Avoiding the SCAMs
2007; Cell Press; Volume: 54; Issue: 3 Linguagem: Inglês
10.1016/j.neuron.2007.04.018
ISSN1097-4199
Autores Tópico(s)Axon Guidance and Neuronal Signaling
ResumoDendrites from the same neuron usually avoid contact with one another, a behavior known as self-avoidance. In this issue of Neuron and in the upcoming May 4, 2007 issue of Cell, a pair of studies by Soba et al. and Hughes et al. and a study by Matthews et al., respectively, identify products from the highly alternatively spliced Dscam gene as central to this behavior in Drosophila. Signaling induced by adhesion between identical isoforms triggers repulsion between sister dendrites. Dendrites from the same neuron usually avoid contact with one another, a behavior known as self-avoidance. In this issue of Neuron and in the upcoming May 4, 2007 issue of Cell, a pair of studies by Soba et al. and Hughes et al. and a study by Matthews et al., respectively, identify products from the highly alternatively spliced Dscam gene as central to this behavior in Drosophila. Signaling induced by adhesion between identical isoforms triggers repulsion between sister dendrites. Tree-like patterns exist widely in nature and are thought to provide optimal flow of information and/or material from a relatively large space to a point. Neuronal dendrites receive and integrate information from multiple sources, often covering large areas. The term dendrite is derived from dendron, the Greek word for tree, reflecting the characteristic patterns of these structures. Dendritic branches from the same neuron would be expected to be most efficient in territorial coverage when they are regularly spaced from one another, and indeed this is what is seen in vivo. This tendency of sister dendrites to avoid contact or crossing one another is known as self-avoidance. In addition, functionally redundant branches from different neurons would also be expected to avoid each other's territories in order to provide a cleanly derived cell-by-cell representation map in the central nervous system. This behavior is referred to as tiling (Jan and Jan, 2003Jan Y.N. Jan L.Y. Neuron. 2003; 40: 229-242Abstract Full Text Full Text PDF PubMed Scopus (265) Google Scholar; Figure 1). In general, tree-like structures are thought to form by either competition between branches for a common limiting resource, or by an as of yet poorly resolved mechanism that involves direct self-avoidance. The branches of plants are guided by competition for light (and hydrodynamic considerations). How then do the sister branches of neurites know their own? In the relative darkness of the developing Drosophila embryo and larva, it turns out that dendritic branches are guided by touch. With the analysis of the first animal genomes came the startling surprise that there are far fewer genes than expected. How can relatively few genes explain the complexity of neuronal circuitry? Alternative splicing of transcripts from a single gene to produce functionally different protein isoforms increases functional diversity without increasing gene number. For example, the use of alternative exons of the Slo Ca2+ activated K+ channel gene produces functionally different proteins that are important for hair cell function in the cochlea (Ramanathan et al., 1999Ramanathan K. Michael T.H. Jiang G.J. Hiel H. Fuchs P.A. Science. 1999; 283: 215-217Crossref PubMed Scopus (154) Google Scholar). The Down's syndrome Cell Adhesion Molecule (Dscam) gene in insects is an exceptional example of alternative splicing; it can potentially code for up to 38,016 isoforms, with the striking observation that isoforms seem to only adhere to their own isoform. These properties of Dscam raised the exciting possibility that Dscam could allow neurons to distinguish their own sister neurites from those of other neurons (Zipursky et al., 2006Zipursky S.L. Wojtowicz W.M. Hattori D. Trends Biochem. Sci. 2006; 31: 581-588Abstract Full Text Full Text PDF PubMed Scopus (51) Google Scholar). Three groups have now provided strong evidence that this is indeed the case for dendritic self-avoidance, but not for dendritic tiling. The evidence is provided by Hughes et al., 2007Hughes M.E. Bortnick R. Tsubouchi A. Bäumer P. Kondo M. Uemura T. Schmucker D. Neuron. 2007; 54 (this issue): 419-429Abstract Full Text Full Text PDF Scopus (188) Google Scholar and Soba et al., 2007Soba P. Zhu S. Emoto K. Younger S. Yang S.-J. Yu H.-H. Lee T. Jan L.Y. Jan Y.N. Neuron. 2007; 54 (this issue): 405-418Abstract Full Text Full Text PDF Scopus (184) Google Scholar in a pair of studies in this issue of Neuron and in a study by Matthews et al., 2007Matthews B.J. Kim M. Flanagan J.J. Hattori D. Clemens J. Zipursky S.L. Grueber W.B. Cell. 2007; 129 (in press. Published online May 3, 2007)https://doi.org/10.1016/j.cell.2007.04.013Abstract Full Text Full Text PDF PubMed Scopus (242) Google Scholar in the upcoming May 4, 2007 issue of Cell. Previous studies on the guidance of sister axon branches in mushroom bodies had strongly suggested that Dscam would be required for recognition of self and subsequent repulsion (Zipursky et al., 2006Zipursky S.L. Wojtowicz W.M. Hattori D. Trends Biochem. Sci. 2006; 31: 581-588Abstract Full Text Full Text PDF PubMed Scopus (51) Google Scholar). These studies were limited by the resolution available for this class of neurons, which have small 3D target volumes. All three new studies focus on the dendritic arborization (da) neurons of the embryonic and larval peripheral nervous system, which form over a relatively large area. da neurons fall into four classes, I–IV, reflecting increasingly complex dendritic arbors, all of which can be examined with exquisite resolution in part due to their 2D nature. All da neuron classes exhibit self-avoidance, but only classes III and IV exhibit intersegmental tiling. Dscam is expressed in all four da classes, and loss of Dscam activity leads to a loss of self-avoidance in all classes. Single-cell mutant clones revealed that Dscam is required cell autonomously. Previous work had identified similarities between self-avoidance and tiling, so it may have been unexpected when Dscam mutants were found not to affect tiling. So what is the effect of the loss of self-avoidance? Sister neurites, once forbidden to touch, now fasciculate and fail to seek independent territories. The global effect is an inefficient branched structure with regions of poor coverage and regions of too much coverage. The neuron could be described as getting less territory for its material investment. Overexpression of a single Dscam isoform had previously been shown to partially rescue axonal branching phenotypes, suggesting that Dscam has roles both dependent and independent of its diversity. When a single Dscam isoform is overexpressed in da neurons, the self-avoidance defect is rescued in the mutant background. This might suggest that diversity is not required for this function. However, examination of da neurons relative to one another revealed dendrites that had previously coexisted within the same territories now repelling one other. These gain-of-function data beautifully complement the loss-of-function experiments, strongly supporting the idea that the molecular diversity of Dscam mediates recognition of self. Individual photoreceptor cells express 14–50 different Dscam isoforms (Neves et al., 2004Neves G. Zucker J. Daly M. Chess A. Nat. Genet. 2004; 36: 240-246Crossref PubMed Scopus (186) Google Scholar), and it seems probable that WT da neurons will also express a small subset rather than a single isoform. For repulsion of sister dendrites, both the extracellular and cytoplasmic domains of Dscam are required. This implies an active signaling pathway and modulation of the cytoskeleton to mediate retraction. Dscam function in axon guidance requires the adaptor protein Dock and the kinase Pak. The tricornered and hippo kinases and the furry gene are required for dendritic tiling. However, none of these potential downstream effectors displayed genetic interactions with Dscam (Hughes et al., 2007Hughes M.E. Bortnick R. Tsubouchi A. Bäumer P. Kondo M. Uemura T. Schmucker D. Neuron. 2007; 54 (this issue): 419-429Abstract Full Text Full Text PDF Scopus (188) Google Scholar, Matthews et al., 2007Matthews B.J. Kim M. Flanagan J.J. Hattori D. Clemens J. Zipursky S.L. Grueber W.B. Cell. 2007; 129 (in press. Published online May 3, 2007)https://doi.org/10.1016/j.cell.2007.04.013Abstract Full Text Full Text PDF PubMed Scopus (242) Google Scholar, Soba et al., 2007Soba P. Zhu S. Emoto K. Younger S. Yang S.-J. Yu H.-H. Lee T. Jan L.Y. Jan Y.N. Neuron. 2007; 54 (this issue): 405-418Abstract Full Text Full Text PDF Scopus (184) Google Scholar). Like the molecular basis of tiling, the identification of the cytoplasmic effectors will require future work. The active nature of the repulsion process was elegantly documented using time-lapse microscopy. Sister dendrites were seen to contact, fasciculate, and then repel, revealing adhesion between sister dendrites to be an intermediate step. The same process has been equally and beautifully documented in zebrafish sensory axon arbors (see the supplemental data in Sagasti et al., 2005Sagasti A. Guido M.R. Raible D.W. Schier A.F. Curr. Biol. 2005; 15: 804-814Abstract Full Text Full Text PDF PubMed Scopus (123) Google Scholar). This neatly resolves the apparent paradox that repellent molecules are rarely seen to colocalize, yet have to physically contact one another to effect repulsion. The complexity of neuronal branching structures has encouraged many models that invoke external patterning stimuli, such as experience. The new Dscam data points to a very sophisticated potential for endogenous patterning. Interestingly, in the fly visual system, external signals seem to play almost no role in patterning a very complicated retinotopic map, and instead cell-autonomous signals regulate axonal synapse specification (Hiesinger et al., 2006Hiesinger P.R. Zhai R.G. Zhou Y. Koh T.-W. Mehta S.Q. Schulze K.L. Cao Y. Verstreken P. Clandinin T.R. Fischbach K.F. et al.Curr. Biol. 2006; 16: 1835-1843Abstract Full Text Full Text PDF PubMed Scopus (77) Google Scholar). While this seems to conflict with what we know of the role of external stimuli in patterning vertebrate sensory maps, some data indicates a critical role for cell autonomy in these same systems (Crowley and Katz, 2000Crowley J.C. Katz L.C. Science. 2000; 290: 1321-1324Crossref PubMed Scopus (224) Google Scholar). In Dscam mutants, the dendrites still know approximately where to go, so there must be other guidance factors, perhaps coupled to cell-autonomous control of dendrite size. In this context, it is worth noting that axon guidance is also subject to cell-autonomous functions linked to cell size (Canal et al., 1998Canal I. Acebes A. Ferrus A. J. Neurosci. 1998; 18: 999-1008PubMed Google Scholar). In addition, distinct and complex in vivo branch patterns of some fly neurons can be recapitulated in isolated culture (Kraft et al., 2006Kraft R. Escobar M.M. Narro M.L. Kurtis J.L. Efrat A. Barnard K. Restio L.L. J. Neurosci. 2006; 26: 8734-8747Crossref PubMed Scopus (39) Google Scholar). Vertebrate Dscams display minimal alternative splicing, indicating that dendritic self-avoidance must function by a different mechanism. In an era when evolutionary conservation of molecular mechanism is thought to be the rule rather than the exception, the lack of alternative splicing of Dscam outside of insects is highly unusual. Has Dscam function been superceded during evolution? Are arthropods really "hard-wired"? The absence to date of neurotrophins in the fly is provocative, and perhaps indicative of their loss during evolution (Bothwell, 2006Bothwell M. Brain Behav. Evol. 2006; 68: 124-132Crossref PubMed Scopus (35) Google Scholar). However, others have confidently predicted the existence of fly neurotrophins (Hidalgo et al., 2006Hidalgo A. Learte A.R. McQuilton P. Pennack J. Zhu B. Brain Behav. Evol. 2006; 68: 173-180Crossref PubMed Scopus (18) Google Scholar). Equally conspicuous by its absence to date is activity-dependent patterning. It would not be surprising if the molecular effectors for Dscam's role in dendritic branching are conserved in vertebrates, and their identification may help solve the riddle. As our knowledge of the formation of the fly brain rapidly advances, just how much patterning will continue to march to the beat of the genetic drum remains to be seen. What seems to be likely, however, is that the answers, like mutant neurites, will stray from the path our preconceived models dictate they should follow. Drosophila Sensory Neurons Require Dscam for Dendritic Self-Avoidance and Proper Dendritic Field OrganizationSoba et al.NeuronMay 03, 2007In BriefA neuron's dendrites typically do not cross one another. This intrinsic self-avoidance mechanism ensures unambiguous processing of sensory or synaptic inputs. Moreover, some neurons respect the territory of others of the same type, a phenomenon known as tiling. Different types of neurons, however, often have overlapping dendritic fields. We found that Down's syndrome Cell Adhesion Molecule (Dscam) is required for dendritic self-avoidance of all four classes of Drosophila dendritic arborization (da) neurons. Full-Text PDF Open ArchiveHomophilic Dscam Interactions Control Complex Dendrite MorphogenesisHughes et al.NeuronMay 03, 2007In BriefAlternative splicing of the Drosophila gene Dscam results in up to 38,016 different receptor isoforms proposed to interact by isoform-specific homophilic binding. We report that Dscam controls cell-intrinsic aspects of dendrite guidance in all four classes of dendrite arborization (da) neurons. Loss of Dscam in single neurons causes a strong increase in self-crossing. Restriction of dendritic fields of neighboring class III neurons appeared intact in mutant neurons, suggesting that dendritic self-avoidance, but not heteroneuronal tiling, may depend on Dscam. Full-Text PDF Open ArchiveDendrite Self-Avoidance Is Controlled by DscamMatthews et al.CellMay 04, 2007In BriefDendrites distinguish between sister branches and those of other cells. Self-recognition can often lead to repulsion, a process termed "self-avoidance." Here we demonstrate that dendrite self-avoidance in Drosophila da sensory neurons requires cell-recognition molecules encoded by the Dscam locus. By alternative splicing, Dscam encodes a vast number of cell-surface proteins of the immunoglobulin superfamily. We demonstrate that interactions between identical Dscam isoforms on the cell surface underlie self-recognition, while the cytoplasmic tail converts this recognition to dendrite repulsion. Full-Text PDF Open Archive
Referência(s)