Biphasic pro-melanogenic and pro-apoptotic effects of all-trans-retinoic acid (ATRA) on human melanocytes: Time-course study

G Model

DESC-2550; No. of Pages 9 Journal of Dermatological Science xxx (2013) xxx–xxx

Contents lists available at SciVerse ScienceDirect

Journal of Dermatological Science journal homepage: www.jdsjournal.com

Biphasic pro-melanogenic and pro-apoptotic effects of all-transretinoic acid (ATRA) on human melanocytes: Time-course study Ioana Baldea a, Gertrude-Emilia Costin b, Yiqun Shellman c, Katerina Kechris d, Elena Diana Olteanu a, Adriana Filip a, Maria Rodica Cosgarea e,**, David Albert Norris c, Stanca Ariana Birlea c,f,* a

Department of Physiology, University of Medicine and Pharmacy ‘‘Iuliu Hatieganu’’, Cluj-Napoca, Romania Institute for In Vitro Sciences Inc. (IIVS), Gaithersburg, MD, USA c Department of Dermatology, University of Colorado Anschutz Medical Campus, Aurora, CO, USA d Department of Public Health, University of Colorado Anschutz Medical Campus, Aurora, CO, USA e Department of Dermatology, University of Medicine and Pharmacy ‘‘Iuliu Hatieganu’’, Cluj-Napoca, Romania f Human Medical Genetics Program, University of Colorado Anschutz Medical Campus, Aurora, CO, USA b

A R T I C L E I N F O

A B S T R A C T

Article history: Received 1 December 2012 Received in revised form 3 May 2013 Accepted 9 June 2013

Background: The effects of retinoids on melanogenesis and their mechanism as depigmenting agents in topical therapy have not been fully elucidated. Conflicting data about their impact on melanogenic pathways have been reported. Objective: To investigate the effects of all-trans-retinoic acid (ATRA) on normal human melanocytes from Caucasian subjects. Methods: We assessed ATRA’s cytotoxicity by measuring viability with a cell proliferation assay, and apoptotic effects using Annexin V and g-H2AX markers. ATRA’s melanogenic activity was investigated based on spectrophotometric measurement of melanin content and tyrosinase enzymatic activity. Tyrosinase expression was assessed by Western blotting. We tested the antioxidant activity of superoxide dismutase (SOD) and catalase (CAT) in melanocytes using a spectrophotometric assay. Results: Of the concentrations tested in this 72 h time-course study, the 1.0 mM ATRA had a well-defined two-stage pro-melanogenic and pro-apoptotic effect on melanocytes. In the first 6 h, treated cells showed significant increase (p  0.01) of melanin content, tyrosinase, SOD, and CAT activities compared to the controls. While overall tyrosinase expression was not affected by ATRA, all other tested parameters decreased progressively beyond the short-term point of 6 h. ATRA treatment of over 6 h induced melanocyte apoptosis, as shown by the time-dependent decrease in cell viability, coupled with significant increase in Annexin V positive cells and nuclear accumulation of g-H2AX foci. Conclusion: The results obtained using this testing platform show a biphasic ATRA action: immediate pro-melanogenic effect and longer-term exposure pro-apoptotic activity. These data qualify ATRA as a potent tool to better understand the mechanisms that regulate the pigmentary system. ß 2013 Japanese Society for Investigative Dermatology. Published by Elsevier Ireland Ltd. All rights reserved.

Keywords: Retinoids ATRA Melanogenesis Proliferation Apoptosis

1. Introduction Retinoids (vitamin A and derivatives) have been utilized as suppressors of excessive keratinization and inflammation in skin disorders (like acne, psoriasis) [1]. They have photo-protective effects on normal skin and showed chemo-preventive effects on

* Corresponding author at: Department of Dermatology, University of Colorado Anschutz Medical Campus, Aurora, CO, USA. Tel.: +1 303 724 6690; fax: +1 303 724 4048. ** Corresponding author. Tel.: +40 264 592 394; fax: +40 264 592 394. E-mail addresses: [email protected] (S.A. Birlea). [email protected] (M.R. Cosgarea),

epithelial skin cancers, based on their capability to diminish the melanocyte and keratinocyte atypia [2]. On pigmentary disorders such as melasma, solar lentigines and post-inflammatory hyperpigmentation, retinoids act as lightening agents by correcting the associated irregular hyperpigmentation [1]. At the cellular and molecular level, retinoids have shown a bimodal effect on melanocytes, acting either as stimulators or inhibitors of melanin production. Their ability to reduce melanin synthesis was supported by various mechanisms such as blocking tyrosinase transcription, dispersion of keratinocyte pigment granules, or enhancement of epidermal cell turnover [3]. In different experimental settings, they modulated cell growth, differentiation and apoptosis of skin epithelial cells, thus

0923-1811/$36.00 ß 2013 Japanese Society for Investigative Dermatology. Published by Elsevier Ireland Ltd. All rights reserved. http://dx.doi.org/10.1016/j.jdermsci.2013.06.004

Please cite this article in press as: Baldea I, et al. Biphasic pro-melanogenic and pro-apoptotic effects of all-trans-retinoic acid (ATRA) on human melanocytes: Time-course study. J Dermatol Sci (2013), http://dx.doi.org/10.1016/j.jdermsci.2013.06.004

G Model

DESC-2550; No. of Pages 9 2

I. Baldea et al. / Journal of Dermatological Science xxx (2013) xxx–xxx

contributing to the maintenance of their homeostasis [4]. An inhibitory melanogenic response to retinoic acid was described by most of the clinical studies reported thus far. It has been shown that an average of 3–10 month application of topical retinoic acid can reduce the irregular hyperpigmentation associated with previous ultraviolet (UV) exposure in Caucasians [5], with postinflammatory hyperpigmentation and melasma in African-Americans [6,7], and with photo-aging in Chinese and Japanese patients [8]. Other experiments showed that a short-term treatment (4 days) with retinoic acid applied topically induced tyrosinase activity on Caucasian skin, but had no effect on African-American skin [9]. To explore more thoroughly the effect of retinoids on the melanogenic pathways, we tested the activity of all-transretinoic acid (ATRA) on Caucasian normal human melanocytes with focus on melanogenesis and apoptosis. Furthermore, we investigated ATRA’s effects on the enzymatic activity of superoxide dismutase (SOD) and catalase (CAT), due to their well-known implication in the defense against oxidative stress which was reported in association with melanin formation [10]. SOD and CAT operate in combination: the role of SOD is to dismute the superoxide anion radicals, generating hydrogen peroxide and oxygen, while CAT removes the hydrogen peroxide. Thus, the purpose of our experiments was to investigate if ATRA, a compound that can influence the level of melanin production, may also change the dynamics of enzymatic activities of antioxidative stress enzymes SOD and CAT. Gaining a better understanding of ATRA’s effects on human melanocytes is essential for supporting its preventive and therapeutic roles in pigmentation disorders. 2. Materials and methods

2.2. Cell culture Adult normal primary epidermal human melanocytes were obtained as previously described [12] from healthy Caucasian subjects (skin phototypes II, III), with ages between 23 and 25 years old. Briefly, normal skin biopsies were trimmed of excess subcutaneous tissue and dermis and were incubated overnight in 2000 IU/mL collagenase. The epidermal cells separated by trypsinization were then resuspended and seeded in serum free KGM. All cultures were fed twice weekly and incubated in a humidified atmosphere at 37 8C and 5% CO2. At first passage, the melanocytes were separated by differential trypsinization and were resuspended in MGM. All cell culture experiments were conducted in triplicate using sub-confluent cultures at the 3rd and 4th passage in MBM. MBM maintains cell survival over a short culture time, thus allowing for experiments performed without exposure to potent non-physiologic stimulants of proliferation and/or melanogenesis [12]. This approach eliminates any bias due to melanin stimulators contained in the media and was used to better assess ATRA’s independent effect on melanin production. For the melanogenic assays (Section 2.4), enzymatic activity assays (Section 2.5), Western blotting (Section 2.6), and cell proliferation (Section 2.7) experiments, melanocytes were seeded in Petri dishes at a density of 104 cm 2 for 24 h in MBM. Upon completion of ATRA exposure and after removal of the compound from the culture media, melanocytes were further cultured in fresh MBM for the next 24 h and then were processed following the specific procedures for each assay performed. Melanocytes were exposed to ATRA (0.01, 0.1, and 1.0 mM) for 6, 12, 24, and 72 h, timepoints at which measurements of the parameters considered in the study were taken. In parallel, the same experiments were performed using untreated melanocytes that served as controls, and for which an additional measurement was performed at the 0 h time-point.

2.1. Materials 2.3. Cell lysis Keratinocyte growth medium (KGM) and melanocyte basal medium (MBM) were from Promocell (Heidelberg, Germany); melanocyte growth medium (MGM) and neutral collagenase were from Cellsystems (Hamburg, Germany). Trypsin (0.05%)/ EDTA (0.02%), gentamicin, amphotericin, Ca++ and Mg++ freephosphate buffered saline (CMF-PBS), fetal calf serum (FCS), dimethyl sulfoxide (DMSO) were all from Biochrom AG (Berlin, Germany). Tretinoin (ATRA), Nonidet, protease inhibitor cocktail, L-3,4-dihydroxyphenylalanine (DOPA), synthetic melanin and 1,5-bis-2-(di-methylamino) ethylamino-4,8-dihydroxyanthracene-9,10-dione (DRAQ5) were all from Sigma Chemical Co. (St. Louis, MO, USA). The Bradford protein assay kit, SDSPAGE gels, SDS Laemli sample buffer, Tris–glycine buffers, nonfat dry milk, Tween 20 were from Biorad (Hercules, California, USA). CellTiter 961Aqueous Non-Radioactive Cell Proliferation Assay was from Promega Corporation (Madison, WI, USA). Annexin V-FITC and propidium iodide (PI) were from BD Pharmingen Biosciences (San Jose, CA, USA). Mouse antihistone g-H2AX (phospho-Ser 139) monoclonal antibody (gF3) was from Stressgen Bioreagents Corporation (Victoria, BC, Canada), and AlexaFluor 488 goat anti-mouse IgG2b (g2b) antibody was from Invitrogen Molecular Probes (Eugene, OR, USA). Rabbit anti-human tyrosinase antibody and goat antirabbit horseradish peroxidase-linked antibody were purchased from Santa Cruz Biotechnology (Heidelberg, Germany). ATRA stock solution (3 mg/mL) was prepared in DMSO, under argon gas, in subdued light and was stored at 70 8C until use. This solution was diluted in MBM immediately before experiments to achieve the final ATRA concentrations used in this study (0.01, 0.1, and 1.0 mM). The final concentration of DMSO in medium (0.01%) was not harmful to the cells [11].

The cell lysates used in all experiments were prepared as follows: the culture medium and melanocytes were collected and briefly centrifuged. The cells were then washed three times with cold PBS, collected by centrifugation and solubilized in extraction buffer (1% Nonidet P-40 in CMF-PBS containing protease inhibitors) for 1 h on ice. The samples were centrifuged at 14,000  g for 30 min at 4 8C and the supernatants were stored at 70 8C for further experiments. Protein concentrations were determined using the Bradford method according to the manufacturer’s specifications and using bovine serum albumin as standard. 2.4. Melanogenic assays Tyrosinase enzymatic activity (DOPA oxidase) was measured as described previously [13]. Briefly, 100 mL of each cell lysate were incubated with 1 mL DOPA (2.5 mg/mL prepared in 10 mM phosphate buffer, pH 7.2.) for 30 min at 37 8C. L-Dopachrome formation was measured spectrophotometrically at 475 nm using mushroom tyrosinase as positive control. The total melanin content was measured as described previously [14]. Following the cell lysis, the melanin pellets were dissolved in 1 M NaOH at 37 8C, overnight. Aliquots of each sample were transferred to 96-well plates and quantitated by absorbance at 475 nm against a standard curve generated using synthetic melanin. Total melanin content was expressed as mg/mL. 2.5. Enzymatic activity assays For SOD enzymatic activity, 50 mL of each cell lysate were mixed with 2.9 mL horse heart C cytochrome in 50 mM phosphate

Please cite this article in press as: Baldea I, et al. Biphasic pro-melanogenic and pro-apoptotic effects of all-trans-retinoic acid (ATRA) on human melanocytes: Time-course study. J Dermatol Sci (2013), http://dx.doi.org/10.1016/j.jdermsci.2013.06.004

G Model

DESC-2550; No. of Pages 9 I. Baldea et al. / Journal of Dermatological Science xxx (2013) xxx–xxx

buffer, pH 7.8, and then 50 mL xanthine oxidase 0.2 U/mL solution were added. Absorbance at 550 nm was recorded spectrophotometrically against a standard curve generated using pure bovine liver SOD [15,16]. For CAT enzymatic activity, 20 mL of each cell lysate were added to 3.0 mL solution of 10 mM H2O2 in potassium phosphate buffer and the absorbance at 240 nm was continuously measured spectrophotometrically for 3 min. 1 U of catalase equals the amount of enzyme which induces a change of 0.43 in the 240 nm absorbance during the 3 min period [15,16]. 2.6. Western blotting For Western blotting, lysates (20 mg protein/lane) were separated by electrophoresis under reducing conditions. Briefly, lysates were mixed 1:2 (v/v) with Laemli sample buffer (BioRad) containing 2-mercaptoethanol and were boiled at 95 8C for 10 min. Samples were separated by 8% SDS PAGE gels (BioRad) and transferred to polyvinylidenedifluoride membranes (BioRad) using Biorad Miniprotean system (BioRad). The blots were blocked in 5% nonfat dry milk in PBS, containing 0.1% Tween 20 (PBS-T) for 1 h at room temperature and were then incubated with rabbit antibody against tyrosinase (Santa Cruz Biotechnology) and GADPH (as protein loading control) (Trevigen-Biotechnology, Gaithersburg, MD, USA), respectively, diluted 1:500 in 1% non-fat milk in PBS-T, overnight at 4 8C. After washing with PBS-T, the blots were incubated with goat anti-rabbit horseradish peroxidase-linked antibody (1:1500) (Santa Cruz Biotechnology) in 1% non-fat milk in PBS-T for 90 min at room temperature. After six further washes (10 min each) with PBS-T, proteins were visualized and detected using Supersignal West Femto Chemiluminiscent substrate (Thermo Fisher Scientific), and a Gel Doc Imaging system equipped with a XRS camera and a Quantity One analysis software (Biorad). 2.7. Cell proliferation assay The melanocytes proliferation was performed using the CellTiter 961Aqueous Non-Radioactive Cell Proliferation Assay as specified by the manufacturer. Briefly, melanocytes were seeded in ELISA 96-well micro titration flat bottom plates for each experimental condition (ATRA concentrations and time-points) at a density of 104/well and were exposed to ATRA as described in Section 2.2. After each individual time-point, the ATRA-treated melanocytes and their counterpart controls were incubated for 2 h with 20 mL of 3-(4,5-dimethylthiazol-2-yl)-5-(3-carboxymethoxyphenyl)-2-(4-sulfophenyl)-2H-tetrazolium MTS/phenazine methosulfate (PMS) mixture in 100 mL culture medium [17]. The absorbance of each sample was read at 490 nm using an ELISA plate reader (Tecan, Ma¨nnedorf, Switzerland). 2.8. Microscopy For phase contrast and confocal microscopy experiments and observations, cells were seeded on glass chamber slides (Nalgene, Rochester, NY, USA) at a density of 5  104 cm 2 as previously described [18]. Pictures were taken using a Zeiss AxioObserver D1 inverted microscope equiped with AxioCam MR camera (phase contrast), and with a Zeiss LSM 710 confocal laser microscope, respectively. The Annexin V-fluorescein isothiocyanate (FITC) apoptosis detection kit was used in conjunction with the vital dye propidium iodide (PI) to identify late apoptotic (necrotic) cells (penetrated by PI); the kit was also used to identify viable cells (Annexin V-FITC negative) with intact cellular membranes as well as early apoptotic cells (Annexin V-FITC positive) which were not penetrated by PI [18]. Cells were stained with Annexin V-FITC according to the

3

manufacturer’s instructions and then fixed in 2% paraformaldehyde. The number of cells positive or negative for Annexin V and PI was counted for each sample, and cells were divided in the following two categories: (1) viable cells: Annexin V ( )/PI ( ); (2) apoptotic cells: Annexin V (+)/PI ( ) (in early apoptosis), and Annexin V (+)/PI (+) (in late apoptosis). At least 200 cells per sample were scored by eye, at a magnification of 63. Following the manufacturer’s protocol and a previous report [18], the same fields observed in phase contrast were subsequently stained with the Annexin V-FITC kit in conjunction with PI and the assessment of viable and apoptotic cells was performed. The total number of cells observed in phase contrast represented the viable and apoptotic cells (in early and late apoptosis). The experiments were performed in triplicate. Formation of g-H2AX foci, a marker for early appearance of high molecular weight DNA fragments (shown to be concomitant with early apoptosis) [19], was detected by immunostaining. Upon completion of exposure to ATRA, the melanocytes were fixed in 4% formaldehyde, incubated with mouse anti-g-H2AX antibody (1/500) and then with goat anti-mouse antibody conjugated with AlexaFluor 488 (green). DRAQ5 staining (red) for nuclei was used for comparison. At least 100 nuclei per condition were scored by eye, at a magnification of 63 using an Olympus fluorescence microscope BX41 equipped with an Olympus E330 camera using software Olympus Master version 1.41EX. The number of g-H2AX foci per nucleus was counted for each sample and the nuclei were divided in four categories: 3; 4–10; 11–30; 30 g-H2AX foci/cell, following guidelines provided by a previous report [20]. All results are reported as mean of three different determinations performed on each sample. 2.9. Statistical methods and analysis Two-way ANOVA, Student’s t test, and Fisher’s exact test were used for statistical analysis and performed with the R statistical software (http://www.r-project.org/). All results were considered significant for p  0.01. 3. Results 3.1. Assessment of ATRA cytotoxicity on normal human melanocytes We assessed ATRA’s cytotoxicity in order to select the concentration and exposure time that minimally impact the cellular proliferation while allowing modulation of the melanogenic response. As presented in Fig. 1a, the control melanocytes proliferated continuously up to 72 h, and the proliferation rate (expressed as OD490 nm values) was directly correlated with the time elapsed. ATRA induced an inhibition of proliferation, observed after 6 h for the highest concentration of 1.0 mM, and after 12 h for the lower ones (p < 1.0E 04). This effect was time- and dosedependent, and was maximal at 72 h. The two-way ANOVA analysis showed significant time, treatment and time–treatment interaction effects (p < 1.0E 03). In parallel with cell proliferation, we also investigated the cell morphology using phase contrast microscopy. Starting at 12 h, ATRA-treated cells exhibited distinctive cellular changes that were more evident in cultures treated with concentration of 1.0 mM. As illustrated in Fig. 1b, treated cells appeared less numerous compared to the controls, and exhibited specific cytotoxic signs of stress such as diminished dendricity, pleiomorphism with bipolar or spherical shapes, and loss of adherence. At 6 h we observed apparent normal cellular morphology for all concentrations used (data not shown for 0.1 mM and 0.01 mM ATRA), indicating that at this time-point ATRA allowed the maintenance of cell viability, with minimal toxicity of the compound.

Please cite this article in press as: Baldea I, et al. Biphasic pro-melanogenic and pro-apoptotic effects of all-trans-retinoic acid (ATRA) on human melanocytes: Time-course study. J Dermatol Sci (2013), http://dx.doi.org/10.1016/j.jdermsci.2013.06.004

G Model

DESC-2550; No. of Pages 9 4

I. Baldea et al. / Journal of Dermatological Science xxx (2013) xxx–xxx

Fig. 1. (a) Graph showing cell proliferation after 6 h, 12 h, 24 h and 72 h exposure to 1.0 mM, 0.1 mM, and 0.01 mM ATRA (vs. untreated melanocytes); mean values and standard deviations (n = 3) are shown; ATRA induced inhibition of proliferation, which was significant after 6 h for the highest concentration of 1.0 mM, and after 12 h for the lower ones (*p < 1.0E 04, proliferation of treated cells versus control, Student’s t-test). (b) Phase contrast images of normal human melanocytes in culture: normal cellular morphology observed in control melanocytes is shown in the left panel; in the next four panels, the melanocytes treated with 1.0 mM ATRA were less numerous compared to the controls; starting with 12 h of exposure, they gradually exhibited specific cytototoxic signs of stress such as diminished dendricity, pleiomorphism with bipolar or spherical shapes, and loss of adherence. Magnification 20.

3.2. Induction of apoptosis by ATRA in normal human melanocytes The induction of apoptosis in melanocytes treated with ATRA was investigated by two methods: (a) assessment of positive cells (%) for Annexin V-FITC and/or PI (Fig. 2a and b), and (b) focal accumulation of g-H2AX (Fig. 3a and b).

Fig. 2. Assessment of apoptosis by positive signal for Annexin V and PI in control melanocytes and in melanocytes treated with 1.0 mM ATRA, 0.1 mM ATRA, and 0.01 mM ATRA for 6 h, 12 h, 24 h, 72 h. (a) Percent of viable cells (no apoptosis). (b) Percent of apoptotic cells identified as Annexin V (+) showing a significant increase in all ATRA-treated cultures versus control melanocytes after 12 h of exposure (p  1.0E 05 (*); 1.0E 05  p  1.0E 03 (#); 1.0E 03  p  1.0E 02 (^); p  1.0E 02 (NS). NS = not significant; Fisher’s exact test was used for statistical comparison). Each bar represents mean and standard deviation (n = 3). The graph displayed in Fig. 2(a) uses a 0–100 scale, while the graph in Fig. 2(b) uses a 0–40 scale.

3.2.1. Evaluation of apoptosis using Annexin V-FITC/PI The Annexin V-FITC positive cells were detected by green fluorescence, while PI positive staining was indicated by the presence of fluorescent red nuclei (not shown). The viable cells (showing no apoptosis) were identified as Annexin V ( )/PI ( ); the apoptotic cells were identified as Annexin V (+)/PI ( ) (early apoptosis), and Annexin V (+)/PI (+) (late apoptosis), respectively. As presented in Fig. 2a (the % of viable melanocyte measured with the time ellapsed), the control cells were viable at all exposure times. By contrast, ATRA-treated cultures showed a gradual decline in the % of viable cells which was paralleled by an increase in the % of apoptotic cells, in direct correlation with the exposure time and ATRA concentration (Fig. 2b). At 6 h exposure (Fig. 2b) the cultures treated with all concentrations of ATRA showed a low % of apoptotic cells. The percentage of apoptotic cells increased at 12 h, although remained unremarkable compared with the controls. Starting with 24 h, all ATRA-treated cultures showed a significantly increased % of apoptotic cells compared to the control melanocytes (p < 1.0E 02), with the maxima at 72 h (p < 1.0E 04). 3.2.2. Evaluation of apoptosis by accumulation of g-H2AX foci The nuclear accumulation of g-H2AX foci was indicated in the confocal microscopy images by yellow fluorescence, resulted from the colocalization of g-H2AX (green) with the nuclear marker

Please cite this article in press as: Baldea I, et al. Biphasic pro-melanogenic and pro-apoptotic effects of all-trans-retinoic acid (ATRA) on human melanocytes: Time-course study. J Dermatol Sci (2013), http://dx.doi.org/10.1016/j.jdermsci.2013.06.004

G Model

DESC-2550; No. of Pages 9 I. Baldea et al. / Journal of Dermatological Science xxx (2013) xxx–xxx

5

Fig. 3. Assessment of apoptosis by formation of g-H2AX foci following exposure to ATRA for 6 h, 12 h, 24 h, 72 h. (a) Confocal microscopy images of normal human melanocytes treated with ATRA [double immunostaining with g-H2AX (green) and DRAQ5 (red)] vs. controls. Colocalized signal displays yellow color. The single left panel: control melanocytes at 0 h time-point show no or few foci/cell. The four right side panels: in melanocytes treated with 1.0 mM ATRA for 6 h, 12 h, 24 h and 72 h the number of g-H2AX foci/cell increased with the time elapsed. Arrows point toward the foci. Magnification 63. (b) The graphs represent the percentage of cells plotted against the four categories of g-H2AX foci in ATRA-treated (red lines for 1.0 mM ATRA and green lines for 0.01 mM ATRA) vs. control melanocytes (blue lines) at 6 h, 12 h, 24 h, 72 h. The green bars indicate increasing difference (D) between the % of cells having 0–3 foci in ATRA-treated vs. untreated melanocytes (and directly correlated with ATRA concentration and time elapsed); the pink bars indicate increasing difference (D) between the % of cells having 30 foci in ATRA-treated vs. untreated melanocytes (and directly correlated with ATRA concentration and time elapsed). For simplification, only data pertaining to the cells treated with the highest (1.0 mM) and lowest (0.01 mM) ATRA concentrations are displayed.

DRAQ5 (red). We observed no or a reduced number of nuclear gH2AX foci in the control population, whereas the 1.0 mM ATRAtreated melanocytes showed a proportionally increased number of foci associated with the exposure time (Fig. 3a). Similar results were obtained for the melanocytes treated with the lower ATRA concentrations (0.1 mM, and 0.01 mM) (data not shown). As shown in Fig. 3b, the % of total cell population of control melanocytes exhibiting a low number of foci (0–3) was high (>77%) at all exposure times, including the 0 time-point (not represented in the figure). At the opposite end of the spectrum, the % of control cells that had a high number of foci (>30) was

constantly low (<7%) at all exposure times, thus describing a viable cell culture. In the melanocytes treated with ATRA, we observed changes specific to a cell population characterized by early appearance of high molecular weight DNA fragments concurrent with early apoptosis, as follows: (a) the % of cells with 0–3 foci decreased progressively compared with control melanocytes following an inverse correlation with exposure time and ATRA concentration (p < 1.0E 05) (as shown in Fig. 3b by the mean of the difference D between the %

Please cite this article in press as: Baldea I, et al. Biphasic pro-melanogenic and pro-apoptotic effects of all-trans-retinoic acid (ATRA) on human melanocytes: Time-course study. J Dermatol Sci (2013), http://dx.doi.org/10.1016/j.jdermsci.2013.06.004

G Model

DESC-2550; No. of Pages 9 6

I. Baldea et al. / Journal of Dermatological Science xxx (2013) xxx–xxx

Fig. 4. (a) Spectrophotometric detection of melanin concentration (mg/mL) in normal human melanocytes treated with ATRA 1.0 mM, 0.1 mM, and 0.01 mM (vs. control) after 6 h, 12 h, 24 h, 72 h exposure. Note the significantly increased melanin content of melanocyte cultures in the first 24 h of exposure to ATRA, with a peak at 6 h. (b) Tyrosinase enzymatic activity (U/mg protein) in melanocytes treated with ATRA 1.0 mM, 0.1 mM, and 0.01 mM (vs. control) after 6 h 12 h, 24 h, 72 h of exposure. The enzymatic activity of tyrosinase was significantly increased at 6 h (vs. control) followed by a gradual decline; at 72 h the tyrosinase activity was below the level observed in controls. p < 1.0E 05 (*); 1.0E 05  p  1.0E 03 (#); 1.0E 03 < p  1.0E 02 (^); p > 1.0E 02 (NS). NS = not significant; Student’s t-test was used for statistical comparison. Each bar represents mean and standard deviation (n = 3).

of cells having 0–3 foci in ATRA-treated vs. untreated melanocytes). (b) the % of cells with 4–10, 11–30, and >30 foci progressively increased following a direct correlation with exposure time and ATRA concentration, the variation being more significant for % of cells with >30 foci (p < 1.0E 05) (as shown in Fig. 3b by the mean of the difference D between the % of cells having 30 foci in ATRA-treated vs. untreated melanocytes). 3.3. Biphasic effect of ATRA on melanin content and tyrosinase enzymatic activity 3.3.1. Melanin content To test the effect of ATRA on melanogenesis we determined the total melanin content in cultured melanocytes. The control melanocytes did not show significant variation in their melanin content with the time elapsed (Fig. 4a). The 1.0 mM ATRA increased the total melanin content in treated cells compared to the controls at all time-points (Fig. 4a); the difference was remarkable during the first 24 h of exposure, with the highest melanin content determined at 6 h. At 72 h we observed a sudden decrease in the melanin content compared to previous time-points, although there was still a slightly significant difference over the controls (p < 7.4E 03). In the melanocytes treated with the lower ATRA concentrations, the increase in melanin content in the first 24 h of exposure was less evident (Fig. 4a), but followed a similar pattern with that observed in 1.0 mM ATRA-treated cultures (Fig. 4a). Twoway analysis of variance (ANOVA) showed a significant time-effect and treatment-effect (p < 1.0E 04), and the lack of time– treatment interaction (following an additive effect). 3.3.2. Tyrosinase activity We verified the implications of ATRA treatment on the enzymatic activity of tyrosinase by detecting the levels of Ldopachrome formation. We did not observe notable variations with exposure time in the enzymatic activity of tyrosinase in control melanocytes (Fig. 4b). In contrast, the 1.0 mM ATRA treatment significantly increased tyrosinase activity at 6 h compared to the controls (p < 1.0E 06) (Fig. 4b), and the difference gradually decreased at 12 h and 24 h. At 72 h we observed a significant decrease of tyrosinase enzymatic activity compared to the controls (Fig. 4b), likely due to the cytotoxicity associated with longer-term exposure to the highest ATRA concentration (see also Fig. 1a). Lower ATRA concentrations

(0.1 mM, and 0.01 mM) showed similar (but more attenuated) effects than 1.0 mM (Fig. 4b). The two-way ANOVA analysis showed a highly significant time and treatment effects (p < 1.0E 05) and significant time–treatment interaction (p < 1.0E 04). 3.4. Modulation of tyrosinase expression by ATRA To further explore the impact of ATRA on tyrosinase expression, we performed Western blot analysis. As shown in Fig. 5, tyrosinase was detected as the mature form (68 kDa) in all samples analyzed. Tyrosinase protein levels were constantly well represented in the control and ATRA-treated melanocytes at all exposure times. Tyrosinase protein levels variation was not significant although a slight increase in protein synthesis was observed in all ATRA-treated cells particularly at 6 h, followed by a minor decrease at 12 h and subsequently at longer exposure times (24 and 72 h). 3.5. Biphasic effect of ATRA on SOD and CAT enzymatic activity In order to investigate the oxidative stress status of the melanocytes treated with ATRA, we assessed the enzymatic activity of two antioxidant enzymes, SOD and CAT using a spectrophotometric assay. 3.5.1. SOD enzymatic activity As shown in Fig. 6a, the increase in SOD enzymatic activity in the control melanocytes with time elapsed was unremarkable compared to the initial time-point measurement. ATRA at 1.0 mM significantly increased SOD enzymatic activity at 24 h, with the highest spike at 6 h. At 72 h, the SOD activity was significantly inhibited compared to the controls (p < 9.0E 06) (Fig. 6a). The melanocytes treated with 0.1 mM ATRA showed a similar response, but more attenuated than the effect of 1.0 mM concentration (Fig. 6a). In cultures treated with 0.01 mM ATRA, SOD activity was slightly inhibited at all time-points compared to the controls (Fig. 6a). 3.5.2. CAT enzymatic activity The CAT enzymatic activity in control melanocytes showed an unremarkable variation with the time elapsed. The highest 1.0 mM ATRA concentration significantly increased the CAT activity compared to controls at all exposure times (Fig. 6b), with a

Please cite this article in press as: Baldea I, et al. Biphasic pro-melanogenic and pro-apoptotic effects of all-trans-retinoic acid (ATRA) on human melanocytes: Time-course study. J Dermatol Sci (2013), http://dx.doi.org/10.1016/j.jdermsci.2013.06.004

G Model

DESC-2550; No. of Pages 9 I. Baldea et al. / Journal of Dermatological Science xxx (2013) xxx–xxx

7

Fig. 5. (a) Expression of human tyrosinase in normal human melanocytes treated with ATRA (vs. untreated melanocytes). Protein levels were analyzed by Western blot. Image analysis of Western blot bands’ intensity was performed and the data are shown as graph insets near each individual cell culture condition (control vs. ATRA treatments) (expressed as intensity/mm2). (b) GAPDH was used as loading control.

Fig. 6. (a) Spectrophotometric detection of SOD enzymatic activity (U/mg protein) in normal human melanocytes treated with ATRA 1.0 mM, 0.1 mM, and 0.01 mM (vs. control) after 6 h 12 h, 24 h, 72 h of exposure. For concentrations of 1.0 mM and 0.1 mM ATRA, increased SOD activity at 6 h (vs. control) was observed, followed by a gradual decline; at 72 h the SOD activity was significantly decreased for all ATRA concentrations compared to the level observed in controls. (b) CAT enzymatic activity (U/mg protein) in melanocytes treated with ATRA 1.0 mM, 0.1 mM, and 0.01 mM (vs. control) after 6 h, 12 h, 24 h, 72 h of exposure. All ATRA concentrations induced a peak of CAT enzymatic activity at 6 h, followed by gradual decline; at 24 h, the CAT activity was still significantly elevated compared to the controls. p < 1.0E 05 (*); 1.0E 05  p  1.0E 03 (#); 1.0E 03 < p  1.0E 02 (^); p > 1.0E 02 (NS). NS = not significant. Student’s t-test was used for statistical comparison. Each bar represents mean and standard deviation (n = 3).

maxima at 6 h (p < 1.0E 06), after which the enzymatic activity gradually decreased. A similar, but more attenuated effect was observed in melanocytes treated with 0.1 mM ATRA at all exposure times, and with 0.01 mM ATRA in the first 24 h of exposure (Fig. 6b). For both SOD and CAT enzymatic activities, the two-way ANOVA analysis indicated a weak treatment effect (p < 1.0E 02), and unremarkable time effect and time–treatment interaction (following an additive effect). 4. Discussion Retinoids have been reported to control the melanogenic pathways, however their precise mechanism of action remains unclear. Several in vitro experiments suggested their bimodal effect of either induction or inhibition of melanin synthesis. For example, studies based on normal human melanocytes and human and mouse melanoma lines which used similar exposure times and

concentrations to our study, showed that retinoic acid had a stimulatory effect on melanin production [21–23]. A different study [24] reported that retinoic acid (0.001–0.1 mM) increased melanin synthesis in mouse embryonic stem cells exposed up to 24 days and showed different effects on melanocyte depending on the stage of differentiation. While our experiments did not investigate the melanocyte differentiation when exposed to ATRA, it would be of interest to expand in the future the scope of our studies toward modulation of differentiation in various lines of human melanocytes. Earlier studies reported that short-term incubation (2–6 days) of mouse melanoma cells with retinoic acid had an inhibitory effect on melanogenesis [25–27]. These conflicting results may reside with the species identity of the cell lines (mouse melanoma) and experimental conditions used in these studies. It would be of importance to gain a better understanding of ATRA’s action on normal vs. melanoma and mouse vs. human cell lines and to identify the specific markers responsible for the opposite effects of retinoids thus far reported in literature.

Please cite this article in press as: Baldea I, et al. Biphasic pro-melanogenic and pro-apoptotic effects of all-trans-retinoic acid (ATRA) on human melanocytes: Time-course study. J Dermatol Sci (2013), http://dx.doi.org/10.1016/j.jdermsci.2013.06.004

G Model

DESC-2550; No. of Pages 9 8

I. Baldea et al. / Journal of Dermatological Science xxx (2013) xxx–xxx

Inconclusive data on the effect of retinoids have been reported by clinical studies as well, albeit leaning toward a lightening effect of ATRA. Most of these clinical studies used retinoic acid for an average of 3–16 months on white or dark phototypes, and showed improvement of the irregular hyperpigmentation following chronic sun exposure [28], liver spots [5], chronic inflammation [6], actinic lentigo [8,29], or melasma [7]. A different study showed lightening effects on darkly pigmented normal skin after shortterm treatment (4 days) with retinoic acid [9], while the lightly pigmented skin responded to the treatment with increased tyrosinase activity. Furthermore, retinoids were reported to significantly reduce the size and pigmentation of atypical nevi (or even lead to their elimination) [30], and to cause some histologic improvement [31], although the treated nevi still met the histological criteria for atypia [32]. To shed light on the bimodal effect of retinoids, we investigated the effect of ATRA on Caucasian human melanocytes with particular focus on proliferation/apoptosis and melanogenesis/ oxidative stress axes. Our data show that ATRA affected cell morphology and decreased cell proliferation in a dose- and timedependent manner (Fig. 1a and b); the significant decline of cell viability in ATRA-treated cultures was observed beyond the 6 h time-point for 1.0 mM ATRA, and beyond 12 h time-point for the lower concentrations. These results were supported by experiments that assessed apoptosis directly (using Annexin V) and indirectly (using g-H2AX foci). As such, in all ATRA-treated cultures the % of apoptotic cells (Fig. 2b) increased progressively in a timeand dose-dependent manner, becoming significantly different compared to the controls after the 12 h time-point. The development of g-H2AX foci was dose-dependent. The assessment of gH2AX foci indicated a progressive accumulation of high molecular weight DNA fragments, which suggest the gearing toward apoptosis beyond the 6 h time-point (Fig. 3a and b). All these results converge in showing that in ATRA-treated melanocytes, the 6 h time-point is indicative of sustained cell viability that allows for interpretation of data related to other endpoints used in our study. The current study showed that ATRA increased both melanin content and tyrosinase activity in treated cells up to 24 h, with a maximal peak at 6 h (Fig. 4a and b). These two parameters were time- and dose-dependent and decreased progressively up to 72 h. Our experiments tested the entire population of melanocytes (affected or not affected by ATRA) in order to mimic and hopefully best capture the outcome of ATRA treatment as assessed in the clinical setting. Thus, the enzymatic activity was assigned to the active tyrosinase synthesized by the viable melanocytes, while the melanin content represented the pigment produced by both viable and apoptotic cells. It would of interest to investigate the individual contribution of the melanocytes affected or not by ATRA to the melanogenic activity which is rather difficult to capture in the clinical studies. In our test system, the tyrosinase enzymatic activity and melanin content variation following ATRA treatment were not paralleled by changes in tyrosinase synthesis, and our results are consistent with other reports. As such, an earlier study [9] showed that induction of tyrosinase activity in white skin treated with 0.1% topical retinoic acid for 4 days was not accompanied by an increase in either tyrosinase mRNA or protein expression. Those data suggested that modulation of tyrosinase enzymatic activity may occur through post-translational mechanisms, or by suppression of a tyrosinase inhibitor or phosphorylation or glycosylation of preexisting tyrosinase molecules [33]. A different experiment showed similar results on melanoma cell cultures incubated with 1 mM ATRA for 6 days [34]. The post-translational regulation of tyrosinase activity, consisting in activation/down-regulation of tyrosinase was previously described after treatment of

melanocyte cultures with other compounds, like linoleic acid [35], phospholipase D2 (PLD2) [36], quercetin [37], N-butyldeoxynojirimycin (NB-DNJ) [38], etc. Different earlier reports showed that in cultured human melanocytes from light-skinned subjects, which are similar to the melanocyte cultures used in the present study, tyrosinase activity and melanin production are controlled primarily by a post-translational regulation of pre-existing enzyme and not by regulating tyrosinase gene transcription [39,40]. Several reports have shown that pigmentation does not always correlate with the expression of tyrosinase mRNA or protein, and thus its function is also regulated at the post-translational level. Thus, tyrosinase does not necessarily exist in a catalytically active state, and its post-translational activation could be an important control point for regulating melanin synthesis [41]. Modulation of tyrosinase activity by retinoic acid may involve multiple complex secondary and tertiary interactions that are not directly amenable to study in simple in vitro cell culture models [9]. In our experiments, the increase in melanin production and tyrosinase activity was also correlated in ATRA-treated cells with increased enzymatic activity of SOD (up to 12 h) and CAT (up to 24 h) with the maxima at 6 h for both parameters. These data reflect that the cells activate their oxidative stress mechanisms in response to ATRA treatment, but perhaps also in response to increased melanin production associated with short-term exposures. Melanin synthesis is an oxygen-dependent process that acts as a potential source of reactive oxygen species inside melanocytes [42]. Our experiments suggest that some antioxidant protection could be triggered when melanogenesis is stimulated, to cope with endogenous and exogenous sources of oxidative stress. The dual effect of ATRA on proliferation/apoptosis and melanogenesis axes shown in our experiments was previously reported in other testing systems. As such, ATRA (0.01–100 mM) inhibited the proliferation of NCC-melb4 mouse cells after 72 h incubation and induced apoptosis, while promoting cell melanization [43]. A different experiment using mouse B16F10 melanoma cell line showed similar effects using 1–6 day exposure to ATRA (0.01–1.0 mM) [23]. Finally, experiments using chick retinal pigment epithelial cells showed that retinoic acid inhibits cell proliferation and induces melanin synthesis [44]. We report herein on a biphasic effect of ATRA on melanocytes within the same testing system: pro-melanogenic – as a response to short-term treatment (up to 24 h, with a peak at 6 h), and proapoptotic – as a response to longer treatment (up to 72 h). In the current study, ATRA’s opposite effects were best reflected by the highest concentration (1.0 mM). Our data model the in vitro events that occur after short-term treatment and that may not be easily observed in clinical studies. While most of the clinical reports describe a long-term effect (3–16 months) of depigmentation under ATRA treatment, they lack the short-term exposure observations (first days of treatment); if the short-term exposure is accompanied by an increase in melanin production in the clinical setting, awaits clarification. Finally, it is difficult to extrapolate the in vitro model to the clinical setting; however, it may be of relevance that concentration of 1.0 mM used in the present study is close to that of the retinoid compound found in human skin treated with a 0.1% retinoic acid formula [45]. This formula currently used for the treatment of hyperpigmentation was identified to have an optimal penetration of the human epidermis, however sometimes accompanied by skin irritation. For this reason, the in vitro observations of melanocyte incubated with 1.0 mM ATRA can provide directions for future clinical trials in search for better treatments of hyperpigmentation disorders. In good correlation with other reported data on retinoic acid, our study shows its very complex mechanistic activity, from coordination of gene transcription, to post-transcriptional and

Please cite this article in press as: Baldea I, et al. Biphasic pro-melanogenic and pro-apoptotic effects of all-trans-retinoic acid (ATRA) on human melanocytes: Time-course study. J Dermatol Sci (2013), http://dx.doi.org/10.1016/j.jdermsci.2013.06.004

G Model

DESC-2550; No. of Pages 9 I. Baldea et al. / Journal of Dermatological Science xxx (2013) xxx–xxx

post-translational actions, likely driven by combination(s) of different experimental and study design parameters such as cell type, culture conditions, skin phototype, and level of melanization. The results reported herein represent the starting point in our experiments to comprehensively investigate the mechanism of ATRA’s action at the transcriptional and translational level expanded to the key proteins involved in the melanogenic pathways. Acknowledgments The authors thank Dr. Pompei Bolfa, Department of Morphopatology, University of Veterinary Medicine and Agricultural Studies, Cluj-Napoca, ROMANIA for the confocal microscopy photographs and Mr. Marius Alupei, Department of Experimental Biology, Faculty of Biology and Geology, Babes-Bolyai University, Cluj-Napoca, ROMANIA, for Western blot photographs. The authors also thank Drs. Richard Spritz (University of Colorado, Aurora, CO, USA) and Andrzej Slominski (University of Tennessee, Memphis, TN, USA) for helpful comments and suggestions. References [1] Orfanos CE, Zouboulis CC, Almond-Roesler B, Geilen CC. Current use and future potential role of retinoids in dermatology. Drugs 1997;53:358–88. [2] Cho S, Lowe L, Hamilton TA, Fisher GJ, Voorhees JJ, Kang S. Long-term treatment of photoaged human skin with topical retinoic acid improves epidermal cell atypia and thickens the collagen band in papillary dermis. J Am Acad Dermatol 2005;53:769–74. [3] Halder RM, Nootheti PK. Ethnic skin disorders overview. J Am Acad Dermatol 2003;48:S143–8. [4] Roberts AB, Roberts AB. Cellular biology and biochemistry of the retinoids. In: Sporn MB, Roberts AB, Goodman DS, editors. The retinoids. New York: Academic Press; 1994. p. 209–86. [5] Rafal ES, Griffiths CE, Ditre CM, Finkel LJ, Hamilton TA, Ellis CN, et al. Topical tretinoin (retinoic acid) treatment for liver spots associated with photodamage. N Engl J Med 1992;326:368–74. [6] Bulengo-Ransby SM, Griffiths CE, Kimbrough-Green CK, Finkel LJ, Hamilton TA, Ellis CN, et al. Topical tretinoin (retinoic acid) therapy for hyperpigmented lesions caused by inflammation of the skin in black patients. N Engl J Med 1993;328:1438–43. [7] Kimbrough-Green CK, Griffiths CE, Finkel LJ, Hamilton TA, Bulengo-Ransby SM, Ellis CN, et al. Topical retinoic acid (tretinoin) for melasma in black patients. A vehicle-controlled clinical trial. Arch Dermatol 1994;130:727–33. [8] Griffiths CE, Goldfarb MT, Finkel LJ, Roulia V, Bonawitz M, Hamilton TA, et al. Topical tretinoin (retinoic acid) treatment of hyperpigmented lesions associated with photoaging in Chinese and Japanese patients: a vehicle-controlled trial. J Am Acad Dermatol 1994;30:76–84. [9] Talwar HS, Griffiths CE, Fisher GJ, Russman A, Krach K, Benrazavi S, et al. Differential regulation of tyrosinase activity in skin of white and black individuals in vivo by topical retinoic acid. J Invest Dermatol 1993;100:800–5. [10] Maresca V, Flori E, Briganti S, Mastrofrancesco A, Fabbri C, Mileo AM, et al. Correlation between melanogenic and catalase activity in in vitro human melanocytes: a synergic strategy against oxidative stress. Pigment Cell Melanoma Res 2008;21:200–5. [11] Kashino G, Liu Y, Suzuki M, Masunaga S, Kinashi Y, Ono K, et al. An alternative mechanism for radioprotection by dimethyl sulfoxide; possible facilitation of DNA double-strand break repair. J Radiat Res 2010;51:733–40. [12] Morelli JG, Yohn JJ, Zekman T, Norris DA. Melanocyte movement in vitro: role of matrix proteins and integrin receptors. J Invest Dermatol 1993;101:605–8. [13] Ohkura T, Yamashita K, Mishima Y, Kobata A. Purification of hamster melanoma tyrosinases and structural studies of their asparagine-linked sugar chains. Arch Biochem Biophys 1984;235:63–77. [14] Gordon PR, Gilchrest BA. Human melanogenesis is stimulated by diacylglycerol. J Invest Dermatol 1989;93:700–2. [15] Pippenger CE, Browne RW, Armstrong D. Regulatory antioxidant enzymes. Methods Mol Biol 1998;108:299–313. [16] Janaszewska A, Bartosz G. Assay of total antioxidant capacity: comparison of four methods as applied to human blood plasma. Scand J Clin Lab Invest 2002;62:231–6. [17] Berridge MV, Herst PM, Tan AS. Tetrazolium dyes as tools in cell biology: new insights into their cellular reduction. Biotechnol Annu Rev 2005;11:127–52. [18] Van Engeland N, Nieland LJW, Ramaekers FCS, Schutte B, Reutelingsperger CPM. Annexin V-affinity assay. A review on an apoptosis detection system based on phosphtidylserine exposure. Cytometry 1998;31:1–9.

9

[19] Rogakou EP, Nieves-Neira W, Boon C, Pommier Y, Bonner WM. Initiation of DNA fragmentation during apoptosis induces phosphorylation of H2AX histone at serine 139. J Biol Chem 2000;275:9390–5. [20] Gire V, Roux P, Wynford-Thomas D, Brondello JM, Dulic V. DNA damage checkpoint kinase Chk2 triggers replicative senescence. EMBO J 2004;23: 2554–63. [21] Lotan R, Neumann G, Lotan D. Characterization of retinoic acid-induced alterations in the proliferation and differentiation of a murine and a human melanoma cell line in culture. Ann N Y Acad Sci 1981;359:150–70. [22] Nakajima M, Shinoda I, Mikogami T, Iwamoto H, Hashimoto S, Miyauchi H, et al. Beta-lactoglobulin suppresses melanogenesis in cultured human melanocytes. Pigment Cell Res 1997;10:410–3. [23] Fernandes SS, Arcuri R, Morgado-Dı´az JA, Benchimol M. Increase of melanogenesis by retinoic acid: an ultrastructural and morphometric study. Tissue Cell 2004;36:95–105. [24] Inoue Y, Hasegawa S, Yamada T, Date Y, Mizutani H, Nakata S, et al. Bimodal effect of retinoic acid on melanocyte differentiation identified by time-dependent analysis. Pigment Cell Melanoma Res 2012;25:299–311. [25] Hosoi J, Abe E, Suda T, Kuroki T. Regulation of melanin synthesis of B16 mouse melanoma cells by 1 alpha,25-dihydroxyvitamin D3 and retinoic acid. Cancer Res 1985;45:1474–8. [26] Edward M, Gold JA, MacKie RM. Different susceptibilities of melanoma cells to retinoic acid-induced changes in melanotic expression. Biochem Biophys Res Commun 1988;155:773–8. [27] Orlow SJ, Chakraborty AK, Pawelek JM. Retinoic acid is a potent inhibitor of inducible pigmentation in murine and hamster melanoma cell lines. J Invest Dermatol 1990;94:461–4. [28] Kligman AM, Grove GL, Hirose R, Leyden JJ. Topical tretinoin for photoaged skin. J Am Acad Dermatol 1986;15:836–59. [29] Draelos ZD. The combination of 2% 4-hydroxyanisole (mequinol) and 0.01% tretinoin effectively improves the appearance of solar lentigines in ethnic groups. J Cosmet Dermatol 2006;5:239–44. [30] Edwards L, Jaffe P. The effect of topical tretinoin on dysplastic nevi. A preliminary trial. Arch Dermatol 1990;126:494–9. [31] Halpern AC, Schuchter LM, Elder DE, Guerry IV D, Elenitsas R, Trock B, et al. Effects of topical tretinoin on dysplastic nevi. J Clin Oncol 1994;12:1028–35. [32] Stam-Posthuma JJ, Vink J, le Cessie S, Bruijn JA, Bergman W, Pavel S. Effect of topical tretinoin under occlusion on atypical naevi. Melanoma Res 1998;8:539–48. [33] Wong G, Pawelek J. Melanocyte-stimulating hormone promotes activation of pre-existing tyrosinase molecules in Cloudman S91 melanoma cells. Nature 1975;255:644–6. [34] Rome´ro C, Aberdam E, Larnier C, Ortonne JP. Retinoic acid as modulator of UVB-induced melanocyte differentiation. Involvement of the melanogenic enzymes expression. J Cell Sci 1994;107:1095–103. [35] Ando H, Funasaka Y, Oka M, Ohashi A, Furumura M, Matsunaga J, et al. Possible involvement of proteolytic degradation of tyrosinase in the regulatory effect of fatty acids on melanogenesis. J Lipid Res 1999;40:1312–6. [36] Kageyama A, Oka M, Okada T, Nakamura S, Ueyama T, Saito N, et al. Downregulation of melanogenesis by phospholipase D2 through ubiquitin proteasome-mediated degradation of tyrosinase. J Biol Chem 2004;279:27774–80. [37] Nagata H, Takekoshi S, Takeyama R, Homma T, Yoshiyuki Osamura R. Quercetin enhances melanogenesis by increasing the activity and synthesis of tyrosinase in human melanoma cells and in normal human melanocytes. Pigment Cell Res 2004;17:66–73. [38] Petrescu SM, Petrescu AJ, Titu HN, Dwek RA, Platt FM. Inhibition of N-glycan processing in B16 melanoma cells results in inactivation of tyrosinase but does not prevent its transport to the melanosome. J Biol Chem 1997;272:15796–803. [39] Halaban R, Pomerantz SH, Marshall S, Lambert DT, Lerner AB. Regulation of tyrosinase in human melanocytes grown in culture. J Cell Biol 1983;97:480–8. [40] Iozumi K, Hoganson GE, Pennella R, Everett MA, Fuller BB. Role of tyrosinase as the determinant of pigmentation in cultured human melanocytes. J Invest Dermatol 1993;100:806–11. [41] Watabe H, Valencia JC, Yasumot K, Kushimoto T, Ando H, Muller J, et al. Regulation of tyrosinase processing and trafficking by organellar pH and by proteasome activity. J Biol Chem 2004;279:7971–81. [42] Smit NP, van Nieuwpoort FA, Marrot L, Out C, Poorthuis B, van Pelt H, et al. Increased melanogenesis is a risk factor for oxidative DNA damage—study on cultured melanocytes and atypical nevus cells. Photochem Photobiol 2008; 84:550–5. [43] Watabe H, Soma Y, Ito M, Kawa Y, Mizoguchi M. All-trans retinoic acid induces differentiation and apoptosis of murine melanocyte precursors with induction of the microphtalmia-associated transcription factor. J Invest Dermatol 2002;118:35–42. [44] Kishi H, Kuroda E, Mishima HK, Yamashita U. Role of TGF-beta in the retinoic acid-induced inhibition of proliferation and melanin synthesis in chick retinal pigment epithelial cells in vitro. Cell Biol Int 2001;25:1125–9. [45] Duell EA, Anstro¨m A, Griffiths CE, Chambon P, Voorhees JJ. Human skin levels of retinoic acid and cytochrome P-450-derived 4-hydroxyretinoic acid after topical application of retinoic acid in vivo compared to concentrations required to stimulate retinoic acid receptor-mediated transcription in vitro. J Clin Invest 1992;90:1269–74.

Please cite this article in press as: Baldea I, et al. Biphasic pro-melanogenic and pro-apoptotic effects of all-trans-retinoic acid (ATRA) on human melanocytes: Time-course study. J Dermatol Sci (2013), http://dx.doi.org/10.1016/j.jdermsci.2013.06.004