Drosophila melanogaster cellularization is a dramatic form of cytokinesis in which a membrane furrow simultaneously encapsulates thousands of cortical nuclei of the syncytial embryo to generate a polarized cell layer. Formation of this cleavage furrow depends on Golgi-based secretion and microtubules. During cellularization, specific Golgi move along microtubules, first to sites of furrow formation and later to accumulate within the apical cytoplasm of the newly forming cells. Golgi movements and furrow formation depend on cytoplasmic dynein. Furthermore, Lava lamp (Lva), a golgin protein that is required for cellularization, specifically associates with dynein, dynactin, cytoplasmic linker protein-190 (CLIP-190) and Golgi spectrin, and is required for the dynein-dependent targeting of the secretory machinery. The Lva domains that bind these microtubule-dependent motility factors inhibit Golgi movement and cellularization in a live embryo injection assay. These results provide new evidence that golgins promote dynein-based motility of Golgi membranes (Papoulas, 2005).
Following thirteen mitotic nuclear divisions, Drosophila syncytial embryos undergo a form of cytokinesis called cellularization. The process occurs during the interphase of nuclear cycle 14 over a 1-h period and relies on both microtubules and Golgi-based secretion. During cellularization, microtubules emanate from apically positioned centrosomes to form 'inverted baskets' around each nucleus at the embryoís surface. The furrows form between adjacent nuclei, generating a honey comb pattern, and ultimately expand laterally at their base and fuse to seal each nucleus off from the inner yolk, creating a multicellular embryo. Cellularization requires an increase in cell surface area of approximately 20-fold, and because the disruption of Golgi function inhibits furrowing, a significant amount of this membrane is believed to come from de novo secretion. A portion of the Golgi undergoes two waves of apically directed movements during furrow formation that are coordinated with the dynamic changes in microtubule organization (Papoulas, 2005).
Golgi body movements can be visualized by injecting live embryos with dilute Cy5-tagged anti-Lava-lamp antibody, which fluorescently labels a small number of Golgi near the site of injection and does not affect Golgi movement or cellularization. Injection of colchicine to depolymerize microtubules blocks the Golgi movements, whereas microfilament depolymerization with cytochalasin D does not, consistent with the idea that microtubule-based Golgi movements support active membrane secretion (Papoulas, 2005).
The mechanism that drives the microtubule-dependent Golgi movements is unknown; however, the trajectory of the Golgi movements suggests that cytoplasmic dynein is involved. Cytoplasmic dynein is a multisubunit minus-end-directed microtubule motor that is known to transport a variety of cargos in animal cells. Several studies have suggested that dynein can associate with mammalian Golgi through Golgi-associated spectrin and dynactin, a protein complex that is required for processive movement of dynein cargo. α-spectrin is present on Drosophila Golgi, raising the possibility that it could recruit dynein/dynactin (Papoulas, 2005).
To determine whether cytoplasmic dynein is responsible for the apically directed Golgi movements, Dynein heavy chain (Dhc) activity was inhibited and the effects on Golgi movement were monitored. The use of transgenic flies that express green fluorescent protein (GFP)-tagged Myosin II (MyoII-GFP) permitted the simultaneous visualization of the furrow tip. Injection of function-blocking anti-Dhc antibodies into live MyoII-GFP embryos disrupted Golgi movements, reducing the number of Golgi that move processively. In addition, disruption of Dhc function blocks the progression and organization of the furrow front. Injection of a control [anti-glutathione-Stransferase (GST)] antibody had no effect. Next, embryos were collected from wild-type or Dhc6 mutant females, fixed, and prepared for indirect immunofluorescence using antibodies against tubulin and Lva. Embryos derived from mutant females that develop to nuclear cycle 14 fail to undergo normal furrow formation. These mutant embryos also accumulate fewer Golgi bodies in the apical cytoplasm compared with wild-type embryos at the same stage, despite the fact that microtubules seem abundant and oriented appropriately to support apically directed Golgi movement (Papoulas, 2005).
To explore the possibility that dynein directly mediates the Golgi movements, it was determined whether Dhc localizes to Golgi bodies. Fixed cellularizing embryos were prepared for immunofluorescence using antibodies against Lva and Dhc. Dhc is abundant throughout the cortical cytoplasm, but areas of Dhc enrichment colocalize with Golgi-associated Lva staining. Corroborating the immunofluorescence results, a portion of membrane-associated dynein and a significant fraction of α-spectrin in membrane extracts are present in Golgi-enriched membrane fractions with Lva on density gradients. Some membrane-associated dynein also sediments with an unknown membrane population that is not associated with Lva (Papoulas, 2005).
The role of spectrin and the dynactin complex in the recruitment of dynein to mammalian Golgi and the previous observation that the golgin Lva and Golgi spectrin associate in Drosophila, prompted an investigation of whether Lva might have a direct role in recruiting and/or regulating dynein function. To determine whether Lva associates with dynein, immunoprecipitations were performed using protein extracts prepared from embryos. Anti-Lva antibodies coimmunoprecipitate α-spectrin, as expected, and also coimmunoprecipitate Dhc, whereas control immunoprecipitations do not. These data demonstrate that a portion of the endogenous Lva and dynein associate in soluble embryo extracts. In conjunction with the colocalization and membrane cofractionation results, these data suggest that Lva and dynein may interact on the surface of Golgi in vivo (Papoulas, 2005).
So far, nearly all dynein-based motility of membrane vesicles requires the dynactin complex, and may also be facilitated by microtubule plus-end tracking proteins (+TIPs). +TIPs bind growing microtubule plus-ends and are believed to regulate microtubule dynamics and the docking of membranes to microtubules. Consistent with this model, the +TIPs p150Glued and CLIP-190 were detected in the Golgi-enriched membrane fractions. p150Glued, a subunit of the dynactin complex, extensively cosediments with Lva; and although anti-p150Glued antibodies coimmunoprecipitate Lva and Dhc, anti-Lva antibodies do not detectably coimmunoprecipitate p150Glued under the conditions used. These observations are consistent with the possibility that antibody binding to Lva disrupts adjacent binding of dynactin and previous data that show that anti-Lva antibody injections disrupt Golgi movements in live embryos. CLIP-190 is the Drosophila orthologue of mammalian CLIP-170, which has been implicated in linking vesicles to microtubules for dynein-dependent transport. CLIP-190 is known to cofractionate and colocalize with Lva, and as expected, CLIP-190 and Lva coimmunoprecipitate. Taken together, the association of Lva with dynein, p150Glued, CLIP-190 and spectrin strongly suggests a role for Lva in the microtubule-dependent movement of Golgi during cellularization (Papoulas, 2005).
The ability of different portions of Lva to interact with these microtubule motility factors and with Golgi membrane was tested. Seven contiguous fragments spanning the full length of Lva were expressed as GST-Lva fusion proteins and used to make affinity columns. Chromatography was performed with native extracts that were derived from cellularizing embryos. Dynein exclusively binds the globular carboxyl terminus of Lva (Lva5), whereas both p150Glued and CLIP-190 bind the coiled-coil central portion of Lva (Lva3) and the globular C terminus. Furthermore, α-spectrin binds to Lva3, but not to the dynein-binding C terminus. Therefore, spectrin and dynactin by themselves are insufficient to recruit dynein under the in vitro conditions. Moreover, dynein binds the C terminus of Lva in the absence of spectrin. Interestingly, Drosophila BicD fails to bind any of the seven Lva segments or colocalize with Golgi-associated Lva by immunofluorescence, despite the established role of its orthologues in dynein recruitment to mammalian Golgi membrane. Membrane-binding assays were used to map the region(s) of Lva that are required for Golgi association. Golgi-enriched membrane preparations were incubated with each of the GST-Lva fusion proteins and then collected by centrifugation and assayed for the presence of Lva fusion proteins by anti-GST immunoblotting. Only GST-Lva2A and 2B were found to associate with Golgienriched membranes with high affinity. No fusion proteins pelleted in the absence of membrane (Papoulas, 2005).
To test the functional significance of the Lva interactions with microtubule motility factors Lva fusion proteins were injected into live D. melanogaster embryos and the effects on Golgi movement and furrow formation were monitored. Injection of GST or Lva1, which do not bind the microtubule motility factors, does not inhibit Golgi movement or furrowing. In contrast, injection of Lva3 or Lva5 severely impairs the Golgi movements and furrowing. The number of Golgi bodies that move is significantly reduced by Lva5 and virtually eliminated by Lva3. The Golgi movements that remain in each case are significantly delayed, occurring exclusively during the time period when the second wave of apical movement is normally observed. The impaired furrowing caused by injection of Lva3 and Lva5 occurs primarily during the late (fast) phase, suggesting that the early Golgi movements are a prerequisite for rapid membrane growth during this later phase. Injection of Brefeldin A (BFA), a potent inhibitor of membrane transport, has a similar effect on the fast phase of membrane growth3. Staining with the lectin concanavalin A reveals normal plasma membrane ruffles at the apical margin of furrows and some subcortical membrane during the slow phase in GST-injected embryos. This plasma membrane topology has been previously described in greater detail and coincides with proposed sites of exocytosis. However, this plasma membrane structure is severely disrupted in Lva3-injected embryos and discontinuity in plasma membrane furrows is evident. These results are consistent with the idea that Golgi targeting to the cell surface is required for new plasma membrane secretion to form cleavage furrows (Papoulas, 2005).
These biochemical experiments suggest that the dominant-negative effect of Lva3 on Golgi body movement might result from displacing endogenous dynactin and CLIP-190 from the Golgi surface. To test this the subcellular localizations of Dhc, CLIP-190 and Lva were studied in fixed embryos after injection with GST or Lva3. The uniform expression of p150Glued in embryos prevented identification of any potential effect on dynactin localization, and Dhc localization was indistinguishable between GST- and Lva3-injected embryos, consistent with the absence of dynein binding to Lva3 in vitro. By contrast, the normal CLIP-190 Golgi-association is disrupted in Lva3-injected embryos, particularly at the furrow front during the slow phase of cellularization. GST-injected embryos are unaffected, and Lva remains Golgi-associated in Lva3-injected embryos; although, Golgi are more dispersed (Papoulas, 2005).
Thus, biochemical and functional analysis of Lva suggests that it has a unique function among golgin proteins. The data are consistent with an adaptor model in which Lva stimulates dynein-dependent Golgi movement by binding cytoplasmic dynein in association with promoters of dynein-dependent motility. This model is based in part on the observation that dynein-dependent Golgi movement and furrow formation are significantly inhibited by injections of Lva3, which has no detectable affinity for dynein. This dominant-negative effect can be explained as a consequence of displacing CLIP-190 from Golgi bodies, but is likely to also displace dynactin, resulting in diminished dynein function. The microtubule motility factors bound to the central domain of Lva could facilitate dynein function indirectly by capturing microtubule plus ends, as has been proposed for dynactin7. Alternatively, Lvaís predicted hinge region could facilitate interactions between microtubule motility factors that are bound to the central region of Lva and the dynein motor that is bound to the C terminus, and/or modulate dynein catalytic activity or microtubule binding. These possibilities are not mutually exclusive, and whether Lva contributes to dynein recruitment in vivo remains an open question, as does the possibility that two distinct Lva/dynein complexes exist, one containing dynactin and the other CLIP-190. Future studies will be required to distinguish between these possibilities and to assess the role of Lva in other developmental contexts (Papoulas, 2005).
To assess whether Lva interacts with other proteins, the native size of Lva was compared with other microfilament/microtubule-associated proteins (MMAPs). The S100 and the final protein (MMAP) fraction were passed separately over a gel filtration column, and fractions were assayed by immunoblot. alpha-Spectrin has been previously shown to copurify with ß- and ßH-Spectrin in two stable heterotetrameric complexes (alpha2ß2 and alpha2ßH2, respectively) and coimmunoprecipitates with ß- and ßH-Spectrin, but information on association of Lva with alpha-Spectrin only is presented for simplicity. Immunoblots show that in both the S100 and the MMAP fraction, Lva, CLIP190, alphaß-, and alphaßH-Spectrin coelute from the column with native molecular weights larger then their predicted molecular weights, indicating that each protein exists in large, stable complexes (Sisson, 2000).
Lva, CLIP190, alphaß-, and alphaßH-Spectrin also cofractionate over two consecutive F-actin affinity columns, indicating that each protein is associated with a stable F-actin-binding activity. S100 was passed over an F-actin column; ABPs were eluted as before, dialyzed against F-actin-binding buffer, and the soluble protein was passed over a second F-actin column. Immunoblots show that Lva, alphaßH-Spectrin, CLIP190, and Anillin each efficiently rebind the second column, while KLP3A does not. The initial binding and subsequent rebinding of alphaß-Spectrin to F-actin is relatively weak. Lva, Spectrins, and CLIP190 elute with a common peak in fractions (Sisson, 2000).
Because Lva, CLIP190, and Spectrins, cofractionate in the above experiments, an assessment was made of whether they interact by immunoprecipitation (IP). Anti-Lva antibody efficiently precipitates Lva protein, and co-IPs alphaßH- and alphaß-Spectrin, as well as CLIP190. Although the anti-alpha-Spectrin and anti-CLIP190 antibodies are inefficient at precipitating their respective antigens, both corroborate the co-IPs obtained with the anti-Lva antibody. Because antibodies to alpha-Spectrin and CLIP190 do not co-IP one another, it is likely that Lva associates with Spectrins and CLIP190 separately (Sisson, 2000).
Blastoderm embryos were treated with fixatives that efficiently preserve both cortical F-actin and MTs and prepared for immunofluorescence. In syncytial and cellularizing blastoderms, Lva, alpha-Spectrin, and CLIP190 colocalize to large cytoplasmic puncta, some of which are found closely apposed to the furrow front in cellularizing blastoderms. Additional CLIP190-specific puncta are also observed at furrow tips. While most Lva is associated with puncta, low levels are seen throughout the cortical cytoplasm. alpha-Spectrin and CLIP190 are also observed elsewhere within the cortical cytoplasm. Although alpha-Spectrin forms stable complexes with ß- and ßH-Spectrin, punctate ß- or ßH-Spectrin localization is not observed. Instead, these proteins are found throughout the cortical cytoplasm. Minimal Spectrin localization is seen along the PM (Sisson, 2000).
Because the punctate colocalization pattern of Lva, alpha-Spectrin, and CLIP190 is reminiscent of that observed for two cis-Golgi markers, ß-coatomer (ß-COP) and a 120-kD integral membrane protein (p120), their distribution was examined relative to the three MMAPs. Indeed, double immunofluorescence shows that p120 and Lva colocalize. alpha-Spectrin and CLIP190 colocalize with the Golgi markers, as with Lva (Sisson, 2000).
Although a special isoform of mammalian ß-Spectrin (ßIsigma-Spectrin) has been shown to associate with Golgi, Golgi-associated Spectrin has not been previously described in Drosophila, where Spectrins have been shown primarily at the PM. To rule out fixation artifacts, three different preparative conditions were tested. In each case, the punctate localization for alpha-Spectrin, Lva, and the Golgi markers was clearly observed, while very weak Spectrin localization was found at the PM. Together, these results suggest that Spectrins reside on both the PM and Golgi bodies of cellularizing blastoderms, and that each Spectrin population is differentially sensitive to the immunofluorescence preparative conditions used (Sisson, 2000).
Brill, J. A., et al. (2000). A phospholipid kinase regulates actin organization and intercellular bridge formation during germline cytokinesis. Development 127: 3855-3864. 10934029
Chan, E., and Fritzler, M.J. (1998). Golgins: coiled-coil-rich proteins associated with the Golgi Complex. Elec. J. Biotech. 1 :1-10
De Matteis, M.A., and Morrow, J.S. (1998). The role of ankyrin and spectrin in membrane transport and domain formation. Curr. Opin. Cell Biol. 10: 542-549. 9719877
Papoulas, O., Hays, T. S. and Sisson, J. C. (2005). The golgin Lava lamp mediates dynein-based Golgi movements during Drosophila cellularization. Nat. Cell Biol. 7(6): 612-8. Medline abstract: 15908943
Sisson, J. C., et al. (2000). Lava lamp, a novel peripheral golgi protein, is required for Drosophila melanogaster cellularization. J. Cell Bio. 151: 905-918. 11076973
Warren, G., and Malhotra, V. 1998. The organisation of the Golgi apparatus. Curr. Opin. Cell Biol. 10: 493-498. 9719870
Waters, M. G., and Pfeffer, S. R. 1999. Membrane tethering in intracellular transport. Curr. Opin. Cell Biol. 11: 453-459. 10449330
date revised: 10 February 2001
Home page: The Interactive Fly © 1997 Thomas B. Brody, Ph.D.
The Interactive Fly resides on the
Society for Developmental Biology's Web server.