Advertisement

Upregulation of CD86 and IL-12 by rhododendrol in THP-1 cells cocultured with melanocytes through ROS and ATP

Open AccessPublished:December 30, 2022DOI:https://doi.org/10.1016/j.jdermsci.2022.12.012

      Highlights

      • A new h-CLATw/M composed of THP-1 cells and melanoma SK-MEL-37 cells was established.
      • RD upregulates CD86 and IL-12 in THP-1 via ROS and ATP production from SK-MEL-37.
      • h-CLATw/M can distinguish between leukoderma-inducible and non-inducible agents.
      • RD is a pro-hapten dependent on tyrosinase, possibly leading to CTL generation.
      • h-CLATw/M would be useful to predict the sensitizing potential to induce leukoderma.

      Abstract

      Background

      The tyrosinase inhibitor rhododendrol (RD), used as a skin whitening agent, reportedly has the potential to induce leukoderma.

      Objective

      Although an immune response toward melanocytes was demonstrated to be involved in leukoderma, the molecular mechanism is not fully understood.

      Methods

      We hypothesized that if RD is a pro-hapten and tyrosinase-oxidized RD metabolites are melanocyte-specific sensitizers, the sensitizing process could be reproduced by the human cell line activation test (h-CLAT) cocultured with melanocytes (h-CLATw/M) composed of human DC THP-1 cells and melanoma SK-MEL-37 cells. Cell surface expression, ROS generation and ATP release, mRNA expression, and the effects of several inhibitors were examined.

      Results

      When RD was added to the h-CLATw/M, the expression of cell-surface CD86 and IL-12 mRNA was greatly enhanced in THP-1 cells compared with those in the h-CLAT. The rapid death of melanoma cells was induced, with ROS generation and ATP release subsequently being greatly enhanced, resulting in the cooperative upregulation of CD86 and IL-12. Consistent with those observations, an ROS inhibitor, ATP receptor P2X7 antagonist, or PERK inhibitor antagonized the upregulation. CD86 upregulation was similarly observed with another leukoderma-inducible tyrosinase inhibitor, raspberry ketone, but not with the leukoderma noninducible skin-whitening agents ascorbic acid and tranexamic acid.

      Conclusion

      RD is a pro-hapten sensitizer dependent on tyrosinase that induces ROS generation and ATP release from melanocytes for CD86 and IL-12 upregulation in DCs, possibly leading to the generation of tyrosinase-specific cytotoxic T lymphocytes. The coculture system h-CLATw/M may be useful for predicting the sensitizing potential to induce leukoderma.

      Abbreviations:

      AA (ascorbic acid), 7-AAD (7-aminoactinomycin D), BBG (Brilliant Blue G), CTL (cytotoxic CD8+T lymphocyte), DC (dendritic cell), h-CLAT (human cell line activation test), h-CLATw/M (h-CLAT cocultured with melanocytes), h-CLATw/H (h-CLAT cocultured with HaCaT cells), GSK (GSK2606414), HQ (hydroquinone), NAC (N-acetyl-L-cysteine), PERK (protein kinase R-like endoplasmic reticulum kinase), RD (rhododendrol), RFI (relative fluorescence intensity), RK (raspberry ketone), ROS (reactive oxygen species), SDS-PAGE (sodium dodecyl sulfate-polyacrylamide gel electrophoresis), TXA (tranexamic acid)

      Keywords

      1. Introduction

      Leukoderma, which is caused by some competitive inhibitors of tyrosinase such as rhododendrol (RD), raspberry ketone (RK), hydroquinone (HQ), or 4-tertiary butyl phenol (4-TBP), is a pigmentary disorder characterized by the loss of functional melanocytes in human skin [
      • Harris J.E.
      Chemical-induced vitiligo.
      ]. Biopsies showed fewer or no melanocytes in depigmented skin from patients with RD-induced leukoderma than in control skin samples [
      • Tanemura A.
      • Yang L.
      • Yang F.
      • Nagata Y.
      • Wataya-Kaneda M.
      • Fukai K.
      • et al.
      An immune pathological and ultrastructural skin analysis for rhododenol-induced leukoderma patients.
      ]. The frequency of CD8+T cells in both lesional skin and peripheral blood was significantly higher in patients with leukoderma than in control participants [
      • Nishioka M.
      • Tanemura A.
      • Yang L.
      • Tanaka A.
      • Arase N.
      • Katayama I.
      Possible involvement of CCR4+CD8+ T cells and elevated plasma CCL22 and CCL17 in patients with rhododenol-induced leukoderma.
      ], with a high frequency of Melan-A-specific cytotoxic CD8+T lymphocytes (CTLs) being detected in patients positive for HLA-A* 0201 [
      • Fujiyama T.
      • Ikeya S.
      • Ito T.
      • Tatsuno K.
      • Aoshima M.
      • Kasuya A.
      • et al.
      Melanocyte-specific cytotoxic T lymphocytes in patients with rhododendrol-induced leukoderma.
      ]. Thus, RD-induced leukoderma is mediated by the cytolysis of melanocytes, as well as by the subsequent immune response toward melanocytes [
      • Tokura Y.
      • Fujiyama T.
      • Ikeya S.
      • Tatsuno K.
      • Aoshima M.
      • Kasuya A.
      • et al.
      Biochemical, cytological, and immunological mechanisms of rhododendrol-induced leukoderma.
      ,
      • Ito S.
      • Wakamatsu K.
      Biochemical mechanism of rhododendrol-induced leukoderma.
      ,
      • Inoue S.
      • Katayama I.
      • Suzuki T.
      • Tanemura A.
      • Ito S.
      • Abe Y.
      • et al.
      Rhododendrol-induced leukoderma update II: Pathophysiology, mechanisms, risk evaluation, and possible mechanism-based treatments in comparison with vitiligo.
      ].
      The immune mechanism by which RD causes skin depigmentation is presumably that RD would be a pro-hapten sensitizer that is not itself an allergen but that is activated metabolically in the skin, as proposed for other leukoderma-inducible tyrosinase inhibitors such as monobenzone [
      • van den Boorn J.G.
      • Picavet D.I.
      • van Swieten P.F.
      • van Veen H.A.
      • Konijnenberg D.
      • van Veelen P.A.
      • et al.
      Skin-depigmenting agent monobenzone induces potent T-cell autoimmunity toward pigmented cells by tyrosinase haptenation and melanosome autophagy.
      ] and N-propionyl-4-S-cysteaminylphenol [
      • Ito S.
      • Nishigaki A.
      • Ishii-Osai Y.
      • Ojika M.
      • Wakamatsu K.
      • Yamashita T.
      • et al.
      Mechanism of putative neo-antigen formation from N-propionyl-4-S-cysteaminylphenol, a tyrosinase substrate, in melanoma models.
      ]. Oxidation of RD by tyrosinase in melanocytes generates its active metabolites, including RD-quinone and RD-melanins [
      • Ito S.
      • Wakamatsu K.
      Biochemical mechanism of rhododendrol-induced leukoderma.
      ,
      • van den Boorn J.G.
      • Picavet D.I.
      • van Swieten P.F.
      • van Veen H.A.
      • Konijnenberg D.
      • van Veelen P.A.
      • et al.
      Skin-depigmenting agent monobenzone induces potent T-cell autoimmunity toward pigmented cells by tyrosinase haptenation and melanosome autophagy.
      ,
      • Westerhof W.
      • Manini P.
      • Napolitano A.
      • d'Ischia M.
      The haptenation theory of vitiligo and melanoma rejection: a close-up.
      ]. Subsequently, RD-quinone covalently binds to tyrosinase or other melanocyte proteins via SH residues to produce so-called neoantigens [
      • van den Boorn J.G.
      • Picavet D.I.
      • van Swieten P.F.
      • van Veen H.A.
      • Konijnenberg D.
      • van Veelen P.A.
      • et al.
      Skin-depigmenting agent monobenzone induces potent T-cell autoimmunity toward pigmented cells by tyrosinase haptenation and melanosome autophagy.
      ,
      • Ito S.
      • Nishigaki A.
      • Ishii-Osai Y.
      • Ojika M.
      • Wakamatsu K.
      • Yamashita T.
      • et al.
      Mechanism of putative neo-antigen formation from N-propionyl-4-S-cysteaminylphenol, a tyrosinase substrate, in melanoma models.
      ]. These neoantigens then trigger a sensitizing response cascade, inducing generation of melanocyte-specific CTLs and resultant melanocyte killing. The RD-melanins are pro-oxidants that induce generation of reactive oxygen species (ROS), with subsequent depletion of cellular antioxidants, eventually leading to the death of melanocytes [
      • Gu L.
      • Zeng H.
      • Takahashi T.
      • Maeda K.
      In vitro methods for predicting chemical leukoderma caused by quasi-drug cosmetics.
      ,
      • Gabe Y.
      • Miyaji A.
      • Kohno M.
      • Hachiya A.
      • Moriwaki S.
      • Baba T.
      Substantial evidence for the rhododendrol-induced generation of hydroxyl radicals that causes melanocyte cytotoxicity and induces chemical leukoderma.
      ].
      Considering that the skin-whitening competitive tyrosinase inhibitors thus have the high potential to induce leukoderma, an in vitro method for the advance evaluation of whether a skin-whitening agent might cause leukoderma must be developed. Given that RD is seemingly a pro-hapten specifically metabolized by tyrosinase in melanocytes, the development of an in vitro method to evaluate the melanocyte-dependent sensitizing potential of a molecule would be the most suitable approach. The human cell line activation test (h-CLAT), an in vitro evaluation method for predicting the sensitizing potential of chemicals, is widely used as a guideline test, OECD TG 442E [
      • OECD Test No
      442E: In Vitro Skin Sensitisation: In Vitro Skin Sensitisation assays addressing the Key Event on activation of dendritic cells on the Adverse Outcome Pathway for Skin Sensitisation.
      ]. h-CLAT quantifies the increase in cell surface expression of the costimulatory molecule cluster of differentiation 86 (CD86), which is critically important for the activation of naive CD4+T cells [
      • Croft M.
      • Dubey C.
      Accessory molecule and costimulation requirements for CD4 T cell response.
      ] and adhesion molecule CD54 on the human dendritic cell (DC), THP-1 monocytic leukemia cell line [
      • Ashikaga T.
      • Hoya M.
      • Itagaki H.
      • Katsumura Y.
      • Aiba S.
      Evaluation of CD86 expression and MHC class II molecule internalization in THP-1 human monocyte cells as predictive endpoints for contact sensitizers.
      ]. However, h-CLAT cannot detect pro-haptens; therefore, coculture systems of THP-1 cells with other cells such as the HaCaT human keratinocyte cell line or co-incubation with liver microsomes have been reported [
      • Eskes C.
      • Hennen J.
      • Schellenberger M.T.
      • Hoffmann S.
      • Frey S.
      • Goldinger-Oggier D.
      • et al.
      The HaCaT/THP-1 Cocultured Activation Test (COCAT) for skin sensitization: a study of intra-lab reproducibility and predictivity.
      ,
      • Chipinda I.
      • Ruwona T.B.
      • Templeton S.P.
      • Siegel P.D.
      Use of the human monocytic leukemia THP-1 cell line and co-incubation with microsomes to identify and differentiate hapten and prohapten sensitizers.
      ].
      In this study, to reproduce the sensitizing process through which pro-haptens such as RD induce leukoderma via the action of tyrosinase-mediated oxidation on melanocytes in vitro, we developed a new coculture system, h-CLATw/M, consisting of THP-1 cells and melanocytes. As a melanocyte, the SK-MEL-37 human melanoma cell line [
      • Carey T.E.
      • Takahashi T.
      • Resnick L.A.
      • Oettgen H.F.
      • Old L.J.
      Cell surface antigens of human malignant melanoma: mixed hemadsorption assays for humoral immunity to cultured autologous melanoma cells.
      ] was used. In that coculture system, RD greatly upregulated CD86 and IL-12 expression in THP-1 cells, indicating that RD may be a melanocyte-specific pro-hapten sensitizer with the potential to promote CTL generation, presumably leading to leukoderma. Analyses using the coculture system clearly revealed that ROS generation and ATP release play critically important roles in the upregulation of CD86 and IL-12 in THP-1 cells. Furthermore, similar upregulation of CD86 was observed with another leukoderma-inducible agent, RK, but not with the leukoderma-noninducible skin-whitening agents ascorbic acid (AA) and tranexamic acid (TXA). Thus, the coculture system h-CLATw/M, consisting of THP-1 and a human melanoma cell line, would be a useful tool for evaluating the sensitizing potential of chemicals that are competitive tyrosinase inhibitors and also pro-haptens to develop melanocyte-specific immune responses resulting in leukoderma.

      2. Results

      2.1 Upregulation of CD86 on THP-1 cells by RD in h-CLATw/M

      Although h-CLAT is useful for predicting the sensitization potential of chemicals, it cannot detect the sensitization potential of pro-haptens that rely on metabolism to activate them [
      • Eskes C.
      • Hennen J.
      • Schellenberger M.T.
      • Hoffmann S.
      • Frey S.
      • Goldinger-Oggier D.
      • et al.
      The HaCaT/THP-1 Cocultured Activation Test (COCAT) for skin sensitization: a study of intra-lab reproducibility and predictivity.
      ,
      • Chipinda I.
      • Ruwona T.B.
      • Templeton S.P.
      • Siegel P.D.
      Use of the human monocytic leukemia THP-1 cell line and co-incubation with microsomes to identify and differentiate hapten and prohapten sensitizers.
      ]. Therefore, to reproduce the sensitizing process by which pro-haptens such as RD induce leukoderma via the action of tyrosinase-mediated oxidation on melanocytes, we established a novel coculture system, h-CLATw/M, consisting of THP-1 cells and SK-MEL-37 cells (Fig. S1). First, we compared the protein expression level of tyrosinase detected by western blotting and the ability to upregulate CD86 in response to RD using different melanoma and melanocyte cell lines (Fig. S2). Human telomerase (hTERT)-immortalized melanocytes showed the highest responsiveness to stimulation with RD, and SK-MEL-37 cells showed the second highest one. However, hTERT-immortalized melanocytes grow very slowly and require expensive special medium and a very high percentage (40%) of fetal bovine serum (FBS). Therefore, also taking into account versatility, we decided to use SK-MEL-37 cells in this study. Instead of adding sample directly to the medium outside the inset in the well, a small volume (100 µl) of sample dissolved in dimethyl sulfoxide (DMSO) diluted in medium or medium alone was added to the center of the scaffold membrane and left for 30–60 min, after which the well was filled with medium (2 ml) (Fig. S1). In this protocol, the sample, at approximately 20 times higher concentration, initially damages melanoma cells, releasing inflammatory cytokines and damage-associated molecular patterns that lead to the upregulation of CD86 and IL-12 in THP-1 cells, which is highly analogous to the in vivo situation [
      • Williams M.A.
      • Bevan M.J.
      Effector and memory CTL differentiation.
      ].
      As expected, RD at a final concentration of 5 mM greatly upregulated CD86 and slightly increased CD54 expression, keeping the viability of THP-1 cells high (>75% [
      • OECD Test No
      442E: In Vitro Skin Sensitisation: In Vitro Skin Sensitisation assays addressing the Key Event on activation of dendritic cells on the Adverse Outcome Pathway for Skin Sensitisation.
      ]) in h-CLATw/M but not in h-CLAT (Fig. 1A–D). Under those conditions, 100 µl of 100 mM RD was initially added to the scaffold membrane containing melanoma cells. In h-CLATw/M, RD dose dependently increased CD86 expression on THP-1 cells, peaking at about 7 mM, and decreasing thereafter (Fig. 1E). By contrast, in h-CLAT, similar time kinetics were observed, but the highest expression was much lower than the expression observed in h-CLATw/M. However, the slightly increased CD54 expression in h-CLATw/M was not significantly different from the expression observed in h-CLAT (Fig. 1F). The viability of THP-1 cells gradually decreased with increasing concentrations of RD (Fig. 1G), but decreased more rapidly at lower concentrations of RD in h-CLAT than in h-CLATw/M. That observation is probably explained by the presence of melanoma cells mitigating the cytotoxic effects of RD on THP-1 cells. According to the criteria of h-CLAT [
      • OECD Test No
      442E: In Vitro Skin Sensitisation: In Vitro Skin Sensitisation assays addressing the Key Event on activation of dendritic cells on the Adverse Outcome Pathway for Skin Sensitisation.
      ], RD at final concentrations of 2–5 mM in h-CLATw/M appeared to be positive for the upregulation of CD86 (relative fluorescence intensity [RFI]>150%) and CD54 (RFI>200%) keeping the viability of THP-1 cells high (>75%). By contrast, RD at final concentrations up to 2 mM, where the viability remains higher than 75%, in h-CLAT appears to be negative for both. These results suggest that h-CLATw/M may predict the sensitizing process by which pro-hapten RD leads to the induction of leukoderma.
      Fig. 1
      Fig. 1Upregulation of CD86 on THP-1 cells by RD in h-CLATw/M. RD, final concentration 5 mM, and control medium, final concentration 0.05% DMSO, were applied in the h-CLATw/M and h-CLAT systems and incubated for 24 h. Upregulation of CD86 and CD54 on THP-1 cells, and cell viability, were analyzed by flow cytometry using anti-CD86 or anti-CD54 (shaded area) or the respective control antibodies (plain line). Representative histograms for cell surface expression of (A) CD86 and (C) CD54, plus (B, D) dot plots for viable cells with RFI of CD86 or CD54 and cell viability are shown. DNCB (2 µg/ml) was used as a positive control. Dose-dependent effects of RD on the RFI of (E) CD86 and (F) CD54, together with (G) their cell viabilities are shown. Data are shown as the mean± standard error of the mean (n = 3–8). P values were determined by one-way analysis of variance with the Dunnett’s test to compare differences relative to 0 mM RD and by the unpaired two-tailed Student t-test to compare differences between h-CLAT and h-CLATw/M. *P < 0.05. **P < 0.01. ***P < 0.001.

      2.2 ROS generation-mediated CD86 upregulation on THP-1 cells by RD in h-CLATw/M

      We next investigated the molecular mechanism by which RD induces the upregulation of CD86 on THP-1 cells in h-CLATw/M. Adding a high concentration of RD (100 mM) to SK-MEL-37 cells rapidly and severely damaged the melanoma cells and induced cell death, which could be clearly observed under a microscope. The cells started to lose their shape, becoming round, detached, and afloat within 15 min (Fig. 2A). Similar severe damage was not observed in HaCaT cells. Supernatants from those experiments, together with RD and control medium, were added to THP-1 cells, and the upregulation of CD86 on THP-1 cells was analyzed by flow cytometry. RD and the supernatant of HaCaT cells treated with RD slightly and similarly increased CD86 expression, but the supernatant of SK-MEL-37 cells treated with RD greatly increased CD86 expression (Fig. 2B, C). However, this different CD86 upregulation has nothing to do with whether the cells were dead or alive, because neither dead cell debris obtained after just freezing/thawing SK-MEL-37 cells three times nor the associated centrifuged supernatant upregulated CD86 (Fig. 2D).
      Fig. 2
      Fig. 2ROS generation-mediated CD86 upregulation on THP-1 cell by RD in h-CLATw/M. RD, final concentration 100 mM, and control medium, final concentration 1% DMSO, were added to HaCaT human keratinocyte cells or SK-MEL-37 melanoma cells in a 12-well plate and incubated for 0 and 15 min (A) Representative images of the resulting cell morphologies under microscopy are shown. (B) After supernatants from those experiments together with RD and control medium were added to THP-1 cells, upregulation of CD86 on THP-1 cells was analyzed by flow cytometry using anti-CD86 (shaded area) or a control antibody (plain line). Representative histograms with RFI of CD86 are shown. (C) The associated RFIs of CD86 were calculated and compared. (D) Cell debris from SK-MEL-37 cells treated with three freeze/thaw cycles and their supernatant (Sup) after centrifugation were added to h-CLAT. RD (final concentration: 4 mM) and control medium (final concentration: 0.04% DMSO) were similarly added to h-CLAT and h-CLATw/M. Their CD86 RFIs of were calculated and compared. (E) RD (final concentration: 4 mM) and control medium (final concentration: 0.04% DMSO) were added to h-CLAT and h-CLATw/M and incubated for 1 and 3 h. The resultant cell supernatants were collected and analyzed for ROS generation. After SK-MEL-37 cells were pretreated with a ROS inhibitor, NAC (0, 5, 10, 20, and 30 mM), for 1 h and then stimulated with RD (final concentration: 4 mM) in the presence of the same concentrations of NAC for a total of 24 h, (F) cell surface expression of CD86 and (G) cell viability were determined by flow cytometry. Data are shown as the mean± standard error of the mean (n = 3–4). P values were determined by one-way analysis of variance with Tukey’s multiple comparison test (C–E), or the Dunnett’s test to compare differences relative to 0 mM NAC (F, G). *P < 0.05. ***P < 0.001. NS = nonsignificant.
      RD induces ROS generation via tyrosinase activity in melanocytes [
      • Kondo M.
      • Kawabata K.
      • Sato K.
      • Yamaguchi S.
      • Hachiya A.
      • Takahashi Y.
      • et al.
      Glutathione maintenance is crucial for survival of melanocytes after exposure to rhododendrol.
      ], and ROS is an inducer of DC maturation [
      • Rutault K.
      • Alderman C.
      • Chain B.M.
      • Katz D.R.
      Reactive oxygen species activate human peripheral blood dendritic cells.
      ,
      • Kantengwa S.
      • Jornot L.
      • Devenoges C.
      • Nicod L.P.
      Superoxide anions induce the maturation of human dendritic cells.
      ,
      • Byamba D.
      • Kim T.G.
      • Kim D.H.
      • Je J.H.
      • Lee M.G.
      The Roles of Reactive Oxygen Species Produced by Contact Allergens and Irritants in Monocyte-derived Dendritic Cells.
      ]. Concomitant with the damage to SK-MEL-37 cells caused by RD, high ROS generation was rapidly detected (as early as 1 h after stimulation) in the culture supernatants of h-CLATw/M; ROS levels tended to decrease thereafter (Fig. 2E). Notably, similar generation of ROS was not observed when RD alone was applied to h-CLAT under the same conditions, consistent with the presence of tyrosinase in melanoma cells being necessary for ROS generation induced by RD [
      • Gu L.
      • Zeng H.
      • Takahashi T.
      • Maeda K.
      In vitro methods for predicting chemical leukoderma caused by quasi-drug cosmetics.
      ,
      • Gabe Y.
      • Miyaji A.
      • Kohno M.
      • Hachiya A.
      • Moriwaki S.
      • Baba T.
      Substantial evidence for the rhododendrol-induced generation of hydroxyl radicals that causes melanocyte cytotoxicity and induces chemical leukoderma.
      ]. Next, to confirm that ROS generation is involved in CD86 upregulation, we examined the effects of N-acetyl-L-cysteine (NAC), a ROS inhibitor [
      • Aldini G.
      • Altomare A.
      • Baron G.
      • Vistoli G.
      • Carini M.
      • Borsani L.
      • et al.
      N-Acetylcysteine as an antioxidant and disulphide breaking agent: the reasons why.
      ], on RD-induced CD86 upregulation. NAC greatly antagonized CD86 upregulation induced by RD in h-CLATw/M in a dose-dependent manner (Fig. 2F, G). These results suggest that ROS generation in SK-MEL-37 cells treated with RD is important for CD86 upregulation on THP-1 cells.

      2.3 ATP release-mediated CD86 upregulation on THP-1 cells by RD in h-CLATw/M

      Because damage-associated molecular patterns and inflammatory cytokines such as high mobility group box 1 (HMGB1), S100A/B, interleukin-1 (IL-1), and ATP are induce DC maturation [
      • Sebastiao A.I.
      • Ferreira I.
      • Brites G.
      • Silva A.
      • Neves B.M.
      • Teresa Cruz M.
      NLRP3 Inflammasome and Allergic Contact Dermatitis: A Connection to Demystify.
      ], we investigated whether factors other than ROS might contribute to RD-induced CD86 upregulation in h-CLATw/M. The supernatant of SK-MED-37 cells treated with RD in h-CLATw/M for 1 h was collected and further separated by ultrafiltration using a centrifugal filter with a 10 kDa cutoff into two fractions containing proteins with molecular weights higher 10 kDa and peptides and others with molecular weights lower than 10 kDa. The supernatant of SK-MED-37 cells treated with RD was also heat-inactivated at 95 °C for 10 min. Untreated SK-MEL-37 cells were frozen and thawed three times, and the resultant supernatant was centrifuged. All of these supernatants were added to THP-1 cells and incubated for 24 h for subsequent flow cytometry analysis (Fig. 3A, B). Notably, the supernatant of SK-MED-37 cells treated with RD was observed to upregulate CD86, while the fraction containing proteins of more than 10 kDa molecular weight failed to increase CD86 expression but the fraction containing peptides or proteins of less than 10 kDa molecular weight greatly increased that expression. Heat inactivation significantly reduced the induction of CD86 upregulation, and the supernatant of SK-MEL-37 cells treated only with freezing/thawing did not increase CD86 expression. These results suggest that proteins with a molecular weight higher than 10 kDa or the supernatant of SK-MEL-37 cells treated with freezing/thawing alone are not involved in the induction of CD86 upregulation, and that peptides or proteins with a molecular weight lower than 10 kDa or small bioactive substances sensitive to heat inactivation contribute to such upregulation.
      Fig. 3
      Fig. 3ATP release-mediated CD86 upregulation on THP-1 cells by RD in h-CLATw/M. (A) SK-MEL-37 cells were treated with RD (100 mM) or control medium (1% DMSO) in h-CLATw/M for 1 h. The culture supernatant of SK-MEL-37 cells treated with RD were further separated into two fractions containing proteins with molecular weights either higher or lower than 10 kDa. In addition, the supernatant was heat-inactivated at 95 °C for 10 min, and a supernatant of SK-MEL-37 cells treated with three freeze/thaw cycles was also prepared. These supernatants were then added to h-CLAT and incubated for 24 h. Flow cytometry analysis using anti-CD86 (shaded area) or control antibody (plain line) was then conducted. Representative histograms with RFI of CD86 are shown. (B) The associated RFIs of CD86 were calculated and compared. The fraction containing proteins with a molecular weight higher than 10 kDa was concentrated approximately 10-foled (10 ×), and the fraction containing proteins with a molecular weight lower than 10 kDa was concentrated approximately 5-fold (5 ×). Those supernatants then underwent sodium dodecyl sulfate-polyacrylamide gel electrophoresis using (C) 15% and (D) 20% polyacrylamide gel, followed by silver staining. (E) ATP release in the supernatant was also determined. (F) THP-1 cells were pretreated overnight with a noncompetitive antagonist of ATP receptor P2Y7, BBG (0.2 µM), and then stimulated with RD (final concentration: 5 mM) in h-CLAT or h-CLATw/M for 24 h followed by flow cytometry analysis for CD86. The RFIs of CD86 were calculated and compared. Data are shown as the mean± standard error of the mean (n = 3). P values were determined by one-way analysis of variance with a Tukey multiple comparisons test. *P < 0.05. ***P < 0.001. NS = nonsignificant.
      To further examine the involvement of the lower-molecular-weight proteins, all of the supernatants underwent silver staining after sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) (Fig. 3C, D). No significant positive band was observed in the fraction containing proteins with a molecular weight lower than 10 kDa, but many strong bands were observed in the fraction containing proteins with a molecular weight higher than 10 kDa and in the supernatant of SK-MEL-37 cells treated with freezing/thawing alone. Given that one of the candidates was small bioactive substances sensitive to heat inactivation, we next examined the contribution of ATP to CD86 upregulation, because ATP induces CD86 upregulation on THP-1 cells and DCs [
      • Miyazawa M.
      • Ito Y.
      • Kosaka N.
      • Nukada Y.
      • Sakaguchi H.
      • Suzuki H.
      • et al.
      Role of TNF-alpha and extracellular ATP in THP-1 cell activation following allergen exposure.
      ,
      • Martins J.D.
      • Silva A.
      • Ferreira I.
      • Goncalo M.
      • Custodio J.B.A.
      • Lopes M.C.
      • et al.
      Adenosine diphosphate involvement in THP-1 maturation triggered by the contact allergen 1-fluoro-2,4-dinitrobenzene.
      ,
      • Schnurr M.
      • Then F.
      • Galambos P.
      • Scholz C.
      • Siegmund B.
      • Endres S.
      • et al.
      Extracellular ATP and TNF-alpha synergize in the activation and maturation of human dendritic cells.
      ]. As expected, RD greatly augmented ATP release into the supernatant of SK-MEL-37 cells, which was only detected in the fraction containing proteins with a molecular weight lower than 10 kDa and which was likely heat-sensitive (Fig. 3E). Importantly, the ATP receptor P2X7 antagonist Brilliant Blue G (BBG) [
      • Jiang L.H.
      • Mackenzie A.B.
      • North R.A.
      • Surprenant A.
      Brilliant blue G selectively blocks ATP-gated rat P2X(7) receptors.
      ] significantly inhibited the RD-induced upregulation of CD86 in h-CLATw/M (Fig. 3F). These results suggest that the release of ATP in SK-MEL-37 cells treated with RD is important for CD86 upregulation on THP-1 cells in h-CLATw/M.

      2.4 ROS- and ATP-mediated cooperative upregulation of CD86 on THP-1 cells by RD in h-CLATw/M

      We further explored the relationship between ROS generation and ATP release from SK-MEL-37 cells in response to RD. SK-MEL-37 cells were pre-treated with ROS inhibitor NAC and/or protein kinase R-like endoplasmic reticulum kinase (PERK) inhibitor GSK2606414 (GSK) [
      • Axten J.M.
      • Medina J.R.
      • Feng Y.
      • Shu A.
      • Romeril S.P.
      • Grant S.W.
      • et al.
      Discovery of 7-methyl-5-(1-{[3-(trifluoromethyl)phenyl]acetyl}-2,3-dihydro-1H-indol-5-yl)-7H-p yrrolo[2,3-d]pyrimidin-4-amine (GSK2606414), a potent and selective first-in-class inhibitor of protein kinase R (PKR)-like endoplasmic reticulum kinase (PERK).
      ] overnight and then stimulated with RD in h-CLATw/M. ATP is released from living cells via PERK in response to endoplasmic reticulum (ER) stress, and GSK is an inhibitor of PERK signaling to suppress ATP release [
      • Garg A.D.
      • Krysko D.V.
      • Verfaillie T.
      • Kaczmarek A.
      • Ferreira G.B.
      • Marysael T.
      • et al.
      A novel pathway combining calreticulin exposure and ATP secretion in immunogenic cancer cell death.
      ]. The RD-induced CD86 upregulation on THP-1 cells was greatly inhibited in h-CLATw/M by either NAC or GSK pre-treatment, and the presence of both inhibitors further, but just slightly inhibited it (Fig. 4A). Next, we assessed ROS generation and ATP release in the supernatants of RD-stimulated SK-MEL-37 cells pretreated with NAC and/or GSK. As expected, NAC suppressed ROS generation and GSK had no effect (Fig. 4B). By contrast, GSK slightly but significantly reduced ATP release, and NAC enhanced it (Fig. 4C). These complicated results might be attributable to mutual regulation between ROS generation and ATP release [
      • Ahmad S.
      • Ahmad A.
      • White C.W.
      Purinergic signaling and kinase activation for survival in pulmonary oxidative stress and disease.
      ,
      • Cruz C.M.
      • Rinna A.
      • Forman H.J.
      • Ventura A.L.
      • Persechini P.M.
      • Ojcius D.M.
      ATP activates a reactive oxygen species-dependent oxidative stress response and secretion of proinflammatory cytokines in macrophages.
      ,
      • Onami K.
      • Kimura Y.
      • Ito Y.
      • Yamauchi T.
      • Yamasaki K.
      • Aiba S.
      Nonmetal haptens induce ATP release from keratinocytes through opening of pannexin hemichannels by reactive oxygen species.
      ]. Then the effects of hydrogen peroxide (H2O2) and ATP on the upregulation of CD86 on THP-1 cells in h-CLAT was examined. Either H2O2 or ATP alone greatly upregulated CD86, and together, they cooperatively upregulated CD86 in h-CLAT (Fig. 4D, E). These results suggest that ROS and ATP cooperatively upregulate CD86 expression on THP-1 cells in response to RD in h-CLATw/M.
      Fig. 4
      Fig. 4ROS and ATP cooperatively upregulate CD86 on THP-1 cells in response to RD in h-CLATw/M. SK-MEL-37 cells were pretreated overnight with a ROS inhibitor, NAC (20 mM), and/or an inhibitor of PERK signaling, GSK (1 µM), which suppresses ATP release. The medium was then exchanged for fresh medium similarly containing the respective inhibitors, and RD (final concentration: 5 mM) or control medium (final concentration: 0.05% DMSO) was added to h-CLATw/M and incubated for 24 h. (A) Then the cells were analyzed for CD86 expression, and the associated RFIs were calculated and compared. (B) ROS generation and (C) ATP release were determined using the culture supernatants. (D) H2O2 (10 mM) and/or ATP (100 µM) were added to h-CLAT, RD (final concentration: 4 mM) was added to h-CLATw/M, and incubated for 24 h. Representative histograms of the flow cytometry analysis of CD86 using anti-CD86 (shaded area) or a control antibody (plain line) with RFI of CD86 are shown. (E) The RFIs of CD86 were also calculated and compared. Data are shown as the mean± standard error of the mean (n = 3). P values were determined by one-way analysis of variance with Tukey’s multiple comparison test (A–C), or the Dunnett’s test to compare differences relative to the absence of both H2O2 and ATP (E). *P < 0.05. ***P < 0.001. NS = nonsignificant.

      2.5 ROS- and ATP-mediated cooperative upregulation of IL-12 mRNA in THP-1 cells by RD in h-CLATw/M

      RD-induced leukoderma is mediated by the direct cytolysis of melanocytes via RD and RD-generated ROS, and by a subsequent immune response to melanocytes through melanocyte-specific CTLs [
      • Tokura Y.
      • Fujiyama T.
      • Ikeya S.
      • Tatsuno K.
      • Aoshima M.
      • Kasuya A.
      • et al.
      Biochemical, cytological, and immunological mechanisms of rhododendrol-induced leukoderma.
      ,
      • Ito S.
      • Wakamatsu K.
      Biochemical mechanism of rhododendrol-induced leukoderma.
      ,
      • Inoue S.
      • Katayama I.
      • Suzuki T.
      • Tanemura A.
      • Ito S.
      • Abe Y.
      • et al.
      Rhododendrol-induced leukoderma update II: Pathophysiology, mechanisms, risk evaluation, and possible mechanism-based treatments in comparison with vitiligo.
      ]. IL-12 is a potent differentiating factor for type 1 helper (Th1) cells and CTLs [
      • Trinchieri G.
      Interleukin-12 and the regulation of innate resistance and adaptive immunity.
      ], and both ROS and ATP induce IL-12 production from macrophages and DCs [
      • Schnurr M.
      • Then F.
      • Galambos P.
      • Scholz C.
      • Siegmund B.
      • Endres S.
      • et al.
      Extracellular ATP and TNF-alpha synergize in the activation and maturation of human dendritic cells.
      ,
      • Aramaki Y.
      • Yotsumoto S.
      • Watanabe H.
      • Tsuchiya S.
      NADPH-oxidase may contribute to IL-12 production in macrophages stimulated with CpG phosphorothioate oligodeoxynucleotides.
      ,
      • Ichikawa S.
      • Miyake M.
      • Fujii R.
      • Konishi Y.
      MyD88 associated ROS generation is crucial for Lactobacillus induced IL-12 production in macrophage.
      ]. Therefore, next we determined whether RD induces IL-12 expression, a heterodimeric cytokine consisting of p35 and p40 subunits, in THP-1 cells. RD (final: 5 mM) and control medium (final: 0.05% DMSO) were applied to h-CLAT and h-CLATw/M and incubated for 24 h. Total RNA was extracted from the THP-1 cells and underwent quantitative PCR (qPCR). As with CD86 upregulation, mRNA expression of both IL-12 p35 and p40 slightly increased after RD stimulation in h-CLAT but was more greatly enhanced in h-CLATw/M (Fig. 5A). As reported [
      • Schnurr M.
      • Then F.
      • Galambos P.
      • Scholz C.
      • Siegmund B.
      • Endres S.
      • et al.
      Extracellular ATP and TNF-alpha synergize in the activation and maturation of human dendritic cells.
      ,
      • Aramaki Y.
      • Yotsumoto S.
      • Watanabe H.
      • Tsuchiya S.
      NADPH-oxidase may contribute to IL-12 production in macrophages stimulated with CpG phosphorothioate oligodeoxynucleotides.
      ,
      • Ichikawa S.
      • Miyake M.
      • Fujii R.
      • Konishi Y.
      MyD88 associated ROS generation is crucial for Lactobacillus induced IL-12 production in macrophage.
      ], H2O2 and ATP increased the mRNA expression of both IL-12 p35 and p40 in h-CLAT, but their cooperative effect was not prominent (Fig. 5B). Pretreatment of SK-MEL-37 cells with the ROS inhibitor NAC (20 mM) or the ATP receptor P2X7 antagonist BBG (1 µM) greatly inhibited upregulation of IL-12 p35 and p40 mRNA expression but combined pretreatment seemed not to further inhibit it (Fig. 5C). These results suggest that RD upregulates IL-12 mRNA expression in a ROS- and ATP-dependent manner in THP-1 cells in h-CLATw/M, an action important for generating melanocyte-specific CTLs.
      Fig. 5
      Fig. 5ROS- and ATP-mediated cooperative upregulation of IL-12 mRNA in THP-1 cells by RD in h-CLATw/M. (A) RD, final concentration 5 mM, and control medium, final concentration 0.05% DMSO, were added to h-CLAT and h-CLATw/M and incubated for 24 h. qPCR was then used to analyze mRNA expression of IL-12 p35 and p40 in the THP-1 cells and of HPRT. (B) After H2O2 (10 mM) and/or ATP (100 µM) were added to h-CLAT and incubated for 24 h, mRNA expression of IL-12 p35 and p40 in the THP-1 cells and of HPRT was analyzed by qPCR. (C) After SK-MEL-37 cells were pretreated overnight with RD (final concentration: 5 mM) or control medium (final concentration: 0.05% DMSO) in the presence or absence of NAC (final concentration: 20 mM), the resulting supernatants were added to THP-1 cells pretreated with BBG (final concentration: 1 µM) for 30 min or untreated in h-CLAT and incubated for 24 h, and mRNA expression of IL-12 p35 and p40 in the THP-1 cells and of HPRT was analyzed by qPCR. Data are shown as the mean± standard error of the mean (n = 3–7). P values were determined by one-way analysis of variance with the Tukey’s multiple comparisons test. *P < 0.05. **P < 0.01. ***P < 0.001.

      2.6 Upregulation of CD86 on THP-1 cells by RK in h-CLATw/M

      Next, we examined the effect of another leukoderma-inducible skin-whitening tyrosinase inhibitor, RK [
      • Fukuda Y.
      • Nagano M.
      • Futatsuka M.
      Occupational leukoderma in workers engaged in 4-(p-hydroxyphenyl)-2-butanone manufacturing.
      ,
      • Fukuda Y.
      • Nagano M.
      • Arimatsu Y.
      • Futatsuka M.
      An experimental study on depigmenting activity of 4-(p-hydroxyphenyl)-2-butanone in C57 black mice.
      ], on CD86 expression in h-CLATw/M. RK at final concentration of 5 mM greatly augmented CD86 expression on THP-1 cells in h-CLATw/M, but much less so in h-CLAT (Fig. 6A, B). In addition, RK dose dependently increased CD86 expression in both h-CLATw/M and h-CLAT, but the upregulation of CD86 was significantly greater in h-CLATw/M than in h-CLAT, whereas cell viability was high in both cases (Fig. 6C, D). According to the criteria of h-CLAT [
      • OECD Test No
      442E: In Vitro Skin Sensitisation: In Vitro Skin Sensitisation assays addressing the Key Event on activation of dendritic cells on the Adverse Outcome Pathway for Skin Sensitisation.
      ], RK at final concentrations of 1.5–3 mM in both h-CLATw/M and h-CLAT appears to be positive for the upregulation of CD86 (RFI>150%) and CD54 (RFI>200%) keeping the viability of THP-1 cells high (>75%). In addition, preliminary data showed that other leukoderma-inducible skin-whitening tyrosinase inhibitors, HQ and 4-TBP, also appear to be positive for CD86 upregulation, although the upregulation efficiencies are likely less than RD (Fig. S3). Thus, h-CLATw/M may be useful to evaluate pro-haptens that are oxidized through tyrosinase to increase CD86 expression.
      Fig. 6
      Fig. 6Upregulation of CD86 on THP-1 cells by raspberry ketone (RK) in h-CLATw/M. (A) CD86 expression was augmented in h-CLATw/M compared with h-CLAT after RK (final concentration: 5 mM) and control medium (final concentration: 0.05% lDMSO) were added. Representative histograms show upregulation levels of CD86 on THP-1 cells analyzed by flow cytometry using anti-CD86 (shaded area) or a control antibody (plain line) with RFI of CD86. (B) Dot plots show viable cells with cell viability. DNCB (2 µg/ml) was used as a positive control. (C) The dose-dependent effects of RK on the RFI of CD86 and (D) cell viability are shown. Data are shown as the mean± standard error of the mean (n = 3). P values were determined by one-way analysis of variance with a Dunnett test to compare differences relative to 0 mM RK, and by the unpaired two-tailed Student t-test to compare differences between h-CLAT and h-CLATw/M. *P < 0.05. ***P < 0.001.

      2.7 No upregulation of CD86 on THP-1 cells by AA and TXA in h-CLATw/M

      Because tyrosinase is a crucial enzyme expressed only by melanocyte cells in synthesizing melanin via melanogenesis, it is any skin-whitening agent’s most prominent and successful target for inhibition of melanogenesis [
      • Pillaiyar T.
      • Manickam M.
      • Namasivayam V.
      Skin whitening agents: medicinal chemistry perspective of tyrosinase inhibitors.
      ]. By contrast, leukoderma-noninducible whitening agents such as AA and TXA are also available. AA, also called vitamin C, is a water-soluble antioxidant that protects skin from ROS and reduces the production of RD-quinones from dihydroxyphenylalanine oxidation by tyrosinase [
      • Pullar J.M.
      • Carr A.C.
      • Vissers M.C.M.
      The Roles of Vitamin C in Skin Health.
      ,
      • Carita A.C.
      • Fonseca-Santos B.
      • Shultz J.D.
      • Michniak-Kohn B.
      • Chorilli M.
      • Leonardi G.R.
      Vitamin C: One compound, several uses. Advances for delivery, efficiency and stability.
      ,
      • Zerbinati N.
      • Sommatis S.
      • Maccario C.
      • Di Francesco S.
      • Capillo M.C.
      • Rauso R.
      • et al.
      The Anti-Ageing and Whitening Potential of a Cosmetic Serum Containing 3-O-ethyl-l-ascorbic Acid.
      ]. TXA inhibits melanin synthesis by suppressing the plasminogen/plasmin pathway, thereby blocking the interaction between melanocytes and keratinocytes [
      • Tse T.W.
      • Hui E.
      Tranexamic acid: an important adjuvant in the treatment of melasma.
      ].
      Our final investigation evaluated the effect of these substances on CD86 upregulation in h-CLATw/M. As expected, both AA and TXA failed to increase CD86 expression on THP-1 cells in h-CLATw/M and in h-CLAT (Fig. S4A, C, E). Cell viability remained high, except during AA treatment in h-CLAT, where cell viability gradually decreased with higher concentrations of AA (Fig. S4B, D, F). Thus, h-CLATw/M can properly distinguish CD86 expression from leukoderma-inducible agents versus leukoderma-noninducible agents.

      3. Discussion

      h-CLAT is widely used to evaluate the sensitizing potential of chemicals in vitro under OECD test guideline 442E [
      • OECD Test No
      442E: In Vitro Skin Sensitisation: In Vitro Skin Sensitisation assays addressing the Key Event on activation of dendritic cells on the Adverse Outcome Pathway for Skin Sensitisation.
      ] as part of a Defined Approach [
      • OECD
      OECD Guideline No. 497: Defined Approaches on Skin Sensitization.
      ]. The molecular mechanisms by which chemical sensitizers induce upregulation of CD86 and CD54, and maturation of DCs are mediated by proinflammatory cytokines such as IL-1, IL-8, and IL-18, and damage-associated molecular patterns such as ROS, uric acid, hyaluronic acid fragments, heat shock proteins, HMGB1, S100A/B, and extracellular ATP, among others [
      • Sebastiao A.I.
      • Ferreira I.
      • Brites G.
      • Silva A.
      • Neves B.M.
      • Teresa Cruz M.
      NLRP3 Inflammasome and Allergic Contact Dermatitis: A Connection to Demystify.
      ]. From an immunologic viewpoint, CD86 upregulation in immature DCs is critically important to properly stimulate antigen-specific naive CD4+T cells through CD28, resulting in the initiation of acquired immunity [
      • Salomon B.
      • Bluestone J.A.
      Complexities of CD28/B7: CTLA-4 costimulatory pathways in autoimmunity and transplantation.
      ]. During T-cell activation, IL-12 production by DCs, which is induced by stimuli such as pathogen-associated molecular patterns, plays a critical role in the differentiation into Th1 cells and the generation of CTLs [
      • Trinchieri G.
      Interleukin-12 and the regulation of innate resistance and adaptive immunity.
      ,
      • Trinchieri G.
      Interleukin-12 and its role in the generation of TH1 cells.
      ,
      • Gately M.K.
      • Wolitzky A.G.
      • Quinn P.M.
      • Chizzonite R.
      Regulation of human cytolytic lymphocyte responses by interleukin-12.
      ,
      • Salem M.L.
      • Salman S.
      • Barnawi I.O.
      Brief in vitro IL-12 conditioning of CD8+ T Cells for anticancer adoptive T cell therapy.
      ]. The antigen specificity is crucial; in allergy, the antigens are proteins bound to the chemical, hapten, and in the RD-induced leukoderma, the antigens could be proteins bound to RD such as tyrosinase in melanocytes. However, h-CLAT is disadvantaged in not being able to evaluate pro-haptens, because THP-1 cells have only limited metabolic capacity [
      • Chipinda I.
      • Ruwona T.B.
      • Templeton S.P.
      • Siegel P.D.
      Use of the human monocytic leukemia THP-1 cell line and co-incubation with microsomes to identify and differentiate hapten and prohapten sensitizers.
      ]. To compensate for that disadvantage and still use that method to predict the potential to induce leukoderma, we established a novel coculture system, h-CLATw/M, consisting of THP-1 cells and melanoma SK-MEL-37 cells as melanocytes. Using the new coculture system, we demonstrated that RD upregulates CD86 and IL-12 in THP-1 cells via ROS generation and ATP release from melanoma cells by causing cytotoxicity against them. Without the melanoma cells, RD only slightly upregulated CD86 and IL-12 expression in h-CLAT. Thus, in h-CLAT, RD may have skin-sensitizing potential that induces contact hypersensitivity, but whether the upregulation observed with RD in h-CLAT has the potential to induce leukoderma cannot be predicted. By contrast, with the melanoma cells in h-CLATw/M, RD upregulated CD86 and IL-12 expression more so than it did in h-CLAT. These results suggest that RD could be a potent pro-hapten sensitizer by acting on melanocytes and consequently generating ROS, releasing ATP and upregulating CD86 and IL-12, possibly leading to the generation of melanocyte-specific CTLs. Both ROS and ATP induce IL-12 production from macrophages and DCs [
      • Schnurr M.
      • Then F.
      • Galambos P.
      • Scholz C.
      • Siegmund B.
      • Endres S.
      • et al.
      Extracellular ATP and TNF-alpha synergize in the activation and maturation of human dendritic cells.
      ,
      • Aramaki Y.
      • Yotsumoto S.
      • Watanabe H.
      • Tsuchiya S.
      NADPH-oxidase may contribute to IL-12 production in macrophages stimulated with CpG phosphorothioate oligodeoxynucleotides.
      ,
      • Ichikawa S.
      • Miyake M.
      • Fujii R.
      • Konishi Y.
      MyD88 associated ROS generation is crucial for Lactobacillus induced IL-12 production in macrophage.
      ]. Complicating matters, mutual regulation between ROS and ATP has also been reported: ROS induces ATP release in the skin [
      • Onami K.
      • Kimura Y.
      • Ito Y.
      • Yamauchi T.
      • Yamasaki K.
      • Aiba S.
      Nonmetal haptens induce ATP release from keratinocytes through opening of pannexin hemichannels by reactive oxygen species.
      ], and extracellular ATP as a danger signal was recently demonstrated to cause ROS production, inflammasome activation, and apoptosis in keratinocytes [
      • Ahn Y.
      • Seo J.
      • Lee E.J.
      • Kim J.Y.
      • Park M.Y.
      • Hwang S.
      • et al.
      ATP-P2X7-Induced Inflammasome Activation Contributes to Melanocyte Death and CD8+ T-Cell Trafficking to the Skin in Vitiligo.
      ]. Thus, the present results greatly support the skin-sensitizing potential of RD after metabolism by tyrosinase in melanocytes and that both ROS and ATP are important for it. CD86 is a critical costimulatory molecule expressed on the cell surface of DCs after activation, and the interaction between CD86 and CD28 is necessary to properly stimulate nave T cells and subsequently induce IL-2 production, which is important to prevent energy, leading to the activation of adaptive immunity and sensitization [
      • Salomon B.
      • Bluestone J.A.
      Complexities of CD28/B7: CTLA-4 costimulatory pathways in autoimmunity and transplantation.
      ]. Although the upregulation of CD86 is a necessary marker and h-CLAT is a validated method for skin sensitization, h-CLAT is not intended to be used alone, but instead, along with other assays in a Defined Approach (OECD 497) [51] to declare something a skin sensitizer. Therefore, further studies are necessary to confirm whether RD has skin-sensing potential, and the h-CLATw/M would be a useful tool to predict the sensing potential.
      To predict the tyrosinase- and melanocyte-specific sensitizing potential of RD, a leukoderma-inducible whitening tyrosinase inhibitor that is a pro-hapten sensitizer activated by metabolism with tyrosinase in melanocytes, we applied melanocytes to the h-CLAT model, establishing a novel coculture system, h-CLATw/M, consisting of THP-1 and SK-MEL-37 melanoma cells. Using h-CLATw/M, we partially clarified the molecular mechanism underlying RD-induced sensitization, finding that ROS generation and ATP release are critically important for CD86 and IL-12 upregulation. Moreover, h-CLATw/M could discriminate leukoderma-inducible whitening tyrosinase inhibitors from leukoderma-noninducible whitening agents by using flow cytometry to measure CD86 upregulation on THP-1 cells. Because this method is similar to the widely used h-CLAT method, and therefore familiar to many, h-CLATw/M could become a useful tool for predicting the leukoderma-inducible potential of chemicals. Nevertheless, further studies are needed to validate studies of h-CLATw/M using additional leukoderma-inducible and leukoderma-noninducible agents and to demonstrate that this method could lead to the formation of CTLs toward melanocytes.

      Funding

      This study was supported by a Grant-in-Aid for Scientific Research (KAKENHI) (Grant No. 20K22885) from the Japan Society for the Promotion of Science (JSPS) and Pola Chemical Industries, Inc.

      CRediT authorship contribution statement

      Yasuhiro Katahira: Conceptualization, Methodology, Formal analysis, Investigation, Validation, Data curation, Writing – original draft, Writing – review & editing, Visualization, Funding acquisition. Eri Sakamoto: Conceptualization, Methodology, Formal analysis, Investigation, Validation, Visualization. Aruma Watanabe: Investigation, Validation. Yuma Furusaka: Investigation, Validation. Shinya Inoue: Investigation, Validation. Hideaki Hasegawa: Investigation, Validation. Izuru Mizoguchi: Formal analysis, Investigation, Validation. Kazuyuki Yo: Conceptualization, Writing – review & editing, Supervision, Project administration. Fumiya Yamaji: Conceptualization, Writing – review & editing, Supervision, Project administration. Akemi Toyoda: Conceptualization, Writing – review & editing, Supervision, Project administration. Takayuki Yoshimoto: Conceptualization, Methodology, Writing – review & editing, Supervision, Project administration, Funding acquisition.

      Declaration of Competing Interest

      Kazuyuki Yo, Fumiya Yamaji, and Akemi Toyoda are employees of POLA Chemical Industries, Inc.

      Acknowledgments

      The authors thank Dr. L.J. Old (Memorial Sloan Kettering Cancer Center, New York, NY, USA) and Dr. H. Nagai (Kobe University) for the SK-MEL-28 and SK-MEL-37 cell lines, and G361 and SBcl2 cell lines, respectively.

      Appendix A. Supplementary material

      References

        • Harris J.E.
        Chemical-induced vitiligo.
        Dermatol. Clin. 2017; 35: 151-161
        • Tanemura A.
        • Yang L.
        • Yang F.
        • Nagata Y.
        • Wataya-Kaneda M.
        • Fukai K.
        • et al.
        An immune pathological and ultrastructural skin analysis for rhododenol-induced leukoderma patients.
        J. Dermatol. Sci. 2015; 77: 185-188
        • Nishioka M.
        • Tanemura A.
        • Yang L.
        • Tanaka A.
        • Arase N.
        • Katayama I.
        Possible involvement of CCR4+CD8+ T cells and elevated plasma CCL22 and CCL17 in patients with rhododenol-induced leukoderma.
        J. Dermatol. Sci. 2015; 77: 188-190
        • Fujiyama T.
        • Ikeya S.
        • Ito T.
        • Tatsuno K.
        • Aoshima M.
        • Kasuya A.
        • et al.
        Melanocyte-specific cytotoxic T lymphocytes in patients with rhododendrol-induced leukoderma.
        J. Dermatol. Sci. 2015; 77: 190-192
        • Tokura Y.
        • Fujiyama T.
        • Ikeya S.
        • Tatsuno K.
        • Aoshima M.
        • Kasuya A.
        • et al.
        Biochemical, cytological, and immunological mechanisms of rhododendrol-induced leukoderma.
        J. Dermatol. Sci. 2015; 77: 146-149
        • Ito S.
        • Wakamatsu K.
        Biochemical mechanism of rhododendrol-induced leukoderma.
        Int J. Mol. Sci. 2018; 19
        • Inoue S.
        • Katayama I.
        • Suzuki T.
        • Tanemura A.
        • Ito S.
        • Abe Y.
        • et al.
        Rhododendrol-induced leukoderma update II: Pathophysiology, mechanisms, risk evaluation, and possible mechanism-based treatments in comparison with vitiligo.
        J. Dermatol. 2021;
        • van den Boorn J.G.
        • Picavet D.I.
        • van Swieten P.F.
        • van Veen H.A.
        • Konijnenberg D.
        • van Veelen P.A.
        • et al.
        Skin-depigmenting agent monobenzone induces potent T-cell autoimmunity toward pigmented cells by tyrosinase haptenation and melanosome autophagy.
        J. Invest Dermatol. 2011; 131: 1240-1251
        • Ito S.
        • Nishigaki A.
        • Ishii-Osai Y.
        • Ojika M.
        • Wakamatsu K.
        • Yamashita T.
        • et al.
        Mechanism of putative neo-antigen formation from N-propionyl-4-S-cysteaminylphenol, a tyrosinase substrate, in melanoma models.
        Biochem Pharm. 2012; 84: 646-653
        • Westerhof W.
        • Manini P.
        • Napolitano A.
        • d'Ischia M.
        The haptenation theory of vitiligo and melanoma rejection: a close-up.
        Exp. Dermatol. 2011; 20: 92-96
        • Gu L.
        • Zeng H.
        • Takahashi T.
        • Maeda K.
        In vitro methods for predicting chemical leukoderma caused by quasi-drug cosmetics.
        Cosmetics. 2017; 4: 31
        • Gabe Y.
        • Miyaji A.
        • Kohno M.
        • Hachiya A.
        • Moriwaki S.
        • Baba T.
        Substantial evidence for the rhododendrol-induced generation of hydroxyl radicals that causes melanocyte cytotoxicity and induces chemical leukoderma.
        J. Dermatol. Sci. 2018; 91: 311-316
        • OECD Test No
        442E: In Vitro Skin Sensitisation: In Vitro Skin Sensitisation assays addressing the Key Event on activation of dendritic cells on the Adverse Outcome Pathway for Skin Sensitisation.
        OECD Guidel. Test. Chem. 2018;
        • Croft M.
        • Dubey C.
        Accessory molecule and costimulation requirements for CD4 T cell response.
        Crit. Rev. Immunol. 2017; 37: 261-290
        • Ashikaga T.
        • Hoya M.
        • Itagaki H.
        • Katsumura Y.
        • Aiba S.
        Evaluation of CD86 expression and MHC class II molecule internalization in THP-1 human monocyte cells as predictive endpoints for contact sensitizers.
        Toxicol. Vitr. 2002; 16: 711-716
        • Eskes C.
        • Hennen J.
        • Schellenberger M.T.
        • Hoffmann S.
        • Frey S.
        • Goldinger-Oggier D.
        • et al.
        The HaCaT/THP-1 Cocultured Activation Test (COCAT) for skin sensitization: a study of intra-lab reproducibility and predictivity.
        ALTEX. 2019; 36: 613-622
        • Chipinda I.
        • Ruwona T.B.
        • Templeton S.P.
        • Siegel P.D.
        Use of the human monocytic leukemia THP-1 cell line and co-incubation with microsomes to identify and differentiate hapten and prohapten sensitizers.
        Toxicology. 2011; 280: 135-143
        • Carey T.E.
        • Takahashi T.
        • Resnick L.A.
        • Oettgen H.F.
        • Old L.J.
        Cell surface antigens of human malignant melanoma: mixed hemadsorption assays for humoral immunity to cultured autologous melanoma cells.
        Proc. Natl. Acad. Sci. USA. 1976; 73: 3278-3282
        • Williams M.A.
        • Bevan M.J.
        Effector and memory CTL differentiation.
        Annu Rev. Immunol. 2007; 25: 171-192
        • Kondo M.
        • Kawabata K.
        • Sato K.
        • Yamaguchi S.
        • Hachiya A.
        • Takahashi Y.
        • et al.
        Glutathione maintenance is crucial for survival of melanocytes after exposure to rhododendrol.
        Pigment Cell Melanoma Res. 2016; 29: 541-549
        • Rutault K.
        • Alderman C.
        • Chain B.M.
        • Katz D.R.
        Reactive oxygen species activate human peripheral blood dendritic cells.
        Free Radic. Biol. Med. 1999; 26: 232-238
        • Kantengwa S.
        • Jornot L.
        • Devenoges C.
        • Nicod L.P.
        Superoxide anions induce the maturation of human dendritic cells.
        Am. J. Respir. Crit. Care Med. 2003; 167: 431-437
        • Byamba D.
        • Kim T.G.
        • Kim D.H.
        • Je J.H.
        • Lee M.G.
        The Roles of Reactive Oxygen Species Produced by Contact Allergens and Irritants in Monocyte-derived Dendritic Cells.
        Ann. Dermatol. 2010; 22: 269-278
        • Aldini G.
        • Altomare A.
        • Baron G.
        • Vistoli G.
        • Carini M.
        • Borsani L.
        • et al.
        N-Acetylcysteine as an antioxidant and disulphide breaking agent: the reasons why.
        Free Radic. Res. 2018; 52: 751-762
        • Sebastiao A.I.
        • Ferreira I.
        • Brites G.
        • Silva A.
        • Neves B.M.
        • Teresa Cruz M.
        NLRP3 Inflammasome and Allergic Contact Dermatitis: A Connection to Demystify.
        Pharmaceutics. 2020; 12: 867
        • Miyazawa M.
        • Ito Y.
        • Kosaka N.
        • Nukada Y.
        • Sakaguchi H.
        • Suzuki H.
        • et al.
        Role of TNF-alpha and extracellular ATP in THP-1 cell activation following allergen exposure.
        J. Toxicol. Sci. 2008; 33: 71-83
        • Martins J.D.
        • Silva A.
        • Ferreira I.
        • Goncalo M.
        • Custodio J.B.A.
        • Lopes M.C.
        • et al.
        Adenosine diphosphate involvement in THP-1 maturation triggered by the contact allergen 1-fluoro-2,4-dinitrobenzene.
        Toxicol. Res (Camb. ). 2016; 5: 1512-1521
        • Schnurr M.
        • Then F.
        • Galambos P.
        • Scholz C.
        • Siegmund B.
        • Endres S.
        • et al.
        Extracellular ATP and TNF-alpha synergize in the activation and maturation of human dendritic cells.
        J. Immunol. 2000; 165: 4704-4709
        • Jiang L.H.
        • Mackenzie A.B.
        • North R.A.
        • Surprenant A.
        Brilliant blue G selectively blocks ATP-gated rat P2X(7) receptors.
        Mol. Pharm. 2000; 58: 82-88
        • Axten J.M.
        • Medina J.R.
        • Feng Y.
        • Shu A.
        • Romeril S.P.
        • Grant S.W.
        • et al.
        Discovery of 7-methyl-5-(1-{[3-(trifluoromethyl)phenyl]acetyl}-2,3-dihydro-1H-indol-5-yl)-7H-p yrrolo[2,3-d]pyrimidin-4-amine (GSK2606414), a potent and selective first-in-class inhibitor of protein kinase R (PKR)-like endoplasmic reticulum kinase (PERK).
        J. Med Chem. 2012; 55: 7193-7207
        • Garg A.D.
        • Krysko D.V.
        • Verfaillie T.
        • Kaczmarek A.
        • Ferreira G.B.
        • Marysael T.
        • et al.
        A novel pathway combining calreticulin exposure and ATP secretion in immunogenic cancer cell death.
        EMBO J. 2012; 31: 1062-1079
        • Ahmad S.
        • Ahmad A.
        • White C.W.
        Purinergic signaling and kinase activation for survival in pulmonary oxidative stress and disease.
        Free Radic. Biol. Med. 2006; 41: 29-40
        • Cruz C.M.
        • Rinna A.
        • Forman H.J.
        • Ventura A.L.
        • Persechini P.M.
        • Ojcius D.M.
        ATP activates a reactive oxygen species-dependent oxidative stress response and secretion of proinflammatory cytokines in macrophages.
        J. Biol. Chem. 2007; 282: 2871-2879
        • Onami K.
        • Kimura Y.
        • Ito Y.
        • Yamauchi T.
        • Yamasaki K.
        • Aiba S.
        Nonmetal haptens induce ATP release from keratinocytes through opening of pannexin hemichannels by reactive oxygen species.
        J. Invest Dermatol. 2014; 134: 1951-1960
        • Trinchieri G.
        Interleukin-12 and the regulation of innate resistance and adaptive immunity.
        Nat. Rev. Immunol. 2003; 3: 133-146
        • Aramaki Y.
        • Yotsumoto S.
        • Watanabe H.
        • Tsuchiya S.
        NADPH-oxidase may contribute to IL-12 production in macrophages stimulated with CpG phosphorothioate oligodeoxynucleotides.
        Biol. Pharm. Bull. 2002; 25: 351-355
        • Ichikawa S.
        • Miyake M.
        • Fujii R.
        • Konishi Y.
        MyD88 associated ROS generation is crucial for Lactobacillus induced IL-12 production in macrophage.
        PLoS One. 2012; 7e35880
        • Fukuda Y.
        • Nagano M.
        • Futatsuka M.
        Occupational leukoderma in workers engaged in 4-(p-hydroxyphenyl)-2-butanone manufacturing.
        J. Occup. Health. 1998; 40: 118-122
        • Fukuda Y.
        • Nagano M.
        • Arimatsu Y.
        • Futatsuka M.
        An experimental study on depigmenting activity of 4-(p-hydroxyphenyl)-2-butanone in C57 black mice.
        J. Occup. Health. 1998; 40: 97-102
        • Pillaiyar T.
        • Manickam M.
        • Namasivayam V.
        Skin whitening agents: medicinal chemistry perspective of tyrosinase inhibitors.
        J. Enzym. Inhib. Med Chem. 2017; 32: 403-425
        • Pullar J.M.
        • Carr A.C.
        • Vissers M.C.M.
        The Roles of Vitamin C in Skin Health.
        Nutrients. 2017; 9
        • Carita A.C.
        • Fonseca-Santos B.
        • Shultz J.D.
        • Michniak-Kohn B.
        • Chorilli M.
        • Leonardi G.R.
        Vitamin C: One compound, several uses. Advances for delivery, efficiency and stability.
        Nanomedicine. 2020; 24102117
        • Zerbinati N.
        • Sommatis S.
        • Maccario C.
        • Di Francesco S.
        • Capillo M.C.
        • Rauso R.
        • et al.
        The Anti-Ageing and Whitening Potential of a Cosmetic Serum Containing 3-O-ethyl-l-ascorbic Acid.
        Life (Basel). 2021; 11
        • Tse T.W.
        • Hui E.
        Tranexamic acid: an important adjuvant in the treatment of melasma.
        J. Cosmet. Dermatol. 2013; 12: 57-66
        • OECD
        OECD Guideline No. 497: Defined Approaches on Skin Sensitization.
        OECD Guidel. Test. Chem., Sect. 4. 2021;
        • Salomon B.
        • Bluestone J.A.
        Complexities of CD28/B7: CTLA-4 costimulatory pathways in autoimmunity and transplantation.
        Annu Rev. Immunol. 2001; 19: 225-252
        • Trinchieri G.
        Interleukin-12 and its role in the generation of TH1 cells.
        Immunol. Today. 1993; 14: 335-338
        • Gately M.K.
        • Wolitzky A.G.
        • Quinn P.M.
        • Chizzonite R.
        Regulation of human cytolytic lymphocyte responses by interleukin-12.
        Cell Immunol. 1992; 143: 127-142
        • Salem M.L.
        • Salman S.
        • Barnawi I.O.
        Brief in vitro IL-12 conditioning of CD8+ T Cells for anticancer adoptive T cell therapy.
        Cancer Immunol. Immunother. 2021; 70: 2751-2759
        • Ahn Y.
        • Seo J.
        • Lee E.J.
        • Kim J.Y.
        • Park M.Y.
        • Hwang S.
        • et al.
        ATP-P2X7-Induced Inflammasome Activation Contributes to Melanocyte Death and CD8+ T-Cell Trafficking to the Skin in Vitiligo.
        J. Invest Dermatol. 2020; 140 (e1794): 1794-1804