Post-Transcriptional Regulation of Melanin Biosynthetic Enzymes by cAMP and Resveratrol in Human Melanocytes

ORIGINAL ARTICLE

Post-Transcriptional Regulation of Melanin Biosynthetic Enzymes by cAMP and Resveratrol in Human Melanocytes Richard A. Newton1, Anthony L. Cook1, Donald W. Roberts1,2, J. Helen Leonard2 and Richard A. Sturm1 Upregulation of microphthalmia-associated transcription factor (MITF) expression has been proposed to mediate melanogenesis stimulated by cAMP, whereas downregulation of MITF has been suggested to underlie the depigmentary effects of resveratrol, a promising chemotherapeutic found in red wine. We have assessed the contribution of MITF to pigmentation regulation by treating primary cultures of normal human melanocytes with the adenylate cyclase activator forskolin and/or resveratrol, then quantifying mRNA and protein levels for MITF, tyrosinase, tyrosinase-related protein-1, and dopachrome tautomerase (DCT). The inhibition of tyrosinase activity by resveratrol was not due to alterations in MITF, but instead was explained by both direct tyrosinase inhibition and a post-transcriptional effect that reduced the amount of fully processed tyrosinase. Glycosidase digestion revealed that the basis for the tyrosinase decrease was the retention of an immature form in the ER and subsequent loss of the mature, Golgi-processed enzyme. Elevation of intracellular cAMP by forskolin markedly increased protein levels for MITF, tyrosinase and DCT, however there was no concomitant increase in tyrosinase or DCT mRNA. This indicated that elevated levels of MITF were not sufficient to promote transcription of these melanogenic genes and that the increase in their protein abundance appeared to be predominantly mediated through post-transcriptional processing events. Journal of Investigative Dermatology (2007) 127, 2216–2227; doi:10.1038/sj.jid.5700840; published online 26 April 2007

INTRODUCTION Melanin synthesis (melanogenesis) is stimulated in human melanocytes by an increase in the intracellular cAMP concentration, which can be facilitated by activation of the melanocortin-1 receptor or following the addition of exogenous cAMP elevating agents such as forskolin (Busca and Ballotti, 2000). Central to this upregulation is increased activity of tyrosinase, the rate-limiting enzyme for melanogenesis. The molecular basis for this elevated activity has been attributed to enhanced transcription of the tyrosinase gene mediated via the binding of microphthalmia-associated transcription factor (MITF) to promoter E-box motifs (Yasumoto et al., 1997; Bertolotto et al., 1998a). MITF has also been shown to bind and transactivate promoter constructs for 1

Melanogenix Group, Institute for Molecular Bioscience, University of Queensland, Brisbane, Queensland, Australia and 2Queensland Radium Institute Research Unit, Queensland Institute of Medical Research, Brisbane, Queensland, Australia Correspondence: Dr Richard A. Sturm, Institute for Molecular Bioscience, The University of Queensland, Brisbane, Queensland 4072, Australia. E-mail: [email protected] Abbreviations: NHM, normal human melanocytes; TYR, tyrosinase; MITF, microphthalmia-associated transcription factor; ER, endoplasmic reticulum; DCT, dopachrome tautomerase; TYRP1, tyrosinase-related protein-1; Q-RTPCR, quantitative real-time polymerase chain reaction; OCA, oculocutaneous albinism; Endo H, endoglycosidase H; PNGase F, peptide: N-glycosidase F; GAPDH, glyceraldehyde-3-phosphate dehydrogenase. Received 11 September 2006; revised 4 February 2007; accepted 19 February 2007; published online 26 April 2007

2216 Journal of Investigative Dermatology (2007), Volume 127

the tyrosinase-related proteins (TYRPs), TYRP1 and dopachrome tautomerase (DCT) (Aksan and Goding, 1998; Bertolotto et al., 1998b). The expression of these two enzymes is important for later steps in the synthesis of eumelanin pigments. The transcription of MITF itself is enhanced by the activation of a cAMP response element contained within its promoter to provide a mechanism by which cAMP can stimulate melanogenic enzyme expression (Bertolotto et al., 1998a). This cascade therefore affords a simplified scheme that has contributed to MITF being considered a master regulator of human pigmentation. Although MITF-mediated transcriptional activation of pigmentation genes is essential for the control of melanocyte cellular differentiation (Vance and Goding, 2004), the upregulation of endogenous tyrosinase, TYRP1, and DCT gene expression in mature melanocytes does not seem to be satisfactorily explained by a simple direct transcriptional control mechanism. For example, a-MSH-elicited cAMP elevation in human melanocytes can increase protein levels for the three melanogenic enzymes without any accompanying change in mRNA levels, suggesting mediation via posttranscriptional events (Abdel-Malek et al., 1995). Support for the concept that as yet unidentified mechanisms underlie the upregulation of tyrosinase elicited by activation of the cAMP pathway was provided by a more recent study utilizing adenovirus encoding wild-type MITF (Gaggioli et al., 2003). Infection of human melanocytes with the recombinant virus showed that the resultant elevation of MITF levels did not & 2007 The Society for Investigative Dermatology

RA Newton et al. Post-Transcriptional Regulation of Tyrosinase

increase expression of endogenous melanogenic enzymes. However, this same study also provided evidence that MITF expression does still have a crucial involvement in the regulation of human melanocyte pigmentation by demonstrating an inhibitory effect of MITF-dominant-negative recombinant adenovirus on endogenous levels of tyrosinase and TYRP1 proteins (Gaggioli et al., 2003). A role for MITF levels in determining the extent of melanocyte pigmentation is also provided by studies utilizing a variety of depigmenting agents that have demonstrated preceding reductions in MITF expression (Jimenez-Cervantes et al., 2001; Kim et al., 2002a, 2004, 2006; Choi et al., 2005). Resveratrol (3,5,40 -trihydroxy-trans-stilbene), a polyphenol found in red wine, has been proposed to represent another such agent because it has been shown to both decrease skin pigmentation and inhibit MITF promoter activity (Lin et al., 2002). Interest in this compound has shown a marked renewal in recent years, being initially triggered by its identification as a chemotherapeutic agent for skin cancer (Jang et al., 1997) and subsequently from its protective effects against heart disease and ability to extend lifespans of lower organisms (reviewed in Baur and Sinclair, 2006). We have examined the depigmentation effects of resveratrol and assessed whether MITF modulation is an essential step in the negative regulation of melanogenesis by this agent. In addition, we have investigated the contribution of elevated MITF expression to the cAMPmediated upregulation of melanogenic enzymes in primary human melanocytes. This has led to the identification of the important role of post-transcriptional regulation of tyrosinase in governing melanin pigment synthesis. RESULTS Resveratrol elicits a time-dependent decrease in tyrosinase activity in human melanocytes that is unaffected by cotreatment with forskolin

To evaluate the ability of resveratrol to decrease melanin synthesis we assayed the tyrosinase activity of melanocytes, both in situ (Figure 1a) and from extracts of treated cultures (Figure 1b, d). Tyrosinase activity was measured because it is known to closely correlate with melanin content of cultured human melanocytes (Halaban et al., 1983; Wakamatsu et al., 2006) and provides a more sensitive indicator of changes in melanin synthesis than determination of total melanin levels. Treatment of normal human melanocytes (NHM) with resveratrol (20 mg/ml) for 24 hours profoundly decreased in situ tyrosinase activity without affecting cell number (Figure 1a). Although concentrations lower than 20 mg/ml resulted in less inhibition of tyrosinase activity in NHM, a significant decrease in activity was still apparent using only 0.25 mg/ml of resveratrol (data not shown) demonstrating the potency and dose dependency of the effect. The resveratrolmediated decrease in tyrosinase activity, quantitated by both 3,4-dihydroxyphenylalanine (DOPA) oxidation and tritium release from [3H]-tyrosine, was strongly time-dependent, requiring 18–24 hours for the maximal 80% inhibition (Figure 1b). This indicated that an indirect modulation of cellular processes were involved in the profound reduction in tyrosinase activity. There was also a smaller decrease in

tyrosinase activity of approximately 25% evident in lysates within 30 minutes of adding resveratrol to melanocyte cultures (Figure 1b), which indicated a possible direct inhibitory effect of resveratrol. Direct inhibition of tyrosinase by resveratrol was anticipated by virtue of its polyphenol structure and was assessed in vitro by adding resveratrol directly to untreated cell lysates and then measuring tyrosinase activity (Figure 1c). DOPA oxidation activity was decreased by 15% when resveratrol was added to melanocyte lysates at 20 mg/ml (88 mM), a finding in accordance with an in vitro study of the effects of resveratrol added to a preparation of murine tyrosinase where the IC50 for resveratrol-mediated inhibition of tyrosine oxidation was in excess of 100 mM (Kim et al., 2002b). However, we also found that the in vitro effect of resveratrol on tyrosinase activity was considerably more potent when determined under the conditions of the tritium release assay resulting in a 75% decrease in activity (Figure 1c) with an apparent IC50 of approximately 2 mM (data not shown). This apparent discordance between the two tyrosinase assays, which contrasts with the excellent agreement obtained when used for the quantitation of the cell culture effects of resveratrol, suggests that the direct, in vitro inhibitory action of resveratrol on tyrosinase is influenced by the nature and/or concentration of substrate. For example, it is possible that resveratrol may compete more efficiently for L-DOPA binding to tyrosinase in the tritium release assay, where L-DOPA is present at a low concentration as a cofactor, rather than in the DOPA oxidase assay where it is present at a high concentration as a substrate. Inhibition of tyrosinase activity in NHM by resveratrol therefore has two components, a direct inhibition of enzyme activity that is expected to be maximal in the early stages of treatment, and a temporal-dependent, indirect downregulation of enzyme activity. It is well established that treatment of cultured NHM with cAMP elevating agents can increase tyrosinase activity and subsequent melanin synthesis (Hunt et al., 1994a, b; AbdelMalek et al., 1995; Im et al., 1998). In concert with these reports, we found that forskolin treatment of cultured NHM increased their tyrosinase activity (Figure 1d). Combined treatment with resveratrol and forskolin, however, resulted in a decrease in tyrosinase activity similar to that seen with resveratrol alone, whether measured from extracts of treated cells (Figure 1d) or in situ (data not shown). The mechanism underlying resveratrol-mediated inhibition of melanin synthesis was therefore anticipated to be equally evident in cells co-treated with forskolin. Melanogenesis is normally coupled with increased melanocyte dendricity (Busca and Ballotti, 2000) and is exemplified by the ability of forskolin to increase the dendricity of NHM (Figure 1e). Quantitation of melanocyte dendricity for Figure 1e revealed that the effect of forskolin was mainly apparent as an increased number of primary dendrites (2.54 per cell) compared with the control average (2.03 per cell). Secondary dendritic branching was also promoted by forskolin exposure, with the proportion of melanocytes having at least one branching dendrite increased from 3 to 21%. Resveratrol has been reported to promote apoptosis of www.jidonline.org 2217

RA Newton et al. Post-Transcriptional Regulation of Tyrosinase

a

Ctrl

100

100

75

75

50

50

25

25

TYR activity (% control)

c

NHM 3 [ H]Tyrosine assay

NHM DOPA oxidation

0

100

0 Ctrl 0.5 1

2

3

4

6

18 24

Ctrl 0.5 1

2

3

4

6

18 24

75 50 25 0

Ctrl Rsvl

Time (hours)

Time (hours)

d

In vitro effect DOPA assay [3H]Tyrosine TYR activity

b

Rsvl

Dendricity

e

NHM DOPA oxidation

TYR activity (% control)

150 100 Ctrl

Fsk

Rsvl

Rsvl+Fsk

50 0 Ctrl

Rsvl

Fsk

Rsvl +Fsk

Figure 1. Resveratrol and cAMP have opposing effects on tyrosinase activity in NHM. (a) In situ tyrosinase activity determined by incubation of cells in L-DOPA, following 24 hours treatment with either vehicle (Ctrl) or 20 mg/ml resveratrol (Rsvl). Images were captured under identical conditions using bright field microscopy. Bar ¼ 40 mm. (b) Time courses for tyrosinase activity measured by both L-DOPA oxidation and tritium release from [3H]tyrosine using lysates obtained from NHM treated for the indicated periods of time with resveratrol. Error bars indicate the average7range of two melanocyte strains. (c) The direct effect of resveratrol on tyrosinase activity was determined by adding resveratrol directly to NHM lysates at an equivalent concentration to that added to cell cultures. Both the L-DOPA oxidation and tritium release assays were used to quantitate the effect. Error bars represent the mean7SEM of three independent experiments. (d) Tyrosinase activity of NHM lysates following treatment of NHM with resveratrol alone, forskolin (Fsk) alone (10 mM), or both agents. Values are given as the mean7SEM (n ¼ 4) of four different melanocyte strains, normalized for total soluble protein and expressed relative to the activity of vehicle-treated cells, as detailed in materials and methods. (e) Changes in dendritic morphology following the same treatment regime described in (d) were assessed by phasecontrast microscopy, with dendrite length quantified using ImageJ software. Bar ¼ 40 mm. All of the above results are representative of at least two independent experiments using at least two melanocyte strains.

human melanoma cells (Niles et al., 2003); however, survival of primary melanocytes was not compromised by resveratrol and surprisingly, treatment with resveratrol resulted in a positive effect on melanocyte dendricity that was mainly evident as an enhancement of dendrite length (Figure 1e). Thus, the total dendrite length per cell was 1.7-fold higher than in control cells despite a similar number of primary dendrites (2.1 per cell). Further, 24-hour treatment with the melanogenesis inhibiting combination of forskolin plus 2218 Journal of Investigative Dermatology (2007), Volume 127

resveratrol elicited a synergistic increase in dendricity (Figure 1e). In particular, the percentage of melanocytes showing secondary dendritic branching (63%) was markedly higher than that seen with either forskolin (21%) or resveratrol (15%) alone, and contributed to the dramatic increase in total dendrite length following combination treatment (2.5-fold greater than control). Therefore, resveratrol has the unusual property of being both a positive regulator of melanocyte dendricity and a negative regulator of melanogenesis.

RA Newton et al. Post-Transcriptional Regulation of Tyrosinase

1 hour

a

3 hours

12 hours – +

Rsvl –

–

+

+

– +

+

Fsk –

+

+

–

+

– +

+

+

24 hours

+ –

+

+

–

– +

+

–

–

P-CREB P-ERK1/2 ERK1/2 MITF

MITF mRNA (% control)

b

400

Ctrl Fsk Rsvl+Fsk Rsvl

300 200 100 0 3 hours

12 hours

24 hours

Figure 2. Effects of forskolin and resveratrol on MITF levels in NHM. (a) Western-blot analysis of whole-cell lysates prepared from NHM treated for the indicated times with resveratrol alone, forskolin alone, or both agents. Phospho-specific antibodies were used to detect phosphorylated CREB and ERK (p44/p42 MAPK) because these activated forms are involved in the transcriptional upregulation and phosphorylation of MITF, respectively. The monoclonal MITF antibody identified the protein as a doublet, with the upper band previously shown to correspond to MITF that has undergone MAPK phosphorylation (Hemesath et al., 1998). Similar protein levels were verified using a phosphorylation-independent ERK antibody. (b) Q-RT-PCR analysis of MITF RNA levels following the same treatment regime applied for protein analysis. Transcript abundance was normalized against 18S ribosomal RNA and expressed relative to the 3 hours control. Each experiment utilized two melanocyte strains and are presented as the mean7SEM (n ¼ 4) of two independent experiments.

MITF levels are increased by forskolin but are unaltered by resveratrol

To ascertain whether decreased MITF expression could underlie the depigmentation effects of resveratrol, the levels of MITF mRNA and protein were determined over the time frame that inhibition of tyrosinase activity occurred (Figure 2). In addition, the phosphorylation status of cAMP-responsive element-binding protein (CREB) and extracellular-regulated kinase (ERK) were also measured (Figure 2a) because these activated forms can modulate the levels of MITF in melanocytes via increased transcription or decreased stability, respectively (Vance and Goding, 2004). Forskolin treatment of NHM induced a rapid increase in phosphorylated CREB that was temporally associated with increased transcription of MITF (Figure 2b), in agreement with the prevailing concept that activated CREB confers cAMP responsiveness to the MITF promoter (Price et al., 1998; Busca and Ballotti, 2000). This increase in RNA was reflected

in a corresponding increase in MITF protein, evident as a doublet with the higher molecular weight band corresponding to an ERK phosphorylated form (Hemesath et al., 1998). Treatment of NHM with resveratrol resulted in only a 25% decrease in MITF transcript levels by 24 hours (Figure 2b), but with no discernable affect on protein abundance (Figure 2a) and importantly did not reduce the elevated MITF levels elicited by forskolin. Furthermore, there was no apparent effect of resveratrol on the ERK-mediated phosphorylation status of MITF that could potentially alter stability and/or DNA binding. This indicated that reduced MITF function does not explain the inhibition of tyrosinase activity by resveratrol. Tyrosinase protein, but not mRNA, is differentially regulated by forskolin and resveratrol

As MITF is widely regarded as the key regulator of tyrosinase transcription, we determined the extent to which the levels of MITF following forskolin and/or resveratrol treatment were reflected in changes in the steady-state expression of tyrosinase mRNA (Figure 3a). Resveratrol did not significantly alter tyrosinase transcript levels in NHM at any of the time points examined, in agreement with the lack of effect of this agent on MITF expression (Figure 2a). Interestingly, there was no significant alteration in tyrosinase RNA expression following either the forskolin or forskolin plus resveratrol treatment regimes (Figure 3a), despite the fact that MITF expression is increased up to 3.5-fold in the presence of forskolin (Figure 2b). Tyrosinase protein was then examined to see if post-transcriptional changes could potentially account for the observed alterations in tyrosinase activity. Western analysis revealed that human tyrosinase appeared mostly as an B80 kDa species (Figure 3b), the size of the mature, fully Golgi-processed form (Halaban et al., 1997). Exposure of NHM to resveratrol led to a time-dependent decrease in abundance of mature tyrosinase and, crucially, this was also seen following co-treatment with forskolin (Figure 3b, c). Therefore, a post-transcriptionally mediated decrease in mature tyrosinase protein can account for resveratrol-elicited reductions in tyrosinase activity. Forskolin treatment alone markedly increased the level of mature tyrosinase after 24 hours indicating that a post-translation regulatory mechanism could potentially explain the elevation in tyrosinase activity produced by this agent. However, the extent of this increase in fully processed tyrosinase only seemed to be partially reflected as elevated tyrosinase activity (Figure 1c). This may reflect the fact that alterations in the catalytic activity of tyrosinase cannot be explained simply by changes in tyrosinase abundance (Iozumi et al., 1993) because mechanisms regulating the activity of tyrosinase within the melanosome, the site of melanin synthesis, are also crucial (Fuller et al., 2001). Retention of immature tyrosinase in the endoplasmic reticulum by resveratrol

Examination of the relative abundance of the different forms of tyrosinase detected by Western analysis indicated potential effects of resveratrol on the trafficking of tyrosinase. Thus, in www.jidonline.org 2219

RA Newton et al. Post-Transcriptional Regulation of Tyrosinase

a

Endo H

Ctrl Fsk Rsvl+Fsk Rsvl

125

Rsvl –

–

+

+

–

–

+

+

–

–

+

+

Fsk –

+

+

–

–

+

+

–

–

+

+

–

TYR mRNA (% control)

100 TYR

75

GAPDH

3 hours 1 hour

12 hours

3 hours

24 hours

12 hours

24 hours

Rsvl – – + + – –

+ +

– – + + – – + +

Fsk – + + – – +

+ –

– + + – – + + – kDa 100 80 70 60 55 40 35

TYR GAPDH

Ctrl Fsk Rsvl+Fsk Rsvl

TYR protein (% control)

200 150 100 50 0

40 35

0

c

100 70 55

50 25

b

PNGase F

Ctrl

1 3 12 24

1 3 12 24

1 3 12 24

hours

hours

hours

Figure 3. Effects of forskolin and resveratrol on tyrosinase levels in NHM. (a) Q-RT-PCR analysis of tyrosinase RNA levels following treatment of NHM with resveratrol alone, forskolin alone, or both agents, for the indicated times. Error bars indicate the mean7SEM (n ¼ 4), relative to the 3 hours control, and were obtained from two melanocyte strains from two independent experiments. (b) Western-blot analysis of tyrosinase protein levels in wholecell lysates prepared from NHM treated as described above. Similar protein levels were verified using an antibody against GAPDH. Samples were run on two gels (indicated by gap) and blots processed in parallel under identical conditions. Vertical lines indicate that single lanes either side of the forskolin lanes were removed during preparation of the Figure. (c) Fully processed tyrosinase, represented by the uppermost band at approximately 80 kDa, was quantified by densitometry of Western-blot autoradiographs. Error bars indicate the mean7SEM (n ¼ 4), relative to the untreated controls, and were obtained using three melanocyte strains from two independent experiments.

addition to the dramatic reduction in fully processed 80 kDa species, evident at 12 and 24 hours following resveratrol, there also appeared to be a small increase in intensity for a broad band detected at approximately 70 kDa at these time points (Figure 3b). This species has previously been shown to correspond to an immature, endoglycosidase H (Endo H)sensitive glycoform of tyrosinase (Halaban et al., 1997) and indicated that resveratrol may cause aberrant processing of tyrosinase. Furthermore, it was also apparent that both 12and 24-hours treatment with resveratrol resulted in the loss of a tyrosinase immunoreactive band at around 60 kDa (Figure 3b), a size expected for the deglycosylated, core polypeptide. 2220 Journal of Investigative Dermatology (2007), Volume 127

Figure 4. Endo H sensitivity of tyrosinase from resveratrol and forskolin treated NHM. Cell extracts obtained following 24 hours treatment with either resveratrol, forskolin, or both agents were subsequently incubated in the presence or absence of the glycosidases, Endo H and PNGase F. Western-blot analysis for tyrosinase was then used to monitor the extent of digestion. Arrows at approximately 80 and 60 kDa indicate bands corresponding to the mature and PNGase F digested forms of tyrosinase, respectively. An Endo H-sensitive band is coincident with the 70 kDa marker. Samples were run on two gels (indicated by gap) and blots processed in parallel under identical conditions. Vertical lines indicate removal of two intervening lanes. Identical results were obtained in a second independent experiment.

The treatment of NHM with proteasomal inhibitors has previously been shown to induce the accumulation 60 kDa tyrosinase protein, indicating that a portion of recently synthesized tyrosinase is targeted for proteolytic degradation (Halaban et al., 1997). The loss of a similarly sized tyrosinase species following resveratrol treatment may therefore represent accelerated transfer of newly synthesized tyrosinase to the proteasomal complex. The 60 kDa band was not seen in the control extracts of Figure 4, which may be a consequence of the lower protein concentration present in these lysates. Glycosidase digestion was utilized to further evaluate aberrant trafficking of tyrosinase following treatment with resveratrol. Digestion with Endo H cleaves oligosaccharides added in the endoplasmic reticulum (ER) but not those matured in the Golgi, whereas peptide: N-Glycosidase F (PNGase F) cleaves both immature and mature oligosaccharides. Inspection of Figure 4 reveals that the majority of tyrosinase protein extracted from resveratrol-treated NHM was Endo H-sensitive and hence was ER-retained protein. This reflects the relative proportions of the 70 and 80 kDa forms that correspond to Endo H-sensitive and -resistant populations, respectively. In contrast, tyrosinase from NHM treated with either vehicle of forskolin alone had mainly the Golgi processed, Endo H-resistant form, consistent with the high proportion of mature tyrosinase detected by Western analysis. Digestion with PNGase F generated a 60 kDa band representing deglycosylated tyrosinase and demonstrated that the total amount of tyrosinase was increased by forskolin whereas it appeared to be reduced by resveratrol (Figure 4). DCT protein is also differentially regulated by forskolin and resveratrol

Although changes in the abundance of tyrosinase can account for the observed changes in tyrosinase activity seen with these agents, it was also necessary to evaluate whether there were also alterations in the two TYRPs, TYRP1 and

RA Newton et al. Post-Transcriptional Regulation of Tyrosinase

a 200

DCT mRNA (% control)

b

Ctrl Fsk Rsvl+Fsk Rsvl

1 hour

100

12 hours

24 hours

DCT

50

GAPDH 3 hours

c

12 hours

24 hours

d

Ctrl Fsk Rsvl+Fsk Rsvl

125

TYRP1 mRNA (% control)

3 hours

Rsvl – – + + – + + – + + – + + – Fsk – + + – + + – + + – + + – –

150

0

TYRP1 protein 1 hour

100

3 hours

12 hours

24 hours

Rsvl – – + + – + + – + + – + + – Fsk – + + – + + – + + – + + – –

75 50

TYRP1

25

GAPDH

0 3 hours

e mRNA (% control)

DCT protein

12 hours

24 hours

SLC45A2

OCA2

250

250

250

200

200

200

150

150

150

100

100

100

50

50

50

0

0 3 hours

12 hours

24 hours

SLC24A5 Ctrl Fsk Rsvl+Fsk Rsvl

0 3 hours

12 hours

24 hours

3 hours

12 hours

24 hours

Figure 5. DCT, TYRP1, SLC45A2, OCA2, and SLC24A5 expression in NHM following treatment with resveratrol, forskolin, or both agents, for the indicated times. Q-RT-PCR and Western immunoblot analysis was used to determine the levels of mRNA and protein, respectively, for DCT (a, b) and TYRP1 (c, d). (e) Q-RT-PCR was also used to quantify expression of the genes SLC45A2, OCA2, and SLC24A5 that encode the proteins MATP, P-protein, and NCKX5, respectively. Transcript abundance was normalized against 18S ribosomal RNA and expressed relative to the 3 hours control. Error bars indicate the mean7SEM (n ¼ 4) of two independent experiments, each utilizing two melanocyte strains.

DCT. These two enzymes are important for the stimulation of eumelanin production, which is likely to be either due to a direct catalytic role and/or an ability to stabilize tyrosinase activity (for review, see Slominski et al., 2004). As was seen for tyrosinase, the transcript levels for both DCT (Figure 5a) and TYRP1 (Figure 5c) showed little correspondence with changes in MITF expression (Figure 2), again implicating the involvement of additional regulatory factors and/or posttranscriptional control. DCT showed extensive differential regulation following resveratrol and/or forskolin treatment (Figure 5a, b). This was particularly apparent at the protein level, 12 and 24 hours after treatment, with forskolin stimulating a profound increase in DCT, and resveratrol markedly decreasing protein abundance (Figure 5b). Whereas the decrease in DCT protein could at least partially be accounted for by reduced levels of mRNA, altered transcription did not seem to be a potential explanation for the forskolin-stimulated increase in protein expression. This again suggested that post-transcriptional regulation might be a crucial mediator of cAMP-induced increases in melanogenic enzymes. Analysis of TYRP1 mRNA levels in NHM revealed that resveratrol treatment caused an approximate 50% decrease in

transcript abundance after 24 hours; however, only a 25% reduction was seen when forskolin was also present (Figure 5c). Treatment with forskolin alone resulted in a small increase in TYRP1 mRNA at 12 and 24 hours that approached 25% above control levels. These changes in RNA expression were reflected by the modest increase in the protein level at 24 hours, as determined by Western analysis (Figure 5d). Protein levels of the melanosomal protein MART-1 were also examined in two melanocyte strains following the same treatment regime and were not altered by resveratrol (data not shown) providing further evidence that the effects of resveratrol were not due to a generalized effect on melanosomal protein stability. siRNA-mediated knockdown of MITF demonstrates that MITF expression is required for the maintenance of melanogenic enzyme transcription

The preceding data indicated that elevation of MITF expression does not necessarily directly regulate melanogenic enzyme levels in cultured NHM. However, MITF may still make an important contribution to the maintenance of melanogenic enzyme transcription. We therefore utilized a siRNA against MITF to evaluate the requirement of this www.jidonline.org 2221

RA Newton et al. Post-Transcriptional Regulation of Tyrosinase

TYRP1

GAPDH

siRNA 1.0

1.0

1.0

0.8

0.8

0.8

0.6

0.6

0.6

0.6

0.4

0.4

0.4

0.4

0.2

0.2

0.2

0.2

0.0

0.0

0.0

1.0

Negative control MITF

0.8

0.0

Negative control MITF

+

–

–

+

c

siRNA

b

TYR

MITF

siRNA

Normalized mRNA level

a

Negative control + – MITF

d

TYR activity

–

siRNA Negative control

+

MITF

MITF

95

TYR 54

TYR 54

95 95

TYRP1

TYRP1

54 54 GAPDH

GAPDH 1

2

1

2

Relative tyrosinase activity

95 1.0 0.8 0.6 0.4 0.2 0.0

Figure 6. Effect of MITF inhibition on melanogenic enzyme levels in NHM. The cells were transfected with siRNAs that were either directed against MITF or consisted of a commercial negative control sequence. 48 hours later, cells were assessed for expression of MITF, tyrosinase, TYRP1, and GAPDH by (a) Q-RT-PCR measurement of mRNA, and (b) Western-blot analysis of protein. (c) Cellular lysates containing tyrosinase were also prepared from identically treated cells then digested with PNGase F to facilitate comparison of core deglycosylated tyrosinase protein levels by Western analysis. (d) Cellular lysates were also assessed for relative tyrosinase activity using the DOPA oxidation assay. Error bars represent the mean7SEM (n ¼ 3) of three independent experiments.

transcription factor for the expression of tyrosinase and TYRP1 in NHM. The effectiveness of the MITF siRNA in NHM was shown by MITF mRNA levels being dramatically reduced to 25% of control levels (Figure 6a), concomitant with a substantial decrease in MITF protein (Figure 6b). Tyrosinase and TYRP1 mRNA expression levels in NHM following siRNA-mediated knockdown of MITF was reduced to approximately 50 and 40% of that seen in negative controls, respectively (Figure 6a). Similarly, mRNA levels for SLC24A5, SLC45A2, and OCA2 were also decreased in NHM transfected with siRNA molecules targeting MITF (data not shown). These decreases were substantially greater than that seen for glyceraldehyde-3-phosphate dehydrogenase (GAPDH) mRNA (Figure 6a). Tyrosinase and TYRP1 mRNA expression level changes were at least partially replicated at the protein level (Figure 6b, c), with TYRP1 protein obviously decreased. For tyrosinase, the decrease in protein appeared to be mainly confined to the lower molecular weight bands (Figure 6b), with decreased protein levels only readily apparent following PNGase F-mediated digestion to the deglycosylated core tyrosinase polypeptide (Figure 6c). However, little difference in tyrosinase activity was detected using the DOPA oxidation assay (Figure 6d), consistent with this catalytic measure reflecting changes in the fully processed form of tyrosinase 2222 Journal of Investigative Dermatology (2007), Volume 127

that showed only minor variation (Figure 6b). Therefore, some level of MITF expression appears to be required for the maintenance of melanogenic enzyme transcription in NHM. DISCUSSION A number of chemical agents can lead to depigmentation of human skin, an effect that is commonly due to phenol-based structures causing direct inhibition of tyrosinase enzymatic activity (Ando et al., 2007; Solano et al., 2006). An additional, indirect depigmenting mechanism invokes a key regulatory role for MITF in governing levels of tyrosinase, with compounds such as sphingosine-1-phosphate and sphingosylphosphorylcholine shown to decrease tyrosinase abundance via a preceding reduction in MITF (Kim et al., 2003; Kim et al., 2006). Such a mechanism was also proposed to underlie the inhibition of melanogenesis by resveratrol (Lin et al., 2002). This suggestion was largely based on the findings that resveratrol was able to decrease the activity of an MITF promoter-luciferase reporter construct and reduce endogenous MITF mRNA levels in NHM. We also found that resveratrol treatment was able to decrease MITF mRNA in NHM, however, this reduction was only by 25% and was not accompanied by any alteration in MITF protein abundance or tyrosinase transcription. Furthermore, we found that resveratrol, in the presence of forskolin, actually

RA Newton et al. Post-Transcriptional Regulation of Tyrosinase

elevated levels of MITF mRNA and protein, despite the fact that the inhibition of tyrosinase activity by this treatment combination is virtually identical to that seen with resveratrol alone. Therefore downregulation of endogenous MITF expression did not account for resveratrol-mediated depigmentation and a new mechanistic explanation was required. Our data demonstrated that at least two mechanisms underlie the inhibition of tyrosinase activity following resveratrol treatment of NHM. The first mechanism caused a reduction of tyrosinase protein levels that was shown to correlate with the major decrease in tyrosinase activity seen after 18–24 hours. Secondly, resveratrol had the capacity to directly inhibit tyrosinase activity when added to cell extracts, an effect that has been previously described (Kim et al., 2002b). Although the decrease in tyrosinase protein can account for the time-dependent reduction in tyrosinase activity measured in lysates from treated NHM, it is unclear to what extent the direct inhibition mechanism contributes to reduced tyrosinase activity in situ. The influence of direct inhibition, however, would be expected to decline with time because studies on resveratrol uptake in cultured human hepatic cells demonstrated that intracellular levels peak within 10 minutes then gradually decrease (Lancon et al., 2004). The high-affinity interaction between resveratrol and tyrosinase revealed by the in vitro studies may also be relevant for the molecular mechanism mediating long-term downregulation of tyrosinase protein. Resveratrol treatment of NHM substantially decreased the amount of tyrosinase protein without affecting steady-state levels of tyrosinase RNA confirming that regulation of tyrosinase transcription does not mediate depigmentation by resveratrol. The basis for the decrease in tyrosinase protein was revealed by Western analysis and glycosidase digestion, with extracts from resveratrol treated NHM shown to consist mostly of ER-retained, immature tyrosinase, whereas control NHM contained mainly the mature, Golgi-processed form. Therefore resveratrol treatment disrupted the trafficking of tyrosinase from the ER to the Golgi leading to a dramatic decrease of functional tyrosinase. Interestingly, retention of tyrosinase in the ER can also occur for mutant forms of tyrosinase that cause type I oculocutaneous albinism (Halaban et al., 2000). Furthermore, the loss of pigmentation that is often seen in human melanomas has also been shown to be the result of aberrant ER retention, with tyrosinase again mainly evident as a 70 kDa Endo H-sensitive, immature form (Halaban et al., 1997). This ER-trapped form is also subject to increased proteolytic degradation (Watabe et al., 2004) accounting for the overall decrease in tyrosinase protein evident following resveratrol exposure. Depigmentation due to ER-retained tyrosinase has recently also been demonstrated in melanoma cells that have been engineered to display enhanced sensitivity to oxidative stress via downregulation of the L-ferritin gene (Maresca et al., 2006). Resveratrol, however, is a free radical scavenger and thus would be expected to decrease oxidative stress. This suggests that the antioxidant properties of resveratrol are unlikely to account for its effects on tyrosinase maturation. Intriguingly, there also appears to be another polyphenol

compound with considerable structural similarity to resveratrol, 2,20 -dihydroxy-5,50 -dipropyl-biphenyl, that appears to downregulate melanogenesis in mouse B16 melanoma cells in a similar manner (Nakamura et al., 2003). Thus, 2,20 dihydroxy-5,50 -dipropyl-biphenyl treatment for 24 hours strongly suppressed tyrosinase activity, an effect that was also explained via ER-retention of tyrosinase and a subsequent acceleration of tyrosinase degradation (Nakamura et al., 2003). Although 2,20 -dihydroxy-5,50 -dipropyl-biphenyl did not show any direct inhibition of tyrosinase, it may share with resveratrol a structural ability to interact with tyrosinase to hinder correct folding/processing of the protein. Misfolding of mutant tyrosinase from albino melanocytes underlies the subsequent increase in ER retention and degradation of tyrosinase (Halaban et al., 2000). The importance of folding in tyrosinase maturation has also been demonstrated in amelanotic melanomas by the ability of the tyrosinase substrates, DOPA and tyrosine, to facilitate the correct folding of the enzyme to the mature, proteolytically resistant form (Halaban et al., 2001). Also crucial for tyrosinase trafficking is the pH within the ER-Golgi compartments, with amelanotic melanoma cells shown to have enhanced tyrosinase maturation and to pigment following treatment with agents that increase the luminal pH of subcellular organelles (Halaban et al., 2002). In this regard, it of interest that resveratrol was able to decrease expression of two genes, SLC45A2 and SLC24A5, which encode transmembrane proteins implicated in the regulation of ion transport (for review, see Sturm, 2006). Therefore, resveratrol, either directly or indirectly, may alter the processing environment with concomitant ER retention of tyrosinase. In contrast to tyrosinase, the profound decrease in DCT protein elicited by resveratrol was also accompanied by a corresponding reduction in mRNA, implicating transcription control. The decrease in DCT following resveratrol treatment could contribute to skin depigmentation and may also have relevance for potential chemotherapeutic applications. Although we did not observe an effect of resveratrol on the proliferation of primary melanocytes, it has been shown to induce apoptosis in melanoma cells (Niles et al., 2003) and increase sensitivity to anticancer therapy (Fulda and Debatin, 2004; Yang et al., 2005). The chemotherapy sensitization effects towards melanoma may relate, in part, to decreased DCT levels, because expression of DCT has been associated with resistance to chemo- and radiotherapy (Chu et al., 2000; Pak et al., 2004). Stimulation of pigmentation following activation of the cAMP signal transduction cascade is at least partially attributable to an upregulation of melanogenic enzymes and studies with gene reporter assays have implicated a key role for MITF in such induction via transcriptional control of melanogenic genes (Bertolotto et al., 1998a, b; Price et al., 1998; Busca and Ballotti, 2000). However, the forskolinelicited upregulation of tyrosinase and DCT proteins in NHM described in this study was not associated with an increase in tyrosinase or DCT transcription, despite a robust increase in MITF expression. Previous evidence for post-transcriptional control of melanogenic enzyme levels in NHM was provided www.jidonline.org 2223

RA Newton et al. Post-Transcriptional Regulation of Tyrosinase

in a report where a-MSH stimulated an increase in tyrosinase, TYRP1, and DCT protein without altering respective mRNA levels, as detected by Northern analysis (Abdel-Malek et al., 1995). The inability of elevated MITF protein to stimulate tyrosinase and DCT transcription in this investigation may be a consequence of the endogenous context of the tyrosinase promoter, a concept supported by a study using adenovirus encoding wild-type MITF which demonstrated that exogenous MITF was not sufficient to increase levels of endogenous tyrosinase and TYRP1 protein in NHM (Gaggioli et al., 2003). This latter study also utilized adenovirus expressing dominant negative MITF to demonstrate that MITF was, however, still required for maintaining endogenous tyrosinase gene expression. Our study employed siRNA-mediated knockdown of MITF to similarly examine the relationship between MITF and constitutive enzyme levels, and we also observed a significant decrease in tyrosinase and TYRP1 expression in NHM following a 75% reduction in MITF mRNA levels. Our data therefore support the concept that ‘MITF is required, but is not sufficient to induce the expression of melanogenic genes’ (Gaggioli et al., 2003). An alternative mechanistic basis for the observed posttranscriptional upregulation of melanogenic proteins by cAMP is therefore required. Insight into the molecular scheme responsible for the increase may be provided by the observation that the two melanogenic proteins with wellcharacterized enzyme activity, tyrosinase, and DCT, were strongly increased by forskolin, in contrast to TYRP1, a protein that may not have an enzymatic role (Boissy et al., 1998). An attractive possible mechanism is that the cAMP signaling pathway can influence the enzyme environment to modulate activity and subsequent processing and/or trafficking. We therefore investigated this possibility by quantitating the expression of the genes SLC45A2, OCA2, and SLC24A5 that encode transmembrane proteins implicated in the regulation of ion transport and/or the processing and trafficking of melanogenic enzymes (Sturm, 2006). The most dramatic effect of forskolin was a twofold increase in expression of the SLC45A2 gene, encoding the MATP protein, which is likely to reflect the ability of SLC45A2 to be transcriptionally modulated by MITF (Du and Fisher, 2002), (our data, not shown). As MATP plays a crucial role in tyrosinase trafficking and stability (Costin et al., 2003), the upregulation of this protein provides a promising candidate to facilitate the increased tyrosinase abundance. However, it is difficult to reconcile the greater upregulation of DCT than TYRP1 by such a MATP-mediated mechanism because DCT trafficking is the least affected melanogenic enzyme in mice with defective MATP function (Costin et al., 2003). Forskolin also induced a 1.5-fold increase in expression of the OCA2 gene encoding the P-protein. The P-protein has been shown to be crucial for normal post-translational processing and trafficking of tyrosinase in the ER (Chen et al., 2002). However, it has also been reported that melanocytes cultured from p-null mice can still elevate tyrosinase levels in response to IBMX, a cAMP phosphodiesterase inhibitor, although it was not established whether this was a transcriptional or post-translational effect (Chen et al., 2004). 2224 Journal of Investigative Dermatology (2007), Volume 127

Post-transcriptional regulation of melanogenic enzyme expression has previously been primarily associated with the inhibition of melanin synthesis, as evident in amelanotic melanoma and classic oculocutaneous albinism. The loss of tyrosinase activity and ER retention of misfolded tyrosinase that underlies such loss of pigmentation has now also been shown to be the main mechanism responsible for the profound loss of pigmentation elicited by the chemopreventative agent resveratrol. Our data also indicated that posttranscriptional upregulation of melanogenic enzyme levels represents an important regulatory control point for melanin synthesis following activation of the cAMP pathway in NHM. As upregulation of tyrosinase and DCT is an important component of the photoprotective delayed tanning response, further understanding of the molecular details of this posttranscriptional control may have significant therapeutic implications. MATERIALS AND METHODS Materials Antibodies against the following human proteins were used: tyrosinase (monoclonal T311, Upstate, Charlottesville, VA), DCT (TRP2, D-18, Santa Cruz Biotechnology, Santa Cruz, CA), TYRP1 (monoclonal B8G3 (Takahashi and Parsons, 1990), a gift from P. Parsons), MITF (Ab-2, monoclonal D5) and MART-1 (NeoMarkers, Melbourne, Australia), phospho-CREB (Upstate), phospho-ERK (phospho-p44/42 MAPK, monoclonal E10, Cell Signaling Technology, Beverly, MA), ERK (p44/42 MAPK, Cell Signaling Technology), and GAPDH (Genzyme, Cambridge, MA). Horseradish peroxidaseconjugated secondary antibodies against rabbit or mouse IgG were purchased from Upstate. The horseradish peroxidase-conjugated anti-goat secondary antibody, forskolin, resveratrol, and activated charcoal were all obtained from Sigma-Aldrich (Sydney, NSW, Australia). L-Tyrosine-3,5-3H (33.9 Ci/mmol) was purchased from Perkin Elmer (Melbourne, VIC, Australia).

Cell culture Queensland foreskin primary melanocyte strains were isolated and propagated from anonymous neonatal foreskin samples as described previously (Leonard et al., 2003) and maintained in RPMI-1640 media containing 3 mM HEPES buffer supplemented with 8% heatinactivated fetal bovine serum, penicillin/streptomycin (40 mg/ml), 1 mM pyruvic acid, 0.2 mM. nicotinamide, 10 ng/ml TPA, 5 ng/ml bFGF, and 0.6 mg/ml cholera toxin. One week before, and for the duration of, experimentation, cholera toxin was removed from the melanocyte growth media to facilitate investigation of cAMPmediated responses. Stimulation of NHM was carried out in culture media that had been present for 3 days since seeding. The melanocyte strains used were all from Caucasian donors with a darkly pigmented and lightly pigmented strain used for all experiments. A further melanocyte strain of intermediate pigmentation was also used for tyrosinase, DCT, and TYRP1 Western immunoblot analysis. The study was conducted according to the Declaration of Helsinki Principles, and ascertainment of foreskin tissue samples was approved by the University of Queensland Medical Research Ethics Committee and QIMR Human Research Ethics Committee. Experiments were conducted with parental consent to the use of tissue for research.

RA Newton et al. Post-Transcriptional Regulation of Tyrosinase

Tyrosinase activity assays and L-DOPA staining To prepare cell extracts following drug treatment, cells were washed in ice-cold phosphate-buffered saline (PBS), lysed in 0.1 M phosphate buffer (pH 6.8) containing 1% Triton X-100 and complete protease inhibitor cocktail (Roche Diagnostics, Sydney, NSW, Australia) for 20 minutes at 41C, then clarified by centrifugation (13,000  g, 5 minutes). The tyrosinase activity of 100 ml aliquots of extract was then determined by either measuring the rate of oxidation of L-DOPA (Takahashi and Parsons, 1992) or the release of tritium from tyrosine-3,5-3H (Pomerantz, 1966). For the DOPA oxidation assay, 100 ml duplicates were incubated with 100 ml L-DOPA (3 mg/ml in 0.1 M phosphate buffer, pH 6.8) at 371C. The absorbance at 490 nm was measured every 30 minutes for at least 2 hours and the tyrosinase activity calculated as OD490/min/mg of lysate protein then expressed relative to the activity of vehicletreated cells. To assess direct inhibitory effects on tyrosinase activity, compounds were added to cell lysates at an equivalent concentration to that used for treating whole-cell cultures, incubated for 5 minutes at room temperature, mixed with L-DOPA solution and incubated for 2 hours at 371C, as described above. The protein concentration of extracts was determined by a BCA assay (Pierce, Rockford, IL, USA). The tritium release assay of Pomerantz (Pomerantz, 1966) was performed essentially as first described. Briefly, 100 ml of cell extract was added to a 900 ml reaction mixture containing 0.1 M phosphate buffer (pH 6.8) and 0.1 mM phenylmethanesulfonylfluoride. For experiments examining the in vitro effects of resveratrol, the drug or DMSO was also added at this stage to give the desired final concentration and incubated at room temperature for 5 minutes. L-DOPA (0.03 mM final concentration) and 1 mCi tyrosine-3,5-3H were then added and the reaction incubated for 1 hour at 371C. The reaction was stopped by adding 50 ml 1 M HCl then mixed with approximately 100 mg activated charcoal for 1 hour at room temperature, centrifuged briefly and 200 ml supernatant mixed with 1.8 ml scintillation fluid, and counted. Background counts were assessed using 100 ml 0.1 M phosphate buffer in place of cell extract. In situ L-DOPA reactivity of NHM was assessed using cultures fixed with 4% paraformaldehyde in PBS for 40 min at room temperature. Following permeabilization in 0.1% triton X-100 in PBS for 2 min, cells were washed with PBS and incubated in 0.1% L-DOPA for 3 hours at 371C. Cells were then rinsed in PBS and images captured using an Olympus CKX41 inverted microscope equipped with an Olympus Camedia C-5060 digital camera.

siRNA transfection A predesigned siRNA directed against MITF (Ambion, Austin, TX, ID No. 115142) was used to downregulate MITF, and a commercial negative control sequence (Ambion, Catalogue No. 4611) was used to monitor for off-target effects. NHMs were seeded in complete growth medium in 10 cm dishes (5  105 cells) or 6-well plates (2  105 cells per well) and allowed to adhere for 48 hours. Transfection of siRNAs was carried out essentially as described previously (Cook et al., 2005). Briefly, siRNA molecules were transfected at a concentration of 100 pmol using 5 ml Lipofectamine 2000 (Invitrogen, Melbourne, VIC, Australia) per 1 ml consisting of 50% OptiMEM and 50% growth medium without antibiotics. After 6 hours, an equal volume of growth medium lacking antibiotics was added and the cells cultured overnight. Twenty-four hours post-

transfection, the liposomes were aspirated and the medium replaced with growth medium lacking antibiotics and the culture continued for a further 24 hours before preparation of total protein lysates, DOPA oxidation assay extracts, or total RNA extracts as appropriate.

Gene expression quantitation RNA extraction, DNase treatment and quantitative real-time polymerase chain reaction (Q-RT-PCR) procedures were carried out essentially as previously described (Newton et al., 2005; Roberts et al., 2006). Ten percent of cDNA synthesis reactions from 0.3 mg total RNA were used in 25 ml Q-RT-PCR reactions containing TaqMan gene expression assays (Applied Biosystems, Foster City, CA, USA) for either tyrosinase (Hs00165976_m1), DCT (Hs00157244_m1), TYRP1 (Hs00167051_m1), MITF (Hs00165 156_m1), GAPDH (Hs99999905_m1), OCA2 (Hs00609341_m1), SLC45A2 (Hs00211813_m1), or SLC24A5 (Hs01385408_m1), together with TaqMan PCR master mix. Transcript levels were normalized against 18S ribosomal RNA that was quantified using a Pre-Developed TaqMan assay for 18S (Applied Biosystems) and 50-fold diluted cDNA. Q-RT-PCR was performed in 96-well plates on an ABI Prism 7500 Sequence Detector system using a two-step thermal cycling protocol of 40 cycles of 951C for 15 s and 601C for 1 minute.

Western immunoblot analysis Following appropriate periods of treatment at 371C, culture plates were placed on ice, rinsed with ice-cold PBS, SDS extraction buffer added and Western blotting performed as described previously (Newton et al., 2005). PageRuler prestained protein standards (MBI Fermentas, Quantum Scientific, Brisbane, QLD, Australia) were used for molecular weight estimation. Membranes were blocked in Tris-buffered saline (TBS) (pH 7.6), 0.05% Tween 20, 5% skim milk powder for 1 hour at room temperature. Primary antibodies were used at the following dilutions: anti-tyrosinase, 1:4,000; antiDCT, 1:500; anti-TYRP1, 1:20; anti-MITF, 1:1,000; anti-GAPDH, 1:10,000; anti-phospho-CREB, 1:1,000; anti-phospho-ERK, 1:2,000; anti-ERK, 1:1,000; anti-MART-1, 1:1,000. Dilutions were all in fresh 5% skim milk blocking buffer, apart from ERK where BSA replaced milk. Incubations were overnight at 41C, except for TYRP1 which had 1.5 hours at room temperature. After washing in TBS, 0.1% Tween 20 (3  10 minutes at room temperature), the appropriate horseradish peroxidase-conjugated secondary antibody was incubated with the blot for 1 hour at room temperature (1:10,000 in 3% blocking buffer). Following washing, as above, the blot was exposed to Supersignal West Pico (Pierce) for detection. The relative intensity of bands was determined by scanning films on a GS800 calibrated densitometer (Bio-Rad, NSW, Australia), followed by volumetric quantitation using Quantity One software (Bio-Rad). The signal obtained from reprobing blots with GAPDH was used to monitor for equal total protein levels.

Glycosylation analysis Protein was extracted from NHM using the same lysis buffer employed for the tyrosinase assay and the protein concentration determined by the BCA procedure. 10 mg aliquots of lysate were then digested with 500 U Endo H or PNGase F (New England Biolabs, Beverly, MA), or incubated with reaction buffer alone, in accordance to the conditions specified by the manufacturer. After incubation, www.jidonline.org 2225

RA Newton et al. Post-Transcriptional Regulation of Tyrosinase

samples were mixed with SDS extraction buffer (5  ), heat denatured, and detected by Western immunoblotting using antityrosinase, as described above. CONFLICT OF INTEREST The authors state no conflict of interest.

ACKNOWLEDGMENTS We thank Professor Peter Parsons for the gift of the TYRP1 antibody, B8G3. RAS is a Senior Research Fellow of the National Health and Medical Research Council of Australia. DWR was supported by a University of Queensland Confirmation Scholarship. The work was funded by project grants NHMRC301173 and ARC-DP0451738. The Institute for Molecular Bioscience incorporates the Centre for Functional and Applied Genomics as a Special Research Centre of the Australian Research Council.

REFERENCES Abdel-Malek Z, Swope VB, Suzuki I, Akcali C, Harriger MD, Boyce ST et al. (1995) Mitogenic and melanogenic stimulation of normal human melanocytes by melanotropic peptides. Proc Natl Acad Sci USA 92:1789–93 Aksan I, Goding CR (1998) Targeting the microphthalmia basic helix-loophelix-leucine zipper transcription factor to a subset of E-box elements in vitro and in vivo. Mol Cell Biol 18:6930–8 Ando H, Kondoh H, Ichihashi M, Hearing VJ (2007) Approaches to identify inhibitors of melanin biosynthesis via the quality control of tyrosinase. J Invest Dermatol 127:751–61 Baur JA, Sinclair DA (2006) Therapeutic potential of resveratrol: the in vivo evidence. Nat Rev Drug Discov 5:493–506

Fuller BB, Spaulding DT, Smith DR (2001) Regulation of the catalytic activity of preexisting tyrosinase in black and Caucasian human melanocyte cell cultures. Exp Cell Res 262:197–208 Gaggioli C, Busca R, Abbe P, Ortonne JP, Ballotti R (2003) Microphthalmiaassociated transcription factor (MITF) is required but is not sufficient to induce the expression of melanogenic genes. Pigment Cell Res 16:374–82 Halaban R, Cheng E, Svedine S, Aron R, Hebert DN (2001) Proper folding and endoplasmic reticulum to golgi transport of tyrosinase are induced by its substrates, DOPA and tyrosine. J Biol Chem 276:11933–8 Halaban R, Cheng E, Zhang Y, Moellmann G, Hanlon D, Michalak M et al. (1997) Aberrant retention of tyrosinase in the endoplasmic reticulum mediates accelerated degradation of the enzyme and contributes to the dedifferentiated phenotype of amelanotic melanoma cells. Proc Natl Acad Sci USA 94:6210–5 Halaban R, Patton RS, Cheng E, Svedine S, Trombetta ES, Wahl ML et al. (2002) Abnormal acidification of melanoma cells induces tyrosinase retention in the early secretory pathway. J Biol Chem 277:14821–8 Halaban R, Pomerantz SH, Marshall S, Lambert DT, Lerner AB (1983) Regulation of tyrosinase in human melanocytes grown in culture. J Cell Biol 97:480–8 Halaban R, Svedine S, Cheng E, Smicun Y, Aron R, Hebert DN (2000) Endoplasmic reticulum retention is a common defect associated with tyrosinase-negative albinism. Proc Natl Acad Sci USA 97:5889–94 Hemesath TJ, Price ER, Takemoto C, Badalian T, Fisher DE (1998) MAP kinase links the transcription factor microphthalmia to c-Kit signalling in melanocytes. Nature 391:298–301 Hunt G, Donatien PD, Lunec J, Todd C, Kyne S, Thody AJ (1994b) Cultured human melanocytes respond to MSH peptides and ACTH. Pigment Cell Res 7:217–21

Bertolotto C, Abbe P, Hemesath TJ, Bille K, Fisher DE, Ortonne JP et al. (1998a) Microphthalmia gene product as a signal transducer in cAMPinduced differentiation of melanocytes. J Cell Biol 142:827–35

Hunt G, Todd C, Cresswell JE, Thody AJ (1994a) Alpha-melanocyte stimulating hormone and its analogue Nle4DPhe7 alpha- MSH affect morphology, tyrosinase activity and melanogenesis in cultured human melanocytes. J Cell Sci 107:205–11

Bertolotto C, Busca R, Abbe P, Bille K, Aberdam E, Ortonne JP et al. (1998b) Different cis-acting elements are involved in the regulation of TRP1 and TRP2 promoter activities by cyclic AMP: pivotal role of M boxes (GTCATGTGCT) and of microphthalmia. Mol Cell Biol 18:694–702

Im S, Moro O, Peng F, Medrano EE, Cornelius J, Babcock G et al. (1998) Activation of the cyclic AMP pathway by alpha-melanotropin mediates the response of human melanocytes to ultraviolet B radiation. Cancer Res 58:47–54

Boissy RE, Sakai C, Zhao H, Kobayashi T, Hearing VJ (1998) Human tyrosinase related protein-1 (TRP-1) does not function as a DHICA oxidase activity in contrast to murine TRP-1. Exp Dermatol 7:198–204

Iozumi K, Hoganson GE, Pennella R, Everett MA, Fuller BB (1993) Role of tyrosinase as the determinant of pigmentation in cultured human melanocytes. J Invest Dermatol 100:806–11

Busca R, Ballotti R (2000) Cyclic AMP a key messenger in the regulation of skin pigmentation. Pigment Cell Res 13:60–9

Jang M, Cai L, Udeani GO, Slowing KV, Thomas CF, Beecher CW et al. (1997) Cancer chemopreventive activity of resveratrol, a natural product derived from grapes. Science 275:218–20

Chen K, Manga P, Orlow SJ (2002) Pink-eyed dilution protein controls the processing of tyrosinase. Mol Biol Cell 13:1953–64 Chen K, Minwalla L, Ni L, Orlow SJ (2004) Correction of defective early tyrosinase processing by bafilomycin A1 and monensin in pink-eyed dilution melanocytes. Pigment Cell Res 17:36–42 Choi H, Ahn S, Lee BG, Chang I, Hwang JS (2005) Inhibition of skin pigmentation by an extract of Lepidium apetalum and its possible implication in IL-6 mediated signaling. Pigment Cell Res 18:439–46 Chu W, Pak BJ, Bani MR, Kapoor M, Lu SJ, Tamir A et al. (2000) Tyrosinaserelated protein 2 as a mediator of melanoma specific resistance to cisdiamminedichloroplatinum(II): therapeutic implications. Oncogene 19:395–402

Jimenez-Cervantes C, Martinez-Esparza M, Perez C, Daum N, Solano F, Garcia-Borron JC (2001) Inhibition of melanogenesis in response to oxidative stress: transient downregulation of melanocyte differentiation markers and possible involvement of microphthalmia transcription factor. J Cell Sci 114:2335–44 Kim DS, Hwang ES, Lee JE, Kim SY, Kwon SB, Park KC (2003) Sphingosine-1phosphate decreases melanin synthesis via sustained ERK activation and subsequent MITF degradation. J Cell Sci 116:1699–706 Kim DS, Kim SY, Chung JH, Kim KH, Eun HC, Park KC (2002a) Delayed ERK activation by ceramide reduces melanin synthesis in human melanocytes. Cell Signal 14:779–85

Cook AL, Smith AG, Smit DJ, Leonard JH, Sturm RA (2005) Co-expression of SOX9 and SOX10 during melanocytic differentiation in vitro. Exp Cell Res 308:222–35

Kim DS, Park SH, Kwon SB, Li K, Youn SW, Park KC (2004) Epigallocatechin3-gallate and hinokitiol reduce melanin synthesis via decreased MITF production. Arch Pharm Res 27:334–9

Costin GE, Valencia JC, Vieira WD, Lamoreux ML, Hearing VJ (2003) Tyrosinase processing and intracellular trafficking is disrupted in mouse primary melanocytes carrying the underwhite (uw) mutation. A model for oculocutaneous albinism (OCA) type 4. J Cell Sci 116:3203–12

Kim DS, Park SH, Kwon SB, Park ES, Huh CH, Youn SW et al. (2006) Sphingosylphosphorylcholine-induced ERK activation inhibits melanin synthesis in human melanocytes. Pigment Cell Res 19:146–53

Du J, Fisher DE (2002) Identification of Aim-1 as the underwhite mouse mutant and its transcriptional regulation by MITF. J Biol Chem 277:402–6

Kim YM, Yun J, Lee CK, Lee H, Min KR, Kim Y (2002b) Oxyresveratrol and hydroxystilbene compounds. Inhibitory effect on tyrosinase and mechanism of action. J Biol Chem 277:16340–4

Fulda S, Debatin KM (2004) Sensitization for anticancer drug-induced apoptosis by the chemopreventive agent resveratrol. Oncogene 23: 6702–11

Lancon A, Delma D, Osman H, Thenot JP, Jannin B, Latruffe N (2004) Human hepatic cell uptake of resveratrol: involvement of both passive diffusion and carrier-mediated process. Biochem Biophys Res Commun 316:1132–7

2226 Journal of Investigative Dermatology (2007), Volume 127

RA Newton et al. Post-Transcriptional Regulation of Tyrosinase

Leonard JH, Marks LH, Chen W, Cook AL, Boyle GM, Smit DJ et al. (2003) Screening of human primary melanocytes of defined melanocortin-1 receptor genotype: pigmentation marker, ultrastructural and UV-survival studies. Pigment Cell Res 16:198–207 Lin CB, Babiarz L, Liebel F, Roydon Price E, Kizoulis M, Gendimenico GJ et al. (2002) Modulation of microphthalmia-associated transcription factor gene expression alters skin pigmentation. J Invest Dermatol 119:1330–40

Slominski A, Tobin DJ, Shibahara S, Wortsman J (2004) Melanin pigmentation in mammalian skin and its hormonal regulation. Physiol Rev 84:1155–228 Solano F, Briganti S, Picardo M, Ghanem G (2006) Hypopigmenting agents: an updated review on biological, chemical and clinical aspects. Pigment Cell Res 19:550–71 Sturm RA (2006) A golden age of human pigmentation genetics. Trends Genet 22:464–8

Maresca V, Flori E, Cardinali G, Briganti S, Lombardi D, Mileo AM et al. (2006) Ferritin light chain down-modulation generates depigmentation in human metastatic melanoma cells by influencing tyrosinase maturation. J Cell Physiol 206:843–8

Takahashi H, Parsons PG (1990) In vitro phenotypic alteration of human melanoma cells induced by differentiating agents: heterogeneous effects on cellular growth and morphology, enzymatic activity, and antigenic expression. Pigment Cell Res 3:223–32

Nakamura K, Yoshida M, Uchiwa H, Kawa Y, Mizoguchi M (2003) Downregulation of melanin synthesis by a biphenyl derivative and its mechanism. Pigment Cell Res 16:494–500

Takahashi H, Parsons PG (1992) Rapid and reversible inhibition of tyrosinase activity by glucosidase inhibitors in human melanoma cells. J Invest Dermatol 98:481–7

Newton RA, Smit SE, Barnes CC, Pedley J, Parsons PG, Sturm RA (2005) Activation of the cAMP pathway by variant human MC1R alleles expressed in HEK and in melanoma cells. Peptides 26:1818–24

Vance KW, Goding CR (2004) The transcription network regulating melanocyte development and melanoma. Pigment Cell Res 17: 318–25

Niles RM, McFarland M, Weimer MB, Redkar A, Fu YM, Meadows GG (2003) Resveratrol is a potent inducer of apoptosis in human melanoma cells. Cancer Lett 190:157–63

Wakamatsu K, Kavanagh R, Kadekaro AL, Terzieva S, Sturm RA, Leachman S et al. (2006) Diversity of pigmentation in cultured human melanocytes is due to differences in the type as well as quantity of melanin. Pigment Cell Res 19:154–62

Pak BJ, Lee J, Thai BL, Fuchs SY, Shaked Y, Ronai Z et al. (2004) Radiation resistance of human melanoma analysed by retroviral insertional mutagenesis reveals a possible role for dopachrome tautomerase. Oncogene 23:30–8 Pomerantz SH (1966) The tyrosine hydroxylase activity of mammalian tyrosinase. J Biol Chem 241:161–8 Price ER, Horstmann MA, Wells AG, Weilbaecher KN, Takemoto CM, Landis MW et al. (1998) Alpha-Melanocyte-stimulating hormone signaling regulates expression of microphthalmia, a gene deficient in Waardenburg syndrome. J Biol Chem 273:33042–7 Roberts DW, Newton RA, Beaumont KA, Leonard JH, Sturm RA (2006) Quantitative analysis of MC1R gene expression in human skin cell cultures. Pigment Cell Res 19:76–89

Watabe H, Valencia JC, Yasumoto K, Kushimoto T, Ando H, Muller J et al. (2004) Regulation of tyrosinase processing and trafficking by organellar pH and by proteasome activity. J Biol Chem 279:7971–81 Yang S, Irani K, Heffron SE, Jurnak F, Meyskens FL, Jr (2005) Alterations in the expression of the apurinic/apyrimidinic endonuclease-1/redox factor-1 (APE/Ref-1) in human melanoma and identification of the therapeutic potential of resveratrol as an APE/Ref-1 inhibitor. Mol Cancer Ther 4:1923–35 Yasumoto K, Yokoyama K, Takahashi K, Tomita Y, Shibahara S (1997) Functional analysis of microphthalmia-associated transcription factor in pigment cell-specific transcription of the human tyrosinase family genes. J Biol Chem 272:503–9

www.jidonline.org 2227