Regulation of mammalian melanogenesis II: the role of metal cations

276

Biochimica et Biophysica Acta, 1035(1990) 276-285

Elsevier BBAGEN 23368

Regulation of mammalian melanogenesis II: the role of metal cations Jose R. Jara, Francisco Solano, Jose C. Garcia-Borron, Pilar Aroca and Jose A. Lozano Departamento de Bioqulmica y Biologia Molecular, Faeultad Medicina, Universidad Murcia, Murcia (Spain)

Received 19 July 1989)

Key words: Mammalian melanogenesls; Metal cation; Dopachrometautomerase; (Mouse melanoma)

Melanogenesis can be divided into two phases. The first one involves two tyrosinase-catalyzed oxidations from tyrosine to dopaquinone and a very fast chemical step leading to dopachrome. The second phase, from dopachrome to melanin, can proceed spontaneously through several incompletely known reactions. However, some metal transition ions and protein factors different from tyrosinase might regulate the reaction rate and determine the structure and relative concentrations of the intermediates. The study of the effects of some divalent metal ions (Zn, Cu, Ni and Co) on some steps of the melanogenesis pathway has been approached using different radiolabeled substrates. Zn(lI) inhibited tyrosine hydroxylation whereas Ni(II) and Co(II) were activators. Ni(II), Cu(II) and Co(II) accelerated chemical reactions from dopachrome but inhibited its decarboxylation. Dopachrome tautomerase also decreased decarboxylation. When metal ions and this enzyme act together, the inhibition of decarboxylation was greater than that produced by each agent separately, but amount of carboxylated units incorporated to the melanin was not higher than the amount incorporated in the presence of only cations. The amount of total melanin formed from tyrosine was increased by the presence of both agents. The action of Zn(II) was different from other ions also in the second phase of melanogenesis, and its effect on decarboxylation was less pronounced. Since tyrosine hydroxylation is the rate-limiting step in melanogenesis, Zn(II) inhibited the pathway. This ion seems to be the most abundant cation in mammalian melanocytes. Therefore, under physiological conditions, the regulatory role of metal ions and dopachrome tautomerase does not seem to be mutually exclusive, but rather complementary.

Introduction The biosynthesis of melanin, the main pigment found in mammalian skin, hair and eyes, proceeds through a complex series of enzymatic and chemical reactions. The first part of the pathway, which is fairly well understood [1-3], leads from L-tyrosine to dopachrome, and requires the presence of a single enzyme, tyrosinase [4]. The second part of the pathway, in the absence of sulphydryl compounds, leads from dopachrome to eumelanin, but the structure and control of the intermediates arising from dopachrome is only partially

Abbreviations: L-dopa, 3,4-dihydroxyphenylalanine;dopachrome, 2carboxy-2,3-dihydroindole-5,6-quinone; DHI, 5,6-dihydroxyindole; DHICA, 5,6-dihydroxyindole-2-carboxylic acid; IQ, indol-5,6° quinone; IQCA, indol-5,6-quinone-2-carboxylicacid; PMSF, phenyl° methylsulfonylfluoride. Correspondence: F. Solano, Departamento de Bioquimica y Biologia Molecular, Facultad de Medicina, Universidad de Murcia, 30100 Murcia, Spain.

known. There are many non-enzymatic steps, and in fact melanin can spontaneously be formed. Dopachrome may undergo an intramolecular rearrangement to give either D H I or D H I C A depending on whether the putative decarboxylation takes place at this stage or not (Fig. 1). Early studies postulated the extensive decarboxylation and the almost exclusive formation of DHI [1,5] but more recent studies have proved that D H I C A is also an important intermediate of the process leading to melanin containing abundant carboxylated indole units [6-10]. D H I and D H I C A might also be oxidized to the corresponding indolquinones, IQ or IQCA, and these reactions can also be catalyzed by tyrosinase [11]. Then, these intermediates polymerize and no stage between indol-quinones and the melanin polymer has been unequivocally characterized. Nevertheless, some dimeric structures have been postulated [12-14] to account for the transient purple melanochrome formed at that phase of melanogenesis. However, in the last decade other factors different from tyrosinase have been proposed to regulate mammalian pigmentation. Supporting their existence,

0304-4165/90/$03.50 © 1990 ElsevierScience Pubfishers B.V. (Biomedical Division)

277 eumelanogenesis only takes place inside the melanosome although the presence of active extramelanosomal tyrosinase isozymes has been reported [15,16]. In turn, there are numerous examples of lack of correlation between the degree of melanization and the tyrosinase content of melanocytes [17,18]. From a chemical point of view, two types of possible regulators of melanogenesis have been reported. On one hand, some transition metal ions have a remarkable effect on the formation rate and the final structure of melanin synthesized ‘in vitro’ [8,9,19]. Therefore, the involvement of metal ions in the control of melanin formation has been suggested 191. On the other hand, the existence of protein regulators different from tyrosinase has also been reported [10,20-221, and the involvement of at least one enzyme which accelerates dopachrome tautomerization has been unequivocally established [10,22]. The ‘in vivo’ effects of metal ions on melanogenesis might be important because of the high affinity of melanin for cations [23,24] and the high content (in the millimolar range) of some transition metal ions in melanin-containing mammalian tissues [25-271. Metal cations can modify both the structure and the conver-

sion rate of some intermediates of melanogenesis by several mechanisms: (i) these intermediates can form complexes with metal ions, either through the oxygenated groups of o-diphenols and o-quinones or through the indolic nitrogen and the proximal carboxylic group of some indoles; (ii) some reactions of the pathway are redox, and metal cations might act as catalysts or as direct reactants; (iii) tyrosinase is a copper-protein, and apotyrosinase can be activated by Cu(I1) ions [15]; (iv) some cations, such as Fe(II), can shorten the lag period of tyrosinase, probably changing the oxidation state of the copper at the enzyme active site [19]. However, few reports have been devoted to analyze the effect of metal cations on mammalian tyrosinase [15,28] and these only focused on the reconstitution of apotyrosinase. The data obtained using non-mammalian tyrosinases might be non-extrapolatable to the mammalian enzyme, since significant differences have been described in the effect of other ions, halides and polyamines, on tyrosinases from different sources [29,30]. Based on the available data, the contribution of metal ions and the new enzyme to the control of melanogenesis is difficult to assess. There is only one

‘

TVROSINE

DOPA

DoPACHRoME / \

DH’ Yt

LEUKODOPACHROME

DH’CA

EUMELANINS

Fig. 1. The eumelanin biosynthetic pathway. The first phase initiates from t-tyrosine and involves two tyrosinase-catalyzed consecutive oxidations and a coupled chemical reaction to yield dopachrome. The second phase initiates with the putative decarboxylation of dopachrome to yield diiydroxyindoles. The proportion of carboxylated to decarboxylated units regulates the formation rate and structure of the final pigment. Dopachrome tautomerase and metal ions enhance the formation of carboxylated units.

278 report comparing the effect of both the enzymatic protein and metal ions on dopachrome [10], concluding that the enzyme seems to be the major factor accounting for the formation of DHICA. In the preceding paper, we describe the isolation and partial characterization of this enzyme, dopachrome tautomerase [22]. This paper is devoted to study the effect of metal cations either alone or together with tyrosinase and dopachrome tautomerase on mammalian melanogenesis from an overall point of view. We report data showing that the regulation by dopachrome tautomerase could be important in the second part of the pathway, but does not exclude the involvement of melanin-associated metal ions to control the activity of tyrosinase to yield dopachrome in the first part of the pathway as well as the rate of the reactions involved in the second part. Furthermore, our results show that both agents, metal ions and tautomerase, could be complementary regulatory factors of mammalian melanogenesis. Materials and M e t h o d s

Reagents. Radioactive substrates, L-[U-14C]tyrosine (spec. radioact., 502.1 mCi/mmol), L-[3,5-3H]tyrosine (specific radioactivity 58.8 Ci/mmol) and DL-[114C]dopa (spec. radioact., 57.5 mCi/mmol) were obtained from New England Nuclear (Boston, U.S.A.). CO 2 absorber was purchased from Packard (Zurich, Switzerland). L-Tyrosine, L-dopa, synthetic melanin, BSA, chloramphenicol, cycloheximide, sodium penicillin G, trypsin, PMSF, Hepes, EDTA, phenylthiourea and Brij 35 were from Sigma Chemical Co. (St. Louis, MO, U.S.A.). Silver oxide, sodium monobasic and dibasic phosphates, sodium hydroxyde, perchloric acid, trichloroacetic acid, sucrose and diethyldithiocarbamate were from Merck (Darmstadt, F.R.G.). Celite 535 was from Koch-Light Lab. Ltd. (Berkshire, UK). Toluene, ethanol, acetone, glycerol, HC1, KOH, activated charcoal and the inorganic salts NalO 4, ZnSO4.7H2 O, NiSO4 • 7H20, CuSO4.5H20 and CoC12 • 6H20 were from Probus (Spain). DEAE-Cellulose and 3MM filter papers were from Whatman (Kent, U.K.). Chelex 100 was from Bio-Rad (Richmond, CA, U.S.A.). CM-Sephadex was from Pharmacia and Ultrogel AcA 34 from LKB (Sweden). All solutions were prepared using double-distilled water passed through a MiUi-Q Waters system. The final resistivity of that water was higher than 10 MI2. cm. All reagents were of the highest purity commercially available (except ethanol and acetone for washing of filter-paper disks) and were used without further purification. Animals and melanomas. B16 mouse melanoma melanocytes were originally a kind gift from Dr. V. Hearing (NIH, Bethesda, U.S.A.). They have been maintained by serial transplantation on hybrid mice

obtained from male DBA and female C57/B1 (Panlab, Spain). Only male mice at 6-8 weeks of age were used for tumor transplantation, and they were injected subcutaneosly with approx. 105 viable cells. After 3-4 weeks visible tumors were excised, and then some were used for new implantation and the others for enzymatic preparations. Preparation of tyrosinase and dopachrome tautomerase. Crude melanosomes were separated by differential centrifugation as described elsewhere [22,31]. All centrifugations were rapidly carried out at 4°C in order to minimize endogenous proteolysis [32]. Briefly, melanomas were homogenized in 10 mM sodium phosphate buffer pH 6.8 containing 0.25 M sucrose and 0.1 mM EDTA (1 : 4 w/v), using a Polytron (power setting at 7). Unbroken cells and nuclei were discarded after centrifugation at 700 x g in a Sorvall SS-34 rotor for 20 min and the supernatant was centrifuged at 11 000 x g for 30 min in the same rotor. The resulting melanosome pellet was resuspended in the same buffer, washed, centrifuged at the same conditions and considered as a source of melanosomal proteins. Tyrosinase was prepared by two different methods. Firstly, tyrosinase was obtained by non-proteolytic solubilization, resuspending a fraction of melanosomes in 10 mM phosphate or Hepes buffer (pH 6.8) containing 1.5% Brij 35 and processed according to [15] except that the purification step on gel filtration chromatography was carried out on a Ultrogel AcA 34 column (52 X 2.6 cm). Alternatively, to solubilize tyrosinase by proteolytic treatment, the melanosomal fraction was resuspended in 10 mM phosphate or Hepes buffer (pH 6.8) and solubilized by incubation for 2 h at 37°C with 0.5 m g / m l trypsin. After incubation, the samples were placed on ice and PMSF was added up to 1 mM to stop the proteolytic digestion. Then, the suspension was centrifuged at 105 000 x g for 1 h in a 60 Ti Beckman rotor. Further purification of tyrosinase was performed according to Jara et al. [31] but using Ultrogel AcA 34 for gel filtration chromatography. The trypsin-solubilized tyrosinase was mostly used to perform the melanin formation assays, because of the inhibition caused by detergents on this technique [33]. To obtain dopachrome tautomerase, melanosomes were resuspended in 10 mM phosphate buffer (pH 6.8) containing 1% Brij 35, and processed as described in the preceding paper [22]. Tyrosine hydroxylase assay. Incubations were carried out as described by Jara et al. [31]. Briefly, each assay (final volume 50 ttl) contained 10 #1 of 0.25 mM Ltyrosine (50 ttCi/ml of L-[3,5-3H]tyrosine), 10 #1 of a solution containing chloramphenicol (1 mg/ml), cycloheximide (1 mg/ml), penicillin G (1000 units/ml), bovine serum albumin (0.1 mg/ml) and L-dopa 0.05 mM, 5-20/~1 of tyrosinase and the volume was brought to 50/~1 with solutions of metal ions in distilled water or

279 other agents to be tested. The assays were routinely run in Eppendorf tubes at 37°C for 1 h. After incubation, 50 mg activated charcoal, 50 mg Celite 535 in 450/tl of 1% trichloroacetic acid were added to the sample tubes. These were shaken at room temperature for 1 h, centrifuged at 8000 x g for 5 min and 100 /xl aliquots removed and counted for radioactivity. Blanks included samples without enzyme. One unit of tyrosinase was defined as the amount that catalyzes the hydroxylation of 1/~mol L-tyrosine/min. Melanin formation assay. Incubations were set up as described for the tyros±he hydroxylase assay, except for the substrate which was 10/~l of 0.25 mM L-tyrosine (25 # C i / m l of L-[U-14C]tyrosine). The assays were routinely run at 37°C for 1 h in wells of a microtiter plate. After incubation, the samples were removed to Whatman 3MM filter-paper discs (2.3 cm diameter). The filters were washed several times with 0.1 M HC1, ethanol and acetone, as described elsewhere [31] then allowed to dry, and counted for radioactivity. Blanks were made and subtracted from sample counts to obtain net clam. One unit of tyrosinase using this assay was defined as the amount of enzyme that catalyzes the incorporation of 1 /~mol L-tyrosine into melanin/min. Determination of dopachrome tautomerase activity. Dopachrome tautomerase was determined spectrophotometrically by monitoring the decrease in absorbance at 475 nm due to the conversion of dopachrome into D H I C A [21,22]. Blanks were made without enzyme and substracted to obtain net activity. One unit of dopachrome tautomerase was defined as the amount of enzyme that catalyzes the isomerization of I/~mol d o p a c h r o m e / min at 30°C. Dopachrome had to be chemically prepared 'in situ' by L-dopa oxidation, because of its unstability. To do that, L-dopa was dissolved in 10 m M sodium phosphate buffer, (pH 6.0) and treated with periodate ( 1 : 2 , d o p a / N a I O 4 stoichiometry). When carboxylic-labeled dopachrome had to be prepared, OL[1-14C]dopa was isotopically diluted with the appropriate amount of L-dopa depending on the required concentration, and the mixture was treated with NaIO 4 as before. Destination of the carboxylic group of [1-t4C]dopa during melanogenesis. Decarboxylation during melanogenesis was determined by measuring the 14CO 2 evolved either from the tyrosinase-catalyzed dopa oxidation or directly from newly synthesized dopachrome. The reactions were carried out at 30°C in sealed tubes with rubber stoppers with two Whatman 3MM paper discs ( 1 / 4 inch diameter) containing 20 #l of CO2 absorber. The total volume of the standard mixture was 100-200 /tl containing 0.5 mM L-dopa and 0.125 /~Ci of OL-[114C]dopa or 0.25 mM dopachrome containing an amount of [1-14C]dopachrome specified in figure legends. When tyrosinase was present in the assay, the reaction was stopped by the addition of 100/xl of 1 M

citric acid. Then, paper discs were immediately removed and counted for radioactivity. The reaction mixtures remaining in the assay tubes were applied on Whatman 3MM filter paper discs (2.3 cm diameter) and washed as described in the melanin formation assay to measure the amount of radioactive carboxylic groups incorporated to the melanin. Protein determination. The protein content was determined by a modified Lowry method [34] using BSA as standard. Melanin determination. The melanin content of tumors was determined by measuring the absorbance at 400 nm of melanin digested with hot 0.85 M K O H [35], using synthetic melanin as standard. R ~ We firstly describe the effect of metal ions on the tyrosinase-catalyzed tyrosine hydroxylation; secondly, the effect on dopachrome decarboxylation and on the destination of the carboxylic group, in the presence or absence of dopachrome tautomerase; and finally the effect on the whole pathway, using tyrosine as initial substrate and tyrosinase with or without dopachrome tautomerase in the reaction media. Effect on tyrosine hydroxylation Table I shows the effect of 1 mM Co(II), Ni(II) and Zn(II) on the melanoma tyrosinase-catalyzed tyrosine hydroxylation in 10 mM sodium phosphate and Hepes buffers, (pH 6.8). Both types of buffer were first used in order to detect any possible artifacts which could mask the results due to precipitation of insoluble phosphates even at the low concentration of buffer used. However, turbidity was not observed in any case and the similarity between both series suggests that this possibility should be discarded. The slightly lower tyrosinase activity observed in Hepes media in comparison to phosphate buffer has been already reported [36], although it remains unexplained. It could be due to a side reaction of the nascent dopaquinone with Hepes, since parallel spectrophotometric experiments showed the appearance TABLE I Effect of metal ions (1 raM) on the tyrosine hydroxylase activity of mouse melanoma tyrosinase All data are expressedin ~tunits as defined in the tyrosinehydroxylase assay. They are the mean± S.D. of three determinations. Activities measured in 50 #l total volumeof 10 mM sodium phosphate or Hepes buffer, (pH 6.8). The amount of tyrosinase was 4.8/tg in each assay. Cation Control Zn(II) Ni(II) Co(II)

Pi 3.43+ 0.90 0.40 ± 0.19 7.05± 0.86 4.52 ± 1.0]

Hepes 3.06± 0.56 0.56 + 0.19 6.60± 0.91 3.96± 0.67

280 in this medium of a compound with an absorbance peak at 408 nm (not shown) instead of the dopachrome ()kma x 475 nm) obtained in phosphate media. About the action of metal ions, Ni(II) and Co(II) stimulated the hydroxylation activity, but conversely Zn(II) inhibited that first step of melanogenesis. In order to characterize the stimulatory effect of cations, the activation produced by Ni(II) was further studied. Incubation of Ni(II) with either substrate or L-dopa before the reaction did not affect the activation under a variety of experimental conditions. Ni(II) could not replace L-dopa as cofactor of the hydroxylase reaction. Moreover, 5 mM EDTA abolished the effect of Ni(lI) and this cation could not restore any activity from tyrosinase previously inhibited by incubation with 5 mM phenylthiourea or diethyldithiocarbamate. The kinetic parameters of tyrosinase also varied in the presence of Ni(II). According to the observed stimulatory effect, 1 mM Ni(II) increased the Vmax. However, the K m also increased from 44 to 216/~M (Fig. 2), decreasing the affinity of the enzyme for the substrate.

J 2.104

o

5

v

c~

,'/i

•

JJ

.

Time (h)

Fig. 3. CO2 evolved from [1-14C]dopachrome spontaneously ( o , blank)

or in the presence of 1 mM Ni(II) (O), 810 /~unit of dopachrome tautomerase (zx) and both agents (A). The concentration of dopachrome was 0.25 mM and the total cpm in each assay were around 30000. The assay buffer was 10 mM phosphate buffer (pH 6.0) at 30°C. The results are the mean of two determinations,with a variability always lower than 5%.

Effect on dopachrome decarboxylation The effect of metal cat ions on the rate and extent of decarboxylation during melanogenesis was studied in two different systems depending on the initial substrate used: firstly, [1-14C]dopachrome; alternatively, [1~4C]dopa plus tyrosinase was used. [1-a4C]Dopachrome was prepared immediately before its use by stoichiometrical oxidation of [1-14C]dopa with NaIO 4. The effect of 1 m M Ni(II), dopachrome tautomerase or both agents on the kinetics of dopachrome decarboxylation is shown in Fig. 3. Interestingly,

o

,

.2

1,~ (OH") Fig. 2. Double reciprocal plots of the tyrosine hydroxylase activity of melanoma tyrosinase in the absence ( o ) or the presence of 1 m M Ni(II) (O). It can be seen that the cation increases the Vm~x (from 0.17 to 0.77 munit) but also the K m for tyrosine (from 44 to 216/~M Thus, the Vmax//Km ratios in both cases were very similar. All parameters were obtained by fitting the data with a linear regression program.

both Ni(II) and the enzyme partially inhibited the amount of CO 2 evolved from dopachrome, but the kinetics was very different at the initial stage of reaction. Ni(II) gave place to a rapid CO 2 release, even higher than the amount of CO 2 released in the blank, but after approx. 2 h an almost complete inhibition of CO 2 release occurred. Dopachrome tautomerase also inhibited the amount of CO 2 evolved, but in this case the process was well-fitted to a first-order kinetics with lower rate than that of the blank. Finally, the presence of both agents gave rise to a higher inhibition of the CO 2 release than the one observed for each agent acting separately. For the other metal ions, the pattern of CO/evolution in the early stage of dopachrome decomposition was similar. Thus, during the first 30 rain, Co(II) and Cu(II) also yielded a release of CO2 higher than in their absence. However, Ni(II) was the ion that yielded the maximal inhibition of CO 2 release at the end of the reaction. The effect of Zn(II) was the most dramatic one, and in the presence of dopachrome tautomerase, the CO 2 release due to Zn(II) was slightly higher than in the absence of the cation during all the time course of the experiment. Fig. 4 shows the CO 2 release, the total amount of carboxylic compounds remaining in the assay mixtures and the insoluble carboxymelanin formed from [114C]dopachrome after 24 h in the absence or the presence of dopachrome tautomerase and several metal ions. Separate experiments adding citric acid to the assay media demonstrated that all the free CO 2 is evolved in these conditions and did not remain in the solution. All assays were performed in duplicate tubes.

281 Thus, the assay mixture of one tube was counted directly to determine the total amount of radioactive carboxylic compounds in solution. The reaction mixture of the duplicate tube was pipetted on a 3MM paper disk and allowed to dry. Then, papers were washed as described for the melanin formation assay to determine the amount of insoluble carboxymelanin formed. The data show that dopachrome tautomerase and all metal ions were able to decrease dopachrome decarboxylation. Moreover, the inhibition was always higher in the presence of both dopachrome tautomerase and ions in comparison to each agent separately, but effects were not totally additive. The action of Zn(II) on the decarboxylation was less pronounced, and in the presence of dopachrome tautomerase the total CO 2 evolved was even higher than in the blank. Logically, the total amount of radioactive carboxylic groups in the assay mixtures was higher as the radioactive CO 2 evolved was lower, the two quantities being complementary. The radioactivity incorporated to melanin was always much lower than the amount of radioactivity associated to the carboxylic compounds remaining in the assay mixtures. Since DHICA is the major intermediate formed in the system, these results would indicate that this compound is fairly stable, slowly incorporated to the polymer, and washable from the paper disks by the acid treatment. With regard to the effect of dopachrome tautomerase and cations on the formation of carboxymelanin, the situation was complex. In the absence of any ion or in the presence of Zn(II), the amount of carboxylated units incorporated to melanin was enhanced by the presence of the tautomerase, but

c p m l O -3

i

2O i

± 2

2

c..t.~

I

c. (.)

3

1

c. (.)

NI

0t)

z.

(.)

Fig. 4. C O 2 evolved (1), carboxylic compounds remained in the assay mixtures (2) and insoluble carboxymelanin (3) formed from [114C]dopachrome after 24 h in the presence (solid bars) or absence (empty bars) of dopachrome tautomerase (405/~unit) and 1 mM metal ions. The concentration of dopachrome was 0.25 mM and the total cpm in each assay were approx. 31000. Assay buffer as in Fig. 3. The results are the mean of three determinations.

TABLE II

Effect of metal ions (1 mM) on decarboxylation during melanogenesis [1J4C]Dopa (0.125 /~Ci, 0.5 mM) was incubated with 4.8 ~g of purified mouse melanoma tyrosinase for 1 h in a total volume of 100 /~1 to determine decarboxylation during the melanin formation. Assay media were 10 mM sodium phosphate buffer (pH 6.8). All data are expressed as cpm and are the mean + S.D. of three determinations. Cation

14CO2

[14C]Carboxymelanin

Control Zn(II) Ni(II) Co(II)

21 119 -t- 923 798+231 9 227 + 923 18699+930

9623 + 1077 1405+ 93 12 386 + 1098 13837+ 692

unexpectingly this enzyme slightly decreased the incorporation of carboxyl groups when Cu(II), Co(II) and Ni(II) were present in the reaction mixture. Thus, no correlation between carboxylic groups retained in the reaction media and formation of carboxylated melanin was observed in these cases. It has been described that hydrogen peroxide is formed during melanogenesis. This endogenously-generated oxidant may disrupt some indole units, and some carbon atoms of the framework molecule might be released as CO 2 [37]. Therefore, the possible effect of catalase (0.1 mg/ml) on the kinetics of the CO2 release and on the amount of carboxymelanin formed was tested. However, in our conditions the presence of this enzyme had no effect, either in the absence or presence of dopachrome tautomerase, as compared with the kinetics shown in Fig. 3. The experiments described above are not totally extrapolatable to the 'in vivo' situation since the substrate used was dopachrome chemically formed. Therefore, we performed another series of experiments using [114C]dopa as initial precursor and tyrosinase to catalyze dopa oxidation. Table II shows the CO 2 evolved and the radioactivity incorporated to melanin after 1 h of incubation in the presence or the absence of 1 mM metal ions. Maximal decarboxylation was observed in the absence of these cations. The incorporation of carboxylic units to the pigment was enhanced in the presence of Co(II) or Ni(II), and inhibited by Zn(II). It is interesting to note that Ni(II) produced a higher inhibition of decarboxylation than Co(II), but the incorporation of carboxylated units to the carboxymelanin was lower. Two effects should be involved in these data. On one hand, it is known that metal ions somehow inhibit decarboxylation and enhance the incorporation of carboxylic groups to the melanin [9,10]. On the other hand, Zn(II) inhibits mammalian tyrosinase. Thus, in the presence of Zn(II), dopa oxidation is partially blocked and further stages of the melanin formation would be inhibited. In addition, the determination of radioactive melanin using [1Jac]dopa or [1J4C]dopachrome estimates the

282 TABLE III

Effect of metal ions (1 mM) and dopachrome tautomerase (DC T-ase) on melanin formation from L-tyrosine

f

All data are expressed in #units as defined in the melanin formation assay. Activities were measured in 50 #1 total volume of 10 mM sodium phosphate buffer (pH 6.8) containing 0.25 #Ci of L-[U14C]tyrosine. The amount of tyrosinase was 30 /~g in each assay. Numbers in brackets indicate the percentage of activity respect to the control of that series, and the column on the fight the percentage of activation produced by the presence of dopachrome tautomerase (648 ~tunit). Cation Control Zn(lI) Ni(II) Co(lI)

DC T-ase

+ DC T-ase

% Act.

2.40 0.24 (10%) 4.12 (172%) 3.80 (158%)

4.43 0.94 (21%) 4.85 (109%) 4.90 (110%)

184 384 117 129

-

carboxymelanin formed, but it is not a good estimation of the total melanin formation. Thus, we performed another series of experiments, using uniformly labeled L-tyrosine as initial substrate, in order to study the effect of metal ions and tautomerase on the whole pathway.

Melanin formation The effects of Co(II), Ni(II) and Zn(II) on the rate of the overall pathway from L-tyrosine to melanin, in the presence or the absence of dopachrome tautomerase, are shown in Table III. With regard to the ions, the data are consistent with those obtained for the tyrosine hydroxylase step (Table I). Ni(II) and Co(II) enhanced melanin formation, whereas Zn(II) inhibited that synthesis. The same pattern could be seen in the presence of dopachrome tautomerase, but in this case the activations produced by Ni(II) and Co(II) were much lower than in the absence of the enzyme. Thus, both dopachrome tautomerase and these metal ions acted as accelerating factors of the total amount of melanin formed. Although both agents shared the same role, their actions were not totally additive. Furthermore, the percentage of activation of both agents in these conditions depended on the tyrosinase activity initially added. The lower was the tyrosinase activity present, the higher was the activation by cations and tautomerase. Accordingly, when the tyrosinase activity was low (4.8 /~g) ions such as Ni(II) could increase approx. 6-fold the melanin formation [33], but when the initial tyrosinase activity was high (30 #g), the same concentration of this ion increased less than two-fold the amount of melanin formed (Table III). Moreover, the activation of dopachrome tautomerase was maximal when the activity of tyrosinase was low because of the inhibition produced by the presence of Zn(II). Thus, it seems that all agents are able to increase melanin formation, but the lower is the polymerization rate, the higher is the activation they can induce.

.-

/ -7.

c LU

+Zn 1

2

Cation concentration

mM

Fig. 5. Effect of Ni(II) (I) and Zn(II) (©) concentrations on the melanin formation activity of mouse melanoma tyrosinase (4.8 gg). Reaction media were 10 mM sodium phosphate buffer, (pH 6.8) containing 0.25 #Ci of L-[U-14C]Tyr, as detailed in Materials and Methods.

The activation by Ni(II) was dependent on its concentration (Fig. 5), showing a maximum around 1 mM. Other effects of Ni(II) on the melanin formation assay were parallel to those above reported on the tyrosine hydroxylation. The inhibition caused by Zn(II) was also depended on its concentration. Bearing in mind the effect of Zn(II) on melanogenesis, and to gain some insight on the possible significance of the regulatory effects of metal ions 'in vivo', the Zn(II) and Cu(II) contents of two heavily and two lightly pigmented melanomas were measured. These melanomas were digested using highly purified nitric acid, and analyzed by atomic absorption spectropho-

TABLE IV

Content of some cations in melanomas with different grade of pigmentation Melanomas were firstly classified as heavily- or lightly-pigmented by visual observation. They were weighed and their melanin content determined by digestion with hot KOH. Metal content was determined by atomic absorption spectrophotometry. Pigmentation grade

Melanin (mg/g wet tissue)

Cation content ( # g / g wet tissue) Cu Zn

Cu Zn

Heavy Light

0.59+0.18 0.17 + 0.06

1.25+0.01 1.30 + 0.05

0.078 0.067

16.12+0.69 19.47 + 0.82

283 tometry. In addition, pieces of these tissues were homogenized and digested in hot KOH for melanin determination. The results expressed in Table IV show that the content of Cu(II) does not change significantly, but the content of Zn(II) appeared to be slightly higher in the poorly pigmented tissues. Discussion

There are significant differences in the effect of metal ions on melanogenesis. These differences depended on the nature of the ion tested, the step of the pathway which is followed, and the presence of dopachrome tautomerase in the reaction mixture. With regard to the first step of the pathway, the tyrosine hydroxylation, Co(II) and Ni(II) enhance this activity, but Zn(II) inhibits it. In the second part of the pathway, the presence of metal ions also increases the rate of dopachrome conversion into indoles, inhibits decarboxylation and enhances the incorporation of carboxylated precursors into the final pigment. At this stage, the effect of Zn(II) is less pronounced and the 14CO2 evolved in its presence is approx, equal to the 14CO2 evolved in the control. In agreement with these data, Ni(II) and Co(II) stimulated the tyrosinase-catalyzed melanin formation from Ltyrosine and from L-dopa, but the inhibition of Zn(II) on the tyrosine hydroxylation yielded an inhibition of the melanin formation. These data are only partially similar to those recently reported [8-10,19]. Beginning in the first step, it has been described that the steady-state rate of the tyrosine hydroxylase activity of cuttlefish tyrosinase was not affected by all these cations [19], but our results indicate a different behaviour for melanoma tyrosinase. However, similar differences in the effects of other ions on mammalian or non-mammalian tyrosinases due to different interactions with the enzyme active site have been reported [29,30]. Moreover, the standard ion concentration used in this work (1 mM) was significantly higher than the one employed for cuttlefish tyrosinase (50 /~M). Experiments using the same conditions and assay technique are needed to clarify this point. The activation caused by Ni(II) and Co(II) might be derived from a direct reduction of the copper at the tyrosinase active site. A similar effect has been reported for Fe (II) and cuttlefish tyrosinase [19]. However, the standard redox potential for the Fe(II)/Fe(III) couple is lower than the potentials of the Ni and Co couples, thus making unlikely this effect. Unfortunately, in our hands Cu(II) and Fe (II) gave very variable results depending on the preparation of tyrosinase used, and they could not help us to clarify this point. In spite of that, the increase in the kinetic parameters of tyrosinase caused by the presence of Ni(II), keeping approx, constant the V m a x / K m ratio, could indicate that the cation might interact with the substrate, being Tyr-Ni(II) c o m -

plexes the true substrates in those conditions. The inhibitory effect of Zn(II) on the reaction would be different, and an inactivating action of this ion on melanoma tyrosinase should not be rule out. To finish this point, metal ions may inhibit (the case of Zn(II)) or activate (the others) the rate-limiting step of the pathway, the tyrosinase hydroxylation. Thus, they may regulate dopachrome formation. With regard to the second part of the pathway, dopachrome tautomerase catalyzes dopachrome tautomerization to DHICA, and inhibits decarboxylation [22]. Metal ions accelerate dopachrome decomposition to yield both indoles, DHI and DHICA [8]. Therefore, they lead to a mixture of both decarboxylated and carboxylated units, the proportion depending on the nature of the ion. The combined action of cations (aside Zn(II)) and the enzyme always produced an inhibition of decarboxylation higher than the one produced by each agent separately. However, the amount of carboxymelanin formed by the action of both agents was not higher than the amount formed in the absence of the tautomerase. Metal ions could increase the number of carboxylated units incorporated to melanin in the presence of the tautomerase, but the opposite was not true. The significance of these data could be important. Let us compare the effects among similar cations. It has been shown, by spectrophotometric measurements, that metal ions increase the rate of dopachrome decomposition in the order Cu(II) > Ni(II) > Co(II) > Zn(II) [8]. It has also been reported, by HPLC analysis, that the rates of the metal-induced tautomerization of dopachrome to DHICA are the order Ni(II) > Cu(II) > Co(II) [8,10]. This permutation of C u / N i indicates that there is no correlation between dopachrome decomposition and DHICA formation. A similar conclusion can be reached from the effects of Ni(II) and Co(II) on [114C]dopa decarboxylation and carboxymelanin formation in the presence of tyrosinase (see Table II). Furthermore, the radiolabeling technique allowed us to estimate both the total carboxylated compounds remaining in solution and the carboxymelanin formed at the reaction media, the difference between both quantities being the amount of carboxylated units that are washable by acid (mostly DHICA monomers). Thus, an estimation of the incorporation of carboxylated indoles to the melanin was obtained. Cations and dopachrome tautomerase inhibited decarboxylation and increased the formation of carboxylated compounds, their effects being partially additive, according to Ref. 10. However, their effects on the incorporation of these carboxylated units to the melanin were not additive at all, and dopachrome tautomerase failed to increase the amount of carboxymelanin formed in the presence of cations. In addition, the yields of carboxylic incorporation were in the order Cu(II)>~ Ni(II)> Co(II) if only the cations

284 occurred at the reaction media, but changed to Co(II) >~ Ni(II) > Cu(II) if the enzyme was also added. Note the change of Cu(II), the most efficient cation in the acceleration of dopachrome decomposition. It appears that the higher is the rate of dopachrome decomposition due to the presence of cations, the higher is the effect of dopachrome tautomerase in decreasing the incorporation of carboxylated units to melanin. To summarize this point, our data strongly suggest that dopachrome tautomerase and Co, Cu and Ni would share only partially the same role in the second part of melanogenesis. These metal ions, obviously to different degrees, accelerate both the tautomerization of dopachrome into DHICA, and the posterior polymerization of indoles. The enzyme also accelerates the tautomerization, but no further reactions. The relative stability of the product, DHICA, justifies the apparent indole blocking factor activity firstly attributed to dopachrome tautomerase [211. The combined effect of cations and tautomerase on total melanin formation from [U-14C]tyrosine was similar. Both agents together produced a higher stimulation of the melanin formation than each one separately, but the effects were not completely additive. However, in these conditions the amount of melanin formed was increased by dopachrome tautomerase even in the presence of Ni(II) and Co(II). Furthermore, the lower the tyrosinase activity, the higher was the stimulation produced by both agents. Zn(II) again differed, and although this ion produced an inhibition of melanin formation, its effect was lower in the presence of the tautomerase. Further studies changing the ratio of concentrations between the enzymes and the ions are needed in order to clarify the role of each agent and the differences between the results obtained by measurement of carboxymelanin formation from dopachrome or dopa and total melanin formation from tyrosine. Zn(II) deserves special discussion from a double point of view. Firstly, its action is different from the other ions. Although the inhibition by Zn(II) on the tyrosine hydroxylase activity of mouse melanoma tyrosinase seems to be its major effect on melanogenesis, its effect on the second part of melanogenesis is also different. Some features of Zn(II) consistent with our results on CO 2 evolution have been previously reported (i) HPLC analysis showed that Zn(II) had minimal influence in the DHICA formation from dopachrome [10]; (ii) the formation of dopa-melanin had a low yield in the presence of Zn(II) in comparison to Ni(II), Cu(II) and Co(II) [9]; (iii) the synthesis of DHI derivatives from dopa can be performed by treatment with an excess of Zn acetate [38]; and (iv) Zn(II) accelerated the tyrosine-catalyzed formation of melanochrome from Ldopa, whereas other ions did not [36]. Secondly, from a physiological point of view, Zn(II) and Cu(II) are probably the most abundant cations in

the mammalian melanosome [25-27]. Although Cu(II) might contribute to control the rate of the second part of melanogenesis, the reconstitution of holotyrosinase should be the major role of this cation [15]. Thus, Cu(II) and Zn(II) should have opposite effects on mammalian melanogenesis, and the ratio Cu(II)/Zn(II) might be an important parameter to regulate melanization in the mammalian melanosome. It has been thought that all metal ions could accelerate melanin formation, but never inhibit the pathway. Taking into account our results, an inverse relationship between the content of Zn(II) and the degree of pigmentation should be observed. Preliminary estimations of the Cu(II) and Zn(II) content in heavily- and lightly-pigmented melanomas showed that the Zn(II) content was slightly higher in the poorly-pigmented tissues, although the significance of these preliminary data is uncertain, since the increase in the Zn(II) content seems to be small. To clarify this point definitively, a more detailed study of the Cu(II) and Zn(II) content in melanotic and amelanotic melanomas should be carried out. In turn, any speculation about the roles of Co(II) and Ni(II) in vivo awaits the measurement of the content of these cations in melanoma. The actions of these two ions, particularly Ni(II) show more similarity to the one of dopachrome tautomerase [10] but their in vivo levels seem to be very low. In summary, mammalian melanogenesis is simultaneously a simple and complex pathway. It is simple since the melanin polymer can be spontaneous obtained from L-dopa, and it is complex because the rate of some steps of the pathway can be regulated by a variety of agents, including metal ions and dopachrome tautomerase. Although activation of some steps of the melanogenesis caused by those agents have been described [8-10,20-22], a knowledge of the metal cations and dopachrome tautomerase contents in a variety of melanotic and amelanotic melanomas is crucial in order to obtain an integrated view of the process and to assess the physiological significance of each agent. So far, metal cations and dopachrome tautomerase do not seem to be mutually exclusive regulatory mechanisms. On the contrary, they could be complementary controls of this biosynthetic pathway.

Acknowledgements This work has been partially supported by grant no. PB87-0698 from the DGICYT, Spain. P. Aroca thanks the 'Comunidad Autonoma de Murcia' for a fellowship. The authors thank the 'Laboratorio Agrario Regional' of Murcia for allowing us to determine the Cu and Zn in melanoma by atomic absorption spectrophotometry (Varian AA-20).

285

References 1 Mason, H.S. (1948) J. Biol. Chem. 172, 83-99. 2 Garcia-Carmona, F., Garcia-Chnovas, F. and Lozano J.A. (1982) Biochim. Biophys. Acta 717, 124-13. 3 Cabanes, J., Garcia-CAnovas, F., Lozano, J. A. and GarciaCarmona, F. (1987) Biochim. Biophys. Acta 923, 187-195. 4 Hearing, V.J. and Jimenez, M. (1987) Int. J. Biochem., 19, 11411147. 5 Bu'Lock, J.D. (1960) Arch. Biochem. Biophys. 91, 189-193. 6 D'Ischia, M., Palumbo, A. and Prota G. (1985) Tetrahedron Lett. 2801-2804. 7 Ito, S. (1986) Biochim. Biophys. Acta 883, 155-161. 8 Palumbo, A., D'Ischia, M., Misuraca, G. and Prota, G. (1987) Biochim. Biophys. Acta 925, 207-209. 9 Palumbo, A., D'Ischia, M., Misuraca, G., Prota, G. and Schultz, T.M. (1988) Biochim. Biophys. Acta 964, 193-199. 10 Leonard, L.J., Townsend, P. and King, R.A. (1988) Biochemistry 27, 6156-6159. 11 Korner, A. and Pawelek, J. (1982) Science 217, 1163-1165. 12 Bu'Lock, J.D. (1960) Arch. Biochem. Biophys. 91,189-193. 13 Palumbo, P., d'Ischia, M. and Prota, G. (1987) Tetrahedron, Lett. 43, 4203-4206. 14 Napolitano, A., Corradini, M.G. and Prota G. (1985) Tetrahedron Lett. 26, 2805-2808. 15 Martinez, J.H., Solano, F., Arocas, A., Garcia-Borr6n, J.C., Iborra, J.L. and Lozano J.A. (1987) Biochim. Biophys. Acta 923 413-420. 16 Hearing, V.J., Nicholson, J.M., Montague, P.M., Ekel, T.M. and Tomecki, K. (1978) Biochim. Biophys. Acta 522, 327-339. 17 Logan, A. and Weatherhead, B. (1978) J. Invest. Dermatol. 71, 295-298. 18 Jimenez, M., Kameyama, K. Maloy, W.L., Tomita, Y. and Hearing, V.J. (1988) Proc. Natl. Acad. Sci. U.S.A., 85, 3830-3834.

19 Palumbo, A., Misuraca, G., D'Ischia, M. and Prota, G. (1985) Biochem. J., 228, 647-651. 20 Pawelek, J., Korner, A., Bergstrom, A. and Bologna, J. (1980) Nature, 286, 617-619. 21 Barber, J.I., Townsend, P., Olds, P. and King, R.A. (1984) J. Invest. Dermatol. 83, 145-149. 22 Aroca, P., Garcia-Borron, J.C., Solan% F. and Lozano, J.A. (1990) Biochim. Biophys. Acta 1035, 266-275. 23 Potts, A.M. and Au, P.C. (1976) Exp. Eye Res. 22, 487-491. 24 Larsson, B. and Tjalve, H. (1978) Acta Physiol. Scand. 104, 479-484. 25 Bowness, J.M. and Morton, R. A. (1953) Biochem. J. 53, 620-626. 26 Dorea, J.G. and Pereira, S.E. (1983) J. Nutr. 113, 2375-2381. 27 Horcicko, J. Borovansky, J. Duchon, J. and Prochazkova, B. (1983) Hoppe-Seyler's Z. Physiol. Chem. 354, 203-204. 28 Lerner, A.B., Fitzpatrick, T.B., Calkins, E. and Summerson, W.H. (1950) J. Biol. Chem. 187, 793-802. 29 Martinez, J.H., Solano, F., Pe/aafiel, R., Iborra, J.L. and Lozano, J.A. (1986) Comp. Biochem. Physiol. 83B, 633-636. 30 Galindo, J.D., Martinez, J.H., Lopez-Ballester, J.A., Pehafiel, R., Solano, F. and Lozano, J.A. (1987) Biochem. Int. 15, 1151-1158. 31 Jara, J.R., Solano, F. and Lozano, J.A. (1988) Pigment Cell Res. 1, 332-339. 32 Martinez, J.H., Solano, F. and Lozano, J.A. (1989) Cell. Biochem. Funct. 7, 21-26. 33 Aroca, P., Jara, J.R., Blazquez, A., Garcia-Borr6n, J.C. and Solano, F. (1989) J. Biochem. Biophys. Methods 19, 327-338. 34 Hartree, E. (1972) Anal. Biochem. 48, 422-427. 35 Lotan, R. and Lotan, D. (1980) Cancer Res. 40, 3345-3350. 36 Vachtenheim, J. Duchon, J. and Matous, B. (1985) Anal. Biochem. 146, 405-410. 37 Swan, G.A. and Wright, D. (1954) J. Chem. Soc. 381-384. 38 Harley-Mason, J. and Bu'Lock, J.D. (1950) Nature 166, 1036-1037.