InteractiveFly: GeneBrief

wrapper: Biological Overview | References

BIOLOGICAL OVERVIEW

Glia play crucial roles in ensheathing axons, a process that requires an intricate series of glia-neuron interactions. The membrane-anchored protein Wrapper is present in Drosophila midline glia and is required for ensheathment of commissural axons. By contrast, Neurexin IV is present on the membranes of neurons and commissural axons, and is highly concentrated at their interfaces with midline glia. Analysis of Neurexin IV and wrapper mutant embryos revealed identical defects in glial migration, ensheathment and glial subdivision of the commissures. Mutant and misexpression experiments indicated that Neurexin IV membrane localization is dependent on interactions with Wrapper. Cell culture aggregation assays and biochemical experiments demonstrated the ability of Neurexin IV to promote cell adhesion by binding to Wrapper. These results show that neuronal-expressed Neurexin IV and midline glial-expressed Wrapper act as heterophilic adhesion molecules that mediate multiple cellular events involved in glia-neuron interactions (Wheeler, 2009).

A sim-Gal4 UAS-tau-GFP transgenic strain and confocal microscopy have been used to study the development of Drosophila CNS midline cells (Wheeler, 2006; Wheeler, 2008). In sim-Gal4 UAS-tau-GFP embryos, GFP is localized to the cytoplasm of all midline cells - both neurons and glia (Wheeler, 2006). When examined in sagittal views, this allows visualization of the morphology of each midline cell type during development. Identification of specific midline cell types employed immunostaining or in situ hybridization with more than 90 validated cell type-specific reagents. The current study used this system to investigate the dynamics of MG development during embryonic stages 12-17 in both wild-type and mutant embryos. MG were identified by their distinct shape, relatively dorsal position within the CNS and expression of wrapper, which is high in AMG, low in PMG, and absent from neurons (Noordermeer, 1998; Wheeler, 2006). PMG were additionally identified by expression of engrailed (en). Antibodies recognizing all neurons (anti-Elav), their axons (MAb BP102), and the midline precursor 1 (MP1) neurons (anti-Lim3) provided a comprehensive view of MG interactions with nearby neurons and axons. At the beginning of CNS axonogenesis {stages 12/3 to 12/0), commissural axons initially converge into a single axon bundle. At stage 12/3, the AMG reside in the anterior of the segment and have an elongated morphology as they migrated toward the axon commissure. Approximately three AMG make contact with the anterior surface of the commissure while the remaining AMG undergo apoptosis. At stage 12/0, the AMG send processes across both the dorsal and ventral surfaces of the commissure. As the commissure separates into AC and PC, the AMG membranes completely ensheath the AC. AC ensheathment concludes with the movement of an AMG cell body between the commissures. After the AC is completely ensheathed, a single, dorsally located AMG migrates across the dorsal surface of the PC during stages 15-16. This AMG extends processes posteriorly to ensheath the PC. During stages 15-17, the AMG also sends cytoplasmic projections into both the AC and PC that become more elaborate as development proceeds. Using electron microscopy, it has been shown that these cytoplasmic projections divide each commissure into distinct subdomains (Wheeler, 2009).

Both the MP1 neurons and PMG are also in proximity to AMG and the commissures. Their positions are constant relative to the migrating AMG. The MP1 neurons remain in close contact with the ventral-most AMG from stages 11-17, and prior to commissure separation the MP1 neurons are closely associated with the commissure along its ventral side. The PMG migrated dorsally and in an anterior direction during stage 12, and at least one PMG abuts the posterior side of the PC from stages 12-16 before undergoing apoptosis. The cell bodies of non-midline-derived neurons also make extensive contacts with the AMG and, together with the PMG and MP1 neurons, they might play important roles in AMG development (Wheeler, 2009).

This paper provides genetic, cellular and biochemical evidence that Nrx-IV and Wrapper physically interact to mediate cell adhesion between MG and neurons. The data presented here on Nrx-IV and wrapper mutant phenotypes are in close agreement with results previously published for wrapper (Noordermeer, 1998), and show that both genes have identical MG phenotypes, a likely outcome for two genes involved in heterophilic cell adhesion. The Nrx-IV-Wrapper interactions occur between MG and three neuronal cell types or structures: (1) MP1s, (2) cell bodies of lateral CNS neurons, and (3) commissural axons. Together, these interactions control the position of MG and the ensheathment and subdivision of commissures (Wheeler, 2009).

This paper provides direct evidence that MP1 neurons closely interact with AMG, suggesting an important role in their development. Beginning at late stage 12, there is a strong accumulation of Nrx-IV at the interface between the MP1s and a subset of AMG, and the Nrx-IV concentration is maintained as the AMG migrate and ensheath the commissures. Nrx-IV accumulation at the MG-MP1 boundaries is abolished in wrapper mutants, and both Nrx-IV and wrapper mutants have gaps between the AMG and MP1s. These results indicate that the MP1s physically adhere to the AMG, and this adhesive interaction is required for proper positioning of MG and ensheathment of the commissures (Wheeler, 2009).

Nrx-IV protein is present in most, if not all, CNS neurons. However, protein levels are generally low. The exceptions are the neurons that flank the MG. These cells show a strong accumulation of Nrx-IV at the interfaces with MG. This indicates an aspect of midline cell biology not commonly considered - that MG interact closely with adjacent lateral CNS neurons. This might act to physically constrain migrating MG at the midline and restrict their lateral movement. In this sense, the lateral CNS neurons, the MP1 neurons and axon commissures work together to construct the MG cytoarchitectural scaffold. Alternatively, the adhesion between lateral CNS neurons and MG might allow developmental signals to pass between these cell types (Wheeler, 2009).

A key functional role of MG is their interaction with commissural axons, and these interactions require complex MG movements and morphological changes. The AMG extend cytoplasmic processes between the commissures, followed by an AMG cell body. These MG structures effectively partition the AC from the PC. Previously, it was proposed that commissure separation is caused by the interposition of MG into the unseparated commissure. However, it is noted that in wrapper mutants, the AC and PC were well separated, even though MG processes were commonly absent between the commissures. It is possible that in wrapper mutants, MG initially cause commissure separation and then quickly retract or undergo apoptosis, indicating that MG function is required only transiently. Alternatively, commissure separation could be independent of MG interposition and the MG partition already-separated commissures. In contrast to wrapper mutants, Nrx-IV mutants have poorly separated commissures. This difference is most likely to reflect an additional function of Nrx-IV because: (1) the MG phenotypes are similar between Nrx-IV and wrapper mutants, (2) the mutants of each gene are null, (3) neither has a recognizable maternal effect, and (4) Nrx-IV is more widely expressed (Wheeler, 2009).

Throughout commissure ensheathment, axons have strong accumulations of Nrx-IV along their interface with the AMG. This suggests a continual requirement of Nrx-IV and Wrapper to mediate MG-axonal adhesion and is consistent with the wide variety of MG-axon adhesion defects observed in both Nrx-IV and wrapper mutants and the inability of elav-Gal4 UAS-Nrx-IV to rescue late Nrx-IV mutant phenotypes. By contrast, MG remained relatively well associated with each other, suggesting that neither wrapper nor Nrx-IV plays an important role in MG-MG adhesion (Wheeler, 2009).

MG projections also subdivide each commissure into discrete compartments. Previous work employing electron microscopy proposed that the MG subdivided each commissure into three dorsoventral regions. This subdivision also requires Nrx-IV and wrapper function because Nrx-IV and Wrapper accumulated in the AMG commissural projections, and the projections are absent in both Nrx-IV and wrapper mutants (Noordermeer, 1998). Both the organizing principles and the significance of these commissural subdomains are unknown, and it remains to be determined whether the MG are a cause of the subdivision or are filling in axonal regions that are already subdivided (Wheeler, 2009).

The view of MG migration presented in this study builds on previous work, but also differs in several aspects. These include nomenclature, MG-neuron interactions and PMG migration. Klambt (1991) proposed a model in which three pairs of MG (MGA, MGM and MGP) arise in the anterior of the segment and, during migration, separate and ensheath the AC and PC. The MGA and MGM migrate posteriorly and ensheath the AC; the MGA ultimately resides anterior to the AC and the MGM between the AC and PC. By contrast, the MGP migrate anteriorly from the adjacent posterior segment and partially ensheath the PC. More recent observations, including some from this paper, point toward a different view. Analysis of 52 genes expressed in MG (Kearney, 2004) indicates that only two distinct MG cell types can be identified, termed AMG and PMG. There are six AMG in the anterior of the segment (this class includes MGA and MGM, which cannot be distinguished molecularly) and four PMG that reside in the posterior of the segment and are identical to MGP in terms of gene expression. Of the six initial AMG, only three survive. These cells migrate posteriorly, ensheath both the AC and PC, and elaborate projections into the commissures. By contrast, all PMG die by stage 17, and therefore do not ensheath the PC. Initially, it was proposed that PMG/MGP migrate from the adjacent posterior segment. In the current experiments, no evidence for this for this. Instead, PMG arise in the En⁺ posterior of the segment and migrate anterodorsally toward the commissure. Before undergoing apoptosis, a single PMG abuts the PC from the posterior side. Thus, the PMG are positioned to influence commissure development (Wheeler, 2009).

The experiments described in this paper strongly support the view that Nrx-IV and Wrapper directly bind and mediate cell adhesion. By contrast, neither protein mediates homophilic cell adhesion. Wrapper is an Ig superfamily protein, and experiments in both flies and vertebrates indicate that Nrx-IV can bind to additional Ig superfamily proteins. In Drosophila septate junctions, Nrx-IV forms a complex with Contactin and Neuroglian (Faivre-Sarrailh, 2004), which are two Ig superfamily proteins. It was proposed that Nrx-IV binds to Contactin at the membrane in a cis configuration, and that Contactin is required for proper Nrx-IV membrane localization (Laval, 2008). Contactin is present in the CNS, and might play a similar role in neurons. It is unknown whether Nrx-IV binds Neuroglian or Contactin in trans, similar to the mechanism proposed here for Nrx-IV-Wrapper binding. However, at paranodal axo-glial junctions in mice, the Nrx-IV homolog Caspr binds in cis to contactin, and in trans to the Neuroglian homolog neurofascin. In summary, Nrx-IV binding to Wrapper indicates a general feature of Nrx-IV, which is its ability to bind diverse Ig superfamily proteins (Wheeler, 2009).

One of the remarkable aspects of Nrx-IV is its strong membrane accumulation at sites where neurons are apposed to MG. In one sense, this resembles the accumulation of Nrx-IV in septate junctions. However, from a mechanistic perspective the situation appears different. In septate junctions, Nrx-IV membrane localization is constrained by interactions with Contactin and Neuroglian (Faivre-Sarrailh, 2004; Laval, 2008), as well as with cytoskeleton-associated proteins important for membrane localization. By contrast, the localization of Nrx-IV in neurons appears relatively fluid and dispersed, only accumulating at high levels when in contact with a Wrapper⁺ membrane. It remains possible that once Wrapper and Nrx-IV bind, additional proteins might bind to Nrx-IV to stabilize its membrane localization. These interactions could further regulate the dynamics of MG-neuron interactions (Wheeler, 2009).

Drosophila Neurexin IV stabilizes neuron-glia interactions at the CNS midline by binding to Wrapper

Ensheathment of axons by glial membranes is a key feature of complex nervous systems ensuring the separation of single axons or axonal fascicles. Nevertheless, the molecules that mediate the recognition and specific adhesion of glial and axonal membranes are largely unknown. This study used the Drosophila midline of the embryonic central nervous system as a model to investigate these neuron glia interactions. During development, the midline glial cells acquire close contact to commissural axons and eventually extend processes into the commissures to wrap individual axon fascicles. This wrapping of axons depends on the interaction of the neuronal transmembrane protein Neurexin IV with the glial Ig-domain protein Wrapper. Although Neurexin IV has been described to be an essential component of epithelial septate junctions (SJ), its function in mediating glial wrapping at the CNS midline is independent of SJ formation. Moreover, differential splicing generates two different Neurexin IV isoforms. One mRNA is enriched in septate junction-forming tissues, whereas the other mRNA is expressed by neurons and recruited to the midline by Wrapper. Although both Neurexin IV isoforms are able to bind Wrapper, the neuronal isoform has a higher affinity for Wrapper. It is concluded that Neurexin IV can mediate different adhesive cell-cell contacts depending on the isoforms expressed and the context of its interaction partners (Stork, 2009).

This study shows that wrapping of commissural axons by midline glia is dependent on the interaction between the GPI-linked protein Wrapper expressed on glial cells and the neuronally expressed Neurexin IV protein. Mutants for Neurexin IV and wrapper show similar wrapping defects at the midline. In both mutants, the midline glia is unable to infiltrate the axonal neuropil of the commissures to ensure ensheathment of individual fascicles. Tissue-specific rescues of Neurexin IV mutants showed that Neurexin IV acts in neurons and functions as an axon-autonomous-specific recognition signal for midline glial processes, as only Neurexin IV-expressing fascicles show restoration of ensheathment. Ectopic expression of Neurexin IV was even able to recruit pronounced midline glial processes to ectopic places far away from their normal localization in wild-type embryos. This suggests that Neurexin IV is a key factor in the axon-glia recognition at the midline, and a model is proposed in which Neurexin IV attracts and stabilizes midline glial processes in a contact-dependent manner (Stork, 2009).

It has been shown that midline glia exhibit thin, highly dynamic cell processes that explore neighboring neuronal substrates. Upon binding to a Neurexin IV-expressing axon fascicle, these initially transient midline glial processes might then be stabilized. This stabilization itself may be required to establish a tight glial wrap or to promote the assembly of further signaling complexes that are required for glial cell development. One possible candidate for neuron-glia communication could be the EGF-receptor ligand Spitz, which is provided by the commissural neurons and needs to be transferred to the midline glia in order to promote their survival. A tight adhesion of the glial processes to the neuronal membranes might facilitate this transfer (Stork, 2009).

Genetic analysis in vivo strongly suggested that Wrapper might act as the glial binding partner for neuronal Neurexin IV in this recognition process. Indeed aggregation assays in S2 cells revealed specific heterophilic binding of Wrapper and Neurexin IV. The Neurexin IV gene generates two distinct isoforms through alternative splicing. This study has shown that the two different isoforms are differentially expressed in the nervous system. Whereas the Nrx-IV^exon3 isoform is predominantly expressed in glial cells that can form septate junctions, the Nrx-IV^exon4-specific isoform is enriched in neurons. Both proteins differ only in the sequence of their N-terminal Discoidin-like domain that mediates interaction with carbohydrates present on many adhesion proteins (Kiedzierska, 2007). Although each isoform alone is able to interact with Wrapper in S2 cell aggregation experiments, a much higher affinity of the neuronally enriched Nrx-IV^exon4 isoform to Wrapper is observed in a competitive aggregation assay. The in vivo rescue experiments corroborate the results obtained by the tissue-specific mRNA isolation and the cell culture experiments. Although both isoforms are able to at least partially rescue the Neurexin IV mutant midline glial wrapping phenotype in the embryo, the rescuing abilities of Neurexin IV^exon3 is less pronounced compared with Neurexin IV^exon4. The alternatively spliced exons 3 and 4 are conserved in all Drosophilidae and Anopheles, and, thus, probably have important functional purposes (Stork, 2009).

Neurexin IV and Wrapper interaction possibly has not only an impact on the midline glial cell but also on the commissural axon. Neurexin IV accumulates in commissural axons, and by recruiting additional adaptors through its cytoplasmic domain it could reorganize the cytoskeleton. In epithelia, Neurexin IV recruits Coracle (Cor), a member of the band 4.1 superfamily, to septate junctions. No expression of Coracle is found at the midline and no mutant midline phenotype is detected in coracle mutant embryos. However, enhanced levels of β-Spectrin and Discs large protein are observed at the midline, which might hint towards a specific cytoskeletal connection established in commissural axons at the CNS midline (Stork, 2009).

In addition to a pure adhesive function of the Neurexin IV-Wrapper complex, Wrapper may also exert signaling properties in the glia cell. However, as Wrapper is a GPI-linked protein it would require a still unknown co-receptor for this function. In this respect, it is also interesting to note that Wrapper is more generally expressed in cortex glia and its binding to Neurexin IV may be a more general property of neuron-glia interaction (Stork, 2009).

Obviously, neuron-glia interaction is not confined to the Drosophila CNS but is also of eminent importance during the insulation of all axonal trajectories in both invertebrates and vertebrates. In vertebrates, Schwann cells wrap axons by either forming a myelin sheath or Remak fibers. Similarly, oligodendrocytes form myelin in the CNS. During myelination, the glial cell membranes form special contact zones with the axon, the paranodes, abutting the nodes of Ranvier. These are characterized by septate-like junctions that prevent current leakage. The ultrastructural architecture of these cell-cell junctions and also the molecules establishing these junctions have been conserved between flies and mammals, suggesting an ancient evolutionary origin of this axonal insulation. As core components of septate or septate-like junctions, the Caspr/Paranodin, Contactin and Neurofascin/155 and their Drosophila counterparts Neurexin IV, Contactin and Neuroglian have been identified. Interestingly, Wrapper appears to be less conserved. Although it is present in all Drosophilidae and the Drosophila genome harbors a Wrapper-related protein, Klingon, no clear Wrapper orthologs can be identified in mammals. However, there are several GPI-linked Ig-superfamily proteins in the mouse genome whose expression profiles need to be determined (Stork, 2009).

Besides prior identification of direct cis-binding partners of Neurexin IV/Caspr, which act in the same cell, this study has identified Wrapper as the first factor that interacts with Neurexin IV in a trans fashion. Based on the tissue culture data, a direct interaction is anticipated but at present the involvement of additional complex partners cannot be excluded. In previous studies it has been shown that Neurexin IV is required to facilitate the secretion of Contactin to the membrane, thereby allowing the generation of adhesive septate junctions. This study shows that Neurexin IV can directly perform adhesive functions by binding to the Wrapper protein decorating opposing cell membranes. Interestingly, Contactin and Wrapper are both similar Ig-domain proteins linked via GPI anchors to the plasma membrane (Stork, 2009).

Within the nervous system, Neurexin IV has been extensively studied for its role in organizing the formation of septate junctions between glial cells, which constitute the major structural component of the Drosophila blood brain barrier. Unlike in the vertebrate paranodes, septate junctions are found extensively at glial-glial cell contacts in the Drosophila nervous system and are only rarely detected between glial cells and axons. The midline glia is not part of this subperineurial glial sheath but rather belongs to the class of wrapping glia that ensures normal insulation of axon fascicles at the midline. In line with this notion, midline glial cells do not form septate junctions visible at the electron-microscopic level. Additionally, major septate junction components such as Coracle, Neuroglian and Lachesin are not enriched at the midline glia, and the corresponding mutants show normal midline glial wrapping behavior. For some septate junction components, midline expression has been reported. This study found that, in these cases, expression is restricted to channel glia, which is part of the subperineurial sheath known to form epithelial-like pleated septate junctions and is not related to the midline glia (Stork, 2009).

The results show that at the Drosophila midline Neurexin IV acts in a novel, septate junction-independent way to ensure neuron-glia adhesion; it will be interesting to determine whether similar adhesive interactions can be attributed to the mammalian homolog Caspr or to other members of the Caspr protein family. Interestingly, it has been recently reported, that Neurexin IV and other canonical septate junction-associated proteins control the adhesive properties of cardial and pericardial cells in the embryonic heart of Drosophila without forming septate junctions. Additionally, these noncanonical adhesive properties of septate junction proteins in the heart, and also the assembly of canonical septate junctions in the Drosophila blood brain barrier, are controlled by different heterotrimeric G protein signaling pathways and possibly Wrapper-Neurexin-IV-mediated adhesion at the CNS midline is also influenced by G protein signaling pathways. In the future, it will be interesting to determine the different roles of the Neurexin IV-Wrapper complex and to dissect the cellular responses triggered by this neuron-glia interaction (Stork, 2009).

Canoe functions at the CNS midline glia in a complex with Shotgun and Wrapper-Nrx-IV during neuron-glia interactions

Vertebrates and insects alike use glial cells as intermediate targets to guide growing axons. Similar to vertebrate oligodendrocytes, Drosophila midline glia (MG) ensheath and separate axonal commissures. Neuron-glia interactions are crucial during these events, although the proteins involved remain largely unknown. This study shows that Canoe (Cno), the Drosophila ortholog of AF-6, and the DE-cadherin Shotgun (Shg) are highly restricted to the interface between midline glia and commissural axons. cno mutant analysis, genetic interactions and co-immunoprecipitation assays unveil Cno function as a novel regulator of neuron-glia interactions, forming a complex with Shg, Wrapper and Neurexin IV, the homolog of vertebrate Caspr/paranodin. These results also support additional functions of Cno, independent of adherens junctions, as a regulator of adhesion and signaling events in non-epithelial tissues (Slováková, 2011).

The midline constitutes a key boundary of bilateral organisms. In vertebrates, it is the floorplate and the functionally equivalent structure in Drosophila is the mesectoderm, which gives rise to all midline cells, neurons and glia, in the most ventral part of the embryo. MG are of great relevance at the midline as an intermediate target during axonal pathfinding, providing both attractive and repulsive guidance cues. These signals allow contralateral axons to cross the midline but never to recross, and they also keep ipsilateral axons away from the midline. In addition to this early function in guiding commissural axons towards the midline, MG are also fundamental later on to separate the commissures by enwrapping and subdividing them. This study shows that the PDZ protein Cno and the DE-cadherin Shg participate in, and contribute to, the regulation of these later stage neural differentiation events, in which neuron-glia interactions play a central role (Slováková, 2011).

In Drosophila, Wrapper and Nrx-IV physically interact to promote glia-neuron intercellular adhesion at the MG. This study proposes that Cno and Shg are important components of this adhesion complex and key to its function. Both Cno and Shg are present at the MG, being highly restricted to the interface between MG and commissural axons. Cno and Shg were detected in a complex in vivo with Wrapper at the CNS MG. Nrx-IV, which is located on the surface of commissural axons, was also consistently found in a complex with Cno, although the amount of Cno protein that was co-immunoprecipitated was much lower than that present in Cno-Wrapper complexes. One plausible explanation is that whereas Cno and Wrapper are present in the same cell (MG), Cno and Nrx-IV are in different cell types (MG and neurons, respectively) and, in addition, Cno is a cytoplasmic protein that is indirectly linked to Nrx-IV through other proteins in the same complex (i.e., Shg and Wrapper). Intriguingly, stronger genetic interactions were found between Cno and Nrx-IV than between Cno and Wrapper (double heterozygote analysis). A possible explanation for this is that Nrx-IV is not only acting through Wrapper-Shg-Cno in the MG but also through other partners. In this way, when the dose of Cno and Wrapper was halved, Nrx-IV could still function fully through these other, putative partners. However, halving the dose of Cno and Nrx-IV would impair not only the Nrx-IV-Wrapper-Cno signal but also the other potential pathways. In vertebrates, the ortholog of Nrx-IV, termed contactin-associated protein (Caspr or Cntnap) or paranodin, is located at the septate-like junctions of the axonal paranodes, where it interacts in cis with contactin (at neurons) and in trans with neurofascin (at the glia). The Drosophila homologs of these Ig superfamily proteins, Contactin and Neuroglian, interact in the same way with Nrx-IV at the septate junctions. However, there are no septate junctions at the neuron-MG interface. Hence, other, as yet unknown partners of Nrx-IV might exist at this location (Slováková, 2011).

Cno and its vertebrate orthologs afadin/AF-6/Mllt4 have been shown to localize at epithelial adherens junctions (AJs), where they regulate the linkage of AJs to the actin cytoskeleton by binding both actin and Nectin family proteins. However, Cno is not exclusively present at the AJs of epithelial tissues. Indeed, it was found that Cno is also expressed in mesenchymal tissues, where it dynamically regulates three different signaling pathways required for muscle/heart progenitor specification. The asymmetric division of these muscle/heart progenitors and of CNS progenitors also requires an AJ-independent function of Cno to asymmetrically locate cell fate determinants and properly orientate the mitotic spindle. Therefore, Cno seems to act through different mechanisms depending on the cell type. This study describes a novel function of Cno during neural differentiation. In the MG, Cno, through Shg, contributes to the tight adhesion between the MG and the commissural axons and perhaps even to the regulation of some intracellular signaling within the MG. Indeed, Cno has been shown to regulate different signaling cascades during development. No AJs or septate junctions (SJs) have been described at the MG-commissural axon interface. This suggests that the function of Cno in the midline is independent of AJs. In fact, the partner of Cno at this location, the Drosophila Nectin ortholog Echinoid, is not detected at the midline. In this context, it is worth pointing out that Shg is an epithelial cadherin key at AJs. This study has shown that Shg can also be found in non-epithelial tissues with an important function independent of AJs. A similar situation occurs with Nrx-IV. Despite Nrx-IV being a very well established component of SJs, no SJs are formed in the midline and no other known components of SJs are expressed there. Thus, different modes of Cno action, either as an AJ protein or as a signaling pathway regulator, are possible and they are not mutually exclusive: it all depends on the cell type and context (Slováková, 2011).

Identification of motifs that are conserved in 12 Drosophila species and regulate midline glia vs. neuron expression

Functional complexity of the central nervous system (CNS) is reflected by the large number and diversity of genes expressed in its many different cell types. Understanding the control of gene expression within cells of the CNS will help reveal how various neurons and glia develop and function. Midline cells of Drosophila differentiate into glial cells and several types of neurons and also serve as a signaling center for surrounding tissues. This study examined regulation of the midline gene, wrapper, required for both neuron-glia interactions and viability of midline glia. A region upstream of wrapper required for midline expression was identified that is highly conserved (87%) between 12 Drosophila species. Site-directed mutagenesis identifies four motifs necessary for midline glial expression: (1) a Single-minded/Tango binding site, (2) a motif resembling a Pointed binding site, (3) a motif resembling a Sox binding site, and (4) a novel motif. An additional highly conserved 27 bp are required to restrict expression to midline glia and exclude it from midline neurons. These results suggest short, highly conserved genomic sequences flanking Drosophila midline genes are indicative of functional regulatory regions and that small changes within these sequences can alter the expression pattern of a gene (Estes, 2008).

To facilitate the identification of sequences responsible for wrapper expression in the midline glia of Drosophila, the genomic region flanking the wrapper transcription unit was examined to determine the degree of conservation between the 12 available Drosophila species. The regions most likely to contain regulatory control elements (motifs) of wrapper are tractable; the genomic regions flanking the transcription unit and the first intron are relatively small. The results of this analysis highlighted a region between -492 and -326 upstream of the transcription start site of wrapper that is highly conserved in all Drosophila 12 species examined, particularly a 70-bp region. To test if these sequences are responsible for the wrapper expression pattern in embryos, this genomic region was amplified within a 884-bp fragment, and then fused it to the green fluorescent protein (GFP) reporter gene within the pHstinger vector, which contains a minimal Hsp70 promoter. This DNA construct (wrapper W:GFP) was injected into D. melanogaster embryos using P element-mediated transformation to generate stable fly lines. Embryos containing this construct express GFP in midline glia beginning at stage 12 of embryogenesis and throughout larval stages. It was confirmed that GFP was expressed in midline glia by staining embryos simultaneously with either (1) wrapper and GFP or (2) sim and GFP. Because wrapper protein is found at the surface of midline glial cells, but the GFP produced by pHstinger localizes to the nucleus, wrapper protein encircles the GFP in these cells. The wrapper W:GFP reporter construct also drives expression in a few additional cells within the lateral CNS and muscles, a pattern that differs from the endogenous wrapper expression pattern. This suggests that the W fragment, although sufficient to drive high levels of expression in midline glia, lacks certain sequences that exclude expression in lateral CNS cells. To confirm the midline expression pattern generated by the reporters, all subsequent experiments were performed by staining embryos with both sim and GFP at stage 16 of embryogenesis. These experiments revealed that GFP generated by the wrapper W:GFP reporter gene was indeed expressed in the midline glia, but not in the cells that develop into midline neurons (Estes, 2008).

Next, to determine the minimal sequences required to provide expression in midline glia, this 884-bp region was divided into several subregions, fused to GFP within the pHstinger vector and tested for the ability to drive midline expression in transgenic embryos. Region E, extending from sequences -756 to -286, is sufficient to drive high levels of GFP expression in midline glia. Moreover, a smaller 166-bp (-492 to -327) G fragment, and an even smaller 119-bp (-492 to -374) internal K fragment, that both include the highly conserved region, are also sufficient to drive GFP expression in midline glia, but the level of expression is reduced compared to that of the E fragment and the intact 884-bp W fragment. None of the other reporter constructs drove GFP expression in the midline. The K fragment is also expressed in a subset of midline neurons, including progeny of the median neuroblast, suggesting that the larger W, E, and G fragments contain a silencer, which is absent from the K fragment and normally represses expression in these midline neurons (Estes, 2008).

Next, to determine if the observed conservation at the sequence level between Drosophila species reflects conservation in function, the corresponding E region from D. virilis was tested to see if it could drive GFP reporter expression in the midline glia of D. melanogaster. The E region is also located upstream of wrapper in D. virilis and is 476 bp in length, while it is 462 bp in melanogaster. The entire E region is 58.4% identical in the two species, and the 70-bp highly conserved section differs by only six nucleotides. The midline expression pattern provided by the D. virilis wrapper E:GFP construct in D. melanogaster flies is indistinguishable from that of the corresponding D. melanogaster E region. These results suggest that the location and function of the regulatory sequences of wrapper have been conserved between D. melanogaster and D. virilis (Estes, 2008).

To determine if previously identified midline transcription factors affect wrapper through these regulatory sequences, the wrapper W:GFP reporter gene was tested in a number of mutant backgrounds. First, the effect of sim mutations on the reporter gene was tested by placing the 884-bp wrapper W:GFP transgene into a sim^H9 mutant background, a mutation that eliminates Sim protein expression. In this background, GFP expression was abolished in most cells, suggesting that sim expression is required for wrapper transcriptional activation in the midline. A few remaining cells did express GFP and these are likely lateral CNS cells also observed in wild-type embryos containing the wrapper W:GFP reporter (Estes, 2008).

Next, the reporter gene was tested in a spitz (spi) mutant background. Spi is a signaling molecule that plays multiple roles during Drosophila development. Wrapper protein is normally found on the surface of midline glia where it mediates direct contact with the lateral CNS axons that cross the midline and promotes survival of midline glia. In wrapper mutant embryos, this intimate interaction cannot occur and additional midline glia die. The amount of spi signaling provided by lateral CNS axons determines how many midline glia survive in each segment. The spi mutation severely disrupted CNS development so that the sim positive cells remained on the ventral surface of the embryo. Only a few of the sim positive cells also express GFP driven by wrapper regulatory sequences, suggesting these are the remaining midline glia. The cells expressing sim, but not GFP, are likely midline neurons, while cells expressing GFP and not sim are lateral glia, because they also express reversed polarity (repo), a marker of lateral CNS glia. These results indicate spi mutations reduce the number of midline glia in the embryo and also reduce expression of the wrapper W:GFP reporter gene (Estes, 2008).

In addition to sim and tgo, the transcription factors Dichaete (D), a Sox HMG protein, and Dfr, a POU domain protein, regulate genes expressed in midline glia. The D protein directly interacts with the PAS domain of Sim and the POU domain of Dfr and all three genes activate expression of slit in midline glial . The wrapper W:GFP construct was tested in both a D and dfr mutant background. In both cases, the number and behavior of midline cells was altered and they did not migrate to the dorsal region of the ventral nerve cord, as they normally do. While development of midline cells was disrupted in these mutant backgrounds and fewer midline glia were present, robust GFP expression was still observed from the reporter construct in the midline cells that remained, suggesting that (1) D and Dfr do not directly activate wrapper via these regulatory sequences, (2) additional, redundant factors exist that can substitute for them, or (3) they can substitute for one another (Estes, 2008).

In summary, midline cell development was disrupted in sim, spi, D, and dfr mutant backgrounds. The sim^H9 mutation eliminated midline glia and neurons, while a mutation in spi eliminated most midline glia. As predicted, both sim and spi mutations severely reduced the number of cells expressing GFP driven by the wrapper W:GFP reporter gene. In the D and dfr mutants, the number of midline glia was reduced and the remaining midline glia expressed high levels of GFP (Estes, 2008).

Ectopic sim expression converts neuroectodermal cells into midline cells and activates downstream, midline genes. To test the effect of ectopic sim on wrapper expression, sim was overexpressed using the UAS/GAL4 system and it was found that wrapper was expressed in neuroectodermal cells outside of the midline, but not in all cells that overexpress sim. In the UAS-sim/da-GAL4 embryos, wrapper is activated in cells that correspond to the lateral edges of the CNS and the cells in the anterior of each segment, with gaps in the expression pattern. Next, whether overexpression of the secreted form of spi could expand wrapper to cells outside the midline was tested. Ectopic expression of secreted spi with the da-GAL4 driver also expanded wrapper expression. To determine if it is possible to expand the expression domain of wrapper further, sim was overexpressed together with spi. This caused additional expansion of the wrapper domain into broad stripes within ectodermal cells. In addition, overexpression of either sim or spi causes severe disruption in embryonic development (Estes, 2008).

Next, the ability of sim and spi, either alone or together, to expand expression of the wrapper reporter genes was tested. Expression from both the full-length reporter construct, wrapper W:GFP, and the smaller wrapper G:GFP construct expanded in the UAS-sim/da-GAL4 embryos to a greater extent than the endogenous wrapper gene. The expression pattern provided by the reporter constructs differs from the endogenous wrapper expression pattern, suggesting that either (1) some of the sequences that normally repress wrapper in tissues outside the midline glia may be missing in these wrapper W and G constructs, or (2) ectodermal cells overexpressing sim may undergo cell death and the GFP marker may be more stable in these dying cells compared to wrapper. Overexpression of spi alone also expanded reporter gene expression driven by both the wrapper W:GFP and wrapper G:GFP constructs. The GFP expression domain was expanded to a greater extent in embryos overexpressing sim together with spi compared to those overexpressing either gene alone. Taken together, the results indicate that (1) limiting the wrapper regulatory sequences and (2) increasing the cells that express sim and spi converts the highly specific expression pattern of wrapper from a single strip of CNS cells to a more general pattern throughout the ectoderm of the embryo. In addition, these results suggest that both the sim transcription factor and spi signaling molecule can activate transcription through these sequences derived from the regulatory region of wrapper (Estes, 2008).

To both (1) identify functionally important motifs needed for wrapper expression and (2) determine if all the invariant nucleotides within the conserved 70-bp region of wrapper are essential for the observed midline glial expression pattern, effects of select mutations within the wrapper G region were tested. Previous studies have demonstrated the importance of sim/tgo, D, dfr, and spi for the expression of midline glial genes and, therefore, possible binding sites for these factors were sought. To examine both predicted binding sites, as well as other conserved sequences that may contain binding sites for novel factors, the region was divided into eight motifs that were tested for their effect on midline glia expression (Estes, 2008).

Each of these conserved motifs was tested by changing 2-3 nucleotides in the context of the D. melanogaster G fragment. The altered G fragments were then inserted independently into the pHstinger vector and injected into fly embryos to test their ability to drive midline expression (Estes, 2008).

Despite the high degree of conservation within this region, only four of the eight mutations that were tested (G1, G2, G5, and G7) caused a noticeable reduction in reporter expression. Two of the mutation sets destroyed midline expression of the G reporter construct. The putative Sim/Tgo binding site (G2: CACGT) was needed for midline expression, because changing this sequence to GAAGT eliminated midline glial expression. In addition, another sequence, ATTTTATC (G5), located upstream of the G2, was required for expression of the reporter gene in wild-type embryos and changing this sequence to ATTGGATC eliminated midline glial expression. Two additional sites within the G fragment of wrapper are needed for midline expression: CGGAGAG (G7) and CACAAT (G1). If either of these motifs is altered, midline glial expression is greatly reduced, but not completely eliminated (Estes, 2008). In contrast, the other four sets of mutations had no detectable negative effect on midline glial expression of the reporter gene, even though these sequences are conserved in all 12 Drosophila species. Mutation sets G4, and G8 did cause a low level of reporter gene activation in some midline neurons, suggesting that repressor proteins present in midline neurons may interact with these regions of the wrapper regulatory region. Finally, mutation G3 had no detectable positive or negative effect on expression of the reporter gene, despite being conserved in all 12 Drosophila species. In summary, the various mutations had three different effects on expression driven by the wrapper regulatory sequences: (1) some reduced midline glial expression, (2) some caused the inappropriate activation of the wrapper reporter in midline neurons, and (3) one was conserved, but apparently had no effect on wrapper regulation, in the context of the experiments presented here (Estes, 2008).

Therefore, these experiments suggest that Sim/Tgo heterodimers may directly regulate wrapper gene expression. (1) Activity of the wrapper W:GFP reporter gene is severely reduced in a sim mutant background, suggesting sim is necessary for expression of this transgene and that sim regulates wrapper by activating transcription through these sequences. (2) Midline activity of the wrapper reporter gene is abolished by eliminating the single CME (CACGT) present within this region. (3) wrapper reporter gene expression is expanded in sim overexpression embryos. Future biochemical studies will determine if Sim/Tgo heterodimers directly interact with the wrapper regulatory motif identified in this study (Estes, 2008).

The studies described in this study demonstrate that the wrapper reporter genes are sensitive to levels of spi signaling. Mutations in spi reduce wrapper reporter gene expression and overexpression of the secreted form of Spi, together with Sim expands, not only the expression domain of the endogenous wrapper gene, but the wrapper reporter genes as well. Spi binds the Epidermal Growth Factor Receptor in midline glia, leading to MAPK activation and subsequent activation of the ETS transcription factor, pnt. Therefore, it may be Pnt that directly activates wrapper transcription through the regulatory sequences studied in this study. One of the identified motifs needed for transcriptional activity of wrapper is: CGGAGAG, which loosely conforms to the consensus binding site for ETS transcription factors (C/A)GGA(A/T)(A/G)(C/T). However, further experiments are needed to determine if Pnt directly interacts with these regulatory sequences, as well as the precise mechanism whereby spi signaling regulates wrapper. Taken together with previous studies, these results suggest that the spi signaling pathway may play at least two roles in promoting survival of midline glia: (1) activating wrapper, needed for neuron-glial interactions and (2) phosphorylating, thereby inactivating Head involution defective, which would otherwise cause programmed cell death in midline glia (Estes, 2008).

Many genes expressed in the CNS of metazoan organisms are regulated through synergistic interactions between Sox HMG-containing proteins and POU domain proteins. Recently, many vertebrate genes expressed in the developing CNS have been shown to contain highly conserved noncoding DNA regions enriched for binding sites for three classes of transcription factors: Sox, POU, and homeodomain proteins. Experiments indicated that Sox and POU proteins work together to activate, while homeodomain proteins repress and limit expression of CNS genes. Interestingly, several motifs identified in this study as important for regulation in midline glia of Drosophila resemble binding sites for Sox (G1: CACAAT), POU (G4: ATGCAAAT, G6: ATGCAACA, and G8: ATGCGTGG), and homeodomain proteins (G5: ATTTTATC) (Estes, 2008).

That the wrapper K:GFP, but not the wrapper G:GFP construct is expressed in certain midline neurons, identifies a midline neural silencer in the 43-bp region present in the G fragment, but absent in the K fragment. Within this region, 27 bp are highly conserved in all 12 Drosophila species and two of the three mutations in the G fragment that cause slight activation of reporter gene expression in midline neurons are found within the 43-bp region. All three sites that lead to activation in midline neurons, G4, G6, and G8, conform to a POU domain binding site, suggesting a POU domain protein expressed in midline neurons may bind to one or more of these sites to keep the wrapper gene silent (Estes, 2008).

One POU domain protein, Dfr, binds to the sequence ATGCAAAT in other gene regulatory regions to activate transcription, including those of two genes expressed in midline glia: dfr itself and slit. This sequence is found at site G8 in the wrapper regulatory region, but when changed to ATGCTAGC, caused a low level of activation in midline neurons, rather than reducing expression in midline glia. Although the number of midline glia is reduced in a dfr mutant background, those that remain express a high level of reporter gene expression driven by wrapper sequences and the results suggest dfr is not absolutely required for wrapper reporter gene expression in midline glia (Estes, 2008).

Mutations in the POU domain motifs within the wrapper regulatory sequences suggest a notable difference between the CNS genes studied previously in vertebrates and the midline glial gene studied here. The POU domain binding sites appear to limit expression in midline neurons (rather than activate expression as in vertebrate CNS genes), and it is the Sox and homeodomain binding sites that are needed for activation. This may reflect a key difference in regulatory control of glial vs. neural genes and it is plausible that other midline glial genes excluded from midline neurons will contain silencer elements similar to the one identified in this study, but further experiments are needed to confirm this (Estes, 2008).

Common motifs shared by conserved enhancers of Drosophila midline glial genes

Coding sequences are usually the most highly conserved sectors of DNA, but genomic regions controlling the expression pattern of certain genes can also be conserved across diverse species. In this study, five enhancers were identified capable of activating transcription in the midline glia of Drosophila melanogaster and each contains sequences conserved across at least 11 Drosophila species. In addition, the conserved sequences contain reiterated motifs for binding sites of the known midline transcriptional activators, Single-minded, Tango, Dichaete, and Pointed. To understand the molecular basis for the highly conserved genomic subregions within enhancers of the midline genes, the ability of various motifs to affect midline expression, both individually and in combination, were tested within synthetic reporter constructs. Multiple copies of the binding site for the midline regulators Single-minded and Tango can drive expression in midline cells; however, small changes to the sequences flanking this transcription factor binding site can inactivate expression in midline cells and activate expression in tracheal cells instead. For the midline genes described in this study, the highly conserved sequences appear to juxtapose positive and negative regulatory factors in a configuration that activates genes specifically in the midline glia, while maintaining them inactive in other tissues, including midline neurons and tracheal cells (Fulkerson, 2010).

The results described in this study indicate that the four genes expressed in the midline glia contain enhancers with subregions conserved in 11 or 12 of the sequenced Drosophila genomes. These conserved subregions contain one or more of the four motifs previously identified in the wrapper regulatory region, are highly A/T rich, and needed for robust expression in the midline. These results confirm the importance of several transcription factor-binding sites for midline glial activation. One of these sites, the CME, binds both Sim/Tgo and Trh/Tgo heterodimers and, when multimerized, can drive reporter gene expression in both midline and tracheal cells. Two lines of evidence indicate that the context of the CME determines whether or not it can be utilized to drive expression in these two tissues. (1) The sequences flanking the CMEs are highly conserved in the four genes discussed in this study, Glec, oatp26f, liprinγ and wrapper, suggesting that the location and sequence of other transcription factor-binding sites are constrained. (2) Changing the sequences flanking the CME in the synthetic multimers can eliminate expression in the midline, trachea, or both tissues (Fulkerson, 2010).

A multimerized CME in the context of the 4Toll:GFP reporter was expressed in both the midline and trachea and quite sensitive to slight modifications in flanking sequences. Changing 5-7 nucleotides on either side of the CME within this multimerized construct either substantially elevated expression in the trachea and eliminated it in the midline (T rich:GFP) or eliminated expression in both tissues (Sox:GFP). Additional combinations between the CME and one of the other midline glial motifs restricted expression to the midline (Pnt:GFP) or the trachea (POU:GFP). These results indicate that testing binding sites for two different factors next to one another can disrupt the endogenous ordering and spacing of the sites within the enhancers. Significantly, the Toll:GFP and Pnt:GFP reporters, unlike the intact enhancers described in this study, drive GFP expression in both midline neurons and glia. This midline expression pattern suggests that the synthetic multimers may lack repressor-binding sites that restrict expression to midline glia. Taken together, these results demonstrate the sensitivity of CME function to flanking sequences within the midline enhancers (Fulkerson, 2010).

Existing experimental evidence suggests that unlike most transcription factors, Sim/Tgo heterodimers (as well as Trh/Tgo heterodimers) preferentially binds one sequence over all others: ACGTG, the CME. Within the enhancers described in this study, sites flanking the CME have remained unchanged over evolutionary time due, in part, to similarities between binding sites for Sim and Trh and the molecular consequences of changing nucleotides adjacent to the CME. This conservation may ensure transcription is restricted to the midline glia and repressed in tracheal cells. In addition to the midline enhancers reported in this study, regions conserved among Drosophila species were found within the known midline enhancers. For instance, a 1.0-kb enhancer present in the first intron of slit drives expression in the midline glia and it contains a single CME and a 32-bp sequence conserved in 11 Drosophila species. It is important to note that the number of midline enhancers described in this study is limited and not all the midline glial enhancers are likely to exhibit such a high degree of conservation. For instance, a midline enhancer of the ectoderm3 gene, was identified that exhibits much less conservation among Drosophila species and presently, the basis for the observed variation among enhancers is unknown (Fulkerson, 2010).

The Ets transcriptional activator, pnt, a downstream effector of EGFR signaling, and Drifter, a POU domain protein, are expressed in both embryonic midline glia and tracheal cells. Previous studies have shown that deleting a POU domain-binding site within an enhancer of rhomboid eliminated expression in tracheal cells, but did not affect its midline glial expression. The results described in this study confirm and extend these results and suggest that the location of a POU domain-binding site relative to the CME can play a role in determining if a gene is expressed in the midline glia, the trachea or both. Moreover, swapping the PAS domains between Sim and Trh proteins indicated that additional, midline or tracheal specific cofactors bind to the PAS domains of the individual proteins and likely to determine which genes are expressed in the two different cell types. This may be the reason sequences adjacent to the CME play such a critical and sensitive role in determining which tissues express the various reporter genes described here. To activate the midline genes, Sim may interact with Drifter and Pnt and bind to sequences flanked by different binding sites compared with sequences bound by Trh, Drifter, and Pnt needed to activate tracheal genes. The simplicity of the multimers studied in this paper raise the possibility that different PAS heterodimers may specifically interact with other factors, such as Drifter and Pnt, in a manner that depends on the relative location and/or distance between each binding site, as has been described for nuclear hormone receptor complexes (Fulkerson, 2010).

The results confirm those of Swanson (2010), who found binding sites can be juxtaposed in different ways within enhancers to favor particular short-range interactions, and, in this way, various combinations of transcription factor binding sites (inputs) can result in more than one output. Similarly, the motifs described in this study can be combined in different ways that result in either midline or tracheal expression. The results indicate the proximity of the CME to activators, one another and/or to repressors could contribute to the level of expression observed in the trachea and midline. This study focused on activator sites, but repressor sites are also likely present and restrict expression to certain cell types. Previous studies in Drosophila embryos have revealed the complexity of the transcriptional regulatory 'grammar' and have shown that the transcriptional output from various genes can be determined by the stoichiometry, affinity, spacing, arrangement, and distance between activator and repressor sites (Fulkerson, 2010).

The high degree of conservation within the midline enhancer subregions examined in this study here belies known properties of transcription factors and their recognition sequences, as well as observations made for many early developmental regulators of Drosophila development. Most transcription factors can vary considerably in the sequences they recognize and tend to bind to related sites with different affinities. This property would suggest that enhancers need not be strictly conserved to function, in contrast to what is reported here. The pattern of conserved sequences within these identified enhancers suggests that the transcription factors that bind these regions do so in a conserved order and spacing pattern. These results suggest that Sim and Trh may interact with other proteins to form an 'enhanceosome'-like complex, similar to that observed in the regulation of the interferon-β gene, in which activators and HMG proteins interact to form a specific multiprotein complex, with a defined structure. This model contrasts with the 'information display/billboard' model of enhancer function. In that model, enhancers are bound by a group of independent factors or group of factors that work together to promote or repress transcription in particular cell types. An important distinction between the two models is the arrangement of binding sites within an enhancer. Within an enhanceosome, the arrangement of binding sites relative to each other is constrained, whereas within a billboard enhancer, the relative arrangement of binding sites is rather flexible as long as a sufficient number of binding sites work together, in many possible configurations, to recruit factors for transcriptional activation (Fulkerson, 2010).

Results obtained with the midline glial genes examined in this study suggest that midline enhancers may consist of a nucleating enhanceosome-like region that combines with an 'information display/billboard' constellation of additional binding sites. This is supported by results obtained with the 70-bp conserved region of wrapper. When tested alone, it only marginally drives midline expression, whereas in the context of the 166-bp enhancer, it works quite well. Moreover, the 166-bp region of virilis cannot function on its own, but drives high levels of expression in the midline glia of melanogaster in the context of the larger, 476 bp region. That the 166-bp region from virilis cannot work efficiently in the midline suggests the transcription complex that binds to this region may be slightly different in virilis compared with melanogaster. For each enhancer described in this study, the presence of the conserved region is required to obtain expression in the midline glia (Fulkerson, 2010).

After comparing vertebrate genomes and generating reporter constructs with highly conserved noncoding sequences, Bailey (2006) noticed that many of these direct expression to regions of the CNS. It is possible that enhancers of CNS genes are more conserved compared with other gene sets, such as early developmental regulators of Drosophila that have been studied in detail. This may be due to the highly conserved nature of the transcription factors that regulate gene expression in this tissue, many of which have analogous functions in flies and mammals). Sox-binding sites are present throughout conserved regions of CNS genes and one of the similarities between these conserved CNS genes, the extensively characterized interferon-β enhanceosome and midline glial genes is the importance of HMG proteins. These proteins may bend the DNA, facilitating binding to highly structured, multiprotein complexes. The enhancers described here likely bind PAS and Sox proteins together with other conserved CNS regulators and it may be this combination of transcription factors that contributes to the similarly conserved arrangement of binding sites (Fulkerson, 2010).

Numerous combinations of transcription factor binding sites can be used to drive expression in many tissue types. Despite the conservation found in this study, binding sites for transcription factors do vary considerably, making it, at times, difficult to identify enhancers based on sequence conservation. In certain cases, changes within enhancers can generate diverse phenotypes between Drosophila populations. The continuing challenge is to understand both the forces constraining the enhancer sequences between Drosophila species, as well as how changes in these regions lead to significant modifications in the expression pattern of a gene, which over the long term, leads to variation among Drosophila populations and eventually, Drosophila species. For the midline genes described in this study, selection has stabilized the constellation of binding sites found within enhancers, resulting in their conservation among Drosophila species over approximately 40 million years of evolution (Fulkerson, 2010).

Wrapper, a novel member of the Ig superfamily, is expressed by midline glia and is required for them to ensheath commissural axons in Drosophila

The midline glia are specialized, nonneuronal cells at the midline of the Drosophila central nervous system (CNS). During development, the midline glia provide guidance cues for extending axons. At the same time, they migrate and help separate the two axon commissures. They then wrap around and ensheath the commissural axons. In many segments, a few of the glia do not enwrap the axons, and these cells die. The wrapper gene encodes a novel member of the immunoglobulin (Ig) superfamily. Wrapper protein is expressed specifically on the surface of midline glia. In wrapper mutant embryos, the midline glia express their normal guidance cues and migrate normally. However, they do not ensheath the commissural axons, and as a result, the glia die. In the absence of Wrapper, the two axon commissures are not properly separated (Noordermeer, 1998).

The wrapper gene was identified in a reverse genetic screen for secreted and transmembrane proteins expressed in the Drosophila CNS. The screen was based on a new method for screening large numbers of cDNAs by whole embryo in situ hybridization. The cDNA library for the screen was prepared from normalized cDNA made from rough endoplasmic reticulum-bound mRNAs and thus was enriched for clones encoding membrane and secreted proteins. One of the cDNAs uncovered is specifically expressed by midline glia. This cDNA led to the identification and genetic analysis of the wrapper gene, named on the basis of its mutant phenotype (Noordermeer, 1998).

This cDNA clone was found to contain only part of the open reading frame (ORF). Rescreening the original library with the initial cDNA led to the identification of four more independent cDNA clones, one of which contains an apparently complete ORF. This cDNA (1.9 kb) recognizes a single transcript on a Northern blot of ~2 kb, suggesting that this clone contains the complete transcription unit (Noordermeer, 1998).

DNA sequence analysis of the wrapper cDNA reveals the presence of a single ORF that encodes 500 amino acids. Sequence comparison with genes in the public databases uncovered several domains characteristic of the Ig superfamily. The ORF starts with a stretch of 23 hydrophobic amino acids constituting a putative signal peptide. In addition, the putative Wrapper protein contains three Ig domains and one Fibronectin type III domain. A triproline sequence is found in the first of the three Ig domains. This sequence is often observed in the hinge region of Ig cell adhesion molecules and is believed to create a turn in protein structure. A hydrophobic region, proceeded by a short hydrophilic domain indicative of a glycosylphosphatidylinositol (GPI) lipid anchor site is present at the C terminus. This domain organization is also observed in two other Ig superfamily members in insects: Klingon in Drosophila and Rega-1 in grasshopper. While their protein domain structures are similar, the overall amino acid identity between these three family members is not very high; the percentage identity between Wrapper and Klingon is 21%, between Wrapper and Rega-1 is 16%, and between Klingon and Rega-1 is 20%. The homology of the individual Ig and Fibronectin III domains between the three proteins ranges from 19% to 34% and from 22% to 28%, respectively (Noordermeer, 1998).

Many Ig superfamily members, including Klingon, have been shown to mediate homophilic cell adhesion in a cell culture assay in which Drosophila S2 cells are transiently transfected with the cDNA. No evidence has been found for homophilic cell adhesion of Wrapper in transfected S2 cells (Noordermeer, 1998).

EM analysis of embryos that lack the wrapper gene shows that the function of wrapper is highly specific; in its absence, the midline glia migrate to their appropriate positions but do not enwrap the commissural axons, thereby vastly reducing the cell surface contact between the midline glia and the axons that cross the midline. This lack of close contact correlates with the increased death of midline glia, confirming earlier observations that intercellular communication between glia and axons is essential for midline glia survival. In addition, some commissural axons stray from their usual trajectories along the commissural tracts, resulting in an incomplete separation and in some cases, partial fusion of the anterior and posterior commissures. Wrapper expression at the midline can still be detected during the third larval instar stage and throughout the larval brain, indicating that Wrapper may also have a role in postembryonic CNS development. Nothing is known about this potential larval function (Noordermeer, 1998).

Wrapper shares structural features with other Ig superfamily members expressed in the CNS of a variety of species. Wrapper is most closely related to Klingon expressed in the Drosophila compound eye and to REGA-1 expressed on sheath cell processes in the grasshopper embryo. All three proteins have a similar overall structure; each has three Ig domains, a single Fibronectin Type III domain, and a putative PI linkage. However, the percent amino acid identity among the three is modest, ranging from 16% to 21% over the entire length of the proteins and between 19% and 34% for individual domains. At present, there is little evidence whether or not the function of these proteins is conserved during development. Given its mutant phenotype of perturbing glial-axonal interactions, it is proposed that Wrapper on midline glia interacts with another protein on the surface of commissural axons. The identity of this putative binding partner is unknown (Noordermeer, 1998).

Many events at the midline involving midline glia and commissural axons occur normally in wrapper mutant embryos. Midline glia appear to produce appropriate guidance cues, and pioneering commissural axons project normally. The midline glia make their normal posterior migrations. However, later in development, after the formation of the commissures is well underway, defects appear in the shape and separation of the two commissures. The two commissures are not completely separated, and some axons are observed abnormally wandering between them. Thus, Wrapper appears to play a role in the development and maintenance of the physical separation between the commissures (Noordermeer, 1998).

EM analysis shows that in wild-type stage 15 embryos, the midline glia start sending out long, intricate processes to cover and, by stage 16, completely enwrap the commissural axons. In wrapper embryos, no glial processes are extended around the axons, and often the commissural axons are completely uncovered dorsally and lack ensheathment. In addition, at this later stage, often many of the midline glia are missing and have presumably died and been removed by macrophages (Noordermeer, 1998).

The role that Wrapper plays in glial cell survival is not completely clear. The data suggest a direct role in glial-axonal interactions and a resulting role in survival versus death. Up to 50% of midline glia die via programmed cell death during normal development, including a subset of cells expressing Wrapper. Clearly, the expression of Wrapper does not in and of itself destine a cell to survive or die. Rather, the expression of Wrapper appears to be permissive of some essential cell-cell interaction that is required for survival. Wrapper appears to permit glial cells to have intimate contact with commissural axons. Those that do have this intimacy survive, and those that do not apparently die. Earlier studies indicated a correlation between axonal contact and glial survival. The results presented in this study confirm and extend that model. An alternative explanation is that Wrapper is part of a trophic signal that is responsible in part for survival of midline glia; in the absence of Wrapper, cells start to die and consequently lose or fail to make contact with commissural axons (Noordermeer, 1998).

Another member of the Ig superfamily has been shown to play a similar role to Wrapper in the control of cell death. In the Drosophila compound eye, the IrreC-rst gene has been shown to play an important role in the selection of which cells live or die in the areas between ommatidia. The IRREC protein is also a member of the Ig superfamily and is closely related to the vertebrate cell adhesion molecule SC1/BEN. IRREC protein accumulates at the border of primary pigment cells and the interommatidial cells (IOC) in the developing retina and facilitates the aligning of the IOC and their subsequent death. In the absence of IRREC, the IOC fail to line up and are not eliminated by apoptosis. The molecular mechanism by which normal IRREC expression facilitates selection of cells for survival has not been elucidated, but both Wrapper and IRREC provide examples of proteins involved in an important developmental process: the regulation of intercellular contacts that facilitate the survival of essential cells versus the elimination of excess cells. How either protein controls cell-cell interactions, and as a result regulates survival versus death, is unknown. But it suggests that normal cell-cell interactions mediated by IRREC in the eye and Wrapper at the midline lead to the suppression of the cell death program. Cells that do not participate in these normal cell-cell interactions are eliminated (Noordermeer, 1998).

Previous genetic analyses have identified two other genes in Drosophila, gliotactin and a Drosophila neurexin-related protein, which participate in the ensheathing of peripheral axons by peripheral glia. In the absence of either of these two genes, the glia form close contact with the axons, wrap around them, and survive, but the glia fail to form a complete blood-nerve barrier, presumably because glial septate junctions do not properly form. The wrapper gene controls an earlier event in glial-axonal contact. In its absence, the midline glia do not begin to wrap commissural axons, and instead a subset of them die (Noordermeer, 1998).

When Wrapper is ectopically expressed using the GAL-4 system, increased cell death is often induced. Panneuronal expression of Wrapper using an elav-GAL4 line leads to extensive cell death of the midline glia and the death as well of other subsets of lateral glia and neurons. In terms of the death of midline glia, this gain-of-function phenotype resembles the wrapper loss-of-function. Ectopic expression at the midline, the site of normal Wrapper expression, did not result in an apparent phenotype, suggesting that the presentation of increased amounts of Wrapper protein on the midline glia does not interfere with normal midline glia cell functions or interactions. Expression by a subset of photoreceptor neurons in the eye or by embryonic muscles also does not result in apparent phenotypes. These results suggest that the induction of glial and neuronal cell death following ectopic Wrapper expression is specific to the CNS and is not the result of a nonspecific cytotoxic activity of the Wrapper protein. Wrapper is not a killer protein. Rather, a simple hypothesis is that ectopic expression of Wrapper protein on neuronal surfaces sequesters a putative axonal Wrapper-binding protein or receptor. This abnormal sequestration possibly prevents normal interactions between glia and axons during development and thus causes the same death as is seen in wrapper mutant embryos. In this way, the wrapper gain-of-function resembles the wrapper loss-of-function. Identification of the Wrapper-binding partner will be required to evaluate this hypothesis (Noordermeer, 1998).

Search PubMed for articles about Drosophila Wrapper

Bailey, P., et al. (2006). A global genomic transcriptional code associated with CNS-expressed genes. Exp. Cell Res. 16: 3108-3119. PubMed ID: 16919269

Estes, P., Fulkerson, E. and Zhang, Y. (2008). Identification of motifs that are conserved in 12 Drosophila species and regulate midline glia vs. neuron expression. Genetics 178(2): 787-99. PubMed ID: 18245363

Faivre-Sarrailh, C., Banerjee, S., Li, J., Hortsch, M., Laval, M. and Bhat, M. A. (2004). Drosophila contactin, a homolog of vertebrate contactin, is required for septate junction organization and paracellular barrier function. Development 131: 4931-4942. PubMed ID: 15459097

Fulkerson, E. and Estes, P. A. (2010). Common motifs shared by conserved enhancers of Drosophila midline glial genes. J. Exp. Zool. B Mol. Dev. Evol. 316(1): 61-75. PubMed ID: 21154525

Kearney, J. B., Wheeler, S. R., Estes, P., Parente, B. and Crews, S. T. (2004). Gene expression profiling of the developing Drosophila CNS midline cells. Dev. Biol. 275: 473-492. PubMed ID: 15501232

Kiedzierska, A., Smietana, K., Czepczynska, H. and Otlewski, J. (2007). Structural similarities and functional diversity of eukaryotic discoidin-like domains. Biochim. Biophys. Acta 1774: 1069-1078. PubMed ID: 17702679

Klambt, C., Jacobs, J. R. and Goodman, C. S. (1991). The midline of the Drosophila central nervous system: a model for the genetic analysis of cell fate, cell migration, and growth cone guidance. Cell 64: 801-815. PubMed ID: 1997208

Laval, M., Bel, C. and Faivre-Sarrailh, C. (2008). The lateral mobility of cell adhesion molecules is highly restricted at septate junctions in Drosophila. BMC Cell Biol. 9: 38-49. PubMed ID: 18638384

Noordermeer, J. N., et al. (1998). Wrapper, a novel member of the Ig superfamily, is expressed by midline glia and is required for them to ensheath commissural axons in Drosophila. Neuron 21(5): 991-1001. PubMed ID: 9856456

Slováková, J. and Carmena, A. (2011). Canoe functions at the CNS midline glia in a complex with Shotgun and Wrapper-Nrx-IV during neuron-glia interactions. Development 138(8): 1563-71. PubMed ID: 21389054

Stork, T., et al. (2009). Drosophila Neurexin IV stabilizes neuron-glia interactions at the CNS midline by binding to Wrapper. Development 136(8): 1251-61. PubMed ID: 19261699

Swanson, C., Evans, N. C. and Barolo, S. (2010). Structural rules and complex regulatory circuitry constrain expression of a Notch- and EGFR-regulated eye enhancer. Dev Cell 18: 359-370. PubMed ID: 20230745

Wheeler, S. R., Kearney, J. B., Guardiola, A. R. and Crews, S. T. (2006). Single-cell mapping of neural and glial gene expression in the developing Drosophila CNS midline cells. Dev. Biol. 294: 509-524. PubMed ID: 16631157

Wheeler, S. R., Stagg, S. B. and Crews, S. T. (2008). Multiple Notch signaling events control Drosophila CNS midline neurogenesis, gliogenesis and neuronal identity. Development 135: 3071-3079. PubMed ID: 18701546

Wheeler, S. R., Banerjee, S., Blauth, K., Rogers, S. L., Bhat, M. A. and Crews, S. T. (2009). Neurexin IV and wrapper interactions mediate Drosophila midline glial migration and axonal ensheathment. Development 136(7): 1147-57. PubMed ID: 19270173

date revised: 28 December 2011

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

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