Research Article 2661 RhoB regulates endosome transport by promoting actin assembly on endosomal membranes through Dia1

Mar Fernandez-Borja1,*, Lennert Janssen1, Desiree Verwoerd1, Peter Hordijk2 and Jacques Neefjes1 1The Netherlands Cancer Institute, Division of Tumour Biology, Plesmanlaan 121, 1066 CX Amsterdam, The Netherlands 2Sanquin Research at CLB, Department of Molecular Cell Biology, Plesmanlaan 125, 1066 CX Amsterdam, The Netherlands *Author for correspondence (e-mail: [email protected])

Accepted 16 March 2005 Journal of Cell Science 118, 2661-2670 Published by The Company of Biologists 2005 doi:10.1242/jcs.02384

Summary Rho GTPases are crucial regulators of the actin endosomes by activated RhoB. Dia1 is required for the cytoskeleton and they play a role in the control of formation of the actin coat around endosomes downstream membrane trafficking. In contrast to the close family of RhoB, connecting membrane trafficking with the members RhoA and RhoC, RhoB localises to endosomes regulation of actin dynamics. and delays epidermal growth factor receptor traffic. Here, we show that activated RhoB induces the peripheral Supplementary material available online at distribution of endosomes, which align along subcortical http://jcs.biologists.org/cgi/content/full/118/12/2661/DC1 actin stress fibres and are surrounded by an actin coat. The Diaphanous-related formin, Dia1, is recruited to Key words: RhoB, Dia1, Endocytosis, Transport

Introduction and Rusk, 2003). Correspondingly, RhoB localises to the Cell morphological changes, such as the extension of cytoplasmic face of endosomal membranes (Adamson et al., lamellipodia during migration, require the coordinated 1992; Robertson et al., 1995) and has been suggested to regulation of membrane trafficking, and actin and microtubule regulate endosome traffic. Specifically, activated RhoB has cytoskeletons. Long-range bidirectional transport of vesicles been shown to delay the transport of internalized epidermal Journal of Cell Science between the plasma membrane and the juxtanuclear area growth factor (EGF)-receptor to lysosomes (Gampel et al., occurs along microtubules. Endocytic vesicles are transported 1999) by an as yet unknown mechanism. RhoB is highly towards the microtubule minus-end located at the perinuclear homologous to the well-studied Rho family member RhoA microtubule-organizing centre (MTOC) where they fuse (83% identity). Due to this high homology, both isoforms with the sorting/recycling endosomes that reside in this exhibit a similar ability to bind specific effector proteins in region (Aniento et al., 1993). However, before they reach vitro. However, it is unclear whether this also applies in vivo. microtubules, endocytic vesicles must cross the cortical actin For example, the serine/threonine protein kinase PRK1 was cytoskeleton underlying the plasma membrane (Aschenbrenner identified as a downstream effector of RhoA (Amano et al., et al., 2004). Actin polymerisation has been shown to play a 1996), but recent work shows that PRK1 is recruited to role in endocytosis (Lamaze et al., 1997; Fujimoto et al., 2000; endosomes by RhoB and appears to be activated by this reviewed in Merrifield, 2004), but its role in post-endocytic GTPase (Mellor et al., 1998). Similarly, the Rho-binding transport is less clear. Endocytic vesicles could be transported domain of the mammalian Diaphanous-related formin (DRF), on subcortical actin microfilaments by a myosin motor until Dia1, interacts with recombinant RhoA, B and C in vitro they reach the microtubular network. An additional mechanism (Watanabe et al., 1999). Dia1 contains an N-terminal Rho- implicates the phosphatidylinositol-4 phosphate 5 kinase- binding domain and two C-terminally conserved FH domains stimulated polymerisation of a Listeria-like actin comet on raft- responsible for actin filament growth (Watanabe et al., 1997; enriched vesicles by the WASP-Arp2/3 pathway (Rozelle et al., Wasserman, 1998) as well as for Arp2/3-independent actin 2000). This actin tail is suggested to contribute to endocytic assembly (Pruyne et al., 2002; Evangelista et al., 2003). transport by propelling newly formed pinosomes towards the Interestingly, Dia1 has been localised to endosomes in HeLa cell interior (Merrifield et al., 1999). cells (Tominaga et al., 2000), which suggests that it may Rho GTPases are critical regulators of actin dynamics and regulate endocytic traffic downstream of RhoB. Notably, have been involved in the control of endocytosis, although the another endosomal Rho GTPase, RhoD, was shown to regulate connection between these two activities is not completely endosome transport through hDia2C, a novel splice variant of understood. Activated RhoA, Rac and Cdc42 localize to the the human DFR2 (Murphy et al., 1996). plasma membrane and consistently have been implicated in the Here, we have investigated the effect of activated RhoB on regulation of receptor internalization (reviewed in Symons endocytosis and the actin cytoskeleton. Activated RhoB 2662 Journal of Cell Science 118 (12)

promotes the polymerisation of an actin coat around Wandinger-Ness (University of New Mexico, Albuquerque, NM). endosomes and the association of these vesicles to subcortical Rabbit polyclonal anti-myosin IIA heavy chain antibody was actin cables, effectively inhibiting further endosomal transport. purchased from Biomedical Technologies Inc. Mouse monoclonal The Rho effector protein Dia1 is recruited to endosomes by antibody anti-EGF receptor was from BD-Transduction Labs. activated RhoB and is critically involved in the assembly of Secondary antibodies labelled with FITC (Dako), Texas Red actin on the vesicle membrane. This exemplifies a mechanism (Molecular Probes) or Cy5 (Jackson Labs) were used. F-actin was detected with Alexa 568-labelled phalloidin (Molecular Probes). of control of endocytosis by a Rho GTPase through the regulation of actin polymerisation on membranes by a DRF protein. Microinjection and immunofluorescence Cells were microinjected on a heated xy-stage of an inverted Olympus IX70 microscope equipped with an Eppendorf Micromanipulator Materials and Methods InjectMan device. Expression plasmids were injected in the nuclei of Cell lines and treatments 100-150 cells at a concentration of 0.1 mg.ml–1 in microinjection The human melanoma cell line Mel JuSo and HeLa cells were buffer (120 mM potassium glutamate, 40 mM KCl, 1 mM MgCl2, 1 cultured in Iscoves medium with 10% fetal calf serum. A Mel JuSo mM EGTA, 0.2 mM Ca Cl2, 10 mM Hepes, 40 mM mannitol, pH 7.2). cell line stably expressing EGFP-actin was generated by calcium For image recordings of living cells, Texas Red labelled dextrane 70 phosphate transfection with an expression plasmid containing Myc- kDa (Molecular Probes) was added as microinjection marker. After GFP-human β actin cDNA (kindly provided by P. Stein, Harvard microinjection, cells were incubated for 4 to 5 hours to allow protein Medical School, Boston, MA). A cell line stably expressing EGFP- expression and were subsequently fixed with 3.7% formaldehyde in FYVE (EEA1∆1-1256) was generated by calcium phosphate PBS for 10 minutes, permeabilized with 0.1% Triton X-100 and transfection of Mel JuSo cells with the corresponding expression immunostained. Primary and secondary antibodies and phalloidin plasmid (kindly provided by H. Stenmark, The Norwegian Radium were diluted in PBS containing 0.5% bovine serum albumin. Confocal Hospital, Oslo, Norway). Expressing cells were selected in Iscoves images were obtained with a Leica TCS SP confocal laser-scanning medium supplemented with 2 mg ml–1 G418 (Gibco). HeLa cells were microscope. transfected with Lipofectamine 2000 (Invitrogen) following the instructions of the manufacturer. Rho kinase activity was blocked by the incubation of cells with 10 µM Y-27632 (Sigma) for 1 hour before Fluorescence live imaging microscopy fixation. Time-lapse recordings of GFP-FYVE expressing cells were done by collecting 36 images with a time interval of 15 seconds on a Leica TCS SP CLSM. The images recorded were maximally projected and Fluid-phase and receptor mediated endocytosis encoded green. This projection image was then combined with a red- To follow receptor-mediated endocytosis of transferrin, cells were coloured image of the cells at t=0 (initial position of vesicles) as serum starved for 1 hour prior to the incubation with 50 µg ml–1 Texas previously described (Wubbolts et al., 1996). Red-labelled human transferrin (Molecular Probes) for 15 minutes. Fluid phase endocytosis was analysed by using sulforhodamine 101 (SR101, Molecular Probes) as a fluid-phase endocytosis marker Results

Journal of Cell Science (Wubbolts et al., 1996). Cells were incubated with SR101 for 15 Activated RhoB induces endosome dispersion and cell minutes, washed to remove non-internalised marker, and chased for elongation up to 60 minutes prior to fixation with 3.7% formaldehyde. Previous studies have shown that RhoB localises to the endosomal compartment (Adamson et al., 1992; Robertson et Expression plasmids al., 1995) and causes the dispersion of EGF receptor- Myc-RhoBG14V cDNA was transferred from the pEXV-myc- containing endosomes (Gampel et al., 1999). Consistently, we RhoBG14V plasmid (Chardin et al., 1988), kindly provided by A. Hall found that the constitutively active GTPase-deficient mutant (ICRF, London, UK), into the EcoRI site of pcDNA3 (Invitrogen). RhoBG14V largely co-localised with the early endocytic marker pcDNA3-myc-RhoB wild type (RhoBwt) was generated by site- EEA1 on endosomes that were relocated from their usual directed mutagenesis using QuickChange (Stratagene) following the G14V juxtanuclear location to the cell periphery and appeared smaller manufacturer’s instructions. pcDNA3-myc-RhoA was kindly than in non-injected cells (Fig. 1b, compare injected cell with provided by W. Moolenaar (The Netherlands Cancer Institute, neighbouring control non-injected cells). In addition, Amsterdam, The Netherlands). pEGFP-mDia1 full length and the G14V ∆N3 and N1 deletion mutants were kindly provided by S. Narumiya RhoB induced F-actin polymerisation in thin parallel stress (Kyoto University Faculty of Medicine, Kyoto, Japan) and are fibres running along the longitudinal axis of the cells (Fig. 1c). described elsewhere (Watanabe et al., 1999). pcDNA3-myc-Rab5Q79L Endosomes aligned along the RhoB-induced actin cables (Fig. (Stenmark et al., 1994) and pEGFP-FYVE were kindly provided by 1a-c, arrowheads). The wild-type form of RhoB (RhoBwt) also H. Stenmark (Norwegian Radium Hospital, Oslo, Norway). co-localised with EEA1 on vesicles, but had no effect on the actin cytoskeleton or the distribution of endosomes, which localised to the juxtanuclear area as in control cells (Fig. 1d- Antibodies f). We observed that (RhoBwt)/EEA1 positive endosomes had Myc- and HA-tagged proteins were detected with the mouse 9E10 and polymerized actin on discrete sites on their membrane 12CA5 monoclonal antibodies, respectively. Rabbit polyclonal (arrowheads in Fig. 1d-f and Fig. 3). These results suggest that antibody sc-119 anti-RhoB was purchased from Santa Cruz G14V Biotechnology. Rabbit polyclonal anti-EEA1 antibodies were kindly RhoB induces the association of endosomes to actin cables provided by H. Stenmark (Norwegian Radium Hospital, Oslo, preventing their transport to the cell centre. Norway) (Simonsen et al., 1998). Mouse monoclonal antibody 4F11 To test for the specificity of the effects induced by anti-Rab5 (Stein et al., 2003) was generously provided by A. RhoBG14V, we expressed an activated mutant of RhoA Dia1 regulation of endosome transport 2663

(RhoAG14V). RhoA is highly homologous to RhoB and their respective effector loops differ only in two residues. Unlike RhoBG14V, RhoAG14V localised to the cell cytoplasm (Fig. 1g) and did not alter endosome distribution (Fig. 1h). RhoAG14V induced actin polymerisation and the formation of parallel stress fibres (Fig. 1i), but it had markedly different effects on cell shape compared to RhoBG14V (compare Figs 1c and 1i, supplementary material Fig. S1). To analyse this, we measured the length of the perpendicular long and short axes of RhoBG14V and RhoAG14V expressing cells and compared them with control cells. The average lengths of the long and short axes in control cells were 47.32±7.61 µm and 25.70±5.93 µm (n=30), respectively. RhoBG14V expressing cells exhibited an elongated, spindle-like shape; the average lengths of the long and short axes were 62.95±8.89 µm and 18.81±3.05 µm (n=30), respectively. Notably, the length of the axes parallel to the stress G14V fibres was reduced in RhoA Fig. 1. Effect of activated RhoB on endosome distribution and the actin cytoskeleton. Cells were expressing cells, being 35.85±9.55 microinjected with myc-RhoBG14V (a-c), myc-RhoBwt (d-f) or myc-RhoAG14V (g-i). After 4 hours µm on average, while the axes incubation, cells were fixed and stained for myc, the endosomal protein EEA1 and F-actin. RhoB perpendicular to the stress fibres proteins largely co-localised with EEA1 on endosomes (a,b and d,e, arrowheads) while RhoAG14V was longer, 31.62±8.87 µm (n=25). was cytosolic (g). EEA1-positive endosomes were located at the cell centre in control non-injected Journal of Cell Science The ratio between the lengths of cells and in cells expressing RhoBwt (e) or RhoAG14V (h). In cells expressing RhoBG14V, EEA1- positive endosomes were peripheral and often appeared aligned on actin filaments (b,c, the long and short axes in µ every condition was 3.43±0.59 arrowheads). Bars: 10 m. for RhoBG14V expressing cells; 1.22±0.46 for RhoAG14V expressing cells; and 1.96±0.52 for control cells (Fig. 2). These location depends on microtubule association (not shown). We results show that, despite the almost complete identity between also observed the formation of circular actin structures the effector loops of the two GTPases, RhoB affects endosome apparently associated to actin fibres in live GFP-actin Mel JuSo distribution and cell morphology in a different manner than cells expressing RhoBG14V. Immunodetection of RhoB in these RhoA. cells showed that these actin structures surrounded RhoBG14V- positive vesicles (Fig. 3b, arrowheads), confirming the results obtained after phalloidin staining for F-actin. Together, these RhoB induces the assembly of an actin coat on results suggest that endogenous RhoB activity controls endosomes endosomal interactions with actin, which transiently occur We next investigated in more detail the apparent association during endocytosis. of RhoBG14V-positive endosomes with actin cables. High magnification microscopic analysis of RhoBG14V expressing cells revealed that RhoBG14V/(EEA1) positive endosomes were RhoB causes reduced motility of endosomes coated with F-actin and apparently associated to F-actin In Mel JuSo cells the transferrin receptor-positive recycling filaments (Fig. 3d-f, arrows). In contrast, juxtanuclear EEA1 compartment and the lysosomes reside in the juxtanuclear area positive endosomes of control cells (non-injected cells in Fig. (Wubbolts et al., 1996). Our data suggest that the association 1) and RhoBwt/(EEA1)-positive endosomes showed F-actin of endosomes to subcortical actin fibres in RhoBG14V- associated to a discrete site on their membrane (Fig. 3a-c, expressing cells prevents their transport to the juxtanuclear arrows) as in control cells and were never encapsulated by area. We examined the fate of molecules internalised either by actin. Treatment of the cells with nocodazole induced the receptor-mediated or fluid phase endocytosis. Receptor- dispersion of these endosomes, indicating that their perinuclear mediated endocytosis was assessed by incubating cells with 2664 Journal of Cell Science 118 (12)

4 endosomes to actin fibres may prevent their transfer to

tio microtubules and thus impair endosome motility. To confirm

a 3

r this, we examined endosome dynamics in a stable cell line s 2 expressing the EEA1 domain responsible for endosome xe

a G14V 1 targeting tagged with GFP (GFP-FYVE). RhoB cDNA

cell was introduced in GFP-FYVE expressing cells together with a 0 control RhoBG14V RhoAG14V 70 kDa Texas Red-labelled dextrane to mark injected cells. Vesicle movement was followed by fluorescence live imaging Fig. 2. The activated forms of RhoB and RhoA have opposite effects confocal microscopy. The images recorded during the time- on cell morphology. The long and short perpendicular axes of control lapse experiment (36 images with a time interval of 15 cells (n=30), RhoBG14V (n=30) and RhoAG14V (n=25) expressing seconds) were maximally projected and encoded green. This cells were measured and the ratio between the two axes ±s.e.m. was image was then combined with a red-coloured image of the calculated. cells at t=0 (initial position of vesicles) (Fig. 4c; see also supplementary material Movie 1). Vesicle movement appears transferrin-Texas Red. The ligand was endocytosed and as green tracks on the projection (Fig. 4c, arrowheads). The reached the endosomes labelled for RhoB, while it average length of the tracks formed by vesicles in control cells concentrated in juxtanuclear vesicles in control non-injected during approximately 9 minutes was of 3.6±0.8 µm, ranging cells, where the recycling compartment resides (Fig. 4a). from 2.6 to 4.5 µm. GFP-FYVE-positive vesicles in RhoBG14V- Subsequently, fluid-phase endocytosis was studied by feeding injected cells exhibited Brownian-like motion (Fig. 4c, cells with the fluid-phase marker sulforhodamine 101 (SR101) arrows). Occasionally, short tracks were observed with an for 15 minutes, followed by a 60 minutes chase period, in average length of 1.3±0.3 µm, ranging from 0.8 to 1.6 µm. which SR101 reaches the lysosomal compartment (Wubbolts These results demonstrate that RhoBG14V reduces endosomal et al., 1996). SR101 was internalised and accumulated in motility. Again, the peripheral location of endosomes marks vesicles that remained in the periphery in RhoBG14V expressing the RhoBG14V-injected cell, and contrasts the juxtanuclear cells, while in control non-injected cells SR101-containing location in control cells (Fig. 4c). These results suggest that vesicles concentrated in the juxtanuclear area (Fig. 4b). These RhoBG14V inhibits microtubule-based endosome motility and results suggest that RhoB is implicated in the regulation of an transport to the juxtanuclear area as a result of endosome endocytic step posterior to the internalization at the plasma association to actin fibres. membrane and prior to the transport of endocytic vesicles to To further confirm the inhibition of microtubule-based the cell centre. endosome motility by RhoBG14V, we studied homotypic fusion Since long-distance transport of endosomes is known to between endosomes. Homotypic fusion of early endosomes occur along microtubules, our data imply that association of requires endosome movement along microtubules and is regulated by the small GTPase Rab5 (Aniento et al., 1993; Stenmark et al., 1994; Nielsen et al., 1999). Expression of a Journal of Cell Science constitutively active Rab5 (Rab5Q79L) causes massive homotypic fusion resulting in the formation of large endosomes (Stenmark et al., 1994). The association of endosomes to subcortical actin cables in RhoBG14V- expressing cells may thus prevent microtubule-based transport and inhibit endosome homotypic fusion. To test this we co-expressed Rab5Q79L and RhoBG14V. When expressed alone, Rab5Q79L localised to

Fig. 3. Activated RhoB induces the assembly of an F-actin coat around early endosomes, which appears associated to actin filaments. (A) High magnification microscopical analysis of endosomes of cells expressing myc-RhoBwt (a-c) or myc- RhoBG14V (d-f) and stained for Myc, EEA1 and F-actin. Both RhoBwt and RhoBG14V localised to EEA1-positive endosomes (a,b and d,e). RhoBwt-positive endosomes showed polymerized actin on a discrete site on their membranes (arrows in a-c and merge). RhoBG14V-positive endosomes were coated with F-actin and apparently associated with actin cables (arrows in d-f and merge). Bars: 2.5 µm. (B) Cells expressing GFP-actin were microinjected with myc-RhoBG14V plasmids and stained for Myc. High magnification images of the RhoBG14V-positive vesicles (indicated by arrowheads) show that these vesicles were coated with actin and apparently associated with actin fibres. Bar: 10 µm. Dia1 regulation of endosome transport 2665

Fig. 4. Activated RhoB inhibits transport of receptor-mediated and fluid-phase endocytosed cargo to the juxtanuclear area and impairs endosome motility. The effect of RhoBG14V on receptor-mediated endocytosis was studied by incubating cells with 50 µg ml–1 Texas Red- labelled human transferrin for 15 minutes. Myc-RhoBG14V was detected with 9E10 and FITC-labelled secondary antibodies (a). Fluid-phase endocytosis was analysed by 15 minutes incubation with the fluid-phase endocytic marker sulforhodamine 101 (SR101), followed by a 60 minutes chase (b). The effect of RhoBG14V on endosome motility was analysed on live cells transfected with pEGFP-FYVE (c). Myc-RhoBG14V cDNA was microinjected together with a 70 kDa Texas-Red-dextrane to mark the injected cells. Endosome dynamics was observed by time-lapse fluorescence confocal microscopy and analysed by projection of the time-lapse images (36 images with a time interval of 15 seconds) where vesicle movement is seen as green tracks. The projected image was then combined with a red-coloured image corresponding to t=0. In the RhoBG14V expressing cell (asterisk) vesicles form much shorter tracks (c, arrows) than in control non-injected cells (c, arrowheads). Nuclei of cells microinjected with myc-RhoBG14V are marked with an asterisk. Nuclei of control non-injected cells are marked with a circle. Bars: 10 µm.

juxtanuclear large endosomes, probably due to the stimulation from RhoA. We investigated whether ROCK was involved in of endosome homotypic fusion (Fig. 5a). However, when co- the effects induced by RhoBG14V by incubating cells with the expressed with RhoBG14V, Rab5Q79L was found on a large ROCK inhibitor Y-27632. This drug did not inhibit the effects number of small peripheral vesicles, partially co-localizing with of RhoBG14V on endosome distribution (Fig. 6a). Furthermore, RhoB (Fig. 5b, arrows). Together, these data demonstrate that diffuse thin F-actin fibres remained after Y-27632 treatment of RhoB regulates an actin-dependent endocytic step following cells expressing RhoBG14V (Fig. 6b). Finally, a constitutively internalisation from the plasma membrane and preceding the active mutant, ROCK∆3, did not cause endosome dispersion microtubule-dependent Rab5-regulated transport. or actin assembly on endosomes (not shown). The presence of diffuse F-actin fibres after Y-27632 incubation has been Journal of Cell Science previously described in cells expressing a constitutively active Dia1 is a downstream effector of RhoB mutant of the Diaphanous-related formin Dia1 (Watanabe et Next, we sought to identify the effector molecule implicated in al., 1999). Therefore, we investigated the possibility that Dia1 the regulation of endosome transport downstream of RhoB. mediated the RhoB effects by expressing GFP-tagged full- Due to the high homology between the effector-binding regions length Dia1 or the GFP-tagged truncation mutants Dia1-N1 (N- of RhoA and RhoB, both GTPases may be able to interact with terminal Rho-binding domains; dominant negative form) and the same effector proteins in solution. However, RhoA is Dia1-∆N3 (C-terminal FH1 and FH2 domains; constitutively largely cytosolic while RhoB is targeted to the endosomal active form) (Watanabe et al., 1999). RhoBG14V was able to compartment. Therefore, the intracellular distribution of these recruit full-length Dia1 to endosomes (Fig. 7b). Notably, Dia1- GTPases may determine their respective roles by the specific N1 partially co-localised with RhoB and effectively blocked activation of similar pathways on different intracellular the RhoBG14V effects on the distribution of endosomes, which compartments. now located to a juxtanuclear position (Fig. 6c,d and Fig. 8f). ROCK proteins mediate stress fibre formation downstream In the presence of Dia1-N1, RhoBG14V-positive vesicles were

Fig. 5. Activated RhoB prevents Rab5-dependent endosome fusion. Cells were microinjected with the cDNA for the activated mutant of Rab5 (myc- Rab5Q79L) alone (a) or in combination with myc- RhoBG14V (b,b′). Rab5 was detected with the mouse 4F11 monoclonal antibody and RhoB with the rabbit sc-119 polyclonal antibody. Expression of myc-Rab5Q79L induced the formation of large endosomes that localised to the juxtanuclear area (a). Co-expressed myc-Rab5Q79L and myc- RhoBG14V co-localised in smaller endosomes scattered over the cell (b,b′, arrows). Bars: 10 µm. 2666 Journal of Cell Science 118 (12) therefore, is critically involved in the regulation of endosomal cargo transport.

Discussion Rho GTPases are essential regulators of filamentous actin reorganisation and have been implicated in the control of endocytosis. However, the link between these two functions of Rho GTPases remains unclear. The results presented in this study show that an activated form of the endosomal GTPase RhoB promotes the polymerisation of an actin coat around early endosomes and induces their association to subcortical actin fibres. Activated RhoB does not inhibit internalization from the plasma membrane, in contrast to other RhoGTPases (Symons and Rusk, 2003). However, endosomes containing the internalised cargo remain in the cell periphery, whereas they reach the juxtanuclear area in control cells. Fig. 6. Involvement of Rho-effector proteins in the regulation of endosome Dispersed endosomes show reduced motility and appear transport downstream of RhoB. Cells expressing myc-RhoBG14V showed to be retained on actin filaments. This might be due to the peripheral RhoBG14V-positive endosomes (a) and diffuse actin filaments (b) crosslinking of the F-actin around endosomes with the after 1h treatment with the ROCK inhibitor Y-27632. Co-expression of actin cytoskeleton by myosin II. Indeed, myosin II myc-RhoBG14V and a GFP-tagged dominant negative Dia1-N1 Dia1 (Dia1- redistributes and co-localises with actin-coated N1) blocked the dispersion of RhoB-positive vesicles (c). Dia1-N1 partially G14V µ RhoB -positive vesicles (supplementary material Fig. co-localised with RhoB (merge of c and d). Bars: 10 m. S2) although we have no direct evidence for a role of myosin II in this process. Importantly, the overexpression of wild type RhoB had no effect on endosome transport, no longer encapsulated by an actin coat but showed F-actin at suggesting that endogenous RhoB transiently inhibits a discrete site on their membranes, like endosomes in control endocytosis under the regulation of the GTPase cycle. cells (Fig. 8h,i). This strongly suggests that Dia1 is required The small GTPase Rab5 is known to regulate homotypic for RhoB-induced actin assembly on endosomes and thus fusion of endosomes and a constitutively active Rab5 mutant crucial for RhoB regulation of endosome traffic. was shown to drive massive endosome fusion in a microtubule- To test whether Dia1 was sufficient to mediate RhoB effects dependent manner resulting in the formation of unusually large on endosomes, a constitutive active mutant of Dia1 (Dia1- endosomes (Stenmark et al., 1994). A consequence of reduced ∆N3) was expressed. Dia1-∆N3 was recruited to vesicles endosome motility would be the impairment of the Rab5- Journal of Cell Science scattered over the cell that were coated with F-actin (Fig. 9c,d dependent endosome fusion. RhoBG14V-positive endosomes and high magnification insets). Dia1-∆N3 also induced the are smaller, suggesting that endosome fusion is inhibited. dispersion of EEA1-positive endosomes (Fig. 9a,b). Similar to Corroborating this observation, we show that a constitutively RhoBG14V, Dia1-∆N3 induced cell elongation and the active mutant of Rab5 co-localises with RhoB on endosomes formation of parallel thin actin cables (Fig. 9b) (Watanabe et but it is no longer able to drive endosome fusion. Altogether, al., 1999). these results suggest that activated RhoB induces retention of RhoB has been previously implicated in endocytosis of the endosomes on actin fibres and prevents their transfer onto EGF receptor in HeLa cells (Gampel et al., 1999). To test microtubules, inhibiting fusion and further transport of whether Dia1 is involved in the regulation of this pathway, endosomal cargo to lysosomes. This might explain the HeLa cells were transfected with RhoBG14V, Dia1-N1 or Dia1- previously reported delay of internalised EGF-receptor ∆N3 or with both RhoBG14V and Dia1-N1. Transfection of transport to lysosomes by activated RhoB (Mellor et al., 1998). RhoBG14V or the constitutively active Dia1-∆N3 resulted in the Despite the high homology between RhoA and RhoB, their accumulation of non-degraded EGF-receptor, which was cellular effects on endosome distribution and cell shape differ reversed by co-transfection of Dia1-N1 (supplementary largely. This may be due to their different intracellular material Fig. S4). Together, these results indicate that the formin protein Dia1 functions downstream of activated RhoB to promote actin polymerization on endosomes and,

Fig. 7. Activated RhoB recruits mDia1 to endosomes. GFP- mDia1 (Dia1) localises to the cytoplasm of injected cells (a). Co-expression of myc-RhoBG14V and GFP-mDia1 results in the recruitment of Dia1 to peripheral RhoBG14V-positive vesicles (b,b′). The area within the square shows multiple vesicles positive for both Dia1 and RhoB. Bars 10 µm. Dia1 regulation of endosome transport 2667

Fig. 8. Activated RhoB effects on endosome distribution and actin coat assembly on endosomes are blocked by the dominant-negative Dia1-N1 mutant. GFP-mDia1-N1 (Dia1-N1) expression did not cause alteration of endosome distribution or the actin cytoskeleton (a-c). Dia1-N1 was recruited to endosomes by RhoBG14V (compare cell co-expressing both proteins with cell expressing only Dia1-N1 (d,e)). Co-expression of Dia1-N1 with myc-RhoBG14V blocked the effects of RhoBG14V on the distribution of EEA1-positive endosomes, which localised to the juxtanuclear area as in non-injected cells (d-f). Inset in 8f shows a magnification of the merged picture of the injected cell in 8d-f showing white-blue vesicles positive for Dia1- N1 (green), RhoB (red) and EEA1 (blue). Dia1-N1 also blocked the assembly of an actin coat around RhoBG14V-positive vesicles which now showed F-actin on a discrete site on their membranes (g-i). Inset in i shows vesicles labelled for RhoB (green) and F-actin (red). Bars: 10 µm.

localisations: mainly cytosolic for RhoA and endosomal for RhoB. Thus, RhoA and RhoB may activate similar pathways, although of different subcellular locations, which may result in different biological outcomes. Alternatively, the subcellular location may determine the accessibility of the GTPase to certain effectors. For example, ROCK is critically involved in RhoA-regulated polymerisation of actin filaments downstream of RhoA (Ishizaki et al., 1996; Kimura et al., 1996), to endosomes in HeLa cells (Tominaga et al., 2000). Here we however, we found that ROCK does not have a role in the effect show that Dia1 is recruited to endosomes by activated RhoB of RhoB on endosome transport by treatment of RhoBG14V and that a deletion mutant lacking the FH domains (Dia1-N1) expressing cells with the specific ROCK inhibitor Y27632 suppresses RhoB effects on endosomes, strongly suggesting (Uehata et al., 1997). Furthermore, cells expressing activated that Dia1 is a downstream effector of RhoB. Consistently, we RhoB showed diffuse F-actin fibres after treatment with Y- show that activated RhoB induces cell elongation and the Journal of Cell Science 27632. A similar effect has been described in cells expressing formation of parallel thin actin fibres, a phenotype also activated Dia1 (Watanabe et al., 1999), which is suggestive of reported for constitutively active mDia1 (Ishizaki et al., 1996). an involvement of Dia1 in the RhoB-induced effects. The fact that Dia1-N1 inhibits actin coat assembly on Dia1 belongs to the Diaphanous-related formin (DRF) endosomes and prevents their dispersion suggests that the subfamily characterized by the presence of an N-terminal Rho- binding domain (RBD) and a C-terminal Diaphanous autoinhibitory domain (DAD) (Alberts, 2001). Binding of activated Rho to the RBD causes the activation of DRF proteins by relieving the intramolecular interaction with the C-terminal DAD. This activation unmasks the two conserved C-terminal formin-homology (FH) domains that are implicated in the control of actin polymerisation (Evangelista et al., 2003). Dia1 has been shown to be a downstream effector of RhoA and to be required for RhoA-induced actin stress fibre formation, microtubule stabilization and SRF activation (Watanabe et al., 1997; Watanabe et al., 1999; Nakano et al., 1999; Tominaga et al., 2000; Geneste et al., 2002; Palazzo et al., 2004). Interestingly, Dia1 was shown to interact with RhoA as well as with RhoB in vitro (Watanabe et al., 1999) and to localise

Fig. 9. An active mutant of Dia1 mimics RhoB effects. Expression of GFP-mDia1-∆N3 (Dia1-∆N3) induces the dispersion of EEA1- positive endosomes and the formation of thin parallel actin fibres (a,b, asterisk marks injected cell). Dia1-∆N3 is targeted to vesicles and induces the assembly of an actin coat around them (c,d and high magnification insets). Bars: 10 µm. 2668 Journal of Cell Science 118 (12)

Fig. 10. Model for the regulation of endosome Plasma membrane transport by RhoB-Dia1. (a) RhoB-GDP is RhoB-GDP internalised from the plasma membrane and activated a on endosomes by an exchange factor, for example RhoB-GDP Vav2 (Gampel and Mellor, 2002). (b) RhoB-GTP Vav2 RhoB-GDP recruits Dia1 and PRK1 (Mellor et al., 1998) to PRK1 endosomes. RhoB activates Dia1 by relieving the Endosome RhoB-GTP b Endosome autoinhibitory intramolecular interaction between the ‘STOP’ R ‘GO’ 1 BD DAD domain and the Rho-binding domain (RBD). H RBD FH1 d (c) Activated Dia1 interacts with unknown factors on F Dynein 2 DAD FH2 endosomes and promotes actin assembly/elongation H on endosomes through FH2-dependent actin Active Dia1 F D Inactive Dia1 nucleation and/or FH1-dependent recruitment of A c Microtubule profilin-ATP-actin. Endosomal actin associates with D F-actin actin fibres that run underneath the plasma membrane, by an as yet unknown mechanism, Plasma membrane preventing the transfer of endosomes to microtubules and inhibiting further transport. (d) Finally, inactivation of RhoB prevents further actin polymerisation, the actin coat depolymerises and endosomes can then bind to microtubules via the minus-end directed motor dynein for transport toward the microtubule minus-end.

RhoB-induced actin coat is necessary for the association of Overexpression of RhoB was shown to cause the endosomes with subcortical actin fibres and for the arrest of accumulation of endocytosed EGF in peripheral vesicles and endosome motility. In our experiments expression of the Dia1- to delay EGF-receptor traffic (Gampel et al., 1999). N1 mutant alone did not affect the F-actin content of cells or Furthermore, PRK1 was found to mediate the effects of RhoB endosome distribution (Fig. 8a-c). This indicates that, under on EGF-receptor traffic. Our results show that RhoB acts our experimental conditions, this mutant does not inhibit through Dia1 to induce actin polymerization on endosomes and endogenous RhoA-dependent pathways. The discrepancy with their association to actin cables. In addition, dominant negative previous studies that show the disappearance of actin stress Dia1 reverses the inhibition of EGF-receptor degradation fibres upon Dia1-N1 expression might be due to the shorter caused by activated RhoB in HeLa cells. Thus, it is possible time lapse between cDNA transfection/microinjection and cell that Dia1 and PRK1 co-operate in the regulation of endosome fixation in our experiments (4 to 5 hours) compared to others traffic downstream of RhoB, Dia1 through the polymerization (Watanabe et al., 1999; Krebs et al., 2001). The role of Dia1 of actin on endosomes and PRK1 through the phosphorylation in endosome transport was further corroborated by the of a yet unknown target. expression of Dia1-∆N3, an active mutant lacking the RBD, RhoB has been proven to be a negative regulator of cellular which localised to endosomes and induced the same effects as transformation (Liu et al., 2001; Jiang et al., 2004a; Jiang et activated RhoB. The fact that Dia1-∆N3 localised to al., 2004b) and a crucial target for the anti-tumour drugs Journal of Cell Science endosomes despite lacking the RBD suggests that the C- farnesyl-transferase inhibitors (FTIs) (Prendergast, 2001). FTIs terminal part of Dia1 interacts with additional factors on deplete cells from the farnesylated form of RhoB with the endosomes (Olson, 2003). Interestingly, another endosomal subsequent enrichment of the geranylgeranylated form, which Rho GTPase, RhoD, has been shown to regulate endosome reduces proliferation and survival of transformed cells (Du et transport through the DRF protein hDia2C, but not through al., 1999; Du and Prendergast, 1999; Liu et al., 2000). Notably, mDia1 (Gasman et al., 2003). Similarly to RhoB, the in HeLa cells geranylgeranylated-RhoB localises to endosomes expression of an activated mutant of RhoD (RhoDG26V) inhibits and inhibits EGF receptor traffic in a greater extent when endosome motility and induces the association of endosomes constitutively activated (Wherlock et al., 2004). Interestingly, with actin filaments. However, unlike RhoBG14V, RhoDG26V inactive RhoB is localised to the plasma membrane and expression causes loss of stress fibres and only induces actin cytoplasm (supplementary material Fig. S3) and it is activated filaments when co-expressed with hDia2C (Murphy et al., on endosomes by Vav2 (Gampel and Mellor, 2002). 1996; Gasman et al., 2003). At any rate, both Gasman et al. Furthermore, wild type RhoB localises to the plasma and our study suggest that endosomal Rho GTPases (RhoB and membrane and to juxtanuclear endosomes (supplementary RhoD) may stabilize endosome association with the actin material Fig. S3), suggesting that endogenous RhoB cycles cytoskeleton through the recruitment and activation of a DRF between these two compartments. Taken altogether, a model protein. In addition to this, our results provide a possible for the biological function of RhoB emerges (Fig. 10) in which mechanism for the control of endosome traffic by RhoB-Dia1 RhoB is internalised from the plasma membrane and activated through the assembly of actin on endosomes, in agreement on endosomes by the Vav2 exchange factor. Subsequently, with the established function of Rho GTPases as promoters of activated RhoB recruits and activates Dia1 at endosomes, actin polymerisation. Furthermore, RhoB has a short half-life exposing the Dia C-terminal domain which then interacts with and its synthesis is upregulated by several growth factors and other endosomal proteins (such as actin or actin- stress stimuli (Prendergast, 2001). Thus, although apparently polymerisation factors). Dia1 induces the assembly/elongation both RhoB and RhoD have similar functions concerning the of actin filaments on endosomes and stabilises the association regulation of endosome traffic, the RhoB-Dia1 pathway may of endosomes with actin filaments. F-actin is a dynamic operate only under certain conditions while the RhoD-hDia2C polymer, which length is determined by the respective rates of pathway may act in a more constitutive manner. polymerisation and depolymerisation. In our model, Dia1 regulation of endosome transport 2669

inactivation of RhoB stops actin polymerisation while GTP-binding protein Rho binds to and activates a 160 kDa Ser/Thr protein depolymerisation continues, resulting in the removal of the kinase homologous to myotonic dystrophy kinase. EMBO J. 15, 1885-1893. actin coat from endosome membranes. Finally, endosomes Jiang, K., Delarue, F. L. and Sebti, S. M. (2004a). EGFR, ErbB2 and Ras but not Src suppress RhoB expression while ectopic expression of RhoB bind to microtubules via the minus-end directed motor protein antagonizes oncogene-mediated transformation. Oncogene 23, 1136-1145. dynein for further transport toward the centre of the cell Jiang, K., Sun, J., Cheng, J., Djeu, J. Y., Wei, S. and Sebti, S. (2004b). Akt (Hirokawa, 1998). mediates Ras downregulation of RhoB, a suppressor of transformation, In conclusion, our results suggest that the RhoB-Dia1 invasion, and metastasis. Mol. Cell. Biol. 24, 5565-5576. pathway induces the transient association of endosomes with Kimura, K., Ito, M., Amano, M., Chihara, K., Fukata, Y., Nakafuku, M., Yamamori, B., Feng, J., Nakano, T., Okawa, K. et al. (1996). Regulation actin fibres and delays endosome maturation into later of myosin phosphatase by Rho and Rho-associated kinase (Rho-kinase). compartments, further establishing the role of DRFs in the Science 273, 245-248. regulation of endocytosis. The tandem RhoB-Dia1 exemplifies Krebs, A., Rothkegel, M., Klar, M. and Jockusch, B. M. (2001). a novel pathway linking membrane trafficking with the Characterization of functional domains of mDia1, a link between the small GTPase Rho and the actin cytoskeleton. J. Cell Sci. 114, 3663-3672. regulation of actin dynamics. Lamaze, C., Fujimoto, L. M., Yin, H. L. and Schmid, S. L. (1997). The actin cytoskeleton is required for receptor-mediated endocytosis in G14V We thank A. Hall for the RhoB cDNA, W. Moolenaar for the mammalian cells. J. Biol. Chem. 272, 20332-20335. pcDNA3-RhoAG14V plasmid; P. Stein for the GFP-actin expression Liu, A., Du, W., Liu, J. P., Jessell, T. M. and Prendergast, G. C. (2000). plasmid; S. Narumiya for the GFP-mDia1 cDNAs; H. Stenmark for RhoB alteration is necessary for apoptotic and antineoplastic responses to the anti-EEA1 antibodies, pEGFP-FYVE and the pcDNA3-myc- farnesyltransferase inhibitors. Mol. Cell. Biol. 20, 6105-6113. Rab5Q79L plasmids; and A. Wandinger-Ness for the anti-Rab5 Liu, A. X., Rane, N., Liu, J. P. and Prendergast, G. C. (2001). RhoB is antibody. We thank F. van Leeuwen and J. Collard for critically dispensable for mouse development, but it modifies susceptibility to tumor reading the manuscript and Lauran Oomen and Lenny Brocks for the formation as well as cell adhesion and growth factor signaling in excellent assistance with confocal microscopy. This work was transformed cells. Mol. Cell. Biol. 21, 6906-6912. Mellor, H., Flynn, P., Nobes, C. D., Hall, A. and Parker, P. J. (1998). PRK1 supported by NKB (grant 99-20254). is targeted to endosomes by the small GTPase, RhoB. J. Biol. Chem. 273, 4811-4814. Merrifield, C. J. (2004). Seeing is believing: imaging actin dynamics at single References sites of endocytosis. Trends Cell Biol. 14, 352-358. Adamson, P., Paterson, H. F. and Hall, A. (1992). Intracellular localization Merrifield, C. J., Moss, S. E., Ballestrem, C., Imhof, B. A., Giese, G., of the P21rho proteins. J. Cell Biol. 119, 617-627. Wunderlich, I. and Almers, W. (1999). Endocytic vesicles move at the tips Alberts, A. S. (2001). Identification of a carboxyl-terminal diaphanous-related of actin tails in cultured mast cells. Nat. Cell Biol. 1, 72-74. formin homology protein autoregulatory domain. J. Biol. Chem. 276, 2824- Murphy, C., Saffrich, R., Grummt, M., Gournier, H., Rybin, V., Rubino, 2830. M., Auvinen, P., Lutcke, A., Parton, R. G. and Zerial, M. (1996). Amano, M., Mukai, H., Ono, Y., Chihara, K., Matsui, T., Hamajima, Y., Endosome dynamics regulated by a Rho protein. Nature 384, 427-432. Okawa, K., Iwamatsu, A. and Kaibuchi, K. (1996). Identification of a Nakano, K., Takaishi, K., Kodama, A., Mammoto, A., Shiozaki, H., putative target for Rho as the serine-threonine kinase protein kinase N. Monden, M. and Takai, Y. (1999). Distinct actions and cooperative roles Science 271, 648-650. of ROCK and mDia in Rho small G protein-induced reorganization of the Aniento, F., Emans, N., Griffiths, G. and Gruenberg, J. (1993). actin cytoskeleton in Madin-Darby canine kidney cells. Mol. Biol. Cell 10, Cytoplasmic dynein-dependent vesicular transport from early to late 2481-2491. endosomes. J. Cell Biol. 123, 1373-1387. Nielsen, E., Severin, F., Backer, J. M., Hyman, A. A. and Zerial, M. (1999). Journal of Cell Science Aschenbrenner, L., Naccache, S.N. and Hasson, T. (2004). Uncoated Rab5 regulates motility of early endosomes on microtubules. Nat. Cell Biol. endocytic vesicles require the unconventional myosin, Myo6, for rapid 1, 376-382. transport through actin barriers. Mol. Biol. Cell 15, 2253-2263. Olson, M. F. (2003). Dispatch. GTPase signalling: new functions for Chardin, P., Maduale, P. and Tavitan, A. (1998). Coding sequence of human Diaphanous-related formins. Curr. Biol. 13, R360-R362. rho cDNA clone 6 and clone 9. Nucleic Acids Res. 16, 2717. Palazzo, A. F., Eng, C. H., Schlaepfer, D. D., Marcantonio, E. E. and Du, W. and Prendergast, G. C. (1999). Geranylgeranylated RhoB mediates Gundersen, G. G. (2004). Localized stabilization of microtubules by suppression of human tumor cell growth by farnesyltransferase inhibitors. integrin- and FAK-facilitated Rho signaling. Science 303, 836-839. Cancer Res. 59, 5492-5496. Prendergast, G. C. (2001). Actin’ up: RhoB in cancer and apoptosis. Nat. Du, W., Lebowitz, P. F. and Prendergast, G. C. (1999). Cell growth Rev. Cancer 1, 162-168. inhibition by farnesyltransferase inhibitors is mediated by gain of Pruyne, D., Evangelista, M., Yang, C., Bi, E., Zigmond, S., Bretscher, A. geranylgeranylated RhoB. Mol. Cell. Biol. 19, 1831-1840. and Boone, C. (2002). Role of formins in actin assembly: nucleation and Evangelista, M., Zigmond, S. and Boone, C. (2003). Formins: signaling barbed-end association. Science 297, 612-615. effectors for assembly and polarization of actin filaments. J. Cell Sci. 116, Robertson, D., Paterson, H. F., Adamson, P., Hall, A. and Monaghan, P. 2603-2611. (1995). Ultrastructural localization of ras-related proteins using epitope- Fujimoto, L. M., Roth, R., Heuser, J. E. and Schmid, S. L. (2000). Actin tagged plasmids. J. Histochem. Cytochem. 43, 471-480. assembly plays a variable, but not obligatory role in receptor-mediated Rozelle, A. L., Machesky, L. M., Yamamoto, M., Driessens, M. H., Insall, endocytosis in mammalian cells. Traffic 1, 161-171. R. H., Roth, M. G., Luby-Phelps, K., Marriott, G., Hall, A. and Yin, H. Gampel, A. and Mellor, H. (2002). Small interfering RNAs as a tool to assign L. (2000). Phosphatidylinositol 4,5-bisphosphate induces actin-based Rho GTPase exchange-factor function in vivo. Biochem. J. 366, 393-398. movement of raft-enriched vesicles through WASP-Arp2/3. Curr. Biol. 10, Gampel, A., Parker, P. J. and Mellor, H. (1999). Regulation of epidermal 311-320. growth factor receptor traffic by the small GTPase rhoB. Curr. Biol. 9, 955- Simonsen, A., Lippe, R., Christoforidis, S., Gaullier, J-M., Brech, A., 958. Callaghan, J., Toh, B-H., Murphy, M., Zerial, M. and Stenmark, H. Gasman, S., Kalaidzidis, Y. and Zerial, M. (2003). RhoD regulates (1998). EEA1 links PI(3)K function to Rab5 regulation of endosome fusion. endosome dynamics through Diaphanous-related Formin and Src tyrosine Nature 394, 494-498. kinase. Nat. Cell Biol. 5, 195-204. Stein, M. P., Feng, Y., Cooper, K. L., Welford, A. M. and Wandinger-Ness, Geneste, O., Copeland, J. W. and Treisman, R. (2002). LIM kinase and A. (2003). Human VPS34 and p150 are Rab7 interacting partners. Traffic. Diaphanous cooperate to regulate serum response factor and actin dynamics. 4, 754-771. J. Cell Biol. 157, 831-838. Stenmark, H., Parton, R. G., Steele-Mortimer, O., Lutcke, A., Gruenberg, Hirokawa, N. (1998). Kinesin and dynein superfamily proteins and the J. and Zerial, M. (1994). Inhibition of rab5 GTPase activity stimulates mechanism of organelle transport. Science 279, 519-526. membrane fusion in endocytosis. EMBO J. 13, 1287-1296. Ishizaki, T., Maekawa, M., Fujisawa, K., Okawa, K., Iwamatsu, A., Fujita, Symons, M. and Rusk, N. (2003). Control of vesicular trafficking by Rho A., Watanabe, N., Saito, Y., Kakizuka, A., Morii, N. et al. (1996). The small GTPases. Curr. Biol. 13, R409-R418. 2670 Journal of Cell Science 118 (12)

Tominaga, T., Sahai, E., Chardin, P., McCormick, F., Courtneidge, S. A. protein for Rho small GTPase and is a ligand for profilin. EMBO J. 16, 3044- and Alberts, A. S. (2000). Diaphanous-related formins bridge Rho GTPase 3056. and Src tyrosine kinase signaling. Mol. Cell 5, 13-25. Watanabe, N., Kato, T., Fujita, A., Ishizaki, T. and Narumiya, S. (1999). Uehata, M., Ishizaki, T., Satoh, H., Ono, T., Kawahara, T., Morishita, T., Cooperation between mDia1 and ROCK in Rho-induced actin Tamakawa, H., Yamagami, K., Inui, J., Maekawa, M. et al. (1997). reorganization. Nat. Cell Biol. 1, 136-143. Calcium sensitization of smooth muscle mediated by a Rho-associated Wherlock, M., Gampel, A., Futter, C. and Mellor, H. (2004). protein kinase in hypertension. Nature 389, 990-994. Farnesyltransferase inhibitors disrupt EGF receptor traffic through Wasserman, S. (1998). FH proteins as cytoskeletal organizers. Trends Cell modulation of the RhoB GTPase. J. Cell Sci. 117, 3221-3231. Biol. 8, 111-115. Wubbolts, R., Fernandez-Borja, M., Oomen, L., Verwoerd, D., Janssen, Watanabe, N., Madaule, P., Reid, T., Ishizaki, T., Watanabe, G., Kakizuka, H., Calafat, J., Tulp, A., Dusseljee, S. and Neefjes, J. (1996). Direct A., Saito, Y., Nakao, K., Jockusch, B. M. and Narumiya, S. (1997). vesicular transport of MHC class II molecules from lysosomal structures to p140mDia, a mammalian homolog of Drosophila diaphanous, is a target the cell surface. J. Cell Biol. 135, 611-622. Journal of Cell Science