Genetic effects on fine structure and development of pigment granules in mouse hair bulb melanocytes

Genetic effects on fine structure and development of pigment granules in mouse hair bulb melanocytes

DEVELOPMENTAL BIOLOGY Genetic 17, 366-381 (1968) Effects on Fine Structure Pigment Granules II. The c in Mouse and p loci, and Developmen...

5MB Sizes 0 Downloads 3 Views

DEVELOPMENTAL

BIOLOGY

Genetic

17,

366-381

(1968)

Effects on Fine Structure

Pigment

Granules

II. The

c

in Mouse

and p loci,

and Development

of

Hair Bulb Melanocytes

and ddpp Interaction’

ELIZABETH RITTENHOUSE~ Department

of

Zoology, University of Michigan, Accepted

September

Ann Arbor,

Michigan

48104

22, 1967

INTRODUCTION

A pale coat color in the house mouse may result from reduction of the quantity of pigment in the hair shaft or from changes in its distribution. Maltese dilution (d) is expressed chiefly in a clumping of pigment granules that appears to be related to the morphology of the melanocyte rather than to the properties of the individual pigment granules ( Markert and Silvers, 1956; Russell, 1949; Rittenhouse, 1968). Pink-eyed dilution (p), on the other hand, markedly decreases the size of individual granules and the total amount of hair pigment, besides producing a clumping of pigment granules that might result from the properties of the granules themselves (Russell, 1949). Pinkeyed dilution also interacts with Maltese dilution to produce in ddpp mice a very high level of tyrosinase activity combined with a very low level of pigmentation (Foster, 1959). Electron microscopy has shown that pink-eyed dilution disrupts the orderly pattern of the melanin-producing granule framework seen in both brown and black melanocytes of mouse retina (Moyer, 1963, 1966). In mouse hair bulb melanocytes, on the other hand, the brown (b) allele thoroughly disrupts the pattern of framework in the absence ’ Part of a thesis submitted in partial fulfillment of the requirements for the degree of Doctor of Philosophy in the University of Michigan. This work was supported in part by USPHS Genetics traineeship 2 G-71-C3 (subsequently GM-7107), USPHS research grant CA-04305 (subsequently NIH grant HD01254), and NIH training grant 2G-989/62-3. *Present address: Department of Genetics, Albert Einstein College of Medicine, New York, New York 10461. 366

EFFECTS

OF

C AND

j’l LOCI

Opi

MELANIN

CRANULES

367

of any dilution factor ( Rittenhouse, 1968). Albino (CC) melanocytrs lack tyrosinase activity (Foster, 1959), but retinal melanocytes of albino mouse embryos contain unpigmented granule framework of apparently normal structure (Moyer, 1963). This paper will deal with the effects of pink-eyed dilution and of albinism on the fine structure of melanin granules in hair bulb melanocytes of black and brown house mice, and also with the interaction of the pink-eyed and Maltese dilution factors. hlATERIALS

AND

METHODS

Dorsal skin from house mice was fixed in Veronal-acetate buffered osmium tetroxide solution, embedded in Epon according to the procedure of Luft ( 1961), and sectioned on a Porter-Blum microtome. Sections were stained in an aqueous solution of many1 acetate and examined in an RCA EMU 3E microscope. All mice came from stocks maintained at the Mammalian Genetics Center of the University of Michigan. The four pp genotypes studied were pink-eyed black ( BBCCDDpp) pink-eyed brown (bbCCDDpp), pink-eyed and Maltese black (BBCCddpp) and pink-eyed and Mal tese brown ( bbCC&!pp). The stocks for all four pp genotypes were originally derived from crosses of G57BL/6J-a’ with STOLI/Lw. In the case of one 5-day-old bbCCDDpp mouse, there had been a further cross with YBR/HeWiHa. Some additional observations were made on C57BL/6J (BBCCDDPP) skin. All pigmented mice were nonagouti (au). The albino mice were from the non-agouti strains ST/J ( acrhbccDDPP) and C57BL/6J-c ( UUBBCCDDPP) and the agouti strain Balb/c ( AAl&ccDDPP). For each pigmented genotype, skin was taken from two or more’ 5-day-old mice having different parents. In addition, skin from three S-day bhCCDDpp mice (two of them littermates and the third from different parents), one S-day-old BBCCDDpp mouse and one 7&yold RRCCddpp mouse was examined. For each of the albino genotvpes, skin was taken from two 5-dav-old mice. RESULTS

The Alehocyte Apart from their pigment granule populations, melanocytes of the four pp genotypes differ from the corresponding PP melanocvtes

368

ELIZABETH

RITTENHOUSE

FIG. 1. BBCCDDpp, five days. Granules are lightly melanized or unmelanized and may show periodicity ( gI g2, g3). Granules g4 and gS are somewhat de-

EFFECTS

OF

C

AND

jl

LOCI

ON

MELANIN

GRANULES

369

(Rittenhouse, 1968) in two respects only. Nuclei of pp melanoctyes are often deeply indented into irregular shapes. Such nuclei seem to be more common in 5-day-old pp melanocytes than in 7- or S-day-old pp melanocytes. m melanocytes also contain membrane-enclosed bodies which bear some resemblance both to multivesicular bodies and to immature pigment granules. These bodies are variable in composition and may include a sparse scattering of strands or small vesicles resembling Golgi vesicles, and, occasionally, small areas having the electron density of melanin (Fig. 7). cc melanocytes contain bodies of similar appearance, but lacking the melanin-like areas. The Granules Pink-eyed dilution. The striking characteristic of both pink-eyed black ( RBCCDDpp) and pink-eyed brown ( bbCCDDpp) melanocytes is the high proportion of lightly melanized and unmelanized granules. Even in the dendrites (Fig. 2), where the granules are more heavily melanized than in the cell body, separate strands of melanization within the granules are easily discernible. There is no size diff ercnce between granules in different stages of melanization. Pink-eyed black granules (Figs. 1 and 2) are similar in size and structure to lightly melanized and unmelanized granules of intense black (BBCCDDPP) melanocytes, but some minor differences were seen. Pink-eyed black granules are slightly smaller, with the decrease in width more pronounced than the decrease in length. Longitudinal sections range from long ovals to slivers, with strands only a little less orderly than the strands of intense black granules. Since the deviation from a pattern of straight parallel lines is confined chieflv to thcs unmelanized granules, the explanation of the variation observed mav be simplv that melanin helps to prevent distortion of the granules during fixation and embedding. Cross sections, however. c’o seem to show fewer instances of simple spirals or concentric circles. formed, possibh bv processes of fixation and embedding. The arrow indicatrs a membrane-enclosed body of uncertain iclentitv. X32,600. FIR 2. RBCCDDpp, fix-e days. Granules in the dendrite of a melanoqtc,. The arrow indicates a melanizing granule with headlike periodieity of 200 A. x 41,000. Frc. 3.

RBCCDDPP, five days. The arrow pointy to apparent pairing of the dense lines along a longitudinal strand. TWO units of a coarse. headlike periodicit\, lie to the right of the arrow. ~80,000.

370

ELIZABETH

BITTENHOUSE

bbCCDDpp, five days. Granule g, is somewhat altered toward the FIG. 4. “black” pattern hut retains some of the disorganization characteristic of brown granules. The unmelanized granule g? seems to contain a membrane lying in the plane of the section. ~48,000.

EFFECTS

OF

C AND

rl

LOCI

ON

MELANIN

GRANULES

‘371

The inte,nal structure of these granules, nevertheless, seems to consist of one or more membranes running the length of the granule in a pattern very much like the pattern seen in intense black granules. The granules in the S-day-old mouse did not differ from the granules in the S-day-old mouse. In S-day-old pink-eyed brown melanocytes (Fig. 9) the majority of granules resemble lightly melanized intense brown (bhCCDDPP) granules. They are usually round, with the greatest diameter not exceeding 0.55 ,A, The framework is a disorderly web of strands. The few unmelanized granules occur in or near the Golgi region. In 5-clay-old pink-eyed brown melanocytes (Fig. 4) however, many granules are oval or elongate, with a greatest diameter comparable to the diameter of heavily melanized intense brown granules (up to O.‘i 1-t). The volume of these elongate granules. however, would 1.c~ substantiallv less than the volume of the more nearly spherical intense brown granules. The framework of many oval and elongate granules tends toward a pattern of longitudinal strands and membranes, but with more forking and crossing than is usual in black granules. Periodic striations of the type seen in black granules sometimes arc evident. Figure 4 shows an unmelanized hods similar to the bodies interpreted in black genotypes as granules sectioned so that a menbrane lies in the plane of section. Unmelanized granules are common and may be seen in any part of the cell. CCDDpp melanocytes tend to contain more granules than CCDDPP, and a frw are as denselv crowded as CCtlrlPP (Maltese) cells. There FL. 5. hbCCt&p, eight days. The 1~~1~~ indicatrcl 1,). the arrow has ( 19,000.

37”

ELIZABETH

RITTENHOUSE

FIG. 9. bbCCDDpp, eight days. g, appears to be a granule framework forming in the midst of a grouping of Golgi membranes and ribosomes. The appear-

EFFECTS

OF

C AND

p

LOCI

OS

hIELAK;IS

CRANIXE3

373

of -distinct -chnqs of ‘granules within the is no indication of knmatkm melanocytes. Clumps of granules in pp hair cells often seem to involve not only crowding, but also some degree of fusion of granules (Figs. 10 and 11) . The small size and light melanization characteristic of pp granules may be significant for the formation of this type of clumping. Pink-eyed and hlaltes, 0 dilution. These melanocytes are usualh densely crowded with granules, in the manner of CCddPP mela&cytes, but the granules are more heavily melanized than CCDDpp granules. In some BBCCddpp melanocytes the heavilv melanized granules may be markedly larger than the unmelanized, although somewhat smaller than mature BBCCDDPP granules (Fig. 8). In others the melanized granules are no larger than the unmelanized (Fig. 7). bhCCddpp granules are oval or elongate. The more heavily melanized granules may be as long as 0.8 II, but the longest ones are usually narrow. Unmelanized granules resemble bbCC DDpp granules in size. Granules showing rather orderly strands with obvious periodicity are more common than in bbCCDDpp cells. Nevertheless, in these cells, as in hbCCDDpp cells, even the most orderly granules show some of the irregularity of structure seen in intense brown granules. Albinism. There seems to be no reliable criterion for distinguishing melanocytes of the three albino genotypes (aaBBccDDPP. ante of granule g?, especially, suggests the participation of ribosomes in granule formation. X 25,000. FIG. 10. hbCCDDpp, five days. A clump of granules in a hair cell. The edges of the granules are indistinct, and the region between the dense centers of granules seems to he rather diffusely melanized. x 15,000. FIG. 11. BBCCDDpp, eight days. A clump of granules in a hair cell. The edges of the granules are indistinct, and the appearance of the chrmp suggests that some fusion of granules may have occurred. ~40,000. showing a gridlike FIG. 12. BBCCddpp, five days. A granule memhrane pattern. The periodicity at right angles to the greatest diameter of the granule is approximately 105 A. X 120,000. Fro. 13. bbCCDDpp, five days. Smooth-walled vesicles appear to he passing from the side of an irregularly granular double membrane to a region of granule framework formation. A clgmp of rihosomes lies to the right of the forming granule, and other ribosomes may be present in the region of fnrming framework.

x 81,200.

374

ELIZABETH

BITTENHOUSE

EFFECTS

OF

C AND

p

LOCI

ON

MELANIK

CRANULES

375

AAbbccDDPP, aubbccDDPP) from one another. Melanocytes

of all three genotypes contain a mixture of unmelanized granules of the “black” type (ovals with longitudinal strands, or small circles with spiral or circular patterns), unmelanized granules of the “brown” type (large circles with a complex, disorderly internal structure ) , multivesicular bodies, and the bodies described above as having an appearance intermediate between multivesicular bodies and melanin granules (Fig. 14). In all three albino genotypes, occasional “black” granules show obvious periodicity in their longitudinal strands, but periodicity was not seen as frequently as in otherwise similar unmelanized granules of pigmented black genotypes. The most orderly “black” granules are often unusually large for unmelanized granules, sometimes more than a micron long (Fig. 15). Granule formation. It has been reported (Rittenhouse, 1968) that ir. brown and black melanocytes without dilution factors or with only Maltese dilution, the unmelanized granules occur at the periphery of the Golgi region where Golgi vesicles, free ribosomes, and occasional rough-surfaced double membranes occur together. The same distribution is seen in BBCCddpp and bbCCddpp melanocytes. In BBCCDDpp and bbCCDDpp melanocytes, unmelanized granules may be seen anywhere in the cell but are most common at the periphery of the Golgi region. Small vesicles of the Golgi type were occasionally seen within what appeared to be unmelanized pigment granules (Figs. 5 and 6). In Fig. 13 smooth-walled vesicles seem to be passing from one side of an irregularly granular double membrane into a region of granule formation. In Fig. 9, however, there is nothing to indicate that Golgi vesicles are incorporated into the granule g,, which appears to be forming in the midst of a configuration of Golgi membranes and ribosomes. The appearance of the other granules suggests a direct participation of ribosomes surrounding, and perhaps included within, the granule. FIG. 14. RBccDDPP, five days. Of the nleml)rane-enclosed bodies indicated black granules, some more closely ret )v arrows, some resemble unmelanized semble m~melanized brown granules, some appear to be multivesicular bodies, and others have an appearance intermediate in some degree. x 19,600. FIG. 15. ~~ccDDPP, five days. This unusuallv large (approximately 1 a) brown albino granule has a pattern of orderly longitudinal strands indistinguishable from the characteristic pattern of black granules. The periodicity along the longitudinal strands is approximately 105 A. x 90,000.

376

ELIZABETH

RI’ITENHOUSE

In both bb and BB genotypes periodicity along longitudinal strands or membranes of unmelanized granules measures close to 105 A (Figs. 12, 13, 15). It was not often possible to measure periodicity accurately after melanization had begun, but a few melanizing granules showed a beadlike periodicity of about 200 A. Figure 3 might be interpreted as showing the formation of units of the coarser periodicity from pairs of units of the finer periodicity. DISCUSSION

The principal effect of pink-eyed dilution in both bb and BB melanocytes is to decrease the amount of pigment deposited on the granule framework. Although BBpp granules are somewhat more variable in shape and structure than BBPP granules, the alterations in structure are minor compared to the reduction of pigmentation. The decrease in pigmentation is accompanied by a failure of the granules to grow beyond the usual size of lightly melanized granules in PP genotypes. In the hair cells the granules are often clumped, and apparently fused, together, but clumping was not seen in the melanocyte. In bb melanocytes pink-eyed dilution shifts the structure of the granules away from the disorderly ball characteristic of intense brown granules and toward the longitudinal membrane pattern characteristic of black granules. Since albinism (cc) has a similar effect, this change may be related to reduced melanization in a manner not specific to the p locus. Maltese dilution counteracts, though incompletely, the effect of pink-eyed dilution on granule melanization. If it is assumed that the effect of pink-eyed dilution is to decrease the rate of melanization, and that the effect of Maltese dilution is to cause prolonged retention of the granules within the melanocyte, the effect of the mutations together could be the simple sum of their separate effects. Foster’s studies of tyrosinase activity of whole skin (1959) suggest that the effect may not be quite so simple, since the activity of bbddpp skin is much higher than the tyrosinase activity of bbDDpp, bbDDPP, or bbddPP skin. Still, if it is assumed that measurable enzyme activity decreases with increasing melanization of the granule (Seiji and Fitzpatrick, 1961), and that the granules which have not yet entered the hair cells are the principal source of activity measured, then the ppdd melanocyte, with its increased number of granules of somewhat reduced melanization, might happen to strike a balance between these IJXJOfactors most favorable for measurable enzyme activity.

EFFECTS

OF

C AND

p

LOCI

ON

hlELA1L’IS

GK4NULES

37:

According to Russell ( 1949) granule clumps of both the flocculent (characteristic of pp) and granular (characteristic of dd) types are seen in ddpp hairs. The characteristic structure of p granules and the appearance of the flocculent clumps themselves suggest that floe culence is the result of a partial fusion of lightly melanized granules. It may be that the diEerence in the melanization of individual melanostriking in BBddpp cvtes of tldp~ hairs, a difference particularly melanocytes, is a significant factor in the appearance of both types of clumps in ddpp hairs. Although Slaltese dilution alone does not, in the hair bulb, shift the disorderly pattern of brown granules toward the “black” pattern ( Rittenhouse, 1968 ) , th e shift caused by pink-eyed dilution is increased bv the addition of Maltese dilution, and this increased shift seems to be as pronounced at 7 and 6 days as at 5 days. Moreover, the granules in bbddPP (strain DBA) dermal melanocytes of the ear show a similar shift (Rittenhouse, unpublished ). In a study of embryonic mouse retinal melanocytes and :3- and 5-day-old hair bulb melanocytes, hloyer (1966) reported that unmelanized granules of both intense black and intense brown genotypes had a framework appearing as bundles of parallel fibers in methacrylate embedded material and as longitudinal sheets of membrane in material embedded in epox!. resins. It appears, then, that the degree of resemhlanct between black and brown granule framework is somrvvhat \ariahlc, and mav be influenced by age, melanocyte location, and the presence of dilution factors. Failure of albino granules to melanize is accompanied by changes in granule structure such that there appears to be no consistent difference between the granule populations of l&cc and B&c melanocytes. Cells of both genotypes contain some granules of the “black” pattern and many round bodies which may resemble brown granules or multivesicular bodies, but often resemble both in some degree. Since the reduction or absence of melanization occurs together with a shift toward the “black” pattern in both bbccPP and bbCCpp melanocytes, it would be convenient to explain the shift toward the “black” pattern as a result of reduced melanization. One might say that if the disorganized framework of l>h granules is not stabilized almost immediately by at least a light melanization, it will tend to acquire the more orderly organization characteristic of black granules. The fact that S-day-old hbCCDDpp melanocytes have a higher percentage of granules showing at least some melanization and a lower percentage of

378

ELIZABETH

BITTENHOUSE

granules with some suggestion of the “black” pattern than 5day-old melanocytes of the same genotype would tend to support this hypothesis. On the other hand, one sometimes sees granules, such as the bbCCddpp granule shown in Fig. 13, that seem to be forming directly in the “black” pattern. Lightly melanized granules of the “black” pattern in both black and brown genotypes are usually elongate, and it may be that the pattern of the framework is determined by the size of the least diameter, or the ratio of the least diameter to the greatest diameter. The causes of the differences in size and shape remain to be explained. The fact that bbCCddpp granules are more elongate, more heavily melanized, and more distinctly shifted toward the “black” pattern than bbCCDDpp granules would point to granule shape or size as the most important factor if it were certain that the heavier melanization implied a more rapid initial melanization. The rather large number of unmelanized and very lightly melanized granules seen in and near the Golgi region of both BBCCddpp and bbCCddpp melanocytes suggests, however, that this may not be the case. Granules in the retinal epithelium of young albino mice (Moyer, 1963, 1966) and young albino rats (Dowling and Gibbons, 1962) have been described as normal in appearance except for the absence of melanin. Dowling and Gibbons also reported absence of granules in retinal epithelium of adult albino rats but did not describe the process of disintegration or loss. If unmelanized granules in the hair bulb melanocytes disintegrate, many of the bodies with a structure of internal strands might be granules in the process of disintegration rather than granules formed abnormally or formed in the “brown” pattern. The hypothesis that brown granules which do not melanize immediately may move toward a more orderly structure is not necessarily incompatible with the hypothesis that unmelanized granules eventually disintegrate. It is possible, for instance, that the small vesicles of the Golgi type sometimes seen within unmelanized granules of light-colored genotypes may contain substances contributing not to granule formation but to granule destruction. Chian and Wilgram ( 1967), on the other hand, attribute albinism to a tyrosinase normal in activity, as measured by DOPA oxidation in vitro, but unable to aggregate into structurally normal granules and therefore vulnerable to the action of an inhibitor normally present in melanocyte cytoplasm. Since the material used in their study was an

EFFECTS

OF

C AND

p

LOCI

ON

MELANIN

GRANULES

379

amelanotic strain of S 91 melanoma, the relationship of the amelanotic condition to the c locus is uncertain. In the occurrence of granules varying considerably in degree of abnormality and especially of granules abnormal chiefly in their lack of periodicity there is similarity to the situation seen in the cc hair bulb melanocyte. Since granules of apparently normal periodicity are occasionally seen both in the amelanotic melanoma cells and in the albino hair bulb melanocytes, one might expect occasional melanization also. No melanization was seen in albino hair bulb granules, however, and (Ihian and Wilgram do not refer to melanization in the “albino” S 91 granules. It was not t,o be expected that the lightly melanized genotypes would clarifv the question of granule formation. In view of the fact that many granules remain unmelanized, and especially in view of the possibility that they eventually disintegrate, the interpretation of any configuration as representing granule formation must be cluestionable. It was to be hoped, however, that the>- would contribute evidence about the nature of granule growth. Unfortunatelv the results from albinism and pink-eyed dilution conflict. In the two CCDDyp genotypes examined, in which all granules were unmelanized or lightly melanized. the size of the granules did not vary with degree of melanization or with position relative to the Golgi region. This would suggest that it is increased melanin deposition, without addition of further granule framework, that causes the increase in size of melanized granules. The fact that albino granules may be as long as 1 I,, considerably longer than unmelanized granules of any of the pigmented genotypes studied, suggests, on the other hand, that addition of new framework might occur. Since large albino granules were seen inside as well as outside the Golgi region, it is possible that they did not grow, but were formed directlv as large granules. Although periodicity along the strands of unmelanized granules is about 105 A, melanizing granules occasionally 410~. a periodicity ot about 200 A. Wellings and Siegel ( 1963) have reported a 200 A periodicitv in granules of melanoma. They did not refer to degree of melanization of the granules, but the pictures shown seem to be of melanizing granules. Since the units of the coarser periodicity are almost twice the size of the units of finer periodicit!-. it seems possible that there is no expansion of the framework involved, but that tn.0 of the smaller units contribute to the formation of one of the lqrr units. Such a doubling might be illustrated in Fig. 3.

380

ELIZABETH

RITTENHOUSE

SUMMARY

This study dealt with the effects of mutation at the p (pink-eyed dilution) and c (albinism) loci separately, and at the p and d ( Maltese dilution) loci together, on the fine structure of brown and black melanin granules in hair bulb melanocytes of the house mouse. Mutation at the p locus affects size, degree of melanization, and, in bb cells, shape of the granules. pp granules are lightly melanized and are smaller than PP granules, apparently because they fail to enlarge after the beginning of melanization. The c allele does not prevent the formation of granule framework, but does prevent melanization. Occasional cc granules are as large as heavily melanized granules of the corresponding CC genotype. The c and p alleles have two effects in common. Both produce in bb granules a shift toward the pattern of longitudinal strands and membranes characteristic of BB granules. Both also cause the occurrence in the melanocyte of round membrane-enclosed bodies of variable internal structure which may be granules in the process of disintegration. In cc melanocytes these effects are so marked that BBcc cells cannot be reliably distinguished from bbcc. The addition of Maltese dilution to pink-eyed dilution counteracts, though incompletely, the decrease of granule melanization, but intensifies the shift toward the “black” pattern. I am especially indebted to Dr. Morris Foster for his unflagging contribution of advice, criticism, and mice throughout the course of this study. I am grateful also to Dr. Norman Kemp for advice, assistance, and the use of laboratory facilities, and to Drs. Tahir Rizki, Erich Steiner, and Alfred Stockard for discussion and criticism. I should like to thank Dr. W. C. Bigelow, Mr. John Alley, Mr. Robert Weymouth, and Nancy Smith Istock for advice and assistance in the technique of electron microscopy, and to thank the photographic department of the Chester Beatty Institute for preparation of prints, REFERENCES its role in L. T. Y., and WILGRAM, G. F. (1967). Tyrosinase inhibition: suntanning and albinism. Science 155, 19S206. DOWLING, J. E., and GIBBONS, I. R. (1962). The fine structure of the pigment epithelium in the albino rat. 1. CeU. Biol. 14, 450474. FOSTER, M. (1959). Physiological studies of melanogenesis. In “Pigment Cell Biology” (M. Gordon, ea.), pp. 301-314. Academic Press, New York. LUFT, J. ( 1961). Improvements in epoxy resin embedding methods. J. Biophys. B&hem. Cytol. 9, 409414. MARKERT, C. L., and SILVERS, W. K. (1956). The effects of genotype and cell

CHIAN,

EFFECTS

OF C AND p LOCI ON MELANIN

GRAHJLES

381

environment on melanoblast differentiation in the house mouse. Genetics 41, 429450. MOYER, F. H. ( 1963). Genetic effects of melanosome fine structure and ontogeny in normal and malignant cells. Ann. N.Y. Acad. Sci. 100, 58P606. MOYER, F. H. (1966). Genetic variations in the fine structure and ontogeny of mouse melanin granules. Am. Zoo?.ogist 6, 43-66. RITTENHOUSE, E. (1968). Genetic effects on fine structure and development of pigment granules in mouse hair bulb melanocytes. I. The b and d loci. Deuelop. Biol. 17, 351-365. RUSSELL, E. S. ( 1949). A quantitative histological study of the pigment found in coat color mutants of the house mouse. IV. The nature of the effects of genie substitution in five major allelic series. Genetics 34, 146-166. SEIJI, ht., and FITZPATRICK, T. B. ( 1961). The reciprocal relationship between melanization and tyrosinase activity in melanosomes (melanin granules ) . J. Biochem. 49, 700-706. WELLI~TGS, S. R., and SIEGEL, B. V. (1963). Electron microscopic studies on the subcellular origin and ultrastructure of melanin. Ann. N.Y. Acad. Sci. 100, 497-539.