0021-972X/99/$03.00/0 The Journal of Clinical Endocrinology & Metabolism Copyright © 1999 by The Endocrine Society
Vol. 84, No. 5 Printed in U.S.A.
Interactions between Growth Hormone, Insulin-Like Growth Factor I, and Basic Fibroblast Growth Factor in Melanocyte Growth* STEPHANIE R. EDMONDSON, VINCENZO C. RUSSO, ANDREW C. MCFARLANE, CHRISTOPHER J. WRAIGHT, AND GEORGE A. WERTHER Centre for Hormone Research, Royal Children’s Hospital, Parkville, Victoria 3052, Australia ABSTRACT Melanocytes, highly differentiated neural crest-derived cells, are located in the basal layer of the epidermis, where they play a role in protecting against UV damage in the skin. Previous studies suggest that both growth hormone (GH) and the insulin-like growth factor I (GH/IGF-I) system may be important for melanocyte growth and function. We have therefore characterized the role of the GH/IGF system in melanocyte growth in vitro and its interaction with the local growth factor basic fibroblast growth factor (bFGF). Analysis of the effects of GH, IGF-I, and bFGF and combinations of these growth factors on melanocyte growth in vitro revealed that 1) GH stimulates the growth of melanocytes when combined with IGF-I, des(1–3)IGF-I [an analog of IGF-I that has a reduced binding affinity for IGF-binding proteins (IGFBPs)], or bFGF, either separately or in combination; 2)
M
ELANOCYTES are specialized neural crest-derived cells situated in the basal layer of the epidermis. They play a protective role against UV damage in the skin by producing melanin in response to UV stimulation, which is passed, via their dendrites, onto adjacent keratinocytes. The relationship between melanocytes and keratinocytes is thus a crucial feature of the epidermis and is known to involve numerous growth factors, including basic fibroblast growth factor (bFGF) (1). Insulin-like growth factor I (IGF-I), a potent mitogen for keratinocytes, may also be involved in this relationship. An essential role for the IGF system in normal epidermal development is indicated by the finding that IGF-I receptor (IGF-IR) knockout mice have an extremely thin epidermis (2). IGF-binding proteins (IGFBPs), modulators of IGF action, have also been identified in skin (3–5), cultured keratinocytes (6, 7), and dermal fibroblasts (8). Furthermore, the IGFBPs have been shown to regulate IGF-I action on keratinocytes (9) and fibroblasts in vitro (10) and show cellspecific regulation by several growth factors present in skin (10, 11). The IGF-I system appears to play a role in melanocyte growth and function. Melanocytes, both in vitro and in vivo, express IGF-I messenger ribonucleic acids (mRNAs) and Received December 11, 1998. Revision received February 3, 1999. Accepted February 9, 1999. Address all correspondence and requests for reprints to: Stephanie R. Edmondson, Centre for Hormone Research, Royal Children’s Hospital, Flemington Road, Parkville, Victoria 3052, Australia. E-mail: edmondss@ cryptic.rch.unimelb.edu.au. * This work was supported by the National Health and Medical Research Council of Australia.
in contrast to the lack of effect of GH or bFGF alone, both IGF-I and des(1–3)IGF-I enhance melanocyte growth in a dose-dependent manner; and 3) IGF-I is more efficacious in eliciting a growth response at low concentrations compared to des(1–3)IGF-I. Using Western ligand blotting, affinity cross-linking, immunoprecipitation, RIA, and Northern analysis, we show that cultured human melanocytes synthesize and secrete minimal amounts of IGFBP. IGFBP-4 is the major IGFBP produced by these cells when cultured in complete growth medium or in the presence of either IGF-I or des(1–3)IGF-I alone. In conclusion, these studies provide support for a role for both GH and IGF-I in the growth of human melanocytes in vitro, involving synergy with bFGF. Low levels of melanocyte-derived IGFBP-4 may play a role in enhancing the modulation of IGF action. (J Clin Endocrinol Metab 84: 1638 –1644, 1999)
IGF-I receptor (IGF-IR) (12–14) and in vitro grow in response to IGF-I (15). IGF-I also plays an important role in the growth of some melanomas (16). Recent reports have noted enhanced growth of melanocytic lesions of patients treated with recombinant human GH (17, 18). The presence of GH receptor (GHR) mRNA in cultured melanocytes suggests that GH may have a direct action on these cells (12). Alternatively, locally synthesized or circulating IGFs may mediate the effects of GH on melanocytes. In the present study, we have characterized the role of GH/IGF system in melanocyte growth in vitro, including its interaction with another key local growth factor, bFGF. We have further examined the potential role of melanocytederived IGFBPs in modulating IGF action. Materials and Methods Materials Melanocyte growth medium [MGM-2; which is melanocyte basal medium (MBM) and includes 20 mg/mL bovine pituitary extract, 5 mg/mL insulin, 1 ng/mL bFGF, 0.5 mg/mL hydrocortisone, 10 ng/mL phorbol 12-myristate 13-acetate, and 10 ng/mL GA], MBM, and two primary melanocyte cell cultures were obtained from Clonetics (San Diego, CA). Hanks’ Balanced Salt Solution, gentamicin (10 mg/mL), geneticin (G418), and 0.025% trypsin-1 mmol/L ethylenediamine tetraacetate (EDTA) were all obtained from Life Technologies (Grand Island, NY). Dispase was purchased from Collaborative Biomedical Products (Bedford, MA). FCS, DMEM, and phosphate-buffered saline (calcium and magnesium free; PBS) were obtained from Trace (Melbourne, Australia). Tissue culture flasks and 24-well plates were obtained from Nunc (Copenhagen, Denmark). RIA grade BSA, protein A-Sepharose, and l-DOPA (3,4-dihydroxyphenylalanine) were purchased from Sigma Chemical Co. (St. Louis, MO). Nitrocellulose filters were purchased from Schleicher & Schuell, Inc. (Dassel, Germany).
1638
GH, IGF-I, AND bFGF IN MELANOCYTE GROWTH [125I]IGF-I (2000 Ci/mmol), 14C-labeled protein molecular mass (14 –22 kDa) markers, [32P]dCTP (3000 Ci/mmol), [32P]ATP (3000 Ci/mmol), and Hybond-N nylon membranes were obtained from Amersham (Aylesbury, UK). [125I]IGF-II (1500 Ci/mmol) was a gift from Dr. L. A. Bach (Department of Medicine, Austin and Repatriation Medical Centre, Melbourne, Australia). The complementary DNA (cDNA) for human IGFBP-4 was provided by Dr. S. Shimasaki (Whittier Institute, La Jolla, CA) (19). Recombinant human IGF-I was a gift from Dr. A. Sko¨ttner (Pharmacia-Upjohn, Stockholm, Sweden). Recombinant human bFGF and the random priming kit were obtained from Boehringer Mannheim (Sydney, Australia). Des(1–3)IGF-I was purchased from Gropep (Adelaide, Australia). 2,3-Bis(2-methoxy-4-nitro-5-sulfonyl)-5[(phenylamino)carbonyl]-2H-tetrazolium hydroxide (XTT) was obtained from Scima R (Victoria, Australia). Disuccinimidyl suberate was obtained from Pierce Chemical Co. (Rockford, IL).
Isolation of melanocytes Melanocytes were purchased from Clonetics (newborn foreskinderived) or were isolated from foreskins (collected after obtaining ethical approval from the Department of Surgery, Royal Children’s Hospital, Melbourne, Australia). Foreskins (newborn; 11 months old; and 8, 9, and 11 yr old) were collected in keratinocyte serum-free medium containing gentamicin (5 mg/mL) and were stored at 4 C overnight. After transfer to 70% ethanol for 10 min, foreskins were rinsed in PBS, cut into 5 3 5-mm pieces, immersed in Hanks’ Balanced Salt Solution containing dispase at 25 caseinolytic units/mL, and incubated at 4 C for 18 h. The epidermal layer was removed and placed in 2 mL 0.025% trypsin/1 mmol/L EDTA at 37 C for 20 –30 min. During this period cells were dispersed by intermittent vigorous pipetting. Trypsin was inactivated by the addition of 0.5 mL FCS, then the cell suspension was filtered through sterile gauze and pelleted at room temperature (1000 rpm) for 10 min. Cells were resuspended in MGM-2 and plated into two 75-cm2 flasks. Flasks were incubated at 37 C in 5% CO2 overnight. Medium was changed every 3 days, during which time the majority of the keratinocytes and occasional fibroblasts had died, and melanocytes remained adhered to the flask. Differential trypsinization was used to completely remove contaminating keratinocytes and fibroblasts. Specifically, cells were incubated in trypsin-EDTA at 4 C for 5 min, and loosely adhered melanocytes were removed, leaving fibroblasts and keratinocytes. The melanocytes were pelleted and replated into a fresh 75-cm2 flask. The following day the melanocytes were assessed for contamination with fibroblasts and keratinocytes. On rare occasions keratinocytes and fibroblasts were still present. For complete removal of fibroblasts, geneticin (G418; 100 mg/mL) was added to the cultures according to the method of Halaban and Alfano (20). Cells were grown in the presence of G418 for only 1 week. In the primary melanocyte cell cultures used for these studies all contaminating keratinocytes had died by passage 3. If keratinocytes were present in melanocyte cultures at passage 3, the cells were discarded.
Identification of melanocytes Melanocytes isolated in this laboratory were positively identified following a modified method described by Hu et al. (21). Primary melanocytes and human skin fibroblasts (negative control) were plated into 24-well tissue culture plates with their appropriate growth medium (fibroblasts were grown in DMEM-10% FCS). When confluent, medium was removed, and cells were washed in PBS and fixed in 5% formalin for 4 C at 30 min. Cells were rinsed with ultrapure water (Continental Water Systems, Australia) and then incubated with 0.1% l-DOPA in PBS at 37 C for 3.5 h. l-DOPA was then removed, and the cells were fixed in 10% formalin for 60 min at 22 C. Formalin was removed, and cells were air-dried. Positive identification of melanocytes was seen as brown staining when cells were microscopically assessed; fibroblasts did not stain brown.
Melanocyte growth assays Primary melanocytes (n 5 3; passages 6 – 8) were plated at 1 3 104 cells/200 mLzwell on 96-well poly-d-lysine-coated plates (Collaborative Biomedical Products, Bedford, MA). After overnight incubation, MGM-2 was removed and replaced with experimental medium. Spe-
1639
cifically, cells were exposed in triplicate to MGM-2, MBM-0.1% BSA, or increasing doses of growth factors diluted in MBM-0.1% BSA [IGF-I, 0 –100 ng/mL; des(1–3)IGF-I, 0 –100 ng/mL; GH, 0 –100 ng/mL; bFGF, 0 –50 ng/mL]. To assess the effect of combinations of growth factors, only the highest concentration of each was used; IGF-I, des(1–3)IGF-I, GH at 100 ng/mL, or bFGF at 50 ng/mL. Assays were carried out for a period of 15 days, with medium (130 mL of the 200-mL volume) replaced every 3 days. At the end of the experimental period cell number was measured using an XTT colorimetric assay (22).
XTT colorimetric assay XTT was dissolved in prewarmed (60 C) PBS at 1 mg/mL. Phenazine methosulfate (PMS; 100 mmol/L) was prepared and stored in the dark at 220 C. Immediately before use, PMS was added to XTT (1 mg/mL), so that the final concentration was 125 mmol/L. Fresh XTT-PMS (40 mL) mix was added to 200 mL medium in each well of the 96-well plate. Cells were incubated at 37 C in 5% CO2 for 30, 60, 90, and 120 min, and absorbency (cell number) was measured at 450 nm.
Statistical analysis of melanocyte growth assay Experiments were conducted on cells obtained from three primary cell cultures. For statistical analysis of melanocyte growth in response to increasing doses of each growth factor alone, triplicate optical density values (at 450 nm; cell number) were averaged, and data were expressed as the percent increase in absorbance compared to that of cells treated with MBM-0.1% BSA. Data for each dose-response curve were statistically analyzed by one-way ANOVA, using Dunnett’s multiple comparison test. Dose responses for IGF-I and des(1–3)IGF-I at comparable concentrations were compared using unpaired t test. Data for melanocyte growth in response to combinations of growth factors were expressed as the percent increase in absorbance compared to cells treated with MBM-0.1% BSA and were statistically analyzed by unpaired t test. Statistical analyses were determined using Prism GraphPad Software (GraphPad Software, Inc., San Diego, CA).
Identification of secreted IGFBPs by Western ligand blot (WLB) Melanocyte-conditioned medium (1 mL of 15 mL total; CM) was concentrated by ethanol precipitation, and samples were electrophoresed as previously described (7) and according to the method of Laemmli (23). Protein samples were run on a 12% SDS-polyacrylamide gel with 1 mL unconditioned melanocyte medium (negative control), 5 mL human serum (positive control), and 14C-labeled mol wt markers (Amersham). Separated proteins were electrophoretically transferred to nitrocellulose filters and probed with [125I]IGF-II (1 3 106 cpm). Filters were exposed to x-ray film (up to 4 weeks; X-Omat AR, Eastman Kodak Co., Rochester, NY) or a phosphorimaging screen (Storm, Molecular Dynamics, Inc., Sunnyvale, CA). X-Ray autoradiographs were scanned using a Bio-Rad model GS-670 Imaging densitometer (Bio-Rad Laboratories, North Ryde, Australia). IGFBP secretion by cultured melanocytes (n 5 7) was initially assessed using complete growth medium (MGM-2). Specifically, samples were assessed for period (days) of conditioning, confluence of cells, passage number, and donor age.
Identification of secreted IGFBPs by [125I]IGF-I binding and cross-linking analysis Melanocyte CM (2 mL) was precipitated as described previously and resuspended in 220 mL 10 mmol/L Tris-HCl (pH 7.4). Fifty-five microliters of these samples were incubated in a final volume of 60 mL with [125I]IGF-II (50,000 cpm) in the presence or absence of 60 ng IGF-I (competitor) in duplicate for 2 h at 37 C with rotation. Controls included [125I]IGF-II (50,000 cpm) in the absence or presence of 60 ng IGF-I. Samples were then incubated on ice for 10 min followed by the addition of disuccinimidyl suberate to a final concentration of 1 mmol/L and a further incubation for 15 min on ice. The cross-linking reaction was quenched by the addition of 5 mL ice-cold 16 3 quenching buffer (80 mmol/L Tris-HCl (pH 7.4) and 80 mmol/L EDTA (pH 8.0)]. Samples were analyzed by SDS-PAGE (as described for WLB above), then dried under vacuum at 80 C for 2 h and autoradiographed.
1640
EDMONDSON ET AL.
JCE & M • 1999 Vol 84 • No 5
Identification of IGFBPs by immunoprecipitation For further identification of melanocyte-derived IGFBPs, CM was immunoprecipitated with antisera specific for human IGFBP-3 (human anti-IGFBP-3 antiserum, donated by Prof. Robert Baxter, Kolling Institute, Sydney, Australia) and human IGFBP-4 (anti-IGFBP-4 antiserum, Hop, donated by Dr. Steven Chernausek, Children’s Hospital Medical Center, OH). Specifically, 2 mL melanocyte CM were preabsorbed with 320 mL of a 20% suspension of protein A-Sepharose [suspended in Tris-buffered saline, i.e. 25 mmol/L Tris-HCl (pH 7.4) and 0.9% (wt/vol) NaCl containing 0.1% BSA) by gentle agitation for 2 h at 4 C. The protein A-Sepharose was pelleted, and the CM was transferred to a new tube. Antiserum (1:100) and 320 mL of a fresh 20% suspension of protein A-Sepharose were then added to the CM, and samples were incubated overnight with gentle agitation for 2 h at 4 C. Fibroblast CM (500 mL) was used as a positive control. Samples were then centrifuged (5000 rpm at 4 C), and the protein A-Sepharose pellet was washed three times in Tris-buffered saline (1 mL; described above) before dissolving in Laemmli sample buffer, boiled for 5 min, run on SDS-PAGE, and ligand blotted as described previously.
Identification of IGFBPs by RIA Melanocyte CM was subjected to RIA for IGFBP-1, -2, -3, and -6. These assays were performed by Ms. Sara Holman in Prof. Robert Baxter’s laboratory (Kolling Institute, Royal North Shore Hospital, Australia) using previously described assays (24 –27).
Northern blot analysis Total RNA from melanocytes at passages 5– 8 grown in MGM-2 was isolated by the guanidium thiocyanate method (28) and quantified by spectrophotometer at 260 nm. Total RNA (30 mg) was run in a 1.2% agarose-formaldehyde gel as described by Sambrook et al. (29). Samples were transferred to Hybond-N nylon membrane and probed as previously described (11) using [32P]dCTP-labeled human IGFBP-4 cDNA. Membranes were exposed to x-ray film (X-Omat, Eastman Kodak Co., Rochester, NY) for 3–12 days.
Results IGFs dose dependently stimulate melanocyte growth
Melanocyte growth was assessed in response to increasing concentrations of IGF-I, des(1–3)IGF-I, GH, or bFGF. As shown in Fig. 1a, IGF-I stimulated melanocyte growth in a dose-dependent manner. IGF-I at 1, 10, and 100 ng/mL induced growth responses of 38 6 11%, 45 6 11%, and 47 6 16%, respectively, compared to basal medium (P , 0.05). Des(1–3)IGF-I, an analog of IGF-I that has minimal affinity for IGFBPs (30), also induced growth responses at 10 and 100 ng/mL of 26 6 15% and 48 6 16%, respectively, over basal medium (P , 0.05; Fig. 1a). The growth responses to des(1– 3)IGF-I at 0.1 and 1.0 ng/mL were significantly less than those for IGF-I at the same concentrations (Fig. 1a; P , 0.05). Neither GH nor bFGF alone significantly enhanced melanocyte growth (Fig. 1, c and d). GH and bFGF in combination stimulate melanocyte growth and potentiate the effects of IGF-I and des(1–3)IGF-I
Melanocyte growth was also assessed in response to combinations of the following growth factors: IGF-I (100 ng/mL), des(1–3)IGF-I (100 ng/mL), GH (100 ng/mL), and bFGF (50 ng/mL). As shown in Fig. 2, GH or bFGF in combination with IGF-I or des(1–3)IGF-I enhanced melanocyte growth 2- to 4-fold compared to any of the growth factors alone (P , 0.05). The addition of GH to IGF-I and des(1–3)IGF-I induced growth responses of 143 6 36% and 201 6 56%, respectively.
FIG. 1. Melanocyte growth assays: dose-response curves for IGF-I, des(1–3)IGF-I, GH, and bFGF. Melanocytes were grown in the presence of increasing concentrations of each growth factor: a) IGF-I (0 –100 ng/mL; black columns) and des(1–3)IGF-I (0 –100 ng/mL; clear columns), b) GH (0 –100 ng/mL), and c) bFGF (0 –50 ng/mL). Melanocyte cell numbers were determined using an XTT assay as described in Materials and Methods. Cell numbers are expressed as the percent increase in absorbance (at 450 nm) compared with that of melanocytes grown in basal medium. Values represent the mean of triplicate determinations from three separate experiments 6 SEM. *, P , 0.05 compared to basal medium; **, P , 0.05 compared to IGF-I at the same concentration.
bFGF in combination with IGF-I and des(1–3)IGF-I induced growth responses of 303 6 150% and 302 6 119%, respectively. Although GH or bFGF alone did not stimulate melanocyte growth (Figs. 1, c and d, and 2) GH and bFGF in combination were able to induce a significant growth response (198 6 90%) compared to either growth factor alone (P , 0.05). This GH/bFGF-stimulated growth response was not significantly different from that seen with GH or bFGF combined with either IGF-I or des(1–3)IGF-I or a combination of both GH and bFGF with either IGF-I or des(1–3)IGF-I. Complete MGM-2 induced melanocyte growth significantly
GH, IGF-I, AND bFGF IN MELANOCYTE GROWTH
greater than that induced by all growth factors alone or in combination (P , 0.05). IGFBP-4, in low abundance, is the primary IGFBP in CM
Des(1–3)IGF-I (at 10 and 100 ng/mL) was no more effective than IGF-I in stimulating melanocyte growth and was less effective at lower concentrations (0.1 and 1.0 ng/mL; Fig. 1a), raising questions about a role for IGFBPs in IGF action on melanocytes. We, therefore, assessed the secretion of IGFBPs by melanocytes grown in complete growth medium (MGM-2). CM was analyzed for the presence of IGFBPs by WLB as described in Materials and Methods. A major band was seen at 24 kDa, and inconsistent bands were seen at approximately 38 and 42 kDa (Fig. 3). Results were obtained after 4 weeks of autoradiographic exposure, indicating that IGFBP abundance in melanocyte CM was extremely low. The results shown in Fig. 3 are typical for all CM analyzed. The IGFBP profile was not altered by age of donor (newborn to 11 yr), passage number (passages 6 – 8), or cell density. The IGFBP profile was maintained when melanocytes were grown in the
1641
presence of IGF-I (100 ng/mL in MBM-0.1% BSA) or des(1– 3)IGF-I (100 ng/mL in MBM-0.1%BSA) alone (data not shown). To further characterize melanocyte-secreted IGFBPs, CM was subjected to affinity cross-linking with [125I]IGF-II followed by SDS-PAGE (Fig. 4). A specific band at approximately 31 kDa (lanes 1 and 2) was identified that was completely abolished with unlabeled IGF-I (lanes 3 and 4). This 31-kDa band is consistent with IGFBP-4 (24 kDa) cross-linked with IGF-I (7 kDa). Immunoprecipitation was employed to confirm the identity of the 24-kDa band detected in melanocyte CM by WLB. Figure 5 shows results for a typical immunoprecipitation of melanocyte CM followed by WLB. As shown in Fig. 5, the 24-kDa protein was consistently immunoprecipitated from melanocyte CM using anti-IGFBP-4 antiserum (lane 5). AntiIGFBP-3 antiserum did not precipitate any protein from the same melanocyte CM (lane 3), but was able to precipitate IGFBP-3 from human fibroblast CM (lane 7). RIA of melanocyte CM was performed to identify IGFBP-1, -2, -3, or -6. RIA did not detect IGFBP-1, -2, -3, or -6 in any melanocyte CM assayed (data not shown). IGFBP-4 mRNA is detected in cultured human melanocytes
To further confirm that melanocytes synthesize IGFBP-4, total RNA (30 mg) from melanocytes grown to complete confluence in MGM-2 was subjected to Northern analysis. As shown in Fig. 6, the predicted 2.4-kb IGFBP-4 mRNA band (31) was clearly identified in melanocyte RNA (lanes 1 and 2) and was also detected in fibroblast total RNA used as a positive control (lane 3). FIG. 2. Melanocyte growth assays in response to combinations of growth factors. Melanocytes were grown in the presence of the growth factor combinations described [IGF-I, des(1–3)IGF-I, and GH were at 100 ng/mL and bFGF was at 50 ng/mL]. Cell numbers are expressed as the percent increase in absorbance (at 450 nm) compared with that of melanocytes grown in basal medium. Values represent the mean of triplicate determinations from three separate experiments 6 SEM. I, IGF-I; D, des(1–3)IGF-I; G, GH; F, bFGF; MGM, melanocyte growth medium. *, P , 0.05; ✹, P , 0.05.
FIG. 3. Identification of IGFBPs present in melanocyte CM by WLB. Representative WLB autoradiograph of melanocyte CM after exposure to cells at the indicated confluence. The major IGFBP produced by melanocytes is indicated at 24 kDa. HS, Human serum; M, 14C mol wt markers.
FIG. 4. Identification of IGFBPs present in melanocyte CM by crosslinking of [125I]IGF-II. Representative autoradiograph of melanocyte CM (in duplicate) cross-linked with [125I]IGF-II and run on SDSPAGE. Lanes 1 and 2, CM and [125I]IGF-II; lanes 3 and 4, CM, [125I]IGF-II, and IGF-I; lane 5, [125I]IGF-II; lane 6, [125I]IGF-II and IGF-I. A specific band at approximately 31 kDa is indicated.
1642
EDMONDSON ET AL.
FIG. 5. Immunoprecipitation of melanocyte CM to identify the 24kDa IGFBP. Melanocyte CM was immunoprecipitated with specific antisera to human IGFBP-4 or human IGFBP-3, and samples were subjected to WLB and autoradiography. Only IGFBP-4 was identified in melanocyte CM (lane 5). Lane 1, 14C-Labeled mol wt markers; lane 2, human serum; lane 3, blank lane; lane 4, melanocyte CM and anti-IGFBP-3 antiserum; lane 5, melanocyte CM and anti-IGFBP-4 antiserum; lane 6, melanocyte CM without immunoprecipitation; lane 7, human fibroblast CM and anti-IGFBP-3 antiserum; lane 8, human fibroblast CM and anti-IGFBP-4 antiserum; lane 9, human fibroblast CM without immunoprecipitation.
FIG. 6. Identification of IGFBP-4 mRNA by Northern analysis of melanocyte total RNA. Representative Northern blot showing 30 mg total RNA isolated from two melanocyte primary cell cultures (at 90% confluence) and probed with a human IGFBP-4 cDNA. IGFBP-4 mRNA is detected at 2.4 kb. Fibroblast total RNA (30 mg) was used as a control. Lane 1, Passage 5 total RNA; lane 2, passage 6 total RNA; lane 3, fibroblast total RNA.
Discussion
We have investigated the effects of IGF-I, GH, bFGF, and combinations of these growth factors on the growth of normal human melanocytes in vitro. Whereas GH is ineffective alone, melanocyte growth is enhanced when GH is combined with either IGF-I, des(1–3)IGF-I (an analog of IGF-I that has a reduced binding affinity for IGFBPs), or the local growth factor bFGF, either separately or in combination. We have also shown that IGF-I and des(1–3)IGF-I both enhance melanocyte growth in a dose-dependent manner, with IGF-I being more potent, further suggesting a role for IGF-I in melanocyte function. Using WLB, affinity cross-linking, immunoprecipitation, RIA, and Northern analysis, we have shown that cultured human melanocytes produce minimal amounts of IGFBP. IGFBP-4 was identified as the major IGFBP secreted by these cells when cultured in complete growth medium or in the presence of either IGF-I or des(1– 3)IGF-I alone. The response of melanocytes to GH in combination with bFGF or IGF-I suggests that GH plays a role in melanocyte
JCE & M • 1999 Vol 84 • No 5
growth in vitro, consistent with earlier findings of GHR mRNA detected in cultured human melanocytes (12). Our data show that these GHRs are responsive in vitro. If our data reflect the in vivo situation, then the observed increase in the size of melanocytic naevi in patients treated with human GH (17, 18) may be due to the direct action of GH on melanocytes. As GH was unable to elicit a growth response when added alone, our findings suggest that GH action on melanocyte growth does not simply involve local production of IGF-I. The mechanisms involved in GH/IGF-I/bFGF synergy are unknown, but might involve up-regulation of GHRs or other postreceptor processes. bFGF is an important melanocyte growth factor produced by neighboring keratinocytes (32) and dermal fibroblasts (33), but not by melanocytes (32). Under our experimental conditions, bFGF was only able to induce a melanocyte growth response when IGF-I or GH was also present in basal medium. Our finding that bFGF alone did not produce a dose-dependent growth response is perhaps not surprising. Previous studies assessing the proliferative effect of bFGF on cultured melanocytes have always included additional growth factors or phorbol esters in the experimental medium (32, 34). For example, Pittelkow and Shipley (15) demonstrated a synergistic effect of insulin and bFGF, in the presence of phorbol 12-myristate 13-acetate, on melanocyte proliferation. Furthermore, bFGF-induced proliferation of neuronal precursor cells was dependent on the presence of IGF in the medium (35), and this may also be the case with melanocytes. Binding of bFGF to its receptor is exquisitely regulated by specific protoeglycans found in extracellular matrix (36). IGF-I can regulate turnover in articular cartilage explants (37) and proteoglycan synthesis in a chondrocyte cell line (38). It is therefore possible that cultured melanocytes produce extracellular matrix components, such as proteoglycans, in response to IGF-I or GH, which, in turn, augment the bFGF response. It is also possible that intracellular signaling events downstream of receptor activation contribute to synergistic growth stimulation by IGF-I/GH, IGF-I/ bFGF, GH/bFGF, or a combination of all three growth factors. In contrast to GH, IGF-I alone was able to stimulate melanocyte growth in a dose-dependent manner. This is in contrast to a report by Herlyn et al. (39) in which IGF-I alone at 10 ng/mL was unable to sustain melanocyte growth, but is supportive of data showing that 5 mg/mL insulin (probably acting via the IGF-IR) can enhance melanocyte growth (40). The differences in IGF-I responses among all studies may relate to the different formulations of experimental medium. IGF-I activity can be augmented or inhibited by a family of at least six IGFBPs that are produced and regulated in a cell-specific manner (41). Our melanocyte growth studies revealed that des(1–3)IGF-I (which has minimal binding to IGFBPs) was not more effective than IGF-I, suggesting that IGFBPs may play a minor role, if any, in the IGF-I response in vitro. This hypothesis is supported by our analysis of conditioned medium, which clearly showed that melanocytes secrete very low levels of IGFBPs, essentially IGFBP-4. Alternatively, the apparent reduced efficacy of des(1–3)IGF-I compared with that of IGF-I at low concentrations may sug-
GH, IGF-I, AND bFGF IN MELANOCYTE GROWTH
gest that IGFBP-4 augments IGF-I growth stimulation in these cells. This observation would be consistent with recent studies in which IGFBP-4 enhanced IGF-I-stimulated survival of rat neuronal cells (42) and IGFBP-4 knockout mice, which are born smaller than their wild-type counterparts (43). Whereas IGFBP-4 is usually inhibitory, these recent findings and our current results in melanocytes all support the possibility that IGFBP-4 can enhance IGF action in some circumstances. IGFBP-4 is the major IGFBP produced by a range of neuronal cell lines (44), and our finding that IGFBP-4 is the major IGFBP produced by melanocytes is consistent with their neural crest origin. IGFBP-4 production may thus play an important role in the regulation of IGF action in these cell types. IGF-I is a primary mitogen for a number of malignant melanoma cell lines (45). Studies on melanoma cell lines (46, 47) indicate synthesis of a range of IGFBPs (IGFBP-2, -3, -4, -5, and -6), with regulation by serum or IGF-I. It is thus possible that aberrantly expressed IGFBPs play a more significant role in regulating the IGF-I response of melanoma cells. The role of melanocyte-derived IGFBPs in modulating IGF action in vivo remains uncertain, particularly since adjacent basal keratinocytes produce abundant amounts of IGFBP-3 (3, 5, 7, 11). The juxtaposition of melanocytes with basal keratinocytes suggests that IGFBP-3 may also play a significant role in targeting or inhibiting IGF-I interaction with melanocyte IGF-IR. In conclusion, these studies provide support for a role for both GH and IGF-I in the growth of human melanocytes in vitro, involving synergy with bFGF. Low levels of melanocyte-derived IGFBP-4 may enhance IGF action in vitro, which may be further modulated in vivo by other locally produced IGFBPs. References 1. Halaban R, Gosh S, Baird A. 1987 bFGF is the putative natural growth factor for human melanocytes. In Vitro Cell Dev Biol. 23:47–52. 2. Lui JH, Baker J, Perkins A, Robertson E, Efstratiadis A. 1993 Mice carrying null mutations of the genes encoding insulin-like growth factor-I (IGF-I) and type I IGF receptor (IGF-IR). Cell. 75:59 –72. 3. Batch JA, Mercuri FA, Edmondson SR, Werther GA. 1994 Localization of messenger ribonucleic acid for insulin-like growth factor binding proteins in human skin by in situ hybridization. J Clin Endocrinol Metab. 79:1444 –1449. 4. Batch JA, Mercuri FA, Werther G. 1995 Identification and localization of insulin-like growth factor-binding protein (IGFBP) messenger RNAs in human hair follicle dermal papilla. J Invest Dermatol. 106:471– 475. 5. Wraight CJ, Edmondson SR, Fortune DW, Varigos G, Werther GA. 1997 Expression of insulin-like growth factor binding protein-3 (IGFBP-3) in the psoriatic lesion. J Invest Dermatol. 108:452– 456. 6. Wraight CJ, Murashita MM, Russo VC, Werther GA. 1994 A keratinocyte cell line synthesizes a predominant insulin-like growth factor-binding protein (IGFBP-3) which modulates insulin-like growth factor-I action. J Invest Dermatol. 103:627– 631. 7. Murashita M, Russo V, Edmondson S, Wraight C, Werther G. 1995 Identification of insulin-like growth factor binding proteins from cultured human epidermal keratinocytes. J Cell Physiol. 163:339 –345. 8. Camacho-Hubner C, Busby W, McCusker R, Wright G, Clemmons D. 1992 Identification of the forms of insulin-like growth factor binding proteins produced by human fibroblasts and the mechanisms that regulate their secretion. J Biol Chem. 267:11949 –11956. 9. Wraight CJ, Werther GA. 1995 Insulin-like growth factor-I and epidermal growth factor regulate insulin-like growth factor binding protein-3 (IGFBP-3) in the human keratinocyte cell line HaCaT. J Invest Dermatol. 105:602– 607. 10. Yateman ME, Claffey DC, Cwyfan-Hughes SC, Frost VJ, Wass JA, Holly JM. 1993 Cytokines modulate the sensitivity of human fibroblasts to stimulation with insulin-like growth factor-I (IGF-I) by altering endogenous IGF-binding protein production. J Endocrinol. 137:151–159. 11. Edmondson SR, Murashita MM, Russo VC, Wraight CJ, Werther GA. Reg-
12. 13. 14. 15. 16. 17. 18. 19. 20. 21. 22. 23. 24. 25. 26. 27. 28. 29. 30. 31.
32. 33. 34. 35.
36. 37. 38.
39.
1643
ulation of insulin-like growth factor binding protein-3 (IGFBP-3) by EGF and TGFb1 in cultured human keratinocytes. J Cell Physiol, in press. Tavakkol A, Elder JT, Griffiths CEM et al. 1992 Expression of growth hormone receptor, insulin-like growth factor I (IGF-I) and IGF-I receptor mRNA and proteins in human skin. J Invest Dermatol. 99:343–349. Fleming MG, Howe SF, Graf Jr LH. 1994 Expression of insulin-like growth factor I (IGF-I) in nevi and melanomas. Am J Dermatopathol. 16:383–391. Hodak E, Gottlieb AB, Colen S, Anzilotti M, Krueger JG. 1996 In vivo expression of the insulin-like growth factor-I (IGF-I) receptor in congenital pigmented nevi. J Cutaneous Pathol. 23:19 –24. Pittelkow MR, Shipley GD. 1989 Serum-free culture of normal human melanocytes: growth kinetics and growth factor requirements. J Cell Physiol. 140:565–576. Furlanetto RW, Harwell SE, Baggs RB. 1993 Effect of insulin-like growth factor receptor inhibition on human melanomas in culture and in athymic mice. Cancer Res. 53:2522–2526. Bourguignon JP, Pierard GE, Ernould C, et al. Effects of human growth hormone therapy on melanocytic naevi. Lancet. 341:1505–1506. Pierard GE, Pierard-Franchimont C, Nikkels A, Nikkels-Tassouadji N, Arrese JE, Bourgignon JP. 1996 Naevocyte triggering by recombinant human growth hormone. J Pathol. 180:74 –79. Shimasaki S, Ling N. 1991 Identification and molecular characterization of insulin-like growth factor binding proteins (IGFBP-1, -2, -3, -4, -5 and -6). Prog Growth Factor Res. 3:243–266. Halaban R, Alfano FD. 1984 Selective elimination of fibroblasts from cultures of normal human melanocytes. In Vitro. 20:447– 450. Hu DN, McCormick SA, Orlow SJ, Rosemblat S, Lin AY, Wo K. 1995 Melanogenesis by uveal melanocytes in vitro. Invest Opthalmol Visual Sci. 36:931–938. Nargi FE, Yang TJ. 1993 Optimization of the L-M cell bioassay for quantitating tumor necrosis factor a in serum and plasma. J Immunol Methods. 159:81–91. Laemmli UK. 1970 Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature. 227:680 – 685. Baxter RC, Martin JL. 1986 Radioimmunoassay of growth hormone dependent insulin-like growth factor binding protein in human plasma. J Clin Invest. 78:1504 –1512. Baxter RC, Martin JL, Wood MH. 1987 Two immunoreactive binding proteins for insulin-like growth factors in human amniotic fluid: relationship to fetal maturity. J Clin Endocrinol Metab. 65:423– 431. Baxter RC, Saunders H. 1992 Radioimmunoassay of insulin-like growth factor binding protein-6 in human serum and other body fluids. J Endocrinol. 134:133–139. Martin JL, Coverley JA, Baxter RC. 1994 Regulation of insulin-like growth factor binding protein-6 production by neonatal and transformed human fibroblasts. J Biol Chem. 269:11470 –11477. Chomczynski P, Sacchi N. 1987 Single-step method of RNA isolation by acid guanidinium thiocyanate-phenol-chloroform extraction. Anal Biochem. 162:156 –159. Sambrook J, Fritsch EF, Maniatis T. 1989 Molecular cloning: a laboratory manual, 2nd ed. Cold Spring Harbor: Cold Spring Harbor Laboratory. Ballard FJ, Francis, GL, Bagley CJ, Szabo L, Wallace JC. 1989 Effects of insulin-like growth factors on protein metabolism: why are some molecular variants more potent? Biochem Soc Symp. 55:91–104. Sheikh MS, Shao ZM, Hussain A, et al. 1993 Retinoic acid and estrogen modulation of insulin-like growth factor binding protein-4 gene expression and the estrogen receptor status of human breast carcinoma cells. Biochem Biophys Res Commun. 193:1232–1238. Halaban R, Langdon N, Birchall N, et al. 1988 Basic fibroblast growth factor from human keratinocytes is a natural mitogen for melanocytes. J Cell Biol. 107:1611–1619. Thomas RA. 1987 Fibroblast growth factors. FASEB J. 1:434 – 440. Krasagakis K, Garbe C, Zouboulis ChC, Orfanos CE. 1995 Growth control of melanoma cells and melanocytes by cytokines. Recent Results Cancer Res. 139:169 –182. Drago J, Murphy M, Carroll SM, Harvey RP, Bartlett PF. Fibroblast growth factor-mediated proliferation of central nervous system precursors depends on endogenous production of insulin-like growth factor I. Proc Natl Acad Sci USA. 88:2199 –2203. Moscatelli D, Flaumenhaft R, Saksela O. 1991 Interaction of basic fibroblast growth factor with extracellular matrix and receptors. Ann NY Acad. Sci. 638:177–181. Fosang AJ, Tyler JA, Handingham TE. 1991 Effect of interleukin-1 and insulin like growth actor-I on the release of proteoglycan components and hyaluronan from pig articular cartilage in explant culture. Matrix. 11:17–24. Takigawa M, Okawa T, Pan H, et al. 1997 Insulin-like growth factors I and II are autocrine factors in stimulating proteoglycan synthesis, a marker of differentiated chondrocytes, acting through their respective receptors on a clonal human chondrosarcoma-derived chondrocyte cell line, HSC-2/8. Endocrinology. 138:4390 – 400. Herlyn M, Mancianti ML, Jambrosic J, Bolen JB, Koprowski H. 1988 Reg-
1644
40. 41. 42.
43.
EDMONDSON ET AL.
ulatory factors that determine growth and phenotype of normal human melanocytes. Exp Cell Res. 179:322–331. Shipley GD, Pittelkow MR. 1988 Growth of normal human melanocytes in a defined medium. Pigment Cell Res. 1(Suppl):27–31. Jones JI, Clemmons DR. 1995 Insulin-like growth factors and their binding proteins: biological actions. Endocr Rev. 16:3–34. Kummer JL, Zawada WM, Freed CR, Chernausek SD, Heidenreich KA. 1996 Insulin-like growth factor binding proteins in fetal rat mesencephalic cultures: regulation by fibroblast growth factor and insulin-like growth factor I. Endocrinology. 137:3551–3556. Pintar JE, Schuller A, Bradshaw S, Cerro J. Genetic analysis of IGFBP function [Abstract]. Proc of the 4th Int Symp on insulin-Like Growth Factors. Tokyo International Forum. 1997.
JCE & M • 1999 Vol 84 • No 5
44. Chernausek SD, Murray MA, Cheung PT. 1993 Expression of insulin-like growth factor-4 (IGFBP-4) by rat neural cells: comparison to other IGFBPs. Regul Pept. 48:123–132. 45. Rodeck U, Herlyn M, Menssen HD, Furlanetto RW, Koprowski H. 1987 Metastatic but not primary melanoma cell lines grow in vitro independently of exogenous growth factors. Int J Cancer. 40:687– 690. 46. Olney RC, Anhalt AE, Neely EK, Wilson DM. 1995 A quantitative assay for IGF-I and IGF binding protein mRNAs: expression in malignant melanoma cells. Mol Cell Endocrinol. 110:213–223. 47. Bellan C, Remacle-Bonnet M, Garouste F, et al. 1996 Serum factors inhibit melanoma cell surface expression of type I and type II IGF receptors. J Receptor Signal Transduction Res. 16:115–134.
Growth Hormone Research Society Conference September 7–9, 2000 Svenska Ma¨ssan, Go¨teborg, Sweden The GRS was founded in Madrid in October 1992. Its mission is to promote basic and clinical research and scientific exchange in the field of growth hormone and related substances. GRS Council: Raymond Hintz, USA (President); Bengt-Åke Bengtsson, Sweden (Vice-President); Michael Thorner, USA (Treasurer); Jens Sandahl Christiansen, Denmark (Secretary); Lena Carlsson, Sweden; David Clemmons, USA; Pinchas Cohen, USA; Ken Ho, Australia; Primus Mullis, Switzerland; lain Robinson, U.K.; Christian Strasburger, Germany; Toshiaki Tanaka, Japan. We invite abstracts from basic and clinical researchers within the field of growth hormone and related substances such as growth hormone binding proteins, growth hormone releasing factors and peptides, insulin-like growth factors (IGFs) and IGF binding proteins. A common theme for the GRS 2000 meeting will be “Basic and Clinical GH and IGF-1 Research in the Post-Genomic Era”. Deadline for abstracts: March 15, 2000. For information, contact the Conference Secretariat at: GRS 2000, c/o Congrex Go¨teborg AB, Box 5078, SE-402 22 Go¨teborg, Sweden. Tel: 146317088200; Fax: 14631203620; E-mail:
[email protected]; Website: www. congrex.se/grs2000.