|
Gene name - Dscam Synonyms - CG17800 Cytological map position - 43A4--B3 Function - surface receptor Keywords - Bolwig's organ, axon guidance, CNS |
Symbol - Dscam FlyBase ID: FBgn0033159 Genetic map position - Classification - multiple Ig domain protein Cellular location - transmembrane |
Dscam, a Drosophila homolog of human Down syndrome cell adhesion molecule (DSCAM), an immunoglobulin superfamily member, was isolated by its affinity to Dreadlocks (Dock), an SH3/SH2 adaptor protein required for axon guidance. Dscam, Dock and Pak, a serine/threonine kinase, act together to direct pathfinding of Bolwig's nerve, which contains a subclass of sensory axons, to an intermediate target in the embryo. Dscam also is required for the formation of axon pathways in the embryonic central nervous system. cDNA and genomic analyses reveal the existence of multiple forms of Dscam with a conserved architecture containing variable immunoglobulin (Ig) and transmembrane (TM) domains. Alternative splicing can potentially generate more than 38,000 Dscam isoforms. This molecular diversity is likely to contribute to the specificity of neuronal connectivity (Schmucker, 2000).
To gain insight into the mechanisms by which growth cones integrate guidance cues, a combined biochemical and genetic analysis of the Dock signal transduction pathway has been pursued. Dock is an adaptor protein containing 3 SH3 domains and a single SH2 domain, and is closely related to mammalian Nck. dock mutants show defects in axon guidance in the adult fly visual system and in the embryonic nervous system. Based on the role of the adaptor protein Grb-2 in linking receptor tyrosine kinases to Ras, it is proposed that Dock links guidance receptors to downstream regulators of the actin cytoskeleton. Pak, a p21-activated serine/threonine kinase, acts downstream of Dock in adult photoreceptor neurons. Dock binds through its second SH3 domain to Pak and Pak binds directly to Rho family GTPases, evolutionarily conserved regulators of the actin-based cytoskeleton. Genetic studies reveal that both Pak's kinase activity and its interaction with Rho family GTPases are essential for axon guidance (Schmucker, 2000 and references therein).
Dscam binds directly to multiple domains of Dock and is widely expressed on axons in the embryonic nervous system. Dscam is required for recognition of an intermediate targeting determinant for Bolwig's nerve: Dock and Pak are required for this step, and Dscam shows dosage-sensitive interactions with both dock and Pak. Based on these studies, it is proposed that Dscam recognizes a guidance signal(s) and translates it into changes in the actin-based cytoskeleton through Dock and Pak (Schmucker, 2000).
Since Dscam mutants are early larval lethal and Dscam is expressed on many axons in the developing embryonic CNS, assessment was made of whether mutations in Dscam disrupt axon pathways. Axon tracts in stage 16 embryos were visualized with monoclonal antibody BP102 and monoclonal antibody 1D4 an antibody to Fasciclin II. In wild-type embryos, mAbBP102 stains in a ladder-like pattern within the ventral nerve cord, highlighting the connectives and commissures. Dscam mutant embryos exhibit mild disorganization of these tracts. mAb1D4 stains three continuous fascicles on both sides of the midline. In Dscam, breaks in the connectives, predominantly of the outer two fascicles, are observed. Defects exhibit variable expressivity. In addition, axon bundles aberrantly cross the midline. These defects are seen in all embryos of two different heteroallelic combinations (Schmucker, 2000).
To gain insight into Dscam's role in guidance at a defined choice point for which both Dock and Pak also are required, focus was placed on Dscam's role in Bolwig's nerve axon guidance. The projection of Bolwig's nerve elaborates a simple and well-studied pathway. Bolwig's organ, a larval photosensitive structure, contains 12 photoreceptor neurons that extend a single bundle of axons during embryogenesis to their targets in the brain. Bolwig's nerve extends posteriorly along the embryonic optic stalk to the surface of the optic lobe. It then grows across the surface of the optic lobe to a presumptive intermediate target neuron, P2. At P2, the growth cones pause before projecting to their synaptic targets in the brain. Phenotypes can be readily assessed in embryos double-stained for Bolwig's nerve and P2, using mAb22C10 and anti-Bsh (brain-specific homeobox protein), respectively. In stage 16 embryos carrying DscamE1 or DscamP over a deficiency, some 53%-58% of Bolwig's nerve projections were defective. In half of the 'abnormal projections' the entire nerve mistargets, whereas in the remainder only a subset of axons does. Mistargeting axons either project past P2 or stop prematurely. Many of the projections scored as wild type, however, show abnormal expansion of Bolwig's nerve terminus at P2. These guidance defects are unlikely to reflect an indirect consequence of defects in Bolwig's organ differentiation, because organ and cellular morphology in Dscam mutants are indistinguishable from wild type. Bolwig's nerve guidance phenotypes in Dscam mutants are similar to those seen in dock and Pak mutants. The penetrance of dock and Pak are higher than in Dscam. The incomplete penetrance of Dscam may reflect residual function due to maternal contribution. Alternatively, this may reflect redundancy: two additional genes encoding Dscam-related proteins have been identified in the fly genome (Schmucker, 2000).
Dscam RNA is expressed in Bolwig's organ as well as more generally within the CNS and PNS. The protein product is exclusively expressed on axon processes. To assess whether selective expression of Dscam in Bolwig's nerve is sufficient to rescue the mutant, a transgene encoding full-length Dscam driven by the GMR promoter (a strong transcriptional driver providing Bolwig's organ-specific expression) was constructed and it was introduced into the germline by P element DNA transformation. Two independent insertions were characterized. In a wild-type (or a Dscam mutant) background, 100% of the embryos carrying one or two copies of GMR-Dscam exhibit strong axon guidance phenotypes. Individual axons project in abnormal directions over the surface of the optic lobe and rarely contact P2. It is unclear whether this reflects the sensitivity of Bolwig's nerve guidance to increased levels of Dscam or misexpression in Bolwig's organ of an inappropriate isoform, or both. Due to the large size of the Dscam locus (61 kb), whether or not the wild-type gene rescues the mutant phenotype could not be assessed. In any case, the dominant phenotype precludes assessing transgene rescue of the mutant phenotype. In contrast, GMR-dock rescues 85% of dock mutant embryos (Schmucker, 2000).
Whether dock and Pak are functional components of a Dscam guidance pathway was assessed through genetic analysis. Mistargeting defects of Bolwig's nerve were observed in some 44% of the embryos heterozygous for both Dscam and dock. In contrast, only 4%-6% and 10%-13% of embryos heterozygous for either dock or Dscam, respectively, show defects. Similarly, whereas some 38% of embryos heterozygous for both Dscam and Pak have an abnormal Bolwig's nerve, only 5% were defective in embryos heterozygous for Pak. The synergistic interactions between Dscam, dock, and Pak, the similarity of complete loss-of-function phenotypes, and the physical interactions between these proteins are consistent with their acting together to mediate recognition between the Bolwig's nerve growth cones and P2 (Schmucker, 2000).
It is envisioned that during Bolwig's nerve guidance a signal from P2, an intermediate target, attracts Bolwig's nerve growth cones. In dock, Pak, and Dscam mutants, Bolwig's nerve frequently comes in close proximity to P2 and projects past it or deflects away from it. Hence, Dscam is likely to play a prominent role in recognizing a local attractive signal at P2. Since the Dscam phenotype in the CNS is more severe than that of dock, it is likely that Dscam also functions through other downstream components in other neurons (Schmucker, 2000).
How might binding of Dscam on a filopodial process to a ligand at P2 promote attraction? Two alternative models are proposed. (1) Dock and Dscam may form a complex in the absence of ligand. This interaction would be mediated by the SH3-1 and SH3-3 domains in Dock and two closely spaced PXXP sites in Dscam's cytoplasmic tail. Dock's SH3-2 domain may bind to the polyproline stretch in the C-terminal tail; this may prevent recruitment of Pak to the membrane in the absence of ligand. Upon encountering an attractive ligand at P2, however, it is envisioned that Dscam becomes tyrosine phosphorylated. This may promote rearrangement of the complex with Dock binding through its SH2 domain to phosphotyrosine. This conformational change cannot strictly depend upon the SH2 domain, as the SH3-1 and SH3-3 domains can functionally compensate for the SH2 domain. This conformational change would unmask the SH3-2 domain, thus facilitating recruitment of Pak to the complex. Pak, in turn, would promote actin reorganization and subsequent movement of the growth cone to P2. Alternatively, in the absence of ligand the cytoplasmic tail of Dscam may be in a conformation not accessible to Dock binding. In this model, ligand binding would promote both a conformational change in Dscam's cytoplasmic domain and tyrosine phosphorylation. Dock may then associate with Dscam and recruit Pak to the membrane. Candidates for additional components in this signaling pathway have been identified through biochemical and genetic studies and include a nonreceptor tyrosine kinase, an adaptor protein linked to actin, and a nonreceptor tyrosine phosphatase. A precise mechanistic understanding of the relationship between Dscam, Dock, and Pak and these additional components will require identification of the Dscam ligand and detailed biochemical studies (Schmucker, 2000).
Does human DSCAM signal through Nck and Pak? While the extracellular region of human and Drosophila Dscam share identical domain structures and highly related amino acid sequences, the intracellular domains appear unrelated. This may indicate that the human and fly proteins signal downstream via different pathways or that the interactions with Nck are mediated by other proline-rich sequences, phosphotyrosine residues or other associated proteins. Alternatively, there may be multiple mammalian DSCAM genes, one of which shares sequence homology with the cytoplasmic domain of the fly protein. While a partial sequence of a second DSCAM-related gene has been identified on human chromosome 11 (accession no. BAA86446), its cytoplasmic domain is conserved with human DSCAM and is unrelated to Drosophila Dscam (Schmucker, 2000).
Multiple forms of Dscam have been identified. These different forms encode proteins with the same architecture but with sequence variations in three different Ig domains and the transmembrane domain. Genomic analysis reveals that exons encoding these alternative domains are tandemly arranged. There are 12 alternative forms of exon 4 (encoding the N-terminal half of Ig2); 48 alternative forms of exon 6 (encoding the N-terminal half of Ig3); 33 alternative forms of exon 9 (encoding Ig7), and 2 alternative forms of exon 17 (encoding transmembrane domains). DNA sequence analysis supports the view that multiple forms of Dscam are generated by alternative splicing. If all combinations are permissible, 38,016 forms of Dscam may be generated. This level of diversity is comparable to that achieved in individual antibody protein subunits through rearrangement prior to further diversification via TdT-mediated nucleotide insertion and somatic mutation. For instance, it is estimated that, in humans, recombination generates some 9,180 variable heavy chains (51 VH segments, 30 D segments and 6 J segments). An extraordinary diversity of Dscam isoforms is expressed during development. As the mechanisms utilized to generate splicing of one of two alternative exons in many other well-characterized genes are poorly understood, it seems premature to entertain models for the molecular mechanisms regulating the 'either-or' splicing patterns for the 12, 48, and 33 densely packed alternative exons in Dscam. Alternative forms of human DSCAM have not been found in searches of available genomic sequences and ESTs. Hence, either alternatively spliced forms of DSCAM are not generated in mammals or diversity exists among DSCAM-related genes (Schmucker, 2000).
Several other cell surface proteins involved in neuronal connectivity or implicated in the process show considerable molecular diversity. The vertebrate olfactory receptors, encoded by a large gene family containing in excess of 1000 different sequences, play a key role in determining glomerular specificity in the olfactory bulb, although they do not appear to be the sole specificity determinants. Multiple forms of neurexins also have been proposed to play a role in synapse specificity. Alternative splicing of neurexins encoded by three different genes may give rise to more than a 1000 different isoforms differing in size and amino acid sequence. In contrast to neurexins, all Dscam isoforms share a common domain architecture. While loss-of-function studies have not yet critically addressed the role of neurexins in forming neuronal connections, their localization to synapses makes them intriguing candidates for contributing to connection specificity (Schmucker, 2000 and references therein).
Two cadherin cell adhesion molecule subfamilies containing multiple forms also have been implicated in specificity. There are some 20 different classic cadherins expressed in mammals. They are localized to synapses and, in some cases, show restricted patterns of expression correlating with specific functional pathways. Recently, a second family of cadherins, cadherin-related neuronal receptors (CNRs), have been identified in mice through a protein interaction screen and are expressed at synapses (Kohmura, 1998). Fifty-two different forms of CNRs have been identified in three different clusters in the human genome (Wu, 1999). The mRNAs encoding the mouse CNRs are expressed in a diffuse and rather general pattern within the nervous system. In the olfactory bulb, two different CNRs are coexpressed in a single neuron while a smaller fraction of neurons express only one or the other CNR. This raises the intriguing possibility that different combinations of CNRs in different neurons could impart connection specificity. Given the conservation between guidance receptors in flies and mammals, it is surprising that CNRs were not identified in the recently published DNA sequence of the Drosophila genome. While it is still unclear whether DSCAM-related genes in mammals may be alternatively spliced, different organisms may have evolved different strategies to generate diversity among related guidance receptors, as a means of specifying complex patterns of neuronal connectivity (Schmucker, 2000 and references therein).
The massive use of alternative exons in a combinatorial fashion to generate thousands of different protein isoforms all sharing a common architecture with differences in amino acid sequence is unprecedented within the nervous system. The physiological importance of these different isoforms remains entirely unclear. Diverse forms of Dscam may recognize distinct isoforms of specific ligands. That such alterations could lead to changes in ligand–receptor interaction are supported by the findings that alternative forms of Ig repeats in murine FGF receptor 1, generated by alternative splicing, exhibit marked differences in affinity for bFGF. Alternatively, Dscam may form dimers with other forms of Dscam or act as a coreceptor to modulate the activities of other guidance receptors. In this context, it has been proposed that accurate targeting of motoneurons to their specific muscle targets requires input from multiple receptor pathways. Future studies will identify proteins binding to the extracellular domain of Dscam, address whether different forms of Dscam are expressed in specific neurons, and test the functional importance of diversity (Schmucker, 2000).
Multiple forms of Dscam are generated by alternative splicing. Extensive cDNA analyses have revealed alternative amino acid sequences for Ig domains 2, 3, 7, and the transmembrane domain. Alternative forms of each Ig domain have been used to identify additional related sequences within the gene. This led to the identification of a total of 12, 48, and 33 potential alternative sequences for Ig2, Ig3, and Ig7, respectively. Two alternative transmembrane domains also were identified. The N-terminal half of Ig2 is encoded by alternative forms of exon 4, the N-terminal half of Ig3 is encoded by alternative forms of exon 6, and the entire Ig7 domain is encoded by alternative forms of exon 9. Exons are used in a mutually exclusive fashion. That is, each cDNA sequence contains only one each of the variable exons encoding Ig2, Ig3, and Ig7 and contains only one of the two alternative transmembrane domains. The protein sequences of alternative exons form highly related families: exons 4 share between 33% and 81% identity; exons 6 (with the exception of exon 6.11) share between 22% and 87% identity; exons 9 share between 23% and 92% identity, and the two alternative transmembrane segments share 25% identity (Schmucker, 2000).
The Dscam gene extends 61,206 nucleotides from the 5' end of exon 1 to the poly(A) addition site in exon 24. Alternative forms of exon 4 are tandemly arranged within 6.5 kb of DNA between invariant exons 3 and 5; alternative forms of exon 6 within 12.7 kb flanked by invariant exons 5 and 7, and finally, the entire Ig7 domain is encoded by alternative exons in 15.7 kb of DNA bracketed by invariant exons 8 and 10. Alternative exons are flanked by conserved splice donor and acceptor sites. Regulatory sequences required for DNA rearrangement of sequences in the mammalian immune system were not identified. These observations support the notion that multiple forms of Dscam are generated by alternative splicing.
If alternative exons can be spliced independently, then the Dscam locus potentially encodes 38,016 isoforms. To gain some additional clues as to the number of alternative exons used and the variety of different combinations, a panel of 50 cDNAs, synthesized by RT-PCR, was prepared from mRNA isolated from 12-24 hr embryos. Inserts 1.8 kb in length and spanning exons 3 through 10 were sequenced. Of these 50 cDNAs, 49 contained unique combinations of variable exons. Representatives of 11 of 12 exons 4; 30 of 48 exons 6; and 25 of 33 exons 9 were found in these cDNAs. Sequencing of other cDNAs from embryos and pupae have revealed that an additional 1, 6, and 3 alternatives for exons 4, 6, and 9, respectively, were used. These data suggest that all alternative exons are utilized in vivo (Schmucker, 2000).
The Dscam cDNA contains an open reading frame of 6,048 nucleotides, encoding from its N terminus the following: (1) a putative signal peptide; (2) nine tandemly repeated immunoglobulin (Ig) domains; (3) four fibronectin type III (FNIII) domains; (4) a single Ig domain; (5) an additional two FNIII domains; (6) a transmembrane domain; and (7) a novel cytoplasmic domain of 374 amino acids containing multiple potential tyrosine phosphorylation sites and several putative SH3 binding sites. Two 33 amino acid repeats are found in the cytoplasmic domain, each containing a consensus Dock/Nck SH2 binding motif. The extracellular region contains Ig and FNIII domains that are frequently found in axon guidance receptors, including Robo and the netrin-receptor Frazzled. Drosophila Dscam is most highly related to human DSCAM, a gene that has been implicated in the mental retardation associated with Down syndrome (Yamakawa, 1998). The number and organization of the extracellular domains of the fly and human proteins are identical. They share 32% sequence identity and 49% sequence similarity. In contrast, the intracellular domains appear unrelated. Based on the similarities in the extracellular domain, p270 is referred to as Drosophila Dscam (Schmucker, 2000).
Two additional genes encoding Dscam-related proteins, CG12536 and CG8619, have been identified in the fly genome (Schmucker, 2000).
The Drosophila Dscam gene encodes an axon guidance receptor and can generate 38,016 different isoforms via the alternative splicing of 95 variable exons. Dscam contains 10 immunoglobulin (Ig), six Fibronectin type III, a transmembrane (TM), and cytoplasmic domains. The different Dscam isoforms vary in the amino acid sequence of three of the Ig domains and the TM domain. The organization of the Dscam gene from three members of the Drosophila subgenus (D. melanogaster, D. pseudoobscura, and D. virilis), the mosquito Anopheles gambiae, and the honeybee Apis mellifera have been compared. Each of these organisms contains numerous alternative exons and can potentially synthesize tens of thousands of isoforms. Interestingly, most of the alternative exons in one species are more similar to one another than to the corresponding alternative exons in the other species. These observations provide strong evidence that many of the alternative exons have arisen by reiterative exon duplication and deletion events. In addition, these findings suggest that the expression of a large Dscam repertoire is more important for the development and function of the insect nervous system than the actual sequence of each isoform (Graveley, 2004).
The Dscam is a protein overexpressed in the brains of Down syndrome patients and implicated in mental retardation. Dscam is involved in axon guidance and branching in Drosophila, but cellular roles in vertebrates have yet to be elucidated. To understand its role in vertebrate development, the zebrafish homolog of Dscam was cloned and it was shown to share high amino acid identity and structure with the mammalian homologs. Zebrafish dscam is highly expressed in developing neurons, similar to what has been described in Drosophila and mouse. When dscam expression is diminished by morpholino injection, embryos display few neurons and their axons do not enter stereotyped pathways. Zebrafish dscam is also present at early embryonic stages including blastulation and gastrulation. Its loss results in early morphogenetic defects. dscam knockdown results in impaired cell movement during epiboly as well as in subsequent stages. It is proposed that migrating cells utilize dscam to remodel the developing embryo (Yimlamai, 2005).
Down syndrome (DS), a major cause of mental retardation, is characterized by subtle abnormalities of cortical neuroanatomy, neurochemistry and function. Recent work has shown that chromosome band 21q22 is critical for many of the neurological phenotypes of DS. A gene, DSCAM (Down syndrome cell adhesion molecule), has now been isolated from chromosome band 21q22.2-22.3. Homology searches indicate that the putative DSCAM protein is a novel member of the immunoglobulin (Ig) superfamily that represents a new class of neural cell adhesion molecules. The sequence of cDNAs indicates alternative splicing and predicts two protein isoforms, both containing 10 Ig-C2 domains, with nine at the N-terminus and the tenth located between domains 4 and 5 of the following array of six fibronectin III domains, with or without the following transmembrane and intracellular domains. Northern analyses reveals the transcripts of 9.7, 8.5 and 7.6 kb primarily in brain. These transcripts are differentially expressed in substructures of the adult brain. Tissue in situ hybridization analyses of a mouse homolog of the DSCAM gene has revealed broad expression within the nervous system at the time of neuronal differentiation in the neural tube, cortex, hippocampus, medulla, spinal cord and most neural crest-derived tissues. Given its location on chromosome 21, its specific expression in the central nervous system and neural crest, and the homologies to molecules involved in neural migration, differentiation, and synaptic function, it is proposed that DSCAM is involved in neural differentiation and contributes to the central and peripheral nervous system defects in DS (Yamakawa, 1998).
Down Syndrome (DS) caused by trisomy 21 is the most common birth defect associated with mental retardation. To understand the cellular function of DSCAM protein, human DSCAM cDNA was transfected into mouse fibroblast L cells and its expression was analysed. On Western blot analysis, antibodies raised against recombinant DSCAM-Ig3 recognize a 198 kDa protein band in the membrane fraction of DSCAM transfected L cells. Stable transformants expressing DSCAM show uniform surface expression. DSCAM-expressing transfectants exhibit enhanced adhesive properties, aggregating with faster kinetics and forming aggregates in a homophilic manner. Divalent cations are not required for this cell aggregation. These results demonstrate that DSCAM is a cell adhesion molecule that can mediate cation-independent homophilic binding activity between DSCAM expressing cells (Agarwala, 2000).
Down Syndrome Cell Adhesion molecule (DSCAM) is a member of the immunoglobulin superfamily, and represents a novel class of neuronal cell adhesion molecules. In order to understand the cellular functions of DSCAM, full-length mouse and human cDNA clones have been isolated, and its expression analysed during mouse development and differentiation. Sequence analysis of the human DSCAM cDNA predicts at least 33 exons that are distributed over 840 kb. When compared to human DSCAM, the mouse homolog shows 90% and 98% identity at the nucleotide and amino acid levels, respectively. In mouse, DSCAM is located on 16C, the syntenic region for human chromosome band 21q22 and also the region duplicated in mouse DS models. DSCAM gene is predicted to encode an approximately 220-kDa protein, and its expression shows dynamic changes that correlate with neuronal differentiation during mouse development. These results suggest that DSCAM may play critical roles in the formation and maintenance of specific neuronal networks in brain (Agarwala, 2001a).
DSCAM, a conserved gene involved in neuronal differentiation, is a member of the Ig superfamily of cell adhesion molecules. A human DSCAM (Down syndrome cell adhesion molecule) paralog, DSCAML1, is located on chromosome 11q23. The deduced DSCAML1 protein contains 10 Ig domains, six fibronectin-III domains, and an intracellular domain, all of which are structurally identical to DSCAM. When compared to DSCAM, DSCAML1 protein shows 64% identity to the extracellular domain and 45% identity to the cytoplasmic domain. In the mouse brain, DSCAML1 is predominantly expressed in Purkinje cells of the cerebellum, granule cells of the dentate gyrus, and in neurons of the cerebral cortex and olfactory bulb. Biochemical and immunofluorescence analyses indicate that DSCAML1 is a cell surface molecule that targets axonal features in differentiated PC12 cells. DSCAML1 exhibits homophilic binding activity that does not require divalent cations. Based on its structural and functional properties and similarities to DSCAM, it is suggested that DSCAML1 may be involved in formation and maintenance of neural networks. The chromosomal locus for DSCAML1 makes it an ideal candidate for neuronal disorders (such as Gilles de la Tourette and Jacobsen syndromes) that have been mapped on 11q23 (Agarwala, 2001b).
Dscam, a novel cell-adhesion molecule belonging to the Ig-superfamily mediates homophilic intercellular adhesion and is expressed abundantly in the nervous system during development. To gain better understanding of the role of Dscam in neuronal differentiation, an antibody was raised and its protein product was characterized. Anti-Dscam antibody detects an approximately 200-kDa protein band in human and mouse brain lysates. Immunohistochemical studies show that during embryonic development of mice, mouse Dscam is expressed throughout the neuronal tissues and also in nonneuronal tissues such as lung, liver, and limb buds. In adult brain Dscam expression is predominant in the cerebellum, hippocampus, and olfactory bulb. Immunofluorescence double labeling of hippocampal and cerebellar primary cultures reveals that Dscam is associated with axonal and dendritic processes. In view of its cellular localization and spatiotemporal expression pattern, it is suggested that Dscam is involved in cell-cell interactions during axonal-dendritic development, and maintenance of functional neuronal networks (Agarwala, 2001c).
CNS development involves neural patterning, neuronal and axonal migrations, and synapse formation. DSCAM, a chromosome 21 axon guidance molecule, is expressed by CNS neurons during development and throughout adult life. DSCAM and its chromosome 11 paralog DSCAML1 exhibit inverse ventral-dorsal expression patterns in the developing spinal cord and distinct, partly inverse, expression patterns in the developing cortex, beginning in the Cajal-Retzius cells. In the adult cortex, DSCAM predominates in layer 3/5 pyramidal cells and DSCAML1 predominates in layer 2 granule cells. In the cerebellum, DSCAM is stronger in the Purkinje cells and DSCAML1 in the granule cells. The predicted DSCAML1 protein contains 60 additional N-terminal amino acids which may contribute to its distinct expression pattern and putative function. It is proposed that the DSCAMs comprise novel elements of the pathways mediating dorsal-ventral patterning and cell-fate specification in the developing CNS (Barlow, 2002).
DSCAM is a member of the immunoglobulin superfamily that maps to a Down syndrome region of chromosome 21q22.2-22.3. Genetic and biochemical studies have shown that in Drosophila, Dscam activates Pak1 via the Dock adaptor molecule. The extracellular domain of human DSCAM is highly homologous to the Drosophila protein; however, the intracellular domains of both human and Drosophila DSCAM share no obvious sequence identity. To study the signaling mechanisms of human DSCAM, the interaction between DSCAM and potential downstream molecules was investigated. DSCAM was shown to directly bind to Pak1 and stimulates Pak1 phosphorylation and activity, unlike Drosophila, where an adaptor protein Dock mediates the interaction between Dscam and Pak1. DSCAM activates both JNK and p38 MAP kinases. Furthermore, expression of the cytoplasmic domain of DSCAM induces a morphological change in cultured cells that is JNK-dependent. These observations suggest that human DSCAM also signals through Pak1 and may function in axon guidance similar to the Drosophila Dscam (Li, 2004).
The development of CNS neuronal networks involves processes including neuroblast migration, axonal pathfinding, and synaptogenesis. To evaluate the role of the axonal guidance molecule DSCAM in CNS connectivity, a lacZ reporter construct, Pr1.8-betagal, was generated containing a 1.8kb fragment of the human DSCAM promoter region, and its expression in four E12.5 transgenic mouse embryos was analyzed. Pr1.8-betagal drives lacZ expression in the choroid plexus and roof of the fourth ventricle, the floor plate of the fourth ventricle, pons and medulla oblongata, and the eye, limb buds, and dorsal root ganglion. This recapitulates a subset of DSCAM expression as demonstrated by in situ hybridization, supporting this 1.8kb fragment as a component of the endogenous DSCAM promoter. The Pr1.8-betagal expression pattern supports a role for DSCAM in CNS development, providing an endogenous promoter to investigate the contribution of DSCAM to Down syndrome neural defects (Barlow, 2002).
To establish functional circuitry, retinal neurons occupy spatial domains by arborizing their processes, which requires the self-avoidance of neurites from an individual cell, and by spacing their cell bodies, which requires positioning the soma and establishing a zone within which other cells of the same type are excluded. The mosaic patterns of distinct cell types form independently and overlap. The cues that direct these processes in the vertebrate retina are not known. This study shows that some types of retinal amacrine cells from mice with a spontaneous mutation in Down syndrome cell adhesion molecule (Dscam), a gene encoding an immunoglobulin-superfamily member adhesion molecule, have defects in the arborization of processes and in the spacing of cell bodies. In the mutant retina, cells that would normally express Dscam have hyperfasciculated processes, preventing them from creating an orderly arbor. Also, their cell bodies are randomly distributed or pulled into clumps rather than being regularly spaced mosaics. These results indicate that mouse DSCAM mediates isoneuronal self-avoidance for arborization and heteroneuronal self-avoidance within specific cell types to prevent fasciculation and to preserve mosaic spacing. These functions are analogous to those of Drosophila DSCAM and DSCAM2. DSCAM may function similarly in other regions of the mammalian nervous system, and this role may extend to other members of the mammalian Dscam gene family (Fuerst, 2008).
date revised: 22 March 2005
Home page: The Interactive Fly © 1995, 1996 Thomas B. Brody, Ph.D.
The Interactive Fly resides on the
Society for Developmental Biology's Web server.