Cell Motility and the Cytoskeleton 38:397?409 (1997) Evidence for Several Roles of Dynein in Pigment Transport in Melanophores Hele?n Nilsson and Margareta Wallin Department of Zoophysiology, University of Go?teborg, Go?teborg, Sweden Melanophores are specialized cells that transport pigment granules to and from the cell center, giving animals the ability to change skin color. A kinesin-related plus-end motor has previously been shown to be responsible for pigment granule dispersion [V.I. Rodionov, F.K. Gyoeva, and V.I. Gelfand. Proc. Natl. Acad. Sci. USA. 1991, 88:4956-4960]. Here, we have microinjected a dynein antibody (70.1) into cultured cod (Gadus morhua) melanophores and used the dynein inhibitor vanadate on permeabilized melanophores in skin pieces, to examine the role of the microtubule minus-end motor dynein in these cells. Both pigment granule aggregation and maintenance of the spherical central pigment mass (CPM) were inhibited by the antibody and by vanadate. Vanadate or antibody treatment of cells with aggregated pigment did not induce pigment dispersion. However, when the antibody-injected cells were induced to disperse pigment, the pigment moved farther to the cell periphery, which resulted in a depletion of pigment in the cell center. Similar superdispersion of previously uniformly distributed pigment was also seen when the antibody was injected in melanophores with dispersed pigment. Our results demonstrate that both pigment aggregation and maintenance of the CPM are dynein-dependent processes. Our data further show that dynein is involved in the homogeneous distribution of dispersed pigment. These results suggest that both dynein and kinesin are active in keeping pigment granules dispersed throughout the cytoplasm, transporting pigment granules in opposite directions. The possibility that dynein is continuously active during both aggregation and dispersion, while kinesin might be the target for regulation, is discussed. Cell Motil. Cytoskeleton 38:397?409, 1997. r 1997 Wiley-Liss, Inc. Key words: melanophore; chromatophore; pigment; vanadate; kinesin; dynein; fish; Gadus morhua INTRODUCTION Chromatophores are regarded as excellent model systems for studies in regulation of intracellular transport of membrane-bound organelles [Haimo and Thaler, 1994] and in organization of microtubules [Rodionov and Borisy, 1997a,b]. Pigment granules can aggregate to the center of the cell to form a spherical central pigment mass (CPM), or disperse evenly throughout the cytoplasm. These movements are controlled by neurohumoral stimuli and determine the pigmentation of the animal [Fujii and Oshima, 1994]. Microtubules are important for maintaining cell morphology and for movement of pigment granules [Schliwa and Bereiter-Hahn, 1973a,b; Murphy and Tilney, 1974; Ochs, 1982; Obika and Negishi, 1985; Visconti and Castrucci, 1985; Chen and Wang, 1993; r 1997 Wiley-Liss, Inc. Nilsson et al., 1996], and McNiven and Ward  showed the first clear evidence that pigment granules were associated to microtubules by using scanning electron microscopy of lysed erythrophores. We have recently shown by immunocytochemistry that both the minus-end directed motor dynein and the Contract grant sponsors: Swedish Natural Science Research Council; Helge Ax:son Johnsons Stiftelse; Nordenskjo?ldska Swedenborgsfonden; Kungl. och Hvitfeldska Stipendieinra?ttningen; Magnus Bergvalls Stiftelse; Lars Hiertas Minne; Hierta Retzius Stipendiefond *Correspondence to: Margareta Wallin, Department of Zoophysiology, Go?teborg University, Medicinaregatan 18, S-413 90, Go?teborg, Sweden; E-mail: email@example.com Received 18 August 1997; accepted 19 August 1997 398 Nilsson and Wallin plus-end motor kinesin co-localize with both aggregated and dispersed pigment in cultured Atlantic cod (Gadus morhua) melanophores [Nilsson et al., 1996]. Another indication that these proteins are bound to melanosomes is that both dynein and a kinesin-like protein are present on Western blots of isolated melanosomes from cultured Xenopus melanophores [Rogers et al., 1997]. Microinjection of a polyclonal kinesin antibody into black tetra melanophores suppressed pigment dispersion [Rodionov et al., 1991], showing that kinesin, or a kinesin-like protein, has an important role in dispersion (for a discussion, see Rogers et al. ). The evidence for a role of dynein in aggregation is more indirect, however. Vanadate and EHNA (erythro-9[3-(2-hydroxynonyl)]adenin), inhibitors of cytoplasmic dynein ATPase in vitro [Paschal and Vallee, 1987; Shimizu et al., 1995; Niclas et al., 1996], inhibit pigment aggregation [Beckerle and Porter, 1982; Clark and Rosenbaum, 1982]. However, caution is necessary, as these substanses are known to inhibit cellular activities other than based on dynein [Kobayashi et al., 1978; Pike et al., 1978; Penningroth, 1986; Porter et al., 1987]. Many different types of microtubule motors have been found in recent years [Moore and Endow, 1996], and the lack of specific inhibitors of dynein have so far made it difficult to assess the contribution of dynein in transport processes, as the different motors cannot be pharmacologically distinguished from one another in crude systems [see Collins, 1994, for discussion]. A monoclonal antibody (mAb) to dynein (70.1) has recently been shown to inhibit dynein-dependent organization of microtubules [Heald et al., 1996], an antibody that makes it possible to determine the role of dynein in melanophores. We have microinjected 70.1 into cultured cod melanophores to study the role of dynein in four different processes: aggregation, maintenance of the CPM, dispersion, and the homogeneous distribution of melanosomes throughout the cytoplasm. We have also used vanadate on permeabilized skin melanophores for similar studies to be able to discuss the use of vanadate as a chemical tool and to make comparisons with older data. MATERIALS AND METHODS Chemicals and Antibodies Sodium orthovanadate and vinblastine (Sigma Chemical Co., St. Louis, MO) were dissolved in H2O, stored at 220蚓 in dark, and diluted 1 : 1,000 to the experimental concentrations in buffer just before use. Mowiol was from Hoechst (Frankfurt, Germany). The 70.1 antibody against chick brain cytoplasmic dynein [Steuer et al., 1990], control ascites fluid, as well as an antibody against b-tubulin, were from Sigma. Secondary antibodies conjugated with Texas red or fluorescein were from Amersham (Stockholm, Sweden). Secondary antibodies conjugated with 12 nm colloidal gold, as well as normal serum from donkey, were from Jacson Immuno Research Laboratories (West Grove, PA). Goat antimouse Ig conjugated to horseradish peroxidase (HRP) and peroxidase substrate kit were from BioRad (Richmond, CA). Melanophore Culture Melanophores were cultured according to a modified procedure described by Nilsson et al. . The cod fin explants formed a monolayer of skin cells including melanophores in medium 199 with Hank?s salts supplemented with 10% fetal calf serum (FCS), 1% antibiotic/ antimycotic solution, and 1% L-glutamine from Life Technologies (Renfrewshire, Scotland). Thereafter, the cells were further cultured in serum-free CO2-independent medium supplied with 1% antibiotic/antimycotic solution and 1% L-glutamine (Life Technologies), 5 痢/ml insulin, 5 痢/ml transferrin, and 5 ng/ml sodium selenite from Boehringer Mannheim (Mannheim, Germany) and 1 mg/ml glucose, to isolate melanophores on the coverslips and induce detachment of other skin cells. Melanophores used in the experiments had been cultured for 1?2 weeks. Microinjection of Dynein Antibody and Immunocytochemistry Microinjections were performed with micropipettes made from borosilicate glass capillar with filament OD 1.0 mm, ID 0.78 mm, and 10-cm length, from Bergstro?m Instrument AB (Go?teborg, Sweden), with Micropipette Puller P-87 from Sutter Instrument (Novato, USA). Observations of the cells were done in a Nikon diaphote microscope (Tokyo, Japan), adapted for microinjection with Eppendorf micromanipulator 5171 and transjector 5246 (Hamburg, Germany). The 70.1 antibody against cytoplasmic dynein and control ascites were transferred to a microinjection buffer made according to Rodionov et al. , by repeated filling and centrifugation at 10,000g 3 3 for 10 min at 4蚓 in Microcon 50 microconcentrator tubes from Amicon (Beverly, MA) and in a Ole Dich centrifuge (Hvidovre, Denmark). The 70.1 antibody was concentrated to an antibody concentration of 6 mg/ml and total protein concentration of approximately 270 mg/ml, or 3 mg/ml antibody and 135 mg/ml proteins, respectively. Both concentrations were tested and found to give the same effects. Only the lower concentration was further used in the experiments. The control ascites was concentrated to a total protein concentration of approximately 140 mg/ml. Working aliquots of the concentrated 70.1 antibody and control ascites were Roles of Dynein in Pigment Transport in Melanophores TABLE III. Effect of Vanadate on Pigment Dispersion in Skin Melanophores TABLE I. Effect of Vanadate on Rate of Pigment Aggregation (MI/min) Treatment Control Control Saponin Saponin Vanadate 10 然 Vanadate 10 然 Vanadate 50 然 Vanadate 50 然 Vanadate 100 然 Vanadate 100 然 ATP Mean 6 s.d. Fish (No) Sample (No) n y n y n y n y n y 1.36 6 0.30 1.49 6 0.20 1.81 6 0.30 1.53 6 0.26 0.75 6 0.10 0.75 6 0.27 0.25 6 0.18 0.23 6 0.18 0.04 6 0.08 0.00 6 0.00 2 2 2 2 2 2 2 2 2 2 8 8 8 7 8 8 8 8 8 8 Treatment Rate (MI/min) Time for onset (min) Mean 6 SD Mean 6 SD Control Saponin Vanadate 10 然 Vanadate 50 然 Vanadate 100 然 0.59 6 0.20 0.79 6 0.21 0.96 6 0.31 0.81 6 0.15 0.82 6 0.28 TABLE II. Effects of Microinjected 70.1 Antibody on Pigment Aggregation* Injected substance Control Ascites 70.1 Antibody 3 mg/ml Control Ascites 70.1 Antibody 3 mg/ml Complete aggregation Dispersion to periphery Total No cells 8 0 0 13 8 14 *Melanophores with dispersed pigment were injected and thereafter allowed to rest for 1 h before addition of noradrenaline (NA). Response to NA was determined after 8 min. Complete aggregation was defined as Melanophore Index 5 1 or 2. There was a slight peripheral aggregation in 2 antibody injected cells. Dispersion to the periphery was analysed 30 min after removal of NA. Numbers indicate No of cells. kept at 220蚓. All experiments were performed at 12蚓 in CO2-independent medium. Cells in every treatment were from at least four different fishes. Control cells were either injected with ascites fluid, microinjection buffer, touched with a microneedle or uninjected. Touched and uninjected control cells were observed simultaneously as injected cells. The responses of the cells were documented by photography. To verify that the microtubule system was unaffected by the microinjections of 70.1 antibody, ascites, or buffer, some cells were immunocytochemically stained for tubulin using methanol as fixative (not shown), according to Osborn and Weber . To confirm that the 70.1 antibody solution had spread throughout the melanophores 30 min after injection, some cells were fixed and labelled with secondary antibody (not shown). Because the cells were of different sizes, the protein concentration of the injected substanses likely variated within the cells. A clearly visible swelling of an injected cell was defined as a proper injection. For aggregation experiments, melanophores were injected in dispersed state of pigment distribution and allowed to rest for 1 h. Noradrenaline (norepinephrine, NA) was thereafter added to the medium to a final concentration of 1 然 to induce aggregation. After 8 and Sample (No) 4 4 4 4 4 16 16 16 16 16 MI 5 Melanophore Index. TABLE IV. Effects of Microinjected 70.1 Antibody on CPM and Pigment Dispersion* Response to NA Fish (No) 9.1 6 4.0 6.2 6 2.7 3.1 6 2.3 3.5 6 1.0 2.2 6 0.8 y 5 included, n 5 not included, MI 5 Melanophore Index. Injected substance 399 Deformed Dispersion Total CPM Dispersion to periphery No cells 1 12 16 12 0 7 16 12 *Melanophores with noradrenaline (NA) induced aggregated pigment were injected and allowed to rest in 1 然 NA for 30 min. Deformed central pigment masses (CPM) were determined after rest. Thereafter dispersion was induced. Cells with dispersed pigment and dispersion to cell periphery were determined after 30 min. Numbers indicate No of cells. 25 min in NA, the pigment distribution in the cells was analyzed. The cells were thereafter replaced in fresh medium without NA to initiate dispersion. For experiments on maintenance of spherical CPM and dispersion, cells were incubated in 1 然 NA for 15 min to obtain spheric CPMs before the cells were injected. Injection was made in close approximity to the CPM. After microinjection, the cells were allowed to rest in the presence of NA for 30 min, and the amounts of deformed or spheric CPMs were analyzed. Thereafter the cells were replaced in fresh medium to initiate dispersion. Cultured melanophores with dispersed or aggregated pigment were also used for immunocytochemical staining against dynein according to Nilsson et al. , with the 70.1 antibody. Immunoblotting Characterization of the dynein antibody (70.1) immunoreactivity in Atlantic cod tissue, was performed on a vertical 7.5% sodium dodecyl sulfate (SDS) polyacrylamide gel (LKB, Bromma, Sweden), according to Laemmli . Cod brains were homogenized in 1 : 2 v/v cold Pipes-MgSO4-EGTA-buffer as described in Nilsson et al. , and centrifuged at 10.000g for 10 min at 4蚓. The crude supernatant was boiled in sample buffer (BioRad) for 5 min, and 100 痞 was loaded on each lane of the gel for electrophoresis and subsequent Western blotting as described before [Nilsson et al., 1996]. 400 Nilsson and Wallin Fig. 1. Skin melanophores were incubated in 10, 50, or 100 然 vanadate for 10 min in lysis buffer with 0.02% saponin. Controls were incubated in lysis buffer with or without saponin. One-half of the samples were also treated with 100 然 ATP. Noradrenaline (NA) was thereafter added to 1 然 to induce pigment aggregation; the process was measured as melanophore index (MI) every 30 sec. Graph shows mean rates of aggregation (MI/min) in each treatment. Statistical analysis on slopes of MI/min to 3 min and on three samples/fish, showed that pigment aggregation was inhibited by 10, 50, or 100 然 vanadate treatment (ANOVA: F4,10 5 133.1, P , 0.05 and SNK-test p , 0.05). No influence of ATP was found. Vanadate and Vinblastine Treatment of Skin Explants 58triphosphate (ATP) (Sigma). Thereafter, NA was added, and the rate of aggregation was measured every 30 s for 6 min. Pigment movement was measured according to the melanophore index (MI), where 1 5 completely aggregated and 5 5 fully dispersed [Hogben and Slome, 1931]. In order to investigate the effects of microtubule and dynein inhibitors on the spherical CPM, melanophores on skin pieces were induced to aggregate pigment granules completely by addition of 1 然 NA in medium. The microtubule disrupting agent vinblastine was thereafter added for 2 h of incubation at a concentration of 3 然 in the presence of NA. Controls were incubated in 1 然 NA only. The skin explants were thereafter fixed in ice-cold methanol for 6 min and rinsed in PBS 2 3 5 min. The effect on maintenance of the spherical shape of CPMs was determined as the amount of deformed CPMs in percentage of the amount of total CPMs counted from each well with skin explants. Cultured cells were treated similarily, and immunocytochemically stained by antibodies against tubulin, to verify disruption of microtubules (not shown). To reveal whether dynein is involved in maintenance of the CPM, vanadate was added at different concentrations. Explants were induced to aggregate melanophore pigment completely, and thereafter incubated with 1 然 NA with 10, 50, or 100 然 vanadate for 10 min in lysis buffer with saponin. Controls were treated with lysis buffer with or without saponin. Thereafter the explants were fixed as previously and the amount deformed CPMs measured. Experiments on pigment translocation and maintenance of circular CPM were carried out on melanophores on skin pieces of about 1 mm2 prepared from the dorsal fins using a diaphote Nikon microscope. Skin pieces were allowed to adhere to the bottom of sterile plastic culture dishes with four wells in 1 drop of medium for about 1 hr before addition of medium 199 and further incubation at 12蚓. The experiments were performed the day after at 12蚓 in medium, or lysis buffer at pH 7.0 made according to Grundstro?m et al. . For mild permeabilization of cell membranes to vanadate, 0.02% saponin was included in the lysis buffer. Every treatment was performed on four fish, and from every fish measurements were done on melanophores on four skin pieces in different wells (total amount of measurements: 16 per treatment). Aggregation of pigment granules was induced by addition of 1 然 NA in the medium. Explants that did not respond by aggregation of the melanophore pigment granules within 3 min were discharged. For aggregation experiments, pigment granules were allowed to redisperse in fresh medium to the next day. Explants were thereafter incubated in 10, 50, or 100 然 vanadate for 10 min in lysis buffer with saponin. The time of treatment was standardized to 10 min, as longer time in 0.02% saponin was found to inhibit NA-induced pigment aggregation (not shown). Controls were treated with lysis buffer with or without saponin. One-half of the samples were also incubated with 100 然 adenosine Roles of Dynein in Pigment Transport in Melanophores 401 Fig. 2. Characterization of the 70.1 dynein antibody in cod. Immunoblot of a crude cod brain extract (a). Cod brain proteins were separated on 7.5% SDS-Page and blotted to a nitrocellulose membrane. The membrane was probed with 70.1 mAb against dynein three light chains of ,70 kD from chicken. Lane 1, Coomassie blue stained gel; lane 2, nitrocellulose membrane stained with 70.1 mAb. Distribution of 70.1 mAb immunoreactivity in cultured melanophores with dispersed (b) and aggregated (d) pigment. c, e: Light micrographs of the two cells, respectively. Bar 5 10 然. For dispersion experiments, melanophores were first induced to aggregate pigment for 4 min by addition of 1 然 NA. They were thereafter incubated for 10 min in 10, 50, or 100 然 vanadate together with 1 然 NA in lysis buffer with saponin, or in lysis buffer only with or without saponin together with 1 然 NA. Measurements of dispersion was initiated by removal of NA by rinsing the samples with lysis buffer including only vanadate, or lysis buffer without vanadate for the controls, and measurements were made every min for 22 min. In experiments with vanadate and vinblastine, measurements were ??blind?? assays in which the observer was unaware of treatment. Determination of differences in amount of spherical shapes of CMPs and differences in 402 Nilsson and Wallin Fig. 3. Effects of microinjection of 70.1 mAb on pigment granule aggregation in cultured melanophores. Melanophore touched with a microneedle (a?d), injected with ascites fluid (e?h), and injected with 70.1 mAb (i?l). a, e, i: Micrographs showing cells before injection. f, j: At 1 h after injection. e, g, k: After 8 min in NA. d, h, l: After 30 min in fresh medium without NA. k: Arrow, no pigment aggregation after 8 min in NA. l: Arrows, pigment translocated to the cellular periphery after replacement of the cells into medium without NA. Bar 5 50 痠. Roles of Dynein in Pigment Transport in Melanophores 403 rates of pigment dispersion was performed with a factorial analysis of variance (ANOVA) with treatment as fixed factor, individual fish as a nested factor and perecentage deformed CPMs or slope of MIs/min in the interval of translocation with maximum 22 min as dependent variables. Differences in rates of pigment aggregation were after random removal of data from one skin explant/fish, analysed with a factorial ANOVA with vanadate and ATP treatments as fixed factors, individual fish as a nested factor and slope of MIs/min in the interval of translocation with a maximum of 3 min as the dependent variable. Data from the studies of maintenance of CPM were transformed by arcsin as recommended for proportions [Sokal and Rohlf, 1995]. All data were tested for homogeneity of variance with Cochran?s C test with a 5 0.05 [Snedecor and Cochran, 1989]. Multiple comparisons of means were made with Student?Newman Keul?s test. Differences in time for onset of dispersion, measured as mean time spent as MI 5 1, were examined with nonparametric Kruskal?Wallis and Tukey-type multiple comparisons as data were noncontinuous [Zar, 1984]. Level for significance was P , 0.05. RESULTS Vanadate Inhibits Aggregation of Pigment in Skin Melanophores The dynein inhibitor vanadate was used as one tool to investigate whether dynein is involved in the aggregation of pigment in melanophores on skin pieces from cod. Aggregation was induced by addition of 1 然 NA in lysis buffer, and dispersion by removal of NA by rinsing with lysis buffer. All vanadate treatments (in the presence or absence of ATP) decreased the aggregation rates in a doseresponse-like appearance compared to controls (Fig. 1, Table I). In the presence of 100 然 vanadate, the aggregation was almost completely inhibited. The same effects was seen after 3 min in the presence of 50 然 vanadate. Vanadate also inhibited aggregation at a concentration of 10 然, but not as efficient as the higher concentrations. Microinjection of Dynein Antibody Inhibits Pigment Granule Aggregation in Cultured Melanophores In our study, we have used a mAb against cytoplasmic chick dynein light chains (70.1), characterized by Fig. 4. Effects of microinjection of 70.1 mAb on pigment granule aggregation in a cultured melanophore in comparison with simultaneously observed uninjected melanophores. a: Micrograph of melanophores with dispersed pigment. b: Cells are photographed just after injection of 70.1 mAb in the melanophore (arrow). c: Cells after 30 min rest after injection. d: Cells at 20 min after addition of 1 然 NA. Bar 5 50 痠. 404 Nilsson and Wallin Fig. 5. Light micrographs of representative skin explants with melanophores with normally dispersed pigment (a), and melanophores with completely aggregated pigment after incubation in 1 然 NA in medium, and thereafter incubated in buffer, vinblastine, or vanadate (c?e). A representative control explant was incubated in lysis buffer with 0.02% saponin and NA for 10 min (b). Skin explant incubated in 3 mM vinblastine in medium with NA for 2 h (c). Explants incubated in lysis buffer with 0.02% saponin, NA, and 50 然 vanadate (d), and lysis buffer with 0.02% saponin, NA and 100 然 vanadate (e). Note the deformed shape of the melanophore pigment masses after treatment with vinblastine (b) and with vanadate (d, e,), in contrast to control (b). Bars 5 100 mm. Steuer et al. , which has recently been shown to be a powerful tool in studies of dynein-mediated processes [Heald et al., 1996]. To show that the 70.1 antibody reacts with cod dynein and is not species-specific, a crude cod brain extract was immunoblotted with the antibody. The antibody reacted specifically with three bands of the appropriate Mr, of which the two upper were very closely aligned(Fig. 2a). Cultured melanophores with dispersed or aggregated pigment were analyzed immunocytochemically with the 70.1 antibody. The antibody labelling showed a punctate distrubution throughout cells with dispersed granules (Fig. 2b), and an accumulated staining of the center of the cells with aggregated pigment (Fig. 2d). To determine whether dynein is involved in pigment granule aggregation, the 70.1 antibody was microin- jected into cultured cod melanophores with dispersed pigment (Table II). Aggregation was induced 1 h after injection by addition of 1 然 NA and the melanophores were observed during 25 min. A complete aggregation was seen within 6 min for control cells, which either were touched with a microneedle, injected with ascites fluid (Fig. 3a?c,e?g), or uninjected (not shown) . These cells aggregated their pigment slightly slower than melanophores that were untreated (not shown). Microinjection of the dynein antibody induced a change in pigmentation of the dispersed cells already during the 1-h rest before NA treatment. The pigment extended farther to the periphery, leaving a spot with a lower density of pigment in the central part of the cell, and a higher density of pigment in the peripheral parts (Fig. 3j). This was in clear contrast to Roles of Dynein in Pigment Transport in Melanophores Fig. 6. Melanophores with completely aggregated pigment were incubated with 1 然 NA in the presence of 10, 50, or 100 然 vanadate, respectively, for 10 min in lysis buffer with 0.02% saponin. Controls were treated with NA and lysis buffer with or without saponin. Mean amount of percentage deformed CPMs of total CPMs were calculated from 16 skin samples in every treatment. Error bars are s.d. The effects of 50 and 100 然 vanadate were significant, as indicated by stars (ANOVA: F4,15 5 16.5, P , 0.05, F15,60 5 16.8, P , 0.05 and SNK-test P , 0.05). the touched control and ascites injected cells, where a slight aggregation was induced by the treatment (Fig. 3b,f). When NA was added to the antibody-injected cell, the aggregation was inhibited and no CPM was formed (Fig. 3k), in contrast to control cells (Fig. 3c,g). In Figure 4, only one of several melanophores has been injected with dynein antibody, clearly showing the difference between uninjected control cells and dynein mAbinjected cells during rest and NA-induced aggregation. Dispersion of pigment (Table II), which was induced by replacing the cells in fresh medium without NA, was unaffected in touched or ascites injected melanophores (Fig. 3d,h), showing that these cells behaved normally. In the dynein antibody- injected cells, which were already in a dispersed state caused by inhibition of aggregation, a further translocation of pigment to the peripheral parts of the cells was seen upon removal of NA stimulation, leading to a very unusual pigment pattern (Fig. 3l). Our results indicate that dynein not only is responsible for aggregation, but also in keeping the pigment uniformly dispersed. Vinblastine and Vanadate Treatments Induce Deformed Shapes of the CPMs in Skin Melanophores To determine whether microtubules are involved in maintenance of the spherical form of CPM, melanophores on skin pieces with pigment aggregated to punctate with 1 然 NA were treated with the microtubule 405 disrupting agent vinblastine at 3 mM in medium with NA for 2 h (Fig. 5c). The results showed that microtubule disruption led to a significant (P , 0.05, ANOVA: F1,6 5 216.8) deformation of 65% of the spherical CPMs, compared with 19% in control skin pieces incubated in medium with only NA. To investigate whether dynein is involved in maintenance of the CPM, skin melanophores with pigment aggregated to punctate with 1 然 NA, were also treated with 10, 50, and 100 然 vanadate in lysis buffer with 0.02% saponin and NA for 10 min. Vanadate induced a significant deformation of the CPMs in a dose-responsedependent way, where 100 然 vanadate induced 82% deformed CPMs, 50 然 55% and 10 然 34% (Fig. 6B) . A significant difference in percentage deformed CPMs was induced by 50 and 100 然 vanadate, compared to the controls with (20%) or without saponin (22%). Vanadate treatment inhibited only pigment aggregation, as no significant difference in dispersion rate was found between vanadate-treated samples, and controls with or without saponin (Fig. 7, Table III). Even if the dispersion rate was unaffected, the time needed for initiation of dispersion decreased significantly in the presence of 100 然 vanadate, compared to controls in lysis buffer (Table III). Microinjection of Dynein Antibody Inhibits Maintenance of Spherical CPMs in Cultured Melanophores To evaluate further the role of dynein in maintenance of the spherical shape of the CPM, the 70.1 antibody against dynein was microinjected into cultured melanophores with NA-induced aggregated pigment granules. After injection, the cells were allowed to rest for at least 30 min in presence of NA to maintain the aggregated state of pigment. Control cells, which were either uninjected or injected with ascites fluid, maintained their spherical CPMs after rest in NA (Fig. 8a,b,d,e). In contrast, microinjection of the dynein antibody induced a loosening of the spherical CPM, as observed after the rest period (Fig. 8g,h). The results show that dynein is important for maintenance of the spherical CPM. Dispersion of pigment was induced by replacing the cells in fresh medium without NA for 20 min (Fig. 8c,f,i, (Table IV)). Pigment moved to a uniformal distribution throughout the control cells (Fig. 8 c,f). In contrast, a further translocation of pigment to the peripherial parts of the cells was seen in the antibody injected cells (Fig. 8i). This was even more visible after prolonged time (not shown). DISCUSSION In this study we have for the first time shown that dynein is involved in three different processes in melanophores; aggregation, maintenance of the central pigment ma ss, and homogeneous distribution of pigment granules 406 Nilsson and Wallin Fig. 7. Skin melanophores were incubated in NA to induce pigment aggregation. Skin explants were then incubated in 10, 50, or 100 然 vanadate with NA for 10 min in lysis buffer. The explants were thereafter rinsed in lysis buffer with vanadate and in absence of NA to induce pigment dispersion. Measurement of dispersion as MI was done every min. The graph shows the mean rate of dispersion (MI/min) in each treatment. No difference in slopes of time/MI between treatments was found (ANOVA: F4,15 5 1.42, P . 0.05). Time for onset of dispersion (detemined as time until MI 2 is reached) differed between treatments (Kruskal?Wallis: P , 0.05). Multiple comparisons detected a quicker onset of dispersion after treatment with 100 然 vanadate compared to controls in lysis buffer (Tukey type: P , 0.05). throughout the cells. The 70.1 antibody against chicken dynein has proved an important tool with broad species specifity for functional studies of dynein in both Xenopus egg extracts [Heald et al., 1996] and cod melanophores. Vanadate, which we also have used, is well known to inhibit pigment granule aggregation, either by microinjection [Beckerle and Porter, 1982] or by permeabilization of the cell membrane [Clark and Rosenbaum, 1982; Grundstro?m et al., 1985; Negishi et al., 1985; Ogawa et al., 1987; Oshima et al., 1990], but its specificity is under debate [Penningroth, 1986; Collins, 1994]. Here we have shown that similar inhibitory effects were achieved with microinjection of the 70.1 antibody and with the addition of vanadate to permeabilized skin melanophores in the studies on aggregation and maintenance of the spherical CPM, while only the use of the dynein antibody could show that dynein has a role in keeping dispersed pigment evenly distributed. Our results on inhibition of aggregation is also similar to the older data with vanadate, making it possible to discuss dynein as a general mediator for pigment aggregation. The mechanism behind maintenance of the spherical CPM has not been examined before, but a lack of microtubule dependency in this process has been discussed (Schliwa and Bereiter-Hahn, 1975; Chen and Wang, 1993). Our results show that disruption of microtubules with vinblastine, inhibition of dynein with vanadate, or microinjection of 70.1 antibody into cod melanophores with aggregated pigment, resulted in loosening of the CPM. This shows clearly that microtubules and dynein are responsible for maintenance of the spherical CPM. Loosening of the CPM did not induce pigment dispersion, implying that an opposite-directed motor must be responsible for the dispersion process. This is in agreement with Rodionov et al. , who showed that a kinesin-related protein was involved in melanophore pigment dispersion in black tetra. The normal dispersion process from the cell center can be explained by two different mechanisms. One possibility is that dynein is inactivated shortly after loosening of the CPM and kinesin is induced to translocate pigment granules towards the cell periphery until dynein is activated again. Together with kinesin, they could accomplish the shuttling and uniformly spreading of the pigment granules. The similar rates of pigment dispersion in skin melanophores treated with vanadate and control melanophores support this model. Another possibility is that dynein is continuously active during aggregation and dispersion, and kinesin is the target of activation during dispersion, as speculated in a review by Haimo and Thaler , leading to overriding of dynein by kinesin dynein during dispersion. It has been suggested from in vitro studies that kinesin could determine the direction of movement of other vesicles by overriding the activity of dynein [Vale et al., 1992; Muresan et al., 1996]. The mechanism of action is unclear, but differences in time that the motors spend attached to microtubules during the ATPase cycle, and in affinity for vesicle membranes, were discussed. This model is in agreement Roles of Dynein in Pigment Transport in Melanophores 407 Fig. 8. Effects of microinjection of 70.1 mAb on maintenance of the spherical central pigment mass and pigment granule dispersion in cultured melanophores. The cells were preincubated in 1 然 NA to induce pigment aggregation before injection. Uninjected melanophore (a?c), injected with ascites fluid (d?f), and injected with 70.1 mAb (g?i). d, g: Cells before injection. e, h: At 30 min after injection. f, i: At 20 min after transfer to fresh medium without NA. k: Arrow, deformed shape of the central pigment mass in mAb injected cell after 1-h rest in NA. i: Arrow, pigment translocated to the periphery of the cell after transfer of the cell into medium without NA to induce pigment dispersion. Bars 5 50 痠. with the irregular pigment movement observed during dispersion [Green, 1968; Murphy and Tilney, 1974]. Dispersed pigment granules are usually homogeneously distributed in melanophores and shuttle short distances to and fro [Green, 1968], suggesting that some force antagonize the transport of pigment completely to the peripheral parts of the cell. This is in clear contrast to the aggregation process, where all pigment granules are transported rapidly to the central part of the cell. The reason for the evenly distributed pigment granules has hitherto been unclear. If only kinesin is active during dispersion, the pigment granules would move to the periphery of the cells and change the color of the animal. Color changes of animals involve aggregation or dispersion of pigment throughout the cells and do not involve pigment dispersion to the cell periphery [for review, see Haimo and Thaler, 1994]. Our data that microinjection of the dynein antibody into dispersed melanophores induced a transport of melanosomes to the cell periphery suggest that the minus end motor dynein not only is active during aggregation, but also in the process of keeping pigment granules evenly distributed in the cell. Vanadate also induced a dispersion of pigment to the periphery (unpublished observations) but prolonged saponin permeabilization of the melanophores affected the melanophores similarly, making it difficult to draw clear conclusions. 408 Nilsson and Wallin distribution of dispersed melanosomes in cod melanophores. In light of our data, the possibility exists that dynein is continuously active and kinesin the target for regulation. ACKNOWLEDGMENTS Fig. 9. Hypothetical depiction of how pigment granule distribution is regulated in melanophores. In the dispersed state of pigment distribution (dark gray melanophore), both dynein and kinesin are active and try to transport pigment granules in opposite directions, leading to uniformly spread pigment granules throughout the melanophore. During aggregation (indicated by the white melanophore), and the subsequent maintenance of the spherical central pigment mass, kinesin is inactivated by a noradrenaline-induced mechanism, while dynein remains active. In the dispersion process, kinesin either becomes active again and dynein could be inactivated until a uniform distribution of dispersed pigment is complete, or kinesin overrides a continuously active dynein. Pigment dispersion requires cAMP, while dephosphorylation seems to be a crucial component regulating pigment aggregation [Rozdzial and Haimo, 1986; Thaler and Haimo, 1990]. Vanadate is known to inhibit phosphatases [Cyboron et al., 1982; VanEtten et al., 1974]. The effect of vanadate might therefore either be a direct effect on dynein, or an inhibition of phosphatase activity, with the result that the melanophores remain dispersed, or hyperdispersed. The 70.1 antibody is specific to dynein, making it a more useful tool in crude systems. Based on the present results, our current hypothesis is that dynein has an active role in aggregation, in formation and keeping of a spherical CPM, and in the homogeneous distribution of pigment throughout the cell, while kinesin only is active in the dispersion process (Fig. 9). To test the hypothesis with continuously active dynein, one would have to block kinesin completely. Rodionov et al.  found that microinjection of kinesin antibodies into melanophores suppressed dispersion, but the inhibition of kinesin did not lead to pigment aggregation in those cells. However, the inhibition was not complete, and one cannot exclude the possibility that kinesin still might be able to override dynein. Further studies focusing on pigment dispersion are therefore needed to clarify the regulatory steps in this process. In conclusion, our results show that dynein is responsible for pigment aggregation and maintenance of the spherical CPM and is involved in the homogeneous We thank Elisabeth Norstro?m for introducing the microinjection technique in our laboratory, microinjection experiments, and other excellent technical assistance; Malin Ba?ckstro?m for assisting vanadate experiments; and Dr. Mattias Sko?ld for statistical help. This work was supported by grants from the Swedish Natural Science Research Council, Helge Ax:son Johnsons Stiftelse, Nordenskjo?ldska Swedenborgsfonden, Kungl. och Hvitfeldska stipendieinra?ttningen, Magnus Bergvalls stiftelse, Lars Hiertas Minne, and Hierta Retzius Stipendiefond. REFERENCES Beckerle, M.C., and Porter, K.R. (1982): Inhibitors of dynein activity block intracellular transport in erythrophores. Nature 295:701? 703. Chen, J.-S., and Wang, S.-W. (1993): The role of microtubules in pigment translocation in goldfish xanthophores. Arch. Histol. Cytol. 56:451?458. Clark, T.G., and Rosenbaum, J.L. (1982): Pigment particle translocation in detergent-permeabilized melanophores of Fundulus heteroclitus. Proc. Natl. Acad. Sci. USA 79:4655?4659. Collins, C.A. (1994): Dynein-based organelle movement. In Hyams, J.S., and Lloid, C.W. (eds.): ??Microtubules.?? New York: Wiley-Liss, pp. 367?380. Cyboron, G.W., Vejins, M.S., and Wuthier, R.E. (1982): Activity of epiphyseal cartilage membrane alkaline phosphatase and the effects of its inhibitors at physiological pH. J. Biol. Chem. 257:4141?4146. Fujii, R., and Oshima, N. (1994): Factors influencing motile activities of fish chromatophores. Adv. Comp. Environ. Physiol. 20:1?54. Green, L. (1968): Mechanism of movements of granules in melanocytes of Fundulus heteroclitus. Proc. Natl. Acad. Sci. USA 59:1179?1185. Grundstro?m, N., Karlsson, J.O.G., and Andersson, R.G.G. (1985): The control of granule movement in fish melanophores. Acta. Physiol. Scand. 125:415?421. Haimo, L.T., and Thaler, C.D. (1994): Regulation of organelle transport: Lessons from color change in fish. BioEssays 16:727?733. Heald, R., Tournebize, R., Blank, T., Sandaltzopoulos, R., Becker, P., Hyman, A., and Karsenti, E. (1996): Self-organization of microtubules into bipolar spindles around artificial chromosomes in Xenopus egg extracts. Nature 382:420?425. Hogben, L.T., and Slome, D. (1931): The pigmentary effector system. VI. The dual character of endocrine coordination in amphibian colour change. Proc. R. Soc. Lond. Ser. B 108:10?53. Kobayashi, T., Martensen, T., Nath, J., and Flavin, M. (1978): Inhibition of dynein ATPase by vanadate, and its possible use as a probe for the role of dynein in cytoplasmic motility. Biochem. Biophys. Res. Commun. 81:1313?1318. Laemmli, U.K. (1970): Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature 227:680?685. Roles of Dynein in Pigment Transport in Melanophores McNiven, M.A., and Ward, J.B. (1988): Calcium regulation of pigment transport in vivo. J. Cell Biol. 106:111?125. Moore, J.D., and Endow, S.A. (1996): Kinesin proteins: a phylum of motors for microtubule-based motility. BioEssays 18:207?219. Muresan, V., Godek, C.P., Reese, T.S., and Schnapp, B.J. (1996): Plus-end motors override minus-end motors during transport of squid axon vesicles on microtubules. J. Cell Biol. 135:383?397. Murphy, D.B., and Tilney, L.G. (1974): The role of microtubules in the movement of pigment granules in teleost melanophores. J. Cell Biol. 61:757?779. Negishi, S., Fernandes, H.R.C., and Obika, M. (1985): The effects of dynein ATPase inhibitors on melanosome translocation within melanophores of the medaka, Oryzias latipes. Zool. Sci. 2:469? 478. Niclas, J., Allen, V.J., and Vale, R.D. (1996): Cell cycle regulation of dynein association with membranes modulates microtubulebased organelle transport. J. Cell Biol. 133:585?593. Nilsson, H., Rutberg, M., and Wallin, M. (1996): Localization of kinesin and cytoplasmic dynein in cultured melanophores from the Atlantic cod, Gadus morhua. Cell Motil. Cytoskel. 33:183? 186. Obika, M., and Negishi, S. (1985): Effects of hexylene glycol and nocadazole on microtubules and melanosome translocation in melanophores of the medaka Oryzias latipes. J. Exp. Zool. 235:55?63. Ochs, L.R. (1982): The role of microtubules in cell shape and pigment distribution in spreading erythrophores. Eur. J. Cell Biol. 28:226?232. Ogawa, K., Hosoya, H., Yokota, E., Kobayashi, T., Wakamatsu, Y., Ozato, K., Negishi, S., and Obika, M. (1987): Melanoma dynein: Evidence that dynein is a general ??motor?? for microtubule-associated cell motilities. Eur. J. Cell Biol. 43:3?9. Osborn, M., and Weber, K. (1982): Immunofluorescence and immunocytochemical procedures with affinity purified antibodies: Tubulin-containing structures. In Wilson, L. (ed.): ??Methods in Cell Biology.?? Vol. 24. New York: Academic Press, pp. 97?132. Oshima, N., Inagaki, H., and Manabe, T. (1990): Evidence for involvement of dynein?tubulin system in pigment aggregation within tilapia melanophores. Comp. Biochem. Physiol. 96A:517? 523. Paschal, B.M., and Vallee, R.B. (1987): Retrograde transport by the microtubule-associated protein MAP 1C. Nature 330:181?183. Penningroth, S.M. (1986): Erythro-9-[3-(2-hydroxynonyl)]adenine and vanadate as probes for microtubule-based cytoskeletal mechanochemistry. Methods Enzymol. 134:477?487. Pike, M.C., Kredich, N.M., and Snyderman, R. (1978): Requirement of S-adenosyl-L-methionine-mediated methylation for monocyte chemotaxis. Proc. Natl. Acad. Sci. USA 75:3928?3932. Porter, M.E., Scholey, J.M., Stemple, D.L., Vigers, G.P.A., Vale, R.D., Sheetz, M.P., and McIntosh, J.R. (1987): Characterization of the microtubule movement produced by sea urchin egg kinesin. J. Biol. Chem. 262:2794?2802. 409 Rodionov, V.I., and Borisy, G.G. (1997a): Microtubule treadmilling in vivo. Science 275:215?218. Rodionov, V.I., and Borisy, G.G. (1997B): Self-centering activity of cytoplasm. Nature 386:170?173. Rodionov, V.I., Gyoeva, F.K., and Gelfand, V.I. (1991): Kinesin is responsible for centrifugal movement of pigment granules in melanophores. Proc. Natl. Acad. Sci. USA 88:4956?4960. Rogers, S.L., Tint, I.S., Fanapour, P.C., and Gelfand, V.I. (1997): Regulated bidirectional motility of melanophore pigment granules along microtubules in vitro. Proc. Natl. Acad. Sci. USA 94:3720?3725. Rozdzial, M.M., and Haimo, L.T. (1986): Bidirectional pigment granule movements of melanophores are regulated by protein phosphorylation and dephosphorylation. Cell 47:1061?1070. Schliwa, M., and Bereiter-Hahn, J. (1973a): Pigment movements in fish melanophores: Morphological and physiological studies. II. Cell shape and microtubules. Z. Zellforsch. 147:107?125. Schliwa, M., and Bereiter-Hahn, J. (1973b): Pigment movements in fish melanophores: Morphological and physiological studies. III. The effects of colchicine and vinblastine. Z. Zellforsch. 147:127?148. Schliwa, M., and Bereiter-Hahn, J. (1975): Pigment movements in fish melanophores: morphological and physiological studies. V. Evidence for a microtubule-independent system. Cell Tissue Res. 158:61?73. Shimizu, T. (1995): Inhibitors of dynein ATPase and ciliary or flagellar motility. In Dentler, W., and Witman, G. (eds.): ??Methods in Cell Biology.?? Vol. 47. New York: Academic Press, pp. 497?501. Snedecor, G.W., and Cochran, W.G. (1989): ??Statistical methods.?? Ames, IA: Iowa State University Press. Sokal, R.R., and Rohlf, F.J. (1995): ??Biometry.?? New York: W.H. Freeman. Steuer, E.R., L. Wordeman, T.A. Schroer, and Sheetz, M.P. (1990): Localization of cytoplasmic dynein to mitotic spindles and kinetochores. Nature 345:266?268. Thaler, C.D., and Haimo, L.T. (1990): Regulation of organelle transport in melanophores by calcineurin. J. Cell Biol. 111:1939? 1948. Van Etten, R.L., Waymack, P.P., and Rehkop, D.M. (1974): Transition metal ion of enzyme-catalyzed phosphate ester displacement reactions. J. Am. Chem. Soc. 95:6782?6785. Vale, R.D., Malik, F., and Brown, D. (1992): Direction instability of microtubule transport in the presence of kinesin and dynein, two opposite polarity motor proteins. J. Cell Biol. 119:1589?1596. Visconti, M.A., and Castrucci, A.M.L. (1985): Microtubule and microfilament disrupting drugs and melanosome migration in melanophores of Papiliochromis ramirezi (Cichlidae, Teleostei). An. Acad. Brasil. Cienc. 57:233?237. Zar, J.H. (1984): ??Biostatistical analysis.?? Englewoods Cliffs, NJ: Prentice-Hall.