Expression and Transmission of Wild-Type Pigmentation in the Skin of Transgenic Orange-Colored Variants of Medaka (Oryzias latipes) Bearing the Gene for Mouse Tyrosinase

Transgenic fish carrying a reconstructed mouse tyrosinase gene, mg-Tyrs-J, were produced by microinjecting the gene into the oocyte nucleus of an orange-colored variant of medaka (Oryzias latipes). Of 64 oocytes microinjected and subsequently inseminated, 13 embryos developed normally beyond hatching and three of them exhibited brown skin pigmentation in the adult as was commonly observed in the wild type of this species. Light and electron microscopic examination disclosed a ubiquitous distribution of typical melanophores in the skin of these transgenic fish. Judging from their population density and distribution pattern, it was presumed that melanogenesis in these fish was elicited in amelanotic melanophores that resided in the skin of the orange-colored fish of this variant. Immunofluorescence with use of the anti-mouse tyrosinase antiserum lacking reactivity to medaka tyrosinase clearly disclosed that the gene introduced was expressed in the melanophores of transgenic fish. Crosses of female transgenic fish and males from an orange-colored variant yielded offspring exhibiting wild-type or orange-colored pigmentation in a ratio of 1:1, thus implying that mg-Tyrs-J integrated into the medaka genome behaves like a dominant gene. Little melanogenesis was observed in xanthophores, leucophores and iridophores in transgenic fish, suggesting possible specificity in recognition of teleostean cell types (i.e., melanophores) by the regulatory region of the mouse tyrosinase gene.     

Transgenic Medaka Fish Bearing the Mouse Tyrosinase Gene: Expression and Transmission of the Transgene Following Electroporation of the Orange-Colored Variant

Transgenic fish bearing the mouse tyrosinase gene (mg-Tyrs-J) were produced by transfection into fertilized eggs of the homozygous normal orange-colored variant of medaka fish, Oryzias latipes, by means of electroporation. Of 589 eggs transfected, 38 fish (6%) exhibited brownish wild-type skin pigmentation, which was discernible from control siblings. Light microscopy of the skin from the founders thus generated disclosed that 1) melanization occurred and was restricted to melanophores formed presumably from preexisting amelanotic melanophores, 2) there was a wide variation in the degree of melanization observed among melanophores, and 3) no melanin deposition was recognized in xanthophores or leucophores. Immunofluorescence using an antibody raised against mouse tyrosinase disclosed that melanophores at varying stages of maturation were reactive. Thus, it was shown that the transgene in medaka fish expressed its action in a cell type-specific manner. Crossing of transgenic founders with homozygous orange-colored variant fish yielded two groups of offspring expressing either the wild-type or the orange-colored skin pigmentation at an approximate ratio of 1:1. Crossing between founders exhibiting wild-type pigmentation yielded only offspring with melanized skin. Skin melanophores in these offspring formed vertical stripes, which are rare in this species. The hereditary basis of melanized skin was demonstrated in matings of Fl progenies, which resulted in similar degrees of melanization over whole skin melanophores. The sum of these findings implied that the transgene is expressed as a dominant character gene and is transmitted through germ cell lines according to the Mendelian law. PCR analysis combined with nested PCR technique strongly suggested that the transgene was integrated into the medaka genome, even though the copy number deduced from gel banding was largely diminished, possibly as a result of fragmentation or instability within the medaka genome.     

Effect of the barring gene on eye pigmentation in the fowl

Pigment cells of the iris, pecten, retinal pigment epithelium, and choroid of the wild-type jungle fowl (JF) and the barred Plymouth rock (BPR) breeds of adult chickens were studied at both light and electron microscopic levels. BPR choroidal tissues had 2.8 times fewer melanophores than the JF choroid, and BPR melanophores also contained 2.4 times fewer melanosomes, which tended to clump together in variously sized clusters. The melanosomes were often irregular in shape, smaller in diameter, and less mature (stage III) than those granules in the JF. The retinal pigment epithelium of both JF and BPR breeds contained a single epithelial layer of columnar cells. Rod-shaped melanosomes were present in the more apical regions of this cell type in both breeds. Both JF and BPR irides contained a multilayered posterior pigmented epithelium of columnar shaped cells that were densely filled with large spherical granules. Intercellular spaces with interdigitating cytoplasmic projections were present between pigment cells of both breeds. The pecten melanophores of both breeds were dendritic with melanosomes that were larger and fewer in numbers than those pigment cells of the iris and choroid. Intercellular spaces were present between cells in both breeds, with numerous villous-like pigment cell extensions. Choroid melanophores contained very little, if any, acid phosphatase activity. Approximately one-half of the retinal pigment epithelial cells observed contained small amounts of diffuse acid phosphatase activity in both breeds. The iris and pecten melanophores of both breeds contained profuse acid phosphatase activity scattered throughout their cytoplasms. Sparse tyrosinase activity was seen in iris and pecten pigment cells, whereas no tyrosine activity was observed in choroid melanophores or in retinal pigment epithelial cells in the two breeds, indicating that little new melanogenesis occurs in adult pigmented eye tissues. The results show that the barring gene reduces the number and melanin content of the choroidal melanophores in homozygous male BPR chickens as compared to the wild-type JF chickens. Whether this gene prevents the initial migration of embryonic neural crest cells (future melanophores) to the choroid or whether some of the choroidal melanophores prematurely degenerate in the embryo of young birds is yet to be determined. If the latter is the case, this choroid system may serve as a model for a genetic hypomelanotic disease such as vitiligo.     

Melanin synthesis activation dependent on inductive influences

Summary Gene activity in melanin-synthesising cells of albino periodic (a^p) mutants ofXenopus laevis is expressed phenotypically in the framework of the following cycle: a period of complete albinism succeeds the short peak of pigmentation, and melanosomes which have formed disappear. Skin and choroid coat melanophores as well as pigmented epithelium melanocytes are involved in this cycle.Parabiosis experiments allowed hormonal regulation of the melanin-synthesising gene activity to be excluded. Neural fold transplantations have shown that there is no inhibitory action on melanophore differentiation from the side of the a^p/a^p recipient.Melanin synthesis in pigmented epithelium of a^p mutants can be activated to level comparable with that of wild-type animals, if eye vesicles of a^p/a^p embryos have been brought into contact with endomesodermal derivatives of +/+ embryos at the early tail bud stage. Contact of eye vesicles of +/+ embryos with the endomesoderm of mutants prevents normal melanogenesis in pigmented epithelium of transplanted eyes. Eye transplantations made after the early tail bud stage have shown that gene expression in pigmented epithelium is independent of any external influences.Data obtained here demonstrate a selective induction of a separate cell type (melanocytes) and the stage-specificity of this process. In the a^p mutant the abnormal melanin synthesis is apparently predetermined by deficiency in the inducer of melanogenesis. Inhibition of melanogenesis by endomesoderm seems to be less probable. Data are discussed in the light of current ideas on the play of gene activity.     

oca2 regulation of chromatophore differentiation and number is cell type specific in zebrafish

We characterized a zebrafish mutant that displays defects in melanin synthesis and in the differentiation of melanophores and iridophores of the skin and retinal pigment epithelium. Positional cloning and candidate gene sequencing link this mutation to a 410‐kb region on chromosome 6, containing the oculocutaneous albinism 2 (oca2) gene. Quantification of oca2 mutant melanophores shows a reduction in the number of differentiated melanophores compared with wildtype siblings. Consistent with the analysis of mouse Oca2‐deficient melanocytes, zebrafish mutant melanophores have immature melanosomes which are partially rescued following treatment with vacuolar‐type ATPase inhibitor/cytoplasmic pH modifier, bafilomycin A1. Melanophore‐specific gene expression is detected at the correct time and in anticipated locations. While oca2 zebrafish display unpigmented gaps on the head region of mutants 3 days post‐fertilization, melanoblast quantification indicates that oca2 mutants have the correct number of melanoblasts, suggesting a differentiation defect explains the reduced melanophore number. Unlike melanophores, which are reduced in number in oca2 mutants, differentiated iridophores are present at significantly higher numbers. These data suggest distinct mechanisms for oca2 in establishing differentiated chromatophore number in developing zebrafish.     

Studies of pigment transfer between Xenopus laevis melanophores and fibroblasts in vitro and in vivo1

Frog melanophores rapidly change colour by dispersion or aggregation of melanosomes. A long‐term colour change exists where melanosomes are released from melanophores and transferred to surrounding skin cells. No in vitro model for pigment transfer exists for lower vertebrates. Frog melanophores of different morphology exist both in epidermis where keratinocytes are present and in dermis where fibroblasts dominate. We have examined whether release and transfer of melanosomes can be studied in a melanophore‐fibroblast co‐culture, as no frog keratinocyte cell line exists. Xenopus laevis melanophores are normally cultured in conditioned medium from fibroblasts and fibroblast‐derived factors may be important for melanophore morphology. Melanin was exocytosed as membrane‐enclosed melanosomes in a process that was upregulated by α‐melanocyte‐stimulating hormone (α‐MSH), and melanosomes where taken up by fibroblasts. Melanosome membrane‐proteins seemed to be of importance, as the cluster‐like uptake pattern of pigment granules was distinct from that of latex beads. In vivo results confirmed the ability of dermal fibroblasts to engulf melanosomes. Our results show that cultured frog melanophores can not only be used for studies of rapid colour change, but also as a model system for long‐term colour changes and for studies of factors that affect pigmentation.     

Actin and Tubulin Arrays in Cultured Xenopus Melanophores Responding to Melatonin

Melatonin induces pigment granule aggregation in amphibian melanophores. In the studies reported here, we have used fluorescence microscopic techniques to test the hypothesis that such melatonin-induced pigment movement is correlated with alterations in either the actin or tubulin cytoskeletal patterns of cultured Xenopus melanophores. In general, the cytoplasmic domains of the cultured melanophores were flat and thin except in the perinuclear region (especially when the pigment was aggregated). The microtubules and microfilaments were usually found in the same focal plane; however, on occasion, microfilaments were closer to the substratum. Microtubules were arranged in arrays radiating from what are presumed to be cytocenters. A small percentage of the melanophores were very large, had actin-rich circular perimeters and did not respond as rapidly to melatonin treatment as did the other melanophores. Melanophores with either aggregated or dispersed melanosomes had low intensity rhodamine-phalloidin staining of actin filaments compared to nonpigmented cells, whereas the FITC anti-tubulin intensities were comparable in magnitude to that seen in nonpigmented cells. When cells were fixed prior to complete melatonin-induced pigment granule aggregation there was no abrupt diminution in either the tubulin or actin staining at the boundary between pigment granule-rich and pigment granule-poor cytoplasmic domains. Nor could the actin and tubulin patterns in cells with partially aggregated melanosomes be reliably distinguished from those in melanophores in which the melanosomes were either completely dispersed or completely aggregated. These data argue against the hypothesis that melatonin causes consistent large-scale rearrangements of tubulin and actin polymers as it induces pigment aggregation in Xenopus melanophores.     

Melanosome Formation in Cultured Amelanotic Melanophores of Rana brevipoda by a Frog Tyrosinase cDNA Transfection

A cDNA encoding tyrosinase of Rana nigromaculata was introduced into cultured, tyrosinase-negative amelanotic melanophores of R. brevipoda by a calcium phosphate precipitation method. Within a few days following transfection, dark pigmentation became visible in a small number of cells. Light microscopic observation revealed that the morphology of these transformed cells was comparable to that of normal melanophores in culture, and their proliferative activity was lower than that of amelanotic cells. Ultrastructural examination verified that amelanotic melanophores possessed a relatively small number of premelanosomes while the transformants contained numerous melanosomes at various stages of pigment deposition. The result indicated that tyrosinase cDNA of R. nigromaculata was expressed in amelanotic melanophores of R. brevipoda inducing the maturation of premelanosomes. It was also suggested that the expression of transfected tyrosinase cDNA had promoted differentiation of the amelanotic cells into fully developed melanophores.     

Identification of Microtubule-Organizing Centers in Interphase Melanophores of Xenopus Laevis Larvae In Vivo

The morphological characteristics of microtubule-organizing centers (MTOCs) in dermal interphase melanophores of Xenopus laevis larvae in vivo at 51-53 stages of development has been studied using immuno-stained semi-thick sections by fluorescent microscopy combined with computer image analysis. Computer image analysis of melanophores with aggregated and dispersed pigment granules, stained with the antibodies against the centrosome-specific component (CTR210) and tubulin, has revealed the presence of one main focus of microtubule convergence in the cell body, which coincides with the localization of the centrosome-specific antigen. An electron microscopy of those melanophores has shown that aggregation or dispersion of melanosomes is accompanied by changes in the morphological arrangement of the MTOC/centrosome. The centrosome in melanophores with dispersed pigment exhibits a conventional organization, and their melanosomes are situated in an immediate vicinity of the centrioles. In melanophores with aggregated pigment, MTOC is characterized by a three-zonal organization: the centrosome with centrioles, the centrosphere, and an outlying radial arrangement of microtubules and their associated inclusions. The centrosome in interphase melanophores is presumed to contain a pair of centrioles or numerous centrioles. Because of an inability of detecting additional MTOCs, it has been considered that an active MTOC in interphase melanophores of X. laevis is the centrosome. We assume that remaining intact microtubules in the cytoplasmic processes of mitotic melanophores (Rubina et al., 1999) derive either from the aster or the centrosome active at the interphase.     

Volume Changes of Individual Melanosomes Measured by Scanning Force Microscopy

Black pigment cells, melanophores, e.g. located in the epidermis and dermis of frogs, are large flat cells having intracellular black pigment granules, called melanosomes. Due to a large size, high optical contrast, and quick response to drugs, melanophores are attractive as biosensors as well as for model studies of intracellular processes; e.g. organelle transport and G‐protein coupled receptors. The geometry of melanosomes from African clawed toad, Xenopus laevis, has been measured using scanning force microscopy (SFM). Three‐dimensional images from SFM were used to measure height, width, and length of the melanosomes (100 from aggregated cells and 100 from dispersed cells). The volumes of melanosomes isolated from aggregated and dispersed melanophores were significantly different (P<0.05, n=200). The average ellipsoidal volume was 0.14±0.01 (aggregated) and 0.17±0.01 μm³ (dispersed), a difference of 18%. The average major diameter was 810±20 and 880±20 nm for aggregated and dispersed melanosomes, respectively. To our knowledge, this is the first time SFM has been used to study melanosomes. This may provide an alternative non‐destructive technique that may be particularly suitable for studying morphological aspects of various melanin granules.     

Enhancement by fibronectin of spreading of isolated melanophores of the medaka,Oryzias latipes

Summary Cell spreading of isolated melanophores in medium containing fibronectin was observed in the wild type and two mutants of the medaka, Oryzias latipes. Isolated and cultured melanophores of the wild type and the mm mutant were different in appearance from those within scales but dendritic in shape and with fully dispersed pigment granules. Isolated melanophores of the cm mutant were stellate with dispersed pigment granules, whereas in scales the pigment granules are condensed. In the presence of fibronectin, spreading of cultured melanophores of wild type and cm mutant was observed. Spreading of melanophores from the mm mutant was observed only among dendritic melanophores, but not among condensed melanophores. The increase of spreading was inhibited by antibody against fibronectin. To test the involvement of cytoskeletal elements, colchicine, vinblastine or cytochalasin B were added to the culture medium; spreading did not increase, even in the presence of fibronectin. These results suggest that fibronectin-induced melanophore spreading is correlated with the state of pigment granule dispersal and that microtubules and microfilaments may play a role in the mechanism of spreading.Department of Zoology NJ-15, University of Washington, Seattle, Wa 98195, USA.     

Purification of Black Moor Goldfish melanophores and responses to epinephrine

Summary Two key modifications of the previously reported method for isolation of goldfish xanthophores allowed the isolation and establishment of primary cultures of terminally differentiated melanophores from the Black Moor goldfish (Carassius auratus). First, pretreatment with 10⁻⁴ M epinephrine causing aggregation of the melanosomes and collapse of the dendrites, prevents damage to the melanophores during tissue dissociation and melanophore isolation. Second, maintenance of these cells in culture was successful only when the culture medium was supplemented with fish serum. The purified melanophores attached, flattened, and were maintained in culture for up to 3 mo. Although the morphology of the cultured melanophores is less dendritic than their in vivo counterparts, the melanophores translocate melanosomes in a normal manner except that they exhibit enhanced sensitivity to epinephrine. This epinephrine-induced pigment aggregation, as well as the redispersion of pigment after the removal of epinephrine, can occur in the presence of ethylene glycol-bis (β-aminoethyl ether)-N, N, N′, N′-tetraacetic acid and absence of Ca²⁺. This work was supported by grant AM13724 from the National Institutes of Health, Bethesda, MD.     

Unusual development of light-reflecting pigment cells in intact and regenerating tail in the periodic albino mutant of Xenopus laevis

Unusual light-reflecting pigment cells, “white pigment cells”, specifically appear in the periodic albino mutant (a ^p /a ^p ) of Xenopus laevis and localize in the same place where melanophores normally differentiate in the wild-type. The mechanism responsible for the development of unusual pigment cells is unclear. In this study, white pigment cells in the periodic albino were compared with melanophores in the wild-type, using a cell culture system and a tail-regenerating system. Observations of both intact and cultured cells demonstrate that white pigment cells are unique in (1) showing characteristics of melanophore precursors at various stages of development, (2) accumulating reflecting platelets characteristic of iridophores, and (3) exhibiting pigment dispersion in response to α-melanocyte stimulating hormone (α-MSH) in the same way that melanophores do. When a tadpole tail is amputated, a functionally competent new tail is regenerated. White pigment cells appear in the mutant regenerating tail, whereas melanophores differentiate in the wild-type regenerating tail. White pigment cells in the mutant regenerating tail are essentially similar to melanophores in the wild-type regenerating tail with respect to their localization, number, and response to α-MSH. In addition to white pigment cells, iridophores which are never present in the intact tadpole tail appear specifically in the somites near the amputation level in the mutant regenerating tail. Iridophores are distinct from white pigment cells in size, shape, blue light-induced fluorescence, and response to α-MSH. These findings strongly suggest that white pigment cells in the mutant arise from melanophore precursors and accumulate reflecting platelets characteristic of iridophores.     

A Role for Spectrin in Dynactin‐dependent Melanosome Transport in Xenopus laevis Melanophores

The bi‐directional movement of pigment granules in frog melanophores involves the microtubule‐based motors cytoplasmic dynein, which is responsible for aggregation, and kinesin  II and myosin  V, which are required for dispersion of pigment. It was recently shown that dynactin acts as a link between dynein and kinesin  II and melanosomes, but it is not fully understood how this is regulated and if more proteins are involved. Here, we suggest that spectrin, which is known to be associated with Golgi vesicles as well as synaptic vesicles in a number of cells, is of importance for melanosome movements in Xenopus laevis melanophores. Large amounts of spectrin were found on melanosomes isolated from both aggregated and dispersed melanophores. Spectrin and two components of the oligomeric dynactin complex, p150^glued and Arp1/centractin, co‐localized with melanosomes during aggregation and dispersion, and the proteins were found to interact as determined by co‐immunoprecipitation. Spectrin has been suggested as an important link between cargoes and motor proteins in other cell types, and our new data indicate that spectrin has a role in the specialized melanosome transport processes in frog melanophores, in addition to a more general vesicle transport.     

Pigmentary function and evolution of tyrp1 gene duplicates in fish

The function of the tyrosinase‐related protein 1 (Tyrp1) has not yet been investigated in vertebrates basal to tetrapods. Teleost fishes have two duplicates of the tyrp1 gene. Here, we show that the teleost tyrp1 duplicates have distributed the ancestral gene expression in the retinal pigment epithelium (RPE) and melanophores in a species‐specific manner. In medaka embryos, tyrp1a expression is found in the RPE and in melanophores while tyrp1b is only expressed in melanophores. In zebrafish embryos, expression of tyrp1 paralogs overlaps in the RPE and in melanophores. Knockdown of each zebrafish tyrp1 duplicate alone does not show pigmentary defects, but simultaneous knockdown of both tyrp1 genes results in the formation of brown instead of black eumelanin accompanied by severe melanosome defects. Our study suggests that the brown melanosome color in Tyrp1‐deficient vertebrates is an effect of altered eumelanin synthesis. Black eumelanin formation essentially relies on the presence of Tyrp1 and some of its function is most likely conserved from the common ancestor of bony vertebrates.     

Changes in Adrenergic Innervation to Chromatophores During Prolonged Background Adaptation in the Medaka,Oryzias latipes

The pattern of adrenergic innervation to scale chromatophores of the wild-type medaka, Oryzias latipes, was examined by autoradiography with ³H-norepinephrine and found for the first time to be changed reversibly during prolonged background adaptation. In scales of the medaka, which was adapted to a black background for 10-15 days, a great number of melanophores and dense networks of varicose fibers were observed: many fibers built up a radial plexus around each melanophore. However, the dense distribution of varicose fibers disappeared with a decrease in the number of melanophores during long-term adaptation to a white background. As to the changes in the innervation pattern to amelanotic melanophores of the medaka, orange-red variety, a similar result was obtained. Although the increase in the number of leucophores was observed in the medaka adapted to a white background, no exact plexuses of labeled fibers were confirmed around leucophores. From these results, it is concluded that the density of chromatic nerve fibers changes in parallel with the variation of the number of melanophores during prolonged background adaptation.     

Brightly colored pigmentation in lower vertebrates: wonder searching its mechanisms and significance in the context of phylogeny

This is a biographical sketch of my research and its related personal episodes with respect to brightly colored pigmentation in lower vertebrates. It includes a brief story of the studies on; (a) pterinosomes as a specific site of pteridine deposition in xanthophores or erythrophores of fish and amphibians, (b) a mosaic phenotype of chromatophores occurring in the reptiles and its implication for their developmental origin and differentiation mechanisms, (c) erythrophoroma as a tumor of erythrophores in goldfish, (d) the pluripotentials of erythrophoroma cells for expression of neural crest-derived characters in vitro, (e) pigment disorders occurring in hatchery-raised flounders and (f) recognition of pigment cell types by murine tyrosinase genes transfected into an orange-colored variant of medaka fish. Some of the personal affairs associated with the history of the Japanese community for pigment cell research were described to illustrate the background of these studies.     

ABCB6 Resides in Melanosomes and Regulates Early Steps of Melanogenesis Required for PMEL Amyloid Matrix Formation

Genetically inheritable pigmentation defects provide a unique opportunity to reveal the function of proteins contributing to melanogenesis. Dyschromatosis universalis hereditaria (DUH) is a rare pigmentary genodermatosis associated with mutations in the ABCB6 gene. Here we use optical and electron microscopy imaging combined with biochemical tools to investigate the localization and function of ABCB6 in pigment cells. We show that ABCB6 localizes to the membrane of early melanosomes and lysosomes of the human melanocytic cell line MNT-1. Depletion of ABCB6 by siRNA impaired PMEL amyloidogenesis in early melanosomes and induced aberrant accumulation of multilamellar aggregates in pigmented melanosomes. PMEL fibril formation and normal maturation of pigmented melanosomes could be restored by the overexpression of wild-type ABCB6 but not by variants containing an inactivating catalytic mutation (K629M) or the G579E DUH mutation. In line with the impairment of PMEL matrix formation in the absence of ABCB6, morphological analysis of the retinal pigment epithelium of ABCB6 knockout mice revealed a significant decrease of melanosome numbers. Our study extends the localization of ABCB6 to melanosomes, suggesting a potential link between the function of ABCB6 and the etiology of DUH to amyloid formation in pigment cells.     

The effect of cytochalasin B on pigment dispersion and aggregation in perfused Xenopus laevis tailfin melanophores

A perfusion technique is described for the study of melanosome response in ventral tailfin melanophores of Xenopus laevis tadpoles. The melanosomes remain aggregated (punctate melanophores) in Ringer's. Theophylline (15 mM) and caffeine (30 mM) cause a reversible dispersion (stellate melanophores) of melanosomes which is partly blocked by cytochalasin B (10 μg/ml). When added with theophylline or caffeine to stellate cells, cytochalasin B causes a disrupted distribution of pigment granules, characterized by a melanosome free central region. C-AMP (20 mM) and dibutyryl c-AMP (1 mM) cause a reversible dispersion of melanosomes which is partly inhibited by cytochalasin. When cytochalasin plus a nucleotide are added to stellate cells, some show the disrupted distribution of melanosomes. Colchicine (5 mM) causes irreversible, while griseofulvin (0.2 mM) causes a slight, but reversible dispersion of melanosomes, and cytochalasin has little effect on these reactions. Perfused tailfin melanophores remain capable of responding to reversible reagents for at least 12 hours and are unresponsive to changes in illumination.     

Novel receptors for melatonin that mediate pigment dispersion are present in some melanophores of the pencil fish (Nannostomus)

1. Comparing the daytime and the night-time pigmentary patterns of the skin of the pencil fish, Nannostomus beckfordi, we noticed that specific regions of dark spots that were part of the night-time pattern became pale during the day.2. Microscopic observations revealed that melanosomes in the melanophores in those regions were aggregated during the day but became dispersed at night.3. These melanophores responded to melatonin by dispersal of melanosomes while the cells on other parts of the body responded to melatonin by aggregation of the pigment in the normal way.4. The melanophores that responded to melatonin by pigment dispersion responded normally to other hormones and neurotransmitters, as did those on other parts of the skin.5. The results indicate that, in addition to the known melatonin receptor that mediates the aggregation of melanosomes, there also exists an unusual receptor which mediates the dispersion of pigment in melanophores. We have tentatively designated this receptor the ‘beta-melatonin receptor’.