[CANCER RESEARCH 45,1474-1478, April 1985]

Regulation of Melanin Synthesis of B16 Mouse Melanoma Cells by 1a,25-Dihydroxyvitamin D3 and Retinoic Acid1

Junichi Hosoi, Etsuko Abe, Tatsuo Suda, and Toshio Kuroki2

Department of Cancer Cell Research, Institute of Medical Science, University of Tokyo, Shirokanedai, Minato-ku, Tokyo 108 [J. H., T. K.J and Department of BiochemistrySchool of Dentistry, Showa University, Hatanodai, Shinagawa-ku, Tokyo 142 [E. A., T. S.], Japan

ABSTRACT

Melanin synthesis of B16 mouse melanoma cells was foundto be stimulated dose and time dependently by 1a,25-dihydroxy-vitamin D3 [1«,25(OH)2D3],the hormonal form of vitamin D3. Thestimulation of melanogenesis resulted from an increase in theactivity of tyrosinase, a key enzyme in melanin synthesis. Theminimum dose required for this stimulation was as low as 0.05ng/ml, or 0.12 nw, a physiological level of plasma 1a,25(OH)aD3.The stimulation by 1a,25(OH)2D3 was specific; other derivativesof vitamin D3 caused no stimulation at a concentration of 500ng/ml. When the cells were plated on agar plates, the proportionof dark or black colonies was not increased by the exposure to1o,25(OH)2D3. Furthermore, this compound did not induce melanin synthesis of an amelanotic variant. Thus, its stimulatoryeffect seemed to be due to stimulation of melanin synthesis ofmelanotic cells, rather than to conversion of amelanotic clonesto melanotic ones. 1a,25(OH)2D3 did not induce intracellularcyclic adenosine 3':5'-monophosphate, while cholera toxin induced cyclic adenosine 3':5'-monophosphate and stimulated

melanin synthesis and tyrosinase activity much more than did1a,25(OH)2D3, suggesting that 1a,25(OH)2D3 stimulates melaninsynthesis by a cyclic adenosine 3':5'-monophosphate-independ-

ent mechanism.B16 melanoma cells contained specific receptors for

1a,25(OH)2D3. Scatchard plot analysis revealed two types ofreceptor; the high-affinity receptor had a Ka of 18.3 PM and anA/ma«of 10.6 fmol/mg of protein. The specificity of receptorbinding was demonstrated by studies showing that, for 50%displacement of la^SfOHfeDg binding, other derivatives wererequired at 500 times higher concentrations or more.

In contrast to 1«,25(OH)2D3,retinoic acid inhibited melaninsynthesis and tyrosinase activity of B16 melanoma cells doseand time dependently. On simultaneous treatment, 1a,25(OH)2D3and retinoic acid caused mutual interference, and a balancebetween their respective stimulating and inhibitory effects wasobtained at a molar ratio of 10:1; I.e., with 10 nw 1«,25(OH)2D3and 1 nw retinoic acid.

INTRODUCTION

It is now well established that vitamin D3 is metabolized firstin the liver to 25(OH)D33 and then in the kidney, mainly to either

24R,25(OH)2D3 or la^OHfeDa (10). Of these metabolites,

1Supported in part by a Grant for Cancer Research from the Ministry of

Education, Science and Culture of Japan.1To whom requests for reprints should be addressed.3The abbreviations used are: 25(OH)D3, 25-hydroxyvitamin D3; 1«¿Si.OHfcDs,

1a, 25-dihydroxy vitamin D3; 24R,25(OH)2D3, 24R,25-dihydroxyvitamin D3;1n(OH)D3. 1«-hydroxyvitamin D3; cAMP, cyclic adenosine 3':5'-monophosphate;AL«,,maximum binding; MSH, melanocyte-stimulating hormone.

Received 7/31/84; revised 11/29/84; accepted 12/19/84.

1a,25(OH)2D3 is considered to be the hormonal form of thevitamin in regulating the blood Ca2+ level by enhancing intestinal

calcium transport and bone mineral mobilization (10).A receptor with high affinity for 1a,25(OH)2D3 has been de

tected in a wide variety of tisssues and cells including cancercells and cultured cell lines (Refs. 5, 7 to 9,11 to 15,19, 28, and33 to 36; for a review, see Refs. 10 and 23). The existence ofthe receptor suggests that these tissues or cells may be targetsfor 1a,25(OH)2D3, although the physiological functions of thiscompound have been demonstrated only in certain tissues (10).

Recently, we have been studying the regulation of growth,differentiation, and tumor promotion by 1a,25(OH)aD3 (23). Wefound that 1a,25(OH)îD3stimulates differentiation of the myeloidleukemic cells of humans (27,35) and mice (1) and the epidermalkeratinocytes of mice (14). This naturally occurring hormone wasfound to induce fusion of mouse alveolar macrophages by adirect mechanism and by an indirect mechanism mediated byspleen cells (2). We also found that this vitamin markedly enhances chemically induced transformation of BALB 3T3 cells(22) but that it paradoxically inhibits induction of ornithine decar-boxylase caused by 12-O-tetradecanoylphorbol-13-acetate andteleocidin B (6). Its antitumor-promoting action in skin carcino-genesis was reported by Wood ef a/. (39).

As for these collaborative studies, we report here that1a,25(OH)2D3 stimulates melanin synthesis in B16 melanomacells and that its effect is counteracted by another liposolublevitamin, retinoic acid.

MATERIALS AND METHODS

Chemicals. Vitamin D3derivatives, i.e.. 1a^OHfeDa, 24R,25(OH)2D3,25(OH)D3, and 1«(OH)D3,were kindly donated by I. Matsunaga, ChugaiPharmaceutical Co. (Tokyo, Japan). They were dissolved in ethanol. Allfrans-retinoic acid was obtained from Sigma Chemical Co. (St. Louis,

MO) and dissolved in acetone. Cholera toxin was purchased from theChemo-Sero-Therapeutic Research Institute (Kumamoto, Japan). [23,24-3H]-1a,25(OH)zD3 (specific activity, 85 Ci/mmol) and L-[3,5-3H]tyrosine

(specific activity, 54.6 Ci/mmol) were purchased from Amersham International, Ltd. (Amersham, Buckinghamshire, United Kingdom). All othercompounds were of the purest grade available.

Cell Culture. A subline of B16 mouse melanoma cells, C2M (31), wasprovided by Dr. A. Oikawa, Tohoku University. The cells were grown inEagle's minimal essential medium supplemented with 10% fetal calf

serum at 37°C in a humidified atmosphere of 5% CO2 in air. Colonies

were formed on 0.5% agar plates (20) by plating 1000 cells and incubating them for 2 weeks. White and dark colonies were counted under adissecting microscope. An amelanotic variant of B16 melanoma cells, H9(16), was provided by Dr. K. Iwata, Central Research Laboratory, JapanTobacco and Salt Public Corporation.

Measurement of Cytotoxic and Cytostatic Effects. Cytotoxic effectsof 1a,25(OH)2D3 or retinoic acid were measured by plating 100 cells in a60-mm Petri dish and adding the test compound 18 to 24 hr later (Day

CANCER RESEARCH VOL. 45 APRIL 1985

1474

REGULATION OF MELANIN SYNTHESIS BY 1«,25(0^03

1). The cultures were fixed and stained with Giemsa on Day 7 forexamination of plating efficiency. In agar plate culture, the cells wereplated directly on an agar layer containing 1<>,25(OH)2D3. For examinationof the cytostatic effect, 5x10* cells were plated in a 35-mm Retri dish,

and the test compound was added 18 to 24 hr later. Medium containingthe compound was renewed on Day 3. The number of living cells wascounted on Day 5 by dye exclusion test using 0.01% erythrocin B.

Measurement of Melanin Content. The melanin content was measured by a modification of the method of Oikawa and Nakayasu (29) andMeyskens and Fuller (26). Cells grown in a 100-mm Retri dish werewashed twice with phosphate-buffered saline (8 g NaCI, 0.2 g KCI, 0.2g KH2PO4, and 1.15 g Na2HP04 •2H2O per liter), scraped off with a rubber

spatula, centrifugea at 300 x g for 10 min, and stocked in a freezer untilmeasurement. The cells were sonicated in 1 ml of water on ice, and themixture was centrifugea at 1200 x g for 20 min. The supernatant wasdecanted, and acid-insoluble materials were recovered by 2 precipitations

with 10% trichloroacetic acid and one washing with ethanol. The precipitate was dried and solubilized by treatment with 1 ml of 1 N NaOH and10% dimethyl sulfoxide at 80°C for 2 hr. Standard solutions of melanin

(a synthetic preparation obtained from Sigma) were incubated in parallel.The absorbance was read at 470 nm, and the melanin content wasexpressed as ng/mg protein. Protein was measured by the method ofBradford (3).

Assay of Tyrosinase Activity. Tyrosinase activity was assayed bythe method of Oikawa ef al. (32) using L-[3H]tyrosine which was previ

ously evaporated to remove preexisting tritiated water. Sixteen h afterplating 105 cells/60-mm dish, the medium was changed to fresh mediumcontaining i_-[3,5-3H]tyrosine at a concentration of 2 ¿iCi/ml.After incu

bation for 3 days, the medium was collected and frozen. A mixture of500 n\ of the medium and 500 ^l of 10% trichloroacetic acid wascentrifuged, and the supernatant was treated with 50 mg of Norit SX3.Radioactivity of tritiated water derived from L-tyrosine was counted in a

liquid scintillation counter.Measurement of Intracellular cAMP. Intracellular cAMP was mea

sured after succinylation of the cyclic nucleotides by radioimmunoassaywith a kit from Yamasa Shoyu Co. (Chiba, Japan). The cells were washedquickly with ice-cold phosphate-buffered saline, treated with 6% perchloric acid, and scraped off with a rubber spatula. The acid-soluble fraction

was sonicated and subjected to radioimmunoassay as described elsewhere (21 ).

Assay of 1«,25(OH)2D3Receptor. The cytosol fraction of B16 melanoma cells was prepared and incubated with 0.015 to 1.0 nM [3H]-1«,25(OH)uD3at 25°C for 60 min in buffer containing 0.3 M KCI, 2 mw

EDTA, 0.5 HIM dithiothreitol, and 10 mw Tris-HCI (pH 7.4). Bound andfree [3H]-1 «,25(01-1^035were separated on hydroxylapatite (37). Details

of the procedure were reported elsewhere (14).

RESULTS

Inhibition of Growth by 1«,25(OH)2D3.1<>,25(OH)2D3slightlyinhibited colony formation of B16 melanoma cells when added18 to 24 h after the cells were plated; at 5 ng/ml, it suppressedcolony formation by 9%. Although 1a,25(OH)zD3 showed littlecytotoxicity, it suppressed the growth dose dependently. It hada negligible cytostatic effect at 0.05 ng/ml but caused about30% inhibition of the growth at 5 ng/ml on Day 5.

In the following experiments, 1a,25(OH)2D3 doses of 0.05 to5 ng/ml were used because the physiological level in plasma isabout 0.05 ng/ml (4) and because at these concentrations1a,25(OH)2D3 was only moderately toxic, and we had foundpreviously that it induced differentiation of mouse epidermalkeratinocytes and of mouse and human myeloid leukemia cells(1,14,27,35).

Stimulation of Melanin Synthesis by 1a,25(OH)2D3. Chart 14

1 2 345Time (day)

0 ' O.O5 OS 5.0 '

Concentration (ng/ml)

Chart 1. Time-dependent (A) and dose-dependent (B) stimulation of melaninsynthesis by la^OHfeDa. In A, inocula of 2 x 105 cells were plated in 100-mm

dishes and cultured for 6 days. The medium was changed every day, and1,i,25(OH)2D3 (4 ng/ml) was added for the indicated period during the last part ofincubation. In B. inocula of 5 x 10* cells were plated in 100-mm dishes, and

medium containing the indicated concentrations of ln.25(OH);,D3 was added onDay 1 and renewed on Day 3. The melanin content was determined on Day 5.Points, means for 3 (A) and 5 (B) samples; oars, SD.

fi 60

O.05 0.5 5 50Concentration ( ng/ml )

500

Charts. Tyrosinase activity after treatment with 1a,25(OHfcD3 (•),25(OH)D3(A), 24R,25<OH)2D3 (D), and 1«(OH)D3(O). On Day 1 after 105 cells were plated in60-mm dishes, the medium was changed to that containing vitamin D3 derivativesand [3H]tyrosine (2 ¿tCi/ml).On Day 4, the medium was collected, and tyrosinaseactivity was assayed as described in "Materials and Methods.' Points, means for

4 measurements; bars, SD. The difference between 0 and 0.05 nM 1,,.25(OH)?D3is significant at P < 2%, and those between 0 and 0.5 nM and between 0 and 5 nuare significant at P < 0.1%.

shows time-dependent stimulation of melanin synthesis by

1a,25(OH)2D3. To avoid possible effects of the length of cultureperiod on melanin synthesis, B16 melanoma cells were culturedfor the same period, i.e., 6 days in all experimental groups, and1a,25(OH)2D3 (4 ng/ml) was added for the indicated periodsduring the last part of cultivation. Melanin synthesis was stimulated time dependently, reaching a maximum of about 3-fold thecontrol level on treatment for 4 days. Chart 1S shows the dose-

dependent stimulation of melanin synthesis on exposure for 4days to 1a,25(OH)2D3 at 0.05 to 5 ng/ml. Even 0.05 ng/ml causeda significant increase. Some variation in the extent of the stimulation was observed among experiments. The stimulation was2.1-fold on the average (SD, 0.5; range, 1.4 to 3.2) in 9 inde

pendent experiments in which 4 or 5 ng of 1a,25(OH)2D3 per mlwere added for 4 days.

Melanin synthesis is catalyzed by tyrosinase which oxidizestyrosine to dopaquinone and converts 5,6-dihydroxyindole to

melanochrome (18). Therefore, tyrosinase is considered as a keyenzyme in melanin synthesis. As shown in Chart 2, the tyrosinaseactivity of B16 melanoma cells was stimulated by 1<i,25(OH)?D3dose dependently at 0.05 to 5.0 ng/ml. This result suggests that

CANCER RESEARCH VOL. 45 APRIL 1985

1475

REGULATION OF MELANIN SYNTHESIS BY 1a,25(OH>2D3

1o,25(OH)2D3 stimulates melanin synthesis through activation oftyrosinase.

For assessment of the specificity of the effect of 1a,25(OH)2D3,B16 melanoma cells were cultured in the presence of otherderivatives of vitamin D3, i.e., 25(OH)D3, a precursor of1a,25(OH)2D3 formed in the liver; 24R,25(OH)2D3> a metaboliteformed in the kidney in parallel with 1a,25(OH)2D3; and 1«(OH)D3,a synthetic derivative of vitamin D3. No increase in tyrosinaseactivity was observed with these derivatives at concentrationsof 500 ng/ml (Chart 2).

The population of B16 melanoma cells is heterogeneous withrespect to melanin synthesis; when plated on the surface of agarplates, white colonies and colonies showing various degrees ofmelanin synthesis were observed under a dissecting microscope.Stimulation of melanin synthesis by 1a,25(OH)2D3 might therefore be due to stimulation of melanin synthesis of melanoticclones and/or population change by conversion of amelanoticclones to melanotic ones. For examination of these possibilities,B16 melanoma cells were plated on agar plates containing1a,25(OH)2D3 at concentrations of 0.05 to 5 ng/ml, and dark andwhite colonies were counted. As seen in Table 1, colony formation decreased dose dependently to a greater degree than inliquid medium as described above. This toxicity might have beendue to plating of the cells directly on plates containing1a,25(OH)2D3. It was found that 1a,25(OH)aD3 did not increasethe proportion of melanin-synthesizing clones. The majority (54

to 69%) of colonies were dark in both the absence and presenceof 1a,25(OH>2D3.Consistent with this observation, we found thatexposure of cells of an amelanotic variant (H9) of B16 melanomacells to 1a,25(OH)zD3 (5 ng/ml) for 4 days did not induce melaninsynthesis. These results provide evidence for the first possibilitythat 1a,25(OH)aD3 stimulates melanin synthesis of melanotic cellsrather than of amelanotic cells.

Intracellular cAMP Level after Treatment with 1«,25(OH)2D3-The amount of intracellular cAMP was measured by radioimmu-

noassay after exposure of the cells to 1a,25(OH)zD3 (5 ng/ml).During incubation for 6 h, there was no increase in the intracellular cAMP level, which remained at 0.8 pmol/mg protein. Incontrast, addition of cholera toxin (1 ng/ml) markedly increasedthe amount of cAMP, and a level of 35 pmol per mg protein, 44-

fold the control value, was observed after 3 h incubation. Choleratoxin also stimulated melanin synthesis to a greater extent thandid 1a,25(OH)2D3. The stimulation was dose dependent at concentrations above 0.01 ng/ml, and about a 6-fold increase was

obtained at 1 ng/ml. These results indicate that there are 2mechanisms of stimulation of melanin synthesis, cAMP-depend-ent and -independent mechanisms, and that 1a,25(OH)2D3 af

fects the latter.

Table 1Colony formation of B16 melanomacells on agarplates containing 1a,25(OH)J33

No. ofcolonies"(ng/ml)00.050.5

5.0Plating

efficiency36.2±2.4"

25.7 ±1.721.1 ±3.110.1 ±3.1White110.4

79.267.246.8Black/dark251.2

178.0143.853.2Proportion

ofblack/dark

colonies(%)69.4

69.2C68.1C53.2"

Presence of Specific Receptors for 1ir,25(OH)2D3. The specific receptor for 1a,25(OH)2D3 was assayed by incubating cy-tosol with [3H]-1a,25(OH)2D3 and then separating the bound and

free vitamin. Binding was corrected for nonspecific binding, whichwas determined by parallel incubation with a large excess ofcold 1a,25(OH)2D3. This assay was repeated 3 times, and atypical result is shown in Chart 3. Specific binding increased withincrease in the dose of radioactive vitamin but did not reach aplateau at a concentration of 1 nw (Chart 3A). This specificbinding was examined by Scatchard analysis. Two straight lineswere obtained (Chart 3B). There are 2 types of receptor in B16melanoma cells: a high-affinity receptor with a Ka of 18.3 pw andA/m«of 10.6 fmol/mg protein; and a low-affinity receptor with a

Ka of 287 pw and A/™«of 19.7 fmol/mg protein. The A/maxof thehigh-affinity receptor is of the same order as but smaller than

those obtained for epidermal keratinocytes (14), BALB 3T3 cells(22), and myeloid leukemia cells (35).

The specificity of the binding to the high-affinity receptor was

assessed by testing the competitive binding of other vitamin D3derivatives; i.e.. 25(OH)D3, 1«(OH)D3,and 24R,25(OH)zD3. Asseen in Chart 4, these derivatives bound to the receptor but hadmuch lower affinities than did 1a,25(OH)2D3, concentrations of500-fold or more being required for 50% displacement.

40

0.6

0.4

0.2

OS 1.0Concentration (nM)

10Bound

20<pM)

* Average for 6 dishes.* Mean ±SO.' Not significantlydifferent from the control without 1a,25(OH)¡D3.1Significantly (P < 0.5) different from the control and other groups.

Charts. Analysis of binding of 1a,25(OH)2D3to cytosol receptor of B16 melanoma cells. A, dose-dependent binding of ['HJ-la^SfOHfeDs. [3H]-1a,25(OH)uD3was bound to the cytosol fraction of cells in the presence (O) or absence (•)of alarge excess (400 nM)of unlabeled10,25(01-1)203.Specific binding (D) was definedas the differencebetween the 2 values.B, Scatchardanalysisof the specificbindingdata from A.

200

1150

:100

SO

0.01 0.1 1 10 100 1000 10000

Concentration ( nM )Chart 4. Competition of vitamin D3derivatives with [3H]-1a,25(OH)2D3for bind

ing. Binding of |3H|1,,,25(OH)2D3was assayed in the presence of 1...25(OH)2D3(•),25(OH)D3(D), 1a(OH)D3(O),or 24R,25(OH)2D3(A). The cytosol fraction of B16mouse melanoma cells was incubated with 0.083 nM |3H1-1,,.25(OH).,D3in thepresence of various concentrations of unlabeled 1a,25(OH)JD3, 25(OH)D3,24R.25(OH)2D3,or 1a(OH)D3.

CANCER RESEARCH VOL. 45 APRIL 1985

1476

REGULATION OF MELANIN SYNTHESIS BY 1a,25(OH)¡,D3

Inhibition of Melanin Synthesis by Retinoic Acid. A numberof reports had indicated the involvement of vitamin A and itsanalogues in the regulation of cellular differentiation. We therefore investigated its effects, alone and in combination with1a,25(OH)2D3, on melanin synthesis under the same conditions.As shown in Chart 5, retinoic acid inhibited melanin synthesisboth dose and time dependently. The inhibition was maximumon treatment with retinoic acid (3 ng/ml) for 3 days or more.Retinoic acid at concentrations of 0.03 to 3 ng/ml was not toxicto the cells, although it slightly inhibited growth dose dependently(data not shown).

Retinoic acid also inhibited melanin synthesis induced by'\a,25(OH)2D3. For the results shown in Chart 6, B16 melanoma

cells were exposed to 1a,25(OH)2D3 (10 nw or 4 ng/ml) in thepresence and absence of retinoic acid (0.1 to 1000 nu or 0.03to 300 ng/ml) for 4 days, and the amounts of melanin synthesizedwere measured. Retinoic acid inhibited melanin synthesis induced by 10 nw 1a,25(OH);>D3 dose dependently. A balance ofstimulation and inhibition by the 2 vitamins, i.e., A and D, wasobtained at a molar ratio of 1:10. Tyrosinase activity induced by1<>.25(OH)2D3and cholera toxin was also inhibited by retinoicacid (data not shown).

DISCUSSION

In the present study, we demonstrate that 1«^(OHfeDs stimulates melanin synthesis of B16 mouse melanoma cells as aresult of increasing the activity of tyrosinase, a key enzyme in

! 60

! 20

04-12345

Time (day)

100

80

60

40

2O

^'0.03 0.3 3.0 30

Concentration (ng/ml)

Chart 5. Time-dependent(/A)and dose-dependent(B) inhibitionof melaninsynthesis by retinoic acid. Retinoic acid (3 ng/ml in A and the indicated concentrationsin B) was added by the same schedule as for Chart 1. Points, means for 3 (A) and5 (B) samples; bars. SD.

§80oa.oí

60»40f

201Retinoic

a_-;_n_-1nnnM0 10 0 0.1 1 10 100100Ca,2S(OH),D3o0 10 10 10 10 10 10

Concentration of Vitamins ( nM )Chart6. Inhibition of li..25(OH)2D3-mducedmelanin synthesis by retinoic acid.

i,i.25(OH),D3 and/or retinoic acid at the indicated concentrations were added onDays 1 and 3. The amount of melaninsynthesized was measuredon Day 5.

melanin synthesis. The presence of a specific receptor, togetherwith specific stimulation by 1«¿SfOH^Ds,suggests a receptor-mediated mechanism of stimulation.

There are 2 previous reports suggesting the possible regulation of melanin synthesis by vitamin D3. Oikawa and Nakayasu(30) reported that tyrosinase activity and melanin synthesis ofB16 melanoma cells were stimulated by vitamin D2 (ergocalci-ferol) and vitamin D3 (cholecalciferol), but only at a much higherconcentration of 1a,25(OH)2D3, i.e., 10 /¿g/ml.Colston ef al. (7)demonstrated that 1a,25(OH)2D3 inhibited the growth and DNAsynthesis of human melanoma cells (Hs 695T) which containeda specific receptor for 1a,25(OH)2D3 with a Ka of 180 pw and anA/m«of 176 fmol/mg protein. These 2 reports, although ondifferent melanoma lines, suggest regulation of melanin synthesisby 1a,25(OH)2D3, and this possibility was supported by thepresent results.

It is known that MSH, a pituitary hormone, stimulates melaninsynthesis of melanocytes by causing activation of tyrosinase.Wong and Pawelek (38) reported that MSH also stimulatesmelanin synthesis of Cloudman S91 melanoma cells. MSH bindsto a membrane receptor of melanocytes or melanoma cells andactivates adenylate cyclase, resulting in elevation of the intracel-

lular level of cAMP. Körnerand Pawelek (17) found that thetyrosinase activity of melanoma cells is activated by cAMP-

dependent protein kinase, possibly due to inactivation of aninhibitor of tyrosinase. Stimulation of melanin synthesis by cholera toxin may be governed by a similar cAMP-mediated mecha

nism. In contrast, la^SfOH^Ds did not increase the intracellularcAMP level of B16 melanoma cells, suggesting that unlike MSHand cholera toxin it stimulates tyrosinase activity by a mechanismnot mediated by cAMP.

Vitamin A and its derivatives are known to act as regulatorsof differentiation in various cell types, including melanoma cells.Lotan and Lotan (24, 25) reported that retinoic acid inducesmelanogenesis in S91 mouse melanoma cells and Hs931 humanmelanoma cells. Inconsistent with this observation, we foundthat melanin synthesis and tyrosinase activity of B16 mousemelanoma cells were suppressed by retinoic acid. This apparentdiscrepancy suggests that the regulatory mechanism of melaninsynthesis differs among different melanoma cell lines. Of particular interest is the finding that retinoic acid counteracted thestimulatory effect of 1<>,25(OH);.D:,and that the balance of these2 lipophilic vitamins, A and D, was attained at a molar ratio of1:10.

It is well known that vitamin D3 is formed from provitamin D3in the skin, principally in the malpighian stratum, on exposure tosunlight. Pigmentation of the skin prevents penetration of sunlight into the epidermis, thus resulting in reduced synthesis ofvitamin D3. Stimulation of melanin synthesis by 1a,25(OH)2D3, ifit occurs in epidermal melanocytes in vivo, may act as a mechanism of negative feedback of the metabolism of vitamin D3 bypreventing conversion of provitamin D3 to vitamin D3. Colston efa/. (7) also proposed the same hypothesis from the receptorstudy of human melanoma cells.

ACKNOWLEDGMENTS

The authors are grateful to Professor A. Oikawa, Tohoku University, Sendai,Japan, and to Dr. K. Iwata, Japan Tobacco and Salt Public Corporation, forproviding cells and for valuablesuggestions.

CANCER RESEARCH VOL. 45 APRIL 1985

1477

REGULATION OF MELANIN SYNTHESIS BY 1a,25(OH)2D3

REFERENCES

1. Abe, E., Miyaura, C., Sakagami, H., Takeda, M., Konno, K., Yamazaki, T.,Yoshiki, S., and Suda, T. Differentiation of mouse myeloid leukemia cellsinduced by 1«,25-dihydroxyvitamin D3. Proc. Nati. Acad. Sci. USA, 78: 4990-

4994,1981.2. Abe, E.. Miyaura, C., Tanaka, H., Shiina, Y., Kuribayashi, T., Suda, S., Nishii,

Y., DeLuca, H. F., and Suda, T. 1a,25-Dihydroxyvitamin D3 promotes fusionof mouse alveolar macrophages both by a direct mechanism and by a spleencell mediated indirect mechanism. Proc. Nati. Acad. Sci. USA 80: 5583-5587,

1983.3. Bradford. M. M. A rapid and sensitive method for the quantitation of microgram

quantities of protein utilizing the principle of protein-dye binding. Anal. Biochem.72:248-254, 1976.

4. Brumbaugh, P. F., Haussler, D. H., Bursac, K. M., and Haussler, M. R. Filterassay for 1a,25-dihydroxyvitamin D3. Utilization of the hormone's target tissue

chromatin receptor. Biochemistry, 13: 4091-4097, 1974.5. Chen, T. L., and Feldman, D. Regulation of 1,25-dihydroxyvitamin D3 receptors

in cultured mouse bone cells. Correlation of receptor concentration with therate of cell division. J. Biol. Chem., 256: 5561-5566,1981.

6. Chida, K., Hashiba, H., Suda, T., and Kuroki, T. Inhibition by 1a,25-dihydroxy-vitamin D3 of induction of epidermal omithine decarboxylase caused by 12-O-tetradecancylphorbol-13-acetate and teteocidin B. Cancer Res., 44: 1387-

1391,1984.7. Colston, K., Colston, M. J., and Feldman, D. 1o,25-Dihydroxyvitamin D3 and

malignant melanoma: the presence of receptors and inhibition of cell growthinculture. Endocrinology, 708: 1083-1086,1981.

8. Colston, K., Colston, M. J., Fieldsteel, A. H., and Feldman, D. 1,25-Dihydroxy-vitamin D3 receptors in human epithelial cancer cell lines. Cancer Res., 42:856-859.1982.

9. Colston, K., Hirst, M., and Feldman, D. Organ distribution of the cytoplasmic1a,25-dihydroxycholecalciferol receptor in various mouse tissues. Endocrinology, Õ07:1916-1922.1980.

10. DeLuca, H. F., and Schnoes, H. K. Vitamin D: recent advances. Annu. Rev.Biochem., 52: 411-439,1983.

11. Eisman, J. A., Martin, T. J., and Macintyre, I. Presence of 1«,25-dihydroxyvitamin D3 receptor in normal and abnormal breast tissues. Prog. Biochem.Pharmacol., 17: 143-150, 1980.

12. Haussler, M. R., Manolagas, S. C., and Deftos, L. T. Evidence for 1,25-dihydroxyvitamcn D3receptor-like macromolecule in rat pituitary. J. Biol. Chem.,255: 5007-5010,1980.

13. Haussler, M. R., Pike, J. W., Dokoh, S., Chandler. J. S., Chandler, S. K.,Donaldson, C. A., and Marion, S. L. 1,25-Dihydroxyvitamin D receptor incultured cell lines: occurrence, subcellular distribution and relationship tobioresponses. In: A. W. Norman, K. Schaefer, D. V. Herrath, and H.-G. Grigoteit

(eds.). Vitamin D, Chemical, Biochemical and Clinical Endocrinology of CalciumMetabolism, pp. 109-113. Berlin: Walter de Gruyter, 1982.

14. Hosomi, J., Hosoi, J., Abe, E., Suda, T., and Kuroki, T. Regulation of culturedmouse epidermal cells by 1«,25-dihydroxyvitamin D3. Endocrinology, 773:1950-1957.1983.

15. Hughes, M. R., and Haussier, M. R. 1,25-Dihydroxyvitamin D3 receptors inparathyroid glands. J. Biol. Chem., 253: 1065-1073, 1978.

16. Iwata, K., and Inui, N. Establishment of melanoma cell lines in chemicallydefined medium. Proc. Jpn. Acad., 56:146-151,1980.

17. Kömer, A., and Pawetek, J. Activation of melanoma tyrosinase by a cAMP-dependent protein kinase in a cell-free system. Nature (Lond.), 267:444-467,

1977.18. Kömer, A., and Pawelek, J. Mammalian tyrosinase catalyzes three reactions

in the biosynthesis of melanin. Science (Wash. DC), 277:1163-1165,1982.19. Kream, B. E., Jose, M.. Yamada, S., and DeLuca, M. F. A specific high-affinity

binding macromolecule for 1,25-dihydroxyvitamin D3 in fetal bone. Science

(Wash. DC), 797: 1086-1088,1977

20. Kuroki, T. Agar plate culture and Lederberg style replica plating of mammaliancells. Methods Cell Biol. 9:157-178,1975.

21. Kuroki, T., Ito, T., Hosomi, J., Munakata.K., Uchida, T., and Nagai, Y. CyclicAMP as a mitotic signal for epidermal keratinocytes but not for dermalfibroblasta. Cell Struct. Fune.. 7: 295-305, 1982.

22. Kuroki, T., Sasaki, K., Chida, K., Abe, E., and Suda, T. 1«,25-DihydroxyvitaminD3 markedly enhances chemically-induced transformation in BALB 3T3 cells.Gann, 74.: 611-614,1983.

23. Kuroki, T., and Suda, T. Similarity and dissimilarity between phorbol estertumor promoters and 1«,25-dihydroxyvitamin D3, an active form of vitamin D3.

In: H. Fujiki, E. Hecker, R. E. Moore, T. Sugimura, and I. B. Weinstein (eds.),Cellular Interactions by Environmental Tumor Promoters, pp. 89-99. Tokyo:Japan Scientific Societies Press/Utrecht, VNU Science Press, 1984.

24. Lotan, R., and Lotan, D. Stimulation of melanogenesis in human melanomacell line by retinoids. Cancer Res., 40: 3345-3350,1980.

25. Lotan, R., and Lotan, D. Enhancement of melanotic expression in culturedmouse melanoma cells by retinoids. J. Cell. Physiol., 706: 179-189, 1981.

26. Meyskens, F. L., Jr., and Fuller, B. B. Characterization of the effects of differentretinoids on the growth and differentiation of a human melanoma cell line andselected subclones. Cancer Res., 40: 2194-2196.1980.

27. Miyaura, C., Abe, E., Kuribayashi, T., Tanaka, H., Konno, K., Nishii, Y., andSuda, T. 1«,25-Dihydroxyvitamin D3 induces differentiation of human myeloidleukemia cells. Biochem. Biophys. Res. Commun., 702: 937-943, 1981.

28. Monolagas, B. C., Haussler, M. R., and Deftos, L. J. 1«,25-DihydroxyvitaminD3 receptor-like macromolecule in rat osteogenic sarcoma cell lines. J. Biol.Chem., 255: 4414-4417, 1980.

29. Oikawa, A., and Nakayasu, M. Quantitative measurement of melanin astyrosine equivalents and as weight of purified melanin. J. Biol. Med., 46: 500-507, 1973.

30. Oikawa, A., and Nakayasu, M. Stimulation of melanogenesis in culturedmelanoma cells by calciferols. FEBS Lett., 42: 32-35, 1974.

31. Oikawa, A., Nakayasu, M., Claunch, C., and Tchen, T. T. Two types ofmelanogenesis in monolayer cultures of melanoma cells. Cell Differentiation,7: 149-155, 1972.

32. Oikawa, A., Nakayasu, M., Nohara, M., and Tchen, T. T. Fate of L-[3,5-3H]tyrosine in cell-free extracts and tissue cultures of melanoma cells; a newmethod for tyrosinase in living cells. Arch. Biochem. Biophys., 748: 548-557,1972.

33. Prowedini, D. M., Tsoukas, C. D., Deftos, L. J., and Manolagas, S. C. 1,25-Dihydroxyvitamin D3 receptors in human leukocytes. Science (Wash. DC), 227:1181-1183,1983.

34. Stumpf, W. E., Sar, M., Clark, S. A., and DeLuca, H. F. Brain target sites for1,25-dihydroxyvitamin D3. Science (Wash. DC), 275: 1403-1405,1982.

35. Tanaka, H., Abe, E., Miyaura, C., Kuribayashi, T., Konno, K., Nishii, Y., andSuda, T. 1«,25-Dihydroxycholecalciferol and a human myeloid leukemia cellline (HL-60). The presence of a cytosol receptor and induction of differentiation.Biochem. J., 204: 713-719, 1982.

36. Wecksler, W. R., Mason, R. S., and Norman, A. W. Specific cytosol receptorsfor 1a,25-dihydroxyvitamin D3 in human intestine. J. Clin. Endocrinol. Metab.,48: 715-717,1979.

37. Wecksler, W. R., and Norman, A. W. A hydroxyapatite batch assay for thequantitation of 1o,25-dihydroxyvitamin D3receptor complexes. Anal. Biochem.,92:314-323,1979.

38. Wong, G., and Pawelek, J. Melanocyte-stimulating hormone promotes activation of pre-existing tyrosinase molecule in Cloudman S91 melanoma cells.Nature (Lond.), 255: 644-646,1975.

39. Wood, A. W., Chang, R. L., Huang, M. T., Uskokovic, M., and Conney, A. H.1«,25-Dihydroxyvitamin D3 inhibits phorbol ester-dependent chemical carcin-ogenesis in mouse skin. Biochem. Biophys. Res. Commun., 776: 605-611,1983.

CANCER RESEARCH VOL. 45 APRIL 1985

1478