR, TYRP1, and DCT, is induced by MITF in melanocytes [2,8]. Melanocortin 1 receptor (MC1R) is stimulated by α-MSH and other proopiomelanocortin-derived peptide hormones, which activates adenylate cyclase [8,38]. The increase in cyclic AMP subsequently activates protein kinase A and induces MITF gene expression [39]. MITF gene expression is also induced by other mechanisms involving the c-kit or WNT pathways [40].
Our team has been exploring natural products and peptides to find more effective melanin inhibitors. As a result, it was shown that the extract of Sasa quelpaertensis effectively inhibited the production of melanin and its main component, p-coumaric acid, inhibited the activity of the TYR enzyme in a manner competing with the substrate [41][42][43]. In addition, in an exploratory study to discover human TYR inhibitors from terrestrial plant extracts, it was observed that the extract of Vitis viniferae and its active ingredient, resveratrol, strongly inhibited human TYR and thereby inhibited cell melanin production [44][45][46]. In a study of marine plants, it was reported that the extract of Phyllospadix iwatensis and its active ingredient, luteolin 7-sulfate, inhibited cell melanin production by inhibition of TYR gene expression as well as inhibition of TYR catalytic activity [47,48]. Recently, by applying the peptide library screening method, we found low molecular peptides that inhibit cell melanin production [49,50]. These peptides share the common sequence with α-MSH and thus can act as an antagonist of MC1R which are involved in the regulation of TYR gene expression [50].
The current study was carried out on an extension of the research to find an effective and safe melanogenesis modulator. This study is different from the previous studies in that an epigenetic drug library was used. The aim at the beginning of this study was to find a substance that modulates TYR expression by epigenetic mechanisms. However, K8, a drug selected to safely and strongly inhibit melanin production through cell-based assay, did not regulate TYR gene expression, but instead directly regulated the catalytic activity of TYR. Our original assumption was that K8 might inhibit melanin synthesis by suppressing TYR expression to the same extent (as monitored by Western blotting and enzyme assay using the cell lysates) but it turned out not to be the case. As a result, although a drug with a mechanism different from the initial hypothesis was found, this study was considered very meaningful and successful because it provided a new drug candidate with very good potency. K8 was originally developed as an HDAC 6-selective inhibitor to enhance the effectiveness of DNA-damaging anticancer drugs [25]. However, its antimelanogenic effect had not been reported prior to this study.
In the first cell-based screening assay using the drug library, the melanin synthesis inhibitory effect of 141 drugs was compared at 1.0 µM. As a result, two drugs, H4 and K8, were selected because of their strong melanin synthesis inhibitory effect without cytotoxicity. Of these, because H4 showed toxicity at 3 µM and higher concentrations and K8 was finally elected as a promising drug candidate.
In the final in vitro screening assay using the drug library, human TYR inhibitory activity of 141 drugs was compared at 1.0 µM. In addition to H4 and K8, three more drugs, A5, C10, and M3, showed excellent inhibitory action. Of these, A5 and M3 showed no melanin inhibitory action and C3 showed strong cytotoxicity in the cell-base experiments. Therefore, among the drugs library, K8 was selected as the best new drug candidate.
The drugs that had shown cytotoxicity at 1 µM ( Figure 1) were eliminated from the active drug candidates. The eliminated drugs might have inhibited melanin synthesis without cytotoxicity if they had been assayed at lower concentrations. Therefore, further studies are needed to examine this possibility. Because only five out of 141 drugs inhibited the catalytic activity of TYR (Figure 1), additional active drugs, if found any, are expected to inhibit melanin synthesis by mechanisms other than inhibition of TYR catalytic activity.
We also examined whether K8 can reduce the gene expression levels of TYR and other related enzymes involved in melanin synthesis via the regulation of transcription factors such as MITF. The results showed that K8 did not significantly reduce the mRNA and protein levels of TYR, TYRP1, DCT, or MITF. Therefore, the inhibitory effect of the drug on cellular melanogenesis could not be explained by the reduced gene expression of TYR or other related enzymes/proteins. We then investigated other potential mechanisms for the antimelanogenic effects of K8 and found that this drug could directly inhibit the catalytic activity of murine TYR (IC 50 , 1.5 µM) much more potently than β-arbutin (IC 50 , 700 µM). Like β-arbutin, K8 showed a competitive type of inhibition against TYR activity, as determined by Lineweaver-Burk plots.
Kojic acid can competitively inhibit TYR activity through a mechanism involving copper cheating activity [24]. In addition, numerous hydroxamic acid derivatives can chelate metal ions and inhibit metalloenzymes such as TYRs and HDACs [53,54]. K8 is a hydroxamic acid derivative and is known to inhibit the activity of HDAC 6 through the chelation of zinc ions [25]. In the present study, K8 and kojic acid were observed to have similar copper chelating activity and β-arbutin has no such activity. Therefore, the mechanism of action of K8 seems to be more similar to kojic acid than β-arbutin.
The skin pigmentation degree will be determined by the relative balance between the pigmentation processes, such as melanin synthesis, and the depigmentation processes, such as the elimination of stratum corneum. We have previously reported that topically applied resveratrol derivatives, which inhibited melanin synthesis through multiple mechanisms including inhibition of TYR catalytic activity [44,46,56], showed depigmentation efficacy in human trials [34,35]. Thus, if the depigmentation process is in the same condition, the inhibition of melanin synthesis may brighten the skin tone, and a potent TYR inhibitor may be a candidate drug. Further research is needed to investigate whether the active drug found in this study can be delivered to melanocytes well and whether it can actually exhibit pigmentation efficacy in vivo.

Reagents and a Drug Library
An epigenetic screening library (Item No. 11076) was purchased from Cayman Chemical (Ann Arbor, MI, USA). It contains 141 cell-permeable small molecules that modulate the activity of epigenetic enzymes, such as DNA methyltransferases, DNA demethylases, histone acetyltransferases, HDACs, or acetylated histone binding proteins. The composition of the library is shown in Table 1

Screening Assay for Overall Melanin Synthesis
For the screening assay, B16-F10 cells were plated in 96-well plates (3 × 10 3 cells per well) and cultured for 24 h. The cells were then treated with a test drug at 1.0 µM for 60 min, and subsequently stimulated with 0.1 µM α-MSH for 72 h. Overall melanin synthesis was estimated by measuring the absorbance of each well (cells plus medium, without cell lysis) at 400 nm using a Spectrostar Nano microplate reader (BMG LABTECH GmbH, Ortenberg, Germany).

Melanin Content Assay
The intracellular melanin retained in cells and extracellular melanin secreted from cells were separately determined [49,57]. B16-F10 cells were plated in 6-well plates (1.2 × 10 4 cells per well), cultured for 24 h, treated with a test drug at various concentrations for 60 min, and stimulated with 100 nM α-MSH for 72 h. The conditioned medium was evaluated to determine the extracellular melanin content. Adherent cells were extracted with 10% dimethyl sulfoxide solution containing 1.0 M sodium hydroxide at 60 • C for 60 min, and melanin content in the extract was estimated by measuring the absorbance at 400 nm and normalized to the total protein content of the cells.

Cellular TYR Activity Assay
After B16-F10 cells were treated with drug and/or α-MSH, the TYR activity of cell extracts was determined by an indirect spectrophotometric method to monitor dopachrome formation with L-tyrosine plus L-DOPA as the substrates [48,58]. Total cellular protein was extracted from the treated cells using a lysis buffer of 10 mM Tris-HCl buffer (pH 7.4) containing 120 mM NaCl, 25 mM KCl, 2.0 mM EDTA, 1.0 mM EGTA, 0.5% Triton X-100, and a protease inhibitor cocktail (Roche, Mannheim, Germany) at 4 • C, followed by centrifugation at 13,000× g for 15 min. The TYR activity assay mixture (200 µL) consisting of 100 mM sodium phosphate buffer (pH 6.8), 1.0 mM L-tyrosine, and 42 µM L-DOPA was incubated with 40 µg cellular protein extract at 37 • C, and changes in absorbance at 475 nm over time were measured using a Spectrostar Nano microplate reader.

qRT-PCR Analysis
After B16-F10 cells were treated with drug and/or α-MSH, cellular RNA was extracted using an RNeasy kit (Qiagen, Valencia, CA, USA) and the complementary DNA (cDNA) was synthesized by reverse transcription using a High-Capacity cDNA Archive Kit (Applied Biosystems, Foster City, CA, USA). qRT-PCR was conducted using a StepOnePlus™ Real-Time PCR System (Applied Biosystems). The reaction mixture (20 µL) comprised SYBR ® Green PCR Master Mix (Applied Biosystems), 60 ng cDNA, and 2 pmol gene-specific primer sets (Macrogen, Seoul, Korea). The sequences of the primers used in this study are listed in Table 2.
The reaction protocol was as follows: initial incubation at 50 • C for 2 min; DNA polymerase activation at 95 • C for 15 min; and annealing and extension at 60 • C for 1 min. In each PCR run, the melting curve was checked to confirm the homogeneity of the PCR product. The mRNA expression levels of the gene of interest were determined using the comparative Ct method, and normalized to the internal reference GAPDH [59]. In this method, Ct is defined as the number of PCR cycles required for the signal to exceed the threshold. Fold changes in the test group compared to the control group were calculated as 2 −∆∆Ct , where ∆∆Ct = ∆Ct (test) − ∆Ct (control) = [Ct (gene, test) − Ct (reference, test) ] − [Ct (gene, control) − Ct (reference, control) ].
After B16-F10 cells were treated with drug and/or α-MSH, whole-cell lysates were prepared using a lysis buffer (pH 7.2) consisting of 10 mM Tris-Cl (pH 7.4), 120 mM NaCl, 25 mM KCl, 2 mM EGTA, 1 mM EDTA, 0.5% Triton X-100 and a protease inhibitor cocktail (Roche, Mannheim, Germany) for foe evaluation of TYR, TYRP1, and DCT. For MITF extraction, cells were extracted with RIPA buffer consisting of 25 mM Tris-HCl (pH 7.6), 150 mM NaCl, 1% NP-40, 1% sodium deoxycholate, 0.1% SDS and protease inhibitor cocktail. Proteins were denatured in the Laemmli sample buffer by heating at 95 • C for 5 min and separated by 7.5% SDS-PAGE gel. Proteins were transferred to a polyvinylidene difluoride membrane (Amersham Pharmacia, Little Chalfont, UK). The membrane was incubated with a primary antibody solution at 4 • C overnight, followed by incubation with a solution of a secondary antibody conjugated to horseradish peroxidase at room temperature for 1 h. The target protein bands were detected by a chemiluminescence method using a picoEPD Western Reagent kit (ELPIS-Biotech, Daejeon, Korea). The blots were analyzed using the NIH Image J program.

In Vitro TYR Catalytic Activity Assay
To examine the direct effect of a drug on the catalytic activity of TYR, in vitro experiments were undertaken. Lysate of B16F10 cells stimulated with α-MSH for 24 h was used as a murine TYR preparation. Human TYR preparation was the lysate of human embryonic kidney 293 cells constitutively expressing human TYR (HEK293-TYR) [51,52]. The cells were lysed in 10 mM Tris-HCl buffer (pH 7.4) containing 120 mM NaCl, 25 mM KCl, 2.0 mM EDTA, 1.0 mM EGTA, 0.5% Triton X-100, and a protease inhibitor cocktail (Roche) at 4 • C and then centrifuged at 13,000× g for 15 min to obtain the supernatants used as TYR preparations. The TYR activity assay mixture (200 µL) consisted of 100 mM sodium phosphate buffer (pH 6.8), TYR preparation (40 µg protein), a drug at various concentrations, 1.0 mM L-tyrosine, and 42 µM L-DOPA. For enzyme kinetic study, the concentrations of L-tyrosine plus L-DOPA were varied while their relative ratio was kept the same as the standard assay condition. The mixture was incubated at 37 • C, and absorbance at 475 nm was monitored using a Spectrostar Nano microplate reader. The reactions were also performed in the absence of a TYR preparation to correct for the effects of non-enzymatic reactions. Enzyme activity was calculated as follows: activity (% control) = (C−D)/(A−B) × 100, where A and B are the absorbance changes of the control group with and without enzyme, respectively, while C and D are the absorbance changes of the test group with and without enzyme, respectively.

Copper Chelating Activity Assay
Copper chelating activity was determined by using pyrocatechol violet (Sigma-Aldrich) [24]. The principle of this method is that, when the pyrocatechol violet is complexed with copper, it shows maximum absorption at 632 nm, and the test material competing with pyrocatechol violet for copper can reduce the absorbance of the copper-pyrocatechol violet complex. The reaction mixture containing a potential chelator at varied concentrations, 200 µM CuSO 4 , and 200 µM pyrocatechol violet was incubated at 25 • C for 20 min, and its absorption spectra were recorded using a Shimadzu UV-1650PC spectrophotometer (Shimadzu Corporation, Kyoto, Japan). Copper chelating activity was evaluated by the change in the absorbance of pyrocatechol violet at 632 nm.

Statistical Analyses
The data are expressed as the mean ± standard deviation (SD) of three independent experiments. The experimental results were statistically analyzed using SigmaStat v. 3.11 software (Systat Software Inc., San Jose, CA, USA), by one-way analysis of variance (ANOVA), followed by Dunnett's test.

Conclusions
In conclusion, the present study identified K8 from an epigenetic drug library as a potent inhibitor of cellular melanin synthesis, suggesting its potential utility in the treatment of hyperpigmentation disorders. The results suggest that K8 inhibits cellular melanin synthesis not through "epigenetic" downregulation of TYR protein levels but by direct inhibition of TYR catalytic activity. Further studies are needed to examine its in vivo and clinical efficacy.