Intracellular Signaling Mechanisms Leading to Synergistic Effects of Endothelin-1 and Stem Cell Factor on Proliferation of Cultured Human Melanocytes

We previously reported that activation of mitogen-activated protein kinase (MAPK) is involved in the mitogenic stimulation of normal human melanocytes (NHMC) by endothelin-1 (ET-1). In the present study, we determined signaling mechanisms upstream of MAPK activation that are involved in ET-1 stimulation and their synergism with stem cell factor (SCF). Pretreatment of cultured NHMC with ETB receptor antagonists, pertussis toxin, a specific phospholipase C inhibitor (U73122), or a protein kinase C inhibitor (calphostine) blocked a transient tyrosine phosphorylation of MAPK induced by ET-1, whereas the addition of a calcium chelator (BAPTA) failed to inhibit that tyrosine phosphorylation of MAPK. Treatment with ET-1 and SCF together synergistically increased DNA synthesis, which was accompanied by synergism for MAPK phosphorylation. The time course of inositol 1,4,5-trisphosphate formation revealed that there is no difference in the level of inositol 1,4,5-trisphosphate stimulated by ET-1 + SCF or by ET-1 alone. Evaluations of the serine phosphorylation of MEK and Raf-1 activity showed a synergistic effect in SCF + ET-1-treated NHMC. Stimulation with SCF + ET-1 induced a more rapid and stronger tyrosyl phosphorylation of proteins corresponding to p52 and p66 Shc than did stimulation with SCF only, and this was accompanied by a stronger association of tyrosine-phosphorylated Shc with Grb2. Interestingly, a more rapid and marked tyrosine phosphorylation of c-kit was also detected in NHMC-treated with SCF + ET-1 than NHMC treated with SCF only. These data indicate that the synergistic cross-talk between SCF and ET-1 signaling is initiated through the pathway of tyrosine phosphorylation of c-kit, which results in the enhanced formation of the Shc-Grb2 complex which leads in turn to the synergistic activation of the Ras/Raf-1/MEK/MAP kinase loop.

In the long course of studying paracrine mechanisms in-volved in epidermal hyperpigmentary disorders, we have found that endothelin-1 (ET-1) 1 plays a central role in UVB-induced pigmentation (1,2) and in the accentuated pigmentation of senile freckles. 2 Furthermore, stem cell factor (SCF) plays an important role in the increased pigmentation of the epidermis overlying benign fibroblastic tumors in dermatofibroma 3 as well as in UVB-induced pigmentation. 4 We have also observed that aged fibroblasts in culture produce a larger amount of SCF than do younger ones (3), which probably provides a basis for the tendency of aged skins to be more sensitive to environmental stimuli and to be easily induced to epidermal hyperpigmentation. SCF is also known as a stimulator for epidermal hyperpigmentation in mastocytosis where mast cells undergo hyperproliferation in response to the soluble type of SCF derived from keratinocytes (4).
In inherited pigmentary diseases such as piebaldism and Hirschsprung disease, mutations of c-kit (5-7) or endothelin B receptor (8,9), respectively, have been documented. In Waardenburg syndrome type 2, mutation of microphthalmia-associated transcription factor is considered to be a central event leading to the dysfunction or loss of melanocytes (10,11). Recently, the c-kit signaling pathway was found to be upstream of microphthalmia-associated transcription factor transcription through phosphorylation by MAP kinase (MAPK) (12). In mast cells, many reports (13,14) have described a strong link between c-kit expression and microphthalmia-associated transcription factor transcriptional function. Furthermore, a disease termed Shah-Waardenburg syndrome, which combines the Waardenburg type 2 and Hirschsprung phenotypes, shows defects in microphthalmia-associated transcription factor and in the endothelin B receptor (15).
Thus, it is likely that signaling pathways stimulated by binding of ET-1 and SCF to their corresponding receptors have at least some common pathways and are working in a coordinated fashion to regulate melanocyte function. In relation to this, we have recently found that the proliferation of cultured human melanocytes (NHMC) induced by ET-1 is synergisti-* The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked "advertisement" in accordance with 18  cally enhanced by the concomitant addition of SCF (16).
Endothelins are unique mitogens and melanogens for human melanocytes (1,2,17). These cellular actions are known to be initiated by binding of ET-1 to G-protein-coupled ET B receptor, followed by sequential signaling processes consisting mainly of protein kinase C (PKC) and MAPK (16,17). SCF stimulation is important for the survival and proliferation of several cell types in the hematopoietic system (18,19), where it may function in combination with other growth factors. In human melanocytes, SCF alone is sufficient to sustain proliferation under serumfree conditions (16). SCF binding to the c-kit receptor mediates dimerization, activation of its intrinsic tyrosine kinase activity, and autophosphorylation (20). The activated receptor then phosphorylates various substrates and associates with various signaling molecules, including phosphatidylinositol 3Ј-kinase (PI 3-kinase), the Shc and Grb2 adaptor proteins, and the guanine nucleotide exchange factor, SOS, all of which lead to the activation of the Ras-MAPK pathway (21)(22)(23). In the epidermis, the most abundant cells which surround melanocytes are keratinocytes, that are known to produce increased amounts of SCF and ET-1 in response to several stimuli (24). Therefore, it is of particular value to clarify the cross-talk mechanism between ET-1 and SCF-mediated signaling in human melanocytes because each of those specific signaling pathways has been associated with the physiological stimulation of melanogenesis (16). The aim of our present study was to evaluate the mechanisms by which ET signaling activates MAPK and its underlying synergism with SCF. We now report that the synergistic cross-talk between SCF and ET-1 signaling is initiated through the pathway of trans-activation of c-kit, including its enhanced tyrosine phosphorylation, during the SCF-induced activation process. This results in an increase in the formation of the Shc-Grb2 complex, which leads in turn to synergistic activation of the Ras/Raf-1/MEK/MAP kinase loop.
Cell Culture-NHMC were maintained in modified MCDB 153 growth medium supplemented with 1 ng/ml recombinant basic fibroblast growth factor, 5 g/ml insulin, 0.5 g/ml hydrocortisone, 10 ng/ml phorbol 12-myristate 13-acetate, antibiotics (50 g/ml gentamycin and 0.25 g/ml amphotericin B), 0.5% fetal calf serum, and 0.2% bovine pituitary extract at 37°C under a 5% CO 2 atmosphere. In experiments to evaluate cellular effects or signaling changes, NHMC were seeded in culture trays at a density of 5-8 ϫ 10 4 cells/ml, and were cultured in keratinocyte-SFM (Life Technologies, Inc.) containing bovine pituitary extract for 48 h, then treated with reagents at various concentrations.
DNA Synthesis-NHMC cultured in 96-well trays were incubated with ET-1 and/or SCF at concentrations of 0 to 100 nM. Twenty hours later, the cells were labeled for 4 h with 1.0 Ci/ml [ 3 H]thymidine. After three washes with phosphate-buffered saline, the cells were trypsinized and harvested on a glass fiber filter, washed three times with distilled water, and twice with ice-cold ethanol, then dried. The radioactivity on the filter was directly measured using MATRIX 96 (Packard Bioscience Co.).
Measurement of Inositol 1,4,5-Trisphosphate (IP 3 )-These techniques were performed as reported previously (2). Briefly, for IP 3 assay, cells were seeded in 24-well culture trays at a density of 3 ϫ 10 4 -10 5 cells/ml and cultured for 24 -48 h. The media were aspirated and the MCDB 153 medium containing 10 mM LiCl was added and incubated for 10 min at 37°C before stimulation. The ligand stimulation was terminated at designed times by adding 10% perchloric acid and the samples were kept on ice for 15 min. After neutralization with ice-cold 1.5 M KOH for 60 min on ice, the samples were centrifuged at 2,000 ϫ g for 10 min to remove KClO 4 precipitate. The supernatants (100 l each) were subjected to IP 3 assay using the IP 3 assay kit (Amersham Pharmacia Biotech, Buckinghamshire, United Kingdom) (2). The content of IP 3 in each sample was quantitatively determined from a calibration curve established using the binding protein specific for IP 3
Immunoprecipitation-These techniques were performed as reported previously (27,28). Grb2, Shc, and c-kit were immunoprecipitated from whole cell lysates by incubation with 4 g of antibodies for 2 h at 4°C. The resultant immune complexes were then precipitated by incubation with protein G-Sepharose (Amersham Pharmacia Biotech) for 1 h at 4°C. The pellets were washed three times with RIPA buffer, once with phosphate-buffered saline, resuspended in SDS sample buffer (125 mM Tris-HCl, pH 6.8, 20%(v/v) glycerol, 4% (w/v) SDS, 100 mM dithiothreitol, 0.1% (w/v) bromphenol blue), and heated at 95°C for 5 min. Immunoprecipitated and associated proteins were detected with immunoblotting as described above.

Synergistic Stimulatory Effects of the Combination of SCF
and ET-1 on DNA Synthesis-When ET-1 and SCF are concomitantly added at constant and varied concentrations, respectively, there is a synergistic stimulation of DNA synthesis in NHMC (Fig. 1). At ET-1 and SCF concentrations of 10 nM each, a marked synergistic stimulation of DNA synthesis is elicited with a 62-fold increase relative to the 1.2-and 7.2-fold increases elicited by SCF or ET-1 treatments alone, respectively.
ET Binding Initiated Signaling Pathway Leading to the Ac-tivation of MAPK-In order to clarify whether the activation of MAPK is mediated via the endothelin-binding ET A or ET B receptor, we looked at the effect of endothelin A and B receptor antagonists, BQ610 and BQ788, respectively, on the tyrosine phosphorylation of ERK2, a hallmark of MAPK activation as assessed by Western blotting using a phosphotyrosine antibody following immunoprecipitation with anti-ERK2. The ET B receptor antagonist BQ788 completely abolishes endothelin-induced tyrosine phosphorylation of ERK2 whereas the ET A receptor antagonist BQ610 fails to inhibit the phosphorylation (Fig. 2), indicating that the activation of MAPK is mediated through the endothelin B receptor. The addition of 1 g/ml pertussis toxin abolishes tyrosine phosphorylation of ERK2 (Fig. 3), showing that the G i protein is associated with endothelin-induced signaling leading to the activation of MAP kinase. Similarly, the phospholipase C inhibitor, U73122, downregulates tyrosine phosphorylation of ERK2 in a dosedependent manner (Fig. 4), indicating that phospholipase C is also involved in the activation of MAPK during the intracellular signaling initiated by ET and ET B receptor binding. The addition of the calcium chelator, BAPTA, has no effect on the phosphorylation of ERK2 (Fig. 5), showing there is no involvement of calcium mobilization in the activation of MAPK. In contrast, the PKC inhibitor, calphostine, abolishes the phosphorylation of ERK2 (Fig. 6), indicating that activation of PKC is required for the activation of MAPK. Lack of Involvement of the PKC Pathway-The time course of the formation of IP 3 following SCF and ET-1 treatment (Fig. 7) revealed that there is no difference in the level of IP 3 following treatment with ET-1 ϩ SCF or ET-1 alone, indicating no synergy occurring in the PKC pathway.
Synergistic Effect on the Pathway of MAPK Including Raf-1 and MEK-In the evaluation of tyrosine phosphorylation of ERK-1 and -2, an indicator of MAPK activation, as assessed by Western blotting using phosphotyrosine antibodies following immunoprecipitation with anti-ERK-1 and -2, we found that there is a synergistic effect on tyrosine phosphorylation in ET-1 ϩ SCF-treated NHMC compared with ET-1 or SCF alone. There were increases both in the duration and intensity of tyrosine phosphorylation, which reached a peak within 10 min after stimulation (Fig. 8). In the evaluation of serine phosphorylation of MEK, an indicator of MAPKK activation, as assessed by Western blotting using a phospho-specific MEK antibody, there is a synergistic effect in SCF ϩ ET-1-treated melanocytes with a peak within 4 min after stimulation (Fig.  9). This indicates that the synergy in the activation of MAPKK precedes the observed synergy in MAPK activation in SCF ϩ ET-1-treated NHMC. Analysis of Raf-1 activity, an indicator of MAPKKK, as assessed by the final phosphorylation of MBP, showed that 10 nM ET-1 significantly stimulates Raf-1 activity, which indicates that Raf-1 is at least one convergence point from the PKC pathway to the MAPK pathway. Furthermore, there is a synergistic effect on Raf-1 activity in SCF ϩ ET-1treated melanocytes (Fig. 10), suggesting the possibility that the synergistic convergence between SCF and ET-1 signaling is located nearby Raf-1.
The Stimulatory Effect on the Formation of an Shc-Grb2 Complex-The mechanism of tyrosine kinase receptor-stimulated MAPK signaling involves the formation of complexes between the guanine nucleotide exchange protein SOS, and the SH2 and SH3 domain-containing adaptor protein Grb2 either with autophosphorylated growth factor receptors or another tyrosine-phosphorylated adaptor protein known as Shc. Since Shc involvement in this signaling pathway requires tyrosyl phosphorylation, we compared the effects of SCF only and SCF ϩ ET-1 on tyrosyl phosphorylation of Shc proteins in NHMC. Western blotting with phosphotyrosine antibodies following immunoprecipitation with Shc antibodies (Fig. 11) demonstrates that stimulation with SCF ϩ ET-1 induces a more rapid and a stronger tyrosyl phosphorylation of proteins corresponding to p52 (1.2-fold in densitometric intensity) and p66 Shc than is elicited with SCF alone, and that this reaches a maximum (p52: 2.0-fold in densitometric intensity) within 5 min after stimulation. Upon activation of tyrosine kinase receptors, tyrosine-phosphorylated Shc associates with Grb2 and the guanine nucleotide exchange factor SOS, thereby leading to Ras activation. Therefore, we next examined whether the association of tyrosine-phosphorylated Shc with Grb2 also becomes synergistically marked following treatment with SCF ϩ ET-1 as compared with SCF only. As shown in Fig. 12, more distinct bands corresponding to p52 and p66 Shc are observed in immunoprecipitates from SCF ϩ ET-1-treated melanocytes than from a single SCF stimulation. This shows that upon SCF ϩ ET-1 stimulation, Grb2 associates more strongly in a complex with increased amounts of tyrosyl-phosphorylated Shc than it does upon a single SCF stimulation.
The Stimulated Activation of c-kit by the Combination of ET-1 and SCF-Finally, we examined whether the tyrosine phosphorylation of c-kit initiated by SCF binding to the c-kit receptor is stimulated by the concomitant addition of ET-1. Time course experiments assessed by Western blotting (Fig.  13) reveal a rapid and more marked tyrosine phosphorylation of c-kit, which reaches a maximum within 5 min after stimulation, with SCF ϩ ET-1 compared with SCF only. In contrast, a single ET-1 stimulation does not elicit any tyrosine phosphorylation of c-kit, a finding which strongly suggests that ET-1 and ET B receptor binding initiated signaling stimulates tyrosine phosphorylation of c-kit only under conditions where c-kit is activated by SCF binding to its receptor. DISCUSSION In human melanocytes, it is well established that, after binding to its receptor, ET-1 triggers hydrolysis of polyphosphoinositide, which generates IP 3 and diacylglycerol, mobilizing intracellular Ca 2ϩ and activating PKC, respectively, which then stimulate proliferation and melanization (2,16). In addition to the PKC pathway, ET has recently been shown to also activate the MAPK cascade in human melanocytes (16). In other types of cells, such as cardiomyocytes of neonatal rats, a similar activation of MAPK by ET-1 has been reported in relation to mechanical stress-induced cardiac hypertrophy (30). However, the mechanism by which the G i protein-coupled ET receptor activates the MAPK cascade is still poorly characterized. The site at which the PKC pathway stimulated by ET signaling enters the tyrosine kinase pathway (including the MAPK cascade) remains to be elucidated in human melanocytes. Such an understanding would provide a basis for cross-talk mechanisms between ET-1 and SCF initiated signaling. Therefore, we have attempted to determine whether the ET signaling cascade, which consists of its receptor binding, G protein, phospholipase C, calcium mobilization, and PKC activation is really associated with the activation of MAPK. The present study demonstrates clearly that ET activates Erk1/2 in a time-dependent manner in NHMC that express the ET B receptor and that the ET-induced activation of Erk1/2 is independent of calcium mobilization, but is largely associated with the PKC pathway via binding to ET B receptor, G i protein, and the activation of phospholipase C. This strongly suggests that activation of MAPK is involved in the molecular mechanism associated with protein-coupled receptors. In contrast to another study which showed that the formation of the Shc-Grb2 complex can be mediated by G protein activation by ET-1 in astrocytes (31), we found that Erk1/2 activation by ET alone was not accompanied by tyrosine phosphorylation of Shc and the formation of Shc-Grb2 complexes, whereas the prior sequential activation of MEK and Raf-1 were detectable. Since in other cells, Raf-1 is reported to be a target for PKC (32-37), Raf-1 activation, probably via serine/threonine phosphorylation which modifies its catalytic activity, appears to be a convergent point between the PKC and MAPK pathways. Regarding the association between those two pathways, the sum of the above findings indicate that activation of MAPK during the ET signaling pathway is mediated through ET B receptor/G i protein/phospholipase C/PKC and the Raf-1 loop.
SCF and ET-1 binding initiated intracellular signaling that leads to melanization and cell growth of NHMC are transmitted by two major classes of cell surface receptors, tyrosine kinase growth factor receptors (38) and G protein-coupled receptors (16), respectively. Therefore, it is of interest to determine which signaling pathway is responsible for this synergistic effect. At first, we assumed that the convergence between ET-1 and SCF signaling occurs at the activation of MAPK, because we had previously found that ET-1-induced stimulation of melanization and proliferation in NHMC is associated with the activation of MAPK, a pathway very similar to that mediated by SCF (21)(22)(23). In this connection, Western blotting using tyrosyl MAPK antibodies demonstrates that the activation of MAPK occurs synergistically between SCF and ET-1initiated signaling. Similar synergisms to the activation of MAPK have been reported between ET-1 and angiotensin II in cultured cardiomyocytes (30) and between SCF and erythropoietin in human erythroid colony-forming cells (39), in relation to mechanical stress-induced cardiac hypertrophy and expanded erythropoiesis, respectively.
Since MAPK activation is generally accompanied by the prior sequential activation of MEK and Raf-1 (40), we determined whether the synergism in the activation of MAPK is reflected by the synergistic activation of MEK and Raf-1. Related experiments using Western blotting and kinase assays revealed that there is a sequential synergistic activation in the MAPK cascade consisting of Raf-1, MEKK, MEK, and MAPK. This suggests that the convergence point between ET-1 and SCF-initiated signaling is located upstream of Raf-1.
As we confirmed a role for Raf-1 in the cross-talk mechanism between ET-1-associated PKC and SCF-associated tyrosine kinase pathways, we next determined signaling mechanisms leading to the synergistic Raf-1 activation by examining the influence of SCF on the ET-1-dependent PKC pathway. The formation of IP 3 following stimulation is a hallmark for evaluation of PKC activation because IP 3 and the protein kinase C activator, diacylglycerol, are simultaneously generated at an equimolar ratios. The time course of formation of IP 3 following SCF and ET-1 treatment revealed that there is no difference in the raised level of IP 3 induced by ET-1 ϩ SCF or ET-1 alone, indicating no synergy occurs in the PKC pathway.
The mechanism of tyrosine kinase receptor-stimulated mitogenic signaling involves the formation between complexes of the guanine nucleotide exchange protein SOS, and the SH2 and SH3 domain-containing adaptor protein Grb2 with another tyrosine-phosphorylated adaptor protein Shc (21)(22)(23). Recent studies have shown that some G protein-coupled receptors utilize the same effectors as the tyrosine kinase receptor pathway (e.g. Shc-Grb-SOS), resulting in Ras and MAPK activation (41)(42)(43). However, it has been suggested that the pertussis toxin-sensitive G i -coupled receptors utilize a pathway that induces Ras activation in a PKC-independent manner (44,45). In this study, the synergism between ET-1 and SCF was found to be accompanied by synergistic tyrosyl phosphorylation of proteins corresponding to p52 and p66 Shc, leading to synergistic FIG. 11. Synergistic effect on tyrosyl phosphorylation of proteins corresponding to p52 and p66 Shc. A, immunoprecipitation and Western immunoblotting. B, densitometric analysis. Human melanocytes were stimulated for the indicated times with 10 nM ET-1 and/or 10 nM SCF. The tyrosine phosphorylation of immunoprecipitated Shc was evaluated by Western immunoblotting as detailed under "Experimental Procedures." Briefly, cell extract containing 0.5 mg of protein/ sample was incubated with 4 g of Shc antibody. Immunocomplexes were precipitated with protein G-Sepharose, separated by SDS-PAGE, and transferred to polyvinylidene difluoride membrane. Blots were immunodetected with antiphosphotyrosine antibodies (1:1000). Similar results were obtained three times. IP, immunoprecipitation; WB, Western immunoblotting.
FIG. 12. Synergistic effect on the association of tyrosine-phosphorylated Shc with Grb2. Human melanocytes were stimulated for 2 min with 10 nM ET-1 and/or 10 nM SCF. The association of phosphorylated Shc and Grb2 was evaluated by immunoprecipitation with Shc or Grb2 antibodies, followed by Western immunoblotting as detailed under "Experimental Procedures." IP, immunoprecipitation; WB, Western immunoblotting. association of tyrosine-phosphorylated Shc with Grb2. It has also been shown that ET-1 signaling through heterotrimeric G protein-coupled receptors stimulates MAPK activity in primary cultures of astrocytes (46) via an increase in the tyrosine phosphorylation of Shc, which is followed by its stable association with Grb2 (31). Those studies suggested that ET-1-induced MAPK activation is a G protein-coupled pathway that involves Shc, Grb2, and probably Raf-1. Thus, the Shc-Grb2 complex may be involved in activation of the MAPK pathway, not only by several receptor tyrosine kinases but also by heterotrimeric G protein-coupled receptors, such as ET-1 receptors. In contrast to those studies using astrocytes, our study using human melanocytes showed that ET-1 does not stimulate tyrosine phosphorylation of Shc and its association with Grb2 even at its mitogenic and melanogenic concentrations. Interestingly, Western blotting analysis of tyrosine phosphorylation of c-kit upstream of Shc-Grb2 association revealed that this synergistic activation with adaptor molecules is initiated by the synergistic tyrosine phosphorylation of the SCF receptor, c-kit. Again, it should be noted that in human melanocytes the combination of ET-1 and SCF (but not ET-1 alone) enhances tyrosine phosphorylation of c-kit. A similar activation of receptor tyrosine kinases through intracellular signal cross-talk with ET-1-associated G-protein-coupled receptors has been documented for the epidermal growth factor receptor of Rat-1 cells (47) and of smooth muscle cells (48) in which only ET-1 can stimulate tyrosine phosphorylation of the receptor.
In conclusion, the sum of the above findings indicates that synergistic cross-talk between SCF and ET-1 signaling is initiated through the pathway of tyrosine phosphorylation of ckit. This results in the synergistically enhanced formation of Shc-Grb2 complex, which leads to the synergistic activation of the Ras/Raf-1/MEK/MAPK loop. ET-1 associated activation of PKC probably plays a role in the enhanced tyrosine phosphorylation of c-kit although the detailed mechanism is not clear. Thus, our results demonstrate a role for the c-kit tyrosine kinase receptor as a downstream mediator in synergistic mitogenic signaling induced by ET-1 ϩ SCF and suggest a ligandindependent mechanism for c-kit activation through a synergistic intracellular signal cross-talk. FIG. 13. Synergistic effect on tyrosine phosphorylation of c-kit. Human melanocytes were stimulated for 2 or 5 min with 10 nM ET-1 and/or 10 nM SCF. c-kit proteins were extracted in RIPA buffer, immunoprecipitated with c-kit antibody and separated by SDS-PAGE. The phosphorylation of c-kit was detected using phosphorylated tyrosine antibody, as detailed under "Experimental Procedures." IP, immunoprecipitation.