HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

This Article Abstract Full Text (PDF) All Versions of this Article: fj.06-6649revv1 21/4/976    most recent Alert me when this article is cited Alert me if a correction is posted Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Costin, G.-E. Articles by Hearing, V. J. Search for Related Content PubMed PubMed Citation Articles by Costin, G.-E. Articles by Hearing, V. J.

Human skin pigmentation: melanocytes modulate skin color in response to stress

Gertrude-E. Costin^*,1 and Vincent J. Hearing

^* Global R&D, Avon Products, Inc., New Technology Department, Suffern, New York, USA; and

Laboratory of Cell Biology, National Cancer Institute, National Institutes of Health, Bethesda, Maryland, USA

¹Correspondence: Avon Products, Inc., New Technology Department, 1 Avon Pl., Suffern, NY 10901, USA. E-mail: gertrude.costin{at}avon.com

	ABSTRACT

TOP ABSTRACT INTRODUCTION SKIN STRUCTURE HYPERPIGMENTATION INDUCED BY... HYPERPIGMENTATION INDUCED BY... CONCLUSIONS REFERENCES

 
All organisms, from simple invertebrates to complex human beings, exist in different colors and patterns, which arise from the unique distribution of pigments throughout the body. Pigmentation is highly heritable, being regulated by genetic, environmental, and endocrine factors that modulate the amount, type, and distribution of melanins in the skin, hair, and eyes. In addition to its roles in camouflage, heat regulation, and cosmetic variation, melanin protects against UV radiation and thus is an important defense system in human skin against harmful factors. Being the largest organ of the body that is always under the influence of internal and external factors, the skin often reacts to those agents by modifying the constitutive pigmentation pattern. The focus of this review is to provide an updated overview of important physiological and biological factors that increase pigmentation and the mechanisms by which they do so. We consider endocrine factors that induce temporary (e.g., during pregnancy) or permanent (e.g., during aging) changes in skin color, environmental factors (e.g., UV), certain drugs, and chemical compounds, etc. Understanding the mechanisms by which different factors and compounds induce melanogenesis is of great interest pharmaceutically (as therapy for pigmentary diseases) and cosmeceutically (e.g., to design tanning products with potential to reduce skin cancer risk).—Costin, G-E., Hearing, V. J. Human skin pigmentation: melanocytes modulate skin color in response to stress.

Key Words: hyperpigmentation • UV radiation • aging • melanocyte photoprotection

	INTRODUCTION

TOP ABSTRACT INTRODUCTION SKIN STRUCTURE HYPERPIGMENTATION INDUCED BY... HYPERPIGMENTATION INDUCED BY... CONCLUSIONS REFERENCES

 
HUMAN SKIN EXISTS IN A WIDE range of different colors and gradations, ranging from white to brown to black. This is due to the presence of a chemically inert and stable pigment known as melanin, which is produced deep inside the skin but is displayed as a mosaic at the surface of the body. Melanin is therefore responsible for the most striking polymorphic traits of humans and for the most obvious and thoroughly discussed aspect of human geographical variability: skin color. Besides its role in defining ethnicity, melanin plays an essential role in defending the body against harmful UV rays and other environmental challenges. Minor changes in the physiological status of the human body or exposure to harmful external factors can affect pigmentation patterns either in transitory (such as in pregnancy) or permanent (e.g., age spots) manners.

To understand pigmentation of the skin and the factors that affect it, one must focus on the intricate cellular and molecular interactions between melanocytes and keratinocytes, which together compose the epidermal melanin unit. All of the other types of cells distributed within different layers of the skin and the intracellular signaling pathways often overlapping and involving cross-talking must be considered also.

In this review, we provide an update of current knowledge regarding the effects of endocrine and environmental factors on skin pigmentation and the mechanisms by which they function. This review emphasizes that the skin reacts to stress through all its cellular and molecular components, which form a complicated, sophisticated, and highly sensitive signaling network.

	SKIN STRUCTURE

TOP ABSTRACT INTRODUCTION SKIN STRUCTURE HYPERPIGMENTATION INDUCED BY... HYPERPIGMENTATION INDUCED BY... CONCLUSIONS REFERENCES

 
The skin plays an extremely important role, providing a vast physical barrier against mechanical, chemical, and microbial factors that may affect the physiological status of the body (1) . In addition to those functions, the skin also acts as an immune network and, through its pigments, provides a unique defense system against UV radiation (UV-R) (2) . Thus, melanocytes transfer melanosomes through their dendrites to keratinocytes, where they form the melanin caps that reduce UV-induced DNA damage in human epidermis. The skin’s layers are represented by the epidermis, the dermis (whose structure will be discussed in more detail below), and the hypodermis, the latter consisting of fatty tissue that connects the dermis to underlying skeletal components (Fig. 1 A).

View larger version (71K): [in this window] [in a new window]   Figure 1. Structure of the skin. A) The different layers and components. Collagen, elastin, and other components of the extracellular matrix (ECM) of the dermis are not represented. B) Layers of the epidermis. The stratum lucidum usually present in thick epidermis such as palms and soles is not included; it is normally located between the stratum granulosum and stratum corneum and consists of flattened cells with no nuclei. Merkel cells (stratum basale) and Langerhans’ cells (stratum spinosum) are not represented either. Panel B based on and modified from A. H. Robins (1991) Biological Perspectives on Human Pigmentation, pp. 2, Cambridge University Press, Cambridge, UK.

Epidermis
The epidermis is an external, stratified epithelium devoid of blood or nerve supplies of 5–100 µm thickness (which can reach 600 µm on palms and soles) (3) . It is composed of several distinct cell populations; keratinocytes and melanocytes are the main constituents, of which the first comprise 95% of the epidermis and are arranged in four layers, as follows (Fig. 1B ).

Stratum basale (also known as the stratum germinativum) is a single layer of cells attached to a noncellular basement membrane that separates the epidermis from the dermis. The stratum basale consists mostly of basal keratinocytes, which have stem cell-like properties, and at least two different types of neural crest-derived cells: Merkel cells (neuroendocrine cells responsible for the transmission of touch sensation through the cutaneous nerves) and melanocytes.

Stratum spinosum contains irregular polyhedral keratinocytes with some limited capacity for cell division. Also found here are the bone marrow-derived sentinel cells of the immune system called Langerhans’ cells, which represent the antigen-presenting cells of the skin and play a vital role in immunological reactions such as allergic contact dermatitis.

Stratum granulosum contains flattened, polyhedral nondividing keratinocytes producing granules of a protein called keratinohyalin. These granules increase in size and number as the cell nuclei gradually degenerate and the cells die. These cells flatten as dividing cells underneath them progressively push them toward the skin surface.

Stratum corneum contains nonviable, but biochemically active cells called corneocytes. The keratinocytes continue to differentiate as they move from the basal layer to the stratum corneum, the result being cornified cells that contain abundant keratin and lack cytoplasmic organelles. It is these cornified cells that provide a barrier against the physical and chemical agents in the environment that may adversely affect the body. More specifically, this epidermal barrier functions to reduce transepidermal water loss from within and to prevent invasion by infectious agents and noxious substances from without (4) .

Dermis
The dermis is a 2 to 4 mm-thick layer of connective tissue and fibroblasts that houses the neural, vascular, lymphatic, and secretory apparatus of the skin (Fig. 1A ). The main cell type, fibroblasts, is required for synthesis and degradation of the extracellular matrix (ECM) (1) . This matrix is a complex structure composed of highly organized collagen, elastic, and reticular fibers. The dermis also hosts multifunctional cells of the immune system such as macrophages and mast cells, the latter being able to trigger allergic reactions by secreting bioactive mediators such as histamine. Structures within the dermis include: 1) Excretory and secretory glands (sebaceous, eccrine, and apocrine). Sebaceous glands secrete triglyceride and cholesterol-rich sebum that lubricate the skin and keep it supple and waterproof. They are often associated with hair shafts. 2) Hair follicles and nails: in addition to generating the hair shaft, the hair follicle provides a protective niche to several stem cell populations in the skin, including keratinocyte stem cells, melanocyte stem cells, a population of epidermal neural crest stem cells, and the dermal stem cell compartment, known as the dermal papilla (5 , 6) . These stem cells are required most visibly during wound healing. 3) Sensory nerve receptors of Merkel and Meissner’s corpuscles (for touch), Pacinian corpuscles (for pressure), and Ruffini corpuscles (mechano-receptors).

In the dermis, collagen provides the skin with tensile strength and tissue integrity whereas elastin provides elasticity and resiliency. Besides collagen and elastic fibers, the dermis contains the extrafibrillar matrix, which is extracellular and composed of a complex mixture of proteoglycans, glycoproteins, glycosaminoglycans, water, and hyaluronic acid. The most significant glycosaminoglycans, which bind to proteins to form the proteoglycans of the skin, are chondroitin sulfate, dermatan sulfate, keratin sulfate, heparan sulfate, and heparin. The most important proteoglycans of the skin are versican, which is involved in assuring the tightness of the skin, and perlecan, which is found in basement membranes. Glycoproteins, such as laminins, matrilins, fibronectin, fibronectin, tenascins, etc., are involved in cell adhesion, cell migration, and cell-cell communication, which are extremely important processes taking place in the skin.

Melanocytes, melanosomes, and melanin
Melanin biosynthesis is a complex pathway that appears in highly specialized cells, called melanocytes, within membrane-bound organelles referred to as melanosomes (7) . Melanosomes are transferred via dendrites to surrounding keratinocytes, where they play a critical role in photoprotection. The anatomical relationship between keratinocytes and melanocytes is known as "the epidermal melanin unit" and it has been estimated that each melanocyte is in contact with 40 keratinocytes in the basal and suprabasal layers (8) .

Several important steps must occur for the proper synthesis and distribution of melanin, as follows (9) .

1. The development of melanocyte precursor cells (melanoblasts) and their migration from the neural crest to peripheral sites
Prospective melanocytes, known as melanoblasts, derive from the neural crest beginning in the second month of human embryonic life and migrate throughout the mesenchyme of the developing embryo. They reach specific target sites, mainly the dermis, epidermis, and hair follicles, the uveal tract of the eye, the stria vasculare, the vestibular organ and the endolymphatic sac of the ear, and leptomeninges of the brain. In humans, this migration process takes place between the 10th and the 12th wk of development for the dermis and 2 wk later for the epidermis (1) .

The survival and migration of neural crest-derived cells during embryogenesis is highly dependent on interactions between specific receptors on the cell surface and their extracellular ligands. For example, steel factor, formerly known as mast cell growth factor, KIT ligand, or stem cell factor (SCF), binds the KIT receptor on melanocytes and melanoblasts. Mutations in the KIT gene decrease the ability of the KIT receptor to be activated by the steel factor and are responsible for at least one type of human piebaldism (10) . See http://albinismdb.med.umn.edu for other examples of genes that regulate pigmentation and, when mutant, are involved in pigmentary disorders.

2. Differentiation of melanoblasts into melanocytes
Once melanoblasts have reached their final destinations, they differentiate into melanocytes, which at about the sixth month of fetal life are already established at epidermal-dermal junction sites (1) .

3. Survival and proliferation of melanocytes
Melanocytes have been identified within fetal epidermis as early as 50 days of gestation. Dermal melanocytes decrease in number during gestation and virtually disappear by birth, whereas epidermal melanocytes established at the epidermal-dermal junction continue to proliferate and start to produce melanin.

4. Formation of melanosomes and production of melanins
Once established in situ, melanocytes start producing melanosomes, highly organized elliptic membrane-bound organelles in which melanin synthesis takes place. They can be detected using electron microscopy (EM) as early as during the fourth month of gestation.

Melanosomes are typically divided into four maturation stages (I–IV) determined by their structure and the quantity, quality, and arrangement of the melanin produced (Fig. 2 ) (11 , 12) . Nascent melanosomes are assembled in the perinuclear region near the Golgi stacks, receiving all enzymatic and structural proteins required for melanogenesis. Stage I melanosomes are spherical vacuoles lacking tyrosinase (TYR) activity (the main enzyme involved in melanogenesis) and have no internal structural components. However, TYR can be detected in the Golgi vesicles, and it has been shown that it is subsequently trafficked to stage II melanosomes. At this point, the presence and correct processing of Pmel17, an important melanosomal structural protein, determine the transformation of stage I melanosomes to elongated, fibrillar organelles known as stage II melanosomes (12 , 13) ; they contain tyrosinase and exhibit minimal deposition of melanin. After this, melanin synthesis starts and the pigment is uniformly deposited on the internal fibrils, at which time the melanosomes are termed as stage III. Their last developmental stage (IV) is detected in highly pigmented melanocytes; these melanosomes are either elliptical or ellipsoidal, electron-opaque due to complete melanization, and have minimal TYR activity. The developmental stages detailed above refer mainly to eu-melanosomes (containing black-brown pigments); however, they are quite similar to pheo-melanosomes (containing yellow-reddish melanin), the only difference being that the latter remain round and are not fibrillar during maturation.

View larger version (38K): [in this window] [in a new window]   Figure 2. Schematic of a melanocyte: the engine for melanin production. The four developmental stages of melanosomes are shown as they move toward the periphery of the cell within the dendrites. The melanin biosynthesis pathway is displayed in the inset showing the main enzymes involved. Asterisks in different shades of gray represent different stages of glycosylation for the melanogenic proteins as they are transported through the ER and Golgi apparatus.

Within melanosomes, at least three enzymes are absolutely required to synthesize different types of melanin. While tyrosinase is responsible for the critical steps of melanogenesis (including the rate-limiting initial step of tyrosine hydroxylation), tyrosinase-related protein 1 (TYRP1) and DOPAchrome tautomerase (DCT) are further involved in modifying the melanin into different types. Besides these, melanosomes contain other melanocyte-specific proteins that have structural functions (e.g., Pmel17, as mentioned above) or probably are involved in regulating the pH within melanosomes, such as P protein- or membrane-associated transporter protein (MATP), or that play as yet unclear roles, such as the melanoma antigen recognized by T cells 1 (MART1) or oculocutaneous albinism-1 (OA-1) protein (14) .

TYR (monophenol, 3,4-ß-dihydroxyphenylalanine oxygen oxidoreductase, EC 1.14.18.1) is a single chain type I membrane glycoprotein catalyzing the hydroxylation of tyrosine to ß-3,4-dihydroxyphenylalanine (DOPA) (which is the initial rate-limiting step in melanogenesis) and the subsequent oxidation of DOPA to DOPAquinone. TYR, TYRP1, and DCT share numerous structural similarities and follow quite similar biosynthetic, processing, and trafficking pathways (15) . Their maturation is assisted by chaperones, calnexin being the most important one due to its involvement in the correct folding of tyrosinase (16 17 18) . The subsequent metabolism of DOPA and its derivatives by various melanocyte-specific enzymes, including TYRP1 and DCT, results in the synthesis of eumelanin, a black-brown pigment. Briefly, 5,6-dihydroxyindole (DHI) melanins are generated from DOPAquinone after several steps of decarboxylation, oxidation, and polymerization. However, in the presence of DCT, the carboxylic acid group of 5,6-dihydroxyindole-2-carboxylic acid (DHICA) is retained when derived from DOPAchrome, and therefore the so-called DHICA melanins are produced. The synthesis of pheomelanin involves the production of cysteinyldopa conjugates from DOPAquinone after the production of DOPA from tyrosine. TYRP1 is important for the correct trafficking of tyrosinase to melanosomes (19) , and DCT also seems to be involved in the detoxification processes (20) taking place within melanosomes.

Melanins are polymorphous and multifunctional biopolymers that include eumelanin, pheomelanin, mixed melanins (a combination of the two), and neuromelanin. Mammalian melanocytes produce two chemically distinct types of melanin pigments: black-brown eumelanin and yellow-reddish pheomelanin (21) . Although they contain a common arrangement of repeating units linked by carbon-carbon bonds, melanin pigments differ from each other with respect to their chemical, structural, and physical properties. Eumelanin is a highly heterogeneous polymer consisting of DHI and DHICA units in reduced or oxidized states, as detailed above; pheomelanin consists mainly of sulfur-containing benzothiazine derivatives (22) . Due to their chemical structure, both eumelanin and pheomelanin are involved in binding to cations, anions, drugs, and chemicals, etc., and therefore play an important protective role within melanocytes (23) . Neuromelanin, which is produced in dopaminergic neurons of the human substantia nigra, can also chelate redox active metals (Cu, Mn, Cr) and toxic metals (Cd, Hg, Pb), and thus protects against their ability to promote neurodegeneration (24) .

Given their complexity, melanosomes can be used as a model to study organelle biogenesis, protein trafficking and processing, organelle movement, and cell-cell interactions (like those occurring during melanin transfer between melanocytes and keratinocytes) (25) . Therefore, even minor changes in the cellular environment affect melanosomes and pigmentation. Numerous intrinsic and extrinsic factors, including body distribution, ethnicity/gender differences, variable hormone-responsiveness, genetic defects, hair cycle-dependent changes, age, UV-R, climate/season, toxin, pollutants, chemical exposure and infestations, are responsible for a whole range of responses in melanosome structure and distribution under different types of stress.

Cutaneous pigmentation is the outcome of two important events: the synthesis of melanin by melanocytes and the transfer of melanosomes to surrounding keratinocytes (26) . Although the number of melanocytes in human skin of all types is essentially constant, the number, size, and manner in which melanosomes are distributed within keratinocytes vary. The melanin content of human melanocytes is heterogeneous not only between different skin types but also between different sites of the skin from the same individual. This heterogeneity is highly regulated by gene expression, which controls the overall activity and expression of melanosomal proteins within individual melanocytes (27) . It has been shown that melanocytes with a low melanin content synthesize TYR more slowly and degrade it more quickly than melanocytes with a higher melanin content and TYR activity (28) . In general, highly pigmented skin contains numerous single large melanosomal particles (0.5–0.8 mm in diameter), which are ellipsoidal and intensely melanotic (stage IV). Lighter pigmentation is associated with smaller (0.3–0.5 mm in diameter) and less dense melanosomes (stages II and III), which are clustered in membrane-bound groups (29) . These distinct patterns of melanosome type and distribution are present at birth and are not determined by external factors (such as sun exposure). They are responsible for the wide variety of skin complexions (Fig. 3 ).

View larger version (41K): [in this window] [in a new window]   Figure 3. Human pigmentation—the main skin types: African-American, Asian, Caucasian, and Hispanic (left to right).

Epidermal melanin unit and the involvement of keratinocytes in melanin production
The epidermal melanin unit is a functional and structural complex within the epidermis consisting of two cell types: melanocytes and keratinocytes. The variation in skin color among various races is determined mainly by the number, melanin content, and distribution of melanosomes produced and transferred by each melanocyte to a cluster of keratinocytes surrounding it (30) . Once in keratinocytes, the melanin granules accumulate above the nuclei and absorb harmful UV-R before it can reach the nucleus and damage the DNA. When melanin is produced and distributed properly in the skin, dividing cells are protected at least in part from mutations that might otherwise be caused by harmful UV (31) . The melanocyte-keratinocyte complex responds quickly to a wide range of environmental stimuli, often in paracrine and/or autocrine manners. Thus, melanocytes respond to UV-R, agouti signaling protein, melanocyte-stimulating hormone (MSH), endothelins, growth factors, cytokines, etc. After UV-R exposure, melanocytes increase their expression of proopiomelanocortin (POMC, the precursor of MSH) and its receptor melanocortin 1 receptor (MC1-R), TYR and TYRP1, protein kinase C (PKC), and other signaling factors (32 , 33) . On the other hand, it is known that UV stimulates the production of endothelin-1 (ET-1) and POMC by keratinocytes and that those factors can then act in a paracrine manner to stimulate melanocyte function (34 , 35) . In addition to keratinocytes, fibroblasts, and possibly other cells in the skin produce cytokines, growth factors, and inflammatory mediators that can increase melanin production and/or stimulate melanin transfer to keratinocytes by melanocytes. Melanocyte growth factors affect not only the growth and pigmentation of melanocytes but also their shape, dendricity, adhesion to matrix proteins, and mobility.

-MSH, ACTH, basic fibroblast growth factor (bFGF), nerve growth factor (NGF), endothelins, granulocyte-macrophage colony-stimulating factor (GM-CSF), steel factor, leukemia inhibitory factor (LIF), and hepatocyte growth factor (HGF) are keratinocyte-derived factors that are thought to be involved in the regulation of the proliferation and/or differentiation of melanocytes (36) , some acting through receptor-mediated signaling pathways (Fig. 4 ). It has been shown that in human epidermis, -MSH (32 , 37) and ACTH (32 , 37 , 38) are produced in and released by keratinocytes and are involved in regulating melanogenesis and/or melanocyte dendrite formation. -MSH and ACTH bind to a melanocyte-specific receptor, MC1-R (39) , which activates adenylate cyclase through G-protein, which then elevates cAMP from adenosine triphosphate (40) . Cyclic AMP exerts its effect in part through protein kinase A (PKA) (41) , which phosphorylates and activates the cAMP response element binding protein (CREB) that binds to the cAMP response element (CRE) present in the M promoter of the microphthalmia-associated transcription factor (MITF) gene (42 , 43) . The increase in MITF-M expression induces the up-regulation of TYR, TYRP1, and DCT (42 , 43) , which leads to melanin synthesis.

View larger version (23K): [in this window] [in a new window]   Figure 4. Scheme of signaling pathways within the epidermal melanin unit and mechanisms by which keratinocyte-derived factors act on human melanocyte proliferation and differentiation. Based on and modified from T. Hirobe (2005) Role of keratinocyte-derived factors involved in regulating the proliferation and differentiation of mammalian epidermal melanocytes. Pigment Cell Res., vol. 18, pp. 2–12.

Keratinocytes also produce and release NGF, which is involved in regulating the melanogenesis and/or dendritogenesis of melanocytes (44) . Its expression is up-regulated by UV-R, suggesting yet another paracrine influence of keratinocytes on melanocytes with possible relevance to the tanning response. It has been shown that normal human melanocytes express the NGF receptor (45) and also high-affinity receptors for NGF (TRK-A) and NT-3 (TRK-C) (44) .

ET-1 is a 2l amino acid peptide with vasoactive properties first isolated from endothelial cells and later found to be synthesized and secreted by keratinocytes as well (46 47 48) , particularly after exposure to UV-R (46 47 48) . The overall effect of ET-1 is the increase of melanocyte dendricity and the enhancement of melanocyte migration and melanization (48) . Binding of ET-1 to its G protein-coupled receptor (ETBR) on melanocytes activates a cascade of signaling pathways, resulting in mobilization of intracellular calcium, activation of PKC, elevation of cAMP levels, and activation of mitogen-activated protein kinase (MAPK) (49 , 50) . UV-R stimulates keratinocytes to produce ET-1 and also induces interleukin-1 (IL-1) production in these cells. IL-l is known to induce ET-1 in keratinocytes in an autocrine manner. Therefore, it has been suggested that these intracellular events in keratinocytes lead to increased TYR mRNA, protein, and enzymatic activity in neighboring melanocytes as well as to an increase in melanocyte number (51) .

Prostaglandin (PG) E2 and PGF2 are known to be produced and released from human keratinocytes by the stimulation of proteinase-activated receptor 2 (PAR-2). PGE2 and PGF2 stimulate the dendritogenesis of human epidermal melanocytes in culture (52) through EP1, EP3, and FP receptors. Their influence on melanocyte dendricity has been suggested to be cAMP-independent and might be mediated through phospholipase C (PLC) (52) .

bFGF (53) and SCF (54) are also expressed by keratinocytes. Those secreted factors are involved in regulating the proliferation and melanogenesis/dendritogenesis of human epidermal melanocytes in normal skin (38 , 44 , 55) and/or in UV-A (56) /UV-B (32 , 46 , 48 , 53 , 54) -irradiated skin.

HGF binds to its specific receptor, c-Met (57) , activates MAPK, and elicits the up-regulation of proteins required for melanocyte proliferation (58 , 59) .

GM-CSF binds to its specific receptor, GM-CSFR (60) , activates the signal transducer and activator of transcription (STAT-1, STAT-3, and STAT-5) (61 , 62) or MAPK (63) , and induces the up-regulation of proteins required for the proliferation of melanocytes as well as of TYR, TYRP1, and DCT.

The epidermis has a complex network that secretes as well as responds to autocrine and paracrine cytokines produced by keratinocytes and melanocytes, respectively. Human melanocyte proliferation requires the cross-talking of several signaling pathways including the cAMP/PKA, PKC, and tyrosine kinase pathways. Therefore, the mechanisms by which various factors increase skin pigmentation are closely inter-related and will be further discussed below.

There are numerous internal and external stresses that affect human skin pigmentation. The list is fairly long, so for this review we decided to focus on just a couple of the more common stresses whose mechanisms of action are known to some extent or are currently under investigation and whose use may affect the discovery of new approaches to reduce hyperpigmentation. We consider external factors (UV-R: tanning and photoaging; drugs, chemicals, etc.) and internal factors (hormonal influences; inflammation: postinflammatory hyperpigmentation).

	HYPERPIGMENTATION INDUCED BY EXTERNAL FACTORS

TOP ABSTRACT INTRODUCTION SKIN STRUCTURE HYPERPIGMENTATION INDUCED BY... HYPERPIGMENTATION INDUCED BY... CONCLUSIONS REFERENCES

 
UV influence on human pigmentation
The skin responds to UV-R exposure by developing two defensive barriers: thickening of the stratum corneum and the elaboration of a melanin filter in cells of the epidermis. The palms and soles are the regions with the thickest stratum corneum, and they are exceptionally resistant to UV damage. The keratins and proteins within the stratum corneum act mainly by scattering and absorbing the UV. UV-R sets in action an integrated mechanism for the formation and delivery of melanin within melanosomes from melanocytes to keratinocytes. This mechanism is probably triggered by keratinocytes, which respond to UV-R with bursts of mitoses and with increased production of ET-1 and POMC, thus creating a new demand for melanosomes. The mitotic rate of basal keratinocytes increases a day after UV exposure, reaches a maximum 2 days later, and maintains this level for 1 wk. It then declines, and the skin regains its original thickness after 1–2 months if there has been no subsequent exposure. After UV-R, the epidermal melanin unit responds with increased levels of TYR activity, increased synthesis of melanosomes, and higher rates of melanosome transfer to keratinocytes to meet the new demand for melanosomes created by the proliferation of keratinocytes (64) .

UV is part of the electromagnetic spectrum and it lies between the visible and X-ray regions. Only 5–10% of the total radiant energy received at the surface of the Earth from the sun is UV, and the rest is divided between the visible (40%) and the infrared (50%) (65) . According to wavelengths, UV radiation is divided in UV-A (320–400 nm), UV-B (280–320 nm), and UV-C (200–280 nm); the latter is normally screened by the ozone layer and does not reach the Earth’s surface, like most wavelengths <280 nm.

UV-A passes through most glass in automobiles, offices, and windows whereas UV-B is blocked by window glass. UV-A also penetrates deep into the dermis; it is estimated that 19–50% of the solar UV-A can reach the depth of melanocytes, whereas only 9–14% of solar UV-B reaches these cells. Therefore, UV-A stimulates melanin pigmentation, but the resultant tan appears to be transient and less protective against UV-induced injury than tans generated after UV-B exposure. Although the amount of UV-A reaching the Earth’s surface is several orders of magnitude greater than the amount of UV-B, UV-A has 1000-fold less erythema-producing effects than UV-B (65) . However, Garland et al. were the first to hypothesize that UV-A could cause melanoma in humans. The potential carcinogenic effect of UV-A has also been demonstrated in cultured human melanocytes (66) . The authors suggested that endogenous pigments and/or melanin-related molecules seem to enhance DNA breakage after UV-A irradiation. UV-A must first react with endogenous photosensitizers (flavins, porphorins, melanins), which in turn generate reactive oxygen species (ROS), which finally causes single-strand breaks or photoadducts.

UV-B is responsible for causing the sunburn reaction within the skin and is absorbed mainly by the epidermis and upper dermis. Like UV-A, UV-B stimulates the production of melanin, which constitutes the basis for tanning. UV-B has great potential to induce erythema, and therefore its influence on the skin has been thoroughly investigated in vitro and in vivo (64) . The UV-B portion of the spectrum can promote skin cancer, especially if the exposure has been repeated and prolonged. However, recent studies have shown that the use of narrow-band UV-B (NB-UVB 311 nm) is a better choice for phototherapy than frequently used psoralen and UV-A therapy (PUVA), which can cause undesirable side effects, including cutaneous cancers (67 , 68 69 70) .

The only known beneficial effect of UV-B is the stimulation of vitamin D synthesis in the epidermis. Vitamin D promotes the absorption of calcium from the intestine and ensures the proper mineralization of bones. However, exposure of just a small area of the body to a small amount of UV-B (5% of that required for erythema) is all that is needed for adequate synthesis of the vitamin D in the skin (64) .

One role of melanin in the skin is to neutralize the ROS generated by a variety of factors, including UV-B (23) , therefore functioning like a natural sunscreen. Until recently it was thought that the higher the melanin content, the less chance of DNA damage resulting from UV-R exposure. In a recent study, the effects of melanin on UV responses in different racial/ethnic groups were investigated for the first time. Despite the general public assumption that dark skin types are UV resistant and therefore not adversely affected by UV, this study showed that even the darkest UV-resistant skin types accumulated significant DNA damage at levels 1 minimal erythema dose (MED) (71) . The authors demonstrated that even very low UV exposures cause measurable damage to DNA in all skin types, although it was obvious that the most severe DNA damage was in lightly pigmented skin.

The influence of UV on human pigmentation will be detailed further from the perspective of tanning as well as photoaging as a perfect example of factors sharing intracellular pathways with slightly different end results on the skin.

Tanning response to UV-R
In humans, an increase of skin pigmentation over the basal constitutive level is called tanning, and this is physiologically stimulated by UV-R. UV-induced skin darkening involves an increase in the number of melanocytes as well as stimulation of melanin synthesis and melanocyte dendricity, a crucial morphological feature required for melanin transfer to keratinocytes.

The tanning response has been shown to have two distinct phases, termed immediate pigment darkening and delayed tanning. Both have strong genetic determinants and are generally more pronounced in individuals with dark baseline (constitutive) pigmentation (72) .

Immediate tanning is a quick but transient brownish tan that follows the exposure of skin to UV-A or visible light. It begins immediately after exposure, reaches a maximum within 1–2 h, then fades between 3 and 24 h after exposure (72) . Certain ultrastructural changes have been observed in melanocytes during immediate tanning: the appearance of thick filaments and microtubules and the translocation of melanosomes from the perinuclear area to the dendritic processes but no actual increase in the size or number of melanosomes. It thus seems possible that the immediate tanning reaction is based on the photoxidation of preexisting melanin, melanin precursors, or even of other epidermal constituents and/or their redistribution in the epidermis.

Delayed tanning gives rise to a durable tan induced by repeated exposure mainly to UV-B, but also to UV-A or to visible light. It is a gradual process in which the skin starts darkening 48–72 h after irradiation, reaches a maximum 3 wk after exposure, and the skin does not return to its original melanin content until 8–10 months later (72) . Delayed tanning is dependent on both qualitative and quantitative changes within melanocytes, which enlarge in size, increase their dendricity, and develop a diffuse distribution of thick filaments in their cell bodies. Ribosomes, ER, and Golgi apparatus are more prominent, reflecting an increase in the synthesis of TYR and melanosomes in all developmental stages, in their melanization, and in the number that are transferred to keratinocytes. Therefore, delayed tanning is due to an increase in melanocyte numbers and melanogenesis.

Like all photobiologic responses, tanning requires direct interaction of UV photons with molecular targets (chromophores) in the skin. The major cellular chromophores that absorb in the UV-B range are nucleic acids (mainly the pyrimidine and purine bases) and proteins (mostly tryptophan and tyrosine); cytotoxicity, mutagenesis, and new protein synthesis are all initiated by photoreactions involving DNA (73) . However, some responses, such as activation of membrane enzymes and induction of early response genes, involve non-nuclear chromophores. Other biomolecules that absorb UV-B include the reduced form of NAD, quinones, flavins, and other heterocyclic cofactors such as tetrahydrobiopterin. In skin, UV-A- and UV-B-absorbing molecules are present in addition to the above-mentioned cellular chromophores and include 7-dehydrocholesterol, urocanic acid, and melanin (73) .

The products formed in DNA after absorbing UV-B radiation have been studied extensively because of their major role in the mechanisms underlying UV carcinogenesis. Two major types of bipyrimidine photoproducts are created following UV-R: cyclobutylpyrimidine dimers (CPDs) and (6–4) photoproducts (Fig. 5 ; ref. 73 ). Cells are equipped with a complicated and highly regulated apparatus for repairing the DNA damage produced in this manner; however, if that does not occur with high efficiency and fidelity, the cells accumulate mutations that can eventually lead to skin cancer.

View larger version (37K): [in this window] [in a new window]   Figure 5. Mechanisms involved in the (hyper)pigmentation induced by UV-R. The tanning response is determined by a complex set of regulatory processes involving direct effects of UV-R on melanocytes and indirect effects through the release of keratinocyte-derived factors.

UV-R also causes peroxidation of lipids in cellular membranes, leading to generation of ROS, which may stimulate melanocytes to produce excess melanin (74) . Usually, only lipids containing two or more conjugated double bonds in their structure absorb UV-B and thus liberate arachidonic acid, which is subsequently metabolized to various species of PGs and leukotrienes, generates previtamin D₃ from 7-dehydrocholesterol with subsequent processing to various photoproducts and the biologically active 1, 25-dihydroxy-vitamin D₃, and releases diacylglycerol (DAG), which in turn activates PKC, among other possible roles in signal transduction (75 ; Fig. 5 ). It was observed that addition of DAG to cultured human melanocytes increases their melanin content severalfold within 24 h (76) , and subsequent work demonstrated that UV-R acts synergistically with DAG to enhance melanogenesis (77) . These data suggest that DAG may be a physiological mediator of the tanning response in UV-irradiated skin.

Direct melanogenic effects of UV on melanocytes might also involve the production of NO, which is considered a major intra- and intercellular messenger molecule. NO elicits its effects through the activation of a soluble guanylate cyclase, leading to an increase in intracellular cGMP content and the activation of cGMP-dependent protein kinase. Furthermore, it has been shown that UV-R increases both NO and cGMP production, suggesting they are both required for UV-B-induced melanogenesis (Fig. 5) .

Melanin plays a major photoprotective role in human skin by absorbing, scattering, photo-oxidizing, and scavenging free radicals and acting as a pseudo-dismutase to minimize the toxic effects of ROS and to prevent damage to DNA, proteins, and cell membrane lipids (78) . It is known that UV-A produces harmful oxygen species such as O^.₂-, ^.OH, and ¹O₂ and that melanin interacts with them, thus protecting the skin against the damage that could occur (78 , 79) .

Overall, UV-R acts on human skin by increasing transcription of the TYR gene and the function of MC1-R on melanocytes; it also increases expression of POMC and its derivative peptides by keratinocytes and cells within the dermis and the release of DAG from the plasma membrane, which in turn activates PKC. UV-R also activates the NO/cGMP pathway and the production of growth factors and ET-1 by keratinocytes, and induces a SOS response to UV-induced DNA damage, which sets into action the cellular DNA repair mechanisms.

Photoaging: the UV-R contribution to the appearance of solar lentigines (age spots)
Aging of the skin is a complex process induced by both chronologic and environmental factors (mainly UV-R). Changes in the skin associated with chronologic aging include dryness, increased fragility, decreased epidermal and dermal thickness (both influenced by circulating levels of estrogens and potentially by androgens), fragmentation of elastic fibers (most likely influenced by circulating levels of estrogens), decreased sebum production, and the number and function of apocrine glands (probably influenced by levels of circulating androgens). Changes in the skin due to photoaging include wrinkles and furrows, solar lentigines (SL), mottled pigmentation, actinic keratoses, basal cell and squamous cell carcinomas, subsets of melanomas, etc. (80) .

Chronologic aging leads to a decrease in the turnover rate of the epidermis as well as the flattening of rete ridge patterns, which results in a decreased surface area of the basement membrane. This is clinically described as an increased fragility of the skin. Chronologic aging of the dermis is reflected by decreases in collagen content and thickness. Additional age-related changes include decreases in the vascularity of the dermis as well as in the number of functional fibroblasts. Clinically, this results in increased times required for wound healing. Dermal elastic fibers are also affected by chronologic aging and undergo fragmentation and granular disintegration by age 70 (mainly those in the upper dermis) (80 , 81) .

In photoaging (sun-induced skin aging), the degree of damage to the epidermis depends on the cumulative dose of sun exposure as well as on the amount of protection provided by its pigmentation. In the dermis, the primary effects of photoaging involve degeneration of collagen and the deposition of abnormal elastotic material. The clinical manifestations of these changes include wrinkles and furrows of the face, as mentioned above; this is in distinction to the laxity (sagging) and fine wrinkling of non-sun-exposed areas that are due to chronological aging of the dermis (81) .

From the viewpoint of pigmentation, aging results in a decline in functional melanocytes in both the skin and hair. Various studies have indicated that the number of functioning DOPA-positive melanocytes in nonexposed human skin decreases with age by 8–20% of the surviving population each decade. However, in UV-irradiated skin there are approximately twice as many melanocytes as in unexposed areas, but there is still a comparable decrease in melanocytes with age. It is surprising that, unlike hair color, there is no loss of skin pigmentation with age. In fact, the chronically sun-exposed skin of an older person is usually more pigmented than that of a younger subject of similar complexion despite the lower melanocyte density in the former. This paradox has been explained by the greater functional activity in older melanocytes after many years of cumulative sun exposure (82) . An interesting finding recently reported (83) is that DCT is not expressed by melanocytes of human hair compared with human skin. This could potentially contribute to the premature loss of melanin production by functional melanocytes in human hair with age due to added cytotoxic stress of melanogenesis in the absence of DCT.

Given the theme of this review, we outline the current understanding of mechanisms responsible for the appearance of SL, also known as senile lentigo, sun-, liver-, or age spots. Solar lentigines are circumscribed, pigmented macules, which are usually light brown, but vary in degree of color to jet black. SL are typically found on UV-exposed areas of the body (the face, dorsum of the hand, extensor forearm, and upper back) (Fig. 6 ). They can range anywhere in size from <1 mm up to a few centimeters in diameter and, in areas of severely sun-damaged skin, may coalesce into even larger lesions (84) (Table 1 ).

View larger version (67K): [in this window] [in a new window]   Figure 6. Pictures of skin at different ages showing differences in pigmentation pattern.

The molecular mechanisms responsible for skin hyperpigmentation in SL have recently been elucidated by the group of Imokawa. They found a 2-fold increase of TYR-positive cells per length of the dermal/epidermal interface in SL lesions compared with unaffected skin (85) . That group also showed that there is a molecular regulatory network between melanocytes/keratinocytes and melanocytes/dermal fibroblasts in which ET-1 and SCF are key regulators in the development of hyperpigmentation in SL (86 , 87) .

Exposure to UV-R induces an increase in the production of ET-1 by keratinocytes, and its secretion therefore stimulates melanocytes to produce melanin. As described above, the actions of endothelins on melanocytes are initiated by the binding of ET-1 to its receptor (ETBR), followed by sequential signaling processes involving PKC and MAPK. On the other hand, SCF (also produced by keratinocytes) binds to the c-KIT receptor on melanocytes, thus mediating dimerization, activation of its intrinsic tyrosine kinase activity, and autophosphorylation (88) . The activated c-KIT receptor then phosphorylates various substrates and associates with various signaling molecules including phosphatidylinositol 3-kinase, the Shc and Grb2 adaptor proteins, and the guanine nucleotide exchange factor, SOS, all of which lead to activation of the Ras-MAPK pathway (89) (Fig. 7 ).

View larger version (18K): [in this window] [in a new window]   Figure 7. Potential mechanisms involved in the appearance of senile lentigines. The dotted arrows suggest there are intermediates involved in those pathways (see sSCF, mSCF, and HGF signaling); solid arrows describe the fully detailed pathway (see ET-1 signaling).

The potential of keratinocytes located in SL lesional epidermis to produce ET-1 is significantly higher than in perilesional normal controls (46) , and there is an accentuated expression of ETBR transcripts as well (85) . The increased production and localization of ET-1 was paralleled by increased amounts of TYR in melanocytes. These findings suggest that stimulation of the epidermal ET cascade, especially with respect to the expression of ET-1 and ETBR, plays an important role in the mechanism involved in the hyperpigmentation of LS. The ET-1-inducible cytokine, tumor necrosis factor , is consistently up-regulated in the SL lesional epidermis, suggesting its involvement in this hyperpigmentation disorder through the induction of ET-1 (85) .

Imokawa et al. (90) also showed that SL lesional epidermis expresses increased levels of SCF mRNA transcripts and protein compared with nonlesional controls. In epidermal keratinocytes, SCF is expressed as a membrane-bound form (mSCF), not in a secretory or soluble cytokine form such as ET-1 (54) , even after UV-B exposure (54) . In contrast, dermal fibroblasts can secrete soluble SCF (sSCF), probably because of the action of proteolytic enzymes capable of cleaving mSCF to release sSCF (91) . Expression of sSCF could not be detected in SL lesional epidermis, suggesting that the increased production of mSCF plays an essential role in stimulating the proliferation and melanogenesis of melanocytes, leading to epidermal hyperpigmentation in SL (Fig. 7) .

Therefore, the mechanism currently proposed for the appearance of SL involves the stimulation of two epidermal cascades, consisting of ET-1/ETBR and SCF/c-kit, and the cross-talk between those two after the UV exposure. However, other cascades in the skin may also contribute to the hyperpigmentation seen in SL; these may be discovered as work in this field progresses.

The only data regarding the analysis of SL in the dermal compartment of the skin became available just recently (92) . That study showed that the number of melanophages is increased in SL compared with unaffected skin in the same subject. These melanophages were identified as FXIIIa⁺ dermal dendrocytes; they seem to be the main cell type that uptakes melanin (most likely from the epidermis, where it is increased in SL) by a mechanism that remains to be determined. It was also reported recently that dickkopf (DKK1), secreted by fibroblasts in the dermis, plays an important role in regulating melanocytes above them (93) ; work is needed to resolve the role of that factor, if any, in hyperpigmentation of the skin.

The action of drugs, chemicals, etc., on human skin pigmentation
Numerous common drugs can stimulate human skin hyperpigmentation such as certain antibiotics (sulfonamides and tetracyclines), diuretics, nonsteroidal antiinflammatory drugs, pain relievers, and some psychoactive medications.

The use of oral contraceptives has been associated with the development of discoloration of the cheeks, forehead, and nose (94) similar to chloasma (further detailed below). Microscopic examination of the epidermis revealed increased melanogenesis and the presence of enlarged melanocytes.

Certain antiepileptic agents (mainly hydantoins) may also cause skin hyperpigmentation (95) . Their long-term use induces a brownish coloration of the face and neck, similar to chloasma of pregnancy. It has been shown that melanin concentration is particularly increased in females; Caucasians especially seem to be more affected (96) .

It is already known that chloroquine has an affinity for melanin and causes skin hyperpigmentation. Different studies have detected melanin in the dermis of patients undergoing chloroquine treatment (97) .

Levodopa, often used to treat Parkinson’s disease, also induces hyperpigmentation of the skin (64) . DOPA is normally transformed into melanin within melanosomes; therefore, DOPA therapy (applied as levodopa treatment) may possibly enhance melanin biosynthesis, perhaps even by extracellular oxidation, although the literature lacks strong evidence to support this hypothesis.

Heavy metals can also elicit hyperpigmentation, which can arise after the extensive use of drugs containing arsenic, bismuth, gold, or silver (98) . The metals are believed to act by binding, and thereby inactivating, sulfydryl compounds in the skin that normally inhibit TYR activity. Removal of this inhibition stimulates melanogenesis. Mercury products inactivate TYR probably by replacing the essential copper in the enzymatic site of that protein.

Some chemotherapy agents also can cause hyperpigmentation, the most common ones being cyclophosphamide, 5-fluorouracil, doxorubicin, daunorubicin, and bleomycin. Their mechanisms of action are currently unknown but may involve direct toxicity, stimulation of melanocytes, and/or inflammation.

	HYPERPIGMENTATION INDUCED BY INTERNAL FACTORS

TOP ABSTRACT INTRODUCTION SKIN STRUCTURE HYPERPIGMENTATION INDUCED BY... HYPERPIGMENTATION INDUCED BY... CONCLUSIONS REFERENCES

 
Hormonal influence on human skin pigmentation
Hyperpigmentation is sometimes seen during pregnancy and this condition is called melasma, chloasma, or mask of pregnancy; it occurs mainly on the cheeks, upper lip, chin, and forehead. It is characterized by a symmetrical hypermelanosis with an irregular coloration, ranging from light brown to gray and dark brown. Although melasma is usually associated with pregnancy, multiple other factors can contribute to its development including UV exposure, hormone therapy, estrogen-containing oral contraceptives, genetic influences, certain cosmetics, endocrine or hepatic dysfunction, and selected antiepileptic drugs (Table 1) . Of the environmental sources, UV-R is obviously the most influential (84 , 99) .

Recent studies have shown that the areas of hyperpigmentation seen in melasma exhibit increased deposition of melanin in the epidermis and dermis (100 , 101) . No increase in the number of melanocytes in those areas was noted, but the melanocytes were larger, more dendritic, and showed increased melanogenesis, producing especially eumelanin (101) . That study confirmed an increased number of melanosomes in keratinocytes, melanocytes, and dendrites in lesional skin compared with nonlesional skin. During pregnancy (especially in the third trimester), elevated levels of estrogen, progesterone, and MSH have often been found in association with melasma (102 , 103) . TYR activity increases and cellular proliferation is reduced after treatment of melanocytes in culture with ß-estradiol (104) . Sex steroids increase transcription of genes encoding melanogenic enzymes in normal human melanocytes, especially those for DCT and TYR (105) . These results are consistent with the significant increases in melanin synthesis and TYR activity reported for normal human melanocytes under similar conditions in culture (106) .

Since melanocytes contain both cytosolic and nuclear estrogen receptors (107) , melanocytes in patients with melasma may be inherently more sensitive to the stimulatory effects of estrogens and possibly to other sex steroid hormones. Recent studies suggest that estrogens exert their effect in skin through the same molecular pathways used in other nonreproductive tissues (Fig. 8 ) (108) . There is evidence that 17ß-estradiol can use both signaling pathways (either genomic or nongenomic) in epidermal keratinocytes (109) .

View larger version (25K): [in this window] [in a new window]   Figure 8. Mechanism(s) of estrogen action. Estrogens can activate both genomic intracellular receptors (ER and ERß) (classical, genomic pathway) and nongenomic cell surface receptors (nonclassical, nongenomic pathway). In the classical mode, the hormone enters the cell by passive diffusion and interacts with its intracellular receptors. Upon ligation, the receptor forms homo- or heterodimers; after that, the steroid receptor complex tightly associates with specific consensus DNA sequences called estrogen response elements (ERE), which consist of 15 base pair inverted palindromes located within the regulatory region of the target gene. The nonclassical pathway works faster and depends on the ability of estrogen to interact with either membrane estrogen receptors or nonsteroid hormone receptors (e.g., G-protein-coupled receptor, GPR30). The nonclassical pathway acts via different conventional second messengers like cAMP and activates mitogen-activated protein kinases, which regulate transcription of specific genes. Based on and modified from M. J. Thornton (2005) Estrogen functions in skin and skin appendages. Expert Opin. Ther. Targets, vol. 9, pp. 617–629, and S. Verdier-Sevrain, F. Bonte, B. Gilchrest, (2006) Biology of estrogens in skin: implications for skin aging. Exp. Dermatol., vol. 15, pp. 83–94.

It is known that estrogens improve skin moisture and also increase its thickness and collagen content. Therefore, estrogen plays a key role in skin aging homeostasis given the fact that skin appearance declines quickly in the postmenopausal years. Despite the knowledge that estrogens have such important effects on skin, their cellular and molecular mechanisms of action are still poorly understood and their influence on pigmentation is still far from clear.

Estrogens mediate their activity by interaction and activation of specific intracellular receptor proteins, the estrogen receptors (ERs) and ß, which often coexist as homo- or heterodimers. ER and ERß are distinct proteins encoded by separate genes located on different chromosomes (110) ; they share 60% homology, bind 17ß-estradiol with nearly equal affinity, and exhibit a similar binding profile for a large number of natural and synthetic ligands (111) . ER is expressed in both male and female reproductive tissues, bone, the cardiovascular system, and regions of the brain (112) . ERß is also expressed in both male and female reproductive tissues and in many nonreproductive tissues such as the lung, bladder, thymus, pituitary, hypothalamus, heart, kidney, adrenals, and skin (113 114 115) . Estrogens have significant effects on different cell types important in skin physiology, including keratinocytes, fibroblasts, and melanocytes. It has been shown that keratinocyte mitotic activity increases in the epidermis of women in response to estrogens (116) . Furthermore, the stimulation of proliferation and DNA synthesis of human epidermal keratinocytes by estrogens has been demonstrated in vitro (117 118 119) , and it has been shown that human keratinocytes in culture have high affinity estrogen binding sites (119) . In addition to increasing the proliferation of keratinocytes, estradiol accelerates the secretion of GM-CSF by 3-fold in cultured human keratinocytes (120) . GM-CSF is secreted by keratinocytes at the wound edge to promote the migration of endothelial cells and keratinocytes in order to advance neovascularization and re-epithelialisation (121) . Treatment with estrogen accelerates cutaneous wound healing in both sexes by decreasing wound size and by increasing collagen and fibronectin levels (122) .

Dermal fibroblasts have an important modulatory role in remodeling the ECM during wound repair. Primary cultures of human dermal fibroblasts from female skin have recently been shown to express both ER and ERß (123) .

Studies using double immunofluorescence staining have shown that epidermal melanocytes in human scalp express both ER and ERß in situ (124) ; ligand binding studies confirmed that cultured human normal epidermal melanocytes contain estrogen receptors (107) . Recently, Im et al. reported the presence of ER in normal human melanocytes using immunocytochemistry and RT-polymerase chain reaction (RT-PCR) (125) . Examination of the effects of estrogen treatment on TYR activity has revealed a stimulation of this melanogenic enzyme (104 , 105) . It was recently demonstrated that androgens modulate TYR activity via regulation of cAMP, a key regulator of skin pigmentation (126) . The sum of these studies emphasizes the importance of both sex hormones in regulating skin pigmentation.

Details of the mechanisms by which sex hormones influence melanogenesis remain to be elucidated, but this is a good example of the concurrent involvement of different cellular factors (secondary messengers, growth factors, etc.) in multiple molecular pathways within the skin eliciting hyperpigmentation as a final result.

Postinflammatory hyperpigmentation of the skin
Postinflammatory hyperpigmentation is manifested by discrete, hyperpigmented macules with hazy, feathered margins, which may involve the epidermis and/or dermis. This usually develops after resolution of inflammatory skin eruptions like acne, contact dermatitis, or atopic dermatitis. Postinflammatory hyperpigmentation is more common in patients with darker skin and, at the cellular level, is characterized by a normal number of melanocytes that have increased melanin production (Table 1) .

Arachidonate-derived chemical mediators, especially leukotrienes such as LTC₄ and LTD₄, and thromboxanes such as TXB₂ may be responsible for the induction of postinflammatory hyperpigmentation of the skin because they can stimulate normal human melanocytes in vitro. These cells become swollen and more dendritic with increased amounts of immunoreactive TYR when cultured for 2 days with LTC₄, LTD₄, or TXB₂. Such morphological changes are thought to be required for the transfer of melanosomes to surrounding keratinocytes. Those effects were stronger than that elicited by PGE₂, which, together with PGE₁ and PGD₂, are known to be important endogenous regulators of inflammatory diseases in the skin and to stimulate mammalian pigment cells in vitro (127) and in vivo (128) . Despite the common frequency of skin hyperpigmentation following inflammation, the mechanisms responsible for melanin synthesis have not yet been completely clarified, but some data have became available recently, as follows.

PGs are synthesized from arachidonic acid by cyclooxygenase and represent a group of potent lipid hormones that activate multiple signaling pathways, which in turn regulate cellular growth, differentiation, and apoptosis. In the skin, PGs (especially PGE₂, PGF2, and small quantities of prostacyclin) are produced (129) and rapidly released by keratinocytes after UV-R (130 , 131) . They are chronically present in inflammatory skin lesions and are involved in wound healing (132) .

It has been shown that melanocytes express several types of receptors for PGs; EP1-EP4 are receptors for PGE₂ (EP3 and EP4 are high affinity whereas EP1 and EP2 are low affinity) (133) ; only EP1 and EP3 are expressed by human melanocytes. The FP receptor mediates the effects of PGF2 (134 , 135) and is a heterotrimeric G-coupled protein receptor that signals through Gq (136 , 137) . The FP receptor couples to Gq and activates phospholipase C-induced phosphoinositide turnover, intracellular Ca²⁺ mobilization, and MAPK/PKC activation (137) . Scott et al. showed that UV-R stimulates production of PGF2 by melanocytes, which in turn stimulates the activity and expression of TYR, suggesting that PGF2 could act as an autocrine factor for melanocyte differentiation (138) .

On the other hand, PAR-2 is an important factor regulating skin pigmentation because its activation in keratinocytes stimulates their uptake of melanosomes through phagocytosis. It has been reported that activation of PAR-2 in keratinocytes stimulates the release of PGE₂ and PGF2, which act as paracrine factors that stimulate melanocyte dendricity (52) . Melanocyte dendrite formation has been linked to the cAMP-dependent activation of Rac and the inhibition of Rho (139 140 141) . However, recent studies demonstrated that neither PGE₂ nor PGF2 stimulates cAMP in melanocytes, thus demonstrating that these PGs stimulate dendrite formation in a cAMP-independent manner (52) . These data suggest that PAR-2 mediates cutaneous pigmentation through regulation of melanosome uptake and production of PGs, which act as paracrine factors to stimulate melanocyte dendricity.

Secretory phospholipases comprise a large family of Ca²⁺-dependent enzymes that liberate arachidonic acid, which is also a precursor of lysophospholipids. The predominant secretory phospholipase (sPL) expressed by keratinocytes is group X secretory phospholipase A2 (sPLA2-X), which liberates large amounts of arachidonic acid and lysophospholipid lysophosphatidylcholine (LPC) from membranes. They are released during inflammation and their expression in the skin is increased by UV-R (142 , 143) . Recent studies have shown that low levels of sPLA2-X stimulate both TYR activity and the dendricity of human melanocytes in culture. These cells have been shown to express the PLA2 receptor (PLA2R) and two G-protein-coupled receptors for LPC (G2A and GPR119) (144) . G2A and GPR119 couple to several G-proteins (including Gs) that regulate TYR activity and dendricity (141 , 145) .

In conclusion, recent studies help explain the mechanisms involved in hyperpigmentation of the skin after inflammation. All factors and pathways described above interact within the skin; the final result is an increase of TYR activity and melanocyte dendricity, which promotes the production of melanin and its distribution to keratinocytes.

Different factors responsible for increasing human skin pigmentation do so via various intracellular pathways, some of which are common and some distinct, as detailed in this review. Table 2 summarizes some of the internal or external stresses and the secondary messengers and effectors that are involved.

Given the complexi