InteractiveFly: GeneBrief

Ced-12: Biological Overview | References

BIOLOGICAL OVERVIEW

Members of the CDM (CED-5, Dock180, Myoblast city) superfamily of guanine nucleotide exchange factors function as RAC activators in diverse processes that include cell migration and myoblast fusion. The SH3, DHR1 and DHR2 domains of Myoblast city (MBC) are essential for it to direct myoblast fusion in the Drosophila embryo, while the conserved DCrk-binding proline rich region is expendable. This study describes the isolation of Drosophila ELMO/CED-12, an ~82 kDa protein with a pleckstrin homology (PH) and proline-rich domain, by interaction with the MBC SH3 domain. ELMO has been shown to modulate the Rac activation by Dock180 by means of at least three distinct mechanisms: helping Dock180 stabilize Rac in its nucleotide-free transition state; relieving a self-inhibition of Dock180; and targeting Dock180 to the plasma membrane to gain access to Rac. Thus, Dock180 and ELMO function together as a bipartite GEF to optimally activate Rac, upon upstream stimulation, to mediate the engulfment of apoptotic cells and cell migration (Geisbrecht, 2008; Lu, 2006).

Mass spectrometry confirms the presence of an MBC/ELMO complex within the embryonic musculature at the time of myoblast fusion and embryos maternally and/or zygotically mutant for elmo exhibit defects in myoblast fusion. Overexpression of MBC and ELMO in the embryonic mesoderm causes defects in myoblast fusion reminiscent of those seen with constitutively-activated Rac1, supporting the previous finding that both the absence of and an excess of Rac activity are deleterious to myoblast fusion. Overexpression of MBC and ELMO/CED-12 in the eye causes perturbations in ommatidial organization that are suppressed by mutations in Rac1 and Rac2, demonstrating genetically that MBC and ELMO/CED-12 cooperate to activate these small GTPases in Drosophila (Geisbrecht, 2008).

Drosophila myoblast city (MBC), Caenorhabditis elegans CED-5, and vertebrate DOCK180, are closely related members of the evolutionarily conserved CDM family of proteins. They serve as key players in a signaling complex that includes the SH2- SH3 domain-containing adaptor protein CrkII/CED-2 and the PH domain containing protein ELMO/CED-12 (reviewed in Meller, 2005). This complex then acts at the membrane to relay signals to the small GTPase Rac1/CED-10. MBC/DOCK180/CED-5 function as non-conventional Guanine nucleotide Exchange Factors (GEFs) for Rac. Conventional GEFs bind to nucleotide-free Rac via a Dbl-homology (DH) domain, thereby facilitating exchange of GDP for GTP. DOCK180/CED-5, which lack DH domains, associate with nucleotide-free Rac through a conserved Dock-Homology Region (DHR2) (Brugnera, 2002; Cote, 2002). Deletion of this domain results in loss of Rac binding and the inability to direct formation of GTP-bound Rac (Brugnera, 2002; Geisbrecht, 2008 and references therein).

In addition to DHR2, CDM proteins have in common an N-terminal SH3 domain, a second Dock Homology Region (DHR1), and a C-terminal proline rich region. The C-terminal region directs interaction with the SH3 domain of CrkII/CED-2. The CrkII SH2 domain can then direct interaction with upstream proteins that are phosphorylated on tyrosine, such as transmembrane receptors and components of focal adhesions. In addition to membrane recruitment through Crk-related interactions, the C-terminal PH domain of ELMO/CED-12 can also mediate DOCK180 membrane localization (deBakker, 2004; Grimsley, 2004). The DHR1 region of DOCK180, which binds to phosphatidylinositol 3,4,5-triphosphate [PtdIns(3,4,5)P₃], is also required for its membrane localization (Cote, 2005; Kobayashi, 2001). The N-terminal SH3 domains of DOCK180 and CED-5 mediate interaction with the C-terminal proline-rich region of ELMO and CED-12, respectively (Lu, 2004; Lu, 2005; Wu, 2001). In vitro studies demonstrate that DOCK180 binding to Rac can be sufficient for its activation (Cote, 2005), but that this activation can be significantly enhanced by DOCK180 bound to ELMO (Brugnera, 2002; Katoh, 2003; Lu, 2004). Thus, the DOCK180/ELMO complex is a key component in CDM signaling to Rac (Geisbrecht, 2008).

In addition to extensive homology and conserved biochemical interactions between the DOCK180/CED-5 and ELMO/CED-12 protein families, complexes of these proteins perform similar biological functions. For example, genetic studies have shown that C. elegans CED-10 acts with CED-2/Crk, CED-5/DOCK180 and CED-12/ELMO to promote cell migration of the distal tip cells during development of the somatic gonad and engulfment of cell corpses following apoptosis. Membrane targeted DOCK180 increases cell spreading, and overexpression of wild-type DOCK180 in mammalian cells enhances cell migration and phagocytosis of apoptotic cells. Lastly, a reduction in wild-type DOCK180 or overexpression of mutant forms of DOCK180 decrease activated Rac and cause defects in cell spreading and cell migration (Geisbrecht, 2008).

As in C. elegans and vertebrates, Drosophila MBC interacts genetically with other molecules required for CDM pathway function. For example, mutations in mbc delay border cell migration and influence PVR mediated F-actin accumulation in the follicle cells during ovary development, reflecting its proposed role in the PVR-Rac pathway (Duchek, 2001). Studies utilizing RNAi constructs have demonstrated a genetic interaction between MBC, Drosophila Crk (DCrk) and ELMO in adult thorax closure (Ishimaru, 2004). In the Drosophila eye, in which Rac is required for proper actin organization, the effects of Rac1 overexpression are suppressed by loss of one copy of mbc. Rac1 was initially implicated in myoblast fusion in the embryo by the demonstration that overexpression of either dominant-negative or constitutively-active Rac1 interferes with myoblast fusion. More recently, loss-of-function studies have established that Rac1 and Rac2 play a redundant but essential role in this process. MBC is absolutely essential for myoblast fusion, where it colocalizes in founder myoblasts with the immunoglobulin superfamily member Duf/Kirre. MBC is also expressed in, and apparently required in, the fusion competent myoblasts. Structure/function analysis of MBC identified the protein domains that are essential for myoblast fusion. This analysis revealed that, like Dock180, the DHR1 domain binds to PtdIns(3,4,5)P₃ and is essential for myoblast fusion. The N-terminal SH3 domain and the putative Rac binding DHR2 domain are essential for MBC transgenes to rescue the myoblast fusion defects in mbc mutant embryos. Surprisingly, however, while the C-terminal proline-rich region directs a strong interaction with DCrk (Balagopalan, 2006; Ishimaru, 2004), this region is not required for MBC function in the embryonic musculature (Balagopalan, 2006). Thus, the canonical Crk-associated CDM pathway that has been described for other organisms and used in other Drosophila tissues does not appear to be used in Drosophila myoblast fusion. This finding raised the broader question of whether other components of the canonical pathway functioned along with MBC in the embryonic muscle, or whether MBC functions in this tissue through different protein interactions (Geisbrecht, 2008).

To address the involvement of Drosophila ELMO in signaling from MBC, its role in myoblast fusion, border cell migration and ommatidial organization has been isolated and characterized. ELMO is expressed in the ovary, where it interacts genetically with MBC and plays a role in border cell migration. Multiple loss-of-function alleles were generate and analyzed to demonstrate the importance of maternal and zygotic ELMO in myoblast fusion. It was confirmed, through a targeted mass spectrometry approach, that these proteins form a stable complex in the embryonic musculature. ELMO and MBC act cooperatively in the mesoderm, such that an excess of both causes serious defects in myoblast fusion reminiscent of those seen with excess Rac activity. Finally, it was established genetically that ELMO and MBC can cooperate to activate Rac GTPases in the adult eye (Geisbrecht, 2008).

Thus, the conserved PH-domain that is normally present within conventional GEFs is actually provided for CDM proteins by a separate protein family represented by ELMO/CED-12 (Brugnera, 2002; Lu, 2004). One member of each of these two protein families can combine to form an unconventional bipartite GEF. While the small adaptor protein Crk often forms a critical component of this complex, recent studies have suggested that both DOCK180 and MBC can function in its absence (Balagopalan, 2006; Tosello-Trampont, 2007]). Moreover, Crk has been shown to function in pathways that are totally independent of this bipartite GEF. Reminiscent of this diversity of interactions, recent studies have suggested that CDM and ELMO/CED-12 family proteins may also function through independent interactions. For example, DOCK180 is capable of activating Rac on its own and has positive effects on cell migration and phagocytosis, albeit enhanced by binding of ELMO (Brugnera; Katoh, 2003]; Lu, 2004), and ELMO interacts with radixin independent of its interaction with DOCK180 (Grimsley, 2005). Drosophila MBC functions in a wide variety of processes that include border cell migration in the ovary and myoblast fusion in the embryo. This study has demonstrated, through its biochemical and genetic analysis, that Drosophila elmo functions in concert with MBC in these processes. Like MBC, decreased levels of ELMO impair border cell migration and myoblast fusion (Bianco, 2007). Moreover, MBC interacts stoichiometrically in the mesoderm with ELMO. Coincident over-expression of this complex impairs myoblast fusion, reinforcing the model from constitutively-active Rac1 that excess active Rac1 also interferes with myoblast fusion. Co-expression of MBC and ELMO also impacts development of the adult eye, resulting in a rough eye phenotype that is suppressed by decreasing endogenous levels of Rac. These data indicate that MBC and ELMO function together in a complex, and as a RacGEF (Geisbrecht, 2008).

CDM family proteins are required for a diverse array of biological processes. If ELMO functions with MBC in these processes, one would expect it to be expressed in a temporal and spatial pattern coincident with that of MBC. Consistent with this expectation, RT-PCR throughout the fly life cycle, embryonic in situ hybridizations and antibody stainings at multiple stages of Drosophila development, reveal that elmo is broadly expressed. Interestingly, however, the MBC/ELMO complex may serve distinct roles in each of the tissues in which it is expressed. MBC and ELMO are required for migration of the border cells in the ovary (Bianco, 2007; Duchek, 2001); however, myoblast migration appears to occur normally in mbc mutant embryos, as the fusion competent cells in mbc mutants can be found clustered and aligned with the founder cells (Geisbrecht, 2008).

The CDM/ELMO(Ced-12) complexes in both vertebrate and C. elegans function upstream of Rac as unconventional bipartite GEFs to promote exchange of GDP for GTP in activation of monomeric GTPases (Meller, 2005). These studies support a similar role for Drosophila MBC/ELMO. First, during ommatidial development, the rough eye phenotype resulting from co-expression of MBC and ELMO can be suppressed by removing half the gene dosage contributed by Rac1 and Rac2. Also, overexpression of the MBC/ELMO complex is sufficient to provide enough GEF activity to overcome the effect of its sequestration by RacN17 in the eye. In both the musculature and eye, expression of neither MBC nor ELMO alone has a phenotypic consequence, yet co-expression of MBC and ELMO phenocopies activated Rac (Geisbrecht, 2008).

Notably, embryos that are completely lacking both maternal and zygotic elmo die before muscle development occurs, possibly reflecting an earlier role for the protein. Interestingly, however, the development of mutant embryos that lack both maternally and zygotically provided mbc continues until myogenesis (Balagopalan, 2006). These data suggest that, like its vertebrate counterparts, Drosophila ELMO has multiple binding partners. In addition to DOCK180, vertebrate ELMOs bind to three of the five additional CDM family members, suggesting that ELMO binding is a general feature of these proteins (Grimsley, 2004; Sanui, 2003). Based upon primary sequence homology, the fly genome contains at least four potential CDM superfamily members in addition to MBC. The predicted transcripts of two of these are most closely related to vertebrate DOCK9/11 and DOCK 7/8. The Drosophila protein most closely related to vertebrate DOCK4 has been reported as the protein product of the sponge locus, but has not been studied extensively. It remains to be seen whether Drosophila ELMO is capable of binding to these other CDM-like molecules, and functions in concert with them in other tissues. One such place to examine in this regard is the embryonic CNS, where ELMO expression is quite strong but MBC is strikingly low. Thus, alternative CDM/ELMO-like complexes may be present and required in different tissues throughout Drosophila development, or in the same tissues to regulate different GTPases (Geisbrecht, 2008).

The above studies, combined with recent reports of ELMO binding to non-CDM family members, may reflect a role for ELMO proteins in integrating signals from different pathways. Interaction of the N-terminal region of ELMO with RhoG is capable of translocating the ELMO/DOCK180 complex to the membrane to regulate neurite outgrowth and cell migration (Katoh, 2003). Simultaneously, the N-terminal region of ELMO can bind to both the inactive and active forms of the ERM protein radixin (Grimsley, 2005). Interestingly, this ELMO/radixin interaction does not affect the ability of the ELMO/DOCK180 complex to promote Rac activation. Hence, ELMO may be functioning at the membrane to regulate the actin cytoskeleton via Rac, while recruiting radixin and ERM family members to perform their recognized roles in cross-linking the actin cytoskeleton to the plasma membrane (Geisbrecht, 2008).

In addition to Rac activation via the CDM/ELMO proteins, the ARF (ADP-ribosylation factor) family of GTPases has been shown to signal through Rac. Both DOCK180 and ELMO colocalize with ARNO (an ARF-GEF) and overexpression of mutant forms of DOCK180 and ELMO mutants block ARNO-induced Rac activation (Santy, 2005). However, RhoG signaling does not seem to be required for the ARNO-ELMO activation of Rac. This suggests Rac activation is differentially regulated in cell or tissue-specific manners or that within a cell there are localized mechanisms defined by crosstalk between signaling pathways that are responsible for Rac membrane localization. Intriguingly, in the Drosophila musculature, expression of a dominant-negative ARF6 results in myoblast fusion defects while the corresponding ARF-GEF Loner/Schizo is required for membrane localization of Rac. More work is required to see if ARF6, Loner/Schizo, and the ELMO/MBC proteins function in a signaling pathway in Drosophila to activate Rac (Geisbrecht, 2008).

Numerous studies have established a crucial role for Rac activation in myoblast fusion. More than a decade ago, it was demonstrated that a dominant-negative form of Rac1 interferes with myoblast fusion. A genetic analysis of the small Rac GTPases proved that the fusion defects observed with dominant-negative Rac1 reflect the Rac1, Rac2 loss-of-function phenotype. The need for activated Rac in myoblast fusion was further supported by the phenotype of embryos mutant for mbc and elmo. Rac, in turn, likely functions to direct rearrangement of the actin cytoskeleton through regulation of the Arp2/3 complex via Kette and WAVE. Surprisingly, however, constitutively-active forms of Rac negatively impact this process in a manner that, on the surface, have the same phenotypic consequence as the absence of active Rac. The specific relevance of excess activated Rac1 to normal myoblast fusion remains unclear, and it cannot be ruled out that excess Rac interferes with other signaling pathways that do not normally impact myoblast fusion. However, perturbation of this phenotype also has the potential to uncover key components and regulators that contribute to the normal process (Geisbrecht, 2008).

These data establish, for example, that monomeric Rac GTPases are in excess, and are therefore available to be activated when the level of the MBC/ELMO GEF complex is increased. Under normal circumstances, then, the ability of the cell to activate this endogenous Rac remains low. Potential mechanisms for this include the direct regulation of MBC and/or ELMO levels through synthesis or turnover. Though little is known about the synthesis of either MBC or ELMO or their levels in the cell, it is intriguing that DOCK180 is ubiquitinated and this modification ensures its rapid turnover (Makino, 2006). Alternatively, GAPs may be present to ensure that Rac does not remain active. Finally, the MBC/ELMO pathway may be integrating with the loner-ARF6 associated pathway, in such a way that perturbation of MBC and ELMO is impacting Rac1 activity through components of the ARF6 pathway. Either way, the fact that this myoblast fusion phenotype is occurring in response to perturbation of the endogenous pathway by wild-type proteins in a stoichiometric manner suggests that it may be possible to modulate it in ways that provide insights into the myoblast fusion process (Geisbrecht, 2008).

Drosophila Importin-7 functions upstream of the Elmo signaling module to mediate the formation and stability of muscle attachments

Establishment and maintenance of stable muscle attachments is essential for coordinated body movement. Studies in Drosophila have pioneered a molecular understanding of the morphological events in the conserved process of muscle attachment formation, including myofiber migration, muscle-tendon signaling, and stable junctional adhesion between muscle cells and their corresponding target insertion sites. In both Drosophila and vertebrate models, integrin complexes play a key role in the biogenesis and stability of muscle attachments through the interactions of integrins with extracellular matrix (ECM) ligands. This study shows that Drosophila Importin7 (Dim7) is an upstream regulator of the conserved Elmo-Mbc-->Rac signaling pathway in the formation of embryonic muscle attachment sites (MASs). Dim7 is encoded by the moleskin (msk) locus and was identified as an Elmo-interacting protein. Both Dim7 and Elmo localize to the ends of myofibers coincident with the timing of muscle-tendon attachment in late myogenesis. Phenotypic analysis of elmo mutants reveal muscle attachment defects similar to that previously described for integrin mutants. Furthermore, Elmo and Dim7 interact both biochemically and genetically in the developing musculature. The muscle detachment phenotype resulting from mutations in the msk locus can be rescued by components in the Elmo-signaling pathway, including the Elmo-Mbc complex, an activated Elmo variant, or a constitutively active form of Rac. In larval muscles, the localization of Dim7 and activated Elmo to the sites of muscle attachment is attenuated upon RNAi knockdown of integrin heterodimer complex components. These results show that integrins function as upstream signals to mediate Dim7-Elmo enrichment to the MASs (Liu, 2013).

Previous studies have shown that Dim7 localizes to developing muscle-tendon insertion sites and removal of Dim7 has severe consequences in muscle attachment maintenance (Liu, 2011). The current studies extend these observations to elucidate the functional contribution of Dim7-Elmo in regulating Drosophila muscle attachment. The results show that Dim7 is an upstream adaptor protein that recruits Elmo in response to integrin adhesion and/or signaling. Thus, it is proposed that the spatial and temporal regulation Elmo-Mbc activity results in regulation of the Rac-mediated actin cytoskeleton changes at the MASs (Liu, 2013).

The 'myospheroid' phenotype in elmo or msk mutants resemble attachment defects first characterized in mutated genes that encode for integrins, ILK and Talin, and is not due to earlier developmental defects in myogenesis. A similar number of cells expressing the muscle differentiation factor DMef2 was present in elmo or msk mutants, indicating that muscle specification was not affected (Geisbrecht, 2008; Liu, 2011). Genes essential for muscle migration and targeting also lead to detached muscles. For example, in kon/perd or grip mutants, the early arrest of migrating myotubes resulting from defective migration eventually leads to a linkage failure between the muscle and tendon cells. In mutant embryos with reduced levels of Elmo or Dim7, the muscle detachment phenotype did not appear to result from muscle migration defects. First, the spatial-temporal accumulation of Elmo and Dim7 is developmentally regulated. Both proteins are not detected at the leading edges of migrating muscles, but begin to accumulate at MASs after stage 15. Second, failure of muscle ends contacting their corresponding attachment sites was not observed in elmo or msk mutants at late stage 15, when muscle migration was almost complete (Liu and Geisbrecht, 2011) (Liu, 2013).

Both membrane localization and Rac-dependent cell spreading of the uninhibited, active version of Elmo is enhanced compared to native WT Elmo in cultured mammalian cells (Patel, 2010). These in vitro results are in agreement with the current in vivo analysis, where ElmoEDE (a mutation that prevents the autoinhibitory interaction of Elmo) is enriched at larval muscle ends compared to the poor accumulation of full-length Elmo-YFP. This may reflect a potential regulatory mechanism controlling the subcellular localization of Elmo from the cytoplasm to the muscle ends upon the release of Elmo autoinhibition. Within different cells or tissues, various proteins may regulate Elmo localization to the cell periphery, or other sites where active Elmo is needed. In cultured mammalian epithelial cells, membrane recruitment of the Elmo-Dock180 complex is dependent on active RhoG for cell spreading. Consistent with a functional role for membrane-targeted Elmo, active Elmo promotes cell elongation in Hela cells, when co-expressed with RhoG (Patel, 2010; Liu, 2013 and references therein).

The data argues that adaptor proteins may be required in muscle cells for activated Elmo membrane recruitment. Decreased levels of ElmoEDE are observed at the polarized ends of muscle insertion sites when Dim7 levels are decreased. It is still not clear if Dim7 binding is required for the conformational change that results in Elmo activation or if an activated Dim7-Elmo complex already exists within the cell and gets recruited as a complex upon integrin activation. Furthermore, complete loss of Elmo-EDE protein levels is not observed, suggesting that either Dim7 protein levels are not depleted enough or other proteins in addition to Dim7 play a role in Elmo membrane recruitment. Alternatively, post-translational modification(s), such as phosphorylation, could be an additional mechanism for the relief of Elmo autoinhibition. Thus, it is concluded that in muscle, Dim 7 is an essential adaptor protein for the polarized membrane localization of active Elmo or the active Elmo-Mbc complex downstream of integrin signaling pathway (Liu, 2013).

What is the relationship between the integrin adhesome and the Dim7-Elmo complex? Two explanations are proposed that are not mutually exclusive. One possibility is that the Dim7-Elmo-Mbc complex assembles at MASs via integrin-mediated 'outside-in' signaling. Upon ligand binding to ECM molecules, integrin activation results in Dim7-Elmo-Mbc complex localization for the spatial-temporal regulation of Rac activity to maintain dynamic actin filament adhesion at the MASs. It is predicted that localization of activated Elmo to the MASs is a prerequisite regulatory mechanism for actin cytoskeleton remodeling via Rac to maintain stable attachments. This hypothesis is supported by three lines of evidence: (1) muscle attachment defects upon loss of Dim7 or Elmo are only observed after the establishment of the integrin adhesion complex and onset of muscle contraction; (2) muscle detachment in msk mutants can be rescued by expressing low levels of activated Rac; and (3) the enrichment of Dim7 and Elmo-EDE proteins at the ends of muscle fibers is greatly reduced in integrin-deficient larvae (Liu, 2013).

It is also possible that accumulation of the Dim7-Elmo complex to the ends of muscles regulates 'inside-out' signaling to dynamically regulate integrin affinity for strong ligand binding and stable muscle attachments. Previously, it was reported that Dim7 acts upstream of the Vein-Egfr signaling pathway in muscle to tendon cell signaling (Liu, 2011). Combined with previous results that muscle-specific Vein secretion is dependent on the adhesive role of βPS integrin, the Dim7-Elmo complex may be internally required for integrins to regulate Vein secretion. A decrease in Vein-Egfr signaling and loss of tendon cell terminal fate results in a reduction in ECM secretion and weakened integrin-ECM attachment. This is consistent with the observation that msk or elmo mutants phenocopy embryos with reduced or excessive amounts of the αPSβPS integrin complex, where pointed muscle ends result in smaller muscle attachments. Future studies analyzing Dim7-Elmo-Mbc complex localization and function in the background of integrin deletion constructs which separate the 'inside-out' and 'outside-in' signaling pathways will be essential to uncover more detailed molecular mechanisms (Liu, 2013).

What is the relationship between the Dim7-Elmo-Mbc-->Rac signaling pathway and the integrin mediated adhesome complex assembly (including the Talin, IPP [integrin linked kinase (ILK)-PINCH-Parvin-α) complex]? It is proposed that actin filaments within the muscle cell are anchored to the muscle cell membrane via the IPP complex, while regulation of MAS-actin remodeling iscontrolled by the Dim7-Elmo-Mbc-->Rac pathway. The data suggests that these two complexes assemble independently at the muscle ends. In msk mutant embryos, both ILK and Talin properly accumulate at the MASs, suggesting that Dim7 is not responsible for their localization (Liu, 2011). Similarly, both MAS-enriched Dim7 and active Elmo can be detected at two ends of the muscles in ILK-deficient larva, even in fully detached muscles. In a vertebrate cell culture model, Elmo2 was found to physically interact with ILK for the establishment for cell polarity (Ho and Dagnino, 2012; Ho, 2009). Thus, it is possible that the current approaches have not fully knocked down Ilk levels or that the Dim7-Elmo recruitment by Ilk is redundant with another attachment site protein. Alternatively, an upstream scaffold protein may function to recruit both the IPP and Dim7-Elmo complex to the MASs. It is likely that these two complexes are temporally regulated in embryogenesis, where the actin remodeling complex is not needed until initial muscle-tendon initiation has been established (Liu, 2013).

Lu, T. Y., Doherty, J. and Freeman, M. R. (2014). DRK/DOS/SOS converge with Crk/Mbc/dCed-12 to activate Rac1 during glial engulfment of axonal debris. Proc Natl Acad Sci U S A. PubMed ID: 25099352

DRK/DOS/SOS converge with Crk/Mbc/dCed-12 to activate Rac1 during glial engulfment of axonal debris

Nervous system injury or disease leads to activation of glia, which govern postinjury responses in the nervous system. Axonal injury in Drosophila results in transcriptional upregulation of the glial engulfment receptor Draper; there is extension of glial membranes to the injury site (termed activation), and then axonal debris is internalized and degraded. Loss of the small GTPase Rac1 from glia completely suppresses glial responses to injury, but upstream activators remain poorly defined. Loss of the Rac guanine nucleotide exchange factor (GEF) Crk/myoblast city (Mbc)/dCed-12 has no effect on glial activation, but blocks internalization and degradation of debris. This study shows that the signaling molecules Downstream of receptor kinase (DRK) and Daughter of sevenless (DOS) (mammalian homologs, Grb2 and Gab2, respectively) and the GEF Son of sevenless (SOS) (mammalian homolog, mSOS) are required for efficient activation of glia after axotomy and internalization/degradation of axonal debris. At the earliest steps of glial activation, DRK/DOS/SOS function in a partially redundant manner with Crk/Mbc/dCed-12, with blockade of both complexes strongly suppressing all glial responses, similar to loss of Rac1. This work identifies DRK/DOS/SOS as the upstream Rac GEF complex required for glial responses to axonal injury, and demonstrates a critical requirement for multiple GEFs in efficient glial activation after injury and internalization/degradation of axonal debris (Lu, 2014).

Two distinct modes of guidance signalling during collective migration of border cells

Although directed migration is a feature of both individual cells and cell groups, guided migration has been studied most extensively for single cells in simple environments. Collective guidance of cell groups remains poorly understood, despite its relevance for development and metastasis. Neural crest cells and neuronal precursors migrate as loosely organized streams of individual cells, whereas cells of the fish lateral line, Drosophila tracheal tubes and border-cell clusters migrate as more coherent groups. This study used Drosophila border cells to examine how collective guidance is performed. It is reported that border cells migrate in two phases using distinct mechanisms. Genetic analysis combined with live imaging shows that polarized cell behaviour is critical for the initial phase of migration, whereas dynamic collective behaviour dominates later. PDGF- and VEGF-related receptor and epidermal growth factor receptor act in both phases, but use different effector pathways in each. The myoblast city (Mbc, also known as DOCK180) and engulfment and cell motility (ELMO, also known as Ced-12) pathway is required for the early phase, in which guidance depends on subcellular localization of signalling within a leading cell. During the later phase, mitogen-activated protein kinase and phospholipase Cγ are used redundantly, and it was found that the cluster makes use of the difference in signal levels between cells to guide migration. Thus, information processing at the multicellular level is used to guide collective behaviour of a cell group (Bianco, 2007).

Border cells perform a well-defined, invasive and directional migration during Drosophila oogenesis. They delaminate from the follicular epithelium at the anterior end of an egg chamber and migrate posteriorly, towards the oocyte, as a compact cluster. They then migrate dorsally towards the oocyte nucleus. The border-cell cluster consists of about six outer migratory border cells and two inner polar cells that induce migratory behaviour in the outer cells but seem to be non-migratory. Two receptor tyrosine kinases (RTKs), PDGF- and VEGF-related receptor (PVR) and epidermal growth factor receptor (EGFR), are guidance receptors for border cells. Both receptors act redundantly during posterior migration towards the oocyte, whereas EGFR and its dorsally localized ligand, Gurken, are essential for dorsal migration. Localized signalling from the RTKs is important and actively maintained, especially early in migration. Rac and the atypical Rac exchange factor Mbc (myoblast city, also known as DOCK180) are important effectors. To determine the contribution of Mbc and related proteins, a loss-of-function allele of their common cofactor ELMO (engulfment and cell motility, also known as Ced-12) was generated by homologous recombination. Clusters of elmo mutant border cells arrested early in migration, a defect that could be rescued by expressing elmo complementary DNA. As for mbc, reduction in elmo function suppressed F-actin accumulation caused by constitutive PVR signalling, placing ELMO downstream of the receptor in this respect (Bianco, 2007).

To determine whether later steps in migration also depend on ELMO, mosaic border-cell clusters consisting of wild-type and mutant cells were investigated. If a mutation does not affect migration, mutant cells should be distributed randomly within the cluster. Mutant cells defective in migration would be in the rear, 'carried along' by normal cells. As expected, Pvr and Egfr double mutant cells were in the rear during posterior migration, as were Egfr mutant cells during dorsal migration, reflecting the requirements at each stage. elmo mutant cells were in the rear during the initial migration, but were equally frequent in the leading position during dorsal migration. This indicates that, although ELMO is essential for the early-phase signalling, the later phase of migration does not require the Mbc-ELMO complex (Bianco, 2007).

To understand late guidance signalling, EGFR signalling, on which dorsal migration depends, was dissected. Uniformly activated EGFR, like PVR, dominantly impairs migration. The carboxy-terminal tail of EGFR was essential for this activity. Systematic mutagenesis of all docking tyrosines to phenylalanine identified Y1357 as being critical, with minor contributions from Y1405 and Y1406. Other tyrosines, including Y1095 in the conserved activation loop (phosphorylated in HER2 (Human EGF Receptor 2), were not required. Twenty Src-homology 2- and phosphotyrosine-binding-containing signalling molecules were tested for binding to active EGFR and tyrosine mutants. Y1357 was necessary and sufficient for binding of the adaptor protein Shc and its phosphotyrosine-binding domain. No other tested interactor behaved in this way. Binding was confirmed by immunoprecipitation. Border cells mutant for Shc showed no dorsal migration and, when PVR signalling was also blocked, these cells showed severely impaired posterior migration. This phenotype is identical to that of Egfr mutant cells, suggesting that Shc is essential immediately downstream of EGFR for guidance signalling (Bianco, 2007).

The Shc adaptor protein links EGFR and other RTKs to mitogen-activated protein kinase (MAPK) kinase signalling as well as to other classical downstream pathways. Raf, phospholipase Cγ (PLC-γ) or phosphatidylinositol-3-OH-kinase are not uniquely required for migration; however, the pathways might act redundantly. Simultaneous perturbation of PLC-γ and Raf impaired migration, with no effect of phosphatidylinositol-3-OH kinase. Double mutant border-cell clusters, cell-autonomously lacking PLC-γ and Raf or lacking PLC-γ and MAPK kinase (MAPKK), initiated migration but were delayed later in posterior migration and showed no dorsal migration. This phenotype is more severe than that of Egfr or Shc alone, suggesting that both RTKs might be affected. Prevention of PVR activity in double mutant cells did not block posterior migration, confirming that the requirement for these pathways was stage-specific and not EGFR-specific. Finally, analysis of mosaic clusters showed that Raf/MAPK and PLC-γ were important in late migration, reciprocal to the requirement for elmo. These results genetically define two migratory phases: an early posterior phase requiring ELMO-Mbc and a later posterior and dorsally directed phase requiring Raf/MAPK or PLC-γ. Both RTKs shift effector-pathway-dependency as migration progresses (Bianco, 2007).

To investigate the different migratory phases, border-cell migration was examined via live imaging. Appropriate conditions were establised for culturing and imaging of egg chambers, considering only active, growing ones. Border cells were selectively labelled with green fluorescent protein (GFP) and all membranes were labelled with the vital dye FM4-64. For all 24 wild-type samples, the identity of the front cell changed during the observation period, confirming the inference from fixed samples that cells change position during migration. This indicates that there is no determined front-cell fate. A clear difference was observed in behaviour of clusters during early (first half) and late phases. Early clusters had one, sometimes two, highly polarized cells clearly leading the migration; once these cells delaminated they moved straight and relatively fast. Weakly stained extensions protrude far from delaminating cells and subsequently shorten during movement, suggesting a 'grapple and pull' mechanism. Midway towards the oocyte, strong polarization was lost and cells rounded and started to 'shuffle' while dynamically probing the environment with short extensions. Occasionally the cluster would rotate or 'tumble' completely. This shuffling behaviour still provided effective movement of the cluster towards the oocyte and dorsally, albeit more slowly. Labelling cells with nuclear GFP allowed visualization of changes in positions within the cluster. The front cell exchanged, on average, every 18 min (Bianco, 2007).

As expected, positions corresponding to the second, slower phase of migration were more represented when cluster position along the migratory path was quantified in fixed samples. Also, border cells expressing dominant negative PVR and EGFR were individually active but provided little net cluster movement, as expected from the lack of guidance information. Finally, uniform overexpression of the attractant PVF1 caused an increased shuffling behaviour in the early phase but allowed slow forward movement, resembling normal late migration. This indicates that migrating clusters can interpret a shallow gradient when using the shuffling mode. It also suggests that the normal change in migratory behaviour midway into posterior migration might be triggered by the higher concentration of ligands closer to the oocyte (Bianco, 2007).

The early phase of migration with a highly polarized front cell corresponds temporally to the genetic requirement for ELMO activity. During the later phase, individual elmo mutant cells can alternate with wild-type cells in the lead position. Genetic analysis showed that Raf and MAPKK and, by inference, MAPK activation was sufficient to convey late guidance information. This was puzzling because MAPK activation appeared uniform in migrating border cells, and localized effects are usually a hallmark of guidance signalling. However, signalling that is not localized within an individual cell could still transmit spatial guidance information to the cell cluster if the cell with higher overall signalling indicates the direction of subsequent migration for the whole cluster, as observed for MAPK signalling in border cells. In this 'collective guidance' scenario, each cell of the cluster can be thought of as being analogous to a sector of an individual guided cell. Different levels of signalling in individual cells of the cluster transform into migration vectors because border cells adhere to each other and these contacts differ from substrate contacts. The occasional tumbling of border-cell clusters emphasizes the ability of these cells to behave as a collective unit. Tumbling may help single guided cells to 'reassess' their environment (Bianco, 2007).

To test this model for guidance, the relative levels of signalling in individual cells of the cluster were manipulated. Dynamic shuffling should allow cells to constantly 'compete' for the front position. None of the manipulations discussed below improved migration if all cells in a cluster were affected. Individual border cells with moderately elevated levels of PVR or EGFR were preferentially in the front relative to wild-type cells. Cells with elevated PVR tended to stay in or near the front position, suggesting that they were not competed away by other cells. This bias was ligand-dependent, because reducing PVF1 levels shifted the bias from PVR to EGFR, as was also shown by analysis of dorsal migration. Thus, increased signalling gives a cell-front bias when measuring an informative ligand. Elevating intracellular signalling levels had similar effects, whether by overexpression of an active form of Raf or by preventing downregulation of signalling as in Hrs mutant cells, in which RTK-mediated MAPK signalling is elevated in enlarged endosomes. The more modest front bias in Hrs mutant cells was reflected in behaviour: they could be displaced from the front. The E3 ubiquitin ligase Cbl negatively regulates RTK signalling and is also required to maintain localized RTK signalling within border cells initiating migration. Cbl mutant cells shifted from being preferentially at the back during early stages to being in the front during later migration. This indicates a transition from a mode requiring Cbl-dependent localization of signalling within the leading cell to a mode based on collective decisions within the cluster, in which Cbl mutant cells have an advantage owing to elevated RTK signalling (Bianco, 2007).

It is suggested that guidance of border-cell migration is achieved by two means: signalling localized within the cell, as used in individual migrating cells, and collective guidance, whereby the cluster uses differences in signalling strength among its constituent cells to determine direction. The two modes use the same guidance cues and receptors, but different downstream effectors. Localized signalling is required for the initial, polarized rapid migration, whereas collective behaviour, though observable throughout, dominates in the later phase. Collective decisions on the basis of differences in RTK signalling strength are important in Caenorhabditis elegans vulval development and in branching of Drosophila tracheal tubes, in which they result in specification of discrete cell fates. This differs from the dynamic situation reported in this study, in which the identity of the leading cell constantly changes. Indeed, the frequent exchange of leading cells suggests that front behaviour is normally temporarily restricted, possibly by induced inactivation of signalling. Such dynamics may allow the cluster to better reassess the environment. For guided migration of cell groups, this analysis indicates that sensing and regulation happens both at the single cell level and at the next level-that of collective cell decisions (Bianco, 2007).

Search PubMed for articles about Drosophila Ced-12

Balagopalan, L., et al. (2006). The CDM superfamily protein MBC directs myoblast fusion through a mechanism that requires phosphatidylinositol 3,4,5-triphosphate binding but is independent of direct interaction with DCrk. Mol. Cell. Biol. 26: 9442-9455. PubMed ID: 17030600

Bianco, A., et al. (2007). Two distinct modes of guidance signalling during collective migration of border cells. Nature 448: 362-365. PubMed ID: 17637670

Brugnera, E., et al. (2002). Unconventional Rac-GEF activity is mediated through the Dock180-ELMO complex. Nat. Cell Biol. 4: 574-582. PubMed ID: 12134158

Cote, J. F. and Vuori, K. (2002). Identification of an evolutionarily conserved superfamily of DOCK180-related proteins with guanine nucleotide exchange activity. J. Cell Sci. 115: 4901-4913. PubMed ID: 12432077

deBakker, C. D., et al. (2004). Phagocytosis of apoptotic cells is regulated by a UNC-73/TRIO-MIG-2/RhoG signaling module and armadillo repeats of CED-12/ELMO. Curr. Biol. 14: 2208-2216. PubMed ID: 15620647

Duchek, P., et al. (2001). Guidance of cell migration by the Drosophila PDGF/VEGF receptor, Cell 107: 17-26. PubMed ID: 11595182

Geisbrecht, E. R., et al. (2008). Drosophila ELMO/CED-12 interacts with Myoblast city to direct myoblast fusion and ommatidial organization. Dev. Biol. 314(1): 137-49. PubMed ID: 18163987

Grimsley, C. M., et al. (2004). Dock180 and ELMO1 proteins cooperate to promote evolutionarily conserved Rac-dependent cell migration. J. Biol. Chem. 279: 6087-6097. PubMed ID: 14638695

Ho, E., Irvine, T., Vilk, G. J., Lajoie, G., Ravichandran, K. S., D'Souza, S. J. and Dagnino, L. (2009). Integrin-linked kinase interactions with ELMO2 modulate cell polarity. Mol Biol Cell 20: 3033-3043. PubMed ID: 19439446

Ishimaru, S., et al. (2004). PVR plays a critical role via JNK activation in thorax closure during Drosophila metamorphosis. EMBO J. 23: 3984-3994. PubMed ID: 15457211

Katoh, H. and Negishi, M. (2003). RhoG activates Rac1 by direct interaction with the Dock180-binding protein Elmo. Nature 424: 461-464. PubMed ID: 12879077

Kobayashi, S., et al. (2001). Membrane recruitment of DOCK180 by binding to PtdIns(3,4,5)P3. Biochem. J. 354: 73-78. PubMed ID: 11171081

Liu, Z. C., and Geisbrecht, E. R. (2011). Moleskin is essential for the formation of the myotendinous junction in Drosophila. Dev. Biol. 359: 176-189. PubMed ID: 21925492

Liu, Z. C., Odell, N. and Geisbrecht, E. R. (2013). Drosophila Importin-7 functions upstream of the Elmo signaling module to mediate the formation and stability of muscle attachments. J Cell Sci. PubMed ID: 24046451

Lu, M., et al. (2004). PH domain of ELMO functions in trans to regulate Rac activation via Dock180. Nat. Struct. Mol. Biol. 11: 756-762. PubMed ID: 15247908

Lu, M., et al. (2005). A Steric-inhibition model for regulation of nucleotide exchange via the Dock180 family of GEFs. Curr. Biol. 15: 371-377. PubMed ID: 15723800

Lu, M. and Ravichandran, K. S. (2006). Dock180-ELMO cooperation in Rac activation. Methods Enzymol. 406: 388-402. PubMed ID: 16472672

Makino, Y., et al. (2006). Elmo1 inhibits ubiquitylation of Dock180. J. Cell Sci. 119: 923-932. PubMed ID: 16495483

Meller, N., et al. (2005). CZH proteins: a new family of Rho-GEFs. J. Cell Sci. 118: 4937-4946. PubMed ID: 16254241

Patel, M., Margaron, Y., Fradet, N., Yang, Q., Wilkes, B., Bouvier, M., Hofmann, K. and Cote, J. F. (2010). An evolutionarily conserved autoinhibitory molecular switch in ELMO proteins regulates Rac signaling. Curr Biol 20: 2021-2027. PubMed ID: 21035343

Santy, L. C., et al. (2005). The DOCK180/Elmo complex couples ARNO-mediated Arf6 activation to the downstream activation of Rac1. Curr. Biol. 15: 1749-1754. PubMed ID: 16213822

Sanui, T., et al. (2003). DOCK2 regulates Rac activation and cytoskeletal reorganization through interaction with ELMO1. Blood 102: 2948-2950. PubMed ID: 12829596

Tosello-Trampont, et al. (2007). Identification of two signaling submodules within the CrkII/ELMO/Dock180 pathway regulating engulfment of apoptotic cells. Cell Death Differ. 14(5): 963-72. PubMed ID: 17304244

Wu, Y. C. and Horvitz, H. R. (1998). C. elegans phagocytosis and cell-migration protein CED-5 is similar to human DOCK180. Nature 392: 501-504. PubMed ID: 9548255

date revised: 25 October 2014

Home page: The Interactive Fly © 2008 Thomas Brody, Ph.D.

The Interactive Fly resides on the
Society for Developmental Biology's Web server.