The Serum/PDGF-dependent Melanogenic Role Of The Minute Level Of The Oncogenic Kinase PAK1 in Melanoma Cells Proven By The Highly Sensitive Kinase Assay

Contact: joanna.jia@wecistanche.com / WhatsApp: 008618081934791

Summary: We previously demonstrated that the oncogenic kinase PAK4, which both melanomas and normal melanocytes express at a very high level, is essential for their melanogenesis. In the present study, using the highly sensitive "Macaroni-Western" (IP-ATP-Glo) kinase assay, we investigated the melanogenic potential of another oncogenic kinase PAK1, which melanoma (B16F10) cells express only at a very minute level. After transfecting melanoma cells with PAK1-shRNA for silencing the PAK1 gene, melanin content, tyrosinase activity, and kinase activity of PAK1 were compared between the wild-type and transfectants. We found that (i) PAK1 is significantly activated by melanogenic hormones such as IBMX (3-isobutyl-1-methyl xanthine) and α-MSH (melanocyte-stimulating hormone), (ii) silencing the endogenous PAK1 gene in melanoma cells through PAK1-specific shRNA reduces both melanin content and tyrosinase activity in the presence of both serum and melanogenic hormones to the basal level, (iii) the exogenously added wild-type PAK1 in the melanoma cells boosts the α-MSH-inducible melanin level by several folds without affecting the basal, and (iv) α-MSH/IBMX-induced melanogenesis hardly takes place in the absence of either serum or PAK1, clearly indicating that PAK1 is essential mainly for serum- and α-MSH/IBMX-dependent melanogenesis, but not the basal, in melanoma cells. The outcome of this study might provide the first scientific basis for explaining why a wide variety of herbal PAK1-blockers such as CAPE (caffeic acid phenethyl ester), curcumin, and shikonin in cosmetics are useful for skin-whitening.

Keywords: PAK1, melanogenesis, melanomas, tyrosinase, MITF, skin-whitening, serum

whitening skincare cistanche products

Introduction

The color of hairs, eye irises, and skin are controlled by a family of pigments called melanins. The intracellular source of melanin is tyrosine and is converted to melanin through a series of enzymatic oxidation (hydroxylation) which involves tyrosinase, TRP (tyrosinase-related protein) 1, and TRP2. The expression of genes encoding for these melanogenic enzymes requires at least two oncogenic/melanogenic transcription factors (MTFs): beta-catenin and MITF (microphthalmia-associated transcription factor). The majority of herbal compounds in skin-whitening cosmetics either inhibit directly these melanogenic enzymes or down-regulate the MTFs. Tyrosine analogs such as kojic acid belong to the first category (tyrosinase inhibitors), while the majority of remainings such as CAPE (caffeic acid phenethyl ester), curcumin, and shikonin belong to the second category (MTF regulators) (1,2). Interestingly, many of these MTF regulators including CAPE, curcumin, shikonin, and cucurbitacin are known to block the oncogenic/aging kinase PAK1 (RAC/ CDC42-activated kinase 1) (3-5). Thus, it is most likely that PAK1 is involved in the activation of these melanogenic/oncogenic transcription factors in hair cells, eye irises, or skin melanocytes. Indeed, at least beta-catenin is among the direct substrates of PAK1 (6). PAK1 phosphorylates beta-catenin at Ser 675 for the activation, leading to malignant transformation. Furthermore, beta-catenin is essential for the activation of MITF, leading to melanogenesis or/and oncogenesis of melanocytes (7).

Interestingly, however, it was recently revealed that knocking-out PAK1 gene per se in pigmented mice fails to produce any albino mice (Hong He et al., unpublished observation), clearly indicating that at least the basal melanogenesis in both hair cells and eye irises is independent of PAK1, although the contribution of PAK1 to skin melanogenesis remains to be clarified.

We have previously shown that the oncogenic kinase PAK4 (CDC42-dependent kinase 4) is highly expressed in both melanoma and normal melanocyte cell lines, and responsible for their melanogenesis, activating CREB/beta-catenin-MIFT-tyrosinase pathway, while PAK2 (RAC/CDC42-activated kinase 2), another highly expressed member of PAK family, plays no role in their melanogenesis (8).

In this study, we provide the first biochemical evidence for a specific role of another oncogenic kinase called PAK1 in skin melanogenesis, despite the fact that its expression level is minute: shRNA-induced silencing of PAK1 gene in melanocytes (B16F10) derived from mouse melanoma significantly reduced the α-MSH/IBMX-inducible melanogenesis to the basal level, while over-expression of PAK1 gene boosted the α-MSH-inducible melanogenesis without affecting the basal. Furthermore, we found that the α-MSH/IBMXinducible melanogenesis requires a serum factor which is most likely PDGF (platelet-derived growth factor). In the serum-free medium, only basal melanogenesis takes place.

cistanche reduce tyrosinase's activity

2. Materials and Methods

2.1. Materials

B16F10 melanoma cells were purchased from American Type Culture Collection (Rockville, MD, USA). PAK1-SureSilencing plasmids, attractive transfection, endofree® plasmid maxi kit were purchased from Qiagen (Valencia, CA 91355, USA). Primary PAK1-antibodies, biotinylated secondary antibodies, signalfireTM ECL reagent were obtained from Cell Signalling Technology (Danvers, MA, USA). Kinase Glo reagent and ATP (ATP_Glo kinase kit) were purchased from Promega (Madison, Wisconsin, USA). Lipofectamine and JM109 competent cells were purchased from Invitrogen and Life Technology. AG1295 and AG1478 were purchased from Calbiochem (San Diego, CA, USA). All other chemicals including human PDGF-bb were purchased from Sigma-Aldrich (St. Louis, MO, USA).

2.2. Silencing of PAK1 gene in mouse melanocytes (B16F10) by stable transfection with mouse PAK1- specific shRNA

2.2.1. Cell culture

B16F10 melanoma cells were cultured in DMEM supplemented with 10% heat-activated FBS and 1% penicillin/streptomycin (10,000 U/mL and 100 µg/mL) at 37°C in a humidified atmosphere containing 5% CO2.

2.2.2. Stable transfection with mouse PAK1-specific shRNA

ShRNA transfection of B16F10 cell line was carried out basically through the same procedure previously described (9), but with mouse PAK1-specific shRNAs (# KM04553, Qiagen). The following four distinct clones of shRNAs were used for the transfection: clone 1 (CCAGAGAAGTTGTCAGCTATT), clone 2 (CATCAAGAGTGACAATATTCT), clone 3 (GTACACACGGTTCGAGAAGAT), and clone 4 (GTACCACCAGTGTCAGAAGAT) as well as shCONTROL nontargeting shRNA as the negative control (WT). However, in this study, clones 1 and 2 turned out to be more effective than clones 3 and 4 for silencing mouse PAK1 gene in this melanocyte line (data not shown). Stable transfectants were selected in the presence of G418 (1 mg/mL) which kills more than 99% of non-transfected cells.

2.2.3. Western-blot analysis of PAK1 protein level and in vitro assay for the kinase activity of PAK1 in transfectants

2.2.3.1. Western blot analysis

In order to quantify the extremely low level of PAK1 in both control (WT) cell line and PAK1-silenced transfectants (G418-resistant), by minimizing the "nonspecific noise" levels, we conducted the western blot analysis using a rabbit polyclonal antibody against PAK1 (#2062, Cell Signaling Technology) as the primary antibody. Briefly, each clone was seeded at the concentration of 2 × 105 cells/well in a 6 well plate, precultured for 48 h, and the cell lysates were boiled in 25 µL of SDS-PAGE buffer for 5 min. Eight µL (containing 15 µg of total protein) of each supernatant per slot was used for SDS-PAGE. The nitrocellulose was blotted with the anti-PAK1 IgG (1:1,000 dilution as a primary antibody), and as secondary antibodies, the HRP-linked anti-rabbit IgG and anti-biotin IgG (#7074, #7075, Cell Signaling) were used to detect both PAK1 and β-actin bands which were eventually visualized with the ECL system from Amersham. The western blot assays were representative of three independent experiments.

2.2.3.2. "Macaroni-Western" (IP- ATP_Glo) kinase assay for PAK1 in melanoma cells

The kinase activity of PAK1 in the WT melanocytes and shRNA transfectants (SHs) was measured by a substantial modification (5) of a decade-old method (which we coined "Macaroni-Western") developed by an Italian group (10). Briefly, B16F10 melanoma cell lines (WT or SHs, 2 × 105 cells/mL) were pre-cultured on 6-well plate for 24 h. Then cells were treated with 100 nM α-MSH or 100 µM IBMX for 48 h. Cells were disrupted by lysis buffer containing 50 mM TrisHCl pH 7.5 and 150 mM NaCl and 1% Triton-X. For the immunoprecipitation (IP) of PAK1, the cleared cell lysates were incubated with anti-PAK1 IgG (1:50 dilution) and protein A-agarose beads for 1-2 h in the cold room with continuous shaking by a rotary mixer (Nissin, Suginami-ku, Tokyo, Japan). The resultant IP (PAK1) from each lysate was incubated with ATPGlo kinase assay kit (Promega) in the presence of ATP and MBP (myelin basic protein) for 1 h at 37°C, and the remaining ATP level was measured by the ATP-dependent Luciferin-Luciferase reaction which eventually generates a luminescence (10). The final suspension was centrifuged, and the supernatant was transferred to a 96-well plate for reading. Luminescence was recorded by MTP-880Lab microplate reader (Corona, Hitachinaka-ku, Ibaraki, Japan) with an integration time of 0.5 s per well.

2.3. Measuring melanin content and tyrosinase activity in transfectants

2.3.1. Measurement of melanin content

Melanin content was determined as previously described (11). In brief, B16F10 cells (wild-type or PAK1-specific shRNA transfectants) were plated at a density of 2 × 104 cells/well in a 24-well plate. After 24 h of culture, 100 µM isobutyl-1-methylxanthine (IBMX) or 100 nM α-MSH was added and incubated for an additional 72 h at 37°C. The cells were washed twice with phosphate buffer, then lyzed with 500 µL of a solution containing 1 M NaOH and 10% DMSO and incubated at 80°C for 1 h, to solubilize the melanin, and the melanin content was measured at 490 nm. To compare the melanin content in the wild-type and transfected cells, the total amount of melanin produced by the wild-type melanocytes was considered as the control (100%), and those by the transfectants were calculated accordingly.

2.3.2. Assay for intracellular tyrosinase activity

Tyrosinase activity was determined as previously described (12), with a slight modification. B16F10 cells (wild-type or the transfectants) were plated at a density of 2 × 104 cells/well, and after 24 h of culture, 100 µM IBMX or 100 nM α-MSH was added and incubated for an additional 72 h at 37°C. The cells were then washed with ice-cold phosphate buffer and lysed with phosphate buffer (pH 6.8) containing 1% Triton-X (500 µL/well). The plates were frozen at −80°C for 30 min. After thawing, 100 µL of 1% L-DOPA was added to each well. Following incubation at 37°C for 2 h, the absorbance was measured at 490 nm.

2.4. Measuring the melanin content in the wild-type and PAK1-overexpressing melanocytes after α-MSH treatment

Using Myc vector (2 µg), melanocytes (B16F10) were transfected transiently with either wild-type (WT) PAK1 or so-called "constitutively activated" (CA) PAK1 carrying T423E mutation. After 3 days of culture, each transfected clone was harvested for further experiment. The effect of PAK1-overexpression on the melanin content was measured through the same procedures described previously (8). Briefly, melanocytes, either the wild-type or PAK1- overexpression which express either the wild-type PAK1 or its CA mutant, were incubated with 100 nM α-MSH for 3 days. Cells were lysed with a solution containing 1 M NaOH and 20% DMSO to dissolve the melanin, and its content was determined at 405 nm.

2.5. Melanogenesis in a serum-free medium

B16F10 cells (wild-type or PAK1-specific shRNA transfectants) were plated at a density of 2 × 104 cells/well in a 24-well plate in the presence of 10% FBS. After 24 h of incubation, the culture medium is replaced with a serum-free medium, and cultured for an additional 72 h with or without 100 µM IBMX or 100 nM α-MSH to measure the melanin content.

2.6. Effect of PDGF/EGF receptor inhibitors on serum dependent melanogenesis

2 × 104 B16F10 cells were seeded in each well for 24 h in 10% FBS-containing medium, and then cultured for an additional 72 h in the fresh medium containing 100 nM α-MSH and followings: (1) no serum, (2) 10% FBS alone, (3) 10% FBS plus 2 µM AG1295 (PDGF receptor inhibitor), and (4) 10% FBS plus 400 nM AG1478 (EGF receptor inhibitor), to measure the melanin content.

2.7. Statistical analysis

Data are expressed as mean values with their standard errors. Statistical comparisons were performed by one-way ANOVA followed by Duncan's multiple range test. Statistical analysis was conducted using SAS (release 9.2; SAS Institute, Cary, NC, USA), and p ≤ 0.05 was considered significant.

maca ginseng cistanche

3. Results

3.1. Activation of PAK1 in a melanoma cell line (B16F10) by melanogenic hormones

Two melanogenic hormones, α-MSH (melanocyte-stimulating hormone) and IBMX (3-isobutyl-1-methyl xanthine) are often used to stimulate melanogenesis in melanocytes such as melanoma cell line B16F10. Thus, we first examined if these hormones can activate PAK1 in this cell line, although the protein level of PAK1 there is extremely low, compared with those of PAK2 and PAK4 (8). To monitor the changes in kinase activity of PAK1 in cells, we recently developed a highly sensitive/ specific kinase assay called "Macaroni-Western" (IPATP-Glo) kinase assay protocol by combining the immuno-precipitation (IP) of PAK1 from cell lysates and Promega's ATP_Glo kinase kit (5). Using this kinase assay, we found that both α-MSH (100 nM) and IBMX (100 µM) activate PAK1 in melanoma cells, with α-MSH being more potent than IBMX (see in Figure 1). However, we should remind readers that the fold activation of PAK1 shown here is only apparent, because during this time-consuming test tube IP and kinase assay (over 3 h in total) taking place without these melanogenic hormones, PAK1 would be gradually normalized over time. In other words, we cannot freeze the exact kinase status of PAK1 at the end of cell culture treated with these stimulators, and the fold activation is only an under-estimated reflection of PAK1 activation taken place during cell culture by these simulators.

3.2. Silencing of PAK1 gene by shRNAs in mouse melanoma cell line

In order to investigate further biochemically if PAK1 contributes to the melanogenesis in skin cells (melanocytes), we stably transfected the melanoma cell line B16F10 with mouse PAK1-specific shRNA to silence PAK1 gene selectively.

3.2.1. Silencing PAK1 gene in melanocytes

Our preliminary western blot analysis suggested that in two distinct PAK1-silenced clones (SH1 and 2) the PAK1 protein levels are significantly (by around 75%) reduced, but the growth rate of this melanoma cell line per se was not significantly affected by the PAK1 silencing (data not shown). However, the accurate quantification of PAK1 protein levels by scanning this western blot was rather difficult, mainly because of the high background noise around the PAK1 band in the effort to visualize its extremely low expression levels. Instead, using the "Macaroni-Western" kinase assay (5,10), we verified that the kinase activity of PAK1 in both SH1 and 2 clones are only around 25-30% of that in the control (WT) cells (see Figure 2A).

3.2.2. Reduction in the melanogenesis by silencing PAK1 gene

Our next critical question was if such a reduction in the kinase activity of PAK1 affects melanogenesis. Thus, the melanin content in these clones (SH1 and 2) was compared with the wild-type (WT) melanoma cells, after treating all these cells with α-MSH, a melanogenic inducer, for 72 h. As shown in Figure 2B, the melanin content in these PAK1-deficient melanocytes (SH1 and 2) is reduced by around 50% (51-55%), reaching the basal level in the WT without melanogenic hormone. To see if this reduction in melanin content is associated with the suppression of tyrosinase activity, we measured the tyrosinase activity in PAK1-deficient cells (clones SH1 and 2), compared with the WT. As shown in Figure 2C, the tyrosinase activity in PAK1-deficient cells is only around 45% (33-58%) of that in the WT, again reaching the basal (non-stimulated) level, clearly indicating that PAK1 is essential for α-MSH-induced melanin production by tyrosinase in melanocytes. Furthermore, in a very similar manner the IBMX-induced melanogenesis was also reduced by silencing PAK1 gene in this melanoma cell line (data not shown). However, taking the following two factors into account, it is most likely that only a half of melanogenesis in melanocytes is PAK1-dependent, and the remaining (largely "basic") is PAK4-dependent: (i) PAK4 is also essential for the α-MSH-induced melanogenesis in melanocytes (8), and (ii) around 25% of PAK1 gene appears to be still expressed in these transfectants (SH1 and 2). However, it remains to be clarified vigorously whether PAK1 is responsible for the hormone-inducible or the basal melanogenesis.

Furthermore, we have confirmed that the endogenous PAK1 is significantly activated by melanogenic inducers such as IBMX and α-MSH in this melanoma cell line (see Figure 1), but without any significant change in its autophosphorylation at Thr 423 as judged by the anti-pPAK1 antibody (data not shown).

3.3. Increase in α-MSH-inducible melanogenesis by over-expressed PAK1

By overexpression of the wild-type (WT) PAK1 gene in melanocytes (B16F10 cell line) by transfection, we revealed that the exogenously added PAK1 gene boosts the α-MSH-inducible melanin level by several folds (see the left of Figure 3, compare lanes 2 and 4), while it hardly affects the independent "basal" melanin level per se (compare lanes 1 and 3 in the left panel of Figure 3), suggesting that PAK1 contributes mainly to α-MSH/IBMX-dependent melanogenesis, and not the basal melanogenesis without exogenous stimulator(s).

3.4. Serum effect on α-MSH/IBMX-inducible melanogenesis

More interestingly, we found that the induction of melanogenesis by α-MSH/IBMX requires 10% FBS (fetal bovine serum) in addition to PAK1. As shown in Figure 4A, in a serum-free medium, either α-MSH or IBMX hardly induced the melanogenesis in the WT cells, although the 10% FBS alone (without α-MSH/IBMX) induced the melanogenesis significantly (by around 60%). Interestingly the melanogenesis in SH transfectants (PAK1-deficient cells) in the presence or absence of serum was the same level as that in WT cells in the absence of serum (see Figure 4B), suggesting that the serum-dependent melanogenesis also requires PAK1. In supporting this notion, serum alone significantly activates the kinase activity of PAK1 in WT cells, but the α-MSH-dependent activation of PAK1 in WT cells requires serum (see Figure 4C).

Regarding the chemical nature of melanogenic serum factor that alone activates PAK1, we speculate that it is PDGF (platelet-derived growth factor) for the following reasons: (i) more than a decade ago we have shown that PDGF is the sole growth factor in serum that activates PAK1 through the transactivation of EGF (epidermal growth factor) receptor by PDGF receptor (13,14), and very recently others have proven that either PDGF or EGF alone indeed stimulates the melanogenesis of melanocytes (15,16).

3.5. The serum/PDGF-dependent melanogenesis requires EGF receptor

In an attempt to identify the specific chemical nature of this melanogenic serum factor, we have tested the effect of either AG1295 (inhibitor specific for PDGF receptor) or AG1478 (inhibitor specific for EGF receptor) on the serum-dependent melanogenesis in melanocytes. As shown in Figure 5, either 2 µM AG1295 or 400 nM AG1478 strongly reduced the serum-dependent melanogenesis to the level equivalent to the basic (serum-free) melanogenesis. Since PDGF is abundantly present in serum, but not EGF and AG1295 does not inhibit EGF receptor, while AG1478 does not inhibit PDGF receptor (13), it is most likely that PDGF in serum activates its receptor that in turn trans-activates EGF receptor that eventually activates PAK1 that is essential for serum-dependent melanogenesis (for detail, see Figure 6). Based on this assumption, we shall discuss later the most likely mechanism underlying the apparent synergy between α-MSH and serum/ PDGF for the activation of PAK1, leading to robust melanogenesis.

Lastly, it should be worth pointing out that the "basal" melanin content without α-MSH was not significantly changed by over-expressed CA (constitutively active) mutant of PAK1 carrying T423E mutation (see the left panel of Figure 3, lane 5) as if it were either a DN (dominant negative) or inactive mutant. In fact, α-MSH did not induce the phosphorylation of WT PAK1 at Thr 423 at all (data not shown). Thus, it could be concluded that the auto-phosphorylation of PAK1 at Thr 423 has nothing to do with α-MSH-induced activation of PAK1, which leads to the robust melanogenesis in melanoma cells. Since T423E mutant of PAK1 is well known to be highly oncogenic (6), perhaps the auto-phosphorylation at Thr 423 might serve a switch from "melanogenic" to "oncogenic" signaling.

4. Discussion

From our observation on both PAK1-dependent and PAK4-dependent melanogenesis in melanocyte/ melanoma cells, there rise two potentially interesting issues to be pointed out: (i) The "specific" melanogenic activity of PAK1 (only minutely expressed) must be far higher than that of PAK4 (highly expressed) in melanoma cells. (ii) It appears that PAK1 is mainly responsible for serum-induced melanogenesis, while PAK4 is mainly responsible for the intrinsic (basal) melanogenesis. In other words, although PAK4- deficiency is embryonically lethal in mice, while PAK1-deficiency alone fails to produce "albino" mice, the apparent difference in the original skin color (basic melanogenesis) between black and white people for instance could be at least partly a reflection of the difference in the expression level of PAK4, as well as difference in the level of melanogenic tyrosinases.

In vivo, using HRM-2 (pigmented but hairless) mice, we have recently shown that a cream containing 10 μM PF3758309 (PAK1/PAK4-inhibitor) reduces the UV-induced melanogenesis in their skin to the basal level, although it still remains to be clarified whether PAK4 or PAK1 is responsible for UV-induction of melanogenesis (8). Since PAK1 is responsible for the inducible, but not the basal, melanogenesis, it would be worth testing if PAK1 significantly contributes to the UV-inducible melanogenesis (sun tanning) as well, using the rare PAK1 KO (knock out) mutant derived from C57B16 strain of mice carrying dark hairs and eyes (Hong He et al., unpublished observation), in an attempt to understand if a variety of herbal PAK1 blockers in cosmetics are useful for sun-screening agents or not. Interestingly PAK1 was reported to be activated by UV irradiation and other DNA-damaging agents (17).

It has been shown that α-MSH activates the Tyrkinase JAK2 (18), which in turn activates PAK1 by the phosphorylation at Tyr 285, (instead of Thr 423), leading to the PIX-PAK1 interaction (19). This could explain why neither α-MSH–dependent activation of PAK1 nor melanogenesis involves the autophosphorylation of PAK1 at Thr 423. Thus, we recently investigated if the PAK1-dependent melanogenesis is blocked by a potent herbal JAK2-inhibitor called cucurbitacin I (CBI) from bitter melon (Goya) that inhibits directly JAK2 (20), and found that CBI inhibits the melanogenesis in the presence of melanogenic hormones by more than 70%, suggesting the possibility that CBI blocks not only PAK1 but also PAK4 (5). In this context, it would be of great interest to note that the herbal PAK1-blockers such as CAPE, curcumin, shikonin, and FTY 720 could inhibit the α-MSH-induced melanogenesis in mouse or human skin cells only by around 50% even at their concentrations where PAK1 is almost completely blocked (1,2), whereas the synthetic pan-PAK-blocker PF3758309, inhibiting both PAK1 and PAK4, abolishes the melanogenesis in skin cells by around 90% at 300 nM (8). In other words, the α-MSH-induced melanogenic system in melanocytes could distinguish the PAK1-specific blockers from panPAK-blockers. So far no herbal PAK4-specific blockers (other than PAK4-specific siRNAs) have been identified. However, a very potent quassinoid called glaucarubinone derived from a bitter tree grown in Amazon jungles has recently been shown to block both PAK1 and PAK4, inhibiting the growth of pancreatic cancer cells both in vitro and in vivo (21). Thus, it would be of great interest to test if this herbal PAK-blocker suppresses the melanogenesis of skin cells almost completely as does PF3758309.

The major aim of this study was to focus on the specific melanogenic role of PAK1, and not to investigate in detail how PAK1 activates the melanogenic signaling pathway including melanogenic enzymes and MTFs. However, it is most likely that PAK1 activates directly beta-catenin by phosphorylating at Ser 675, leading to the activation of MITF which is essential for expression of genes encoding for melanogenic enzymes such as tyrosinase (Tyr).

Recently we found that PAK4 (CDC42-dependent kinase 4) is also involved in melanogenesis of the same melanoma cell line by activating two transcription factors, CREB and beta-catenin, both of which are essential for MIFT activation (8). Since CREB is known to be activated by LIM kinase (22,23) which like beta-catenin, is among the common direct substrates of both PAK1 and PAK4 (6,18), it is most likely that PAK1 and PAK4 share the same CREB/beta-catenin-MITF signaling pathways to activate the melanogenic enzyme genes.

However, the detailed signal pathways leading to the α-MSH/IBMX-dependent activation of PAK1 could differ significantly from those leading to the activation of PAK4, mainly because the former involves the serum factor (PDGF). The serum alone is capable of activating PAK1 (13) and boosts the α-MSH/IBMXdependent activation of PAK1 as well. In other words, there is a clear synergy between the serum and these melanogenic hormones in both PAK1 activation and melanogenesis.

The following is our working hypothesis as to how this synergy could take place. The major signaling pathway leading to the PAK1 activation is the oncogenic EGFR (epidermal growth factor receptor)-RAS-PI 3 kinase-RAC/CDC42-PAK1 cascade. However, this cascade alone is not sufficient for the full activation. It needs another factor called PIX, an SH3 adaptor protein that binds directly PAK1 through its Pro-rich motif of 18 amino acids called PAK18. The PIX-PAK1 interaction needs a third protein called JAK2, a Tyr-kinase, that phosphorylates PAK1 at Tyr 285. RAS up-regulates JAK2 through prolactin. According to a few previous findings by us and others (13-16), PDGF (platelet-derived growth factor) is the major melanogenic serum factor(s) essential for the α-MSH/IBMX-induced melanogenesis, because it activates its receptor (PDGFR) Tyr-kinase that in turn trans-activates EGFR (epidermal growth factor receptor=ErbB1) Tyr-kinase, leading to the activation of PAK1 through the oncogenic RAS-PI3 kinase-RAC/ CDC42 pathway (13). Thus, PDGF in serum alone can activate PAK1 and induce melanogenesis up to halfway. However, for the full activation of PAK1-dependent melanogenesis, α-MSH/IBMX is needed to activate the JAK2 through a cAMP-dependent pathway (which might involve PKA) that eventually secures the PIX-PAK1 interaction (for detail, see Figure 6).

In conclusion, here we present the very first biochemical evidence which might explain how a variety of herbal PAK1-blockers such as CAPE, curcumin, and shikonin in cosmetic creams could contribute to the "skin-whitening" effects. A series of analogous studies with human melanocytes are awaited for further proof or confirmation.

maca ginseng cistanche

References

1. Lee JH, Jang JY, Park C, Kim BW, Choi YH, Choi BT. Curcumin suppresses alpha-melanocyte-stimulating hormone-stimulated melanogenesis in B16F10 cells. Int J Mol Med. 2010; 26:101-106.

2. Lee JY, Choi HJ, Chung TW, Kim CH, Jeong HS, Ha KT. Caffeic acid phenethyl ester inhibits alpha-melanocyte-stimulating hormone-induced melanin synthesis through suppressing transactivation activity of microphthalmia-associated transcription factor. J Nat Prod. 2013; 76:1399-1405.

3. Maruta H. Herbal therapeutics that block the oncogenic kinase PAK1: A practical approach towards PAK1- dependent diseases and longevity. Phytother Res. 2014; 28:656-672.

4. Nguyen BCQ, Taira N, Tawata S. Several herbal compounds in Okinawa plants directly inhibit the oncogenic/aging kinase PAK1. Drug Discov Ther. 2014; 8:238-244.

5. Nguyen BCQ, Be Tu PT, Tawata S, Maruta H. Combination of immunoprecipitation (IP)-ATP_Glo kinase assay and melanogenesis for the assessment of potent and safe PAK1-blockers in cell culture. Drug Discov Ther. 2015; 9:289-295.

6. He H, Maruta H. Oncogenicty of PAKs and their substrates. In: PAKs, RAC/CDC42 (p21)-activated kinases: Towards the cure of cancers and other PAKdependent diseases (Maruta H, ed.). Elsevier, Oxford, 2013; pp. 23-51.

7. Widlund HR, Horstmann MA, Price ER, Cui J, Lessnick SL, Wu M, He X, Fisher DE. β-Catenin-induced melanoma growth requires the downstream target Microphthalmia-associated transcription factor. J Cell Biol. 2002; 158:1079-1087.

8. Yun CY, You ST, Kim JH, Chung JH, Han SB, Shin EY, Kim EG. p21-activated kinase 4 critically regulates melanogenesis via activation of the CREB/MITF and β-catenin/MITF pathways. J Invest Dermatol. 2015; 135:1385-1394.

9. Huynh N, Liu KH, Baldwin GS, He H. P21-activated kinase 1 stimulates colon cancer cell growth and migration/invasion via ERK- and AKT-dependent pathways. Biochim Biophys Acta. 2010; 1803:1106-1113.

10. Tagliati F, Bottoni A, Bosetti A, Zatelli MC, degli Uberti EC. Utilization of luminescent technology to develop a kinase assay: Cdk4 as a model system. J Pharm Biomed Anal. 2005; 39:811-814. 11. Yoon NY, Eom TK, Kim MM, Kim SK. Inhibitory effect of phlorotannins isolated from Ecklonia cava on mushroom tyrosinase activity and melanin formation in mouse B16F10 melanoma cells. J Agric Food Chem. 2009; 57:4124-4129.

12. Li X, Guo L, Sun Y, Zhou J, Gu Y, Li Y. Baicalein inhibits melanogenesis through activation of the ERK signaling pathway. Int J Mol Med. 2010; 25:923-927.

13. He H, Levitzki A, Zhu HJ, Walker F, Burgess A, Maruta H. Platelet-derived growth factor requires epidermal growth factor receptor to activate p21-activated kinase family kinases. J Biol Chem. 2001; 276:26741-26744.

14. Saito Y, Haendeler J, Hojo Y, Yamamoto K, Berk BC. Receptor hetero-dimerization: Essential mechanism for platelet-derived growth factor-induced epidermal growth factor receptor transactivation. Mol Cell Biol. 2001; 21:6387-6394.

15. Hirobe T, Shibata T, Fujiwara R, Sato K. Platelet-derived growth factor regulates the proliferation and differentiation of human melanocytes in a differentiationstage-specific manner. J Dermatol Sci. 2016; 83:200-209.

16. Garcez RC, Teixeira BL, Schmitt Sdos S, Alvarez-Silva M, Trentin AG. Epidermal growth factor (EGF) promotes the in vitro differentiation of neural crest cells to neurons and melanocytes. Cell Mol Neurobiol. 2009; 29:1087- 1091.

17. Roig J, Traugh JA. p21-activated protein kinase gammaPAK is activated by ionizing radiation and other DNAdamaging agents. Similarities and differences to alphaPAK. J Biol Chem. 1999; 274:31119-31122.

18. Buggy JJ. The binding of the alpha-melanocyte-stimulating hormone to its G-protein-coupled receptor on B-lymphocytes activates the Jak/STAT pathway. Biochem J. 1998; 331:211-216.

19. Hammer A, Oladimeji P, De Las Casas LE, Diakonova M. Phosphorylation of tyrosine 285 of PAK1 facilitates βPIX/ GIT1 binding and adhesion turnover. FASEB J. 2015; 29:943-959.

20. Blaskovich MA, Sun J, Cantor A, Turkson J, Jove R, Sebti SM. Discovery of JSI-124 (cucurbitacin I), a selective Janus kinase/signal transducer and activator of transcription 3 signaling pathway inhibitor with potent antitumor activity against human and murine cancer cells in mice. Cancer Res. 2003; 63:1270-1279.

21. Yeo D, Huynh N, Beutler JA, Christophi C, Shulkes A, Baldwin GS, Nikfarjam M, He H. Glaucarubinone and gemcitabine synergistically reduce pancreatic cancer growth via down-regulation of p21-activated kinases. Cancer Lett. 2014; 346:264-272.

22. Dan C, Kelly A, Bernard O, Minden A. Cytoskeletal changes regulated by the PAK4 serine/threonine kinase are mediated by LIM kinase 1 and cofilin. J Biol Chem. 2001; 276:32115-32121.

23. Yang EJ, Yoon JH, Min DS, Chung KC. LIM kinase 1 activates cAMP-responsive element-binding protein during the neuronal differentiation of immortalized hippocampal progenitor cells. J Biol Chem. 2004; 279:8903-8910.