Materials and methods

Cell lines and maintenance

Metastatic melanoma cell line WM266-4 obtained from a female patient was purchased from Dainippon Pharmaceutical (Osaka, Japan). HS15 and NM26 were removed from metastatic lymph nodes of male patients, and were established by a monolayer system as described (^{Carey et al, 1976;Baker et al, 1986}). Briefly, biopsy specimens were dissected free of adherent normal tissues, and finely minced. The resulting cell suspensions were washed, resuspendend in culture medium, then inoculated into 35 mm diameter dishes, cultured at 37°C in a humidified atmosphere of 5% CO₂, and fed twice weekly. Confluent culture was trypsinized, and expanded into larger dishes or flasks. The subculture was performed once a week. Human breast cancer MCF-7 cells were purchased from Dainippon. These cell lines were cultured in Dulbecco's modified Eagle medium (Gibco-BRL, Grand Island, NY) supplemented with 10% fetal bovine serum (Gibco-BRL), 1% nonessential amino acids, 1 mM sodium pyruvate (ICN Biomedicals, Aurora, OH), and 100 U per ml penicillin G, 100 g per ml streptomycin, 0.25 g per ml amphotericin B (Gibco-BRL). Human prostate cancer LNCaP cells were purchased from Dainippon, and were maintained in RPMI 1640 (GIBCO-BRL) supplemented with 10% fetal bovine serum.

Reagents

E2, 17-estradiol, progesterone, pregnenolone, DHT, and -dihydrotestosterone were purchased from Sigma (St Louis, MO). ICI 182,780, bicalutamide were obtained from Zeneca Pharmaceuticals (Macclesfield, U.K.). RU486 was from Schering AG (Berlin, Germany). These agents were dissolved in ethanol as 10 mM stock solution and were kept in the dark until used. Recombinant human IL-8 was from Sigma. Recombinant human GRO- was from Pepro Tech EC (London, U.K.). Mouse IgG monoclonal anti-human IL-8 antibody was from BioSource International (Camarillo, CA), and was specific to natural and recombinant human IL-8, with no cross-reactivity to human or mouse GRO-, GRO-, or GRO-. Human recombinant PDGF-AB was from Carbiochem-Novabiochem Corp. (San Diego CA). Recombinant human IL-6 was from Boehringer Mannheim (Indianapolis, IN). Recombinant human bFGF was from Becton Dickinson (San Jose, CA). Control mouse IgG was from Dako Corp. (Carpinteria, CA).

Hormone-binding studies

Melanoma cells were harvested and lyzed. The cytosolic fractions were separated, and used for radiolabeled ligand binding assays as described (^{Feucht et al, 1988;Pottratz et al, 1994}). The protein concentration of cytosolic preparation was determined using a Bio-Rad protein assay kit (Hercules, CA). Aliquots of the fractions were incubated at 4°C for 3 h with 0.05–1.0 nM ³H-E2 (Amersham Corp., Arlington Heights, IL), ³H-R5020 (New England Nuclear, Cambridge, MA), or ³H-R1881 (New England Nuclear) for ER, PR, or AR measurements, respectively. One hundred-fold excess of nonradioactive hormone was used to correct for nonspecific binding. Bound and free hormones were separated using dextran-coated charcoal, and specific binding was determined as described (^{Feucht et al, 1988}). Binding constants, K_d and B_max, were calculated according to the method of Scatchard (^{Scatchard, 1949}).

Western blot

Aliquots (50 g protein) of the extracts from melanoma, MCF-7, and LNCaP cells were resolved by sodium dodecyl sulfate–polyacrylamide (8%) gel electrophoresis under nonreducing conditions. The proteins were electrotransferred on to nitrocellulose membranes as described (^{Zhu et al, 1999;Swami et al, 2000}). After blocking with 5% nonfat dry milk, membranes were probed for 2 h with 10 g per sheet primary antibodies: mouse monoclonal IgG1 anti-ER C-314 (Santa Cruz Biotechnology, Santa Cruz, CA), rabbit polyclonal IgG anti-PR H-190 (Santa Cruz Biotechnology), or rabbit polyclonal IgG anti-AR N-20 (Santa Cruz Biotechnology). The blots were washed and incubated for 60 min with second antibodies: peroxidase-conjugated goat anti-mouse IgG or anti-rabbit IgG (Pierce, Rockford, IL). After three washes, immunoreactive bands were detected using an enhanced chemi luminescence kit (Amersham).

Proliferation assays

WM266-4, HS15, and NM26 cells were plated at 5 10³ cells per well in triplicate in flat-bottomed 96-well plates in 100 l of culture medium and adhered for 18 h. The medium was discarded and the plates were washed with phosphate-buffered saline three times, then incubated with sex hormones and/or recombinant cytokines at the indicated concentrations in 100 l per well of serum-free, phenol red-free Dulbecco's modified Eagle medium for 20 h. Then 0.5 Ci per well of ³H-thymidine (Amersham) was added and the cells were incubated for an additional 4 h prior to harvest. The incorporation of ³H-thymidine was assayed by liquid scintillation.

Measurement of cytokines

To measure the secretion of cytokines, melanoma cells were plated in triplicate at 5 10⁴ per well in 24-well plates, adhered overnight, washed, and incubated with sex hormones in 1 ml per well of phenol red-free, serum-free medium for 24 h, and the culture supernatants were harvested and stored at -70°C until used. The activity of IL-6 and IL-8 in the supernatants was measured by an enzyme-linked immunosorbent assay (ELISA) kit (Biosource), that of GRO- and bFGF was measured by ELISA kits (R&D Systems, Minneapolis, MN), and that of PDGF-AB was measured by ELISA kit (Genzyme Techne, Minneapolis, MN), according to the manufacturers' instructions. The sensitivity of the assay for IL-6, IL-8, GRO-, bFGF, or PDGF-AB was 2, 5, 10, 3, or 8.4 pg per ml, respectively.

For the analysis of cytokine mRNA expression, northern blot was performed as described (^{Zachariae et al, 1991}). Melanoma cells were incubated with sex hormones as above for 8 h, and were harvested. Total RNA was extracted from the harvested cells by guanidium isothiocyanate method (Ultraspec, Houston, TX). RNA (25 g) was electrophoresed in a 1% agarose gel and was transferred to nitrocellulose membranes. Blots were hybridized with a ³²P-labeled cDNA probe for human IL-8, which was 0.45 kb EcoRI–EcoRI fragment (^{Mukaida et al, 1989}). To control for differences in RNA sample loading and transfer, the blots were also hybridized with a ³²P-labeled, 1.8 kb HindIII–HindIII fragment of the gene encoding human -actin. The membranes were exposed to X-ray films (Hyperfilm MP; Amersham) for 17 h at -80°C. Autoradiograms were scanned using a Molecular Dynamics computing densitometer (model 300 A; Molecular Devices, Menlo Park, CA). IL-8 mRNA levels were normalized to those of -actin.

Plasmids and transfection

pCAT3-basic vector carrying two SV40 poly(A) signals, one downstream of the chloramphenicol acetyl transferase (CAT) reporter gene, and the other upstream of the multicloning site was purchased from Promega (Madison, WI). The HincII–HindIII fragment of the genomic IL-8 DNA, which spans nucleotides -546 to +44 bp relative to the transcriptional start site (^{Mukaida et al, 1989}), was subcloned into PUC19, treated with appropriate restriction endonucleases and further subcloned into pCAT3-basic vector (^{Dorn and Derse, 1988;Mukaida et al, 1989}). cDNA encoding human ER (1.8 kb) (^{Green et al, 1986}), PR (4.4 kb) (^{Kastner et al, 1990}), and AR (3.2kb) (Brinkmann et al, 1989) were subcloned into EcoRI site of pSG5 (Stratagene, La Jolla, CA), downstream of T7 promoter as described (^{Green et al, 1988}) to produce expression plasmids for ER, PR, and AR, respectively. The carboxyl-terminal-truncated mutant plasmids for ER, PR, and AR were generated by cutting the wild-type plasmids with restriction enzymes, inserting oligonucleotides with translation stop codons, and religating the vectors as described (^{Kastner et al, 1990;Simental et al, 1991;Stein and Yang, 1995}). Transfection of melanoma cells was carried out by the calcium coprecipitation method using CellPhect transfection kit (Amersham) according to the manufacturer's instruction. Briefly, 1 d before the experiment, confluent cultures of melanoma cells were trypsinized, and cells were seeded at 1 10⁶ cells per 100 mm diameter dish and incubated overnight at 37°C. The cultures were replenished with fresh medium and kept at 37°C for 4 h before transfection. The cells were incubated for an additional 4 h with the DNA-calcium precipitate containing 10 g pIL-8 CAT. In some experiments, the cells were cotransfected with either 5 g of empty vector pSG5 or 5 g of wild-type or mutant hormone receptor expression plasmid together with pIL-8 CAT. The cultures were then glycerol shocked and replenished with fresh medium. After 3 h, the cells were trypsinized and subdivided into 24-well plates at 5 10⁴ cells per well and incubated in 1 ml per well of culture medium overnight. This procedure eliminates differences in transfection efficiency as the same construct is used for the transfection of separate cultures. Then the medium was discarded and the cells were washed with phosphate-buffered saline, and maintained in serum-free, phenol red-free medium in the presence or absence of sex hormones at indicated concentrations. After 24 h, the cells were harvested and lyzed by three freeze/thaw cycles. The cell lysate was centrifuged and supernatant was assayed for CAT expression by CAT-ELISA (Roche Diagnostics, Tokyo, Japan) according to the manufacturer's instructions. Total protein amount was measured by a Bio-Rad protein assay kit. The CAT expression was presented as picograms of CAT enzyme synthesized per microgram of total protein. pCAT3-control vector (Promega) containing SV40 early promoter and enhancer sequences was used as a positive control, and promoterless pCAT3-basic vector was used as a negative control. The expression of wild-type or mutant hormone receptors was analyzed by Western blot using cell lysate and specific antibodies. The hormone-binding activity of the expressed receptors was analyzed by radiolabeled ligand binding assays.

Statistical analyses

One-way analysis of variance with Dunnet's multiple comparison test was used for the data in Figure 2. One-way analysis of variance with Scheffe's multiple comparison test was used for the data in Figure 3, Figure 5, Figure 7, and Figure 8, and Table II.

Figure 2.

The inhibitory effects of E2, progesterone, and DHT on the proliferation of melanoma cell lines. (a, d) E2, (b, e) progesterone, (c, f) DHT. WM266-4, HS15, and NM26 cells were incubated with E2, progesterone, or DHT at the indicated concentrations and were pulsed with ³H-thymidine before harvesting as described in Materials and Methods. The data are shown as percentage vs ³H-thymidine uptake of control cultures with medium alone. In parallel experiments, 17-estradiol (a), pregnenolone (b), or -dihydrotestosterone (c) was added to the melanoma cells, and ³H-thymidine uptake was analyzed. In (d–f), melanoma cells were treated with E2 (10^-9 M) in the presence or absence of indicated concentrations of ICI 182,780, or treated with progesterone (10^-9 M) in the presence or absence of indicated concentrations of RU486, or treated with DHT (10^-8 M) in the presence or absence of indicated concentrations of bicaltamide. Results represent the mean SEM of four separate experiments. *p < 0.05 vs control cultures with medium alone, by analysis of variance with Dunnet's multiple comparison test in (a–c). *p < 0.05 vs cultures with E2, progesterone, or DHT alone, by analysis of variance with Dunnet's multiple comparison test in (d–f). The values of ³H-thymidine uptake in control cultures with medium alone were mean SEM (n = 4) 18,536 1523, 15,231 1324, and 10,324 1094 cpm for WM266-4, HS15, and NM26 cells, respectively.

Full figure and legend (72K)

Figure 3.

The inhibitory effects by combination of E2, progesterone, or DHT on the proliferation of melanoma cell lines. WM266-4 (a), HS-15 (b), or NM26 (c) cells were cultured with the indicated combination of E2 (10^-9 M), progesterone (10^-9 M), or DHT (10^-8 M) before examining ³H-thymidine uptake. The data are shown as percentage vs ³H-thymidine uptake of control cultures with medium alone. Results represent the mean SEM of four separate experiments. *p < 0.05 vs control cultures, by analysis of variance with Scheffe's multiple comparison test. The values of ³H-thymidine uptake in control cultures were mean SEM (n = 4) 19,236 1523, 15,231 1224, and 10,354 1194 cpm for WM266-4, HS15, and NM26 cells, respectively.

Full figure and legend (59K)

Figure 5.

The inhibitory effects by combination of E2, progesterone, and/or DHT on the proliferation of melanoma cell lines. WM266-4 (a), HS-15 (b), or NM26 cells (c) were cultured with the indicated combination of E2 (10^-9 M), progesterone (10^-9 M), and/or DHT (10^-8 M) for 24 h, and IL-8 secretion in culture supernatants was analyzed by ELISA. The data are shown as percentage vs IL-8 secretion in control cultures with medium alone. Results represent the mean SEM of four separate experiments. *p < 0.05 vs control cultures, by analysis of variance with Scheffe's multiple comparison test. The amounts of IL-8 secretion in control cultures were mean SEM (n = 4) 8.1 1.2, 7.6 0.9, and 5.2 0.6 ng per ml for WM266-4, HS15, and NM26 cells, respectively.

Full figure and legend (67K)

Figure 7.

The inhibitory effects of E2, progesterone, or DHT on IL-8 promoter activities in melanoma cells. WM266-4 (a), HS15 (b), and NM26 cells (c) were transiently transfected with IL-8 promoter CAT reporter constructs and cultured with indicated combinations of E2 (10^-9 M), progesterone (10^-9 M) or DHT (10^-8 M). The IL-8 promoter activity was assessed by CAT expression of the cell lysate. The data are shown as percentage vs CAT expression in control cultures with medium alone. Results represent the mean SEM of four separate experiments. *p < 0.05 vs control cultures, by analysis of variance with Scheffe's multiple comparison test. The CAT expression in control cultures was mean SEM (n = 4) 32.4 3.5, 30.6 2.9, and 20.3 1.8 pg CAT per g protein for WM266-4, HS15, and NM26, respectively. The CAT expression in pCAT3-control vector was 130 11, 151 14, and 145 13 pg CAT per g protein for WM266-4, HS15, and NM26 cells, respectively. The CAT expression by promoterless pCAT3-basic vector was less than detectable level in all three cell lines.

Full figure and legend (70K)

Figure 8.

The inhibition by E2, progesterone, or DHT on IL-8 promoter activity in wild-type or mutant ER, PR, or AR-transfected melanoma cells. WM266-4 (a), HS15 (b), or NM26 (c) cells were cotransfected with IL-8 promoter CAT reporter construct and either empty vector pSG5 or wild type (WT) or hormone-binding domain-deleted mutant (MU) AR (a), ER (b), or PR (c) expression vector, and incubated with or without DHT 10^-8 M (a), E2 10^-9 M (b), or progesterone 10^-9 M (c). The IL-8 promoter activity was assessed by CAT expression of the cell lysate. Results represent the mean SEM of four separate experiments. *p < 0.05 vs wild-type receptor-transfected cells cultured with medium alone, by analysis of variance with Scheffe's multiple comparison test. The transfected wild-type and mutant AR (d), ER (e), and PR (f) were verified for expression by western blot. The structure of wild-type and mutant receptors is presented in (g–i). Regions A/B, C, D, and E represent the amino-terminal, DNA-binding, hinge, and the hormone-binding domains, respectively. The number of amino acid residues and restriction enzyme-cut sites are indicated.

Full figure and legend (71K)

Table II - The effects of E2, progesterone, or DHT on the secretion of cytokines other than IL-8 in melanoma cells.

Full table

Results

ER, PR, and AR expression in melanoma cells

First, the presence of sex hormone receptors was analyzed in three different metastatic melanoma cell lines. As examined by binding assays, WM266-4 expressed ER and PR, but not AR Table I. HS15 expressed PR and AR, but not ER. NM26 expressed ER and AR, but not PR. Thus the presence or absence of each hormone receptor differed among cell lines; however, the affinity of ER, PR, or AR as expressed as a K_d value, was similar between each receptor-positive cell line. The results in western blot Figure 1 were consistent with those in binding assays; a 65 kDa band for ER protein was detected in WM266-4 and NM26 but not in HS15. A 114 kDa band for PR was detected in WM266-4 and HS15, and not in NM26. A 110 kDa band for AR was detected in HS15 and NM26, but not in WM266-4. We next examined if each hormone may alter the growth of melanoma cell lines in a receptor-dependent manner.

Figure 1.

Western blot analysis for ER, PR, and AR in melanoma cells. The lysates from melanoma WM266-4, HS15, or NM26 were analyzed for the expression of hormone receptors. The ER, PR, and AR were identified as 65 kDa, 114 kDa, and 110 kDa band, respectively. The positive controls were the lysates from human breast cancer MCF-7 cells for ER and PR, and those from prostate cancer LNCaP cells for AR.

Full figure and legend (18K)

Table I - The presence of ER, PR, and AR in metastatic melanoma cell lines.

Full table

The inhibitory effects of E2, progesterone, and DHT on the growth of melanoma cells

As shown in Figure 2(a), E2 inhibited ³H-thymidine uptake of ER-positive WM266-4 and NM26 in a concentration-dependent manner, but not that of ER-negative HS15; percentage inhibition vs control by 10^-9 M of E2 was 52%, 48%, and 2% in WM266-4, NM26, and HS15, respectively. E2 stereoisomer 17-estradiol, known to be biologically inactive as an estrogen, did not alter the growth of any melanoma cell lines. In addition, ER antagonist ICI 182,780 counteracted the growth-inhibitory effect of E2 in both ER-positive WM266-4 and NM26 cells in a concentration-dependent manner, but did not alter the growth of ER-negative HS15 treated with E2 Figure 2d. These results suggest that E2 may inhibit the growth of melanoma cells dependently on the binding to ER. Progesterone also inhibited ³H-thymidine uptake of PR-positive WM266-4 and HS15 but not that of PR-negative NM26 Figure 2b; percentage inhibition by 10^-9 M of progesterone was 47%, 45%, and 3% in WM266-4, HS15, and NM26, respectively, and the growth-inhibitory effect was not manifested by the progesterone stereoisomer, pregnenolone. The PR antagonist RU486 counteracted the growth inhibition by progesterone in both WM266-4 and HS15 cells in a concentration-dependent manner Figure 2e. DHT also inhibited ³H-thymidine uptake of AR-positive NM26 and HS15 in a concentration-dependent manner, but not that of AR-negative WM266-4; percentage inhibition by 10^-8 M of DHT was 49%, 46%, and 2% in NM26, HS15, and WM266-4, respectively, and the growth-inhibitory effect was not mediated by DHT stereoisomer, -dihydrotestosterone. The AR antagonist bicaltamide counteracted the growth inhibition by DHT in both HS15 and NM26 cells in a concentration-dependent manner Figure 2f. As compared with the other two hormones, the dose–response curve for DHT shifted to the right by 1 order, which may reflect the lower affinity of AR, i.e., larger K_d value Table I. As 10^-9 M for E2 and progesterone, and 10^-8 M for DHT appeared to be the optimal concentrations for growth inhibition, respectively, these concentrations were used in further experiments.

We next examined if the combination of different hormones may give additive or synergistic inhibition on the growth of melanoma cells. In ER and PR-positive and AR-negative WM266-4 Figure 3a, the inhibitory effect by combination of E2 and progesterone was not higher than that of either hormone alone. The addition of DHT did not alter the inhibitory effect of E2 or progesterone alone, or that of E2 plus progesterone. Similar results were obtained in HS15 and NM26 Figure 3b, c. Thus the combination of more than two different hormones neither gave additive nor synergistic inhibition, indicating that the mechanism for the growth inhibition may be common to E2, progesterone, and DHT, and that the response of melanoma may be saturated by an optimal concentration of either hormone.

The counteracting effects of IL-8 on growth inhibition by E2, progesterone, and DHT

As metastatic melanoma cells constitutively produce various growth factors and autonomously proliferate through such growth factors, it is indicated that the growth-inhibitory effects of E2, progesterone, and DHT may be mediated by inhibiting the autocrine growth factor production in melanoma. To identify the cytokine(s) that may be involved in the growth inhibition by sex hormones, various cytokines reported as autocrine growth factors of melanoma (^{Herlyn, 1990}) are added to the melanoma cells cultured with hormones, and their influence on E2, progesterone, or DHT-induced growth inhibition was examined. Among various cytokines, only IL-8 overcame the growth inhibition by E2 in WM266-4 Figure 4a. When endogenous IL-8 activity was neutralized by anti-IL-8 antibody, the ³H-thymidine uptake of WM266-4 was reduced by 70% compared with control, indicating that IL-8 may act as an autocrine growth factor in WM266-4. In the presence of this antibody, E2 did not further reduce the ³H-thymidine uptake of WM266-4. Similar results were obtained in NM26 cultured with E2 (data not shown). These results suggest that the growth-inhibitory effect of E2 may be mediated via IL-8 in WM266-4 and NM26. As shown in Figure 4(b, c), IL-8 overcame the growth inhibition by progesterone in HS15 and that by DHT in NM26. Anti-IL-8 antibody reduced the ³H-thymidine uptake of both cell lines, suggesting that IL-8 may act as an autocrine growth factor in both cell lines. When endogenous IL-8 was neutralized by anti-IL-8 antibody, the growth-inhibitory effects of progesterone and DHT were not detected in HS15 and NM26, respectively. Similar results were obtained in WM266-4 cultured with progesterone and in HS15 cultured with DHT (data not shown). These results suggest that IL-8 may be involved in the growth-inhibitory effects of progesterone in HS15 and WM266-4 and those of DHT in NM26 and HS15. We then examined if E2, progesterone, or DHT may alter IL-8 production in the three different melanoma cell lines.

Figure 4.

Cytokine-induced recovery from the inhibitory effects of E2, progesterone, or DHT on the proliferation of melanoma cells. WM266-4 (a), HS15(b), or NM26 cells (c) were cultured in the presence or absence of E2 (10^-9 M) (a), progesterone (10^-9 M) (b), or DHT (10^-8 M) (c) with or without various cytokines (each 5 ng per ml) before examining ³H-thymidine uptake. In some experiments, the culture of melanoma was performed in the presence of mouse anti-IL-8 antibody or control mouse IgG (each 10 g per ml). The data are shown as percentage vs ³H-thymidine uptake of control cultures with medium alone. Results represent the mean SEM of four separate experiments. *p < 0.05 vs control cultures and †p < 0.05 vs cultures of E2, progesterone, or DHT alone, by analysis of variance with Scheffe's multiple comparison test. The values of ³H-thymidine uptake in control cultures were mean SEM (n = 4) 18,289 1626, 15,253 1329, and 11,385 1096 cpm for WM266-4, HS15, and NM26 cells, respectively.

Full figure and legend (83K)

The effects of sex hormones on IL-8 production in melanoma

WM266-4, HS15, and NM26 constitutively secreted large amounts of IL-8 (5.2–8.1 ng by 5 10⁴ cells). E2 reduced IL-8 secretion in ER-positive WM266-4 and NM26 by 62.5% and by 60% compared with controls, respectively, whereas it did not alter that in ER-negative HS15 Figure 5. Progesterone also reduced IL-8 secretion in PR-positive WM266-4 and HS15 by 56.3% and by 61.6%, respectively, and did not affect that in PR-negative NM26. DHT also reduced IL-8 secretion in AR-positive HS15 and NM26 by 65.3% and by 56%, respectively, and did not alter that in AR-negative WM266-4. In parallel with the growth inhibition, anti-estrogen ICI 182,780 counteracted the E2-induced inhibition of IL-8 production in WM266-4 and NM26 cells (data not shown), and similar results were obtained for anti-progesterone RU486 on progesterone-induced inhibition, and anti-androgen bicaltamide on DHT-induced inhibition of IL-8 production (data not shown). Thus E2, progesterone, or DHT inhibited IL-8 secretion in melanoma cells in a respective receptor-dependent manner. In each melanoma cell line, the combination of more than two different sex hormones neither gave additive nor synergistic inhibition on the IL-8 secretion, which correlated with the hormone-induced inhibition of ³H-thymidine uptake Figure 3. Though WM266-4, HS15, and NM26 constitutively secreted large amounts of GRO-, none of E2, progesterone, and DHT altered the secretion of GRO- in any cell lines. The secretion of PDGF-AB, bFGF, and IL-6 were lower than that of IL-8 or GRO-, and were not affected by any hormones in any cell lines Table II.

We then analyzed if IL-8 mRNA expression in melanoma cells may be altered by E2, progesterone, or DHT. As analyzed by northern blot (Figure 6a, b), WM266-4, HS15, and NM26 constitutively expressed IL-8 mRNA, and the expression was suppressed by sex hormones in a respective receptor-dependent manner; E2 reduced steady-state IL-8 mRNA levels of ER-positive WM266-4 and NM26 and did not affect that of ER-negative HS15. Progesterone reduced IL-8 mRNA levels of PR-positive WM266-4 and HS15, but did not alter that of PR-negative NM26. DHT reduced IL-8 mRNA levels of AR-positive HS15 and NM26, and did not affect that of AR-negative WM266-4. Thus the inhibitory effects of E2, progesterone, and DHT on IL-8 mRNA expression closely correlated with those on IL-8 protein secretion Figure 5, suggesting the pretranslational inhibition. We then examined if E2, progesterone, and DHT may exert their effects at the transcriptional level by analyzing the effects of the hormones on the activity of IL-8 promoter.

Figure 6.

Northern blot analysis for the effects of E2, progesterone, or DHT on IL-8 mRNA expression in melanoma cells. WM266-4, HS15, or NM26 cells were cultured with E2 (10^-9 M), progesterone (10^-9 M), or DHT (10^-8 M) for 8 h. Cells were then harvested and total RNA (25 g) from each sample was subjected to northern blot analysis and probed for IL-8 and -actin mRNA (a) and the intensity of the bands were analyzed by densitometry, and the intensity ratio was calculated (b). The data are representative of four different experiments.

Full figure and legend (27K)

The effects of E2, progesterone, and DHT on CAT expression driven by IL-8 promoter

WM266-4, HS15, and NM26 were transiently transfected with plasmid containing human IL-8 promoter driving CAT reporter gene, and the promoter activity was assessed by the expression of CAT enzyme. Though it is known that IL-8 promoter activity is upregulated by IL-1 or phorbol 12-myristate 13-acetate (^{Mukaida et al, 1989}), the melanoma cell lines showed considerable levels of CAT expression driven by IL-8 promoter (20.3–32.4 pg CAT per g protein) even in unstimulated state. As shown in Figure 7, the constitutive activity of IL-8 promoter in each melanoma cell line was inhibited by E2, progesterone, or DHT in a respective receptor-dependent manner; E2 reduced IL-8 promoter activity of ER-positive WM266-4 Figure 7a and NM26 Figure 7c by 65% and by 60% compared with controls, respectively, and did not alter that of ER-negative HS15 Figure 7b. Progesterone reduced IL-8 promoter activity of PR-positive WM266-4 and HS15 by 61% and by 60%, respectively, but did not affect that of PR-negative NM26. DHT reduced IL-8 promoter activity of AR-positive HS15 and NM26 by 62% and by 58%, respectively, and did not affect that of AR-negative WM266-4. The combination of more than two different sex hormones neither gave additive nor synergistic inhibition on IL-8 promoter activity in each melanoma cell line, which was consistent with the sex hormone-induced inhibition of IL-8 secretion Figure 5. These results suggest that E2, progesterone, and DHT may inhibit IL-8 production at the transcriptional level in a respective receptor-dependent manner. To confirm that the hormone-induced transcriptional inhibition may be mediated via the respective hormone receptor, we transiently cotransfected ER, PR, or AR-negative melanoma with the respective receptor expression plasmid together with IL-8/CAT construct, and examined if E2, progesterone, or DHT may inhibit the IL-8 promoter activity in the respective receptor-transfected cells. Transient transfection of AR-negative WM266-4 with wild-type AR vector resulted in the measurable ³H-R1881 binding; K_d = 1.1 nM and B_max = 17.3 fmol protein per mg. On the other hand, WM266-4 transfected with hormone-binding domain-deleted mutant AR Figure 8g did not show significant binding to ³H-R1881. By western blotting, the expression of wild-type and mutant AR was confirmed as 114 kDa and 87 kDa band, respectively Figure 8d. The IL-8 promoter activity in wild-type AR-transfected WM266-4 was not reduced compared with empty vector pSG5-transfected cells Figure 8a, suggesting that the overexpression of wild-type AR itself may not inhibit IL-8 promoter activity. In the wild-type AR-transfected WM266-4 Figure 8a, however, DHT reduced IL-8 promoter activity by 71.4% compared with control cultured with medium alone whereas in WM266-4 transfected with hormone-binding domain-deleted mutant AR, DHT did not reduce the IL-8 promoter activity. Thus wild-type AR, transfected into melanoma, could repress IL-8 promoter activity in the presence of DHT, and the repression required the binding of DHT to AR. Similarly, transient transfection of ER-negative HS15 with wild-type ER resulted in the measurable ³H-E2 binding; K_d = 0.4 nM and B_max = 15.1 fmol per mg protein. On the other hand, HS15 transfected with hormone-binding domain-truncated mutant ER Figure 8h did not show significant binding to ³H-E2. By western blotting, the expression of wild-type and mutant ER was confirmed as 66 kDa and 37 kDa band, respectively Figure 8e. In wild-type ER-transfected HS15 Figure 8b, E2 inhibited IL-8 promoter activity by 63.6% compared with control, but did not alter that in HS15 transfected with hormone-binding domain-deleted mutant ER, indicating that E2-induced transrepression was dependent on E2-binding to ER. Transient transfection of PR-negative NM26 with wild-type PR resulted in the measurable ³H-R5020 binding; K_d = 0.45 nM and B_max = 16.5 fmol protein per mg. On the other hand, NM26 transfected with hormone-binding domain-truncated mutant PR Figure 8i did not show significant binding to ³H-R5020. By western blotting, the expression of wild-type and mutant PR was confirmed as 110 kDa and 79 kDa band, respectively Figure 8f. Progesterone inhibited IL-8 promoter activity by 68.2% compared with control in the wild-type PR-transfected NM26, but did not affect that in NM26 transfected with hormone-binding domain-truncated mutant PR, indicating that progesterone-induced transrepression was dependent on progesterone binding to PR. Figure 8c. These results suggest that the transcriptional repression by E2, progesterone, and DHT may be specifically mediated by binding to the respective receptor.

To characterize the DNA sequences involved in hormone-induced transcriptional repression on the IL8 gene, we used CAT expression plasmids linked to serially 5'-deleted IL-8 promoters, and compared the effects of hormones on the activities of the deleted promoters. In transfected WM266-4 cells, the basal promoter activity of -272 CAT (containing -272 to +44 bp of IL-8 promoter) was not different from that of -546 CAT, and both were repressed by E2 Figure 9a. The basal promoter activity of - 98 CAT was reduced as compared with that of -546 or -272 CAT, indicating that the sequences between -272 to -99 bp may partially direct the constitutive promoter activity. The promoter activity of -98 CAT, however, was still repressed by E2; percentage inhibition by E2 was 61% and 61.5% in -272 and -98 CAT-transfected cells, respectively. These results indicate that the sequences up to -98 bp may be dispensable for the transcriptional repression by E2. When WM266-4 was transfected with -62 CAT, the basal promoter activity was further reduced, and was not at all repressed by E2. These results suggest that the sequences between -98 and -63 bp may partially direct the constitutive IL-8 transcription and may be mainly involved in the E2-induced transcriptional repression. Similar results were obtained in the 5'-deleted IL-8/CAT-transfected and E2-incubated NM26 cells (data not shown). In the 5'-deleted IL-8/CAT-transfected and progesterone-treated HS15 Figure 9b or DHT-treated NM26 Figure 9c, the deletion up to -98 bp of IL-8 promoter preserved the repression by progesterone or DHT, respectively, whereas deletion up to -62 bp resulted in the complete loss of the repression. Similar results were obtained in the 5'-deleted IL-8/CAT-transfected and progesterone-treated WM266-4 or DHT-treated HS15 (data not shown). These results suggest that the sequences between -98 and -63 bp are mainly involved in the hormone-induced repression of IL-8 transcription in melanoma.

Figure 9.

The influence of 5'-deletion on E2, progesterone, or DHT-induced inhibition of IL-8 promoter activity. WM266-4 (a), HS15 (b), or NM26 cells (c) were transfected with CAT reporter vectors into which various 5'-deleted IL-8 promoters were cloned. After transfection, the cells were incubated with or without E2 (10^-9 M) (a), progesterone (10^-9 M) (b), or DHT (10^-8 M) (c) for 24 h. CAT expression was then analyzed.

Full figure and legend (38K)

Discussion

This study demonstrated that E2, progesterone, and DHT inhibited the growth of metastatic melanoma WM266-4, HS15, and NM26 by inhibiting IL-8 transcription. The inhibitory effect of each hormone was mediated via the respective receptor. The growth of the other melanoma cell lines was also receptor-dependently inhibited by sex hormones via the suppression of IL-8 production; among melanoma TO 23, HO 14, and YN 115 established in our laboratory, the presence of ER and E2-induced growth inhibition via IL-8 were only seen in TO 23, but not in the others. The presence of PR and progesterone-induced growth inhibition via IL-8 were only seen in HO 14 and not in the others. The presence of AR and DHT-induced growth inhibition via IL-8 were only seen in YN 115 and not in the others (data not shown). The previous study also reported that progesterone inhibited basal and IL-1-induced IL-8 mRNA expression in rat uterine fibroblasts (^{Ito et al, 1994}); however, it is unknown how E2, progesterone, or DHT may suppress IL-8 transcription. It is likely that the three different hormones may exert their effects by a common mechanism as the combination of more than two hormones gave neither synergistic nor additive effects. One possible mechanism is that each hormone-bound receptor may inhibit the transcriptional activity and/or DNA binding of transcriptional activators on the IL-8 promoter. It is known that several transcriptional activators co-operate with one another through protein–protein interaction to activate the IL-8 promoter (^{Roebuck, 1999}). The sequences from -98 to -63 bp appeared to direct the constitutive IL-8 transcription in melanoma, and may be the main target for hormone-induced transcriptional repression. This region includes adjacent elements for C/EBP (-94 to -81) and NF-B (-80 to -70). It is reported that NF-B subunit p65 and C/EBP form a ternary complex with this region of the IL-8 promoter, which results in synergistic transcriptional activation (^{Kunsch et al, 1994}). It is thus hypothesized that each sex hormone/receptor complex may interact with NF-B and/or C/EBP family proteins, and alter their conformation, which may repress their DNA-binding and/or the cooperative transcriptional activity on IL-8 promoter. The previous studies reported the direct protein–protein interaction of ER with NF-B p65 and with C/EBP(^{Stein and Yang, 1995}). PR and AR directly interacted with NF-B p65 and repressed its transcriptional activity on several gene promoters (^{Kalkhoven et al, 1996;McKay and Cidlowski, 1999}). E2 and DHT also inhibited the transcriptional activity of NF-B via IkB (^{Keller et al, 1996;Sun et al, 1998}) IkB is constitutively associated with NF-B in cytosol and inhibits its translocation to the nucleus where DNA binding occurs. E2 and DHT repressed the degradation of IkB, and maintained the IkB protein level, and thus inhibited the transcriptional activity of NF-B. As dexamethasone and 1,25-dihydroxyvitamin D₃ suppressed IL-8 transcription by inhibiting the DNA binding of NF-B in a glioblastoma cell line (^{Mukaida et al, 1994;Harant et al, 1997}), a similar mechanism is implicated for E2, progesterone, or DHT. DNA-binding activity and mRNA levels of NF-B subunits p65 and p50 are constitutively enhanced in melanomas compared with normal melanocytes (^{Meyskens et al, 1999}), which may be related to the constitutively high level of IL-8 production in melanoma. Taken together, it is hypothesized that NF-B may be one of the main targets for the sex hormone-induced transcriptional repression. It is also likely that C/EBP may be the target. C/EBP is known to mainly mediate the constitutive IL-8 transcription without any stimuli (^{Wu et al, 1997}). DHT-bound AR also suppressed the C/EBP-mediated transcription on rat dehydroepiandrosterone sulfotransferase promoter (^{Song et al, 1998}). In the present study, the deletion of - 272 to - 99 bp on IL-8 promoter greatly reduced the constitutive promoter activity Figure 9. As this region contains AP-1 element (-126 to -120) (^{Mukaida et al, 1989}), AP-1 may partially confer the constitutive IL-8 transcription in melanoma. A previous study also reported that the constitutive DNA-binding activity of AP-1 was higher in melanomas than in melanocytes (^{Meyskens et al, 1999}). It is known that AP-1 as well as C/EBP physically interacts with NF-B and synergistically activates IL-8 transcription (^{Stein et al, 1993;Roebuck, 1999}). AP-1, however, may not be involved in hormone-induced transcriptional repression as the repression was not abolished by the deletion of the region containing -126 to -120 Figure 9.

In addition to the inhibition of transcriptional activators, sex hormones may promote the synthesis and/or activity of repressor(s) for IL-8 transcription. C/EBP site (-94 to -81) on IL-8 promoter overlaps with an element for Oct-1 (-91 to -84) (^{Roebuck, 1999}), and the binding of Oct-1 displaces C/EBP from this element and thus represses the constitutive IL-8 transcription (^{Wu et al, 1997}). Thus sex hormones may promote such transcriptional repression by Oct-1. On the other hand, Oct-1 acts as a transcriptional activator on mouse mammary tumor virus promoter (^{Prefontaine et al, 1999}). It is known that progesterone/PR complex binds to the hormone response element on this promoter, and facilitates the binding of Oct-1 to its consensus element, and thus promotes its transcriptional activation (^{Prefontaine et al, 1999}). There have been no reports, however, that the hormone-bound receptor may enhance the repressor activity of Oct-1, and this possibility should further be investigated. Another possible mechanism is that the sex hormone-bound receptor may directly interact with the IL-8 promoter and thus displace certain transcriptional activator(s) from the respective element(s). This possibility is, however, rather unlikely as the sequences responsible for transcriptional repression (-98 to -63) contain no consensus hormone response elements. Previous studies also reported that E2-bound ER repressed IL-6 promoter activity (^{Pottratz et al, 1994}) and DHT-bound AR repressed rat dehydroepiandrosterone sulfotransferase promoter activity (^{Song et al, 1998}), both without direct interaction with DNA. IL-8 promoter contains the binding site for PR and AR at -330 to -325 (^{Mukaida et al, 1989}); however, this element did not appear to mediate the transcriptional repression by sex hormones as the repression was not abolished by the deletion of this element-containing region (-546 to -273) Figure 9. It is also speculated that the other undefined transcriptional activators or repressors may be the target for sex hormones. Further studies should elucidate the target molecule(s) and the precise mechanism for transcriptional repression by sex hormones on the IL-8 promoter.

In this study, WM266-4, HS15, and NM26 appeared to proliferate by an autocrine loop of IL-8 as their growth was greatly inhibited by the anti-IL-8 antibody Figure 4. Previous reports also confirmed the role of IL-8 as an autocrine growth factor of melanoma, and the correlation of its expression with metastatic potential (^{Schadendorf et al, 1993;Andrew et al, 1995;Luca et al, 1997}). Certain melanomas, however, may grow independently from IL-8. E2, progesterone, or DHT may enhance the growth of certain melanomas by promoting the production of growth factors different from IL-8, such as transforming growth factor-, the production of which is enhanced by DHT (^{Pederson et al, 1999}). Thus further studies should elucidate the hormonal effects on the IL-8-independent melanomas.

This study shows that these sex hormones can be used for the treatment of patients with melanoma, and that the therapeutic efficiency may depend on the respective receptors. The combination of sex hormones with other chemotherapeutic agents may prevent postoperational recurrence or metastasis of melanoma. This study, however, was performed on the established melanoma cell lines and thus the results may be valid only in the in vitro limited settings. Similar examination should be extended into primary tumor cell cultures from freshly biopsied material as the hormonal effects on such material may be more close to those in vivo. To date, several clinical trials have been done on advanced melanomas by hypophysectomy or the administration of testosterone, ethinylestradiol (^{Bodenham and Hale, 1972}), 6-methylpregn-4-ene,3,11,20-trione (^{Johnson et al, 1966}), medroxyprogesterone acetate (^{Baretta et al, 1979}), or recently tamoxifen (^{Neifeld, 1996;Chapman et al, 1999;Creagan et al, 1999;Propper et al, 2000}); however, only limited subsets of patients were responsive to hormonal therapy, and the response did not correlate with the presence of hormone receptor in their melanoma cells (^{Linder and Borden, 1997;Karakosis et al, 1980;Neifeld and Lippman, 1980;Neifeld, 1996}). These indicate nonreceptor-mediated hormonal effects or the effects via alternate receptors other than conventional intracellular counterparts and/or indirect growth regulation via the other hormone-responding cells. It is reported that E2 disrupts the cytoplasmic microtubule network and thus inhibited the growth of both ER-positive and -negative human breast cancer cell lines, indicating the receptor-independent effect (^{Aizu-Yokota et al, 1994}). The recent studies also suggest that membrane-associated receptors, structurally similar to (^{Papas et al, 1995}) or different from conventional intracellular counterparts (^{Bression et al, 1986}), may exist and mediate some rapid, nontranscriptional actions such as calcium influx (^{Blackmore et al, 1990}) or the activation of tyrosine kinase (^{Migliaccio et al, 1996}). In addition to the high-affinity receptors, different hormone-binding sites are identified; type II estrogen binding sites exhibit lower affinity but higher capacity for estrogen compared with ER, and are detected in several human melanoma cells, and may be involved in the in vitro hormonal growth regulation on those cells (^{Piantelli et al, 1995}). The unconventional receptor-mediated hormonal effects in vivo should further be clarified.

Melanoma-derived growth factors act on adjacent stromal cells such as endothelial cells, macrophages, fibroblasts, or lymphocytes, whereas these cells in turn produce various factors that regulate tumor growth, angiogenesis, adhesion, motility, or metastasis (^{Shih and Herlyn, 1993}; Laza-Molnar et al, 2000). Sex hormones may alter the autocrine and/or paracrine production of angiogenic factors, such as vasculoendothelial growth factor (^{Hyder et al, 1996}). Sex hormones may also alter the expression of adhesion molecules (^{Lupetti et al, 1996}), or activity of proteolytic enzymes in melanoma (^{Herlyn, 1990;Lazar-Molnar et al, 2000}). Sex hormones also affect the host immune responses to melanoma by natural killer cells (^{Hanna and Schneider, 1983}) or macrophages (^{Ackermann et al, 1986}). We are now studying the in vivo effects of sex hormones on the invasion and metastasis of presently used melanomas in athymic nude mice. Further studies should also elucidate the growth-regulatory effects of sex hormones on the coculture of melanoma with endothelial cells, macrophages, or fibroblasts in vitro.

References

References

1.	Ackermann MF, Mark R, Chakraborty PR & Morahan PS. Effects of diethylstilbestrol on propionibacterium acnes. Immunomodulation: inhibition of macrophage activation and antitumor activity. J Leukocyte Biol (1986) 40: 549–559. | PubMed | ISI | ChemPort |
2.	Aizu-Yokota E, Ichinoseki K & Sato Y. Microtubule disruption induced by estradiol in estrogen receptor-positive and -negative human breast cancer cell lines. Carcinogenesis (1994) 15: 1875–1879. | PubMed | ChemPort |
3.	Andrew PJ, Harant H & Lindley IJD. Nitric oxide regulates IL-8 expression in melanoma cells at the transcriptional level. Biochem Biophys Res Commun (1995) 214: 949–956 10.1006/bbrc.1995.2378. | Article | PubMed | ISI | ChemPort |
4.	Baker FL, Spitzer G & Ajani JA et al. Drug and radiation sensitivity measurements of successful primary monolayer culturing of human tumor cells using cell-adhesive matrix and supplemented medium. Cancer Res (1986) 46: 1263–1274. | PubMed | ISI | ChemPort |
5.	Baretta G, Taibadon D & Fossati P. Clinical evaluation of medroxyprogesterone acetate (MAP) in malignant melanoma. Cancer Treat Rep (1979) 63: 1200.
6.	Blackmore PF, Beebe SJ, Danforth DR & Alexander N. Progesterone and 17-hydroxyprogesterone. Novel stimulators of calcium influx in human sperm. J Biol Chem (1990) 265: 1376–1380. | PubMed | ISI | ChemPort |
7.	Bodenham DC & Hale B. Malignant melanoma. In: Stoll BA, ed.Endocrine Therapy in Malignant Disease (1972) London: W.B. Saunders pp 377–383.
8.	Bression D, Michard M, le Dafniet M, Pagesy P & Peillon F. Evidence for a specific estradiol binding site on rat pituitary membranes. Endocrinology (1986) 119: 1048–1051. | PubMed | ISI | ChemPort |
9.	Carey TE, Takahashi T, Resnick LA, Oettgen HF & Old LJ. Cell surface antigens of human malignant melanoma: Mixed hemadsorption assays for humoral immunity to cultured autologous melanoma cells. Proc Natl Acad Sci USA (1976) 73: 3278–3282. | PubMed | ChemPort |
10.	Chapman PB, Einhorn LH & Meyers ML et al. Phase III multicenter randomized trial of the Dartmouth regimen versus dacarbazine in patients with metastatic melanoma. J Clin Oncol (1999) 17: 2745–2751. | PubMed | ISI | ChemPort |
11.	Cobb JP & McGrath A. Brief communication: In vitro effects of melanocyte-stimulating hormone, adrenocorticotropic hormone, 17-estradiol, or testosterone propionate on Cloudman S91 melanoma cells. J Natl Cancer Inst (1974) 52: 567–570. | PubMed | ChemPort |
12.	Cohen C, DeRose PB, Campbell WG, Schlosnagle DC & Sgoutas D. Estrogen receptor status in malignant melanoma. Am J Dermatopathol (1990) 12: 562–564. | PubMed | ISI | ChemPort |
13.	Creagan ET, Suman VJ & Dalton RJ et al. Phase III clinical trial of the combination of cisplatin, dacarbazine, and carmustine with or without tamoxifen in patients with advanced malignant melanoma. J Clin Oncol (1999) 17: 1884–1890. | PubMed | ISI | ChemPort |
14.	Dorn PL & Derse D. cis- and trans-acting regulation of gene expression of equine infectious anemia virus. J Virol (1988) 62: 3522–3526. | PubMed | ISI | ChemPort |
15.	Driscoll MS, Grin-Jorgensen CM & Grant-Kels JM. Does pregnancy influence the prognosis of malignant melanoma? J Am Acad Dermatol (1993) 29: 6119–6630.
16.	Duncan LM, Travers RL, Koerner FC, Mihm MC, Jr & Sober AJ. Estrogen and progesterone receptor analysis in pregnancy-associated melanoma: absence of immunohistochemically detectable hormone receptors. Hum Pathol (1994) 25: 36–41. | Article | PubMed | ISI | ChemPort |
17.	Ferno M, Borg A, Ingvar C & Jonsson PE. Estrogen receptor and binding site for estramustine in metastatic malignant melanoma. Anticancer Res (1987) 7: 741–744. | PubMed | ISI | ChemPort |
18.	Feucht KA, Walker MJ, Das Gupta TK & Beattie CW. Effect of 17 -estradiol on the growth of estrogen receptor-positive human melanoma in vitro and in athymic mice. Cancer Res (1988) 48: 7093–7101. | PubMed | ISI | ChemPort |
19.	Green S, Walter P, Kumar V, Krust A, Bornert J-M, Argos P & Chambon P. Human oestrogen receptor cDNA. sequence, expression and homology to v-erb-A. Nature (1986) 320: 134–139. | Article | PubMed | ISI | ChemPort |
20.	Green S, Issemann I & Sheer E. A versatile in vivo eukaryotic expression vector for protein engineering. Nucleic Acids Res (1988) 16: 369. | PubMed | ISI | ChemPort |
21.	Hanna N & Schneider M. Enhancement of tumor metastasis and suppression of natural killer cell activity by -estradiol treatment. J Immunol (1983) 130: 974–980. | PubMed | ISI | ChemPort |
22.	Harant H, Andrew PJ, Reddy S, Foglar E & Lindley IJD. 1,25-Dihydroxyvitamin D3 and a variety of its natural metabolites transcriptionally repress nuclear-factor-kB-mediated interleukin-8 expression. Eur J Biochem (1997) 250: 63–71. | Article | PubMed | ISI | ChemPort |
23.	Herlyn M. Human melanoma: Development and progression. Cancer Metastasis Rev (1990) 9: 101–112. | PubMed | ISI | ChemPort |
24.	Holly EA. Melanoma and pregnancy. Recent Results Cancer Res (1986) 102: 118–126. | PubMed | ChemPort |
25.	Hyder SM, Stancel GM, Chiapetta C, Murthy L, Boettger-Tong HL & Makela S. Uterine expression of vascular endothelial growth factor is increased by estradiol and tamoxifen. Cancer Res (1996) 56: 3954–3960. | PubMed | ISI | ChemPort |
26.	Ito A, Imada K, Sato T, Kubo T, Matsushima K & Mori Y. Suppression of interleukin 8 production by progesterone in rabbit uterine cervix. Biochem J (1994) 301: 183–186. | PubMed | ISI | ChemPort |
27.	Johnson RP, Bisel H & Andrews N et al. Phase I clinical study of 6-methylpregn-4-ene-3,11,20-trione (NSC-17256). Cancer Chemother Rep (1966) 50: 671–673.
28.	Kalkhoven E, Wissink S, van der Saag PT & van der Burg B. Negative interaction between the RelA (p65) subunit of NF-kB and the progesterone receptor. J Biol Chem (1996) 271: 6217–6224. | PubMed | ISI | ChemPort |
29.	Karakosis CP, Lopez RE, Bhakoo HS, Rosen F, Moore R & Carison M. Estrogen and progesterone receptors and tamoxifen in malignant melanoma. Cancer Treat Rep (1980) 64: 819–827. | PubMed |
30.	Kastner P, Krust A, Turcotte B, Stropp U, Tora L, Gronemeyer H & Chambon P. Two distinct estrogen-regulated promoters generate transcripts encoding the two functionally different human progesterone receptor forms A and B. EMBO J (1990) 9: 1603–1614. | PubMed | ISI | ChemPort |
31.	Keller ET, Chang C & Ershler WB. Inhibition of NFkB activity through maintenance of IkB levels contributes to dihydrotestosterone-mediated repression of the interleukin-6 promoter. J Biol Chem (1996) 271: 26267–26275. | Article | PubMed | ISI | ChemPort |
32.	Kuenen-Boumeester V, van der Kwast TH, Claassen CC, Look MP, Liem GS, Klijn JGM & Henzen-Logmans SC. The clinical significance of androgen receptors in breast cancer and their relation to histological and cell biological parameters. Eur J Cancer (1996) 32: 1560–1565. | Article |
33.	Kunsch C, Lang RK, Rosen CA & Shannon F. Synergistic transcriptional activation of the IL-8 gene by NF-kB p65 (RelA) and NF-IL-6. J Immunol (1994) 153: 153–164. | PubMed | ISI | ChemPort |
34.	Lazar-Molnar E, Hegyesi H, Toth S & Falus A. Autocrine and paracrine regulation by cytokines and growth factors in melanoma. Cytokine (2000) 12: 547–554. | Article | PubMed | ISI | ChemPort |
35.	Linder DJ & Borden EC. Synergistic antitumor effects of a combination of interferon and tamoxifen on estrogen receptor-poritive and receptor-negative human tumor cell lines in vivo and in vitro. J Interferon Cytokine Res (1997) 17: 681–693. | PubMed |
36.	Lopez RE, Bhakoo H, Paolini NS, Rosen F, Holyoke ED & Goldrosen MH. Effect of estrogen on the growth of B-16 melanoma. Surg Forum (1978) 29: 153–154. | PubMed | ISI | ChemPort |
37.	Luca M, Huang S, Gershenwald JE, Singh RK, Reich R & Bar-Eli M. Expression of interleukin-8 by human melanoma cells up-regulates MMP-2 activity and increases tumor growth and metastasis. Am J Pathol (1997) 151: 1105–1113. | PubMed | ISI | ChemPort |
38.	Lupetti R, Mortarini R, Panceri P, Sensi M & Anichini A. Interaction with fibronectin regulates cytokine gene expression in human melanoma cells. Int J Cancer (1996) 66: 110–116 10.1002/(sici)1097-0215(19960328)66:1<110::aid-ijc19>3.3.co;2-5. | Article | PubMed | ISI | ChemPort |
39.	Mackie RM, Bufalino R, Morabito A, Sutherland C & Cascinelli N. For the world health organization melanoma programme: Lack of effect of pregnancy on the outcome of melanoma. Lancet (1991) 337: 653–655. | Article | PubMed | ISI | ChemPort |
40.	McKay LI & Cidlowski JA. Molecular control of immune/inflammatory responses: interactions between nuclear factor-kB and steroid receptor-signaling pathways. Endocr Rev (1999) 20: 435–459. | Article | PubMed | ISI | ChemPort |
41.	Meyskens FL, Jr, Buckmeier JA, McNulty SE & Tohidan NB. Activation of nuclear factor-kB in human metastatic melanoma cells and the effect of oxidative stress. Clin Cancer Res (1999) 5: 1197–1202. | PubMed | ISI | ChemPort |
42.	Migliaccio A, Di Domenico M, Castoria G, de Falco A, Bontempo P, Nola E & Auricchio F. Tyrosine kinase/p21ras/MAP-kinase pathway activation by estradiol-receptor complex in MCF-7 cells. EMBO J (1996) 15: 1292–1300. | PubMed | ISI | ChemPort |
43.	Mukaida N, Shiroo M & Matsushima K. Genomic structure of the human monocyte-derived neutrophil chemotactic factor IL-8. J Immunol (1989) 143: 1366–1371. | PubMed | ISI | ChemPort |
44.	Mukaida N, Morita M, Ishikawa Y, Rice N, Okamoto S, Kasahara T & Matsushima K. Novel mechanism of glucocorticoid-mediated gene repression. J Biol Chem (1994) 269: 13289–13295. | PubMed | ISI | ChemPort |
45.	Neifeld JP. Endocrinology of melanoma. Semin Surg Oncol (1996) 12: 402–406 10.1002/(sici)1098-2388(199611/12)12:6<402::aid-ssu5>3.3.co;2-b. | Article | PubMed | ISI | ChemPort |
46.	Neifeld JP & Lippman ME. Steroid hormone receptors and melanoma. J Invest Dermatol (1980) 74: 379–381. | Article | PubMed | ISI | ChemPort |
47.	Pack GT & Scharnagel M. The prognosis for malignant melanoma in the pregnant woman. Cancer (1951) 4: 324–334. | PubMed | ISI | ChemPort |
48.	Papas TC, Gametchu B & Watson CS. Membrane estrogen receptors identified by multiple antibody labeling and impeded-ligand binding. FASEB J (1995) 9: 404–410. | PubMed | ISI | ChemPort |
49.	Pederson L, Kremer M, Judo J, Pascoe D, Spelsberg TC, Riggs BL & Oursler MJ. Androgens regulate bone resorption activity of isolated osteoclasts in vitro. Proc Natl Acad Sci USA (1999) 96: 505–510. | Article | PubMed | ChemPort |
50.	Piantelli M, Maggiano N & Ricci R et al. Tamoxifen and quercetin interact with type II estrogen binding sites and inhibit the growth of human melanoma cells. J Invest Dermatol (1995) 105: 248–253. | Article | PubMed | ISI | ChemPort |
51.	Pottratz ST, Bellido T, Mocharia H, Crabb D & Manolagas SC. 17-estradiol inhibits expression of human interleukin-6 promoter-receptor constructs by a receptor-dependent mechanism. J Clin Invest (1994) 93: 944–950. | PubMed | ISI | ChemPort |
52.	Prefontaine GG, Walther R, Giffin W, Leimieux ME, Pope L & Hache RJG. Selective binding of steroid hormone receptors to octamer transcription factors determines transcriptional synergism at the mouse mammary tumor virus promoter. J Biol Chem (1999) 274: 26713–26719. | Article | PubMed | ISI | ChemPort |
53.	Proctor JW, Auclair BG & Stokowski L. Brief communication: Endocrine factors and the growth and spread of B16 melanoma. J Natl Cancer Inst (1976) 57: 1197–1198. | PubMed | ChemPort |
54.	Propper DJ, Braybrooke JP & Levitt NC et al. Phase II study of second-line therapy with DTIC, BCNU, cisplatin and tamoxifen (Dartmouth regimen) chemotherapy in patients with malignant melanoma previously treated with dacarbazine. Br J Cancer (2000) 82: 1759–1763 10.1054/bjoc.2000.1141. | Article | PubMed | ISI | ChemPort |
55.	Rampen F. Malignant melanoma: sex differences in survival after evidence of distant metastasis. Br J Cancer (1980) 42: 52–57. | PubMed | ISI | ChemPort |
56.	Roebuck KA. Regulation of interleukin-8 gene expression. J Interferon Cytokine Res (1999) 19: 429–438 10.1089/107999099313866. | Article | PubMed | ISI | ChemPort |
57.	Rosenberg JC, Assimacopoulos C & Rosenerg SA. The malignant melanoma of hamsters: III. Effects of sex and castration on the growth of the transplanted tumor. Ann N Y Acad Sci (1963) 100: 297–296. | PubMed | ISI | ChemPort |
58.	Scatchard G. The attractions of proteins for small molecules and ions. Ann N Y Acad Sci (1949) 51: 660–672. | ISI | ChemPort |
59.	Schadendorf D, Moller A, Algermissen B, Worm M, Sticherling M & Czarnetzki BM. IL-8 produced by human malignant melanoma cells in vitro is an essential autocrine growth factor. J Immunol (1993) 151: 2667–2675. | PubMed | ISI | ChemPort |
60.	Schleicher RL, Hitsberger MH & Beattie CW. Inhibition of hamster growth by estrogen. Cancer Res (1987) 47: 453–459. | PubMed | ISI | ChemPort |
61.	Shaw HM, Milton GW, Farrago G & McCarthy WH. Endocrine influences on survival from malignant melanoma. Cancer (1978) 42: 669–677. | PubMed | ISI | ChemPort |
62.	Shih I-M & Herlyn M. Role of growth factors and their receptors in the development and progression of melanoma. J Invest Dermatol (1993) 100: 196S–203S. | Article | PubMed | ChemPort |
63.	Shiu MH, Schottenfeld D, Maclean B & Fortner JG. Adverse effect of pregnancy on melanoma. A reappraisal. Cancer (1976) 37: 181–187. | PubMed | ISI | ChemPort |
64.	Simental JA, Sar M, Lane MV, French FS & Wilson EM. Transcriptional activation and nuclear targeting signals of the human androgen receptor. J Biol Chem (1991) 266: 510–518. | PubMed | ISI | ChemPort |
65.	Song CS, Jung MH, Kim SC, Hassan T, Roy AK & Chattejee B. Tissue-specific and androgen-repressible regulation of the rat dehydroepiandrosterone sulfotransferase gene promoter. J Biol Chem (1998) 273: 21856–21866. | Article | PubMed | ISI | ChemPort |
66.	Stein B, Baldwin AS, Jr, Ballard DW, Greene WC, Angel P & Herrlich P. Cross-coupling of the NF-kB p65 and Fos/Jun transcription factors produces potentiated biological function. EMBO J (1993) 12: 3879–3891. | PubMed | ISI | ChemPort |
67.	Stein B & Yang MX. Repression of the interleukin-6 promoter by estrogen receptor is mediated by NF-kB and C/EBP. Mol Cell Biol (1995) 15: 4971–4979. | PubMed | ISI | ChemPort |
68.	Sun WH, Keller ET, Stebler BS & Ershler WB. Estrogen inhibits phorbol ester-induced IkB transcription and protein degradation. Biochem Biophys Res Commun (1998) 244: 691–695 10.1006/bbrc.1998.8324. | Article | PubMed | ISI | ChemPort |
69.	Swami S, Krishnan AV & Feldman D. 1a,25-dihydroxyvitamin D3 down-regulates estrogen receptor abundance and suppresses estrogen actions in MCF-7 human breast cancer cells. Clin Cancer Res (2000) 6: 3371–3379. | PubMed | ISI | ChemPort |
70.	Walker BJ, Beattie CW, Patel MK, Ronan SM & Das Gupta TK. Estrogen receptor in malignant melanoma. J Clin Oncol (1987) 5: 1256–1261. | PubMed | ISI |
71.	Wong JH, Sterns EH, Kopald KH, Nizze JA & Morton DL. Prognostic significance of pregnancy in stage I melanoma. Arch Surg (1989) 124: 1227–1231. | PubMed | ISI | ChemPort |
72.	Wu GD, Lai EJ, Huang N & Wen X. Oct-1 and CCAAT/enhancer-binding protein (C/EBP) bind to overlapping elements within the interleukin-8 promoter. J Biol Chem (1997) 272: 2396–2403. | Article | PubMed | ISI | ChemPort |
73.	Zachariae COC, Thestrup-Pedersen K & Matsushima K. Expression and secretion of leukocyte chemotactic cytokines by normal human melanocytes and melanoma cells. J Invest Dermatol (1991) 97: 593–599. | Article | PubMed | ISI | ChemPort |
74.	Zhu W, Smith A & Young YF. A nonsteroidal anti-inflammatory drug, flufenamic acid, inhibits the expression of the androgen receptor in LNCaP cells. Endocrinology (1999) 140: 5451–5454. | Article | PubMed | ISI | ChemPort |

Acknowledgments

We thank Dr. Tomoo Tanase (Takara Gene Analysis Center) for the synthesis of pIL-8 CAT. This work is supported in part by the grant from the Japanese Ministry of Education (12770459) and the aid from KANZAWA Medical Research Foundation.