Cyclic AMP promotes a peripheral distribution of melanosomes and ...

The FASEB Journal express article 10.1096/fj.03-1240fje. Published online April 1, 2004.

Cyclic AMP promotes a peripheral distribution of melanosomes and stimulates melanophilin/Slac2-a and actin association Thierry Passeron, Philippe Bahadoran, Corine Bertolotto, Christine Chiaverini, Roser Buscà, Gaëlle Valony, Karine Bille, Jean-Paul Ortonne, and Robert Ballotti INSERM U597, Biologie et pathologie des cellules mélanocytaires, Faculté de Médecine, Nice cedex 2, France Corresponding author: R. Ballotti, INSERM U597, Biologie et pathologie des celllules mélanocytaires, Faculté de Médecine, Avenue de Valombrose 06107, Nice cedex 2, France. E-mail: [email protected] ABSTRACT Melanosomes are melanin-containing organelles that belong to a recently individualized group of lysosome-related organelles. Recently, numerous reports have dissected the molecular mechanisms that control melanosome transport, but nothing was known about the possible regulation of melanosome distribution by exogenous physiological stimulus. In the present report, we demonstrate that a physiological melanocyte-differentiating agent such as αmelanocyte-stimulating hormone, through the stimulation of the cAMP pathway, induces a rapid centrifugal transport of melanosomes, leading to their accumulation at the dendrite tips of melanocytes. Interestingly, the small GTP binding proteins of the p21Rho family and one of their effectors, p160 Rho-associated kinase, but not PKA, play a key role in redistribution of melanosomes at the extremities of the dendrites. Further, we have investigated, at the molecular level, the effect of cAMP on the different proteins involved in the control of melanosome transport. We demonstrate that cAMP stimulates the expression of Rab27a and rapidly increases the interaction of the melanophilin/Slac2-a with actin. Thus, we propose that the stimulation of the interaction between melanophilin/Slac2-a and actin would allow the rapid accumulation of melanosomes in the actin-rich region of the dendrite extremities. Key words: melanocytes ● alpha-melanocyte stimulating hormone

M

elanosomes belong to a recently individualized group of subcellular organelles called lysosome-related organelles or secretory lysosomes, which also includes lytic granules, MHC class II compartments, platelet-dense granules, basophil, and azurophil granules (1, 2). Melanosomes are melanocyte and pigment epithelial cell specific organelles in which the synthesis of melanin takes place. In human skin, melanosomes transport and transfer melanin to keratinocytes. Then, melanin is gathered in the supranuclear area of basal keratinocytes and thereby protects DNA from the mutagenic effects of ultraviolet radiation of the solar light.

Page 1 of 19 (page number not for citation purposes)

Recently, numerous reports have dissected the molecular mechanisms governing melanosome transport. The data gathered in these studies indicate that melanosomes are transported within the melanocytes on both microtubule and actin networks (3). The microtubule dependent transport is bi-directional and mediated through kinesin and dynein molecular motors (4, 5). The actin network allows the transport of melanosomes in the dendrite outgrowths and their docking at the dendrite tips (3, 6). Unequivocal data have demonstrated that at least three proteins, myosin-Va (7, 8), Rab27a (9-11), and melanophilin/Slac2-a (12-14), play a pivotal role in the actindependent transport and docking of melanosomes. Rab27a is a small G protein anchored in the melanosomal membrane (10, 15). Myosin-Va is a molecular motor that binds through its Nterminal head domain with the actin filaments (16) but was also reported to interact, indirectly, with the microtubule network (7, 17, 18). Melanophilin/Slac2-a is a Rab27a effector that belongs to the synaptotagmin-like protein family (19). Melanophilin/Slac2-a makes the link between Rab27a and myosin-Va and allows the interaction of melanosomes with the actin network (2022). It is noteworthy that several other Rab27a effectors have been described, but their involvement in melanosome transport remains to be elucidated (13, 14, 23). Taken together, these observations have shed light on the unexpected complexity of the molecular mechanisms that control the transport of melanosomes and the other lysosome-related organelles. Interestingly, several reports have documented a regulation of melanosome transport in melanophore pigment cells of lower vertebrates (24-26). In these cells, exogenous signals, which increase the cAMP levels, lead to a rapid movement of melanosomes away from the cell center. However, in mammalian melanocytes, a regulation of melanosome distribution upon physiological stimuli has never been clearly demonstrated. In the present report, we have undertaken a detailed study of the regulation of melanosome transport by αMSH and cAMP in B16 mouse melanoma cells. Further, we have investigated, at the molecular level, the effect of cAMP on the different proteins involved in the control of melanosome transport. We demonstrate that cAMP increases the interaction of melanophilin/Slac2-a with actin, thereby leading to the rapid accumulation of melanosome in the actin-rich region of the dendrite extremity. It is tempting to propose that this rapid docking of melanosomes at the dendrite tips allows an early transfer of melanosomes from this active zone to the surrounding keratinocytes. In humans, this process might account for the immediate skin darkening, observed just after sun exposure, which could be a rapid, adaptive, and protective response against the carcinogenic effect of the solar light. MATERIALS AND METHODS Cell lines Murine melanoma cell line B16-F10 and human kidney embryonic cells (A293) were grown at 37°C and 5% CO2 in DMEM supplemented with 7% fetal bovine serum (Hyclone, Logan, UT), 100 U/ml penicillin, and 50µg/ml streptomycin. Antibodies Rabbit polyclonal antibody directed to the C-terminal portion of Tyrp1 (pep1) was a gift from Dr. Hearing (Bethesda, MD; ref 27). Anti-melanophilin antibody was obtained by injection of a Page 2 of 19 (page number not for citation purposes)

rabbit with a peptide corresponding to the C-terminal part of melanophilin/Slac2-a (28). Antibody to Rab27a was a monoclonal antibody (PharMingen BD Biosciences, San Diego, CA); and antibody to myosin-Va was a goat polyclonal antibody raised against a peptide mapping near the N terminus of myosin-Va: myosin-Va (N-20) sc-17706 (Santa-Cruz, Santa-Cruz, CA). AntiCREB and anti-phospho-CREB (s133) were from Cell Signaling Technology, Inc. (Beverly, MA). Secondary antibodies used were as follows: Texas Red goat polyclonal antibody raised against mouse (Molecular Probes, Leiden, The Netherlands); FITC goat polyclonal raised against mouse (Dakopatts, Glostrup, Denmark); Texas Red goat polyclonal raised against rabbit (Dakopatts); and FITC goat polyclonal raised against rabbit (Molecular Probes). Transfections B16 and A293 cells were transfected with lipofectamine (Invitrogen) according to the protocol of the manufacturer. In brief, 6 x 106 cells, in 150 cm2 culture dishes, were transfected with 55 µl of lipofectamine and 20 µg of plasmid in optimen. After 6 h, the cells were replaced in DMEM supplemented with 7% fetal bovine. The plasmids used encode GST-melanophilin/Slac2-a (gift from Dr. M. Fukada, Riken Brain Science Institute, Saitama), GFP-myosin-Va (28), and GFPactin (gift from Dr. C. Ballestrem) fusion proteins. Immunofluorescence B16 melanoma cells grown on glass coverslips (20x103 cells per point) in 12-well dishes were fixed in freshly prepared 3% paraformaldehyde at room temperature for 20 min and permeabilized by a 2 min treatment with PBS, 1% BSA, and 0.2% saponin. Cells were incubated with primary antibodies for 1 h at room temperature and then for 1 h with secondary antibody. Phalloïdin-Texas Red or -FITC (Dakopatts) 1/200 was used for actin labeling. Then, cells were washed with PBS and examined with the x40 objective using a Zeiss Axiophot microscope equipped with epifluorescence illumination. Time-lapse microscopy Cells grown in with 2-well chambers were observed using an inverted microscope Zeiss® with halogen illumination. During the observation, the cells were placed in an incubation chamber at 37°C with 5% CO2. Bright field or Differential Interference Contrast (DIC) images were collected (1 frame each 20 s) with a digital camera and a PC computer. Frames were treated by Metamorph® software. Parameters of mounting were 10 frames/s. Finally, movies were saved in Quicktime® format. As previously reported (3), the black vesicles observed in DIC images have been identified as melanin-containing melanosomes since they can also be observed in bright field images. Melanosome immunopurification B16 cells, grown on 10 cm dishes, were washed three times with cold PBS and scraped in a buffer containing 50 mM Tris pH 7.4, 250 mM sucrose, and 3% nonfat dry milk. Cells were lyzed by 3 freeze-thaw cycles (liquid nitrogen/37°C), followed by 10 passages through a 26 G needle, and centrifuged for 10 min at 400 g to remove nuclei. The supernatants were incubated for 2 h at 4°C with the polyclonal anti-Tyrp1 antibody or with a preimmune serum, both previously fixed onto magnetic protein A-Sepharose beads (Dynal Biotech, Oslo, Norway). After Page 3 of 19 (page number not for citation purposes)

incubation, the beads were washed three times with PBS and then resuspended in buffer containing 10 mM Tris, pH 7.4, 1% Triton X-100, and protease inhibitors. Solubilized proteins were analyzed by SDS page and Western blot. GST pull down A293 cells, transfected with GST-melanophilin/Slac2-a plasmid, were lyzed in buffer containing 1% NP 40, 50 mM Tris pH8, 150 mM NaCl, and protease inhibitors. After centrifugation, supernatants were incubated with GSH-Sepharose for 3 h. B16 melanoma cells treated or not with 20 µM forskolin were lyzed in the same buffer. Soluble proteins were incubated with GSHSepharose/GST-melanophilin/Slac2-a complex for 2 h at 4°C. After three washes with the same buffer, associated proteins were eluted in 40 µl of buffer containing 3% SDS, 5% bromophenol blue, 5% β-mercaptoethanol. Finally, proteins were analyzed by SDS-PAGE and Western blot. Western blot assays For Western blot analysis, samples were subjected to SDS-PAGE and transferred to nitrocellulose membranes (Amersham Pharmacia Biotech, Buckinghamshire, UK). Membranes were saturated in a saline buffer containing 5% of nonfat dry milk and then incubated with the appropriate primary antibody diluted in the saturation buffer for 1 h. After three 10 min washes in the buffer containing 0.5% Triton X-100, 0.5% nonfat dry milk in a saline buffer, blots were incubated for 1 h with the peroxydase conjugated secondary antibody and washed again as described. The antigen-antibody complex was detected with the ECL kit (Amersham Pharmacia Biotech). RESULTS cAMP induces rapid melanosome docking at the cell periphery: Role of the actin network The regulation of melanosome transport, by differentiating agents such as αMSH and cAMP elevating agents, has not been clearly documented so far. In the present study, we used timelapse video-microscopy to study melanosome movements in B16 melanoma cells. In control conditions, we observed a bi-directional movement of melanosomes with no change in the global melanosome repartition after 70 min (Fig. 1A). Then we added forskolin 20 µM and observed the same cell for an additional 100 min period (Fig. 1B, movie 1). As soon as 14 min after forskolin addition, we noticed that melanosomes started to gather at the extremities of dendrite outgrowths. This process was even more obvious after longer exposure to forskolin. Interestingly, αMSH, a physiological cAMP-elevating agent, also induced melanosome docking at dendrite tips (Fig. 1C). Further, the docking of melanosomes induced by cAMP is a reversible phenomenon, since after forskolin removal, we observed that melanosomes left the dendrite tips and spread all over the cell (Fig. 1D, movie 2). Thus, αMSH and cAMP regulate melanosome transport to promote the accumulation of melanosomes at the dendrite extremities. Both microtubule and actin networks have been involved in melanosome transport. Thus, in a first approach, we studied the effect of actin filaments or microtubule depletion on melanosome movement and localization. DIC images showed that addition of cytochalasin D, an agent that disrupts actin filaments, had a tendency to decrease melanosome docking at the dendrite tips Page 4 of 19 (page number not for citation purposes)

(Fig. 1E, Cyto-D, white arrows). After disruption of the microtubule network by nocodazole, we observed a very modest effect that rather favors the accumulation of melanosomes at dendrite tips (Fig. 1E, Noco, black arrows). Addition of cytochalasin-D after nocodazole treatment (Fig. 1E, lower panels) or nocodazole after cytochalasin-D treatment (not shown) completely blocked melanosome transport. Taken together, these observations suggest that the actin network appears to be required for melanosome docking at dendrite tips. Further, it appears that the retrograde transport of melanosomes on microtubules is more efficient than the anterograde transport. RhoA and p160ROCK, but not protein kinase A, are involved in the redistribution of melanosomes induced by cAMP Next, we investigated the role of protein kinase A (PKA), which mediates most of the cAMP effects, in the peripheral docking of melanosomes induced by cAMP elevating agents. In basal conditions or in cells exposed to H89, an efficient PKA inhibitor, immunofluorescence studies with anti-Tyrp1 antibody showed that melanosomes were spread all over the cell. In cells treated by forskolin, we observed a reinforcement of the peripheral labeling with a clear docking of melanosomes at dendrite tips that was not affected by PKA inhibition with H89 (Fig. 2A). In the same conditions, H89 completely blocked the forskolin-induced phosphorylation of CREB and ATF1 transcription factors that are direct targets of PKA (Fig. 2B). Thus, PKA does not mediate the effect of cAMP on melanosome transport. Interestingly, it has been shown that RhoA, a key regulator of the actin cytoskeleton, is inhibited by cAMP (29, 30). Taking into account the role of actin in the transport of melanosomes, we investigated the role of this Rho GTPase in the regulation of melanosome transport by cAMP. DIC images of B16 melanoma cells exposed for 300 min to C3 toxin showed that the inhibition of Rho induced an accumulation of melanosomes at the dendrite tips (Fig. 3A, upper panels). Similarly, the inhibition of the RhoA associated kinase (p160ROCK) by Y27632 also promoted a peripheral docking of melanosomes (Fig. 3A, lower panels and movie 3). A clear accumulation of melanosomes at the dendrite tips induced by C3 toxin or Y27632 was also observed after immunofluorescence studies with anti-Tyrp1 antibody (Fig. 3B). These results indicate that the inhibition of RhoA and p160ROCK are involved in the cAMP-induced relocalization of melanosomes. cAMP does not regulate the interaction between Rab27a, melanophilin/Slac2-a, and myosin-Va Rab27a, melanophilin/Slac2-a, and myosin-Va constitute a molecular tripartite complex that allows the interaction of melanosomes with the actin network. Therefore, we studied the role of these proteins in the regulation of melanosome distribution by cAMP. Western blot analysis of Rab27a, melanophilin/Slac2-a, and myosin-Va proteins showed that the levels of Rab27a, melanophilin/Slac2-a, and myosin-Va were not modified by a short-term (2 h) treatment with forskolin (Fig. 4A). Interestingly, longer exposure to forskolin (20 h) clearly stimulated Rab27a and myosin-Va expression, while the level of melanophilin/Slac2-a remained stable. Since the accumulation of melanosomes at the dendrite tips induced by cAMP takes 85% of cells out of 150 cells analyzed in each condition. B) B16 cells were exposed to 5 µM H89, 20 µM forskolin, or both. Solubilized proteins were analyzed by SDS PAGE and Western blot with antibodies to phosphoCREB (Ser 133) (upper panel) or to CREB (lower panel).


Fig. 3

Figure 3. RhoA and p160ROCK are involved in redistribution of melanosomes induced by cyclic AMP. A) DIC images of B16 cells, before and 300 min after addition of 2mM Clostridium botulinium C3 toxin (Tox-C3) or treated by 10 µM Y27632. Three independent experiments were performed. Images are representative of ~75% of cells out of 100-120 cells analyzed in each condition (B) B16 cells were exposed to 2 mM Clostridium botulinium C3 toxin (Tox-C3) or treated by 10 µM Y27632. Melanosomes were labeled polyclonal antibody to Tyrp1 (pep1, 1/200) and goat anti-rabbit Texas Red (1/30). Bar = 2 µm (A); bar = 16 µm (B). Three independent experiments were performed. Images are representative of >75% of cells out of 150 cells analyzed in each condition.


Fig. 4

Figure 4. Cyclic AMP does not regulate the interaction between Rab27a, melanophilin/Slac2-a, and Myosin-Va. A) B16 cells were exposed for the indicated time to 20 µM forskolin. Then proteins were solubilized and analyzed by Western blot with antibody to myosin-Va (Mva, 1/100), melanophilin/Slac2-a (Mlph, 1/100) or Rab27a (1/200). B) A293 cells were transfected with GST-melanophilin/Slac2-a plasmid. Then proteins were solubilyzed and GST-melanophilin/Slac2-a was purified by incubation with GSH-Sepharose. Then solubilized proteins from control or forskolin (20 µM) treated B16 cells were incubated with GSH-sepharose/GST-melanophilin/Slac2-a complex for 2h at 4°C. After 3 washes, associated proteins were eluted and analyzed by SDS-PAGE and Western blot with antibodies to Rab-27a (1/200) and melanophilin/Slac2-a (1/100) (right panel). Total B16 cells extracts were analyzed by SDS-PAGE and Western blot with the same antibodies (left panel). C) B16 cells were exposed to 20 µM of forskolin for indicated time. Then solubilized proteins were incubated with nonimmune serum (NIS) or with anti-melanophilin/Slac2 serum (Anti-Mlph) fixed on protein A-Sepharose beads. Precipitated proteins were analyzed by SDS-PAGE and Western blot with antibodies to melanophilin/Slac2-a (Mlph) or with antibody to Rab27a. D) B16 cells were transfected with a GFP-tagged myosin-Va tail construct and exposed to 20 µM forskolin for indicated times. Total cell extracts (total) and proteins immunoprecipitated with anti-melanophilin/Slac2-a (IP) were analyzed by Western blot with anti-melanophilin/Slac2-a (Mlph) or anti-GFP antibodies (Mva-GFP). Results are representative of 3 independent experiments.


Fig. 5

Figure 5. Cyclic AMP increases interaction between melanophilin/Slac2-a and actin. A) Intact melanosomes were immunopurified from control and forskolin (20 µM) treated B16 cells as described in Methods. Total (total) or melanosome-associated proteins (pep1) were analyzed by SDS page and Western blot with antibodies to myosin-Va (1/100), melanophilin/Slac2-a (1/100), Tyrp1 (1/500), and Rab27a (1/200). Middle panel shows Western blot experiments after purification with nonimmune serum. B) B16 cells transfected with an actin-GFP encoding vector were exposed to forskolin 20 µM for indicated times. Total protein extract (total) and proteins immunoprecipitated with antimelanophilin/Slac2-a (IP) were analyzed by Western blot with anti-GFP (actin-GFP) and anti-melanophilin/Slac2-a (Mlph) antibodies. These results represent 3-5 independent experiments. Page 19 of 19 (page number not for citation purposes)