SEARCH

SEARCH BY CITATION

Keywords:

  • Drosophila;
  • myoblast fusion;
  • adhesion;
  • podosome F-actin;
  • Dumbfounded/Kirre;
  • Sticks and stones;
  • Rolling pebbles;
  • Blow;
  • Titin;
  • C2C12 cells

Abstract

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. RESULTS
  5. DISCUSSION
  6. EXPERIMENTAL PROCEDURES
  7. Acknowledgements
  8. REFERENCES
  9. Supporting Information

During myogenesis in Drosophila embryos, a prominent adhesive structure is formed between precursor cells and fusion-competent myoblasts (fcms). Here, we show that Duf/Kirre and its interaction partners Rols7 (found in founder myoblasts and growing myotubes) and Sns (found in fcms) are organized in a ring-structure at the contact points of fcms with precursor cells, while cytoskeletal components like F-actin and Titin are centered in this ring in both cell types. The cytoplasmic protein Blow colocalizes with the actin plugs in fcms after cell adhesion. Furthermore, the requirement of additional as yet unidentified components was demonstrated by using mammalian C2C12 myoblasts. In this study, we propose that the fusion-restricted myogenic-adhesive structure (FuRMAS) is pivotal in linking cell adhesion as well as local F-actin assembly and dynamics to downstream events that ultimately lead to plasma membrane fusion. Moreover, we suggest that the FuRMAS may restrict the area of membrane breakdown. Developmental Dynamics 236:404–415, 2007. © 2006 Wiley-Liss, Inc.


INTRODUCTION

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. RESULTS
  5. DISCUSSION
  6. EXPERIMENTAL PROCEDURES
  7. Acknowledgements
  8. REFERENCES
  9. Supporting Information

In addition to mitosis without cytokinesis, cell–cell fusion is the main process that generates multinuclear tissues in metazoans. In contrast to the growing amount of data on the mechanisms of intracellular membrane fusion and virus–cell fusion, only little is known about the mechanisms of plasma membrane fusion and the proteins involved in this process.

To date, several pathways leading to the final fusion event have been characterized. For example, if fusing cells are not ordered in an epithelial sheet like the cells of the Caenorhabditis elegans hypodermis, they first have to contact each other. In such cases, cell–cell recognition and subsequent adhesion are essential for membrane fusion. Previously characterized complexes mediating cell–cell adhesion in metazoans include adherens junctions like the zonula adherens in epi- and endothelia that are built up mainly by cadherins. Transient adhesion is mediated by focal adhesion points harboring molecules like integrins and cytoplasmic plugs containing vinculin, talin, and alpha actinin connected to actin. To date, none of these classic cell adhesion molecules have been shown to be essential for myoblast fusion in Drosophila (Bunch et al.,1992; Prokop et al.,1998). However, integrins are known to be necessary for myoblast fusion in vertebrates (Schwander et al.,2003; Lafuste et al.,2005).

During Drosophila myoblast fusion, adhesion occurs between two different cell types: the founder cells and the fusion competent myoblasts (fcms; reviewed in Taylor,2003; Chen and Olson,2004; Abmayr et al.,2005). Heterophilic adhesion is mediated by founder cell-specific proteins such as the Ig-superfamily (IgSF) protein Dumbfounded (Duf)/Kin of IrreC (Kirre) in redundancy with Roughest/IrreC (Ruiz-Gomez et al.,2000; Strünkelnberg et al.,2001) and the fcm-specific IgSF member Sticks and stones (Sns; Bour et al.,2000; Galletta et al.,2004). After initial cell adhesion, the GEF Schizo/Loner (Chen et al.,2003; Oenel et al.,2004) and the cellular adapter protein Rolling pebbles (Isoform 7; Rols7; Chen and Olson,2001; Menon and Chia,2001; Rau et al.,2001) are required in precursor cells. In fcms, the cytoplasmic protein Blown fuse (Blow) is involved in the fusion process (Artero et al.,2003; Schröter et al.,2006). During the past decade, an increasing number of fusion-relevant proteins has been described. Some are expressed by both cell types like the GEF Myoblast city (Mbc) (Rushton et al.,1995), the small GTPases Drac1 and Drac2 (Hakeda-Suzuki et al.,2002), and the IgSF protein Roughest/IrreC (Strünkelnberg et al.,2001). Mutations of the corresponding genes lead to severe fusion defects and embryonic lethality (Rushton et al.,1995; Bour et al.,2000; Ruiz-Gomez et al.,2000; Chen and Olson,2001; Menon and Chia,2001; Rau et al.,2001; Hakeda-Suzuki et al.,2002; Chen et al.,2003). Moreover, fusion-relevant proteins such as Rols7 (Chen and Olson,2001) and Schizo/Loner (Chen et al.,2003) depend on the protein Duf/Kirre to be correctly localized near the cell membrane. The cytoplasmic domain of Duf/Kirre is known to interact with the C-terminal domain of Rols7 (Kreisköther et al.,2006), thus it can be proposed that this interaction subsequently localizes rols.

Also for Sns, other features in addition to cell adhesion, such as cell signaling, have been proposed (Galletta et al.,2004). At the ultrastructural level, cell adhesion is accompanied by the formation of a so-called prefusion complex consisting of aligned pairs of vesicles at opposing plasma membranes that later form electron-dense plaques followed by membrane breakdown in this area (Doberstein et al.,1997).

To obtain further insight into the process of myoblast adhesion and to learn how cellular adhesion is linked to later events such as plasma membrane merging, we studied the subcellular localization of Sns as well as Duf/Kirre and its interaction partner Rols7. At contact points, we observed a ring-shaped distribution of Sns in the filopodia of fcms, with Duf/Kirre and Rols7 concentrated at the opposing site of the precursor cell. For the purpose of this manuscript, we use the term filopodia to describe the long processes extending from the fcms toward the growing muscles. In addition, F-actin and Titin were found to localize at the center of this ring in both cell types. Blow, which genetically interacts with the actin regulator Kette/Hem (Schröter et al.,2004), is solely expressed in fcms in the somatic mesoderm (Schröter et al.,2006). Here, we show that Blow expression overlaps with the actin plugs observed in fcms. We have termed this structure FuRMAS (fusion-restricted myogenic-adhesive structure) as several proteins known to be necessary for fusion are concentrated in this ring-structure in close proximity. We also investigated whether a FuRMAS(-like) structure can be formed in mammalian cells. Finally, we propose a model where the formation of the FuRMAS structure is essential for myoblast fusion in Drosophila melanogaster as it links cell adhesion with subsequent events that ultimately lead to cell membrane fusion.

RESULTS

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. RESULTS
  5. DISCUSSION
  6. EXPERIMENTAL PROCEDURES
  7. Acknowledgements
  8. REFERENCES
  9. Supporting Information

Duf/Kirre, Rols7, and Sns Show a Ring-Shaped Arrangement at the Contact Points of fcms With Precursor Cells While Blow Has a Plug-Like Distribution in the Filopodia-Like Processes of fcms

To analyze the distribution of the IgSF member Duf/Kirre and its interaction partner Rols7 (both founder specific) as well as of Blow and Sns (both fcm specific), fusion stages of wild-type embryos were labeled with antibodies against Duf/Kirre (Kreisköther et al.,2006), Rols7 (this work), Blow (Schröter et al.,2006), and Sns (this work). These proteins exhibit a spot-wise pattern at the contact points of fcms with precursor cells in the somatic mesoderm during myoblast fusion (Fig. 1A–C). For Blow, we first observe a uniform cytoplasmic distribution and from stage 14 onward a spot-wise pattern (Fig. 1D). Higher magnification revealed that Duf/Kirre (Fig. 1E) and Rols7 (Fig. 1F) are not uniformly distributed, but build up a ring-shaped structure at the contact point of the precursor cell with a fcm. This finding also holds true for the distribution of the Sns protein in the fcms, where a large fraction of the protein assembles in a similar structure at the contact area of the cells (Fig. 1G). The rings at the contact points of the cells are best visualized at late stages of fusion (stage 15), when only a small number of fcms fuses to nearly mature myotubes. At these stages, high levels of Duf/Kirre, Rols7 in the growing myotube as well as Blow and Sns in the fcm are observed at those points where the filopodium of a fcm contacts the growing myotube (Fig. 1I–K, see M–O for corresponding drawings). Interestingly, Blow is not distributed in a ring-shape, but it is centered in the ring-structure where the filopodium of the fcm touches the growing muscle (Fig. 1L, see P for corresponding drawing). A similar protein distribution for Duf/Kirre, Rols7, Blow, and Sns can also be observed during development of the pharynx and the visceral musculature. Moreover, Duf/Kirre and Sns are detectable in the garland cells and the central nervous system of later stages (data not shown).

thumbnail image

Figure 1. AD: In stage 13 embryos, the proteins Duf/Kirre (A), Rols7 (B), Sns (C), and Blow (D) exhibit a dot-like expression pattern in the somatic mesoderm. EH: Higher magnification reveals that Duf/Kirre, Rols7, and Sns build up a ring-shaped structure with a protein-free center. Note that Blow is not involved in this process. The protein expression in the somatic mesoderm persists until stage 15/16, when the final fusions take place. IL: In stage 15 embryos, protein expression is restricted to the contact points of the filopodia of fcms with growing myotubes. MP: Below I–L, schematic drawings of myotubes and fcms are provided.

Download figure to PowerPoint

The Observed Circular Structure Surrounds an F-Actin Core in the Growing Myotube and fcm

Previously, we reported that regulators of F-actin filament assembly are essential for myoblast fusion. In particular, we showed that Kette (Hem), which genetically interacts with Blow (Schröter et al.,2004) and Wasp are essential for myoblast fusion (Schaefer et al., in revision). Therefore, we analyzed F-actin distribution during fusion using fluorescent double-labeling with phalloidin–tetrarhodamine isothiocyanate (TRITC) to detect F-actin and an anti-Rols7 antibody to visualize the ring structure.

At stage 13, F-actin appears in a dot-like pattern in the somatic mesoderm (Fig. 2B). This dot-like distribution can also be observed for Rols7 (Fig. 2A, overlay see Fig. 2C). At higher magnification, it became apparent that F-actin does not have a ring-shaped distribution but rather looks like a plug (Fig. 2D,F). The highest F-actin concentration is detected in the Rols7-free center of the ring (Fig. 2E,G) on the growing myotube. The plug-like F-actin distribution in the growing muscle and filopodium of the fcms can be best visualized in a lateral view in comparison to Rols (located on the growing myotube; Fig. 2I–K). Of interest, Blow shows the same distribution as the plug-like F-actin in fcms (Fig. 1L), whereas alpha-actinin, a protein known to interact with Rols7 in Z-discs (Kreisköther et al.,2006), is not detectable in these structures (data not shown).

thumbnail image

Figure 2. B: In stage 13 embryos, F-actin is expressed in a wide variety of tissues, but in the somatic mesoderm, relatively strong expression can be observed in a dot-like pattern in the somatic mesoderm. A: This dot-like pattern was also detected for Rols expression. C: Both proteins colocalize in the somatic mesoderm, indicated by the yellow color in the overlay. DG: Higher magnification of this area of coexpression in a stage 15 embryo reveals that Rols7 has a ring-shaped expression pattern with a protein-free center (E), while F-actin expression is more laminar (F) with the highest expression in the center of the Rols7 ring (G, overlay). The colocalization of F-actin and β1-integrin in the somatic mesoderm is restricted to the muscle attachment site. H: This area is indicated by the yellow color of overlay of an anti–F-actin/anti–β1-integrin double-labeled stage 15 embryo (arrowhead). At the contact points of cells, only the laminar F-actin staining was observed (H, arrow). In IK, the localization of Rols7 (I) in growing myotubes in comparison to F-actin (J,K) in myoblasts is shown.

Download figure to PowerPoint

In vertebrates, β1 integrin is involved in myoblast fusion (Schwander et al.,2003). Double-labeling of stage 15 embryos with phalloidin-TRITC and an anti-βPS-integrin antibody failed to produce βPS-integrin staining at the actin spots in the mesoderm (Fig. 2H, arrow); however, colocalization at the muscle attachment sites was observed (Fig. 2H, arrowhead). This finding is supported by the fact that Drosophila integrin mutants show defects in muscle attachment to the epidermis but not in myoblast fusion (for review, see Bökel and Brown,2002).

The Ring-Shaped Adhesive Structure Varies in Size During the Course of Myoblast Fusion

Measurements of the size of the ring-shaped structure marked with the Duf/Kirre antibody show that the initial diameter of the ring is around 1 μm at the time when the filopodium of a fcm first anchors to a precursor cell (Figs. 1I, 3A) at around embryonic stages 11/12. Later, the filopodium is no longer visible as the fcm has a broader contact area with the growing muscle (Fig. 3C,D). At stages 13/14, larger rings with a maximal diameter of 4–5 μm (Fig. 3B–D) can be observed that were absent during the early stages. Assuming that this enlargement does not simply reflect cell size variations, these results indicate a widening of the ring-structure during fusion. In addition, when the diameter of the structure is enlarged, the fcms can be observed in very tight contact to the growing precursor cell (Fig. 3C,D) and the filopodium has shrunken. We also saw a similar expansion of the distribution for F-actin and D-Titin/Kettin/Zormin (not shown). These findings are in agreement with data from Zhang et al. (2000).

thumbnail image

Figure 3. A,B: View on the surface of a growing myotube. Focus is at a Duf/Kirre containing ring (=FuRMAS, fusion-restricted myogenic-adhesive structure). C,D: Lateral view. The diameter of the ring-shaped structure can vary from 1 μm (A) to 4–5 μm (B–D) indicated by the Duf/Kirre distribution in stage 14 embryos. Rings with a larger diameter are closely associated with fcms (C,D), and it seems that the fcms are more flattened (D), indicating that the fusion proceeds.

Download figure to PowerPoint

The Assembly of the Adhesive Structure Depends on Cell–Cell Contact

Previous observations during myoblast fusion or cell culture experiments led to the conclusion that membrane localization of Duf/Kirre and Sns is independent of cell recognition and adhesion but that assembly of the complex might depend on cell contact (Chen et al.,2003; Galletta et al.,2004). To test whether this is indeed the case or whether the adhesive complex is precasted, the protein distribution of Duf/Kirre in sns-mutant embryos was analyzed.

In sns-mutant embryos, in which the fusion process is already stopped at the level of the founder cell, the recognition of the founder cells and the fcms is disturbed. We clearly show that founder cells accumulate Duf/Kirre at the membranes, but do not exhibit a ring-shaped distribution at any area of the cell (Fig. 4A). In later stages, Duf/Kirre accumulates at the contact points between elongating founder cells (not shown), which may result from homophilic interaction of Duf/Kirre, as seen in cell culture (Chen et al.,2003; Galletta et al.,2004; Menon et al.,2005).

thumbnail image

Figure 4. The formation of the rings in founder cells depends on Sns expression in fusion competent myoblasts (fcms). A: In sns20-2 mutant embryos, the Duf/Kirre protein is localized to the membrane of founder cells, but no ring structure can be observed (black arrow). B,C: In duf/kirre; rst/IrreC and sns20-2 mutants, Blow does not accumulate at the contact points between fcms and myotubes. Founder cells are marked by f, while the asterisk marks some fcms.

Download figure to PowerPoint

In fcms of wild-type embryos, we observed a broad cytoplasmic distribution of Blow followed by concentration in the filopodia, which are in contact with growing muscles (Fig. 1D,H,L). We next addressed the question whether this accumulation depends on the cell adhesion between fcms and the growing myotube. In the deletion mutant BL6018, where both duf and rst are absent and the sns null mutant sns202 (formerly rost202 Paululat et al.,1995; Bour et al.,2000), Blow is scattered in speckles over the cytoplasm (Fig. 4B,C). Thus we conclude that successful cell adhesion is essential for the localization of Blow to the area of the F-actin plug in the filopodia of the fcms.

The Assembly and the Stability of the Adhesive Structure Does Not Involve Blow, Kette, Mbc, or Rols7

To clarify whether any other known fusion protein is involved in the assembly and the stability of the adhesive structure, the protein distribution of Duf/Kirre, Rols7, and Sns in amorphic mutants for the genes blow, kette, and mbc as well as a deficiency deleting the rols gene was analyzed. The blow, kette, and mbc mutants all build rings harboring Duf/Kirre (Fig. 5E–H) and Rols7 (Fig. 5I–K) on the founder/precursor cell and with Sns (Fig. 5A–D) on the fcm. The proteins accumulate mostly at the contact points of the fcm filopodia with the founder/precursor cell (Fig. 5B,G,I,J), building up a complex with a diameter of 1 μm as in wild-type embryos (Fig. 5A–K). Larger complexes are rare in all fusion mutants, but sometimes the ring-shaped expression domain of the proteins seems to be broader than in wild-type embryos. This finding is especially the case when more than one filopodium was in contact with the same membrane area of a founder cell. In later stages, when the fcms were phagocytosed by macrophages, ring-shaped complexes on founder/precursor cells without contact to a fcm were observed in embryo mutants for the fusion genes (Fig. 5E,F). The total protein level of Duf/Kirre, Rols7, and Sns in all these fusion mutants seems to be elevated as was also recently described by Menon et al. (2005). Homozygous rols-deficient embryos also still exhibit ring-shaped adhesive structures formed by Sns on fcms (Fig. 5D) and by Duf/Kirre on founder cells (Fig. 5H); however, the total number of adhesive complexes is greatly reduced.

thumbnail image

Figure 5. AK: The localization of Sns (A–D), Duf/Kirre (E–H), and Rols7 (I–K) is indistinguishable to wild-type in blow2 (A,E,I), kettej4-48 (B,F,J), mbcc1 (C,G,K), and rols (D,H) stage 16 mutant embryos. In all mutants, fusion competent myoblasts (fcms) adhering to founder/precursor cells were observed and the proteins were seen to mostly form a ring-shaped structure at the contact point of the filopodium-like extension of a fcm with a founder/precursor cell (e.g., in B,G,I,J). E,F: In some cases, the ring-shaped expression of Duf/Kirre was observed on founder/precursor cells that had no contact to a fcm. The diameter of the ring is always approximately 1 μm.

Download figure to PowerPoint

Duf Is Able to Recruit Rols to the Membrane in C2C12 Myoblasts

To obtain further insight into the key players central to myoblast fusion, we studied some factors known to be involved in Drosophila myoblast fusion in vertebrate myoblasts. To this end, we tested whether homologs of identified Drosophila myoblast fusion proteins are expressed within C2C12 myoblasts (Yaffe and Saxel,1977). Antibody stainings against mouse Crk, Dock180 (Mbc-homolog), and Drosophila Kette/Hem were all positive (data not shown). In contrast, no staining using antibodies against the Drosophila proteins Blow, Duf/Kirre, Rols, and Sns was observed (data not shown). However, as no close homologs of these proteins have been described in vertebrate genomes, it is possible that they have a very special role in Drosophila.

Therefore, we transfected cells with full-length constructs of either Duf/Kirre plus Rols or Blow plus Sns to mimic the situation in founder cells or fusion competent myoblasts, respectively. Using appropriate antibodies, all constructs were found to be expressed within the cell lines after transfection. In detail, Duf/Kirre localizes to the cell membrane when it is expressed in C2C12 myoblasts as it does in Drosophila myoblasts. Additional staining can be observed throughout the cytoplasm, which is probably due to the excess amount of protein present (Fig. 6A–C). Overall, the Duf/Kirre-transfected cells have a severely altered morphology: the Duf/Kirre-positive cells formed thin filopodia with high accumulation of Duf/Kirre, which extend over several cell diameters (arrow in Fig. 6B), and from these thin filopodia, small Duf/Kirre-positive vesicles seemed to be released (arrowheads in Fig. 6B′). When Rols is transfected in the C2C12 cells alone, it forms a dot-like pattern in the cytoplasm of the cell (Fig. 6D–F) and no localization to the cell membrane is observed. However, upon cotransfection with Duf, Rols relocalizes to the cell membrane where it seems to be always more toward the interior of the cell than Duf (Fig. 6G–I). Thus, the interaction between Rols and Duf/Kirre, which had previously been shown in immunoprecipitations (Chen and Olson,2001) and in a yeast two-hybrid assay (Kreisköther et al.,2006) as well as in S2 cells (Menon et al.,2005), also works in the environment of a mammalian cell.

thumbnail image

Figure 6. Duf-expressing cells form thin filopodia and are able to recruit Rols and Sns to the membrane of C2C12 myoblasts. The actin cytoskeleton is labeled with phalloidin-tetrarhodamine isothiocyanate (TRITC; red in A,C,D,F,Q,S). The nuclei of the cells are stained with Höchst (blue in C,F,I,M,P); the expressed proteins were detected using antibodies as indicated. AC: C2C12 myoblasts transfected with Duf change their morphology and form long, thin filopodia that seem to release small Duf-positive vesicles (B′). DF: When C2C12 cells were transfected with an HA-tagged Rols construct, the protein can be detected in a dot-like pattern throughout the cytoplasm. GI: In contrast, transfection with Rols and Duf leads to a localization of Rols at the cell membrane especially in the formed filopodia (G′). KM: Sns does not localize at the membrane of C2C12 cells either when transfected alone or together with Blow. Nevertheless, Sns seems to be transported into the filopodia of the cells (K′). NP: When cells that were transfected with Sns/Blow are mixed with Duf/Rols-expressing cells, Sns can be found at the membrane (O′). QS: In cells that are transfected with Blow, a strong colocalization with the actin cytoskeleton can be observed as visualized by phalloidin staining.

Download figure to PowerPoint

Sns Only Localizes to the Cell Membrane When Duf-Positive C2C12 Myoblasts Are Present

We transfected cells with Sns/Blow (Fig. 6K–M) or Sns alone (data not shown) and observed accumulation of Sns in the cytoplasm of the cells in close proximity to the nucleus, which very likely represents protein trapped in the ER. When the transfected cells formed lamellipodia to make contact with other cells, Sns protein was detected within them. Even though we could detect this directed transport of Sns into the lamellipodia (arrow in Fig. 6K), we never observed Sns localization at the cell membrane.

Therefore, to check whether the interaction with the known interaction partner in Drosophila myoblast fusion, Duf/Kirre, might be sufficient to induce the localization of Sns within the cell membrane, we mixed cells that were transiently transfected with Duf/Kirre and Rols with cells that were transiently transfected with Sns and Blow. In some transfected cells, we observed Sns localization in the cell membrane (Fig. 6N–P). This localization of Sns seems not to depend on direct contact with a Duf/Kirre-positive cell, as none of them was detectable in close proximity to the Sns-positive cells. Therefore, this localization of Sns may be directed by the long, thin filopodia extending from the Duf/Kirre-expressing cells or by secreted Duf/Kirre. This process could mediate a signal over a distance of several cell diameters.

Blow Colocalizes With the Actin Cytoskeleton of Vertebrate Cells

We transfected C2C12 myoblasts with a Blow construct and observed that Blow colocalizes with the actin cytoskeleton (Fig. 6Q–S). It has already been proposed that Blow together with the actin-regulating factor Kette (Bogdan and Klämbt,2003; Bogdan et al.,2004) is involved in actin rearrangement during myoblast fusion (Schröter et al.,2006). This colocalization is not only observed in myoblasts but also in fibroblast cell culture (C3H10T1/2 cells; data not shown) or in Hek293 cells (Supplementary Figure S1, which can be viewed at http://www.interscience.wiley.com/jpages/1058-8388/suppmat) and is not altered upon cotransfection with Sns (data not shown).

In Transfection Studies, no Formation of FuRMASs Can Be Observed in Vertebrate Cells

To analyze whether the Drosophila proteins expressed in vertebrate cell culture are also able to establish FuRMASs, we transfected Hek293 cells (Graham et al.,1977) with either Rols/Duf or Sns/Blow. We used Hek293 cells because we could not obtain stable C2C12 cell clones that expressed our constructs. All four proteins localize in Hek293 cells as they do in C2C12 myoblasts (Supplementary Figure S1). Nevertheless, when a Rols/Duf-expressing cell was in contact with a Sns/Blow-expressing cell after mixing the two cell populations, we never observed protein accumulation in the area of contact (Supplementary Figure S1). Moreover, in the rare cases when Rols/Duf- and Sns/Blow-expressing C2C12 myoblasts were in contact with each other, we had similar observations (data not shown). Taken together, our data suggest that additional proteins must be involved in FuRMAS formation. We propose that in vertebrate cells at least the mediator between Sns and Blow is missing so that FuRMASs cannot be assembled.

DISCUSSION

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. RESULTS
  5. DISCUSSION
  6. EXPERIMENTAL PROCEDURES
  7. Acknowledgements
  8. REFERENCES
  9. Supporting Information

Membrane fusion is a fundamental biological process and occurs between (1) membranes of different cell compartments (intracellular fusion), (2) the plasma membrane of a cell and a virus, or (3) the plasma membranes of two cells (intercellular fusion). For membranes to fuse, the first critical events are cell recognition followed by cell adhesion. In most cases, cell adhesion does not lead to plasma membrane fusion; however, it remains to be resolved whether—in the case of membrane fusion—the merging of the membranes is intimately linked to the preceding event of cell adhesion.

The Adhesive Complex Preceding Cell Fusion During Drosophila Myogenesis Exhibits a Bipartite Architecture

During myogenesis in D. melanogaster, we identified a ring-shaped structure that connects the growing myotube with the fcm and leads to the accumulation of Rols on the site of the growing muscle. In the center of this ring, plugs of F-actin are present at both sites, that is, where the myotube meets the fcm, and the growing muscle. We termed this Drosophila complex FuRMAS, which stands for fusion-restricted myogenic-adhesive structure. Here, we demonstrated that the FuRMAS is a structure that assembles at cell-contact points and connects founder/precursor cells and fcms. The FuRMAS apparently expands during the course of myoblast fusion, because in wild-type embryos, we found FuRMASs of up to 5 μm in diameter, while in mutants with arrested myoblast fusion after adhesion large rings were rarely observed and the average diameter was less than 2 μm. The characteristic morphological feature of the FuRMAS is the bipartite architecture with a core of F-actin, the F-actin–interacting protein Titin as well as regulators of the actin filament system (e.g., Blow in fcms) ringed by adhesive proteins of the IgSF (Duf/Kirre and Sns) and associated proteins (e.g., Rols7 in founder cells).

As the ring widens, the contact area between growing myotubes and fcms is enlarged. We propose that F-actin branching within the FuRMAS forces the adhesion ring apart. A notion that is supported by our data that show that the diameter of the ring is not significantly increased in the mesoderm of embryos mutant for blow that genetically interacts with the F-actin regulator Kette (Schröter et al.,2004). Thus, the FuRMAS links adhesion molecules to molecules relevant for subsequent events and the rearrangement of the actin cytoskeleton. These events finally lead to merging of the plasma membranes (see below and Fig. 7 for a model).

thumbnail image

Figure 7. A model for myoblast fusion in Drosophila melanogaster. (1, 2a, 3a, and 4a show a lateral view; 2b, 3b and 4b, view from above). First, recognition between founder cell and fcm is mediated by Duf/Kirre and Sns (1). The cells adhere to each other, the FuRMAS assembles (boxed area 2a, 3a), and the vesicles of the prefusion complex are recruited to the membranes of both cell types (2a, 2b). During the expansion of the FuRMAS, the prefusion complex moves to a more lateral position with the adhesive and adapter molecules (4a), while the older (and most inner) parts of the prefusion complex build up the electron-dense plaques by exocytosis (3a, 3b). Membrane modifications most likely have to take place before the fusion pore can be formed (white area in 4b) with the vesiculating plasma membrane in the center of the complex at the next step (4b). While the adhesive molecules and the prefusion complex are localized at the lateral edge of the FuRMAS, the fusion pore should open in the center of the complex with the electron-dense plaques in close proximity (4a, 4b). The proteins involved in FuRMAS formation and localization of Blow and actin are indicated in 2a to 4a.

Download figure to PowerPoint

Dumbfounded Induces Filopodia Formation in Mammalian Cells

Upon transfection into mammalian cells, Duf/Kirre localizes to the cell membrane, where it induces the formation of multiple, thin filopodia that are not observable in cells that do not express Duf/Kirre. We propose that the Duf/Kirre-induced filopodia formation, which is also observable in the living Drosophila embryo during the stages of myoblast fusion (Schnorrer and Dickson,2004), is necessary to attract the fcms for the fusion process. The filopodia extend over several cell diameters, and in cell culture experiments, it seems as if small Duf/Kirre-positive vesicles are released from their tips. Considering the situation in the Drosophila embryo, this might be a mechanism that allows the somatic founder cells to even attract the fcms of the visceral mesoderm as seen in Jelly belly (Jeb) or Alk mutants (Lee et al.,2003; Stute et al.,2004). Furthermore, Duf/Kirre when expressed ectopically in epidermal cells of the Drosophila embryo lacking Duf/Kirre in the mesoderm is able to attract fcms (Ruiz-Gomez et al.,2000).

Additional Factors Seem to Be Necessary for FuRMAS Formation

In this report, we show that the Drosophila proteins Duf/Kirre and Rols localize in the mammalian cells in the same way as they do in the Drosophila embryo. For Sns to localize to the membrane, the presence of Duf/Kirre appears to be required. This finding is in contrast to what is observed in S2 cells where Sns protein is uniformly distributed on the surface in unaggregated cells (Galletta et al.,2004). In addition, in contrast to the behavior in myoblasts in the Drosophila embryo, none of the proteins concentrates at the points where a Duf/Rols-expressing “founder-like cell” is in contact with a Sns/Blow-expressing “fcm-like cell” in our experiments. Previously, in S2 cells first transfected with Sns or Duf and subsequently cocultured, the proteins concentrate heterotopically at the contact sites between two different protein-expressing cells (Galletta et al.,2004) or homotopically at the sites were two Duf/Kirre-expressing cells contact each other (Menon et al.,2005). The colocalization of Blow with the actin cytoskeleton, which we observed in our cell-culture experiments, fits well with the localization of Blow as a plug in the center of the FuRMAS where also actin is found in a plug-like pattern. Nevertheless, Blow is also not recruited to the points of cell–cell contact. We propose that at least the mediator between Sns and Blow in the mammalian cells is either absent or has diverged during evolution.

The Adhesive Complex Preceding Cell Fusion During Drosophila Myogenesis Exhibits a Bipartite Architecture With Striking Structural Similarities to Vertebrate Adhesive Complexes

A ring-shaped distribution of adhesive proteins was described for other structures that are involved in cell–cell and cell–matrix adhesion—the podosomes and the immunological synapse of vertebrate cells, and the FuRMAS in Drosophila myoblast fusion may represent a third structure.

The podosome is a structure known from vertebrate cell culture and is involved in cell migration, adhesion of the cell to the substratum, and the degradation of extracellular matrix components by the localized secretion of matrix metalloproteases (Linder and Aepfelbacher,2003; Johansson et al.,2004; Burgstaller and Gimona,2005; Linder and Kopp,2005). The podosome is characterized by its bipartite architecture that consists of a ring of integrins and associated proteins and has a F-actin core as well as by the rapid turnover of the structure (half-life 2–20 min). An additional feature of podosomes is that besides F-actin itself, the core of the complex is mainly composed of F-actin–interacting proteins (Linder and Kopp,2005). This observation parallels that of Titin accumulation in FuRMASs (Zhang et al.,2000; Menon and Chia,2001).

The immunological synapse also exhibits a striking bipartite architecture with an outer ring structure (pSMAC) composed of integrins (LFA-1) and associated proteins (Talin) or ICAM as well as a core (cSMAC) of either TCRs or MHCs. F-actin is dynamically distributed, as it can be observed at the center as well as the periphery of an activated synapse (Monks et al.,1998; Tskvitaria-Fuller,2003). In the case of the immunological synapses between CTL cells and their target cells, the actin-polymerization factor Arp2/3 clearly localizes to the center of the structure. For this special immunological synapse, plasma membrane fusion inside the borders of the structure has been reported (Stinchcombe et al.,2001a,b). Further research is essential to clarify whether the FuRMASs create a reaction center for membrane fusion analogous to the reaction center for T-cell activation in the immunological synapse.

A New Model for Plasma Membrane Fusion During Myogenesis in D. melanogaster

To establish a model for the process of plasma membrane fusion, we have combined our data on the FuRMAS with those obtained from examination of ultrastructural features, such as a prefusion complex consisting of electron-dense vesicles, the appearance of electron dense plaques, and membrane vesiculation in D. melanogaster (Doberstein et al.,1997; Rau et al.,2001; Schröter et al.,2004). Furthermore, we have also integrated information on the function of proteins known to be essential (for review, see Chen and Olson,2004; Abmayr et al.,2005) in membrane fusion events into our new model as presented here.

The first step in myoblast fusion is cell–cell recognition and adhesion mediated by Duf/Kirre and Sns (Fig. 7.1). In a second step, the FuRMAS assembles at the interface of the adhering cells (Fig. 7.1, 2a). In the third step, the prefusion complex is recruited to the contact point of the precursor cell with the fcm (Fig. 7, 2a,b). This step requires the action of Rols7, because in rols7 mutants, no prefusion complex was observed and the cell contacts are not stable (Rau et al.,2001). We obtained no evidence that the localization of Duf/Kirre itself depends on Rols7 as previously reported (Menon et al.,2005). In both founder cells and fcms, an adhesive ring with a plug of F-actin in its center was formed. In fcms, the F-actin area also contains Blow, which seems to be essential for the fourth step of myoblast fusion, the exocytosis of the electron-dense vesicles (Fig. 7, 3a, b), because in blow mutants, the fusion process is arrested after the formation of the prefusion complex (Fig. 7, 3a; Doberstein et al.,1997). Exocytosis most likely precedes the fusion of membranes and might deliver the content of the electron-dense vesicles to the membranes and the extracellular space. Recently, we showed that Blow genetically interacts with the actin regulator Kette/Hem (Schröter et al.,2004). Further data indicate that Blow possibly interacts with Crk, an interaction partner of the DOCK-180 family of GEFs for the small GTPase Rac that controls, for example, the exocytosis of vesicles in neurogenic synapses (Cote and Vuori,2002; Humeau et al.,2002; Giot et al.,2003). As Blow is solely expressed in fcms, actin filaments have to be differently regulated in founder and precursor cells as well as in the growing muscle. The GEF Schizo/Loner might have a similar function together with ARF6 by localizing active Rac at the plasma membrane in founder precursor cells (Chen et al.,2003; reviewed by Chen and Olson,2004).

The generation of the electron-dense plaques is followed by merging of the membranes and the establishment of a fusion pore at the contact sites of the cells. For membrane fusion to proceed, the fusion pore has to expand and the membranes must vesiculate into the pore (Fig. 7.3a–4b). On the ultrastructural level, this seems to be very similar to hypodermal cell fusion in C. elegans, were the expansion of the fusion pore follows the movement of the adherens junction and the membrane vesiculates inside the pore (Mohler et al.,1998). We suggest that the adhesive molecules of the FuRMAS also limit the width of the fusion pore, and it is very likely that the expansion of the fusion pore follows the observed expansion of the FuRMAS. Therefore, we propose that the adhesive molecules of the FuRMAS not only connect the cells with each other, but also build a network with other fusion-relevant components. In this model, the fusion machinery should be localized to the inner edge of the FuRMAS and expand with this structure, while in the center of the complex the fusion pore enlarges. We propose that this process depends on Blow and Kette (Schröter et al.,2004). The Kette protein is involved in F-actin polymerization (Bogdan and Klämbt,2003), so that expansion of the fusion pore might be directly connected to the reorganization of the F-actin cytoskeleton. In addition, in the growing myotube, Schizo and Arf6 have been proposed to induce actin rearrangement (Chen and Olson,2004).

The proposed model here, where the fusion of the membranes of myoblasts is constricted by the boundaries of an adhesive complex, makes the fusion event a highly controlled and ordered process leading to localized membrane fusion and integration of the fcm in the growing myotube. Thus, the FuRMAS concept of membrane fusion presented in this study will be a useful working hypothesis for further research in the field of plasma membrane fusion.

EXPERIMENTAL PROCEDURES

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. RESULTS
  5. DISCUSSION
  6. EXPERIMENTAL PROCEDURES
  7. Acknowledgements
  8. REFERENCES
  9. Supporting Information

Fly Stocks

The following fly strains were used in this study: Oregon R (wild-type), blow2/Cyo,hg-lacZ obtained from Doberstein et al. (1997); Df(3L)BK9, ru1red1cv-c1Sbsbd-1sr1e1/TM3, Sb1 (rols-Deficiency, Bloomington Stockkeeping Center), red1e1mbcC1/TM3, Sb1ry1e1 (mbc null allele; Rushton et al.,1995), sns202/CyO, hg-lacZ (sns null allele), kettej4-48/TM3 (kette null allele; Hummel et al.,2000), Df(1)w67k30, lz1ras1v1/FM7c (duf/kirre+rst/irreC-Deficiency; Bloomington Stockkeeping Center), If/CyO, hg-lacZ (Blue-Balancer for the II. Chromosome), and w; Dr/Sb, TDlZ (Blue-Balancer for the III. Chromosome).

Immunocytochemistry and Phalloidin Labeling of F-Actin on Drosophila Embryos

For immunocytochemistry, embryos of stages 12–16 where collected from grape juice agar plates, rinsed with TNX (0.7% NaCl and 0.01% Triton), dechorionized with 50% bleach, hot fixed in boiling phosphate buffered saline (PBS) for 10 sec, and then immediately cooled by adding ice-cold PBT. After fixation, embryos where devitellinized by shaking in 1:1 vol/vol methanol/heptane. After rehydration in PBT, embryos where incubated overnight at 4°C with primary antibodies in PBT at the following concentrations: anti-Rols (anti-Rolo2, this work), 1:250; anti-Duf/Kirre (Kreisköther et al.,2006), 1:1,000; anti-Sns, 1:5,000 (this work), and anti-Blow (Schröter et al.,2006), 1:250 preadsorbed. The primary antibody was detected with biotinylated anti-rabbit secondary antibodies (1:500) for 2 hr at room temperature. After amplifying the reaction using an ABC kit from Vectastain, stainings were carried out with diaminobenzidine, H2O2, and NiCl2. Specimens were mounted in Epon and examined under a light microscope.

For double labeling with fluorescent antibodies and phalloidin-TRITC, embryos of stages 12–15 where collected from grape juice agar plates and rinsed with TNX. After the fixation (20 min in 1:1 vol/vol heptane/4% formaldehyde in PBS), the embryos were mounted at the bottom of a black well with heptane glue, shortly dried, covered with PBT, and then the vitelline membrane was manually dissected with a sharp glass needle. The devitellinized embryos were incubated with phalloidin-TRITC for 10 min at room temperature. Antibodies were applied overnight at 4°C in the following concentrations: anti-Rols (Menon and Chia,2001), 1:1250; anti-Titin, 1:1,000 (Machado et al.,1998), and anti-βPS-integrin (DSHB, Iowa), 1:200. The primary antibodies were detected by using the corresponding fluorescence-labeled secondary antibodies. Specimens were mounted in Vectashield and examined under a confocal microscope.

Antibody Generation

To generate polyclonal antibodies specific for Sns, a prokaryotic His-tagged protein was generated by cloning the nucleotide region from codon 1225 to 1482 of the Sns cDNA sequence into the expression vector pRSET-B (Invitrogen). The fusion protein was expressed and purified using Ni-NTA agarose columns as recommended in the manufacturer's instructions (Qiagen). The fusion proteins were further purified by sodium dodecyl sulfate-polyacrylamide gel electrophoresis and blotted onto a nitrocellulose membrane. The immunogen was excised from the membrane, dissolved in dimethyl sulfoxide, and injected intraperitoneally into BALB/c mice and into a rabbit (New Zealand) in the presence of lipopeptide adjuvant P3CSK4 (vd Esche et al.,2000). Polyclonal sera were obtained 15 days after the third immunization. The anti–Rolo-2 antibody was generated by the Pineda Antibody service (Berlin) against the peptide NH2-AEEQLQEQDEDEEE-CONH2 in rabbits.

Constructs

For cell culture transfections, full-length coding regions from Sns, Rols, Blow, and Duf/Kirre were amplified by polymerase chain reaction and cloned into pIREShrGFP2a (Sns and Rols, Invitrogen) or pcDNA3 (Blow and Duf/Kirre, Stratagene).

Cell Culture

C2C12 cells were grown on plastic coverslips (Sarstaedt) in DMEM (low glucose, PAA) supplemented with 5% fetal bovine serum (Sigma) and transfected using the calcium phosphate precipitation method with the constructs described above. After 48 hr, cells were fixed with 3% paraformaldehyde/2% sucrose for 10 min, washed in PBS supplemented with 0.1 M glycine and blocked in PBS/2% normal goat serum/0.4% Triton for 20 min at room temperature. After washing in PBS/0.2% Triton/0.2% bovine serum albumin, the primary antibody (mouse anti-HA antibody [HA.11, Covance, Freiburg] 1:500; rabbit anti-Blow antibody [Schröter et al.,2006] 1:1000; rabbit anti-Kirre antibody [Kreisköther et al.,2006] 1:1,000; anti-Kette-antibody [Bogdan and Klämbt,2003] 1:200; anti Dock180 (N19) 1:50 and anti-Crk antibody (C20) 1:10 [Santa Cruz Biotechnology]) was incubated overnight at 4°C. After several washes, the Cy3- or Cy5-coupled secondary antibodies (Dianova, Hamburg) where applied for 2 hr at room temperature at a dilution of 1:100. After additional washes, the cells were either incubated with Hoechst (1:200, Sigma) and/or phalloidin- TRITC (1:20, Sigma) for 5 to 10 min or immediately mounted in Fluoromount G (Southern Biotechnology Associates, Birmingham, AL). The samples were examined using a Zeiss Axioplan 2 Imaging System.

Acknowledgements

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. RESULTS
  5. DISCUSSION
  6. EXPERIMENTAL PROCEDURES
  7. Acknowledgements
  8. REFERENCES
  9. Supporting Information

We thank Roxane Schröter for drawings and Ruth Hyland for technical assistance. We thank Barbara Winter, Hans-Henning Arnold (Braunschweig), Thomas Braun (Bad Nauheim), and Martin Klingenspor (Marburg) for cell lines. The anti-Kette antibody was a generous gift from Sven Bogdan and Christian Klämbt (Münster). We also thank Monika Hassel for sharing cell culture facilities. The anti–integrin-PSβ antibody developed by D. Brower was obtained from the Developmental Studies Hybridoma Bank developed under the auspices of the NICHD and maintained by the University of Iowa, Department of Biological Sciences, Iowa City, IA 52242. This research was supported by the Deutsche Forschungsgemeinschaft Re628/14-2 and the EU's Network of Excellence MYORES to R.R.-P. and the SFB 505 to K.F.F.

REFERENCES

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. RESULTS
  5. DISCUSSION
  6. EXPERIMENTAL PROCEDURES
  7. Acknowledgements
  8. REFERENCES
  9. Supporting Information
  • Abmayr SM, Balagopalan L, Galetta BJ, Hong SJ. 2005. Myogenesis and muscle development. In: GilbertLI, IatrouK, GillS, editors. Comprehensive molecular insect science. Vol. 2. Oxford: Elsevier Ltd. p 145.
  • Artero R, Furlong EE, Beckett K, Scott MP, Baylies M. 2003. Notch and Ras signaling pathway effector genes expressed in fusion competent and founder cells during Drosophila myogenesis. Development 130: 62576272.
  • Bogdan S, Klämbt C. 2003. Kette regulates actin dynamics and genetically interacts with Wave and Wasp. Development 130: 44274437.
  • Bogdan S, Grewe O, Strunk M, Mertens A, Klambt C. 2004. Sra-1 interacts with Kette and Wasp and is required for neuronal and bristle development in Drosophila. Development 131: 39813989.
  • Bökel C, Brow N. 2002. Integrins in development: moving on, responding to, and sticking to the extracellular matrix. Dev Cell 3: 311321.
  • Bour BA, Chakravarti M, West JM, Abmayr SM. 2000. Drosophila SNS, a member of the immunoglobulin superfamily that is essential for myoblast fusion. Genes Dev 14: 14981511.
  • Bunch TA, Salatino R, Engelsgjerd MC, Mukai L, West RF, Browner DL. 1992. Characterization of mutant alleles of myospheroid, the gene encoding the β subunit of the Drosophila PS integrins. Genetics 132: 519528.
  • Burgstaller G, Gimona M. 2005. Podosome-mediated matrix resorption and cell motility in vascular smooth muscle cells. Am J Physiol Heart Circ Physiol 288: 30013005.
  • Chen EH, Olson EN. 2001. Antisocial, an intracellular adaptor protein, is required for myoblast fusion in Drosophila. Dev Cell 1: 705715.
  • Chen EH, Olson EN. 2004. Towards a molecular pathway for myoblast fusion in Drosophila [review]. Trends Cell Biol 14: 452460.
  • Chen EH, Pryce BA, Tzeng JA, Gonzalez GA, Olson EN. 2003. Control of myoblast fusion by a guanine exchange factor, loner, and its effector ARF6. Cell 114: 751762.
  • Cote JF, Vuori K. 2002. Identification of an evolutionary conserved superfamily of DOCK180-related proteins with guanine nucleotide exchange activity. J Cell Sci 115: 49084913.
  • Doberstein SK, Fetter RD, Metha AY, Goodman CC. 1997. Genetic analysis of myoblast fusion: blown fuse is required for progression beyond the prefusion complex. J Cell Biol 136: 12491261.
  • Galletta BJ, Chakravarti M, Banerjee R, Abmayr SM. 2004. SNS: adhesive properties, localization requirements and ectodomain dependence in S2 cells and embryonic myoblasts. Mech Dev 121: 14551468.
  • Giot L, Bader JS, Brouwer C, Chaudhuri A, Kuang B, Li Y, Hao YL, Ooi CE, Godwin B, Vitols E, Vijayadamodar G, Pochart P, Machineni H, Welsh M, Kong Y, Zerhusen B, Malcolm R, Varrone Z, Collis A, Minto M, Burgess S, McDaniel L, Stimpson E, Spriggs F, Williams J, Neurath K, Ioime N, Agee M, Voss E, Furtak K, Renzulli R, Aanensen N, Carrolla S, Bickelhaupt E, Lazovatsky Y, DaSilva A, Zhong J, Stanyon CA, Finley RL Jr, White KP, Braverman M, Jarvie T, Gold S, Leach M, Knight J, Shimkets RA, McKenna MP, Chant J, Rothberg JM. 2003. A protein interaction map of Drosophila melanogaster. Science 302: 17271736.
  • Graham FL, Smiley J, Russell WC, Nairn R. 1977. Characteristics of a human cell line transformed by DNA from human adenovirus type 5. J Gen Virol 36: 5974.
  • Hakeda-Suzuki S, Ng J, Tzu J, Dietzl G, Sun Y, Harms M, Nardine T, Luo L, Dickson BJ. 2002. Rac function and regulation during Drosophila development. Nature 416: 438442.
  • Humeau Y, Popoff MR, Kojima H, Doussau F, Poulain B. 2002. Rac GTPase plays an essential role in exocytosis by controlling the fusion competence of the release sites. J Neurosci 22: 79687981.
  • Hummel T, Leifker K, Klambt C. 2000. The Drosophila HEM-2/NAP1 homolog KETTE controls axonal pathfinding and cytoskeletal organization. Genes Dev 14: 863873.
  • Johansson MW, Lye MH, Barthel SR, Duffy AK, Annis DS, Mosher DF. 2004. Eosinophils adhere to vascular cell adhesion molecule-1 via podosomes. Am J Respir Cell Mol Biol 31: 413422.
  • Kreisköther N, Reichert N, Buttgereit D, Hertenstein A, Fischbach KF, Renkawitz-Pohl R. 2006. Drosophila Rolling pebbles colocalises and putatively interacts with alpha-Actinin and the Sls isoform Zormin in the Z-discs of the sarcomere and with Dumbfounded/Kirre, alpha–Actinin and Zormin in the terminal Z-discs. J Muscle Res Cell Motil 27: 93106.
  • Lafuste P, Sonnet C, Chazaud B, Dreyfus PA, Gherardi RK, Wewer UM, Authier FJ. 2005. ADAM12 and alpha9beta1 integrin are instrumental in human myogenic cell differentiation. Mol Biol Cell 16: 861887.
  • Lee HH, Norris A, Weiss JB, Frasch M. 2003. Jelly belly protein activates the receptor tyrosine kinase Alk to specify visceral muscle pioneers. Nature 425: 507512.
  • Linder S, Aepfelbacher M. 2003. Podosomes: adhesion hot spots of invasive cells. Trends Cell Biol 13: 376385.
  • Linder S, Kopp P. 2005. Podomes at a glance. J Cell Sci 118: 20792082.
  • Machado C, Sunkel CE, Andrew DJ. 1998. Human autoantibodies reveal titin as a chromosomal protein. J Cell Biol 141: 321333.
  • Menon SD, Chia W. 2001. Drosophila Rolling Pebbles: a multidomain protein required for myoblast fusion that recruits D-Titin in response to the myoblast attractant Dumbfounded. Dev Cell 1: 691703.
  • Menon SD, Osman Z, Chenchill K, Chia W. 2005. A positive feedback loop between Dumbfounded and Rolling pebbles leads to myotube enlargement in Drosophila. J Cell Biol 169: 909920.
  • Mohler WA, Simske JS, Williams-Masson EM, Hardin JD, White JG. 1998. Dynamics and ultrastructure of developmental cell fusions in Caenorhabditis elegans hypodermis. Curr Biol 8: 10871090.
  • Monks CR, Freiberg BA, Kupfer H, Sciaky N, Kupfer A. 1998. Three-dimensional segregation of supramolecular activation clusters in T-cells. Nature 395: 8286.
  • Oenel S, Bolke L, Klämbt C. 2004. The Drosophila ARF6-GEF Schizo controls commissure formation by regulating Slit. Development 90: 25872594.
  • Paululat A, Burchard S, Renkawitz-Pohl R. 1995. Fusion from myoblasts to myotubes is dependent on the rolling stone gene (rost) of Drosophila. Development 121: 26112620.
  • Prokop A, Martin-Bermundo MD, Bate M, Brown NH. 1998. Absence of PS integrins or laminin A affects extracellular adhesion, intracellular assembly of hemiadherens and neuromuscular junctions in Drosophila embryos. Dev Biol 196: 5876.
  • Rau A, Buttgereit D, Holz A, Fetter R, Doberstein SK, Paululat A, Staudt N, Skeath J, Michelson AM, Renkawitz-Pohl, R. 2001. rolling pebbles (rols) is required in Drosophila muscle precursors for recruitment of myoblasts for fusion. Development 128: 50615073.
  • Ruiz-Gomez M, Coutts N, Price A, Taylor MV, Bate M. 2000. Drosophila dumbfounded: a myoblast attractant essential for fusion. Cell 102: 189198.
  • Rushton E, Drysdale R, Abmayr SM, Michelson AM, Bate M. 1995. Mutations in a novel gene, myoblast city, provide evidence in support of the founder cell hypothesis for Drosophila muscle development. Development 121: 19791988.
  • Schaefer G, Weber S, Holz A, Bogdan S, Renkawitz-Pohl R, Oenel S. 2006. The Wiskott-Aldrich Syndrome Protein WASP is essential for myoblast fusion in Drosophila. Dev Biol (in revision).
  • Schnorrer F, Dickson BJ. 2004. Muscle building; mechanisms of myotube guidance and attachment site selection [review]. Dev Cell 7: 920.
  • Schröter RH, Lier S, Holz A, Bogdan S, Klämbt C, Beck L, Renkawitz-Pohl R. 2004. kette and blown fuse interact genetically during the second fusion step of myogenesis in Drosophila. Development 131: 45014509.
  • Schröter RH, Buttgereit D, Beck L, Holz A, Renkawitz-Pohl R. 2006. Blown fuse regulates morphogenesis but not myoblast fusion of the circular visceral muscles in Drosophila. Differentiation 74: 608621.
  • Schwander M, Leu M, Stumm M, Dorchies OM, Ruegg UT, Schnittny J, Muller U. 2003. Beta1 integrins regulate myoblast fusion and sarcomere assembly. Dev Cell 4: 673685.
  • Stinchcombe JC, Barral DC, Mules EH, Booth S, Hume AN, Machesky LM, Seabra MC, Griffiths GM. 2001a. Rab27a is required for regulating secretion in cytotoxic T lymphocytes. J Cell Biol 152: 825834.
  • Stinchcombe JC, Bossi B, Booth S, Griffiths GM. 2001b. The immunological synapse of CTL contains a secretory domain and membrane bridges. Immunity 15: 751761.
  • Strünkelnberg M, Bonengel B, Moda LM, Hertenstein A, de Couet HG, Ramos RG, Fischbach KF. 2001. Rst and its paralogue kirre act redundantly during embryonic muscle development in Drosophila. Development 128: 42294239.
  • Stute C, Schimmelpfeng K, Renkawitz-Pohl R, Palmer RH, Holz A. 2004. Myoblast determination in the somatic and visceral mesoderm depends on Notch signalling as well as on milliways (miliAlk) as receptor for Jeb signalling. Development 131: 743754.
  • Taylor MV. 2003. Muscle differentiation: signalling cell fusion. Curr Biol 13: 964966.
  • Tskvitaria-Fuller I, Rozell AL, Yin HL, Wülfing C. 2003. Regulation of sustained actin dynamics by the TCR and costimulation as a mechanism of receptor localization. J Immunol 171: 22872295.
  • vd Esche U, Ayoub M, Pfannes SD, Muller MR, Huber M, Wiesmuller KH, Loop T, Humar M, Fischbach KF, Strunkelnberg M, Hoffmann P, Bessler WG, Mittenbuhler K. 2000. Immunostimulation by bacterial components: I. Activation of macrophages and enhancement of genetic immunization by the lipopeptide P3CSK4. Int J Immunopharmacol 22: 10931102.
  • Yaffe D, Saxel O. 1997. A myogenic cell line with altered serum requirements for differentiation. Differentiation 7: 159166.
  • Zhang Y, Featherstone D, Davis W, Rushton E, Broadie K. 2000. Drosophila D-titin is required for myoblast fusion and skeletal muscle striation. J Cell Sci 113: 31033115.

Supporting Information

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. RESULTS
  5. DISCUSSION
  6. EXPERIMENTAL PROCEDURES
  7. Acknowledgements
  8. REFERENCES
  9. Supporting Information

The Supplementary Material referred to in this article can be found at http://www.interscience.wiley.com/jpages/1058-8388/suppmat

FilenameFormatSizeDescription
jws-dvdy.21035.tif8255K Supplementary Figure 1 Duf, Rols, Sns, and Blow localize in Hek293 cells in the same way as they do in C2C12 cells. The actin cytoskeleton is labeled with Phalloidin-TRITC (red in B, C, E, F, Q, and R), the expressed proteins detected using antibodies as indicated. The nuclei of the cells are stained with H?chst (Blue in C, F, I, L, O and U). (A-C) Hek293 cells that are transfected with Duf change morphology and form long, thin filopodia that seem to release small Duf-positive vesicles (arrows in A and A?). (D-F) When Hek293 cells are transfected with a HA-tagged Rols construct, the protein can be detected in a dot-like pattern throughout the cytoplasm (arrow in F?). (G-I) In contrast, cotransfection with Rols and Duf leads to localization of Rols at the cell membrane (arrows in G and I?). (J-L) Sns does not localize at the membrane of Hek293 cells either when transfected alone or together with Blow. Nevertheless, Sns can be detected in the filopodia of the cells (arrow in K and K?). (M-O) When cells that are transfected with Sns and Blow are mixed with Duf- and Rols-expressing cells, Sns can be found at the membrane (arrow in M and M?). (P-R) In cells that are transfected with Blow, colocalization with the actin cytoskeleton can be observed as visualized by Phalloidin staining especially at the points where the cells form protrusions (arrow in P and R?). (S-U) When a Duf/Rols-expressing cell (marked with an asterix in T and U?) is in contact with a Sns/Blow-expressing cell (marked with an arrowhead in T and U?), the proteins do not concentrate at the points of cell-cell contact (arrow in S and U?).

Please note: Wiley Blackwell is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.