|Home | About | Journals | Submit | Contact Us | Français|
During embryonic development large numbers of apoptotic cells are generated and subsequently removed rapidly and efficiently by phagocytes. In this issue of Cell, Kurant et al. describe a novel Drosophila transmembrane protein, Six Microns Under (SIMU), expressed on the surface of glial cells and macrophages that recognizes apoptotic neurons and promotes phagocytosis upstream of another phagocytic receptor Draper in the developing CNS.
The terminal step in the apoptotic program is the phagocytic removal of the dying cell. The clearance of apoptotic cells achieves the orderly disposal of potentially harmful cellular remains from the dying cells, and many features of the apoptotic cell engulfment process are evolutionarily conserved. Apoptotic cell clearance is performed by tissue-resident neighboring cells and/or “professional” phagocytes such as macrophages and dendritic cells. Receptors on the phagocyte recognize specific markers (‘eat-me’ signals) on the surface of the apoptotic cell and this recognition transmits signals that lead to reorganization of the phagocyte cytoskeleton, engulfment and degradation of the dying cell. Over the past few years, a number of receptors that mediate engulfment of apoptotic cells have been identified in C. elegans, Drosophila, and in mammalian cells (Ravichandran and Lorenz). The multiple receptors and ligands that are linked to this process demonstrate the complexity of apoptotic cell engulfment and the need to further define the signaling pathways involved in this fundamental process.
In this issue of Cell, Kurant et al. report that glial cells in the developing Drosophila CNS efficiently engulf apoptotic neurons through the use of a novel engulfment receptor SIMU. Once the ventral nerve cord is ensheathed macrophages are excluded, but the sessile glial cells extend their cellular membranes to capture and efficiently clear the dying neurons. Similar bystander engulfment in the CNS has been reported (Freeman et al.; Sonnenfeld and Jacobs), but Kurant et al. demonstrate quite convincingly that apoptotic cell clearance in the developing CNS is accomplished chiefly by glia with little role for macrophages. To elucidate the genes involved in glial engulfment, a transcriptional profile of isolated glia was performed and a novel transmembrane protein they dubbed SIMU (“six microns under”) was identified. SIMU is strongly expressed on the surface of glial cells and macrophages and is upregulated during major waves of developmental apoptosis. In vivo, SIMU-deficient glia show a decreased ability to make contact with and engulf apoptotic cells; an attendant increase in the number of unengulfed apoptotic cells in the CNS is also observed, pointing to a role for SIMU in recognition of apoptotic neurons. In vitro, purified SIMU can bind to apoptotic but not viable S2 insect cells, indicating that the extracellular portion of SIMU recognizes a specific, as yet unidentified, moiety on apoptotic cells.
The simu gene is predicted to encode a 377 aa type I transmembrane protein possessing a short cytoplasmic tail, a transmembrane (TM) domain, along with four NIMROD-type EGF repeats and an EMILIN (EMI)-like domain in the extracellular portion. simu is a member of the recently identified nimrod family of genes, some of which are known to function in the uptake of bacteria. NIMROD proteins are characterized by a number of conserved cysteine-rich EGF-like repeats (called NIM repeats) separated by variable loops of 6–11 residues (Kurucz et al.). The presence of a transmembrane domain is variable within this family and thus these proteins may be membrane-anchored or secreted. Among the transmembrane nimrod genes, the cytoplasmic tail is usually short and devoid of known protein interaction motifs. Two other engulfment receptors, CED-1 in C. elegans and Draper in Drosophila, display similar domain arrangements, including a number of tandem EGF-like extracellular domains, with the primary difference being that these proteins have longer cytoplasmic tails that have been shown to serve as docking sites for signaling proteins (Awasaki et al.; Su et al.). simu was originally identified as ORF CG16876 within a cluster of 10 nimrod genes on chromosome 2 of Drosophila (Kurucz et al.). To date two other Drosophila nimrod genes, eater and NimC1, have been shown mediate hemocyte phagocytosis of Staphylococcus and E. coli in in a NIM repeat-dependent manner (Kocks et al.; Kurucz et al.). Interestingly, NIMROD-like features are also found in the mammalian engulfment proteins MEGF10 and Jedi, raising the possibility that NIMROD proteins play an evolutionarily conserved role in mammalian engulfment.
An interesting finding of Kurant et al. is that SIMU appears to function upstream of the engulfment protein Draper. Draper is expressed on the surface of glia and macrophages, and its function is required for the engulfment of apoptotic neurons and degenerating axons by glia in the fly CNS (Awasaki et al.; Freeman et al.; MacDonald et al.). At this point it is not clear how SIMU might cooperate with Draper during engulfment. Kurant et al. were unable to detect a physical association between SIMU and Draper, so it seems likely that other factors are required to link these proteins. However, these authors also make another intriguing observation that might provide a clue. In genetic rescue experiments of simu mutant flies with different forms of simu, they find that the extracellular EMI domain of SIMU is essential for optimal apoptotic cell clearance; interestingly, a mutant of SIMU lacking the transmembrane domain but carrying the EMI domain and is therefore soluble/secreted (SIMUΔTM) could also promote engulfment in these rescue studies. Since SIMU would normally be expressed on the phagocyte cell surface, this soluble SIMU (carrying the EMI domain required for recognition of apoptotic cells) would be expected to compete with recognition, rather than promoting it. An interesting explanation for this is that the EMI domains could be involved in homotypic adhesion or oligomerization between EMI-domain containing proteins (Figure 1). In such a case, the SIMUΔTM mutant could represent a ‘gain of function’ mutation that binds/activates other EMI-domain proteins and thereby promotes apoptotic cell clearance. Since the stoichiometry of SIMUΔTM is not known, it is possible that the SIMUΔTM itself may be an oligomer, and could help bridge a marker on apoptotic cells and another EMI domain receptor (such as Draper). Although the authors state that no apparent interaction between SIMU and Draper were found, it is unclear how this was tested and whether the isolated EMI domains of the two proteins were examined. Future experiments should be able to directly test this possibility.
Draper is the Drosophila orthologue of the C. elegans engulfment receptor CED-1 (Zhou et al.). Genetic and physical interactions place CED-1 and a cytoplasmic adapter CED-6 in the same engulfment pathway (Su et al.), leading to downstream signaling that promotes engulfment (Zhou et al.). Draper also interacts with the CED-6 orthologue in Drosophila (dCED-6) to promote apoptotic cell removal. However, the ligand for Draper, and precisely how Draper recognizes dying targets has been unclear (Awasaki et al.). Now Kurant et al also present an intriguing set of data and suggest an additional role for Draper that does not involve direct corpse recognition/internalization. They observe that SIMU-deficient glia fail to efficiently recognize and contact apoptotic neurons and have few internalized corpses; in contrast, Draper-deficient glia appear to touch and presumably engulf apoptotic neurons but show a perplexing deficiency in post-engulfment corpse degradation. The loss of both SIMU and Draper in glia leads to a phenotype similar to that of the SIMU-only mutant with unrecognized corpses, leading Kurant et al. to propose that SIMU functions as a corpse recognition/tethering receptor, while Draper has a role primarily in corpse degradation rather than recognition. This is a provocative notion that needs further clarification. Draper has clearly been shown to localize around dying neurons, and Draper deficient glia fail to move toward injured axons, strongly suggesting a role for Draper in recognition of corpses (Awasaki et al.; MacDonald et al.). Moreover, CED-1 in C. elegans has clearly been shown to localize around corpses, and a tailless form of CED-1 can encircle the corpses but fail to internalize them (Zhou et al.). Additionally, recent intriguing work by Zhou and colleagues suggest that CED-1 (and its downstream adapter CED-6) could play a role not only in the early steps of corpse recognition/internalization, but also in the subsequent steps of corpse degradation (Yu et al.). Thus, Draper could similarly play a role in recognition as well as subsequent steps of degradation. Then how could one reconcile the data of Kurant et al with a role for SIMU in tethering and the apparent role for Draper only post-internalization? The answer may lie in the unique properties of SIMU. It is possible that SIMU may function upstream of multiple engulfment receptors, with Draper being one of them. It is well documented that there are multiple engulfment receptors in mammals which function alone or in combination. Similarly, there are two redundant pathways in C. elegans, with CED-1 providing the receptor in one of the two pathways, and a second yet to be identified receptor upstream of the cytoplasmic CED-12/CED-5/CED-2 module. If the multiple-receptor paradigm can be extended to Drosophila, Draper-deficient flies could still have some level of engulfment that occurs through other receptors, but a critical role for Draper in corpse degradation cannot be fully overcome, hence the resultant phenotype.
Certainly the engulfment pathways as defined in model organisms and in mammals are becoming fascinatingly complex, and many more pieces of the puzzle remain to be discovered. The work of Kurant et al presented herein and the identification of SIMU as a new player provides some tantalizing possibilities toward understanding this fundamentally important problem of apoptotic cell clearance.