Incorporation of sulfhydryl compounds into melanins in vitro

Biochimica et Biophysica Acta 964 (1988) 1-7

1

Elsevier BBA22867

Incorporation of sulfhydryl compounds into melanins in vitro Shosuke Ito

a,

Yoichiro Imai

a,

Kowichi Jimbow b and Keisuke Fujita a

a School of Hygiene and Institute for Comprehenswe Medical Science, Fujita-Gakuen Health University, Aichi and b Sapporo Medical College, Sapporo (Japan)

(Received 29 June 1987)

Key words: Melanin; 3,4-Dihydroxyphenylalanine; Cysteine; Glutathione; Tyrosinase

Cysteine is known to be involved in pheomelanin synthesis. Available evidence, however, indicates that glutathione may also play an important role in melanogenesis. This in vitro study clarifies how cysteine and glutathione participate in melanogenesis. Melanins were prepared by tyrosinase oxidation of 3,4-dihydroxyphenylalanine (dopa) with various amounts of cysteine or glutathione, and were subjected to HCI hydrolysis. Synthetic melanins and the hydrolyzed melanins gave N / S molar ratios corresponding to those calculated from the ratio of dopa to cysteine or glutathione. On hydriodic acid hydrolysis, dopa plus cysteine-melanins and dopa plus glutathione-melanins gave aminohydroxyphenylalanineand cysteine, respectively, the yields of which were proportional to the sulfur content. The results indicate that cysteine is integrated into benzothiazine units (pheomelanins) while glutathione is connected to dihydroxyindole units (eumelanins) with the retention of glutathione moiety. Analysis of natural melanins, prepared by HCI hydrolysis of Sepia, BI6, and Harding-Passey melanosomes, indicates that sulfur (0.4-1.4%) in these natural melanins may be derived artificially from the reaction of melanins with cysteine or cystine in the course of HCI hydrolysis.

Introduction

In melanocytes, a specific enzyme tyrosinase (monophenol, L-dopa : oxygen oxidoreductase, EC 1.14.18.1) converts tyrosine to 3,4-dihydroxyphenylalanine (dopa) and then to dopaquinone, which is cyclized and further oxidized to give rise to eumelanins [1]. If dopaquinone encounters cysteine, pheomelanins are formed via cysteinyldopas [2]. However, glutathione (GSH), but not cysteine, is the major cellular SH compound. Thus, unless there exists a specific mechanism to take up cy-

Abbreviations: dopa, 3,4-dihydroxyphenylalanine; GSH, glutathione; PTCA, pyrrole-2,3,5-tricarboxylic acid; AHP, aminohydroxyphenylalanine. Correspondence: S. Ito, School of Hygiene, Fujita-Gakuen Health University, Toyoake, Aichi 470-1, Japan.

steine into melanosomes, it is likely that glutathionyldopas are formed first [3] and then converted to cysteinyldopas by the action of Tglutamyl transpeptidase and a peptidase [4]. Several groups have presented data indicating the significance of GSH and related enzymes in melanogenesis [5-7]. Furthermore, T-glutamyl transpeptidase was found to be inactivated in the course of melanogenesis [8]. Thus, there is a possibility that glutathionyldopas may become incorporated into melanins in vivo. Another interesting fact is that any natural eumelanin contains a trace to a modest amount of sulfur [1]. This has been ascribed to the copolymerization of dopa and cysteinyldopas, leading to the production of mixed-type melanins [1,9,10]. However, there is a possibility that the sulfur may be derived from GSH, but not from cysteine. It appears thus necessary to examine how cysteine

0304-4165/88/$03.50 © 1988 Elsevier Science Publishers B.V. (Biomedical Division)

2

KMn04 (COOH)

HOOC * HOOC

COOH H

Dopa-derived

units

PTCA

“ooc~~~cooH~~~~“~H 2 Cysteinyldopa-derived

units

(COOH)

h

2 AHP

HI

HSCH CHCOOH 21 NH2 Cysteine

I

Glu-Cys-Gly Glutathionyldopa-derived

Fig. 1. Methods arrows represent

units

for the estimation of dopa-derived, cysteinyldopa-derived, and glutathionyldopa-derived units in melanins. The the points of connections with adjacent monomer units. The GSH moiety can also be connected to the C-4 position.

and GSH are incorporated into synthetic melanins and to establish methods to estimate the cysteinyldopa-derived (cysteine-derived) units and the glutathionyldopa-derived (GSH-derived) units in synthetic and natural melanins (Fig. 1). In this melanins were prepared by tyrosinase study, oxidation of dopa in the presence of cysteine or GSH. These melanins were also subjected to HCl hydrolysis, the method commonly used to isolate melanins from natural sources. Methods were developed for quantitation of the cysteinyldopa-derived and glutathionyldopa-derived units in melanins. Materials and Methods Materials L-cysteine, GSH, and mushroom L-Dopa, tyrosinase (2000 units/mg) were purchased from Sigma Chemicals (St. Louis, MO). Other chemicals were of analytical grade from Wako Pure Chemicals (Osaka, Japan). The melanosomes were prepared from the ink sac of Sepia (cuttlefish) and from B16 and Harding-Passey mouse melanomas [ll]. The Sepia

melanosome preparation was a generous Prof. G. Prota (University of Naples).

gift from

Preparation and HCI hydrolysis of synthetic melanins A solution of r_.-dopa (1.00 mmol) and L-cysteine or GSH (O-l.00 mmol) in 0.05 M sodium phosphate buffer (pH 6.8, 100 ml) was oxidized by mushroom tyrosinase (10 mg) under oxygen current at 37 o C. After 4 h of incubation, the mixture was acidified to pH 3 with 1% acetic acid and left for 1 h at 4OC. The resulting precipitate was collected by centrifugation and washed twice with 1% acetic acid (40 ml). The melanin was dried in a desiccator over P,O, and NaOH, and weighed after equilibration with moisture in the air. When 0.50 mm01 of GSH was used, the melanin formed remained mostly in solution at pH 3; therefore, the pH 3 mixture was concentrated to approx. 20 ml in a rotary evaporator and the precipitated melanin was processed as above. 50 mg of melanin was heated in 6 M HCl (20 ml) under reflux for 24 h. The hydrolyzed melanin was collected by centrifugation, washed twice with 0.1 M HCl (10 ml), and dried in a desiccator. The hydrolysate and washings were combined and

evaporated to dryness in a rotary evaporator. The residue was dissolved in 10 ml of a pH 2.2 buffer for amino-acid analysis.

Preparation of natural melanins 50 mg of Sepia melanosomes was heated in 6 M HC1 (20 ml) under reflux for 24 h. The hydrolyzed melanin was collected by centrifugation, washed twice with 0.1 M HCI, twice with acetone, then once with 0.1 M HC1, and dried in a desiccator. The hydrolysate was subjected to amino-acid analysis to determine the protein content. The hydrolysis was also carried out in the presence of 100 mg of cysteine or cystine. Melanosome preparations from B16 and Harding-Passey mouse melanomas were similarly hydrolyzed. Yields of melanins and protein contents were: Sepia melanosomes, melanin 59% and protein 10%; B16 melanosomes, melanin 22% and protein 45%; Harding-Passey melanosomes, melanin 11% and protein 60%.

HPLC analysis of melanins after chemical degradation Permanganate oxidation and HI hydrolysis of melanins were performed as previously reported [12] except that the quantity of melanins was

increased to 1 mg. Thus, pyrrole-2,3,5-tricarboxylic acid (PTCA) in the oxidation products was analyzed by high-performance liquid chromatography (HPLC) with ultraviolet detection. Aminohydroxyphenylalanine (AHP) in the HI hydrolysate was analyzed by HPLC with electrochemical detection. Cysteine in the HI hydrolysate was analyzed by our newly developed HPLC method [13] after reduction of cystine to cysteine as follows. 10 ~tl of the HI hydrolysate (dissolved in 0.4 M HC104) was mixed with 10 /xl of 5 mM 2mercaptoethanol and 10/~1 of 0.4 M Na2CO3, and the mixture was left at room temperature. After 30 min, the mixture was diluted with 70/~1 of 0.6 M HC104 and subjected to the cysteine determination. Standard solution of 100 n m o l / m l cysteine and 50 n m o l / m l cystine gave almost identical values. Results

Preparation and HCI hydrolysis of synthetic and natural melanins Table I shows the yields of the melanins prepared and the results of HC1 hydrolysis. Insoluble, dark brown to black melanins were obtained by tyrosinase oxidation of dopa (1.00 mmol)

TABLE I P R E P A R A T I O N A N D HC1 H Y D R O L Y S I S O F S Y N T H E T I C M E L A N I N S Dopa (1.00 mmol) was oxidized by tyrosinase (10 mg) in the presence of cysteine (Cys) or GSH, the a m o u n t of which is shown in parentheses. No.

1 2 3 4 5 6 7 8 9 10 a b c d

Addition (mmol)

Yield (mg)

none Cys (0.05) Cys (0.10) Cys (0.25) Cys (0.50) Cys (1.00) d G S H (0.05) G S H (0.10) G S H (0.25) G S H (0.50)

190 174 193 184 178 234 201 237 242 205

HC1 hydrolysis a w / w (%)

/Lm o l / m g melanin

melanin

amino acid b

Gly

GIu

Cys ¢

76 76 75 69 60 55 78 70 66 53

1.8 4.2 4.2 4.4 5.1 5.2 7.5 9.4 16 23

0.01 0.03 0.03 0.02 0.03 0.03 0.26 0.42 0.87 1.28

0.01 0.03 0.03 0.02 0.03 0.03 0.14 0.24 0.52 0.70

0.00 0.01 0.02 0.03 0.04 0.03 0.05 0.07 0.14 0.24

Heated in 6 M HC1 under reflux for 24 h. Sum of the a m o u n t s of amino acids (minus water) determined by amino-acid analysis of the hydrolysate. Analzyed as cystine and the value doubled. Prepared by using 20 m g of tyrosinase.

in the presence of cysteine (0.05-1.00 mmol) or G S H (0.05-0.25 mmol). Dopa + GSH(0.50)melanin was soluble in 0.1 M HC1 and a p H 6.8 buffer. Dopa + GSH(1.00)-melanin was easily soluble in water at any pH, and thus could not be isolated. As the reaction of dopaquinone with sulfhydryl compounds is a very rapid one [14], it is very likely that these melanins were formed via cysteinyldopas [2] or glutathionyldopas [3]. By acid hydrolysis in 6 M HC1, dopa +

cysteine-melanins were decomposed to considerable extents to give brown-colored hydrolysates, the color of which increased in response to the increase in the cysteine content. The percentage recovery of melanin also decreased as the cysteine content increased. Our previous study has shown that approx. 80% ( w / w ) of melanin is recovered after HC1 hydrolysis of dopa-melanins with loss of CO 2 [15]. Dopa + GSH-melanins were hydrolyzed to give the amino-acid constituents of GSH, i.e.,

T A B L E II E L E M E N T A L C O M P O S I T I O N S O F S Y N T H E T I C A N D N A T U R A L M E L A N I N S A N D C O N T E N T S O F A H P A N D CYST E I N E A F T E R HI H Y D R O L Y S I S Melanins No. 1-10 correspond to those in Table I. Melanins No. 11-20 were prepared by HCI hydrolysis of melanins No. 1-10. Cys, cysteine; Cys2, cystine. No.

Melanin

Elemental composition (%) a

N / S molar ratio

Content ( b t m o l / m g melanin)

C

H

N

S

found

calcd,

sulfur b

AHP c

Cysc

1 2 3 4 5 6 7 8 9 10

Synthetic melanin Dopa-melanin Dopa + Cys(0.05)-melanin Dopa + Cys(0.l 0)-melanin Dopa + Cys(0.25)-melanin Dopa + Cys(0.50)-melanin Dopa + Cys(1.00)-melanin Dopa + GSH(0.05)-melanin Dopa + GSH(0.10)-melanin Dopa + GSH(0.25)-melanin Dopa + GSH(0.50)-melanin

53.50 53.10 49.87 48.69 51.98 46.17 51.63 48.05 46.98 46.53

4.11 3.80 4.17 4.24 4.49 4.41 4.02 4.37 4.48 4.35

7.06 7.81 7.60 7.92 9.03 8.78 7.94 7.83 9.03 10.09

0.06 1.05 1.55 3.71 7.05 7.96 0.95 1.28 2.80 4.51

17.0 11.2 4.89 2.93 2.52 19.1 14.0 7.38 5.12

21 11 5 3 2 23 13 7 5

0.02 0.33 0.48 1.16 2.20 2.48 0.30 0.40 0.87 1.41

0.000 0.04 0.10 0.25 0.51 0.60 0.000 0.001 0.001 0.001

0.02 0.03 0.02 0.01 0.03 0.02 0.17 0.25 0.48 0.84

11 12 13 14 15 16 17 18 19 20

Synthetic melanin after HCI hydrolysis d Dopa-melanin-HCl 48.82 Dopa + Cys(0.05)-melanin-HCl 49.09 Dopa + Cys(0.10)-melanin-HCl 48.06 Dopa + Cys(0.25)-melanin-HCl 47.74 Dopa + Cys(0.50)-melanin-HCl 44.52 Dopa + Cys(1.00)-melanin-HCl 43.77 Dopa + GSH(0.05)-melanin-HC1 48.02 Dopa + GSH(0.10)-melanin-HCl 48.29 Dopa + GSH(0.25)-melanin-HCl 47.94 Dopa + GSH(0.50)-melanin-HCl 44.13

4.04 4.11 3.98 3.91 4.10 4.00 3.96 3.91 3.85 4.00

6.87 7.23 7.12 7.70 7.96 8.08 7.05 7.27 7.60 7.42

0.15 0.79 1.98 4.03 6.26 7.32 0.76 1.35 2.96 4.73

20.9 8.23 4.37 2.91 2.53 21.5 12.3 5.88 3.59

21 11 5 3 2 21 11 5 3

0.05 0.25 0.62 1.26 1.95 2.28 0.25 0.42 0.92 1.48

0.000 0.03 0.07 0.16 0.20 0.25 0.001 0.001 0.001 0.001

0.02 0.02 0.02 0.02 0.02 0.02 0.08 0.12 0.33 0.45

21 22 23 24 25

Natural melanin prepared by HC1 hydrolysis d from Sepia melanosomes 51.09 3.65 from Sepia melanosomes, + Cys e 50.73 3.74 from Sepia melanosomes, + Cys 2 e 51.12 3.56 from B16 melanosomes 50.68 4.56 from Harding-Passey melanosomes 54.15 4.66

7.60 7.92 7.70 7.02 6.37

0.43 1.58 0.81 0.67 1.38

40.4 11.5 21.7 24.0 10.6

0.13 0.49 0.25 0.21 0.43

0.000 0.000 0.000 0.000 0.008

0.01 0.02 0.02 0.01 0.02

a b c d e

Analyzed after equilibration with moisture present in the air and corrected for ash content. Calculated from the elemental composition. Average of two separate determinations. Heated in 6 M HC1 under reflux for 24 h. Prepared in the presence of cysteine or cystine.

glycine, glutamic acid, and cysteine, in an average molar ratio of approx. 1.0 : 0.6 : 0.2 (Table I). The reason for the partial loss of glutamic acid as compared to glycine is not clear at present. Amino acids in the hydrolysates of dopa + cysteine-melanins originated from the tyrosinase used for the oxidation, whereas those in dopa + GSH-melanins came also from GSH connected to the melanins. The low recovery of cysteine in dopa + GSH-melanins indicates that GSH is bound to the melanins via the cysteine residue. This was proved by HI hydrolysis of dopa-GSHmelanins, giving cysteine in reasonably high yields (see below). The elemental compositions of synthetic melanins and those obtained by HC1 hydrolysis are shown in Table II. The N / S molar ratios found in synthetic melanins correspond fairly well to the ratios calculated from the molar ratios of dopa to cysteine or GSH. This result indicates that cysteine and GSH are incorporated into synthetic melanins with the retention of their original ratio to dopa. The N / S ratios in dopa + GSHmelanins relative to the calculated ones increased by HC1 hydrolysis; this may be due to the partial loss of sulfur in the form of cysteine. Natural melanins, prepared by HC1 hydrolysis of melanosomes, contained 0.43 to 1.38% of sulfur

KMnO4 Oxidation

120

1.0

(Table II). The sulfur content increased significantly when the hydrolysis was carried out in the presence of cysteine or cystine.

HPLC analysis of synthetic and natural melanins after chemical degradation In Fig. 2, PTCA content after permanganate oxidation and AHP and cysteine contents after HI hydrolysis are plotted against the sulfur content. The PTCA content decreased gradually as the sulfur content increased. The rate of decrease was lower in dopa + GSH-melanins than in dopa + cysteine-melanins. This fact can be explained as follows. It is likely that the glutathionyldopa-derived units in dopa + GSH-melanins consist mainly of the 7-S-glutathionyl-5,6-dihydroxyindole units (Fig. 1) [16]. These units, in addition to the dopa-derived 5,6-dihydroxyindole units, should give rise to PTCA on permanganate oxidation. The A H P content in dopa + cysteine-melanins was proportional to the sulfur content. The AHP content in dopa + cysteine-melanins after HC1 hydrolysis was also proportional to the sulfur content, although the yield decreased to approx, a half. On the other hand, dopa + GSH-melanins did not give appreciable amounts of AHP. The cysteine content in dopa + GSH-melanins was fairly proportional to the sulfur content. The

100

/

0.8 E~. 0,6

8O DOPO+G

D°D°+7

N 6o

3

ES~

=t60

el

(-) 0.4 I.o. 0,2

I00 HIHydrolysis

HIHydrolysis/

DODo+GSH-HCI ~ 40 D O P G % ~

20

I

I

I

I

2

4

6

8

~HCI

20

I

0

I

2

I

~

4

I

6

I

8

0

II~0 I

I

2

llC I 4

o--lk>-e I

I

6

8

S content (°/o)

Fig. 2. Relationships between the sulfur content and the contents of PTCA after permanganate oxidation and of A H P and cysteine after HI hydrolysis, e , dopa + cysteine-melanin; O, dopa + cysteine-melanin obtained by HCI hydrolysis; II, dopa + GSH-melanin; 1:3, dopa + GSH-melanin obtained by HCI hydrolysis.

hydrolyzed melanins also gave cysteine in approx. half the yields. On the other hand, d o p a + cysteine-melanins gave only trace amounts of cysteine. The same data as shown in Fig. 2 are also presented in Table II in which data are calculated on the molar basis. The yields of AHP from dopa + cysteine-melanins and those after HCI hydrolysis were rather constant when calculated on the basis of the sulfur contents; the yields were 20-24% (mol/mol) for dopa + cysteine-melanins and 11-13% (mol/mol) for the hydrolyzed melanins. On the other hand, the yields of cysteine from dopa + GSH-melanins and those after HC1 hydrolysis were also fairly constant; the yields were approx. 60% (mol/mol) dopa + GSHmelanins and approx. 30% (mol/mol) for the hydrolyzed melanins. Trace amounts of cysteine came also from tyrosinase used for the oxidation. These results indicate that AHP and cysteine after HI hydrolysis are indicators of the cysteinyldopa-derived units and the glutathionyldopa-derived units, respectively, although AHP is highly specific whereas cysteine is less specific. Natural melanins, obtained by HC1 hydrolysis of Sepia, B16, and Harding-Passey melanosomes, gave only trace amounts of both AHP and cysteine (Table II). These results suggest that not only Sepia melanin but also B16 and HardingPassey melanins are almost pure eumelanins in spite of the sulfur contents of 0.67 and 1.38%. Sepia melanins obtained by HC1 hydrolysis in the presence of cysteine or cystine did not give higher yields of cysteine on HI hydrolysis, although the sulfur contents increased significantly by 0.38 and 1.15%. Discussion

This study has shown that cysteine and GSH are incorporated into melanin polymers as long as they are present in the oxidation medium. Cysteine is integrated into the benzothazine units (pheomelanins), while GSH is connected to the dihydroindole units (eumelanins) with the retention of GSH moiety (Fig. 1). Consistent with these results is the report that oxidation of glutathionyldopa gives a red, dopachrome-type chromophore which develops into a black pigment [16]. The

cysteinyldopa-derived units and the glutathionyldopa-derived units can be estimated as AHP and cysteine, respectively, after HI hydrolysis. It is also shown that cysteine and cystine present in the hydrolysis medium is incorporated into Sepia melanin. The incorporation of cysteine may take place through the nucleophilic addition of cysteine to the o-quinone form of the dihydroxyindole units, while that of cystine may take place through the electrophilic substitution of cysteine sulfeny| cation [17] on the dihydroxyindole units. Three possibilities appear to emerge as the origin of sulfur in natural eumelanins: (1) participation of cysteine in melanogenesis leading to mixed-type melanins [1,9,10]; (2) participation of GSH in melanogenesis leading to eumelanin-like pigments; and (3) incorporation of cysteine or cystine into pre-formed melanins during HC1 hydrolysis to remove contaminating proteins. The origin of sulfur in natural eumelanins examined in this study can hardly be ascribed to the participation of cysteine or GSH in melanogenesis. Most of the sulfur appears to be derived artificially from the reaction of melanins with cysteine or cystine present in proteins in the course of HCI hydrolysis. It should be noted that, although B16 and Harding-Passey melanosomes revealed quite distinctive colors and ultrastructures [11,18], the melanins formed in these melanosomes are almost pure eumelanins. Cysteine and GSH are quantitatively incorporated into melanins formed, while B16 and Harding-Passey melanosomes produce almost pure eumelanins. Thus, an interesting problem has emerged: how these melanosomes prevent the incorporation of cysteine and GSH into melanins produced therein. Two explanations appear to be possible: (1) these melanosomes possess a specific mechanism to exclude these SH compounds from melanosomes; and (2) cysteinyldopas and glutathionyldopas are actually formed, but are excreted from melanosomes before incorporated into melanins. Whichever mechanism is operating, it may be relevant to the switch mechanism leading to eumelanogenesis and pheomelanogenesis: both eumelanin and pheomelanin are produced within the same follicular melanocytes to form the agouti band of wild-type mice [19]. Oxidation of dopa takes place not only in

m e l a n o c y t e s , b u t also o u t s i d e of m e l a n o c y t e s w h e r e t y r o s i n a s e is a b s e n t [20,21]. A s G S H is the major cellular SH c o m p o u n d , the oxidation should r e s u l t m o s t l y i n the f o r m a t i o n of g l u t a t h i o n y l d o p a s [3,22]. T w o m e t a b o l i c p a t h w a y s are p o s s i b l e for g l u t a t h i o n y l d o p a s , o n e l e a d i n g to the f o r m a t i o n of c y s t e i n y l d o p a s b y e n z y m i c h y d r o l y s i s a n d o t h e r i n v o l v i n g the o x i d a t i o n of g l u t a t h i o n y l d o p a s [16]. A l t h o u g h w h i c h p a t h w a y p r e d o m i n a t e s rem a i n s to b e clarified, this s t u d y has s h o w n that the latter p a t h w a y gives ' s o l u b l e ' m e l a n i n w h i c h c a n h a r d l y b e detected.

References 1 Prota, G. (1980) J. Invest. Dermatol. 75, 122-127 2 Ito, S. and Prota, G. (1977) Experientia 33, 1118-1119 3 Ito, S., Palumbo, A. and Prota, G. (1985) Experientia 41, 960-961 4 Agrup, G., Falck, B., Kennedy, B.-M., Rorsman, H., Rosengren, A.-M. and Rosengren, E. (1975) Acta Dermatovenereol. (Stockholm) 55, 1-3 5 Halprin, K.M. and Ohkawara, A. (1967) in Advances in Biology of Skin, Vol. 8, The Pigmentary System (Montagna, W. and Hu, F., eds.), pp. 241-251, Pergamon Press, New York 6 Benedetto, J.-P., Ortonne, J.-P., Voulet, C., Khatchadourian, C., Prota, G. and Thivolet, J. (1982) J. Invest. Dermatol. 79, 422-424 7 Hu, F. (1982) J. Invest. Dermatol. 79, 57-61

8 Mojamdar, M., Ichihashi, M. and Mishima, Y. (1985) in Pigment Cell 1985 (Bagnara, J., Klaus, S.N., Paul, E. and Schartl, M., eds.), pp. 713-720, University of Tokyo Press, Tokyo 9 Ito, S., Novellino, E., Chioccara, F., Misuraca, G. and Prota, G. (1980) Experientia 36, 822-823 10 Novellino, E., Ortonne, J.-P., Voulet, C., Chioccara, F., Misuraca, G. and Prota, G. (1981) FEBS Lett. 125, 101-103 11 Jimbow, K., Jimbow, M. and Chiba, M. (1981) J. Invest. Dermatol. 77, 213-218 12 Ito, S. and Fujita, K. (1985) Anal. Biochem. 144, 527-536 13 Imai, Y., Ito, S. and Fujita, K. (1987) J. Chromatogr. 420, 404-410 14 Tse, D.C.S., McCreery, R.L. and Adams, R.N. (1976) J. Med. Chem. 19, 37-41 15 Ito, S. (1986) Biochim. Biophys. Acta 883, 155-161 16 Carstam, R., Edner, C., Hansson, C., l_indblach, C., Rorsman, H. and Rosengren, E. (1986) Acta Dermatovenereol. (Stockholm) 66, Suppl. 126 17 Ito, S., Inoue, S., Yamamoto, Y. and Fujita, K. (1981) J. Med. Chem. 24, 673-677 18 Jimbow, K., Miyake, Y., Homma, K., Yasuda, K., Izumi, Y., Tsutsumi, A. and Ito, S. (1984) Cancer Res. 44, 1128-1134 19 Silvers, W. (1979) The Coat Color of Mice, pp. 6-28, Springer-Verlag, New York 20 Carstam, R., Hansson, C., Krook, G., Rorsman, H., Rosengren, E. and Wirestrand, L.E. (1985) Acta Dermatovenereol. (Stockholm) 65, 435-437 21 Ito, S. and Fujita, K. (1984) Biochem. Pharmacol. 33, 2193-2197 22 Fehling, C., Hansson, C., Rorsman, H. and Rosengren, E. (1981) Acta Dermatovenereol. (Stockholm) 61,339-342