Chapter 72. Biology of Melanocytes

Melanocytes are pigment-producing cells that originate from the dorsal portions of the closing neural tube in vertebrate embryos1 (eFig. 72-0.1). They derive from pluripotent neural crest cells that differentiate into numerous cell lineages including neurons, glia, smooth muscle, craniofacial bone, cartiledge, and melanocytes.2,3 Progenitor melanoblasts migrate dorsolaterally between the mesodermal and ectodermal layers to reach their final destinations in the hair follicles and the skin as well as inner ear cochlea, choroids, ciliary body, and iris.2,4 Pigment-producing cells can be found in fetal epidermis as early as the 50th day of gestation.

Melanoblast migration and differentiation into melanocytes are influenced by a number of signaling molecules produced by neighboring cells. These include Wnt, endothelin (ET)-3, bone morphogenetic proteins (BMPs), steel factor (SF) (stem cell factor, c-Kit ligand), and hepatocyte growth factor (HGF/scatter factor).5–10 By interacting with their specific cell surface receptors, these molecules induce intracellular and intranuclear signaling to influence gene transcription and protein synthesis. Genetic defects in some of these molecules are associated with human genetic diseases: ETs (Waardenburg syndrome and Hirschsprung disease) and c-Kit and stell factor with piebaldism. Detailed discussion of each signaling molecule is available online.

Wnt

Bone Morphogenetic Proteins

Endothelins

Steel Factor

Hepatocyte Growth Factor

Cadherins

Melanocyte Stem Cells

Generally, stem cells are defined by their undifferentiated state and their capacity to develop into several differentiated cell types. They are quiescent, slow-cycling cells that frequently are found in niches where they are surrounded by differentiated cells that affect their behavior through the secretion of cytokines and growth factors.29,30

Melanocyte stem cells reside in the hair follicle bulge (Fig. 72-1). They express TRP-2 as well as the neural crest stem cell intermediate filament nestin in addition to other neural crest stem cell markers including the transcription factors Sox10 and Pax5 that participate in the regulation of microphthalmia-associated transcription factor (MITF) and TRP-2. Melanocyte stem cells can leave the bulge region and migrate/differentiate in the epidermis or the hair follicle.

Cutaneous Melanocytes

The largest number of melanocytes are present in the skin and hair follicles. While in most furred mammals melanocytes are found only in the hair follicle, in humans, melanocytes are also present in interfollicular epidermis, specifically in the basal layer.28 There is approximately one melanocyte per five or six basal keratinocytes. Melanocytes synthesize melanin, a pigmented polymer that is stored in cytosolic organelles called melanosomes that are transferred to keratinocytes through melanocyte dendritic processes (Fig. 72-2). As keratinocytes are continuously being desquamated, there is a constant need for synthesis and transfer of melanosomes from melanocytes to keratinocytes in order to maintain cutaneous pigmentation.

The term “epidermal melanin unit” describes a single epidermal melanocyte surrounded by several epidermal keratinocytes31 (Fig. 72-3). Interestingly, signals from keratinocytes substantially regulate epidermal melanocyte survival, dendricity, melanogenesis and the expression of cell surface receptors32 (see below). Most keratinocyte-derived signals are induced by ultraviolet (UV) irradiation.

Melanocyte density/mm2 ranges from approximately 550 to >1,200 with the highest concentrations found in the genitalia and face.33,34 Melanocyte density is the same in individuals of different racial backgrounds,34 and thus cutaneous pigmentation does not depend on melanocyte number, but rather upon melanogenic activity within the melanocyte, the proportion of mature melanosomes, and/or their transfer and distribution within the keratinocytes.35 Indeed, in light-skinned individuals, melanosomes are smaller and are present in clusters within the keratinocytes, while in ethnic groups with darker complexion, melanosomes are larger, darker, and are individually dispersed within the keratinocytes.36

Hair Follicle Melanocytes

In contrast with interfollicular epidermal melanocytes, the follicular melanin unit undergoes cyclic modifications in coordination with the hair cycle (Fig. 72-4). Melanocytes are located in the proximal hair bulb during anagen and there is a ratio of 1:5 between melanocytes and keratinocytes and 1:1 between melanocytes and basal layer keratinocytes.37 Melanocytes proliferate, migrate, and undergo maturation during early to mid anagen. Melanogenesis and melanin transfer to keratinocytes occurs throughout anagen. Melanocytes eventually apoptose during late catagen. In mice, melanocyte proliferation and differentiation during anagen depends on c-Kit expression by melanocytes and SF synthesis by keratinocytes.38

Similar to their role in the epidermis, in hair, melanocytes transfer melanin to differentiated keratinocytes that ultimately form the hair shaft. They thus determine hair color by the amount of melanin transferred, as well as by the ratio of eumelanin (black–brown) to pheomelanin (red–yellow) (see below).39

In hair, melanin does not appear to have a protective effect, since UV irradiation does not reach the hair follicle. Still, in furry animals hair color plays an important role in camouflage, mimicry, and social communication.40 It is also speculated that melanocytes restrain keratinocyte proliferation, affect calcium homeostasis, and protect against reactive oxygen species (ROS) during the rapid proliferation and differentiation of the hair follicle.37

Ocular Melanocytes

Otic Melanocytes

Cephalic Melanocytes

The major differentiated function of melanocytes is to synthesize melanin in specialized organelles within the melanocytes, the melanosomes, and to transfer melanosomes to neighboring keratinocytes in order to provide protection from UV irradiation (Fig. 72-5). Melanogenesis, the synthesis and distribution of melanin in the epidermis, involves several steps: transcription of proteins required for melanogenesis; melanosome biogenesis; sorting of melanogenic proteins into the melanosomes; transport of melanosomes to the tips of melanocyte dendrites and transfer of melanosomes to keratinocytes. Disruption in any of these events results in hypopigmentation.

Melanosome Biogenesis

The melanosome is a unique membrane-bound organelle in which melanin biosynthesis takes place. Because melanosomes contain enzymes and other proteins also present in lysosomes, they are thought to represent a modified version of the latter. Proteins common to both organelles include the lysosome-associated membrane proteins (LAMPs) that participate in autophagy and regulation of intravesicular pH,56 as well as acid phophatase, a marker enzyme for lysosomes.56 Also like lysosomes, melanosomes can endocytose receptors that are targeted for degradation.56

Depending on the type of melanin synthesized, melanosomes can be divided into eumelanosomes and pheomelanosomes (Fig. 72-6). Eumelanosomes are large (∼0.9 × 0.3 mm), elliptical in shape and contain a highly structured fibrillar glycoprotein matrix required for eumelanin synthesis.40 Pheomelanosomes are smaller (∼0.7 mm in diameter), spherical in shape and their glycoprotein matrix appears disorganized and loose40 Although both eumelanosomes and pheomelanosomes may be present within a single melanocyte,57 once committed, they do not change.58

Melanosomes display four maturation stages (Fig. 72-6). Stage I melanosomes or premelanosomes likely develop from the endoplasmic reticulum (ER).40 They have an amorphous matrix and display internal vesicles that form as a result of membrane invagination. Premelanosomes already contain the glycoprotein Pmel17 (gp100) but it requires further processing to become a component of the final fibrillar matrix.59 Stage II eumelanosomes have organized structured fibrillar matrix, but no active melanin synthesis, whereas in stage II pheomelanosomes melanin synthesis already takes place. Although no active melanogenesis takes place in stage II eumelanosomes, they already contain the enzyme tyrosinase. Deposition of melanin on the fibrillar matrix is found in stage III eumelanosomes, while stage IV eumelanosomes are fully melanized and their internal matrix is masked by melanin deposits.60,61

Melanogenic Proteins

The timely and organized sorting of melanogenic enzymes and structural proteins to melanosomes is an integral part of melanosomal maturation. Melanosome proteins express sorting signals at their amino-terminus and these direct them into the ER and eventually into the melanosomes.40,60,61

Enzymes

Tyrosinase

Tyrosinase is present in plants, insects, amphibians, and mammals. It was initially identified in the early 1900s in mushroom extracts and was subsequently isolated and purified in 1949 from murine melanoma cells.62 Mouse and human tyrosinase genes are 60–70 kb, and 50 kb long, respectively. The murine tyrosinase gene maps to chromosome 7, whereas human tyrosinase gene maps to chromosome 11. The human tyrosinase gene is composed of five exons and four introns,63 and tyrosinase mRNA is approximately 2 kb long (genebank access number NM_000372).

Tyrosinase is synthesized in the ER as a precursor protein whose nascent chain is processed in the Golgi complex where sialic acid and neutral sugars are added to the peptide via N- and O-glycosidic linkages through a process called glycosylation64 (Fig. 72-7). At least four forms of tyrosinase, all differing with regards to their degree of glycosylation, have been identified. The glycolsylation steps have been shown to be important for proper association of tyrosinase with melanosomes, as well as for its activity.64 Following the glycosylation steps, mature tyrosinase is folded in the ER, a step required for appropriate trafficking/sorting of tyrosinase into the Golgi apparatus and ultimately into endosomes and finally into melanosomes. A strict control mechanism guarantees the elimination of defective tyrosinase.

Within the melanosome, tyrosinase spans melanosomal outer membrane (eFig. 72-7.1). It has three domains: (1) an inner melanosomal domain, (2) a melanosomal transmembrane domain, (3) and a cytoplasmic domain. The inner domain that contains the catalytic region is approximately 90% of the protein. It is followed by a short transmembrane domain, and a cytoplasmic domain composed of approximately 30 amino acids.65 Histidine residues present in the inner (catalytic) portion of tyrosinase bind copper (Cu) ions and the latter are required for tyrosinase activity.

The biological function of the tyrosinase cytoplasmic domain was not known for a long time. In a mouse model where the entire cytoplasmic domain is missing, tyrosinase protein is inserted into the cellular plasma membrane instead of into the melanosomal membrane, suggesting that tyrosinase cytoplasmic domain is required for proper trafficking of tyrosinase into melanosomes. Indeed, it was found that the motif EXXQPLL (glutamic acid-X-X-glutamine-proline-leucine-leucine, where “X” stands for any amino acid) in the cytoplasmic domain is responsible for tyrosinase trafficking into the melanosomes.66 In addition, protein kinase C-β (PKC-β) (see below) must phosphorylate two serine residues on this cytoplasmic domain to activate tyrosinase,65 and in the absence of those phosphorylation events pigmentation does not occur.

Tyrosinase mutations including missense, nonsense frameshift, and deletion mutations that lead to inactivation of the enzyme are found in oculocutaneous albinism type I (see Chapter 73 and Albinism database: http://albinismdb.med.umn.edu/). Such mutations may affect tyrosinase glycosylation interfering with enzyme maturation, or may involve Cu-binding sites disrupting tyrosinase activity or premature termination of tyrosinase protein that causes truncation of cytoplasmic domain.64

Tyrosinase-Related Proteins

Two tyrosinase-related proteins, (1) TRP-1 and (2) TRP-2, play important roles in melanogenesis.67–69 They are structurally related to tyrosinase and share ∼40% amino acid homology. Also, similar to tyrosinase, TRP-1 and TRP-2 are glycoproteins located within the melanosomes and span the melanosomal membrane.70 The conserved nucleotide and amino acid sequences among these three melanogenic enzymes suggest that they originated from a common ancestral gene.71,72 Mutations of TRP-1 and TRP-2 in mice after coat color (“brown” and “slaty” mice, respectively) and polymorphisms of these gene products are implicated in lighter hair and skin color in European population studies. Their functions are incompletely understood, but the proteins complex with tyrosinase and may stablilize it.

TRP-1

TRP-2

Protein Kinase C-β

PKC constitutes a family of at least 12 isoforms86 among that PKC-β has been shown to be involved in regulating tyrosinase activity.87 The mechanisms through which PKC mediates a wide range of membrane generated signals and their relevance to melanocyte biology are further discussed below (see Section “Signaling Pathways Regulating Melanocyte Function”). PKC-β phosphorylates serine residues on the cytoplasmic domain of tyrosinase, thus activating tyrosinase.88 Still, the means by which PKC-β-mediated phosphorylation of tyrosinase leads to the enzyme activation is not well elucidated. It has been suggested that phosphorylation of tyrosinase causes a complex to form between tyrosinase and TRP-1,89 an event known to stabilize tyrosinase and increase its enzymatic activity.79

In melanocytes, activated PKC-β is associated with melanosomes and the enzyme is found in close proximity to the melanosomal membrane.88 While structural differences among PKC isoforms may contribute to their associations with particular subcellular fractions, receptors for activated C-kinase (RACK), unique for each PKC isoform, primarily determine the translocation of specific PKC isoforms to specific cellular compartments to activate its target on the membrane90–92 (Fig. 72-8). RACK-I is the PKC-β partner,90 and in human melanocytes, the activated PKC-β/RACK-I complex translocates to the melanosome membrane to allow tyrosinase phosphorylation (Fig. 72-8).93

Structural Proteins

Fibrillar matrix proteins within the melanosomes are required for proper deposition of melanin. Pmel17 and MART-1 are two such melanosomal structural matrix proteins.

Pmel17

Pmel17, also known as gp100 and the silver locus product, is a glycoprotein recognized by the antibodies HMB45, HMB50, and NKI-β.94 It plays a critical role in fibril matrix formation within eumelanosomes.59,95 Pmel17 transcription is induced by α-MSH through MITF and it is synthesized as a precursor protein in the ER and the protein undergoes glycosylation and eventual cleavage (Fig. 72-7). After its synthesis, Pmel17 is transported to stage I melanosomes to form a fibrillar structure that is the backbone of eumelanosome matrix,94 contributes to melanosome ellipsoid shape and promotes melanin polymerization.94 Melanosomes lacking Pmel17 cannot transit to stage II and have no active melanogenesis.96 It has been suggested that loss of functional Pmel17 results in melanin cytotoxicity, perhaps through leakage of melanin intermediates from abnormal melanosomes into the cytosol.94,97 However, there are no known hypopigmentary disorders in humans linked to mutations of Pmel17.

MART-1/Melan A

MART-1, also known as Melan A, is a membrane-associated protein98 that is present in stage I and stage II melanosomes and forms a complex with Pmel17 (Fig. 72-7). MART-1 affects the expression, stability, trafficking, and processing of Pmel17 within the melanosomes.98 To date, no hypopigmented phenotypes associated with nonfunctional MART-1 have been identified.

Transport Proteins

Heterotetrameric Adaptor Protein Complexes

Sorting of membrane-associated proteins including tyrosinase, TRP-1, TRP-2, and Pmel17 and directing them to the appropriate cytosolic organelles is facilitated by heterodimeric adapter protein complexes (APs).99,100 AP-3 and possibly also AP-1 facilitate tyrosinase transport from endosomes to melanosomes101 (Fig. 72-7). Patients with Hermansky–Pudlak syndrome—an autosomal recessive disorder of oculocutaneous albinism, platelet dysfunction, and pulmonary disease (see Chapter 73)—have defects in specific subunits of the AP-3 adaptor protein complex and as a result display several anomalies associated with cellular transport of molecules.102 Studies suggest that a molecule of the kinesin family, microtubule-associated motor proteins, is involved in endosomal sorting and positioning of melanosomal proteins via interaction with AP-1.103,104

P-Protein

The P-protein (pink-eyed dilution) is a transmembrane protein with 12 membrane-spanning domains whose sequence is homologous to that of other transmembrane transport proteins, including anion transporters97,105,106 thought to function as a transport protein.107 Studies have identified the protein as an ATP-associated proton pump responsible for maintaining acidic environment within the melanosomes.108 Other proposed functions of P-protein include stabilizing the tyrosinase/TRP-1/TRP-2 complex and/or transporting tyrosine into the melanosomes.105 Individuals lacking functional P-protein display occulocutaneous albinism type 2, largely due to improper melanosomal pH.108–110 Also, Angelman and Prader–Willi syndromes display deletion mutations that include the P-locus on chromosome 15.

SLC24A5

SLC24A5 is a melanosomal protein whose structure and homology to cation exchange proteins suggests that it is a melanosome-associated cation exchanger.111 Mutations in slc24a5 in zebrafish lead to hypopigmentation of the organism.111 The ancestral human homolog is expressed by darker complexioned individuals including Africans and Asians, while lighter complexioned Europeans tend to express a variant allele.111

Scavenger Protein

Lysosome-Associated Membrane Proteins

LAMPs are linked to melanosome membranes and/or matrix. They are thought to protect melanosomal integrity by acting as scavengers of free radicals that are produced during melanin biosynthesis.112 Because LAMPs are also present in lysosomes, it is thought that melanosomes and lysosomes share a common ancestral origin.112

Regulatory Proteins

Microphthalmia-Associated Transcription Factor

Gene and Protein

MITF, a basic-helix-loop-helix (bHLH) and leucine zipper transcription factor, has been termed the master gene for melanocyte survival and is a key factor regulating the transcription of the major melanogenic proteins, tyrosinase, TRP-1, TRP-2,12 PKC-β113 and MART-1.114 In melanocytes, it is the MITF-M isoform that stimulates transcription of tyrosinase and PKC-β.114 MITF binds to conserved consensus elements in gene promoters, specifically the M- (AGTCATGTGCT) and E- (CATGTG) boxes.115 It can bind as a homodimer or a heterodimer with another related family member.116 MITF appears to be a key regulator determining cell fate, as transfection of human MITF cDNA into mouse fibroblasts converts these cells into dendritic cells expressing melanocyte-specific genes.117

MITF comprises a family of nine isoforms: (1) MITF-M, (2) -A, (3) -B, (4) -H, (5) -C, (6) -D, (7) -E, (8) -J, and (9) -Mc.118,119 MITF-M expression is highly specific for melanocytic cells.120 Melanocytes express in addition other MITF isoforms specifically, MITF-A, -B, and, -E.114 The biologic role of other MITF isoforms in normal melanocytes is not known.

Regulation of MITF Activity and Expression

The activity and stability of MITF are modulated by phosphorylation of the protein. MITF activity is increased upon its phosphorylation by the mitogen-activated protein kinase-2 (MAP kinase-2), whose activity is in turn induced by binding of SF/kit/stem cell factor to c-Kit receptor121 (Fig. 72-8). Phosphorylated MITF binds to another protein, p300/CBP, which belongs to a coactivator family of proteins and acts to enhance MITF transcriptional activity.121,122 Another kinase that is activated by SF/c-Kit interaction is p90RSK that also phosphorylates MITF but at a different site from that phosphorylated by MAP kinase-2.123 These phosphorylation events both activate MITF and at the same time decrease the stability of the protein, as phosphorylated MITF is targeted for degradation by proteosomes (eFig. 72-8.1A).123,124

The expression of MITF is under the control of several transcription factors, including Sox10 (mutated in Waardenburg's syndrome type 4, see Chapter 73) and Pax3. MITF expression is also controlled by the cAMP-response element binding protein (CREB) and Lef1 transcription factor that participates in Wnt-signaling. These transcription factors bind to specific sites within MITF promoter regions to induce MITF transcription.116 The promoter region of the MITF gene contains a cAMP-response element (CRE) that interacts with CREB when the cAMP-dependent pathway is activated.125,126 Therefore, cAMP-elevating agents like α-MSH induce the expression of MITF (eFig. 72-8.1B).

MITF Role in Melanocyte Proliferation and Survival

MITF promotes melanoycte survival by upregulating the expression of a major antiapoptotic protein BCl2.127 It is frequently overexpressed or amplified in melanomas, contributing to their increased survival.128–131 Mutations in MITF are found in the pigmentary disorder Waardenburg syndrome type 2132(see Chapter 73).

A role for MITF in melanocyte proliferation has also been proposed, as under certain conditions, MITF induces the expression of the cell cycle-associated kinase Cdk2 that is involved in the progression of cells from G1 into S phase of the cell cycle.133 MITF also suppresses the expression of p21, a protein that inhibits Cdk2 activation.133,134 Conversely, under different conditions, MITF can induce p21 expression and it can also stimulate the expression of p16INK4a, a protein that inhibits the activation of kinases required for progression through the cell cycle, thus promoting cell cycle arrest.135,136 Because MITF cooperates with other transcription factors to induce its effects, it is to be expected that these transcription factors would influence MITF activity, resulting in either stimulation or inhibition of melanocyte proliferation.

Mice bearing null mutations of MITF display loss of melanocytes, deafness and failure of differentiation of retinal pigment epithelium.12

Melanocortin 1 Receptor

Melanocortin receptors (MCRs) comprise a family of five related receptors (MC1R, MC2R, MC3R, MC4R, and MC5R). Each has seven transmembrane domains and they belong to the G-protein-coupled receptor superfamily.137 MC3R and MC4R are mainly found in the central nervous system, are absent in melanocytes,138 and are thought to control energy intake. MC2R is expressed in the adrenal cortex and MC5R is expressed in peripheral adipocytes.139 MC1R is expressed in a number of cells such as endothelial cells, fibroblasts,140 and keratinocytes,140 but the highest expression is found in melanocytes.140

α-MSH and adrenocorticotropic hormone (ACTH), a 39 amino acid proopiomelanocortin-derived peptide that contains the α-MSH sequence (Fig. 72-9A),141–143 activate MC1R (Fig. 72-9B) Receptor–ligand interaction leads to G-protein-dependent activation of the enzyme adenylate cyclase followed by increased intracellular cAMP level141 inducing MITF transcription and upregulating the level of melanogenic proteins including tyrosinase40 promoting the synthesis of brown/black eumelanin.141 Agouti, a protein expressed in both humans and mice, whose expression in mice leads to yellow coat color, antagonizes α-MSH by competitive binding to MC1R. It thus blocks adenylate cyclase activation144–146 and favors pheomelanin over eumelanin synthesis (Fig. 72-10). However, the role of agouti in human pigmentation is poorly documented.

Polymorphisms within the MC1R gene are largely responsible for the different skin/hair color among different racial groups.147 At least 30 MC1R variants have been identified and nine of them display loss of function148,149 (Fig. 72-9B), not being able to induce intracellular cAMP production in response to α-MSH despite adequate receptor/ligand binding. Other MC1R variants have reduced affinity for α-MSH.148,149 Three MC1R variants, each with only a single amino acid substitution, have been associated with red/yellow hair and fair skin150 of Northern Europeans and Australians.151–156 Mice expressing a loss-of-function MC1R variant receptor also fail to respond to UV irradiation with increased pigmentation despite an increased level of epidermal α-MSH,157 but do tan if provided forskolin, a chemical enhancer of pigmentation that bypasses the receptor to directly increase cAMP, demonstrating that the intracellular melanogenic pathway is functional in such individuals.155

Two types of melanins are synthesized within melanosomes: (1) eumelanin and (2) pheomelanin.158 Eumelanin is dark, brown–black, and insoluble, whereas pheomelanin is light, red–yellow sulfur-containing, and soluble.158 Melanins are indole derivatives of 3,4 di-hydroxy-phenylalanine (DOPA) and they are formed in melanosomes through a series of oxidative steps159 (Fig. 72-11). Melanosomal pH affects the activity of the melanogenic enzymes and influences melanin polymerization.

The synthesis of both types of melanin involves the rate-limiting catalytic step in which the amino acid tyrosine is oxidized by the enzyme tyrosinase (also called tyrosine oxidase, DOPA oxidase, monophenol, DOPA: oxygen oxidoreductase) to DOPA, a first step in a reaction known as the Raper–Mason pathway160 (Fig. 72-11). Conversion of tyrosine to DOPA is thought to be the critical rate-limiting step in melanogenesis, as inhibition of this reaction blocks melanin synthesis.161 In both reactions, DOPA acts as a cofactor and also as a substrate for tyrosinase. Although the exact interaction between tyrosinase and its substrates is not completely understood, in vitro kinetic studies suggest that distinct sites mediate tyrosinase binding to tyrosine and to DOPA, and that binding to DOPA causes a conformational change in tyrosinase, resulting in increased affinity for both tyrosine and DOPA.

DOPA is oxidized into DOPAquinone,162 DOPAquinone is further converted to DOPAchrome and DOPAchrome can be converted to 5,6-dihydroxyindole (DHI) or to 5,6-dihydroxyindole-2-carboxylic acid (DHICA). The latter reaction is catalyzed by the enzyme DOPAchrome tautomeras or TRP-2. The level of brown versus black eumelanin appears to correlate with DHI/DHICA ratio, with a higher ratio leading to the formation of black eumelanin and a lower ratio to brown eumelanin.163 DOPAquinone can also combine with glutathione or cyteine to form cysteinylDOPA, which then becomes the yellow/red, soluble, low-molecular-weight pheomelanin.163 Interestingly, tyrosinase also catalyzes a more distant step in melanin biosynthesis, namely DHI conversion to indole-5,6-quinone. In mice, the enzyme TRP-1 (also called DHICA oxidase) converts DHICA to indole-5,6-quinone carboxylic acid. However, the TRP-1 role in human melanin biosynthesis is not well established.

The main function of melanin is to provide protection against UV-induced DNA damage by absorbing and scattering UV radiation (280–400 nm). Accordingly, energy absorption by melanin is maximal in this portion of the electromagnetic spectrum, and decreases gradually across the visible light spectrum. UV absorbed by melanin is converted into heat, a less toxic form of energy.164 Still, in vitro studies conducted by several investigators suggest that melanin's capacity to act as a sunscreen is limited and that melanin, when incorporated into a cream and spread over the skin, absorbs only 50%–75% of incident sunlight. Naturally, it is possible that in vivo, by virtue of localizing above the nucleus, melanin in melanosomes achieves a higher level of protection.

Melanin intermediates as well as melanin itself can also be harmful to the cell because, depending on their molecular weight and polymerization state, they can promote UVA (320–400 nm)-induced DNA damage, most likely through the generation of ROS.165 It has been suggested that the increased incidence of UV-induced melanomas in light-skinned, red-hair individuals is not only due to decreased ability of pheomelanin to protect against UV-induced DNA damage, but may also be due to mutagenic capacity of pheomelanin and possibly other melanin intermediates as a result of their pro-oxidant capacity.166

Melanocyte dendrites are branching protoplasmic processes that interact with keratinocytes. Actin is a major structural component of melanocyte dendrites, and actin filament disruption leads to dendrite loss.167 Cocultures of keratinocytes and melanocytes demonstrate that keratinocyte-derived factors play a role in melanocyte dendricity.168 These factors include ET-1, nerve growth factor (NGF), α-MSH, ACTH, prostaglandins (PGs) E2 and F2α168 and β-endorphin.169 Integrins, receptors that mediate actin-extracellular matrix contact are likely to play a role in dendrite formation as well.170

Another group of proteins, the Rho family, also plays a role in melanocyte dendrite formation. Rho proteins become active when they bind GTP and inactive when binding GDP.171,172 It appears that when Rho is activated, dendrites retract; while when its family member Rac is activated, dendrites form.172 Indeed, it is currently assumed that by increasing cAMP levels, α-MSH inhibits Rho, enhancing melanocyte dendricity. Thus, the equilibrium between Rho and Rac appears to be an important factor influencing melanocyte dendricity.

Within Melanocytes

Melanosomes are transferred from their site of origin in melanocyte perikaryon to the tips of melanocyte dendrites. Melanosome transport takes place on microtubules that are arranged parallel to the long axis of the dendrite and is controlled by two classes of microtubule-associated motor proteins: (1) kinesins173–175 and (2) cytoplasmic dyneins176–180 (Fig. 72-12). Both motor proteins act as short cross-bridge structures connecting the organelle to the microtubules.

Centrifugal, anterograde organelle movement is mediated primarily by kinesin, whereas centripetal movement is controlled by cytoplasmic dynein. Studies examining melanosomal transport suggest that their microtubule-dependent movement is bidirectional,181 consistent with a cooperative forward and backward pull of kinesin173 and dynein,176 respectively. For melanosomes with net centrifugal movement, the bidirectional movement appears to terminate with myosin-Va (encoded by dilute locus)-dependent melanosomal capture in the actin-rich periphery of the dendrite (Fig. 72-12).181

Additional proteins that participate in melanosome transport include Rab27a (encoded by ashen locus) that mediates myosin-Va binding to melanosomes through another linker protein-melanophilin (encoded by leaden locus) (Fig. 72-12).182 In the absence of myosin-Va, melanosomes do not collect in dendrite tips.

Mutations in any of the above gene products results in decreased cutaneous pigmentation. Griscelli syndrome, a rare autosomal recessive disorder in which individuals display dilute skin and hair color, is the result of mutations of myosin-Va, Rab27a, or melanophilin182 (see Chapter 73). Myosin-Va and Rab27a are closely located on chromosome 15.183–186 Because myosin-Va is also expressed in the brain, mutations of this gene may also cause neurological abnormalities. Rab27a also plays a role in immuno-regulation and individuals with mutations of this gene display abnormalities of the immune system. Mutations of melanophilin result only in the distinctive hypopigmentation that characterizes the syndrome.186

To Keratinocytes

Transfer of melanosomes from melanocytes to neighboring keratinocytes is a critical step in normal pigmentation. Studies suggest several ways for melanosomal transfer, including exocytosis, cytophagocytosis, fusion of plasma membranes, and transfer by membrane vesicles.187

The exocytosis pathway of melanosomal transfer involves fusion of the melanosomal membrane with the melanocyte plasma membrane, melanosome release into the intercellular space and phagocytosis by surrounding keratinocytes. Cytophagocytosis is a term indicating the phagocytosis of a live cell or a portion of it. With regard to keratinocytes, they cytophagocytose the tip of a melanocyte dendrite, which then fuses with lysosomes inside the keratinocyte, is transported to a supranuclear location where the phagolysosome membranes break up releasing the melanosomes. Fusion of keratinocyte and melanocyte plasma membranes creates a space through which melanosomes are transferred from the melanocyte to the keratinocyte. Indeed, high-resolution photography shows the presence of filopodia—slender, filliform, pointed cytoplasmic projections at the tip of melanocyte dendrites.188 These filopodia adhere and fuse with keratinocyte plasma membrane prior to melanosome transfer. Another way of melanosomal transfer involves shedding of melanosome-filled vesicles followed by phagocytosis of these vesicles by keratinocytes, or their fusion with keratinocyte plasma membrane.

The molecular and cellular mechanisms involved in melanosome phagocytosis have been partially elucidated. It appears that keratinocytes express a seven-transmembrane G-protein coupled receptor called protease activated receptor-2 (PAR-2). PAR-2 is activated when serine proteases cleave the extracellular portion of the receptor, exposing a new segment that acts as a tethered (attached) ligand.189,190 Activation of PAR-2 increases keratinocyte phagocytic activity.189,190

Interestingly, and consistent with its role in melanosome phagocytosis, UV induces the activity and expression of PAR-2.191 UV effect on PAR-2 activity and expression is more pronounced in individuals with skin phototypes II and III than in those with skin phototype I.191 Keratinocyte growth factor receptor has also been implicated in enhancing phagocytosis of melanosmes by keratinocytes.192

Melanocyte behavior in skin is largely influenced by signals from neighboring keratinocytes as well as autocrine signals and environmental factors such as UV irradiation (also see Section “UV Irradiation and Melanocytes”). The synthesis and secretion of most keratinocyte-derived factors is increased by UV irradiation, but it is also evident that UV can directly stimulate melanocyte dendricity and melanin production.171,193 Melanocytes receive both positive and negative paracrine signals that modulate their proliferation and differentiated function.

Melanogenic Stimulators

Proopiomelanocortin and Derived Peptides

It is well documented that MSH and ACTH are potent stimulators of melanogenesis. They belong to a family of peptides derived from the precursor proopiomelanocortin (POMC) that is synthesized, in addition to the pituitary gland, also by epidermal keratinocytes. Interestingly, POMC expression in keratinocytes is induced by UV, phorbol esters, and interleukins.194,195 In rodents, α-MSH stimulates melanogenesis and favors eumelanin over pheomelanin production, but systemic administration of α-MSH, β-MSH, and ACTH to people increases skin pigmentation predominantly in sun-exposed areas.196,197 However, in certain disease conditions characterized by abnormally high levels of ACTH, such as Addison disease198 or Nelson's syndrome199 (ACTH secreting pituitary adenoma), more generalized hyperpigmentation of the skin has been observed.200

Aside from its effect on melanogenic proteins and eumelanin synthesis, α-MSH was also reported to enhance the repair of UV-induced DNA damage in melanocytes, specifically the repair of pyrimidine dimers, and also to reduce the level of UV-induced hydrogen peroxide in the cell.201 In addition, α-MSH was shown to regulate melanosomal pH.202 These data suggest a role for POMC-derived peptides beyond merely stimulating melanogenesis.

Endothelin-1

ET-1 appears to play a role in mature melanocytes, inducing melanogenesis by activating tyrosinase and increasing TRP-1 levels.203,204 ET-1 also leads to melanocyte proliferation203,204 and promotes dendrite formation.205 Cultured keratinocytes synthesize and secrete ET-1,204–206 and UV irradiation stimulates ET-1 production by keratinocytes.204,205 ET-1 can also cooperate synergistically with other growth factors/cytokines to further influence melanocyte function.

ET-1 upregulates MC1R level and increases MC1R affinity for α-MSH.207,208 Similar to α-MSH, ET-1 displays photoprotective effects on melanocytes, enhancing thymine dimer repair, decreasing the level of UV-induced hydrogen peroxide, and inducing the level of antiapoptotic proteins.201,209

Steel Factor (SF)

Like other keratinocyte-derived factors, SF is induced by UV-irradiation, and in guinea pigs, anti-Kit antibodies block UV-induced tanning. SF can also act synergistically with other cytokines such as IL-3, IL-6, IL-7, IL-9, and granulocyte-macrophage-colony stimulating factor to regulate UV-induced melanogenesis and melanocyte survival.210,211

Inflammatory Mediators

Several inflammatory mediators can affect skin pigmentation. PGs—arachidonic acid-derived metabolites, and leukotrienes—lipid compounds related to PGs, both mediators of inflammatory responses, affect melanocyte function. Their level is elevated in sunburned skin212 and in a variety of inflammatory dermatoses, including atopic dermatitis213 (see Chapter 14) and psoriasis214 (see Chapter 18).

Human melanocytes express several PG receptors including the receptors for PGE2 and PGF2α.215,216 Indeed, PGF2α stimulates melanocyte dendrite formation and activates tyrosinase,215,216 and UV irradiation upregulates the level of PG receptors on melanocytes.215,216 Similarly, leukotrienes B4 and C4 increase melanin synthesis and stimulate melanocyte proliferation and motility.217 Interestingly, melanocytes also contribute to cutaneous inflammatory responses, as they synthesize and release IL-8 when stimulated by the proinflammatory cytokines IL-1 and TNF-α.218

Melanocytes also respond to histamine released by mast cells during cutaneous inflammation. Histamine binds H1 and H2 receptors to induce melanocyte dendricity and upregulate tyrosinase level.219,220 These effects are decreased when melanocytes are pretreated with the H2 receptor antagonist famotidine.220

Neurotrophins

Neurotrophins (NTs) are a family of molecules that enhance neuronal survival in the central and peripheral nervous systems. They include NGF,221 NT-3,222–224 NT-4,225 and brain-derived neurotrophic factor.226,227 Melanocytes express the low-affinity receptor common to all NTs, p75NTR,228 as well as the high-affinity receptors for NGF (TrkA) and NT3 (TrkC).229 Keratinocyte-derived NGF, whose expression is upregulated by UV irradiation, is chemotactic for melanocytes and induces their dendricity.230 Both NGF and NT-3, the latter expressed by dermal fibroblasts, increase melanocyte survival. Specifically, after UV irradiation, NGF supplementation increases the level of the antiapoptotic Bcl-2 protein, reducing melanocyte apoptotic cell death.231,232 Thus, in addition to other keratinocyte-derived cytokines, NGF may help preserve the population of cutaneous melanocytes that would otherwise be depleted by UV damage.

Basic Fibroblast Growth Factor

Basic fibroblast growth factor (bFGF), named for its ability to stimulate the growth of fibroblasts, was one of the first identified melanocyte mitogens.233,234 It is produced by keratinocytes, but lacks a secretory signal and hence is presumed to affect melanocytes through cell–cell contact. It binds tyrosine kinase transmembrane receptors to induce its mitogenic effect in the presence of cAMP elevating factors. Like other keratinocyte-derived cytokines, it is upregulated in response to UV irradiation.234,235 Keratinocyte growth factor, another member of the FGF family of proteins, has been shown to promote melanosome transfer from melanocytes to keratinocytes.192

Nitric Oxide

Nitric oxide (NO) is a diffusible free radical displaying pleiotropic bioregulatory effects in diverse cells and tissues.236,237 Melanocytes and keratinocytes produce NO in response to inflammatory cytokines,238–241 and NO production in keratinocytes is induced by UV irradiation.242 NO increases tyrosinase activity and melanogenesis242 and is thus an autocrine as well as paracrine molecule that affects melanocyte behavior in skin.

Catecholamines

Catecholamines are a group of signaling molecules, primarily functioning as neurotransmitters and as endocrine hormones.243 Catecholamines bind either α1-adrenergic receptors (AR) or β2-AR and can induce melanogenesis through PKC-β or cAMP-dependent pathways.169,244

Melanogenic Inhibitors

Numerous reports have suggested the existence of endogenous melanogenic inhibitors,245,246 but only few specific molecules have been identified. One group of inhibitors includes sphingolipids, a class of membrane-associated lipids247 that act as signal transducers. Sphingolipids were shown to decrease melanogenesis, at least in part by enhancing MITF degradation via ubiquitin-meditated pathways.248,249 Another melanogenic inhibitor, BMP-4, downregulates tyrosinase expression in melanocytes,250 also in part via its effects on MITF.251 Interestingly, physiologic doses of UV irradiation, a potent melanogenic stimulator, decrease the expression of BMP receptors on melanocytes,250 presumably eliminating its inhibition during UV-induced tanning. Mice that transgenically overexpress the physiologic BMP antagonist noggin have a darker coat color than wild-type mice and their hairs have a higher eumelanin to pheomelanin ratio.252

Growth factors, cytokines, hormones, and other ligands for receptors expressed on melanocytes exert their biologic effect by interacting with their specific cell surface receptors, generating a signaling cascade involving activation or inhibition of protein kinases and leading to distinct patterns of protein phosphorylation. Two types of kinases participate in cellular signaling: (1) serine/threonine kinases and (2) tyrosine kinases that by definition phosphorylate serine and/or threonine residues and tyrosine residues, respectively, on their specific target proteins. This section reviews the major signaling pathways that affect melanocyte behavior in skin.

cAMP/PKA-Dependent Pathway

cAMP, one of the first identified intracellular second messengers, plays a key role in diverse biological functions such as cellular metabolism, growth, and differentiation.253,254 It also mediates α-MSH effect255 and was one of the first recognized regulators of mammalian pigmentation256 The intracellular level of cAMP is upregulated by a membrane-associate enzyme called adenylate cyclase that is activated upon receptor–ligand interaction in receptors that are coupled to GTP-binding proteins like MC1R257 (eFigs. 72-8.1 and 72-12.1). cAMP is also elevated by reagents such as choleragen or isobutylmethylxanthine. Providing melanocytes with dibutyryl cAMP, a cAMP analog, increases the intracellular level of cAMP and induces signaling that leads to melanogenesis.258

cAMP-dependent protein kinase (PKA) mediates most of the biologic actions of cAMP.257 PKA is a serine/threonine kinase consisting of two regulatory subunits and two catalytic subunits.257 It exists in the cytosol in an inactive form and binding of cAMP to its regulatory subunits releases the catalytic subunits, activating the enzyme.257 PKA phosphorylates the cAMP responsive element-binding protein (CREB) that binds its DNA consensus sequence CRE in the MITF promoter to induce MITF transcription (eFig. 72-8.1). cAMP elevation also affects other target genes by increasing or decreasing their transcription259 (eFig. 72-12.1). In vitro, PKA effect can be antagonized by the inhibitor PKI that acts as a pseudosubstrate for the catalytic subunit of PKA and thus prevents it from phosphorylating its endogenous substrates.260

PKC-Dependent Pathway

PKC is a serine/threonine kinase involved in diverse cellular functions, including growth, transformation, and differentiation.261 PKC resides as an inactive enzyme in the cytoplasm, and it is activated by diacylglycerol (DAG), a component cleaved from the plasma membrane when cell surface receptors interact with their ligands. DAG can also be released from the membrane by UV irradiation (eFig. 72-12.1). DAG induces PKC translocation to membranes where the latter is activated261 to induce phosphorylation of serine/theonine residues on target proteins like tyrosinase. Phorbol esters mimic DAG action and initially activate PKC.261 However, within 24 hours the entire cellular reserves of PKC are depleted, and when melanocytes are treated with phorbol esters they can no longer signal through PKC.87

The critical role of PKC in melanogenesis was first suggested by the observation that addition of DAG, the endogenous activator of PKC, to cultured human melanocytes rapidly increased total melanin content,262 and this increase was blocked by a PKC inhibitor.262 Moreover, topical application of DAG to guinea pig skin increased epidermal melanin content.263

The expression of the 12 PKC isoforms varies among different tissues.86 Each isoform is thought to carry out a distinct biological function. Human melanocytes express PKC-α, -β, -ϵ, -δ, and -ς264,265 and the PKC-β isoform is specifically involved in regulating tyrosinase activity (see above). ET-1 and histamine also utilize the PKC-dependent pathway (in addition to cAMP-dependent pathway) to exert their regulatory effects on melanocyte function.266,267

Receptor Tyrosine Kinases

Melanocytes express several distinct tyrosine kinase receptors that bind BMP, bFGF, HGF, and c-Kit ligand. Receptor–ligand interaction activates an intracellular tyrosine kinase domain on the receptor, phosphorylating the receptor and subsequently activating a series of kinases called mitogen-activated protein (MAP) kinases, or other intracellular signaling molecules (eFig. 72-12.1). Through a chain reaction involving phosphorylation of proteins like MITF, the signals are transferred to the nucleus to activate or suppress the transcription of genes that participate in melanocyte proliferation, melanogenesis, and/or survival.

cGMP Pathway

β2- and α1-Adrenergic Receptors

Studies suggest that pathways that increase intracellular cAMP are also involved in regulation of melanogenesis. POMC-deficient mice (POMC-/−) that lack melanocortin ligands still display normal black coat color.269 Histological and electron paramagnetic resonance spectrometry of the hair follicles showed normal structure and eumelanin pigmentation.269 This study suggests that either MC1R has adequate basal activity to induce pigmentation or that pathways that do not involve melanocortin can also induce melanogenesis. Indeed, melanocytes express both β2-AR and α1-AR, respectively.270,244 α1-AR interacts with melanocyte derived-norepinephrine and increases the level of DAG169,244 inducing melanogenesis in a PKC-β-dependent pathway. Also, keratinocytes produce epinephrine, which then binds to β2-AR expressed on melanocytes and increases the level of cAMP, leading to melanin synthesis.244 Therefore, numerous pathways may act in tandem to regulate melanogenesis.

Tanning Response

Melanocyte survival, proliferation, and differentiated function are influenced by environmental factors, the most important of which is UV irradiation. UV irradiation induces tanning, the so-called facultative skin color, an increase above baseline or constitutive skin pigmentation that provides protection against future UV irradiation.271 Tanning is divided into immediate tanning and delayed tanning.

Immediate Tanning

Immediate tanning or immediate pigment darkening occurs within 5–10 minutes of exposure and fades within minutes to days depending on the UV dose and the complexion of the individual (Fig. 72-13B). As summarized in Table 72-1, immediate tanning does not provide photoprotection and does not increase epidermal melanin level.272 It is primarily produced by UVA irradiation, although visible light can also induce immediate tanning.273 Immediate tanning is only visible in darker individuals, and it is thought to represent melanosomal relocation from the perikaryon to melanocyte dendrites.274

Delayed Tanning

Delayed tanning, summarized in Table 72-1 and shown in Figure 72-13A, occurs within 3–4 days after UV exposure.271,272 UV is arbitrarily divided into UVC (100–280 nm), UVB (280–320 nm), and UVA (320–400 nm). The UVC portion of the spectrum is generally not present in terrestrial sunlight because it is absorbed by the atmospheric ozone layer. Delayed tanning is affected by both UVB and UVA. The action spectrum that produces delayed tanning is the same as for UV-induced erythema (sunburn), with UVB wavelengths far more effective than UVA.264 Especially in darker skinned individuals, suberythemogemic UV doses may be effective as well. Delayed tanning peaks between 10 days and 3–4 weeks depending on the absorbed UV dose and the individual's skin type, then fades gradually over a few weeks. Histologically, there are increased epidermal melanocytes, melanocyte dendriticity, and melanosome transfer to keratinocytes with greater melanization of individual melanosomes.272,274 Overall, total epidermal melanin is increased, providing additional photoprotection from UV irradiation.

Direct and Indirect Effects of UV Irradiation

UV irradiation affects melanization, melanocyte proliferation and survival both directly and indirectly through its effect on keratinocytes, inducing the synthesis and secretion of paracrine keratinocyte factors.

Direct Effects

UV irradiation triggers several biological reactions through its interaction with cellular chromophores that absorb photons. Photochemical reactions affect proliferation, survival, and the differentiated function of melanocytes. Most UVA effects are assumed to be the result of oxidative damage mediated through UVA absorption by cellular chromophores like melanin precursors that act as photosensitizers leading to the generation of ROS and free radicals.275 UVB irradiation is directly absorbed by cellular DNA, leading to the formation of DNA lesions, mainly cyclobutane dimers and pyrimidine (6–4) pyrimidone photoproducts.276 DNA damage repair systems are activated, at least in part through the tumor suppressor p53 protein. It has been shown that plasma membrane lipids are also affected by UV irradiation to release DAG,277 which then activates PKC-β to stimulate melanogenesis by activating tyrosinase (see above).

Indirect Effects

Key keratinocyte paracrine factors induced by UV irradiation and their effects on melanocytes are summarized in eTable 72-1.1. These factors can act alone and/or synergistically to modulate melanocyte function. Interestingly, UV irradiation also induces the level of TNF-α and IL-1, cytokines that inhibit melanogenesis, suggesting a fine-tuned epidermal equilibrium between melanogenic stimulators and inhibitors after UV irradiation, with the final outcome of increased melanogenesis and melanocyte proliferation.

The Role of DNA Damage in Melanogenesis

Interestingly, the action spectrum for tanning is virtually the same as that for the formation of thymine dimers,278,279 and UV-induced melanogenesis can be augmented in pigment cells by treatment with T4 endonuclease V,276 an enzyme that acts exclusively to enhance the repair of UV-induced DNA damage. Moreover, treatment of melanocytes with agents that act exclusively by damaging DNA, unlike UV that has multiple cellular targets, also stimulates melanogenesis.280

A central role for DNA damage and/or its repair in stimulating melanogenesis is further suggested by the fact that p53, a tumor-suppressor protein and transcription factor termed the Guardian of the Genome, when activated, upregulates the level of tyrosinase mRNA and protein, enhancing melanogenesis.281–284 Thus, tanning may be viewed as part of a p53-mediated DNA damage adaptive response that protects the skin during subsequent exposure to UV irradiation.

P53 in UV-Induced Melanogenesis

A central role for DNA damage and/or its repair in stimulating melanogenesis is further supported by the fact that the tumor-suppressor protein p53, when activated, upregulates the level of tyrosinase mRNA and protein, enhancing melanogenesis.281–285 In p53 knockout mice, it was observed that UV irradiation of the ears (that contain interfollicular melanocytes) minimally stimulates melanogenesis, compared to the far greater “tanning” response in p53 wild-type mice.282 These findings were expanded by Cui et al,286 who found a p53 consensus sequence in the POMC gene promoter, thus establishing a continuous signaling pathway from UV-induced DNA damage to the final tanning response. It was shown in mice keratinocytes that following UV irradiation p53 activation stimulates POMC transcription, thereby increasing the release of POMC-derived α-MSH, a key physiologic inducer of melanogenesis. Keratinocyte-derived α-MSH then stimulates MC1R on melanocytes, resulting in increased production of eumelanin.286

It has also been shown that p53 transcriptionally upregulates the hepatocyte nuclear factor-1α (HNF-1α) that is a transcription factor for tyrosinase.287 Thus, even in keratinocyte absence, UV directly activates p53 and HNF-1α in melanocytes to increase tyrosinase transcription. Furthermore, UV is known to increase H2O2 that activates p53.288 Thus, tanning may be viewed as part of a p53-mediated DNA damage adaptive response aimed at protecting the skin from subsequent UV irradiation.282,283,289

Epidermal melanocyte aging is affected by both genetic and environmental factors. With aging, there is a decrease in the density of epidermal melanocytes (number per unit area of skin surface), approximately 10% per decade.290 However, the number of DOPA-positive melanocytes is greater in chronically sun-exposed skin than in sun-protected skin,290 possibly due to melanocyte proliferation after sun exposure and/or UV-induced keratinocyte-derived paracrine factors. Melanocyte loss is especially notable in hair follicles with age, with total loss of melanocytes in approximately half of all scalp follicles by middle age.37 Hair graying (depigmentation) occurs over the entire body but is usually first noted on the scalp, perhaps because of the long anagen (growth) cycle and resulting requirement for melanocyte proliferation and sustained high level of melanogenesis.

In vitro melanocytes derived from older individuals show decreased proliferative capacity compared to those derived from younger individuals. Also, with aging in vitro, there is a general increase in the levels of total melanin as well as in the level differentiation-associated proteins such as MITF, TRP-1, and TRP-2291,292 and decrease in the level of proliferation-associated proteins such as cyclin D1 and E.291,292