SEARCH

SEARCH BY CITATION

Keywords:

  • keratinocyte;
  • melanin;
  • melanocyte;
  • melanosome;
  • melanin transfer;
  • PAR-2;
  • pigment transfer;
  • pigmentation synapse

Abstract

  1. Top of page
  2. Abstract
  3. Modes of Transfer Revisited
  4. Molecular Players in Melanin Transfer
  5. Melanin Transfer and Disease
  6. Conclusion
  7. Acknowledgments
  8. References

Skin pigmentation is accomplished by production of melanin in specialized membrane-bound organelles termed melanosomes and by transfer of these organelles from melanocytes to surrounding keratinocytes. The mechanism by which these cells transfer melanin is yet unknown. A central role has been established for the protease-activated receptor-2 of the keratinocyte which effectuates melanin transfer via phagocytosis. What exactly is being phagocytosed – naked melanin, melanosomes or melanocytic cell parts – remains to be defined. Analogy of melanocytes to neuronal cells and cells of the haemopoietic lineage suggests exocytosis of melanosomes and subsequent phagocytosis of naked melanin. Otherwise, microscopy studies demonstrate cytophagocytosis of melanocytic dendrites. Other plausible mechanisms are transfer via melanosome-containing vesicles shed by the melanocyte or transfer via fusion of keratinocyte and melanocyte plasma membranes with formation of tunnelling nanotubes. Molecules involved in transfer are being identified. Transfer is influenced by the interactions of lectins and glycoproteins and, probably, by the action of E-cadherin, SNAREs, Rab and Rho GTPases. Further clues as to what mechanism and molecular machinery will arise with the identification of the function of specific genes which are mutated in diseases that affect transfer.

Melanocytes are neural crest-derived cells that synthesize and store the melanin pigment in unique membrane-bound organelles termed melanosomes. Melanocytes of the skin and hair transfer their melanin to surrounding keratinocytes. With the help of its dendritic projections, each epidermal melanocyte provides melanin to approximately 36 keratinocytes (1). After transfer, melanin is transported to the apical face of the keratinocyte nucleus, where it protects the genetic material against UV damage. This symbiotic system is called the epidermal melanin unit (2). Transfer of melanin in the hair follicle from mature melanocytes residing at the hair bulb to cortical and medullary keratinocytes is presumed to involve the same mechanisms as in the epidermal melanin unit (3).

Transport of melanosomes inside the pigment cell has been extensively studied for the past years. Melanosomes are transported between the cell centre and the cell periphery along microtubules via the action of the motor proteins kinesin and dynein (4–7). Beneath the plasma membrane, melanosomes undergo short-range movements along the sublemmal actin network via association with the motor protein myosinVa, which attaches to melanosomes through interaction with melanophilin and Rab27a (8–10). Mutation in one of the molecules of this tripartite actin-tethering complex leads to Griscelli syndrome characterized by perinuclear accumulation of melanosomes and hypopigmentation of skin and hair (see below).

It has appeared difficult to unravel the events from attachment of melanosomes to the sublemmal actin network to the point where melanin arrives at the cytoplasm of keratinocytes ready for transport to the supranuclear region. For half a century, different hypotheses have been clung to: release of melanin in the intercellular space by exocytosis, cytophagocytosis and fusion of plasma membranes with tunnel formation. Recent research has postulated phagocytosis by the keratinocyte via activation of the protease-activated receptor-2 as a necessary step. Other molecules with an established or probable role in transfer will be discussed as well as diseases that present pigmentation abnormalities due to defective transfer, as they form an important tool for identification of new players.

Modes of Transfer Revisited

  1. Top of page
  2. Abstract
  3. Modes of Transfer Revisited
  4. Molecular Players in Melanin Transfer
  5. Melanin Transfer and Disease
  6. Conclusion
  7. Acknowledgments
  8. References

Exocytosis

Regulated exocytosis is a process in which the membranes of cytoplasmic organelles fuse with the plasma membrane in response to stimulation. Various types of regulated secretory exocytosis exist – the exocytosis of synaptic vesicles and dense core vesicles at the presynaptic compartment of neuronal cells; the exocytosis of these vesicles (all over the plasma membrane) by neuroendocrine and endocrine cells; exocytosis of secretory granules at the apical plasma membrane of exocrine cells and the exocytosis by haemopoietic cells of various types of secretory organelles (11).

According to the exocytosis theory, melanin transfer is accomplished by fusion of the melanosomal membrane with the melanocyte plasma membrane, resulting in extracellular melanin and ensuing phagocytosis of released melanin by a neighbouring keratinocyte (Figure 1B). This hypothesis has been suggested by electron microscopic observations of human skin and hair follicles, depicting ‘naked melanin’ in the intercellular space and the enfolding of these pigment granules by keratinocyte pseudopods or clathrin-coated pits (12,13) (Figure 2A). Melanocytes discharge melanin in the extracellular space in vitro and do so increasingly when stimulated with α-melanocyte-stimulating hormone or with the soluble domain of the β-amyloid precursor protein (14,15). The process involved is considered to be exocytosis, as on electron microscopy discharged melanin is not surrounded by membrane and moreover, melanocytes express molecules with a well-known role in regulated exocytosis as SNAREs and Rab GTPases. SNAREs comprise three conserved families of membrane-associated proteins (the synaptobrevin/VAMP, syntaxin and SNAP-25 families) that act late in the events of membrane fusion. They associate into core complexes, and usually SNAP25 and syntaxin on the plasma membrane bind VAMP on the vesicle membrane. Different SNAREs have been identified in melanosome-enriched fractions: SNAP23, SNAP25, VAMP2, syntaxin4 and syntaxin6 (16,17). Immunoprecipitation shows association of VAMP2 and SNAP23, but not syntaxin4. Possibly, VAMP2 on melanosomes interacts with SNAP23 and an as yet to be identified syntaxin on the melanocyte plasma membrane to achieve fusion. Another family of proteins that acts in membrane fusion, particularly in tethering and docking of membranes before actual fusion, is Rab GTPases. The Rab3 proteins consisting of Rab3a-d are the central Rabs in regulated exocytosis. Rab3a is expressed in melanocytes and downregulation of its expression is effected by UV irradiation (16,18). Interestingly, downregulation of Rab3 in other cell types has proved to stimulate regulated exocytosis (19,20). Rab27a is also involved in regulated exocytosis of various types of organelles (21). As mentioned before, this molecule has an established role in melanosome transport at the cell periphery before actual transfer occurs. Synaptotagmin-like protein2-a (Slp2-a) is a new Rab27a effector which has been discovered in melanocytes (22). It links Rab27a with phosphatidylserine, thereby attaching melanosomes to the plasma membrane. This suggests that Rab27a has a role in docking melanosomes at the plasma membrane, which is an essential step in exocytosis. Interestingly, Slp2-a is structurally homologous to synaptotagmin, the putative regulator of exocytosis at the neuronal synapse.

image

Figure 1. Different modes of melanin transfer.(A)Cytophagocytosis: a melanocytic dendrite is pinched off and phagocytosed, leading to a phagolysosome from which melanin granules disperse throughout the cytoplasm. (B)Exocytosis: melanin is externalized by fusion of the melanosomal membrane with the plasma membrane and is then taken up by endocytosis or phagocytosis. (C)Fusion: plasma membranes of both cells merge creating a channel which allows passage of melanosomes. (D)Membrane vesicles: melanosomes are shed in vesicles which either fuse with the keratinocyte plasma membrane or are ingested by phagocytosis. M: melanocyte, K: keratinocyte.

image

Figure 2. Transfer at the ultrastructural level.(A)Melanin, which is not surrounded by a membrane (arrow), is seen in the intercellular space of co-cultures of human melanocytes (M) and keratinocytes (K). Scale bar = 2 µm. (B)A cross-section of a dendrite (arrow) surrounded by keratinocyte cytoplasm. It is not clear whether the dendrite is still attached to the melanocyte or whether it has been pinched off and is now residing in a phagolysosome. Scale bar = 1.5 µm. (C)A keratinocyte enfolds part of a melanocyte with pseudopods. The dashed line follows the keratinocyte plasma membrane. Scale bar = 2 µm. (D)Image of a dendrite (arrow) which is surrounded by keratinocyte cytoplasm but is still attached to the melanocyte cell body. Scale bar = 3 µm. (E)Filopodia arise from the dendrite tip and contact a neighbouring keratinocyte, possibly they form intercellular bridges (arrowheads). Scale bar = 2 µm. (F)Melanosomes surrounded by a membrane (arrowheads) are seen in the vicinity of a melanocyte, they could represent filopodia as well as shed vesicles. Scale bar = 3 µm.

Further support for the exocytosis hypothesis was provided by the finding that melanin in the keratinocyte cytoplasm presents in the same way, be it exogenously administered melanin or melanin supplied by melanocytes (14,23–26). Small melanin granules form aggregates after ingestion, while larger granules are dispersed singly in the keratinocyte cytoplasm. The distribution pattern seems to depend on size: phagocytosis of small or large latex beads also results in aggregates and singly dispersed beads, respectively. Aggregates of melanin granules are considered as indicators of cytophagocytosis of a whole dendrite tip with dispersion occurring in a later stage (see below). These studies demonstrate that melanin granules aggregate after being ingested as single granules and suggest that aggregates represent a final stage in the life cycle of melanin in keratinocytes.

Lastly, melanocytes are closely related to both neuronal and haemopoietic cells: to the former because of their neural crest origin and to the latter because melanosomes belong to the family of secretory lysosomes (specialized organelles that display substantial homology to lysosomes) (27). The relation between secretory lysosomes shows in genetic diseases affecting the biogenesis, transport and/or secretion of these organelles. For example, mutation of Rab27a in Griscelli syndrome impairs the exocytosis of secretory granules in cytotoxic T lymphocytes, resulting in immunodeficiency (28). As both synaptic vesicles and secretory lysosomes undergo regulated exocytosis on stimulation, it would be tempting to speculate that melanin transfer occurs by similar mechanisms.

Cytophagocytosis

Phagocytosis is defined as the cellular engulfment of particles with a diameter of more than 0.5 µm. In mammals, phagocytosis is essential in triggering host defences against invading pathogens, as well as in the elimination of damaged, senescent and apoptotic cells. It is usually accomplished by ‘professional phagocytes’, such as macrophages, dendritic cells and granulocytes. These are highly phagocytic and mobile cells capable of infiltrating a wide variety of tissues. ‘Non-professional or amateur phagocytes’ are also capable of phagocytosis but at much slower kinetics. In addition, they are less mobile resident cell types and are limited in the range of particles that they can take up (29). Keratinocytes belong to the latter group, their phagocytic nature has been shown in vitro (23) as well as in vivo (24).

Cytophagocytosis stands for phagocytosis of a viable cell or an intact part of a viable cell. The cytophagocytosis hypothesis of melanin transfer describes phagocytosis of an intact melanocytic cell part, namely the tip of a dendrite, by the keratinocyte. At the first stage, the melanocyte extends its dendrite towards and makes contact with a surrounding keratinocyte. The keratinocyte reacts with extensive membrane ruffling and engulfing of the dendrite tip with villus-like cytoplasmic projections. In the second stage, the dendrite tip is being squeezed and pinched off resulting in the formation of a cytoplasmic poach filled with melanosomes. In the third stage, a phagolysosome is formed by fusion of lysosomes, degradation of the melanocyte membranes and cytoplasmic constituents takes place and, meanwhile, the phagolysosome is transported to the supranuclear region. In the fourth and last stage, the phagolysosome disintegrates into smaller vesicles containing a single melanin granule or aggregates of melanin granules, which are then dispersed over the cytoplasm (30) (Figure 1A). The hypothesis of cytophagocytosis has basically been supported by electron microscopy (13,31–33) and time-lapse video microscopy studies (30,34–36). Some limitations should be borne in mind when considering these studies, as all of them stem from a time when less advanced techniques were used. Standard electron microscopy, for example, has the limitation of being static, and it produces two-dimensional images. A cross-section of a dendrite surrounded by keratinocyte cytoplasm could as well depict an enfolded dendrite that is still attached to the melanocyte as a phagolysosome containing a dendrite that has been pinched off, implicating the possibility that a similar phagolysosome does not exist. Nonetheless, this image, together with the image of a dendrite enfolded with the cytoplasmic projections or pseudopods of a keratinocyte, forms the most important evidence in support of the cytophagic theory (30) (Figure 2B–D). As for time-lapse light microscopy, advances in the field enable acquisition in optimal conditions (temperature and pH) and result in high-resolution pictures with far better membrane discrimination (37). Despite the optimized incubation and acquisition techniques, different authors describe hours of recording without observing interruption of a dendrite or phagocytosis of a pinched-off tip (14,38) (own observation, personal communication with Wu X).

Fusion of plasma membranes

The melanocyte plasma membrane fuses with the keratinocyte plasma membrane, resulting in a pore or a channel that connects the cytoplasm of both cells and through which melanosomes are transported (Figure 1C). The fusion mode of pigment transfer has been suggested in pigmented basal cell carcinoma and in the skin of black guinea pig ear (39,40). More recently, transfer was assumed to occur by fusion of a filopodium with the keratinocyte membrane allowing passage of melanosomes (38). Filopodia extend from the dendrite tips and cell body of melanocytes, adhere to the surface of neighbouring keratinocytes and allow transport of melanosomes towards the keratinocyte membrane. In some instances, transfer was seen via these protrusions, although definite proof of membrane fusion could not be given. The main limitations are the similar optical properties of melanocyte and keratinocyte membranes, which make it difficult to distinguish fusion via light microscopy. Conversely, electron micrographs of co-cultures demonstrate thin projections directly connecting the melanocyte and keratinocyte cytoplasm (Figure 2E).

If true membrane fusion occurs, the structures described by Scott et al. could be considered as tunnelling nanotubes (38). A recent article describes a method of cell–cell communication based on nanotubes that could network various cultured cells and function as channels for organelle transport (41). Filopodia contact and fuse with neighbouring cells, resulting in a tubular structure with a diameter of 50–200 nm directly connecting cell cytoplasms. The tube is composed of actin filaments and admits unidirectional transport of organelles and plasma membrane molecules as opposed to soluble cytoplasmic molecules. Similarly, interconnecting channels, with a diameter of 50–95 nm, can be seen between cytotoxic T lymphocytes or natural killer cells and their respective targets on disassembly of the immunological synapse. They are held responsible for membrane transfer that occurs from target cells to cytotoxic T lymphocytes and natural killer cells, as well as from antigen-presenting cells to B cells (42–44). Tunnelling nanotubes are found in different cell types, suggesting that they present a general tool of intercellular communication.

In neuronal growth cones, filopodia sample the environment and search for guidance cues that allow the growing axon to find its appropriate target. Once the target has been found, the initial tip-like contact flattens and tethers with formation of an early synaptic zone (45). Here, filopodia function in the early stage of synapse formation; they do not end in cell–cell fusion. More research is required to determine whether melanocytic filopodia are conduits for melanin transfer or whether they are rather used as vanguards for dendrite extension and melanocyte–keratinocyte adhesion.

Transfer via membrane vesicles

Shedding of melanosome-containing membrane vesicles by the melanocyte followed by phagocytosis of these vesicles by the keratinocyte or fusion with the keratinocyte plasma membrane is usually not considered as a possible mode of pigment transfer, although it has been described that pieces of membrane can travel from cell to cell (Figures 1D and 2F). Proteins and lipids destined for transfer are concentrated on the plasma membrane, subsequent pinching off results in the formation of an extracellular vesicle, which travels to distant cells.

Two studies suggest the shedding of melanosome-containing membrane vesicles as a model of melanin transfer. Flow cytometry analysis of a human melanoma cell line reveals vesicles that can be distinguished by their small size and fluorescent properties upon neoglycoprotein binding (46). It seems that melanoma cells actively shed these vesicles in culture and that the shedding involves segregation of membrane lectins so that 6-phospho-β-D-galactose-specific receptors are selectively concentrated on the vesicle membrane. Vesicles adhere to keratinocytes and this can be partially inhibited by addition of neoglycoproteins, suggesting a role for carbohydrates in this interaction. As shown by electron microscopy, the vesicles are finally taken up by keratinocytes, thereby delivering melanin. Another study describes transfer of melanin from melanophores to fibroblasts (47). Melanin is surrounded by a double membrane and is transferred to distinct groups of recipient cells, some of which are located at a distance from the melanophore. This suggests that melanophores release melanin by shedding of vesicles that are subsequently recognized by fibroblasts through specific interactions.

It has been established that all living cells shed membrane fragments called microparticles or microvesicles (0.05–1 µm). These vesicles are shed upon the induction of cell stress including cell activation and apoptosis. In this manner, they report cellular activation and tissue degeneration and can be considered as true vectors of information exchange between cells (48). The possibility that melanocytes use this ubiquitous mode of material transfer for delivery of melanosomes should be considered.

In brief, none of the hypotheses presented thus far can be considered conclusive. Given the importance of PAR-2 in the process (see below), phagocytosis seems to be a necessary step. It could involve phagocytosis of either melanin set free by exocytosis or a dendrite tip (in which case the process is called cytophagocytosis), or a vesicle that has been shed by the melanocyte and contains melanosomes. Naturally, these mechanisms are not mutually exclusive. A fourth possible mechanism is fusion of plasma membranes, which, if any, presumably plays a secondary role, as it does not include phagocytosis.

Molecular Players in Melanin Transfer

  1. Top of page
  2. Abstract
  3. Modes of Transfer Revisited
  4. Molecular Players in Melanin Transfer
  5. Melanin Transfer and Disease
  6. Conclusion
  7. Acknowledgments
  8. References

PAR-2

The family of protease-activated receptors (PAR-1-PAR-4) consists of G-protein-coupled transmembrane receptors that are activated by serine proteases that cleave the extracellular amino terminal domain. The newly formed amino termini are tethered ligands, they undergo a conformational change and bind the receptors leading to activation. In the past 5 years a role for PAR-2 in melanin transfer has been established (49). PAR-2 is expressed in keratinocytes (50), not in melanocytes (51). Stimulation of this receptor enhances the phagocytosis rate of keratinocytes and leads to increased melanin transfer, which has been proved in vitro as well as in vivo(51–54). Melanocyte–keratinocyte contact is a prerequisite for this function (51). Ultraviolet irradiation induces PAR-2, and, the other way round, blocking of the PAR-2 receptor inhibits UV-induced pigmentation. Further, PAR-2 expression and induction by UV seem to depend on skin type, with a higher expression and more pronounced induction in dark-skinned individuals (55,56). The putative tethered ligand that cleaves and activates PAR-2 in vivo has yet to be identified. It has been shown in vitro that activation of PAR-2 leads to serine protease secretion by keratinocytes, creating a positive feedback loop. UVB induces a similar effect (55). PAR-2 signals to Rho as Rho-GTP is upregulated upon PAR-2 stimulation and, conversely, inactivation of Rho or its downstream effector Rho kinase abolishes PAR-2-stimulated phagocytosis (57). Meanwhile, commitment in phagocytosis in other cell types has put forward PAR-2 as a genuine phagocytic receptor. A leading role for phagocytosis in melanin transfer is thereby practically guaranteed. But then again, transfer cannot be completely inhibited by treatment with serine protease inhibitors (53). Either phagocytosis by PAR-2 is not the only mechanism for transfer or the receptor is re-activated when inhibitor levels are reduced. Most probably, PAR-2 is one molecular player in transfer together with many others. Redundancy exists as is the case for phagocytosis in other cells. In fact, just now the keratinocyte growth factor receptor (KGFR) has been allocated a similar role (58). Activation of KGFR enhances keratinocyte phagocytosis of latex beads and addition of KGF to co-cultures induces transfer of tyrosinase-positive granules. Phagocytosis via KGFR is dependent upon the PAR-2-Rho pathway, as well as on the activation of Rac and Cdc42 (58).

Apart from phagocytosis, PAR-2 effectuates skin pigmentation by stimulation of melanocyte dendricity. Prostaglandins, PGE2 and PGF2α, are released by the keratinocyte upon stimulation. They bind the surface of melanocytes, thereby inducing dendrite formation (59).

Adhesion molecules: cadherins and lectins

The melanocyte–keratinocyte adhesion site is comparable to the specialized junction between neurons (neuronal synapse), between immune cells (immunological synapse) and between phagocytes and their targets (phagocytic or engulfment synapse) and therefore will be addressed as pigmentation synapse (60–62). The establishment of these synapses first involves ligation of surface-expressed adhesion receptors, which leads to remodelling of the actin cytoskeleton. Eventually, mature and stable adhesion structures are formed. Neuronal synapses primarily form on axonal filopodia and then stabilize by the transmembrane anchoring of the cytoskeleton via cadherins and by the accumulation of scaffold proteins (63,64). Likewise, cadherins play a major role in the establishment of melanocyte–keratinocyte adhesion. They are a family of glycoproteins that function in promoting Ca2+-dependent cell–cell adhesion and serve as the transmembrane components of cell–cell adherens junctions. E-cadherin and P-cadherin are expressed in human melanocytes and both mediate melanocyte adhesion to keratinocytes. P-cadherin seems to play a minor role as opposed to E-cadherin which is the major mediator of melanocyte–keratinocyte adhesion (65). A role for E-cadherin in pigment transfer has been suggested based on findings in guttate leucoderma in Darier's disease (see below). This role awaits further proof. Further analogy of the pigmentation synapse with the neuronal synapse stems from recent data pointing in the direction of melanocytic filopodia being vanguards for adhesion and eventually secretion sites.

In analogy to the phagocytic synapse, the pigmentation synapse engages lectins (62). They are adhesion receptors that bind sugar residues presenting on the surface of delimiting membranes. When added to keratinocyte–melanocyte cultures, lectins and neoglycoproteins inhibit melanin transfer, as shown by flow cytometry and electron microscopy (66). The inhibition is reversible and can be enhanced by addition of niacinamide (67). Lectins that bind galactose residues are more effective than lectins binding mannose residues, while α-l-fucose receptors display a variable effect. Cerdan et al. studied the role of these molecules in binding of melanin-containing vesicles shed by melanoma cells to keratinocytes (46). Binding is inhibited by neoglycoproteins, showing a role for α-l-fucose-specific lectins of keratinocytes on the one hand and for 6-phospho-β-D-galactose-specific lectins of melanocyte-derived vesicles on the other hand. In brief, lectins and plasma membrane glycoproteins are involved in melanocyte–keratinocyte recognition and facilitate melanin transfer.

Other molecules involved (SNAREs, Rab GTPases and Rho GTPases) have been discussed in previous sections. A hypothetical model of the pigmentation synapse is shown in Figure 3.

image

Figure 3. The pigmentation synapse. A hypothetical model of the pigmentation synapse depicting molecules with an established role in transport or transfer: PAR-2 and its downstream effectors, the Rab27a–melanophilin–MyosinVa complex, Slp2-a and lectins; and molecules with a putative role in transfer: E-cadherin and SNAREs and Rabs mediating fusion of the melanosomal membrane with the plasma membrane.

Melanin Transfer and Disease

  1. Top of page
  2. Abstract
  3. Modes of Transfer Revisited
  4. Molecular Players in Melanin Transfer
  5. Melanin Transfer and Disease
  6. Conclusion
  7. Acknowledgments
  8. References

Some genetic diseases specifically and primarily affecting pigment transport and transfer are discussed below and summarized in Table 1.

Table 1. Genetic diseases presenting with abnormal pigmentation due to defective transport or transfer
DiseaseMouse modelGeneFunction related to transfer
Griscelli type IdiluteMyosinVaTransport of melanosomes along the sublemmal actin network
Griscelli type IIashenRab27aTransport and tethering of melanosomes at the sublemmal actin network
Griscelli type IIIleadenMelanophilinTransport and tethering of melanosomes at the sublemmal actin network
Hermansky PudlakgunmetalRGGTPrenylation of Rab proteins functioning in transport/transfer
Chediak-HigashibeigeLystDelivery of TGN-derived vesicles to premelanosomes? Specification of plasma membrane sites destined for transfer?
Guttate leucoderma in Darier's disease SERCA2Mediation of E-cadherin-based melanocyte–keratinocyte adhesion?

Genetic diseases affecting pigmentation, immunity and blood coagulation

A number of diseases arise from mutations in genes that regulate biogenesis, transport and/or secretion of secretory lysosomes. Clinically, this results in the appearance of immunodeficiency along with defects in pigmentation and in blood coagulation (68,69). Five mouse models of these diseases lack the capacity to pass on melanosomes to keratinocytes. They are dilute, ashen, leaden, gunmetal and beige.

Dilute, ashen and leaden are mouse models of Griscelli syndrome (70). Discovery of the affected genes has led to the identification of the tripartite complex involved in melanosomal transport along the sublemmal actin cytoskeleton: Rab27a (ashen)–melanophilin (leaden)–myosinVa (dilute) (8–10). Griscelli patients display silvery grey hair and discrete hypopigmentation of the skin (Figure 4A), due to perinuclear accumulation of melanosomes (Figure 4B). Interestingly, a loss-of-function mutation in dilute suppressor restores the coat colour of dilute and, to a lesser extent, of ashen and leaden (71). It does not involve rescue of actin-based transport at the dendritic tip and hence functions in a mode of transfer that is independent of peripheral distribution of melanosomes. The exact function of dilute suppressor has not been deciphered yet.

image

Figure 4. Transport and transfer diseases.(A)A Griscelli patient presents silvery grey hair and discrete hypopigmentation of the skin. (B)Melanosomes aggregate around the nucleus (arrow) resulting in decreased transfer. Scale bar = 5 µm. (C)Accumulation of melanosomes in gunmetal melanocytes (kindly provided by M Wei). Scale bar = 1.5 µm. (D)Chediak-Higashi skin presenting a melanocyte with giant melanosomes (arrow) and paucity of melanosomes in surrounding keratinocytes (kindly provided by R Boissy). Scale bar = 3 µm. (E)Hypopigmented macules or guttate leucoderma in a dark-skinned individual with Darier's disease. (F)Keratinocytes are relatively empty despite the presence of dendrites filled with mature melanosomes (arrowheads) (kindly provided by BK Goh). Scale bar = 5 µm.

Rab27a is also affected in the gunmetal mouse model of Hermansky–Pudlak syndrome. The affected gene encodes the α subunit of Rab geranylgeranyl transferase which controls membrane association and thus activation of Rab proteins. Melanosomes are fully melanized, though smaller in size, and are normally distributed throughout the cell (72). The number of intracellular melanosomes is increased, suggesting defects in secretion (73) (Figure 4C). The different phenotypes of ashen and gunmetal melanocytes suggest that other Rabs requiring Rab geranylgeranyl transferase are involved in transport and/or transfer. Recently, Rab8 has been assigned as a regulator of actin-based melanosome movement (74). Rab3a is another candidate.

Beige is the mouse model of Chediak–Higashi syndrome. Characteristic is the formation of giant melanosomes that are relatively hypomelanized and are unable to be transferred to surrounding keratinocytes (Figure 4D). The defective gene is Lyst that encodes an unknown cytosolic protein of 430 kDa. Different speculative functions are assigned to the Lyst protein (75). It would anchor vesicles derived from the trans-Golgi network to microtubules for directed delivery to and fusion with late endosomes or, in the case of melanocytes, premelanosomes. Defective transport of vesicles that contain melanogenic enzymes and regulatory proteins prevents normal maturation and results in enlargement and/or fusion of premelanosomes. Another hypothesis comes from the large size which could place Lyst in the same family as piccolo and bassoon, components of the cytoskeleton of presynaptic nerve terminals. It would then function in specifying regions of the plasma membrane destined for transfer. Further, Lyst has been shown to interact with a number of proteins that regulate vesicle docking and fusion (76).

Guttate leucoderma in darier's disease

Darier's disease is an autosomal dominant skin disorder caused by mutations in the ATP2A2 gene that encodes the sarco/endoplasmic reticulum calcium-ATPase isoform 2 (SERCA2). Aberrant intracellular calcium signalling has been postulated to lead to a disturbance in keratinocyte adhesion resulting in acantholysis and dyskeratosis. In dark-skinned individuals, an unusual manifestation of guttate leucoderma (confetti-like hypopigmented macules) may be found in addition to the classical features of greasy keratotic papules, palmoplantar keratoderma and nail changes (Figure 4E). Histological examination of these leucodermic macules shows a reduction in epidermal melanin and may be accompanied by acantholysis (77–79). On ultrastructural examination, the melanocytes display a normal morphology along with mature melanosomes. However, when compared to normally pigmented perilesional skin, there is a significant reduction of melanin granules within the basal and suprabasal keratinocytes despite being surrounded by melanosome-filled dendrites (78,80) (Figure 4F). This observation supports the postulate that loss of keratinocyte–melanocyte adhesion results in loss of pigment transfer. As E-cadherin is the major mediator of melanocyte adhesion to keratinocytes and as it is dissociated in acantholytic lesions of Darier's disease, this molecule is likely involved (65,81). Furthermore, trafficking of E-cadherin to the cell surface is disturbed on chemical inhibition of the SERCA2 pump by thapsigargin (82). The role of E-cadherin in reduced transfer of melanin in hypopigmented macules associated with Darier's disease is currently being studied (80).

Conclusion

  1. Top of page
  2. Abstract
  3. Modes of Transfer Revisited
  4. Molecular Players in Melanin Transfer
  5. Melanin Transfer and Disease
  6. Conclusion
  7. Acknowledgments
  8. References

Although the study of melanin transfer has proven to be a difficult task (83), recent studies have begun to unravel its intricacies. With the discovery of PAR-2 involvement, phagocytosis as a necessary step becomes undeniable. Much can be learned from studying transfer means used by other cell systems and, vice versa, with elucidation of the process governing distribution of melanin in the epidermal melanin unit, a valuable model for intercellular transfer will arise.

Acknowledgments

  1. Top of page
  2. Abstract
  3. Modes of Transfer Revisited
  4. Molecular Players in Melanin Transfer
  5. Melanin Transfer and Disease
  6. Conclusion
  7. Acknowledgments
  8. References

We thank Dr B-K. Goh and the members of our laboratory for critical reading of the manuscript. We also thank Professor R. Boissy for the electron micrograph of Chediak-Higashi skin, Dr M. Wei for the micrograph of gunmetal skin and Dr B-K. Goh for the pictures of guttate leucoderma in Darier's disease.

References

  1. Top of page
  2. Abstract
  3. Modes of Transfer Revisited
  4. Molecular Players in Melanin Transfer
  5. Melanin Transfer and Disease
  6. Conclusion
  7. Acknowledgments
  8. References