By continuing to browse this site you agree to us using cookies as described in About Cookies
Wiley Online Library is migrating to a new platform powered by Atypon, the leading provider of scholarly publishing platforms. The new Wiley Online Library will be migrated over the weekend of February 24 & 25 and will be live on February 26, 2018. For more information, please visit our migration page: http://www.wileyactual.com/WOLMigration/
Desmond J. Tobin, Medical Biosciences Research, School of Life Sciences, University of Bradford, Bradford, West Yorkshire BD7 1DP, UK, Tel.: +44 (0)1274 233585, Fax: +44 (0)1274 309742, e-mail: firstname.lastname@example.org
Abstract: Melanosomes are melanocyte-specific lysosome-related organelles that are transferred to keratinocytes of the epidermis and anagen hair bulb. Transferred melanin forms supra-nuclear caps that protect epidermal keratinocytes against UV irradiation. The mechanism(s) responsible for melanosome transfer into keratinocytes and their subsequent intra-keratinocyte distribution has long remained one of the most enigmatic of heterotypic cell interactions. Although there have been many attempts to study this process, significant progress has been hindered by the absence of an adequate in vitro model. During our ongoing study of melanocyte–keratinocyte interactions in skin and hair follicle, we have developed a novel in vitro assay that exploits the specificity of Silv/Pmel17/gp100 expression for melanosome/melanin granules. Using matched cultures of keratinocytes and melanocytes isolated from normal healthy epidermis together with double immunofluorescence, we have determined that gp100 is a surprisingly useful tracker of transferred melanin. Moreover, transferred gp100 stained melanin granules emit a bright fluorescence signal, facilitating ready quantification of melanin transfer levels between melanocytes and keratinocytes. This quantitative approach was validated using known inducers and inhibitors of the melanocyte phenotype. This assay further confirmed that cytophagocytosis of melanocyte components (e.g. dendrite tips) by keratinocytes is one route for melanin incorporation into keratinocytes. Lastly, a role for the recently proposed filopodium as a direct conduit for melanin transfer was substantiated using this novel approach. In conclusion, this assay promises to significantly aid our investigations of the molecular basis of melanosome transfer and offers a new tool for the clinical evaluation of melanocyte modulators.
Melanocytes are neural crest-derived cells that synthesize and store the melanin pigment within unique lysosome-related organelles termed melanosomes (1). Eumelanosomes biogenesis progresses via four sequential morphologic steps as they mature (2). Stage I (pre-)eumelanosomes are membrane-bound and electron-lucent spherical vesicles produced at the peri-nuclear area of melanogenically active melanocytes. The transition to stage II melanosomes involves an elongation of the stage I vesicle and the appearance within it of distinct fibrillar structures, whose production and maturation depend on the presence of the structural protein Silv/gp100/Pmel17 (pre-melanosomal protein-17). Shortly after its processing in stage II melanosomes (recognized by HMB45), gp100 is further cleaved into several fragments (recognized by αPEP13), which form the fibrillar matrix of the organelle (3–5). NKI/beteb (used in the current study), recognizes the cleaved formed of gp100 when present in maturing/mature (stage III/IV) melanosomes (5). In pigmented cells, melanins are deposited on these melanosomal fibres, resulting in a progressively pigmented internal matrix (stage III). In highly pigmented tissues, melanin synthesis and deposition continue until little or no internal structure is visible (stage IV).
Visible pigmentation in mammals requires the transfer of melanin granules from melanocytes to keratinocytes. For this intercellular transfer to be effective, melanosomes must first accumulate at the distal end of the melanocyte dendrites using cooperative transport mechanisms. These include long-range, bidirectional, microtubule-dependent melanosome movements along the dendrite length before coupling to myosin Va-dependent capture machinery within distal actin-rich regions of the dendrite (6). Myosin Va is recruited onto the melanosome surface by a receptor complex containing Rab27a and melanophilin in a Guanosine triphosphate (GTP)-dependent fashion (7–9). The absence of any one of these three proteins collapses the myosin Va-dependent capture of melanosomes in the periphery, causing instead their accumulation in the perikaryon of the cell.
Although much progress has been made relating to the intra-melanocyte movement of melanosomes, it has been difficult to unravel the mechanisms involved in the actual transfer of melanosomes to keratinocytes, and also the nature of the melanocyte–keratinocyte interactions during this transfer process. Recently, an interesting study revealed that the interactions between melanocyte and keratinocyte plasma membranes induced a transient intracellular calcium signal in keratinocytes that is required for melanosome transfer (10). The majority of studies, however, have been based on melanocyte–keratinocyte co-cultures using electron microscopy and time-lapse video microscopy. Based on these studies, several different mechanisms (which may co-exist) have been proposed over the years, including: (i) exocytosis of individual melanosomes into the extracellular space and their subsequent endocytosis by keratinocytes (11,12); (ii) cytophagocytosis of melanocyte dendritic tips by keratinocytes (13) and (iii) direct ‘injection’ of melanosomes into keratinocytes via tunnelling nanotubes or filopodia (14).
A major obstacle to elucidating the contribution of each or all of these proposed mechanisms of melanosome transfer has been the lack of a suitable assay to allow rapid assessment, both quantitatively and qualitatively, of transfer events in vitro or in vivo. Electron microscopy of co-cultures or reconstructed epidermis has been used to visualize transfer. Melanosome transfer has been observed, albeit only qualitatively, after mixing dark skin-derived melanocytes with light skin-derived keratinocytes (15,16). Time-lapse light microscopy has also been used, although actual transfer in these co-cultures has not been seen (or at least reported), making this assay unsuitable for quantification (14,17). Where quantification was attempted, this has focused on measuring melanin exocytosis into the culture medium in melanocyte monoculture (17) or phagocytosis of purified melanosomes by keratinocytes (18). Alternatively, some researchers have used fluorescence labelling (e.g. with CFDA, Dil and Dio) of melanocytes and subsequently assessed the fluorescence label transferred to keratinocytes (14,16). Although this technique has its advantages, it is not suitable for quantification of melanosome transfer, as these fluorophores are not melanosome-specific. Other groups have transfected melanocytes with GFP-labelled melanocyte-specific proteins (e.g. TYR, TYRP1, MART1, GP100 and RAB27a) or by incubation of melanocytes with fluorescent-labelled melanin intermediates or homologues (19). In most cases, a weak cytoplasmic fluorescence was observed, which probably reflects sub-optimal sorting or deficient incorporation of the fluorescent protein.
Thus, all currently available approaches have significant drawbacks in terms of their specificity, reproducibility, quantification and assay complexity. An ideal quantification assay should label melanosomes in a highly sensitive and specific manner and use melanocytes and keratinocytes in a two- or three-dimensional culture system. Moreover, this assay should allow easy identification of melanin in donor and recipient cell types. In the current study, we have determined that the detection of gp100 protein is a very useful quantifiable tracker of transferred melanin to keratinocytes in vitro. Using this assay, the evaluation of agents to assess their influence on melanin transfer efficacy can be readily achieved. Using this novel gp100 positive-melanosome co-culture assay, the detection of melanocyte fragment phagocytosis by keratinocytes as well as the detection of melanosomes within melanocyte filopodia promises to aid understanding the molecular basis of melanosome transfer.
Results and discussion
Dissection of the molecular mechanisms that underlie melanosome transfer to keratinocytes lies at the heart of understanding many aspects of human pigmentation. The expression of Silv/gp100/Pmel17 protein is limited to pigment cells, including cutaneous melanocytes, uveal tract melanocytes and retinal pigment epithelium (20). Gp100 appears to play a role in generating the melanosome fibrillar matrix and in facilitating melanin deposition – an event that appears to be unique to melanosomes. Several groups have developed monoclonal antibodies that react specifically with gp100, including the antibodies NKI/ beteb (21), HMB-45 (22), HMB-50 (23) and ME20 (24). These antibodies recognize a 100 kDa glycoprotein (gp100). NKI/beteb, HMB-45 and HMB-50 were subsequently shown to identify the product of the same cDNA (25). An analysis of these cDNA sequences identified a predicted product almost identical to human Pmel17 except for the deletion of seven amino acid residues from the transmembrane domain (26). For this study, we have used the NKI/beteb antibody to detect both pre-melanosomes (stages I/II) and melanosomes (stage III/IV), as indicated in the original study that generated this antibody clone (21).
Detection of melanosome transfer to keratinocytes using gp100 in matched melanocyte–keratinocyte co-cultures
The process of melanosome transfer from melanocytes to keratinocytes has been widely studied in vitro using co-culture models of these two cell types (27,28). To investigate the inherent mechanisms of melanosome transfer from melanocytes to keratinocytes, we have developed an in vitro assay involving full matched (i.e. both cell types were isolated and cultivated from the same piece of skin tissue) co-cultures of human epidermal melanocytes and keratinocytes in a ratio of approximately 1:10. Melanocyte and keratinocyte identity was confirmed using NKI/beteb and anti-cytokeratin antibodies, respectively. Melanocytes in the co-culture were recognized by their bright homogeneously green gp100-positive immunofluorescence, while melanosomes transferred to recipient keratinocytes were detected as bright green gp100-positive spots [Fig. 1a(i) and b(i, ii)]. As a parallel control, the cytokeratin-positive keratinocytes used to establish the co-cultures did not reveal any gp100-positive melanosomes at passage 2 [Fig. 1a(ii)], thereby establishing the base-line levels of previously transferred melanin at zero. Gp100-positive melanosomes/melanin granules were detected in keratinocytes located both close and at some considerable distance to the melanosome-donating melanocyte. Gp100-positive spots were located predominantly in the peri-nuclear region of recipient keratinocytes (Fig. 1b). Moreover, confocal microscopic views of these co-cultures confirmed the peri-/supra-nuclear localization of transferred melanin in these keratinocytes (Fig. S1). As this assay is based on the detection of gp100 – a melanocyte-specific protein found in skin of all photo-types, this approach therefore has wide applicability to studies of pigmentation in all skin types.
Quantitative analysis of gp100-positive melanosome transfer using modulators of melanocyte phenotype
Gp100 immunostaining provides a powerful tracking method for the global assessment of melanin transfer to keratinocytes (with the caveat than NKI/beteb may not detect all of the most mature melanosomes) and for the evaluation of melanocyte phenotype modulators. Double immunolabelling (NKI/beteb and anti-cytoskeleton antibody) of the co-cultures after treatment with phosphodiesterase inhibitor (thus, cAMP inducer) 3-Isobutyl-1 methylxanthine (IBMX), soybean trypsin inhibitor (STI), niacinamide, nocodazole and cytochalasin D revealed clear changes in the number of fluorescent green gp100-positive melanin granules in keratinocytes (Fig. 2a). In this assay, melanosome transfer from melanocytes to matched keratinocytes under basal (i.e. unstimulated) conditions was determined at an average of 20.2 gp100-positive spots per keratinocyte [Fig. 2a(i) and b]. However, this was increased almost fourfold after 24-h stimulation of co-cultures with IBMX [76.2 gp100-positive spots per keratinocyte, P < 0.001; Fig. 2a(ii) and b]. Our assay system can also be applied to assess the effects of inhibitors of melanosome transfer. Niacinamide also reduced melanosome transfer by 27% [14.8 gp100-positive spots per keratinocyte, P < 0.01; Fig. 2a(iii) and b], a result that correlates with its clinical use as a lightener of cutaneous pigmentation (29,30). Soybean trypsin inhibitor, an inhibitor of protease-activated receptor-2 (27,31,32), significantly reduced melanosome transfer in our matched melanocyte–keratinocyte co-cultures, as evidenced by a reduction of >40% in melanosome transfer [12 gp100-positive spots per keratinocyte, P < 0.01; Fig. 2a(iv) and b].
We have also validated this assay for the study of the role of cytoskeletal components actin filaments and microtubules in melanosome transfer, including the microtubule disruptor nocodazole and an actin disruptor cytochalasin D (6). Nocodazole and the cytochalasin D both almost completely inhibited melanosome transfer (4 gp100-positive spots per keratinocyte, P < 0.01; 3.4 gp100-positive spots per keratinocytes, P < 0.01, respectively) [Fig. 2a(v), a(vi) and b].
To further evaluate the utility of our gp100 melanocyte–keratinocyte co-culture assay to assess dose-dependent effects of test compounds on melanosome transfer, we used increasing concentrations of the prototypic cAMP inducer IBMX (1 × 10−6 to 2 × 10−4m). Cyclic AMP promotes the distribution of melanosomes to the melanocyte periphery and can also stimulate melanosomes to associate with the melanophilin/Slac2a/actin complex prior to transfer to keratinocytes (33). A dose-dependent increase in the number of gp100-positive spots in keratinocytes was observed (Fig. 3a and b). Maximal transfer of melanosomes to keratinocytes was observed when co-cultures were stimulated with 1 × 10−4m IBMX (an increase of 377%), while IBMX at 1 × 10−6m was also effective, stimulating a 43% increase in melanosome transfer compared with unstimulated basal levels (Fig. 3a and b). This IBMX-facilitated melanosome transfer was predominantly to the peri-nuclear region of the recipient keratinocytes. This localization is highly reminiscent of melanin capping in keratinocytes after stimulation with ultraviolet radiation and IBMX in reconstructed pigmented epidermis (34). IBMX is also routinely used at 1 × 10−4m in melanocyte biology research to stimulate melanogenesis (35). Previous studies have also reported the stimulation of melanosome release and uptake by other cAMP modulators (e.g. α-melanocyte-stimulating hormone) (17). However, in those assay systems, only melanosome exocytosis by melanocytes and melanin granule phagocytosis by keratinocytes were assessed, and so limits their usefulness in assessing the discharge mode of melanin transfer.
Detection of melanocyte dendrite cytophagocytosis by keratinocytes
Ultrastructural and time-lapse video microscopy have long supported the hypothesis that melanin transfer to keratinocytes can occur, at least in part, by cytophagocytosis of melanocyte dendrite tips by adjacent keratinocytes (13,36–39). Immunolabelling with NKI/beteb antibody detected large (5–7 μm) melanocytic elements within some keratinocytes (Fig. 4). These melanocytic components were immunolabelled with NKI/beteb in a broadly similar manner to intact and discrete melanocytes within these co-cultures (Fig. 4a), i.e. with relatively intense fluorescence. This staining pattern would not be expected in the case of keratinocytes when phagocytosing cellular elements of a melanin-containing keratinocyte. The NKI/beteb signal in the latter would be much lower at the exposures used in this study.
In some cases, these melanocytic elements were contained within phagolysosomes distributed close to the keratinocyte nucleus (Fig. 4b). These NKI/beteb-positive melanocytic elements corresponded exactly with highly pigmented melanocytic fragments visible by phase-contrast microscopy (Fig. 4a and b, right panel). In this way, we have described a readily accessible and powerful assay to investigate this mode of melanin transfer (cytophagocytosis) between melanocytes and keratinocytes that is a significant advance on currently employed systems (e.g. use of beads, particles or isolated melanin granules).
Detection of melanosome transfer to keratinocytes via filopodia
Our current findings and those of previous studies (16,18,27) provide evidence of phagocytosis of melanocyte dendrites by keratinocytes as a major mode of melanosome transfer. However, our gp100-positive melanosome transfer assay may also provide an opportunity to examine the potential role of melanocyte filopodia in the process of melanosome transfer. This is a recent addition to the potential modes of melanin transfer originally proposed by Scott et al. (14). These investigators provided evidence in support of melanosome transfer via filopodia interacting with keratinocyte plasma membranes. Cell–cell communication based on nanotubes that form complex networks have been reported in several cell types in vitro where they appear to function as channels for organelle transport (40). Tubular filopodia (50–200 nm in diameter) composed of actin filaments make contact with and fuse with the targeted cell, thereby forming a direct cytoplasmic bridge that permits unidirectional transport of organelles and plasma membrane molecules (but interestingly not soluble cytoplasmic molecules) between the cells. In the case of melanocytes, the filopodia extend from the dendrite tips and cell body and adhere to the surface of neighbouring keratinocytes to facilitate the transport of melanosomes into the keratinocyte cytoplasm (14).
Immunolabelling with NKI/beteb antibody detected green fluorescent melanosomes within melanocyte filopodia [Fig. 5a(i)]. The location of these melanosomes corresponded exactly with highly pigmented melanosomes within filopodia, as seen in paired phase-contrast images [Fig. 5a(ii)]. We have exploited the power of gp100-positive immunostaining in order to investigate the role these filopodia play in transferring melanosomes to keratinocytes (Fig. 5b). Using this tracker system, individual melanin granules were identified within narrow cellular tracks emerging from melanocyte dendrite tips where they were immunolabelled with green fluorescence. These gp100-positive melanin granules (green fluorescence) were also detected in the cytoplasm of keratinocytes. These data strongly support the view that filopodia play an important role in transferring melanosomes to keratinocytes in co-culture. Using time-lapse digital photography, we have obtained the confirmation that filopodia are conduits of melanosome transfer to keratinocytes (Fig. 5c). The melanosomes appeared to fill the filopodial lumen completely and to be transferred only in single file. However, as Scott et al. (14) also found, it was difficult to monitor the fate of the transferred granules after these granules made contact with the keratinocyte cell membrane due to differences in the optical characteristics of the filapodia and the keratinocytes cell membrane (including its ruffling).
Our time-lapse photomicrography of matched melanocyte and keratinocyte co-culture allowed us to estimate the approximate rate of melanosome transfer within a single melanocyte filopodium per unit time. In our assay system, eight melanosomes within a single filopodium were seen to cross from the donating melanocyte to the receiving keratinocytes in 25 min [Fig. 5c(i), movie file Fig. S2]. This approximates to a single melanocyte transfer in less than 5 min [Fig. 5c(ii)]. These data provide strong evidence for the direct transfer of melanosomes into the keratinocytes via fusion of melanocyte nanotubes with recipient keratinocytes. The role of filapodia in melanin transfer in vivo needs to be more fully assessed, although structures consistent with filapodia, arising from the sides and tips of melanocyte dendrites and containing melanosomes within their lumina, have already been reported in human skin in situ (14).
In summary, our gp100-positive-matched melanocyte–keratinocyte co-culture assay is a very powerful and specific tool for the evaluation of multiple melanosome transfer routes to keratinocytes in the epidermal melanin unit of the skin and hair follicle, both in healthy and disease states. This assay may also prove to be a powerful tool to evaluate the potential agents that influence melanin transfer. The ready detection of both cytophagocytosis of melanocytic elements (containing several melanosomes) and the transfer of single melanosomes via filopodia suggest that we can utilize this assay for the study of several aspects of melanosome fate including interaction with the filopodial environment, at membrane fusion and the proteins involved in phagocytosis. Certainly, the utilization of gp100 immunolabelling has provided us with an excellent tool to explore many questions relating to the complexity of melanosome transfer in the future.
Materials and methods
Isolation and culture of matched epidermal keratinocytes and epidermal melanocytes
Human abdomen skin was obtained with informed consent and local research ethics committee approval from normal healthy Caucasian donors with skin photo-type II (female 51y, 55y, 65y; mean 57y) after elective plastic surgery. All cell culture reagents were obtained from Invitrogen Ltd (Paisley, Scotland) unless otherwise stated. Skin samples were collected in RPMI 1640 medium and were processed within 5 h of surgery. Epidermal melanocyte cultures were established as previously described (41) and grown in a mixture of keratinocyte serum-free medium (K-SFM) and Eagle’s minimal essential medium (EMEM) supplemented with 1% FBS, 1 × concentrated non-essential amino acid mixture, penicillin (100 U/ml)/streptomycin (100 μg/ml), 2 mm l-glutamine, 5 ng/ml basic fibroblast growth factor and 5 ng/ml endothelin-1 (Sigma, Poole, Dorset, UK). Matched epidermal keratinocytes were established as described previously (41) and grown in K-SFM supplemented with 25 μg/ml bovine pituitary extract (BPE), 0.2 ng/ml rEGF, penicillin (100 U/ml)/streptomycin (100 μg/ml) and 2 mm l-glutamine. Culture medium was replenished every second day. Primary cultures of keratinocytes and melanocytes were identified using anti-cytokeratin antibody (Abcam, Cambridge, UK) and the melanocyte-specific NKI/beteb antibody (Monosan, Uden, the Netherlands) to gp100, respectively.
For co-culture studies, melanocytes (passage 4) and keratinocytes (passage 2) were seeded onto eight-well Lab-Tek® chamber slides (ICN Biomedicals, Inc., Aurora, OH, USA) at a cell density of 1 × 104 cells/well and in a ratio of 1 melanocyte to 10 keratinocytes. This higher ratio of melanocyte to keratinocytes was selected on the basis that the latter lacked the third dimension as found in situ, where a ratio of 1 melanocyte to 36 viable keratinocyte is the norm. Co-cultures were maintained overnight (16 h) in a mixture of K-SFM and EMEM (co-culture medium) to allow cell attachment, followed by medium replenishment for a further 24 h.
The co-cultures were incubated for 24 h in serum-starved medium (i.e. lacking fetal calf serum and BPE) to remove exogenous sources of growth factors, before assessing the effects of phenotypic modulators. Co-cultures were incubated for 24 h with IBMX (100 μm), soybean trypsin inhibitor (0.01%), niacinamide (10 μm), the microtubule disruptor nocodazole (100 nm) and an actin disruptor cytochalasin D (500 nm). In a separate experiment, co-cultures were also treated with different concentrations of IBMX (1, 10, 50, 100 and 200 μm) for 24 h. IBMX is also sometimes used as a supplement of melanocyte culture medium routine media (although not in our laboratory) to promote melanocyte proliferation (42).
Melanocyte and keratinocyte monocultures, and matched co-cultures of these cells were grown on eight-well Lab-Tek® chamber slides for the detection of melanosome transfer by cytophagocytosis, and filopodia. Cells were fixed in ice-cold methanol for 10 min at −20°C, washed in PBS and then blocked with 10% donkey serum.
For single labelling experiments, the primary antibody NKI/beteb (1:30) was applied overnight at 4°C, followed by incubation with FITC-conjugated secondary antibody (1:100) (Jackson Immunoresearch Laboratories, Inc., West Grove, PA, USA) for 1 h at room temperature. For double-labelling experiments, the first primary antibody NKI/beteb (1:30) was applied overnight at 4°C, followed by incubation with fluorescein-conjugated secondary antibody (1:100) for 1 h at room temperature. The second primary antibody against cytokeratin (1:100) or actin (1:100) (Santa Cruz, CA, USA) was applied for 1 h at room temperature followed by a Tetramethyl Rhodamine Isothiocyanate (TRITC)-conjugated secondary antibody (1:100) (Jackson Immunoresearch Laboratories, Inc.). DAPI (Vector Laboratories, Burlingame, CA, USA) was used to stain nuclei. Images were captured with a cooled Hamamatsu digital camera (Hamammatsu, Japan) using a 100 × objective and post-processed using Paint Shop Pro (Jasc Software Ver. 7, Corel, MN, USA). Negative controls included the omission of primary antibody and replacement with non-immune serum from secondary antibody host and inclusion of secondary antibodies.
Quantitative analysis of melanosome transfer
Matched melanocyte–keratinocyte co-cultures were stimulated with IBMX, STI, niacinamide, nocodazole and cytochalasin D and compared with unstimulated cells. Evaluation of melanosome transfer was performed by counting fluorescent gp100-positive spots within recipient keratinocytes in five random microscopic fields per well at ×100 magnification (oil immersion) in three independent experiments. To avoid counting melanin granules that may still be associated with melanocytes, we only counted gp100-positive spots within keratinocytes that were not in direct contact with melanocytes. Analysis by us (data not shown) and others (28) using other melanogenesis-related proteins (e.g. tyrosinase and Trp1) was not suitable for the assessment of melanosome transfer, as the transfer of these melanogenesis marker proteins yielded non-quantifiable diffuse staining, and so was not pursued.
Time-lapse digital microscopy
Matched co-cultures of melanocytes and keratinocyte in a ratio of 1:10 were sub-cultured onto μ-Dishes (Ibidi, Integrated BioDiagnostics, Munich, Germany). This 1:10 ratio has been widely adopted in similar studies (28), and it is also operationally useful, as it increases the opportunity to see both melanocyte and keratinocyte partner cells in single images at appropriate magnifications. The cells were viewed on an Eclipse TE2000 inverted research microscope (Nikon, Tokyo, Japan) under phase contrast with a ×100 objective. Sequential images were obtained at 10 s intervals using a Hamamatsu digital camera. The resulting eight bit/pixel megapixel (544 × 416) images yielded a resolution of 10 pixels/micron when combined with the ×100 microscope objective.
Statistical significance between groups and treatment was assessed using one-way ANOVA and Dunnett’s post-test using Prism v. 4.00 (GraphPad Software, Chicago, IL, USA). Statistically significant differences are denoted with asterisks: **P < 0.01, ***P < 0.001.
We thank Jamie Fearnley, Medical Biosciences Research, University of Bradford, for his technical assistance.