InteractiveFly: GeneBrief

extra-extra: Biological Overview | Regulation | Developmental Biology | Effects of Mutation | Evolutionary Homologs | References

Gene name - extra-extra

Synonyms - dHb9

Cytological map position - 66A19

Function - transcription factor

Keywords - CNS, motor neuron cell fate, differentiation

Symbol - exex

FlyBase ID: FBgn0041156

Genetic map position -

Classification - homeodomain

Cellular location - nuclear

NCBI links: Precomputed BLAST | Entrez Gene
Recent literature
Banerjee, S., Toral, M., Siefert, M., Conway, D., Dorr, M. and Fernandes, J. (2016). dHb9 expressing larval motor neurons persist through metamorphosis to innervate adult-specific muscle targets and function in Drosophila eclosion. Dev Neurobiol [Epub ahead of print]. PubMed ID: 27168166
The Drosophila larval nervous system is radically restructured during metamorphosis to produce adult specific neural circuits and behaviors. Genesis of new neurons, death of larval neurons and remodeling of those neurons that persistent collectively act to shape the adult nervous system. This study examined the fate of a subset of larval motor neurons during this restructuring process. A dHb9 reporter (exex), in combination with the FLP/FRT system, was used to individually identify abdominal motor neurons in the larval to adult transition using a combination of relative cell body (CB) location, axonal position and muscle targets. Segment specific cell death of some dHb9 expressing motor neurons was found to occur throughout the metamorphosis period and to continue into the post-eclosion period. Many dHb9>GFP expressing neurons however persist in the two anterior abdominal hemisegments, A1 and A2, which have segment specific muscles required for eclosion while a smaller proportion also persist in A2-A5. Consistent with a functional requirement for these neurons, ablating them during the pupal period produces defects in adult eclosion. In adults, subsequent to the execution of eclosion behaviors, the NMJs of some of these neurons were found to be dismantled and their muscle targets degenerate. These studies demonstrate a critical continuity of some larval motor neurons into adults and reveal that multiple aspects of motor neuron remodeling and plasticity that are essential adult motor behaviors.


dHb9 (FlyBase designation: Extra-extra [Exex]), the Drosophila homolog of vertebrate Hb9, encodes a factor central to motorneuron (MN) development. Exex regulates neuronal fate by restricting expression of Lim3 and Even-skipped (Eve), two homeodomain (HD) proteins required for development of distinct neuronal classes. Exex and Lim3 are activated independently of one another in a virtually identical population of ventrally and laterally projecting MNs. Surprisingly, Exex represses Lim3 cell nonautonomously in a subset of dorsally projecting MNs, revealing a novel role for intercellular signaling in the establishment of neuronal fate in Drosophila. Evidence is provided that Exex and Eve regulate one another's expression through Groucho-dependent crossrepression. This mutually antagonistic relationship bears similarity to the crossrepressive relationships between pairs of HD proteins that pattern the vertebrate neural tube (Broihier, 2002).

To identify genes required for proper neuronal fate specification in the Drosophila embryonic CNS, EMS saturation mutagenesis of the third chromosome was conducted and a screen was performed for changes in the CNS expression pattern of Eve. Changes in Eve expression were assayed because Eve is expressed in a stereotyped pattern of eight dorsally projecting MNs and 12 interneurons in each abdominal hemisegment, and because eve is a known regulator of neuronal fate (Broihier, 2002).

Four alleles of exex were identifed. exex mutant embryos display a highly specific phenotype in which two ectopic Eve-expressing neurons develop per hemisegment. These ectopic Eve-positive neurons appear during late stage 11 in the vicinity of the Eve-positive neurons aCC/pCC. By stage 14, one ectopic Eve-expressing neuron is found adjacent to aCC/pCC, while the other migrates posteriorly and laterally to adopt a stereotyped mediolateral position (Broihier, 2002).

To examine more closely the cell fate changes that occur in exex mutant embryos, the lineal origin of the ectopic Eve-positive neurons was determined. Since in exex mutants, the ectopic Eve-expressing neurons arise immediately adjacent to the sibling aCC/pCC neurons, it was hypothesized that, like aCC/pCC, the ectopic Eve-positive neurons derive from the NB1-1 lineage. To test this, assays were carried out to determine whether an Eve-ß-gal reporter gene normally expressed solely by the aCC/pCC and RP2 neurons is also expressed by the ectopic Eve-positive neurons in exex mutant embryos. In support of the model, both ectopic Eve-positive neurons express ß-gal in exex mutant embryos, indicating that the ectopic Eve-positive neurons likely arise within the NB1-1 lineage. These data indicate that exex regulates neuronal fate by repressing eve expression in the NB1-1 lineage (Broihier, 2002).

These data support a role for exex in the cell-autonomous and nonautonomous regulation of several factors required for the development of distinct neuronal fates. Expression of murine Hb9, a vertebrate homolog of exex is restricted to MNs whose axons exit from the ventral side of the neural tube (v-MNs) (Thaler, 1999). v-MNs and V2 interneurons arise from common progenitors characterized by coexpression of Lim3 and Gsh4 (Sharma, 1998). This shared lineage necessitates the presence of factors that differentiate v-MNs and V2 interneurons. Hb9 activity contributes to the v-MN/V2 interneuron distinction, since V2 interneuron-specific gene expression is derepressed in Hb9 mutant mice (Arber, 1999; Thaler, 1999). Interestingly, MNs whose axons emerge from the dorsal side of the neural tube (d-MNs) and arise from an MN-specific progenitor pool do not require Hb9 function (Thaler, 1999). The restriction of Hb9 expression to those MNs arising from Lim3/Gsh4-positive progenitors suggests that Hb9 function is required only in MNs that need to actively suppress an alternate genetic program (Broihier, 2002 and references therein).

In Drosophila, many NBs produce both MNs and interneurons, suggesting a widespread requirement for factors that function to arbitrate between alternate genetic programs. The data suggest that exex acts cell autonomously to repress Eve in neurons in the NB1-1 lineage, whereas exex acts cell nonautonomously to repress lim3 in dorsally projecting U MNs. Inappropriate expression of eve and lim3 in exex mutants is consistent with exex contributing to proper neuronal fate by suppressing the expression of key determinants of neuronal identity. These results also hint at the possibility that Exex regulates cell fate in a manner analogous to its vertebrate homologs (Broihier, 2002).

Several cell fate changes have been characterized in exex mutant embryos, and these phenotypes have been paired with exex function in distinct neurons. However, Exex is expressed in approximately 30 neurons, and regulatory targets have been identified in only a handful of these cells, strongly suggesting that additional targets exist. Given the enormous complexity of the genetic regulatory network that dictates neuronal fate, the power of Drosophila genetics should provide an indispensable tool for identifying Exex-interacting genes -- as well as other key determinants of neuronal identity (Broihier, 2002).

In the vertebrate neural tube, Hb9, Lim3/4, and Isl1 are elements of a combinatorial code directing neuronal identity and axonal pathfinding (Arber, 1999; Sharma, 1998; Thaler, 1999). Hb9 and Lim3/4 have been shown to be expressed in all MNs exiting the neural tube ventrally, though Lim3/4 are only transiently expressed in these MNs. In the Drosophila CNS, functional analysis and reporter construct expression data have supported roles for Lim3 and Islet in regulating the projections of ventrally projecting neurons. Islet expression has been proposed to be required for the identities of ISNb and ISNd MNs, while Lim3 expression in only ISNb MNs is thought (Thor, 1997; Thor, 1999) to resolve ISN neurons into ISNb and ISNd classes (Broihier, 2002).

This analysis of the Lim3 protein expression pattern argues against its proposed role in distinguishing ISNb trajectories from those of ISNd. Lim3 is expressed much more broadly than suggested by a lim3 reporter gene (Thor, 1999). Lim3 is coexpressed with Exex in five of the six major motor axon branches. In addition, Lim3 but not Exex is expressed in the TN motor axon branch (Thor, 1999). Lim3 is then expressed in neurons that populate all motor axon branches. Thus, differential expression of Lim3 is insufficient to explain how neurons choose between ISNb and ISNd (Broihier, 2002).

One question that then arises is why the motor axon phenotypes of exex and lim3 mutants are specific to the ISNb nerve branch when these factors are expressed widely in MNs. It is possible that the ISNb is generally more sensitive to genetic perturbations than other motor axon branches. Consistent with this, guidance molecules with broad CNS expression patterns display motor axon phenotypes largely confined to ISNb. Alternatively, the axonal phenotypes may be ISNb specific because Exex and Lim3 are expressed in a higher percentage of ISNb-projecting neurons than neurons projecting in other nerve branches. For example, eight MNs that project dorsally in the ISN are Eve positive and Exex/Lim3 negative (Broihier, 2002).

While these data argue against the simple combinatorial code proposed to regulate axon pathway choice, it is still certainly true that a neuron's fate is established largely by the combination of transcription factors it expresses. However, the fact that Exex, Lim3, and Isl are coexpressed in a large number of neurons with different identities indicates that individual neuronal identities are not defined by the mere presence or absence of these factors. Clearly additional as yet unidentified factors are required to create the tremendous cellular diversity found in the CNS (Broihier, 2002).

Additional layers of complexity also likely exist within the combinatorial code. For example, the levels and timing of expression of individual transcription factors may play important roles in directing different cellular fates. Consistent with this possibility, while Exex and Lim3 have largely overlapping expression patterns, their relative levels and duration of expression vary between neurons. The data establish that these two factors act largely in parallel to establish neuronal identity. It will therefore be critical to determine whether Exex and Lim3 act independently on distinct targets or together as members of one transcriptional complex. In this context, it is possible that changes in the relative levels of Exex/Lim3 would alter the composition and functional properties of these complexes. Clearly, future research that identifies additional genes with roles in neuronal fate determination and integrates their functions into the regulatory network that controls neuronal diversity will provide a more lucid picture of the genetic and molecular basis of neuronal diversity (Broihier, 2002).

The data demonstrate that a crossinhibitory interaction between Exex and Eve contributes to their mutually exclusive expression patterns -- Eve is expressed in dorsally projecting MNs, and Exex is expressed in more ventrally projecting MNs. Furthermore, functional studies demonstrate that Eve and Exex regulate axonal trajectories of dorsally and ventrally projecting axons, respectively. Together, these results suggest that the crossrepressive relationship between Exex and Eve helps to ensure that neurons in these two populations acquire distinct identities (Broihier, 2002).

The mutual antagonism of Eve and Exex is similar to the relationship between pairs of HD factors whose crossrepressive interactions are central to neural tube patterning. In the vertebrate neural tube, domains of HD protein expression in distinct progenitor domains are established in response to a Shh gradient. Crossrepressive interactions between these HD factors then appear to refine and maintain the progenitor domains. These proteins likely function as transcriptional repressors and may require the corepressor Groucho (Gro) (Broihier, 2002).

The results suggest that Eve and Exex also mediate their crossrepressive interaction in a Gro-dependent manner. The ability of Eve to repress Exex depends on its Gro-interaction domain, implicating Gro in the Eve side of this crossinhibitory interaction. In support of the idea that Exex acts through Gro to repress Eve, a potential Gro-interaction domain has been identified in Exex. Clearly, the significance of this conserved domain with respect to Exex function must be tested in vivo. Nonetheless, these results highlight the significant mechanistic conservation of neuronal fate specification between Drosophila and vertebrates (Broihier, 2002).

The mutually exclusive expression patterns of Eve and Exex arise in part through a crossinhibitory interaction between the two proteins. exex mutant embryos display several additional Eve-positive neurons, and eve mutants exhibit several additional Exex-positive neurons, arguing that the Eve and Exex expression patterns are established largely independently and then refined by the mutually repressive interaction. In the future, it will be important to identify upstream regulators of eve and exex to understand the manner in which these distinct patterns of gene expression arise. Research in this area is likely to be of general relevance since in Drosophila and vertebrates, Hb9 and Lim3 are coexpressed in nearly identical populations of MNs (Arber, 1999; Thaler, 1999; Broihier, 2002). These data argue that Hb9/Lim3-positive MNs constitute an evolutionarily conserved MN population. Given this, significant overlap is expected between the upstream regulators of Exex/Lim3 in Drosophila and vertebrates (Broihier, 2002).


Transcriptional Regulation

The mutually exclusive expression patterns of Eve and Exex and the ability of Exex to repress Eve led to an investigation of whether Eve exhibits a reciprocal ability to repress Exex. Whether eve represses exex was tested by following Exex in eve1D mutant embryos. This temperature-sensitive allele allowed the circumvention of the early requirement for eve during embryonic segmentation. On average, two ectopic Exex-positive neurons were observed in each hemisegment of eve mutant embryos. The position of these neurons identifies one as RP2 and the other as likely aCC or pCC. Therefore, eve exhibits a reciprocal ability to repress exex in a subset of dorsally projecting MNs (Broihier, 2002).

During segmentation, Eve has been shown to act as a transcriptional repressor and contains two domains with repressive capability -- one dependent on the corepressor Groucho (Gro) and one Gro independent. To determine whether Eve requires Gro to repress Exex in the CNS, Exex expression was assayed in eve null embryos that contain an eve transgene deleted for the Gro-interaction domain. In these embryos, Exex is derepressed in RP2 and one of the corner cells. Since this phenotype is essentially identical to that of eve1D mutants, it is concluded that Eve represses Exex in a Gro-dependent manner. These results demonstrate that Eve/Evx proteins act through Gro to regulate cell fate in the CNS (Broihier, 2002).

To investigate if Eve is also sufficient to repress Exex, Eve was misexpressed in all postmitotic neurons. In these embryos, Exex expression is abolished, demonstrating that Eve is a potent repressor of Exex expression in the CNS. Thus, Eve is both necessary and sufficient to repress Exex. Taken together, these genetic studies demonstrate crossrepressive interactions between exex and eve function to delimit the expression of Exex to ventral and lateral MNs—and Eve to dorsal MNs. Since both Exex and Eve are key cell fate determinants, this mutually repressive relationship likely helps to consolidate distinct MN fates (Broihier, 2002).

Nervous system-specific eve mutants were created by removing regulatory elements from a 16 kb transgene capable of complete rescue of normal eve function. When transgenes lacking the regulatory element for either RP2+a/pCC, EL or U/CQ neurons were placed in an eve-null background, eve expression was completely eliminated in the corresponding neurons, without affecting other aspects of eve expression. Many of these transgenic flies are able to survive to fertile adulthood. In the RP2+a/pCC mutant flies: (1) both RP2 and aCC show abnormal axonal projection patterns, failing to innervate their normal target muscles; (2) the cell bodies of these neurons are positioned abnormally; and (3) in contrast to the wild type, pCC axons often cross the midline. The Eve HD alone is able to provide a weak, partial rescue of the mutant phenotype, while both the Groucho-dependent and -independent repressor domains contribute equally to full rescue of each aspect of the mutant phenotype. Complete rescue is also obtained with a chimeric protein containing the Eve HD and the Engrailed repressor domain. Consistent with the apparent sufficiency of repressor function, a fusion protein between the Gal4 DNA-binding domain and Eve repressor domains is capable of actively repressing UAS target genes in these neurons. A key target of the repressor function of Eve is Drosophila Hb9 (Extra-extra), the derepression of which correlates with the mutant phenotype in individual eve-mutant neurons. Finally, homologs of Eve from diverse species are able to rescue the eve mutant phenotype, indicating conservation of both targeting and repression functions in the nervous system (Fujioka, 2003).

The requirement for Eve in axonal guidance is somewhat more stringent in aCC than in RP2 neurons. Although a significant fraction of mutant RP2s initially extend axons in the same direction as wild-type RP2s, essentially none of mutant aCCs do so. In addition, unlike for RP2s, the aCC phenotype is not significantly rescued by either the HD alone or the HD with the N terminus (which provides no detectable repression activity, but might stabilize the protein). In aCC, as in RP2, the phenotype is partially rescued by including either repressor domain, and the Engrailed repressor domain is able to provide full activity. Furthermore, Eve repressor domains are able to actively repress a UAS target gene in aCC neurons. These data indicate that the primary function of eve in aCC is to actively repress target genes. The more stringent requirements in aCC versus RP2 suggest that there may be different target genes in these two motoneurons, although Drosophila Hb9 is a common target (Fujioka, 2003).

Drosophila Hb9 and Eve are expressed in a non-overlapping pattern in the wild-type CNS, and ectopic Eve expression represses Hb9, indicating that Hb9 is a target gene of Eve. Hb9 is derepressed in the RP2 mutant in both RP2 and aCC, but not in pCC neurons (the RP2 mutant lacks Eve in all three cell types), showing that there are significant differences in target gene regulation in different neurons, even in those derived from the same GMC (in the case of aCC and pCC) (Fujioka, 2003).

When the Eve HD alone is used to rescue the RP2 mutant, Hb9 is repressed in many of the RP2 neurons, and this seemingly stochastic repression correlates with a more normal axonal morphology. However, effective repression, particularly in aCC, requires active repression domains, with either of the repressor domains of Eve alone providing partial activity (in the context of the Eve HD). Although there is a strong correlation in situations of partial rescue between the axonal phenotypes of individual neurons and derepression of Hb9, this correlation is not 100%. This suggests that there may be other key target genes that mediate Eve neuronal function in addition to Hb9. The level of expression of the antigen (Futsch) of the monoclonal antibody 22C10 is reduced in RP2 and aCC in the absence of Eve. However, the gene encoding this antigen is likely to be an indirect target of Eve, because its expression is activated rather than repressed by Eve (Fujioka, 2003).

Either of the repressor domains of Eve is sufficient to give a similar degree of partial rescue of each of the phenotypes studied in the nervous system, including the repression of Hb9, showing that these repressor domains provide a similar function. In fact, two copies of a transgene expressing either EveDeltaC or EveDeltaR are able to rescue to a similar degree as that of one copy of the wild-type transgene. Thus, the recruitment of either of two apparently distinct co-repressors, Groucho or Atrophin, produces the same net result. The two are used in these neurons in an additive fashion to generate the appropriate level of Eve repressor activity, with no apparent target gene specificity (Fujioka, 2003).

dbx mediates neuronal specification and differentiation through cross-repressive, lineage-specific interactions with eve and hb9

Individual neurons adopt and maintain defined morphological and physiological phenotypes as a result of the expression of specific combinations of transcription factors. In particular, homeodomain-containing transcription factors play key roles in determining neuronal subtype identity in flies and vertebrates. dbx belongs to the highly divergent H2.0 family of homeobox genes. In vertebrates, Dbx1 and Dbx2 promote the development of a subset of interneurons, some of which help mediate left-right coordination of locomotor activity. This study identifies and shows that the single Drosophila ortholog of Dbx1/2 contributes to the development of specific subsets of interneurons via cross-repressive, lineage-specific interactions with the motoneuron-promoting factors eve and hb9 (exex). dbx is expressed primarily in interneurons of the embryonic, larval and adult central nervous system, and these interneurons tend to extend short axons and be GABAergic. Interestingly, many Dbx+ interneurons share a sibling relationship with Eve+ or Hb9+ motoneurons. The non-overlapping expression of dbx and eve, or dbx and hb9, within pairs of sibling neurons is initially established as a result of Notch/Numb-mediated asymmetric divisions. Cross-repressive interactions between dbx and eve, and dbx and hb9, then help maintain the distinct expression profiles of these genes in their respective pairs of sibling neurons. Strict maintenance of the mutually exclusive expression of dbx relative to that of eve and hb9 in sibling neurons is crucial for proper neuronal specification, as misexpression of dbx in motoneurons dramatically hinders motor axon outgrowth (Lacin, 2009).

As beautifully illustrated by Ramon y Cajal, the nervous system is remarkable for its diversity of cellular phenotypes. In fact, recent physiological and expression studies suggest the presence of thousands of distinct neuronal subtypes in the mammalian brain. The genetic and molecular basis through which individual or small groups of neurons adopt and maintain specific, often unique, morphological and physiological characteristics (neuronal specification) remains poorly understood (Lacin, 2009).

Studies in mice, Drosophila and C. elegans highlight the importance of a large and growing number of transcription factors, which act in a combinatorial manner to govern neuronal specification. Most of these transcription factors are expressed in complex, partially overlapping patterns of neurons, with the specific differentiated phenotype of a neuron being largely dictated by the precise complement of transcription factors it expresses. As detailed below, much of this work has focused on the specification of distinct motoneuron subtypes due in part to the relative ease of distinguishing individual or groups of motoneurons from each other based on axonal trajectory. Less is known about the factors that govern the specification of interneuron subtypes, even though interneurons outnumber motoneurons and can also be grouped based on morphology. For example, interneurons in the Drosophila central nervous system (CNS) outnumber motoneurons by about 10-fold, and can be roughly divided into intersegmental interneurons, which extend projections that span more than one segment, and local interneurons, which terminate their axons within the segment of origin (Lacin, 2009).

Regulatory interactions, often cross-repressive in nature, between transcription factors that govern neuronal specification help ensure that individual neurons adopt the appropriate cellular phenotype. For example, in vertebrates cross-repressive interactions between the LIM-homeodomain (LIM-HD) proteins LIM-1 (Lhx1: Mouse Genome Informatics) and ISLET1 (Isl1: Mouse Genome Informatics) establish and maintain the non-overlapping expression of these proteins in lateral and medial neurons of the lateral motor column, respectively. Lhx1 and Isl1 direct their respective groups of motoneurons to extend axons dorsally or ventrally into the limb mesenchyme in part by regulating the expression of the repulsive guidance receptor, Epha4. Mutually exclusive expression of two sets of transcription factors also defines distinct motoneuron subtypes in the Drosophila CNS. Here, all motoneurons that project axons to dorsal muscle targets express the homeodomain protein even-skipped (eve). By contrast, most motoneurons that project axons to ventral muscles co-express the LIM-HD proteins Lim3 and Islet (Tailup: FlyBase), and the homeodomain proteins Hb9 (Exex: FlyBase) and Nkx6 (HGTX: FlyBase). Cross-repressive interactions between eve and hb9/nkx6 help maintain these mutually exclusive expression patterns, and these genes in turn help direct the projection patterns of their respective motoneurons along different routes (Lacin, 2009).

Functional studies are beginning to tease apart the transcriptional regulatory networks that govern the specification of postmitotic neurons. However, such analysis is complicated, at least in Drosophila, by the context-dependent nature of many of these regulatory interactions. For example, eve and hb9 are each expressed in about a handful of distinct groups of neurons per hemisegment, yet eve is necessary to repress hb9 expression in only two of the roughly 20 neurons that normally express eve. Similarly, hb9 is required to inhibit eve expression in only one or two of its expressing neurons and is sufficient to repress eve in only a subset of Eve+ neurons in the CNS (Lacin, 2009).

Context-dependent regulatory interactions may reflect the underlying organization of the Drosophila CNS. Essentially all cells in the CNS derive from one of a limited set of stem-cell-like precursors, called neuroblasts. Thirty neuroblasts develop per hemisegment, with each neuroblast dividing in a stem-cell-like manner to produce a largely invariant family or clone of neurons. Many different transcription factors are expressed within the neurons of any one lineage, with such factors exhibiting overlapping or mutually exclusive expression in the lineage in a transcription-factor-dependent manner. In addition, most transcription factors that govern neuronal specification are expressed in multiple different groups of neurons, with each group of neurons likely to derive from a different neuroblast lineage. Thus, context-dependent regulatory interactions between two such transcription factors may often reflect lineage-specific interactions, with these interactions preferentially occurring in lineages in which both factors are expressed versus lineages in which one or the other is expressed. It is presently difficult to test this model, as in general the individual lineages from which distinct groups of neurons marked by the expression of a given transcription factor arise have not been delineated (Lacin, 2009).

In the vertebrate neural tube, Dbx1 and Dbx2, two paralogous genes that encode homeodomain-containing proteins of the H2.0 class, are expressed within the p0, p1 and pD6 progenitor domains, with Dbx1 expression nested within that of Dbx2. The ventral limits of Dbx1 and Dbx2 expression are maintained via cross-repressive interactions with Nkx6-2 and Nkx6-1, respectively. Moreover, whereas Dbx1 and Dbx2 are expressed in neural progenitors but not postmitotic neurons, Dbx+ progenitors give rise to v0, v1 and dl6 interneurons, with a subset of V0 interneurons expressing Evx, the vertebrate ortholog of eve, in a Dbx1-dependent manner. Functional analysis of V0 interneurons reveals that a subset of Evx-negative commissural interneurons makes inhibitory connections with contralateral motoneurons that innervate hindlimb muscles; Dbx function is required in these interneurons to regulate left-right alternation of motoneuron firing, required for proper walking movements (Lanuza, 2004; Lacin, 2009 and references therein).

This paper reports the identification and characterization of the Drosophila dbx gene. Lineage tracing reveals that many Dbx+ neurons share a sibling relationship with Eve+ or Hb9+ motoneurons. The cellular phenotype of these pairs of sibling neurons is strikingly distinct - Dbx+ interneurons are small and extend short axons; Eve+ or Hb9+ motoneurons are large and extend long axons. Notch/Numb-mediated asymmetric divisions establish the non-overlapping expression of dbx and eve, or dbx and hb9, within each pair of sibling neurons. Cross-repressive interactions between dbx and eve, and dbx and hb9, then help maintain the complementary expression profiles of these transcription factors in the relevant sibling neurons, a process crucial for the ability of these neurons to adopt and maintain their distinct differentiated phenotypes (Lacin, 2009).

dbx is expressed in a dynamic pattern of neuroblasts, intermediate precursors termed ganglion mother cells (GMCs), and postmitotic cells. dbx is first expressed in a few neuroblasts and GMCs in gnathal segments during early stage 11. Shortly thereafter, dbx expression initiated in multiple cells within each segment as well as in the brain; based on size, relative sub-epidermal position and marker gene expression, Dbx+ cells appeared to identify a stereotyped subset of neuroblasts, GMCs and postmitotic cells. After stage 12, most Dbx+ cells are postmitotic, with some cells expressing dbx transiently and others retaining dbx expression throughout embryogenesis. At the end of embryogenesis, approximately 33 Dbx+ cells resided per thoracic hemisegment and about 20 Dbx+ cells per abdominal hemisegment. Gnathal segments contained even more Dbx+ cells. Such segment-specific differences in dbx expression probably reflect homeotic gene input (Lacin, 2009).

Many CNS neurons retain dbx expression into larval stages, during which time additional Dbx+ neurons arise in the nerve cord. Dbx+ neurons are also found in the CNS of adults. A few Dbx+ are observed in neuroblasts in thoracic segments of third instar larvae. These Dbx+ neuroblasts appeared to bud off Dbx+ GMCs and neurons, as small clusters of Dbx+ neurons resided immediately adjacent to these neuroblasts. Thus, many neurons maintained dbx expression for extended periods of time in larvae, pupae and adults, consistent with dbx helping to maintain the specific differentiated phenotype of these neurons (Lacin, 2009).

Double-label studies with molecular markers that label all neurons (ELAV), all glia (Repo) or most motoneurons (Eve/Hb9) revealed that essentially all postmitotic Dbx+ cells are interneurons. For example, by the end of embryogenesis all postmitotic Dbx+ cells express ELAV but not Repo. In addition, no Dbx+ neurons express eve or hb9, which together mark most embryonic motoneurons, identifying Dbx+ neurons as interneurons. Together with the data detailed above, these results indicate that as in vertebrates, dbx expression labels progenitor cells of interneurons, and in contrast to vertebrates many interneurons maintain dbx expression in flies (Lacin, 2009).

Dbx+ neurons identify a largely uncharacterized population of interneurons; little if any co-expression is observed between dbx and markers of different subsets of interneurons, such as engrailed, dachshund, nmr-1 (H15 - FlyBase), nmr-2 (mid - FlyBase) and eagle. However, many Dbx+ interneurons are GABAergic; significant co-expression occurs between Dbx and glutamic acid decarboxylase (GAD), a marker of GABAergic neurons. By contrast, no Dbx+ interneurons appeared to be seratonergic, dopaminergic or glutamatergic by the first larval instar stage, and only one Dbx+ interneuron is cholinergic. Since GABAergic interneurons are inhibitory in Drosophila as in vertebrates, it is inferred that many embryonic Dbx+ cells are inhibitory GABAergic interneurons (Lacin, 2009).

dbx expression thus marks a poorly defined population of interneurons. To characterize Dbx+ interneurons in greater detail a modified version of the FLP/FRT system was used to map their lineage. Random clones of tau-myc-GFP+ cells were generated in otherwise wild-type embryos, and then GFP+ lineage clones were screened for that contain Dbx+ neurons. Comparison of the morphology and location of such clones relative to the location and morphology of identified neuroblast lineages as determined by Dil-labeling enabled mapping of essentially all Dbx+ neurons in abdominal segments to five neuroblast lineages. This approach also revealed that at least two additional neuroblasts, preliminarily identified as NBs 2-4 and 6-4, produced Dbx+ neurons in thoracic segments, and that many Dbx+ neurons exhibited at most short axonal projections (Lacin, 2009).

The NB4-2 lineage produces four small Dbx+ interneurons in abdominal segments and seven Dbx+ neurons in thoracic segments. This lineage contains the CoR motoneurons, which project axons out of the segmental nerve (SN), and the Eve+ RP2 motoneuron, which projects its axon out of the inter-segmental nerve (ISN). The four Dbx+ neurons included RP2 sib and at least two CoR sibs. These Dbx+ neurons extended at most short axons, consistent with previous observations that all interneurons in this lineage are local interneurons (Lacin, 2009).

The NB5-2 lineage produces two medial Dbx+ neurons within a large family of over 20 cells. Axons from this clone cross the midline via the anterior and posterior commissures, with a single motoneuron, the AC motoneuron, projecting its axon across the midline and out of the SNb nerve branch (Lacin, 2009).

The NB6-1 lineage produces two Dbx+ neurons in abdominal segments and four Dbx+ neurons in thoracic segments at the end of embryogenesis. Abdominal NB6-1 clones analyzed at stage 13 included three additional Dbx+ neurons as well as dbx expression in NB6-1. Thus, some cells expressed dbx transiently in this lineage. The neurons that retain dbx expression appear to be late-born neurons; they resided at the extreme ventral surface of the NB6-1 family of neurons (Lacin, 2009).

The NB6-2 lineage contains three Dbx+ neurons in abdominal segments and five Dbx+ neurons in thoracic segments. Axonal projections from these clones cross the midline as two bundles through the posterior commissure; these bundles arch anteriorly, mirroring the morphology of NB6-2 clones. These neurons expressed dbx at levels below those observed in other lineages (Lacin, 2009).

The NB7-1 lineage produces eight small Dbx+ neurons in thoracic and abdominal segments, with many of these neurons residing next to the well-characterized and lineally related Eve+ U motoneurons, the close proximity of U motoneurons and Dbx+ neurons reflect sibling relationships in multiple cases. In accord with the previous observation that all interneurons in this lineage are local interneurons, all Dbx+ neurons in this lineage extended at most short axons (Lacin, 2009).

Expression and additional lineage-tracing studies confirmed that the juxtaposition of Dbx+ interneurons next to Eve+ or Hb9+ motoneurons reflects sibling relationships in many cases. For example, double-labeling studies in wild-type embryos revealed transient co-expression of eve and dbx in RP2 sib and in the siblings of the U1-U3 motoneurons immediately after their birth from Eve+ GMCs. Whereas eve expression is quickly extinguished in these cells, dbx expression is maintained in RP2 sib until the end of embryogenesis, and in the U1-U3 sibs for an extended period of time (the presence of eight Dbx+ neurons in the 7-1 lineage rendered it difficult to follow individual Dbx+ neurons throughout embryogenesis) (Lacin, 2009).

To ascertain if all Eve+ U motoneurons share a sibling relationship with Dbx+ neurons, two-cell clones were generated marking each U motoneuron and its sibling. Each U motoneuron can be unambiguously identified based on its relative position. Thus, this approached revealed that, like RP2, the U1 and U5 sibs expressed dbx from shortly after their birth until the end of embryogenesis. Similarly, the U2 and U3 sibs expressed dbx from their birth until stage 14, at which point they begin to downregulate dbx expression. Although the U4 sib did not express dbx after stage 15, two-cell U4 clones were not obtained before stage 15. Thus, the U4 sib may transiently express dbx. It is concluded that most of the sibling interneurons of the RP2 and U motoneurons express dbx transiently or continuously during embryogenesis (Lacin, 2009).

The generation of many two-cell clones within the NB4-2 lineage that contained one Hb9+ CoR motoneuron and one Dbx+ interneuron confirmed sibling relationships between Dbx+ interneurons and Hb9+ motoneurons. In contrast to U motoneurons, individual CoR motoneurons could not be identified unambiguously by position. Thus, the clones may not have marked all CoR motoneurons. However, four five-cell clones were identified that contain three Hb9+ CoR motoneurons and two Dbx+ interneurons, pointing to a one-to-one sib relationship between at least two CoR motoneurons and Dbx+ interneurons. The 5-2 lineage is the only other lineage in abdominal segments that contained Dbx+ interneurons and an Hb9+ motoneuron. Although it was not possible to create two-cell clones containing this motoneuron, sublineage clones are consistent with a sibling relationship between this motoneuron and a Dbx+ interneuron. Thus, lineage and expression studies reveal that dbx expression labels the sibling interneuron of at least seven motoneurons in each abdominal segment; all of these interneurons are local interneurons that extend short axons or no axons (Lacin, 2009).

Analysis of dbx, eve and hb9 expression in embryos homozygous mutant for numb or spdo - two genes that exert opposite effects on Notch/Numb-mediated asymmetric divisions - confirmed the observed sibling relationships between Dbx+ neurons and Eve+ or Hb9+ motoneurons. For example, loss of numb function led to reciprocal effects on eve and dbx expression in the 4-2 and 7-1 lineages, with duplication of Eve+ U motoneurons occurring at the expense of Dbx+ interneurons in the 7-1 lineage, whereas RP2 sib (Dbx+) was duplicated at the expense of the Eve+ RP2 in the 4-2 lineage. Removal of spdo function elicited the reciprocal effect on dbx and eve expression in these lineages. Similarly, loss of numb (or spdo) function led to reciprocal effects on dbx and hb9 expression in the 4-2 and 5-2 lineages. The spdo and numb mutant phenotypes each displayed essentially 100% penetrance and expressivity with respect to the groups of neurons assayed. It is concluded that Notch/Numb-mediated asymmetric divisions direct the expression of dbx and eve (or dbx and hb9) to opposite siblings within multiple sib pairs, and in so doing establish the non-overlapping nature of dbx and eve, and dbx and hb9 expression in different pairs of sibling neurons (Lacin, 2009).

The transient co-expression of dbx and eve in RP2 sib and U sib interneurons raises the possibility that negative regulatory interactions between dbx and eve help maintain the non-overlapping nature of the expression profiles of these factors in distinct sets of sibling neurons. To assay for regulatory interactions between dbx, eve and hb9, a null allele of dbx was created via imprecise P element excision, and a full-length UAS-dbx transgene was also prepared. Systematic loss of function and misexpression studies uncovered cross-repressive interactions between dbx and eve, and dbx and hb9, that were largely specific to those lineages that produce sib pairs of Dbx+/Eve+ or Eve+/Hb9+ neurons (Lacin, 2009).

In embryos homozygous mutant for dbxδ48 inappropriate retention of eve expression was observed in the normally Dbx+ RP2 sib. A majority of thoracic, but a minority of abdominal (21%), segments exhibited the RP2 sib phenotype. Conversely, elav-Gal4-mediated misexpression of dbx in all postmitotic neurons was sufficient to repress eve expression in the RP2 and U motoneurons but not in aCC/pCC or the EL neurons, the other Eve+ neurons in the CNS. Thus, dbx is necessary and sufficient to repress eve in the RP2/RP2 sib pair of sibling neurons, and sufficient but not necessary to repress eve expression in U motoneurons, revealing that the ability of dbx to repress eve is restricted to those lineages that produce Eve+ and Dbx+ sibling neurons (Lacin, 2009).

In reciprocal experiments, embryos that lacked eve function specifically in the CNS displayed normal dbx expression. By contrast, eve misexpression throughout the CNS specifically repressed dbx expression in RP2 sib (and other neurons in the 4-2 lineage) and the U sib neurons (and other neurons of the 7-1 lineage); dbx expression in all other lineages was grossly normal. Thus, the ability of eve to repress dbx also appears restricted to those lineages that normally produce Dbx+ and Eve+ sibling neurons (Lacin, 2009).

Analogous tests revealed similar cross-repressive interactions between dbx and hb9. Loss of dbx or hb9 function had no effect on hb9 or dbx expression, respectively. However, dbx misexpression repressed hb9 expression in the CoR motoneurons of the 4-2 lineage as well as in the Hb9+ neurons of the 5-2 lineage. Also reduced hb9 expression was observed in other neurons; however, the lineages to which these cells belong are unknown. Conversely, generalized hb9 misexpression repressed dbx expression in the RP2 sib and CoR sibs of the 4-2 lineage as well as the Dbx+ neurons of the 5-2 and 7-1 lineages. Thus, dbx and hb9 also exhibit cross-repressive interactions that are largely restricted to those lineages that produce Dbx+ and Hb9+ sibling neurons. Together these functional studies suggest that negative regulatory interactions between dbx and eve, and dbx and hb9, help maintain the mutually exclusive expression patterns of these factors in different pairs of sibling neurons (Lacin, 2009).

Since many Dbx+ neurons extend short axons, it was asked whether dbx regulates axonal growth. Although axonal projections are grossly normal in embryos homozygous mutant for dbx, dbx misexpression in all postmitotic neurons led to a decrease in the ability of many neurons to extend axons. For example, 93% of hemisegments exhibited a significant decrease in motor axon projection into the periphery, with thinning or loss of the ISN nerve (within which U and RP2 motoneurons project axons), the SNc nerve (CoR motoneurons) and the SNb nerve (U and AC motoneurons). Within the nerve cord thin and broken longitudinal connectives were observed, at 100% penetrance, between neuromeres as well as an occasional increase in the apparent size of the axonal scaffold within neuromeres, raising the possibility that dbx misexpression induces neurons that normally project axons into the periphery or between segments to project axons locally. It is concluded that dbx expression must be strictly repressed in Hb9+ and Eve+ motoneurons for these neurons to extend axons to their appropriate targets. Note dbx misexpression is unlikely to convert motoneurons into interneurons, since expression of the vesicular-Glutamate receptor, a relatively specific motoneuron marker, is grossly normal in the face of generalized dbx expression (Lacin, 2009).

The ability of dbx misexpression to inhibit motor axon outgrowth in the ISN, SNb and SNc branches generally correlates with its ability to inhibit eve or hb9 expression in specific motoneurons that extend axons in these branches. However, loss of hb9 or eve function elicits significantly weaker motor axon phenotypes than that observed upon generalized dbx misexpression. nkx6 has been shown to promote axon outgrowth in flies, and dbx and nkx6 exhibit cross-repressive interactions in the vertebrate neural tube. Thus, the effect of dbx misexpression on nkx6 expression was assayed, and it was found that dbx was sufficient to reduce nkx6 expression in all, and to eliminate expression in some, Nkx6+ neurons, supporting the model that dbx limits axon growth at least in part by repressing nkx6. However, in contrast to vertebrates, loss of dbx function had no obvious effect on nkx6 expression, and neither loss of nkx6 function nor generalized Nkx6 misexpression grossly disrupted dbx expression (Lacin, 2009).

Homozygous mutant dbxδ48 adult flies are viable, fertile and exhibit locomotor phenotypes. For example, dbx mutant flies are sessile and uncoordinated, and perform poorly in simple climbing assays, and although they can initiate the flight response, they can not maintain flight. It is inferred that Dbx+ neurons function within neural circuits crucial for specific locomotor functions. Of note, in vertebrates (Lanuza, 2004), Dbx+ neurons are known to regulate proper left-right alternation of motoneuron firing required for proper walking movements (Lacin, 2009).

In conclusion, many Dbx+ interneurons share a sibling relationship with Eve+ or Hb9+ motoneurons, and the cellular phenotypes of these sibling neurons are highly disparate: Dbx+ interneurons extend short axons; Eve+ or Hb9+ motoneurons extend long axons. This work suggests a model for the establishment and maintenance of distinct cellular phenotypes between sibling neurons. Initially, Notch-mediated asymmetric divisions establish distinct transcription factor expression profiles in sibling neurons, demonstrated in this study by the ability of such asymmetric divisions to direct dbx expression to one neuron and eve (or hb9) expression to its sibling in multiple pairs of sibling neurons. Once sibling neurons establish distinct transcription factor expression profiles, cross-repressive interactions between these factors help maintain gene expression differences between sibling neurons, implied in this study by the lineage-specific, cross-repressive regulatory relationships observed between dbx and eve, and dbx and hb9. Such cross-repressive interactions are crucial for proper neuronal differentiation, since inappropriate dbx expression in motoneurons impairs motor-axon projections. Transcription factors, such as dbx, eve and hb9, which partake in cross-repressive interactions, also contribute more directly to neuronal differentiation via regulation of downstream effector genes. For example, eve upregulates the netrin receptor, Unc-5, in RP2 and other dorsal motoneurons, which in turn helps guide the motor-axons of these neurons to their appropriate targets (Lacin, 2009).

Genetic redundancy is a common theme between transcription factors that exhibit cross-repressive regulatory interactions during neuronal specification, and may in fact increase the robustness of this process. For example, loss of function in dbx, eve or hb9 yields subtle effects on the expression of the other two genes, whereas generalized misexpression of each gene leads to clear changes in the expression of the other two. In addition, because loss of dbx function induces inappropriate retention of eve expression in about 50% of the normally Dbx+ RP2 sib neurons, other transcription factors must partially compensate for loss of dbx within RP2. Such factors might also compensate for loss of dbx function during axon growth. Moreover, hb9 and nkx6, which are expressed in nearly identical patterns of CNS neurons, also act redundantly to repress eve in a specific set of neurons. Within the bristle lineage, the Su(H) and Sox15 transcription factors act redundantly in the socket cell to repress the expression of Shaven, a Pax-family transcription factor, thus restricting shaven expression to the sibling shaft cell. This event is crucial for socket cell differentiation, since derepression of shaven in the socket cell transforms socket cells towards shaft cells (Lacin, 2009).

This work also highlights the lineage-specific nature of regulatory relationships between transcription factors that govern neuronal specification. For example, dbx is competent to inhibit eve expression only in the two lineages that produce Dbx+ and Eve+ sib neurons. Similarly, eve can inhibit dbx expression in the same two lineages, but not in the three other lineages that produce Dbx+ but not Eve+ neurons. Similar lineage-specific, cross-repressive regulatory interactions occur between transcription factors expressed in neurons born at different times within the same lineage. The Hb9+ CoR and Eve+ RP2 motoneurons probably arise from sequentially born GMCs within the 4-2 lineage, with loss-of-function studies indicating that eve represses hb9 in RP2, consistent with hb9 repressing eve in the CoR motoneurons. Thus, cross-repressive interactions between individual transcription factors during neuronal specification often reflect lineage-specific, rather than CNS-wide, regulatory relationships (Lacin, 2009).

Notch/Numb-mediated asymmetric divisions also exhibit lineage-specific effects on dbx and eve expression. During these asymmetric divisions, high-level Notch signaling in one daughter cell induces it to adopt Notch-dependent 'A' fate, whereas the absence of Notch signaling in the other daughter permits it to adopt the 'B' fate. In the 4-2 lineage Notch signaling promotes the development of Dbx+ interneurons and inhibits the formation of the Eve+. By contrast, in the 7-1 lineage Notch signaling inhibits the formation of Dbx+ neurons and promotes the development Eve+ U motoneurons. Thus, the presence/absence of Notch signaling exerts opposite effects on dbx and eve expression in a lineage-specific manner (Lacin, 2009).

In contrast to dbx and eve, Notch signaling inhibits the formation of nearly all Hb9+ motoneurons. Notch signaling also inhibits the motoneuron fate during the asymmetric divisions that produce the RP2, aCC and three VUM motoneurons. The U motoneurons, which require Notch activity to develop, are the only exception to Notch-mediated inhibition of the motoneuron fate in flies. These observations are interesting in light of a previous model that speculated that vertebrate motoneurons share a common evolutionary ancestry with Drosophila Hb9+ motoneurons, based on the common expression of hb9, lim3 and islet in most vertebrate motoneurons and all fly motoneurons that project to ventral body wall muscles. Might Notch signaling generally inhibit the motoneuron fate in vertebrates? Recent work in zebrafish reveals that Notch inhibits the motoneuron fate during the asymmetric divisions of some pMN progenitors that yield sibling motoneuron and interneurons. However, the fraction of pMN progenitors that divide asymmetrically in this manner remains unclear. Thus, whether Notch signaling strictly inhibits the motoneuron fate during asymmetric divisions in vertebrates awaits further investigation (Lacin, 2009).

This work on dbx, eve and hb9 indicates that the transcriptional networks that control neuronal specification are organized in a lineage-specific manner. Support for this model comes from the identification of three largely lineage-specific enhancers in the eve regulatory region: one enhancer drives expression in the U neurons, one drives expression in the lineally related EL neurons and a third drives expression in the RP2 and a/pCC neurons. Might Dbx, Hb9 and the Notch pathway exert their lineage-specific effects on eve directly via the appropriate cis-regulatory module? Preliminary data support this notion, as consensus and evolutionarily conserved Dbx-binding sites (Noyes, 2008) reside in the RP2 and U motoneurons enhancers. More generally, are the regulatory regions of dbx, hb9 and other similarly functioning genes also organized in a lineage-specific manner? And, do the regulatory regions of the effector genes through which transcription factors such as Dbx, Hb9 and Eve regulate neuronal differentiation reflect a similar organization? Future work that addresses these questions should help clarify the cis-regulatory and transcriptional logic that governs neuronal specification and differentiation in Drosophila (Lacin, 2009).

In contrast to most transcription factors that govern neuronal subtype identity, dbx is not an essential gene. Adult flies that lack dbx function exhibit defects in flight and ambulatory movement. In vertebrates, interneurons derived from Dbx+ progenitors coordinate left-right alternation of motoneuron firing required for proper walking movements via direct, probably inhibitory synaptic input to motoneurons (Lanuza, 2004). The sibling relationship between Dbx+ interneurons and many motoneurons suggests that Dbx+ interneurons perform similar functions in Drosophila. Such speculation is supported by work in other insects, where small axonless interneurons modulate motoneuron function. In this light, the non-essential nature of dbx in flies may facilitate charting of the neural circuits through which Dbx+ neurons regulate distinct locomotor functions in Drosophila (Lacin, 2009).

Targets of Activity

exex mutant embryos display several ectopic Eve-positive neurons. Using the protein-positive ExexJJ154 allele, it was found that these ectopic Eve cells arise from cells that normally express Exex, suggesting that Exex represses Eve cell autonomously. The nonoverlapping expression patterns of Exex and Eve further indicate that Exex acts operationally as an Eve repressor in the CNS. To investigate whether Exex is sufficient to repress Eve, Exex was misexpressed in all postmitotic neurons, and it was found that Exex represses Eve in all Eve-positive neurons except the EL neurons. By late stage 14, only one or two weakly Eve-positive neurons remain in the positions normally occupied by the U, RP2, a/pCC, and fpCC neurons, while the cluster of Eve-positive EL interneurons appears normal. Thus, Exex expression is sufficient to repress Eve expression in all dorsally projecting MNs. The inability of Exex to repress Eve expression in the ELs suggests that the relative ability of Exex to repress Eve is controlled by factors expressed specifically in different neuronal types (Broihier, 2002).

During analysis of Lim3 expression in exex mutant embryos, the presence of a group of ectopic Lim3-positive neurons was observed. Since all Exex-positive neurons normally coexpress Lim3, the presence of ectopic Lim3-positive neurons suggests a cell-nonautonomous effect of Exex on the regulation of Lim3. Surprisingly, double label experiments identify the ectopic Lim3-positive neurons as the six Eve-positive U MNs. This phenotype was visualized using a lim3-tau-myc transgene due to the perdurance of transgene expression in U MNs relative to the more transient expression of endogenous Lim3 in these cells. The transient nature of Lim3 expression in the U MNs is attributed to the ability of Eve to repress Lim3(Broihier, 2002).

The ectopic expression of Lim3 in the U MNs in exex mutants is exciting because neither the U MNs nor their progenitors ever express Exex. These data further support the model that Exex acts cell nonautonomously to repress Lim3 expression in the U MNs. Consistent with this, several groups of Exex-positive neurons surround the U MNs during their development. One or more of these groups of Exex-positive neurons likely serves as the source of the signal received by the U MNs. Taken together, these results uncover a novel role for intercellular signaling in the establishment of neuronal fate in Drosophila (Broihier, 2002).

The homeodomain protein Nkx6 is a key member of the genetic network of transcription factors that specifies neuronal fates in Drosophila. Nkx6 collaborates with the homeodomain protein Hb9/ExEx to specify ventrally projecting motoneuron fate and to repress dorsally projecting motoneuron fate. While Nkx6 acts in parallel with hb9 to regulate motoneuron fate, Nkx6 plays a distinct role to promote axonogenesis; axon growth of Nkx6-positive motoneurons is severely compromised in Nkx6 mutant embryos. Furthermore, Nkx6 is necessary for the expression of the neural adhesion molecule Fasciclin III in Nkx6-positive motoneurons. Thus, this work demonstrates that Nkx6 acts in a specific neuronal population to link neuronal subtype identity to neuronal morphology and connectivity (Broihier, 2004).

In many model systems, MNs that extend axons along common trajectories express similar sets of transcriptional regulators, which in turn regulate key aspects of the differentiation of these MN subtypes. Drosophila MNs are classified by the location of the body wall muscles they innervate. MNs that innervate dorsal body wall muscles in Drosophila express the homeodomain (HD) transcription factor Even-skipped (Eve). Furthermore, genetic analyses indicate that Eve is a key determinant of the fate of dorsally projecting MNs. Eve engages in a cross-repressive interaction with the HD protein Hb9, a determinant of ventrally projecting MNs (Broihier, 2004 and references therein).

Ventrally projecting MNs also express the HD transcription factors Lim3 and Islet. Functional analyses have demonstrated that these three HD factors are required for proper axon guidance of ventrally projecting MNs. The genetic hierarchy governing the fate of ventrally projecting neurons has, however, remained elusive as Lim3, Islet, and Hb9 are expressed independently of each other (Broihier, 2004 and references therein).

To explore the genetic networks behind neuronal diversification in Drosophila, the role of the Drosophila Nkx6 homolog in regulating distinct MN fates was investigated. Genetic interactions were characterized between Nkx6 and factors essential for neuronal fate acquisition. Evidence that Nkx6 collaborates with hb9 (exex – FlyBase) to regulate the fate of distinct neuronal populations. This analysis of hb9 Nkx6 double mutant embryos indicates that ventrally projecting MNs fail to develop properly in these embryos, while expression of eve, a key determinant of dorsally projecting MN identity, expands. In addition, Nkx6 promotes axonogenesis of Nkx6-positive neurons. Consistent with a direct regulatory role in this process, Nkx6 activates the expression of the neural adhesion molecule Fasciclin III in ventrally projecting motoneurons. These data suggest that Nkx6 is a primary transcriptional regulator of molecules essential for axon growth and guidance in a specific neuronal population (Broihier, 2004).

The findings that Nkx6 has roles in both the specification and differentiation of ventrally projecting MNs places Nkx6 in the regulatory circuit that specifies distinct postmitotic neuron fates in the Drosophila CNS. In the mouse, Nkx6 protein function in MN progenitors regulates Hb9 expression in postmitotic MNs. Drosophila Nkx6 is expressed in neural precursors and postmitotic neurons while Hb9 expression is nearly exclusive to postmitotic neurons. However, in contrast to the linear relationship of Nkx6.1/2 and Hb9 in vertebrates, Nkx6 and hb9 were found to act in parallel to specify neuronal fate in Drosophila. Nkx6 and hb9 act in concert both to repress expression of the dorsal MN determinant Eve and to promote expression of Lim3 and Islet in ventrally projecting RP MNs. It will be of interest to extend this genetic analysis to other groups of ventrally projecting MNs. It will be also be important to examine the directness of these genetic interactions. Both Nkx6 and hb9 contain conserved TN domains that in vertebrate HD proteins have been shown to interact with the Groucho co-repressor, suggesting that Nkx6 and hb9 function as transcriptional repressors. This raises the possibility that Nkx6 and hb9 bind to sequences in the eve enhancer and directly repress its transcription. In addition, Nkx6 and hb9 activate lim3/islet gene expression within ventrally projecting MNs, raising the possibility that they do so by repressing an unidentified repressor of ventrally projecting MN identity (Broihier, 2004).

eve represents an appealing candidate for the unidentified repressor in this model. Ectopic Eve expression in RP MNs in hb9 Nkx6 double mutants may repress Lim3 and Islet. Consistent with this, though it was not possible to unambiguously identify the ectopic Eve neurons in hb9 Nkx6 mutants, many of them are situated close to the midline, suggesting they may represent mis-specified RP MNs. Furthermore, pan-neuronal eve expression represses Lim3 and Islet expression in the RP MNs demonstrating that Eve can repress Lim3 and Islet (Landgraf, 1999). A direct test of this model will require resolving the identity of the ectopic Eve neurons in hb9 Nkx6 mutant embryos (Broihier, 2004).

The Homeodomain Transcription Factor Hb9 Controls Axon Guidance in Drosophila through the Regulation of Robo Receptors

Transcription factors establish neural diversity and wiring specificity; however, how they orchestrate changes in cell morphology remains poorly understood. The Drosophila Roundabout (Robo) receptors regulate connectivity in the CNS, but how their precise expression domains are established is unknown. This study shows that the homeodomain transcription factor Hb9 acts upstream of Robo2 and Robo3 to regulate axon guidance in the Drosophila embryo. In ventrally projecting motor neurons, hb9 is required for robo2 expression, and restoring Robo2 activity in hb9 mutants rescues motor axon defects. Hb9 requires its conserved repressor domain and functions in parallel with Nkx6 to regulate robo2. Moreover, hb9 can regulate the mediolateral position of axons through robo2 and robo3, and restoring robo3 expression in hb9 mutants rescues the lateral position defects of a subset of neurons. Altogether, these data identify Robo2 and Robo3 as key effectors of Hb9 in regulating nervous system development (Santiago, 2014).

Combinations of transcription factors specify the tremendous diversity of cell types in the nervous system. Many studies have identified requirements for transcription factors in regulating specific events in circuit formation as neurons migrate, form dendritic and axonal extensions, and select their final synaptic targets. In most cases, the downstream effectors through which transcription factors control changes in neuronal morphology and connectivity remain unknown, although several functional relationships have been demonstrated (Santiago, 2014).

Conserved homeodomain transcription factors regulate motor neuron development across phyla. Studies in vertebrates and invertebrates have shown that motor neurons that project to common target areas often express common sets of transcription factors, which act instructively to direct motor axon guidance. In mouse and chick, Nkx6.1/ Nkx6.2 and MNR2/Hb9 are required for the specification of spinal cord motor neurons, and for axon pathfinding and muscle targeting in specific motor nerves. In Drosophila, Nkx6 and Hb9 are expressed in embryonic motor neurons that project to ventral or lateral body wall muscles, and although they are not individually required for specification, they are essential for the pathfinding of ventrally projecting motor axons. Axons that project to dorsal muscles express the homeodomain transcription factor Even-skipped (Eve), which regulates guidance in part through the Netrin receptor Unc5. Eve exhibits cross-repressive interactions with hb9 and nkx6, which function in parallel to repress eve and promote islet and lim3. Hb9 and Nkx6 act as repressors to regulate transcription factors in the spinal cord; however, guidance receptors that act downstream of Hb9 and Nkx6 have not been characterized. Interestingly, in both flies and vertebrates, Hb9 and Nkx6 are also expressed in a subset of interneurons, and knockdown experiments in Drosophila have suggested a role for hb9 in regulating midline crossing (Santiago, 2014).

Roundabout (Robo) receptors regulate midline crossing and lateral position within the developing CNS of invertebrates and vertebrates. Two recent studies in mice have also identified a role for Robos in regulating motor axon guidance in specific motor neuron populations. The three Drosophila Robo receptors have diversified in their expression patterns and functions. Robo2 is initially expressed in many ipsilateral pioneers and also contributes to Slit-mediated repulsion. Subsequently, robo2 expression is more restricted, and it is required to specify the medio-lateral position of axons. Robo3 is expressed in a subset of CNS neurons and also regulates lateral position (Santiago, 2014).

Characterization of the expression domains of the Drosophila Robos revealed an intriguing pattern, in which Robo1 is expressed on axons throughout the width of the CNS, Robo3 is found on axons in intermediate and lateral zones, and Robo2 is enriched on the most lateral axons. These patterns are transcriptional in origin, as replacing any robo gene with the coding sequence of another Robo receptor results in a protein distribution that matches the endogenous expression of the replaced gene (Spitzweck, 2010). A phenotypic analysis of these gene-swap alleles revealed the importance of transcriptional regulation for the diversification of robo gene function (Spitzweck, 2010). Robo2 and robo3's roles in regulating lateral position are largely dependent on their expression patterns, although unique structures within the Robo2 receptor are also important for its function in lateral position (Evans, 2010; Spitzweck, 2010). In the peripheral nervous system, the Atonal transcription factor regulates robo3 in chordotonal sensory neurons, directing the position of their axon terminals. In the CNS, the transcription factors lola and midline contribute to the induction of robo1. However, how the expression patterns of robo2 and robo3 are established to direct axons to specific medio-lateral zones within the CNS remains unknown (Santiago, 2014).

This study identifies a functional relationship between Hb9 and the Robo2 and Robo3 receptors in multiple contexts. Hb9 acts through Robo2 to regulate motor axon guidance and can direct the medio-lateral position of axons in the nerve cord through its effects on robo2 and robo3. Furthermore, hb9 interacts genetically with nkx6 and requires its conserved repressor domain to regulate robo2. Together, these data establish a link between transcriptional regulators and cell surface guidance receptors, providing an example of how upstream factors act through specific guidance receptors to direct circuit formation (Santiago, 2014).

This study has demonstrated a functional relationship between Hb9 and the Robo2 and Robo3 receptors in multiple contexts in the Drosophila embryo. In the RP motor neurons, hb9 is required for robo2 expression, and genetic rescue experiments indicate that robo2 acts downstream of hb9. Hb9 requires its conserved repressor domain and acts in parallel with Nkx6 to regulate robo2 and motor axon guidance. Moreover, hb9 contributes to the endogenous expression patterns of robo2 and robo3 and the lateral position of a subset of axons in the CNS, and can redirect axons laterally when overexpressed via upregulation of robo2. Finally, restoring Robo3 rescues the medial shift of MP1 axons in hb9 mutants, indicating that hb9 acts through robo3 to regulate medio-lateral position in a defined subset of neurons (Santiago, 2014).

Hb9 and nkx6 are required for the expression of robo2 in motor neurons, and rescue experiments suggest that the loss of robo2 contributes to the phenotype of hb9 mutants. However, nkx6 mutants and hb9 mutants heterozygous for nkx6 have a stronger ISNb phenotype than robo2 mutants, implying the existence of additional downstream targets. One candidate is the cell adhesion molecule FasIII, which is normally expressed in the RP motor neurons and appears reduced in nkx6 mutant embryos. Identifying the constellation of effectors that function downstream of Hb9 and Nkx6 will be key to understanding how transcription factors expressed in specific neurons work together to drive the expression of the cell surface receptors that regulate axon guidance and target selection (Santiago, 2014).

Robo2's activity in motor axon guidance appears distinct from the previously described activities of the Drosophila Robo receptors. Although Robo1 can replace Robo2's repulsive activity at the midline (Spitzweck, 2010), Robo2's function in motor axon guidance is not shared by either Robo1 or Robo3. Moreover, Robo2's antirepulsive activity at the midline and its ability to shift axons laterally when overexpressed both map to Robo2's ectodomain, whereas this study has found that Robo2's activity in motor axon guidance maps to its cytodomain (Evans, 2010; Spitzweck, 2010). The signaling outputs of Robo2's cytodomain remain unknown, as it lacks the conserved motifs within Robo1 that engage downstream signaling partners. How does Robo2 function during motor axon guidance? In mice, Robo receptors are expressed in spinal motor neurons and prevent the defasciculation of a subset of motor axons (Jaworski, 2012). Does Drosophila Robo2 regulate motor axon fasciculation? The levels of adhesion between ISNb axons and other nerves must be precisely controlled during the different stages of motor axon growth and target selection, and several regulators of adhesion are required for ISNb guidance. Furthermore, whereas Slit can be detected on ventral muscles, it is not visibly enriched in a pattern that suggests directionality in guiding motor axons, making it difficult to envision how Robo2-mediated repulsive or attractive signaling might contribute to ISNb pathfinding. Future work will determine how Robo2's cytodomain mediates motor axon guidance, whether this activity is Slit dependent, and whether Robo2 signals attraction, repulsion, or modulates adhesion in Drosophila motor axons (Santiago, 2014).

Elegant gene-swap experiments revealed the importance of transcriptional regulation in establishing the different expression patterns and functions of the Drosophila Robo receptors (Spitzweck, 2010). By analyzing a previously uncharacterized subset of axon pathways, this study has uncovered a requirement for Hb9 in regulating lateral position in the CNS. Although Hb9 can act instructively to direct lateral position when overexpressed, its endogenous expression in a subset of medially projecting neurons suggests that its ability to shift axons laterally is context dependent. A complex picture emerges in which multiple factors act in different groups of neurons to regulate robo2 and robo3. In a subset of interneurons, hb9 is endogenously required for lateral position through the upregulation of robo3 and likely robo2. In other neurons, such as those that form the outer FasII tracts, the expression patterns of robo2 and robo3 rely on additional upstream factors. What might be the significance of a regulatory network in which multiple sets of transcription factors direct lateral position in different groups of neurons? One possibility is that hb9-expressing neurons may share specific functional properties, such as the expression of particular neurotransmitters or ion channels. Alternatively, hb9 may regulate other aspects of connectivity. Indeed, Robo receptors mediate dendritic targeting in the Drosophila CNS, raising the exciting possibility that hb9 regulates both axonal and dendritic guidance through its effects on guidance receptor expression (Santiago, 2014).

What is the mechanism by which Hb9 regulates the expression of robo2, robo3, and its other downstream effectors? This study has found that Hb9 requires its conserved putative repressor domain and acts in parallel with Nkx6 to regulate robo2 and motor axon guidance. It has previously been shown that hb9 and nkx6 function in parallel to regulate several transcription factors. Hb9, nkx6 double mutants show decreased expression of islet and lim3 and upregulation of eve and the Nkx2 ortholog vnd. Are Hb9 and Nkx6 regulating robo2 or robo3 through any of their previously identified targets? Hb9 and nkx6 single mutants show no change in islet, lim3, or vnd expression, arguing that hb9 and nkx6 do not act solely through these factors to regulate robo2 or robo3. Eve expression is unaffected in nkx6 mutants, and whereas it is ectopically expressed in two neurons per hemisegment in hb9 mutants, these do not correspond to RP3 or MP1, the identifiable cells in which changes can be detected in robo2 and robo3. Therefore, the data do not support the hypothesis that Hb9 and Nkx6 regulate robo2 or robo3 primarily through their previously identified targets islet, lim3, vnd, or eve (Santiago, 2014).

Gain-of-function experiments in vertebrates suggest that Hb9 and Nkx6 act as repressors to regulate gene expression in the spinal. The finding that Hb9's Eh domain is required for motor axon pathfinding and robo2 regulation suggests that Hb9 acts as a repressor in this context as well, most likely through a previously unidentified intermediate target. In contrast, the Eh domain is not required for Hb9's ability to regulate robo3 or lateral position in hb9GAL4+ neurons that project to intermediate zones of the CNS. The finding that Hb9;delta;Eh retains significant activity in rescuing lateral position and robo3 expression indicates that Hb9 may regulate robo2 and robo3 via distinct mechanisms, perhaps involving different transcriptional cofactors or intermediate targets. In support of this hypothesis, hb9 overexpression in the ap neurons can induce robo2, but not robo3. These data raise the intriguing possibility that Hb9's ability to regulate robo2 and robo3 via different mechanisms contributed to the diversification of their expression patterns in the CNS. Determining how Hb9 and Nkx6 regulate their effectors will be key to achieving a complete understanding of how these conserved transcription factors control changes in cell morphology and axon pathfinding during development. Of note, Hb9 mutant mice exhibit defects in a subset of motor nerves, including the phrenic and intercostal nerves, which are also affected in Robo mutants. It will be of great interest to determine if despite the vast divergence in the evolution of nervous system development between invertebrates and vertebrates, Hb9 or Nkx6 has retained a role for regulating Robo receptors across species (Santiago, 2014).



To investigate the role of Exex during CNS development, Exex-specific antibodies were raised. Embryonic expression of Exex initiates in the posterior midgut primordium at stage 7. By stage 9, Exex protein is present in the primordia of the anterior and posterior midgut and persists in anterior and posterior regions of the endoderm throughout embryogenesis. In the CNS, Exex protein expression is first detected during stage 11 in one-to-two mitotic GMCs and approximately 15 neurons per hemisegment. Exex expression in the CNS peaks at stage 14, when it is strongly expressed in approximately 30 neurons per hemisegment, including the well-characterized RP1, RP3-5 MNs, and dMP2 and MP1 interneurons. Thus, in the CNS, Exex expression is expressed almost exclusively in a distinct population of postmitotic MNs and interneurons, consistent with Exex regulating neuronal identity (Broihier, 2002).

A key distinguishing trait of neurons is their axonal trajectory. Thus, the axonal trajectories of Exex-positive neurons were traced to investigate whether Exex identifies specific subpopulations of CNS neurons. To create an Exex-dependent axonal marker, targeted transposition was employed to convert a exexLacZ enhancer trap to a exexGal4 enhancer trap. The exexGal4 driver was used to express GAP-GFP and it was confirmed that GFP expression faithfully recapitulates the Exex expression pattern with the exception that the peripheral LBD neuron expresses ExexGal4 but not Exex. Within the CNS, it was found that Exex-positive interneurons project axons in three distinct longitudinal fascicles (Broihier, 2002).

The trajectory of Exex-positive MNs was traced into the periphery and it was found that Exex-positive neurons populate five of the six motor axon branches. In Drosophila, motor axons exit the CNS in the ISN, SN, and transverse nerve (TN). The main branch of the ISN innervates the dorsal and lateral body wall musculature. Axons in two branches of the ISN (ISNb and ISNd) defasciculate from the ISN to innervate distinct groups of ventral body wall muscles. Similarly, the primary branch of the SN, SNa, innervates a lateral muscle group, and axons in its minor branch, SNc, extend along SNa until their choice point where they defasciculate and innervate ventral muscles. Exex-positive motor axons were found to extend in the ISN, ISNb, ISNd, Sna, and SNc nerves. While Exex-positive motor axons are observed in the ISN, they do not project to the most dorsal muscle regions. In addition, no Exex-positive axons were detected in the TN. These data demonstrate that Exex-positive axons populate five of the six major nerve branches and that Exex is expressed in the majority of ventrally and laterally projecting MNs. Interestingly, dorsally projecting MNs express Eve, but apparently not Exex, suggesting that Exex and Eve identify distinct populations of MNs (Broihier, 2002).

Expression of HB9 mRNA and protein is indistinguishable during embryogenesis and is restricted to the gut and CNS. Endoderm expression is conserved from sea urchins to vertebrates, and neuronal expression is conserved from amphioxus to vertebrates. Drosophila HB9 CNS expression begins during late stage 10 in ~12 cells per hemisegment and increases to ~30 cells per hemisegment at stage 15 and later. None of these cells express the glial marker Repo, indicating that all HB9 cells are neurons or GMCs. The timing of expression of HB9 and the size of cells expressing HB9 indicate that cells are likely to be postmitotic. To confirm this, a double-labeling of HB9 and phosphohistone H3, a marker for cells undergoing mitosis, was performed. One or two cells were found in stereotyped mediolateral positions that express phosphohistone H3 and express HB9 weakly; it is concluded that HB9 can be expressed just before GMC division but is usually restricted to postmitotic neurons. HB9 is also detected in the third instar larval brain (Odden, 2002).

On the basis of vertebrate studies, it was expected that HB9 would be expressed specifically in motoneurons. Thus, embryos were double-labeled for HB9 and various motoneuron markers, including Eve (labels all dorsally projecting motoneurons), Islet-Tau:myc (labels all ventrally projecting motoneurons), and Eagle (labels a subset of ventrally projecting motoneurons). HB9 was found to be coexpressed widely with Islet in most ventrally projecting neurons, including the well characterized RP1, RP3, RP4, and RP5 motoneurons that project via segmental nerve B (SNb) and segmental nerve D (SNd) to ventral muscles 6, 7, 12, and 13. To help distinguish exactly which Islet+ motoneurons are HB9+, a lim3:lacZ transgene was used to identify all Islet-positive motoneurons that project via SNb. HB9 is coexpressed with Lim3 and Islet in the four RP motoneurons, two lateral SNb-projecting motoneurons, and a single ventral neuron that has not been described previously. HB9 is not expressed in all ventrally projecting motoneurons, however, because it is not detected in the Eagle+ GW motoneuron that projects to muscle 15 or the Eagle+ motoneuron derived from neuroblast 2-4 that projects via segmental nerve a (SNa) to muscle 8. In addition, HB9 is not expressed in any of the Eve+ motoneurons that innervate dorsal muscle targets. Thus, HB9 is not expressed in all motoneurons but rather is restricted to a small subset of motoneurons that includes those projecting via SNb and SNd to ventral muscle targets (Odden, 2002).

To determine whether HB9 expression is limited to motoneurons like its counterparts in vertebrates, expression was examined in several identified interneurons. The best-characterized interneurons in the CNS are the serotonergic interneurons derived from neuroblast 7-3, for which there are numerous markers. Using several of these markers in combination with HB9, it was found that HB9 is expressed in the two serotonergic interneurons (EW1/EW2) and a third lineally related interneuron (EW3); each of these interneurons projects contralaterally across the posterior commissure. In addition, FasII, a transmembrane protein that labels all motoneurons and a few interneurons, was assayed, reasoning that any HB9+, FasII- cells would be interneurons. Only a ventral cluster of six HB9+ cells fails to express FasII, indicating that these HB9+ cells are interneurons. In summary, HB9 is expressed in at least nine interneurons (three Islet+ Lim3- Eagle+ EW neurons and a ventral cluster of six Islet- Lim3- Eagle- FasII- neurons) (Odden, 2002).

Postmitotic specification of Drosophila insulinergic neurons from pioneer neurons

Insulin and related peptides play important and conserved functions in growth and metabolism. Although Drosophila has proved useful for the genetic analysis of insulin functions, little is known about the transcription factors and cell lineages involved in insulin production. Within the embryonic central nervous system, the MP2 neuroblast divides once to generate a dMP2 neuron that initially functions as a pioneer, guiding the axons of other later-born embryonic neurons. Later during development, dMP2 neurons in anterior segments undergo apoptosis but their posterior counterparts persist. Surviving posterior dMP2 neurons no longer function in axonal scaffolding but differentiate into neuroendocrine cells that express insulin-like peptide 7 (Ilp7) and innervate the hindgut. The find that the postmitotic transition from pioneer to insulin-producing neuron is a multistep process requiring retrograde bone morphogenetic protein (BMP) signalling and four transcription factors: Abdominal-B, Hb9, Forkhead, and Dimmed. These five inputs contribute in a partially overlapping manner to combinatorial codes for dMP2 apoptosis, survival, and insulinergic differentiation. Ectopic reconstitution of this code is sufficient to activate Ilp7 expression in other postmitotic neurons. These studies reveal striking similarities between the transcription factors regulating insulin expression in insect neurons and mammalian pancreatic beta-cells (Miguel-Aliaga, 2008).

The observed death of some Drosophila pioneer neurons has been used to argue that their function is transient, but persistence in other cases suggested that, either they continue to play an axonal-scaffolding role, or that they adopt some other identity. The current findings resolve this long-standing issue by clearly demonstrating that, for dMP2 neurons, the axonal scaffolding function is only transient. After this role is no longer required, surviving dMP2 neurons become insulinergic and innervate the hindgut. The other known innervation of the Drosophila gut occurs much more anteriorly, in the foregut and anterior midgut, from neuronal cell bodies located in the peripheral ganglia of the stomatogastric nervous system. Unlike dMP2 neurons, however, the individual identities of the stomatogastric neurons and their cell lineages remain to be clearly defined. Thus, dMP2 neurons may provide a simple and well-characterised system for studies of the guidance cues involved in enteric innervation. Future studies, however, will be needed to determine the functions of Ilp7 in dMP2 neurons. It will be important to distinguish if this posterior neural source of insulin acts humorally to promote growth, like the more anterior brain mNSCs, or if it has more local effects in abdominal tissues. In this regard, the presence of Ilp7-expressing neurites in close proximity to the Ilp2-producing mNSCs is intriguing (Miguel-Aliaga, 2008).

The transition from pioneer to neuroendocrine neuron is not unique to dMP2 neurons, as Drosophila MP1 pioneer neurons also become neuropeptidergic at larval stages (Wheeler, 2006). In the grasshopper, segment-specific survival of pioneer neurons has also been reported, raising the possibility that they too may become neuroendocrine. Studies in other species, including vertebrates, will be needed to reveal the extent to which the linkage between pioneer and neuroendocrine functions is conserved. Identifying pioneer neurons with an 'ancestral' neuroendocrine identity in other phyla would lend further support to the proposal that pioneer neurons are highly conserved in evolution (Miguel-Aliaga, 2008).

Apoptosis of postmitotic neurons is a widespread feature of normal VNC development, but few developmental regulators of core pro-apoptotic genes such as grim, hid, and rpr have been identified. This study uncovers roles for Fkh and Hb9. Hb9, at least, appears linked to cell death in neurons other than dMP2: in Df(3L)H99 mutant embryos, where apoptosis is blocked, ectopic Hb9-positive RP motor neurons are observed in segments A7-A8. Hb9 is an important regulator of motor neuron identity in both Drosophila and vertebrates. Finding of a pro-apoptotic function for Hb9 in Drosophila, together with the neurotrophic requirement for motor neuron survival in vertebrates, raises the possibility that the same genetic programs specifying the identities of motor neurons also sensitize them for postmitotic editing via apoptosis (Miguel-Aliaga, 2008).

Fkh function in CNS development has not been characterized. Fkh is expressed in segmentally repeated clusters of midline neurons, including dMP2, vMP2, MP1 neurons, and the VUM interneurons. Within the MP2 lineage, Fkh is first expressed in the MP2 neuroblast at stage 9-10 and continues to be expressed in both the dMP2 and vMP2 daughters throughout embryonic and larval stages. In fkh mutants, 95% of anterior dMP2 neurons fail to undergo apoptosis, and 95.3% of posterior dMP2 neurons (and 100% of ectopic anterior counterparts) fail to express Ilp7. Both of these dramatic phenotypes could be rescued to near wild-type levels by reintroducing Fkh under odd-GAL4 regulation, indicating a cell-autonomous requirement for promoting dMP2 apoptosis and Ilp7 expression (Miguel-Aliaga, 2008).

Hb9 and Fkh expression in many neurons that do not die suggests a combinatorial mechanism for the control of developmental apoptosis. One possibility is that several transcription factors function in combination to activate the core pro-apoptotic genes. Given the proposed role for Foxa proteins in chromatin accessibility, Fkh expression in dMP2 neurons may render the promoters of core pro-apoptotic genes responsive to activation by Hb9. An alternative but not mutually exclusive mechanism involves individual transcription factors activating different pro-apoptotic genes such that a combination of these would then be required to trigger neuronal death. For example, Hb9 could be required for rpr/skl but not grim expression. Some support for this idea comes from the observation that loss of hb9 activity blocks rpr/skl-mediated death of dMP2 neurons but not the largely grim-dependent apoptosis of anterior MP1 neurons (Miguel-Aliaga, 2008).

An important conclusion from this study is that the combinatorial transcription factor code controlling apoptosis partially overlaps with that regulating insulinergic identity. Thus, Fkh and Hb9 are both essential components of the codes for anterior apoptosis and also Ilp7 expression, illustrating that these transcription factors play surprising dual roles as pro-apoptotic and pro-differentiation factors within the same neuronal subtype. Importantly, the results also show that the segment-specific Hox protein Abd-B acts as a postmitotic switch, converting the pro-apoptotic Fkh+ Hb9+ code into an insulinergic Fkh+ Hb9+ Abd-B+ code (Miguel-Aliaga, 2008).

Three Ilp7 regulators (Hb9, Abd-B, and Fkh) are expressed at least 12 h before Ilp7 is first activated: from the time when the MP2 neuroblast exits the cell cycle. In the case of Hb9, it was not possible to uncouple two temporally separable functions. Early postmitotic expression of Hb9 is important for its death-activating function, whereas later expression suffices for activating Ilp7. Similarly, the Hox protein Abd-B generates a segment-specific neuropeptide pattern via postmitotic regulation of posterior dMP2 survival and also Ilp7 activation. As vertebrate neuropeptides are also expressed in restricted neuronal populations within specific rostrocaudal domains, they may be similarly regulated by Hox survival/neuroendocrine inputs. In the case of Fkh, it is required for many different aspects of the progression from the early to the late postmitotic dMP2 fate. Fkh expression is restricted to VNC midline neurons and its vertebrate orthologue Foxa2 functions in the differentiation of the floor plate and ventral dopaminergic and serotonergic neurons (Ferri, 2007; Jacob, 2007; Norton, 2005). Thus, in both the Drosophila midline and its vertebrate counterpart, the floor plate, Fkh proteins play a conserved role in the differentiation of ventral neuronal subtypes (Miguel-Aliaga, 2008).

The other two dMP2 regulators identified in this study, Dimm and the BMP pathway, are switched on shortly before the onset of Ilp7 expression. The timing of onset of these two broad neuroendocrine regulators is likely to specify when Ilp7 is first activated, whereas the earlier factors Fkh, Hb9, and Abd-B may contribute more specifically to insulinergic identity. Together, the genetic and expression analyses in this study demonstrate that the combinatorial code of genetic inputs required for Ilp7 expression is assembled in a step-wise manner during postmitotic maturation. Importantly, this allows a subset of the components to be shared (such as Fkh and Hb9) between sequential neuronal programmes (survival and Ilp7 expression) without losing output specificity (Miguel-Aliaga, 2008).

Two observations from this study indicate that insulinergic combinatorial codes can vary from cell-to-cell and also from one Ilp to another. (1) None of the regulators of Ilp7 in dMP2 neurons appear to regulate it in DP neurons. (2) The dMP2 insulinergic code is sufficient to trigger ectopic expression of Ilp7 but not Ilp2 or other neuropeptides such as FMRFa. These findings suggest the existence of additional, as yet unidentified, insulinergic factors in DP neurons and also in the brain mNSCs where Ilp2 is expressed. Identification of the neural progenitor for these mNSCs (Wang, 2007) should facilitate characterization of the Ilp1/Ilp2/Ilp3/Ilp5 combinatorial codes and thus clarify the extent to which different insulinergic transcriptional programmes overlap (Miguel-Aliaga, 2008).

The finding that an Ilp7-expressing neuron derives from the MP2 lineage reveals that at least some insulinergic regulators are similar in insects and mammals. Three apparent similarities may not be very insulin-specific but reflect more general processes shared by neural and endocrine programmes in many species. (1) Notch signalling singles out the MP2 neuroblast and distinguishes its two progeny neurons, while in mammals, it limits pancreatic expression of the 'proneural' gene Ngn3 to prospective endocrine cells. (2) The survival and pro-Ilp7 functions mediated by Abd-B in the dMP2 neuron could also have their postmitotic counterparts in ß-cells, either mediated by related Hox genes or via another homeobox gene, Pdx-1, following its early input into pancreatic induction. (3) Nerfin-1 is required for dMP2 pioneer function (Kuzin, 2005), while its mammalian orthologue Insm1/IA1 is important for pancreatic ß-cell specification (Miguel-Aliaga, 2008).

Several more specific regulatory similarities exist between the insulinergic differentiation factors active in postmitotic dMP2 neurons. For example, the role of fkh in dMP2 neurosecretory differentiation described in this study is similar to the functions of HNF3b/Foxa2 in islet maturation and insulin secretion (Sund, 2001). In addition, mammalian Nkx2.2 is important for pancreatic ß-cell specification and is known to activate transcription of the insulin regulator Nkx6.1: an important late event in ß-cell differentiation. Intriguingly, the Drosophila orthologue of Nkx2.2, Vnd, is required for dMP2 formation. Drosophila Nkx6.1, the orthologue of mammalian Nkx6 (FlyBase name HGTX), is expressed by postmitotic dMP2 neurons, and it will be interesting to determine whether it too functions downstream of Vnd during Ilp7 regulation. Most strikingly, mammalian equivalents of two of the insulinergic inputs identified in this study, Hb9 and BMP signalling, are also required for several aspects of late ß-cell differentiation including the expression of Nkx6.1 and insulin. Together, these insect-mammalian comparisons provide evidence that, although the cell types involved look very different, some of the genetic circuitry regulating insulin is conserved between arthropods and chordates. This suggests that the power of fly genetics can now be harnessed to identify additional mammalian regulators of neuroendocrine cell fates and insulin expression (Miguel-Aliaga, 2008).


The ability of exex to regulate neuronal fate by repressing eve places exex within the genetic regulatory network that governs neuronal fate. To begin to illuminate the role exex plays in this network, exex was characterized at the molecular level. Standard meiotic mapping positioned exex between ru and h on the genetic map, and deficiency analysis localized exex to cytological position 66B1-2. The subsequent completion of sequencing of the Drosophila genome facilitated a candidate gene approach to identify exex. Predicted genes in the region were screened for a CNS expression pattern by RNA in situ hybridization, and one gene, CG8254, expressed in the embryonic CNS, was identified. To determine if this gene corresponds to exex, the CG8254 coding region was sequenced from larvae homozygous for each exex allele. It was found that each exex allele contains a distinct nonsense mutation in the CG8254 coding region. These data and the finding that exexKK30 homozygous mutant embryos fail to produce detectable Exex protein demonstrate that the exex locus corresponds to CG8254 (Broihier, 2002).

The identical phenotypes of ExexKK30 homozygous and ExexKK30/Df(pblNR) transheterozygous embryos identify ExexKK30 as a null allele. Interestingly, Exex protein is present at wild-type levels in embryos homozygous for ExexJJ154, an allele predicted to encode the entire protein except the C-terminal 32 amino acids. Since ExexJJ154 embryos exhibit similar, albeit more severe, CNS phenotypes than ExexKK30embryos, the ExexJJ154 allele likely has dominant-negative activity (Broihier, 2002).

The widespread expression of Exex in MNs suggested that exex regulates MN differentiation. To address this, MAb 1D4 against Fasciclin II was used to visualize MN projections in embryos mutant for the null allele, ExexKK30. The overall organization of motor axon projections is normal, and pathfinding aberrations were not detected in either SN branch or in the ISN or ISNd. However, the ISNb branch exhibits two predominant phenotypes both resulting in a lack of innervation of the ventral muscle field. In 41% of hemisegments, the ISNb defasciculates from the ISN and enters the ventral musculature, where the axons stall and growth cones accumulate. In 19% of hemisegments, the ISNb fails to defasciculate from the ISN and extends dorsally with the ISN. Since Exex is expressed in the ISNb-projecting RP MNs, the aberrant pathfinding of ISNb in exex mutants suggests that exex promotes the differentiation of these neurons (Broihier, 2002).

Loss-of-function analysis indicates that exex is necessary for the proper axonal trajectories of a subset of ventrally projecting MNs. To test whether Exex misexpression is sufficient to reroute motor axons, Exex was misexpressed via the UAS/Gal4 system. Embryos in which Exex is misexpressed in all postmitotic neurons via the elavGal4 driver display highly penetrant axonal phenotypes. In these embryos, all motor axons fuse with the ISN prior to exiting the CNS. Thus, only a single nerve branch, a thickened ISN, forms in these embryos. The thickness of the ISN decreases dramatically in the lateral muscle region, suggesting that most axons acquire a laterally projecting ISN identity. Consistent with this, the ISN terminates prematurely in the dorsal body wall and often branches aberrantly in this region. The defects in dorsal MN projections likely arise as a result of the ability of Exex misexpression to abolish Eve in dorsally projecting MNs. It is concluded that Exex misexpression forces MNs to acquire an ISN-projecting identity and preferentially induces these MNs to project to the lateral body wall region. In combination with the loss-of-function analysis, these data demonstrate that proper levels of Exex activity are required to direct the normal pattern of motor axon outgrowth (Broihier, 2002).

The ISNb MN phenotypes of Exex exhibit similarity to those of Lim3 and Islet. Lim3 and Islet are two LIM-HD proteins that are required for the development of ISNb-projecting axons (Thor, 1997; Thor, 1999). As noted, ISNb-MNs express Exex and require Exex function for their differentiation, suggesting that Exex might interact with Lim3 and Islet to regulate neuronal fate. To investigate this, the relative expression patterns and genetic interactions between these genes were examined. To this end, Lim3- and Islet-specific antibodies were generated because prior expression analyses of Lim3 and Islet used gene-specific reporter constructs (Thor, 1997; Thor, 1999) and such reporter constructs often identify only a subset of a gene's expression profile (Broihier, 2002).

It was found that Lim3 is expressed in about 40 neurons per hemisegment -- this is many more neurons than previously identified by reporter gene expression. Of particular interest, Lim3 is expressed in all Exex-positive neurons as well as in several lateral Exex-negative neurons, including the Eve-positive EL interneurons. Therefore, like Exex, Lim3 is expressed in MNs projecting in the primary and secondary branches of both the SN and ISN. Since previous work has demonstrated that Lim3 is expressed in the TN nerve (Thor, 1999), it is concluded that Lim3 is expressed in all motor axon branches. These results suggest that all ventrally and laterally projecting MNs may express Lim3 (Broihier, 2002).

Despite the near identity of the Exex and Lim3 expression patterns, Exex and Lim3 do not activate each other's expression in these cells. Exex expression initiates normally in lim3 mutants and Lim3 expression in Exex-expressing cells also initiates normally in exex mutants. These data demonstrate that Exex and Lim3 are activated independently of one another in coexpressing cells and suggest that they act in parallel to specify neuronal identity. In addition, the striking similarity of the Exex and Lim3 expression patterns suggests coregulation of Lim3 and Exex by a largely overlapping set of transcriptional regulators (Broihier, 2002).

More limited overlap is found in the expression patterns of Exex and Islet. Islet is expressed in roughly 30 neurons per hemisegment, the majority of which are located laterally in the CNS. Exex and Islet are coexpressed in three discrete neuronal populations: the medial ISNb-projecting RP MNs, a pair of mediolateral interneurons corresponding to the serotonergic interneurons of the CNS, and a compact cluster of six lateral neurons. As observed for Exex and Lim3, Exex and Islet do not regulate each other's expression -- Islet expression is normal in exex mutant embryos and Exex expression is normal in isl mutant embryos. These results indicate that exex and isl do not fall into a simple linear hierarchy and suggest they act in parallel to specify neuronal fate (Broihier, 2002).

To investigate whether exex and Islet act in parallel, isl;exex double mutants were constructed and axonal organization was analyzed in these embryos. isl or exex single mutant embryos exhibit no overt defects in the overall architecture of the CNS. In contrast, isl;exex double mutant embryos exhibit clear defects in the organization of the axonal scaffold. For example, the anterior and posterior commissures are thinner than in wild-type and frequently only one commissure forms per segment. In addition, the longitudinal connectives are thinner than in wild-type and often veer toward or away from the midline (Broihier, 2002).

The defects in axonal organization in isl;exex double mutants have suggested these embryos might exhibit pronounced defects in motor axon projections. Whereas the axonal phenotypes of both single mutants are confined to the ISNb nerve branch, double mutant embryos display widespread defects. In isl;exex double mutants, the organization of motor axons into five nerve branches usually occurs, though axonal outgrowth is substantially delayed relative to wild-type. In addition, the penetrance of ISNb phenotypes in isl;exex double mutant embryos is dramatically higher than in exex single mutants. In 96% of hemisegments, the ISNb either bypasses the ventral muscle domain and extends along the ISN, or stalls shortly after it defasciculates from the ISN. Furthermore, defects are observed in the main ISN branch. In 32% of hemisegments, ISN axons defasciculate inappropriately, giving the ISN a 'frayed' appearance. At lower frequency (5%), the ISNs from adjacent hemisegments fuse. The ISN phenotypes are consistent with the presence of Exex-positive axons in the ISN and demonstrate that like ISNb, the ISN is sensitive to exex levels. Since it is unclear whether Isl is expressed in ISN-projecting neurons, the ISN phenotype in isl;exex embryos may result from loss of isl and exex activity either in common or distinct neuronal populations. In conclusion, the widespread axonal phenotypes in isl;exex double mutant embryos indicate that isl and exex act in parallel to regulate neuronal differentiation. Furthermore, the fact that the isl;exex double mutant reveals a role for exex in regulating ISN-projecting axons suggests that exex may genetically interact with other factors to control the outgrowth of additional motor axon branches (Broihier, 2002).

Expression analyses indicate that Exex and Lim3 are expressed widely in ventrally and laterally projecting MNs. In contrast, Eve has been shown to be expressed in dorsally projecting MNs, suggesting that Exex/Lim3 and Eve might label nonoverlapping MN populations. This is, in fact, what is observed since Exex and Eve label mutually exclusive neuronal subsets. Lim3 and Eve also identify nonoverlapping sets of MNs, since they are only coexpressed in the EL interneurons. Together with other expression analyses, these data show that Exex/Lim3 are expressed in the majority of Eve-negative MNs and demonstrate that Exex/Lim3 and Eve identify distinct MN classes (Broihier, 2002).

Drosophila HB9 is detected in a subset of motoneurons with ventral muscle targets and in a small group of interneurons, including the well characterized serotonergic interneurons. RNA interference knockdown of HB9 levels leads to defects in motoneuron ventral muscle target recognition, ectopic expression of a marker for dorsally projecting motoneurons (Even-skipped), and defects in serotonergic interneuronal projections. Conversely, ectopic HB9 expression causes an expansion of ventral motoneuron projections and repression of Even-skipped. Thus, Drosophila HB9 is required in a subset of motoneurons and interneurons for establishing proper axon projections but does not have a general role in distinguishing motoneuron and interneuron cell types (Odden, 2002).

The CNS contains three primary cell types: motoneurons, interneurons, and glia. Drosophila genes expressed specifically in all motoneurons have not been described, although a growing number of genes are known to be expressed in subsets of motoneurons. The Even-skipped (Eve) homeodomain transcription factor is expressed in dorsally projecting motoneurons and a subset of interneurons; loss of function and misexpression experiments show that it is necessary and sufficient for dorsal axon projections in motoneurons. The Huckebein (Hkb) zinc finger transcription factor is expressed in a subset of dorsally and ventrally projecting motoneurons and a subset of interneurons; it is required for motoneuronal pathfinding and target recognition. Islet and Lim3 are LIM (lin-11, isl-1, mec-3) homeodomain transcription factors that are expressed in overlapping subsets of ventrally projecting motoneurons and a subset of interneurons, in which they regulate motoneuronal pathfinding and target recognition. Together, these studies have led to the model that the motoneuron population consists of small groups of motoneurons that are each specified by a distinct 'combinatorial code' of transcription factors. It is unknown whether any additional transcription factors promote a general motoneuron identity (Odden, 2002).

On the basis of recent vertebrate studies, the HB9/MNR2 gene family is a prime candidate for a general determinant of somatic motoneuron cell type. In vertebrates, chick MNR2 and mouse HB9 are expressed in presumptive somatic motoneuron progenitors. Mouse HB9 maintains expression in somatic motoneurons, whereas chick HB9 is expressed only in postmitotic somatic motoneurons. Together, vertebrate HB9/MNR2 transcription factors are expressed in all somatic motoneurons and excluded from interneurons. Chick MNR2 or HB9 misexpression in interneurons causes a decrease in interneuronal markers; in addition, chick HB9 misexpression causes an increase in motoneuron markers. Mice lacking HB9 have somatic motoneurons with a hybrid motoneuron/interneuron fate; they extend motoneurons into the muscle field but transiently express interneuronal markers. Thus, vertebrate HB9/MNR2 genes are expressed specifically in somatic motoneurons and are essential for distinguishing motoneuron/interneuron cell types (Odden, 2002 and references therein).

Drosophila HB9 is the sole fly ortholog of the HB9/MNR2 gene family. Drosophila HB9 differs from its vertebrate orthologs in several ways: it is not expressed in all somatic motoneurons, it is expressed in a subset of interneurons, and it is required for the proper development of both interneurons and motoneurons (Odden, 2002).

BLAST searches of the Drosophila genome with the entire coding region of chick HB9 and chick MNR2 protein sequences identified a single related Drosophila gene (CG8254). PCR was used to amplify a 751 nt genomic DNA fragment, which was used to screen a pNB40 cDNA library and obtain one full-length clone; conceptual translation of the cDNA yielded a protein identical to that predicted by the Drosophila genome sequence project (Odden, 2002).

To determine the function of HB9 in ventrally projecting motoneurons, RNAi was used to knock down the levels of HB9. Greater than 90% of injected embryos showed extremely low HB9 protein levels; thus, these HB9RNAi embryos can be considered strong hypomorphs for HB9 expression. Control RNAi injections of buffer or other CNS genes showed no effect on HB9 expression (Odden, 2002).

HB9RNAi embryos were assayed for expression of the ventrally projecting motoneuron determinant Islet and the dorsally projecting motoneuron determinant Eve. No change was observed in the expression of the number of cells expressing the islet-tau:myc transgene, despite strong coexpression of Islet and HB9 in a subset of ventrally projecting motoneurons. In contrast, HB9RNAi embryos show a derepression of the Eve dorsal motoneuron determinant: there are consistently two ectopic Eve+ neurons, one located adjacent to the Eve+ dorsally projecting aCC motoneuron and the other located near the Eve+ dorsally projecting U1-5 motoneurons. It was reasoned that if loss of HB9 leads to ectopic Eve expression, then perhaps misexpression of HB9 would inhibit Eve expression. The Gal4/UAS system was used to misexpress HB9, using either ptc,en-Gal4 (which drives HB9 expression in all neuroectoderm and newly formed neuroblasts) or sca-Gal4 (which drives HB9 expression in all newly formed neuroblasts and the initial GMCs and neurons in their lineages). Both drivers result in precocious and ubiquitous expression of HB9 throughout the CNS, and both result in a nearly complete elimination of Eve expression in all dorsally projecting motoneurons (aCC, RP2, U1-5) and the pCC intersegmental interneuron. Interestingly, there is an increase in the number of Eve+ EL local interneurons, showing that the inhibition of Eve expression is not a general effect but depends on the cell type in which HB9 is expressed. There is no change in Islet expression after misexpression of HB9. It is concluded that HB9 restricts Eve expression to a specific subset of motoneurons, in which Eve promotes innervation of dorsal muscle targets. However, no role is found for HB9 in regulating islet gene expression in ventrally projecting motoneurons (Odden, 2002).

HB9 is expressed in ventrally projecting motoneurons: HB9RNAi embryos were assayed for defects in ventral motoneuron axon outgrowth, pathfinding, and muscle target recognition. Focus was placed on the four HB9+ RP1/3/4/5 motoneurons, which project out of the CNS via the SNb to form three distinct synaptic endings between the ventral muscles 7, 6, 13, and 12. In wild-type embryos, three well defined synaptic endings were observed at the 7/6, 6/13, and 13/12 muscle clefts in 90% of the hemisegments assayed. In HB9RNAi embryos, general CNS morphology is normal; axons project out of the SNb nerve at approximately the normal time and terminate within the appropriate ventral muscle field but fail to establish synaptic endings at the 7/6, 6/13, and 13/12 muscle clefts in 65% of the hemisegments assayed. In the affected segments, axons have expanded growth cones and terminate in an abnormally broad pattern. In addition to defects in ventral muscle target recognition, a slight increase was observed in the thickness of the intersegmental nerve (ISN) innervating dorsal muscles, which may reflect ectopic dorsal motor projections from the two extra Eve+ neurons, because it is known that ectopic Eve induces dorsal axon projections. Thus, HB9 is required for ventral muscle target recognition by the well characterized RP motoneurons (Odden, 2002).

To determine whether misexpression of HB9 leads to an increase in ventral muscle target innervation at the expense of dorsal muscle projections (the opposite of the HB9RNAi phenotype), scabrous-Gal4 was used to misexpress HB9 throughout the CNS (but not muscles) and motoneuron projections were scored with FasII. Indeed, a striking increase in the thickness of the SNb projections to ventral muscles was observed, with many hemisegments showing the normal synaptic endings in three muscle clefts. A clear decrease was observed in the number of motoneuron projections to dorsal muscles. The loss of dorsal projections is a result of multiple defects, including stalling of the ISN in the ventral muscle field or ISN neurons forming apparent synapses on the more ventral transverse nerve, the ISN and the SNb were never observed fasciculating together. It is concluded that HB9 promotes the targeting of motoneuron projections to ventral muscles and inhibits motoneuronal projections to dorsal muscle targets (Odden, 2002).

Drosophila HB9, unlike vertebrate HB9/MNR2 genes, is expressed in a subset of interneurons. To determine whether HB9 plays a role in interneuronal axon targeting similar to its role in motoneurons, the eagle-kinesin:lacZ transgene was used as an axonal marker for the HB9+ EW1/EW2 serotonergic interneurons and the lineally related EW3 interneuron. In wild-type embryos, the EW1-EW3 axons are fasciculate tightly, project anteriorly, turn medially to cross the midline, and synapse within the contralateral neuropil. In HB9RNAi embryos, the EW1-EW3 neurons project anteriorly as usual but are occasionally defasciculated and fail to cross the midline. No change was observed in the contralateral projections of other interneurons (e.g., the HB9- EL interneurons), suggesting that EW1-EW3 axon pathfinding defects are not caused indirectly by midline defects. It is concluded that HB9 is required in EW1-EW3 interneurons to promote contralateral axon projections. Misexpression of HB9 causes many general defects in the CNS, some of which, such as broken longitudinal connectives, can be attributed to interneuronal defects (Odden, 2002).

Thus HB9 is expressed in a subset of ventrally projecting motoneurons but not in their target muscles, and HB9 is required for proper muscle target recognition by these motoneurons. It is possible that the HB9 phenotype is caused by loss of Islet or Lim3 expression, because both Islet and Lim3 are required to establish normal ventral axon projections, and HB9 expression overlaps with Islet and Lim3. However, the HB9 loss and gain of function has no effect on Islet expression, and the HB9 and lim3 phenotypes are clearly different (lim3 mutants show a rerouting of SNb motoneurons into the SNd; this is not observed in HB9RNAi embryos). Instead, a model is favored in which HB9, Islet, and Lim3 have independent functions in establishing RP motoneuronal axon projections. lim3 mutants have pathfinding defects in which SNb motoneurons are diverted into the SNd nerve and terminate outside their normal muscle field; islet mutants also show pathfinding and fasciculation defects, in addition to target recognition defects. HB9 RNAi causes target recognition defects but not fasciculation defects or pathfinding defects (Odden, 2002).

Misexpression of HB9 is sufficient to promote thickening of ventral motor projections and reduced innervation of dorsal muscles. This phenotype could arise in several ways. (1) HB9 could transform interneurons into ventrally projecting motoneurons. This is thought to be unlikely, because no transformation is seen of apterous- or islet-expressing interneurons into motoneurons after HB9 misexpression; moreover, HB9 is normally expressed in some interneurons without turning them into motoneurons. (2) HB9 could transform dorsal motoneurons into ventral motoneurons. This seems unlikely, because dorsal motoneurons are clearly extending in the ISN, and although it is truncated, it is not fused with the SNb nerve. (3) The SNb motoneurons could be slightly defasciculated, which would make them appear thicker than normal. Transmission electron microscopy to count axons in the SNb would be necessary to test this model. Finally, the model is favored that HB9 induces SNa motoneurons to fasciculate into the SNb nerve root and innervate the SNb muscle target field, because misexpression of HB9 leads to the loss of the SNa nerve root in parallel with the thickening of the SNb nerve root. The SNa normally extends past the SNb target muscles en route to its own target muscles, so misexpression of HB9 might lead to precocious termination at the SNb target muscles (Odden, 2002).

HB9 is expressed in the serotonergic EW1 and EW2 interneurons and a third lineally related EW3 interneuron and is required to establish their normal contralateral projections. It is suspected that HB9 is acting autonomously in these interneurons rather than leading to defects at the midline that block contralateral projections, because HB9 is not expressed in midline cells of the CNS, and other interneurons (ELs) and motoneurons (RPs) show normal contralateral projections. The generation of hb9 mutant clones in the serotonergic interneurons will be necessary to distinguish between a cell autonomous or cell nonautonomous function of HB9 in regulating interneuronal axon projections. Interestingly, many transcription factors known to regulate motoneuron development, including Eve, Islet, Lim3, and Hkb, are also expressed in a subset of interneurons. islet and hkb are both expressed in the serotonergic interneurons, and each is required for proper axon pathfinding and neurotransmitter synthesis; lim3 is required for axon pathfinding of a different subset of interneurons. An open question is how interneurons can maintain expression of motoneuronal determinants such as HB9, Islet, Lim3, and Eve without fasciculating with motoneurons or exiting the CNS (Odden, 2002).

The expression and function of HB9 in interneurons is unexpected, because all known vertebrate HB9/MNR2 genes are expressed only in motoneurons within the CNS. Perhaps there are small groups of HB9/MNR2+ interneurons in vertebrates that have evaded detection; alternatively, Drosophila may have co-opted an ancestral motoneuronal determinant, HB9, for a parallel function in interneurons. Detailed analysis of HB9 expression patterns in additional organisms will be necessary to resolve this question (Odden, 2002).

Drosophila HB9 is expressed primarily in postmitotic neurons, whereas vertebrate HB9 family members are expressed in progenitor cells and in differentiated cells. This difference in the timing of HB9 expression may reflect differences in the timing of motoneuron cell fate commitment in each organism. In mouse and chick embryos, a domain of ventral spinal cord neural precursors goes through a phase in which they generate only motoneurons; during this period, the precursors express HB9. In Drosophila, there are no known precursors that generate solely motoneurons; virtually every motoneuron derives from a terminal cell division that produces one motoneuron and a non-motoneuron sibling cell. Thus, it is not surprising that Drosophila motoneuron determinants either are restricted to postmitotic cells (Islet and HB9) or are expressed just before the terminal division but downregulated rapidly in the non-motoneuronal sibling (Eve) (Odden, 2002).

The expression of Drosophila HB9 may help illuminate how Drosophila motoneurons are related evolutionarily to vertebrate motoneurons. It has been proposed that Drosophila motoneurons fall into two nonoverlapping groups: Islet+ motoneurons that are homologous to vertebrate somatic and visceral motoneurons and Eve+ motoneurons that have no vertebrate motoneuron counterpart. The Drosophila Islet+ motoneurons contain somatic motoneurons projecting to ventral muscle targets and a visceral TMNp motoneuron that projects to the heart; the Eve+ motoneurons project to dorsal muscles. Drosophila HB9 expression provides additional support for this model. Drosophila HB9 is expressed in many Islet+ somatic motoneurons but not in the Islet+ TMNp visceral motoneuron, similar to the observed restriction of vertebrate HB9/MNR2 expression to somatic motor neurons but not visceral motor neurons. These data are consistent with a model in which Drosophila Islet+ HB9+ somatic motoneurons are homologous to vertebrate Islet+ HB9+ somatic motoneurons, the Drosophila Islet+ TMNp visceral motoneuron is homologous to vertebrate Islet+ visceral motoneurons, and the Drosophila Eve+ motoneurons have no vertebrate motoneuron counterpart. It is noted, however, that murine Even-skipped (Evx1/2) CNS expression is restricted to the locally projecting commissural V0 interneurons. Perhaps this class of vertebrate interneurons has unrecognized similarities to Drosophila Eve+ motoneurons and/or interneurons (Odden, 2002).

Regulation of locomotion and motoneuron trajectory selection and targeting by the Drosophila homolog of Olig family transcription factors

During the development of locomotion circuits it is essential that motoneurons with distinct subtype identities select the correct trajectories and target muscles. In vertebrates, the generation of motoneurons and myelinating glia depends on Olig2, one of the five Olig family bHLH transcription factors. This study investigated the so far unknown function of the single Drosophila homolog Oli. Combining behavioral and genetic approaches, this study demonstrates that oli is not required for gliogenesis, but plays pivotal roles in regulating larval and adult locomotion, and axon pathfinding and targeting of embryonic motoneurons. In the embryonic nervous system, Oli is primarily expressed in postmitotic progeny, and in particular, in distinct ventral motoneuron subtypes. oli mediates axonal trajectory selection of these motoneurons within the ventral nerve cord and targeting to specific muscles. Genetic interaction assays suggest that oli acts as part of a conserved transcription factor ensemble including Lim3, Islet and Hb9. Moreover, oli is expressed in postembryonic leg-innervating motoneuron lineages and required in glutamatergic neurons for walking. Finally, over-expression of vertebrate Olig2 partially rescues the walking defects of oli-deficient flies. Thus, these findings reveal a remarkably conserved role of Drosophila Oli and vertebrate family members in regulating motoneuron development, while the steps that require their function differ in detail (Oyallon, 2012).

The generation of coordinated muscle contractions, enabling animals to perform complex movements, depends on the assembly of functional neuronal motor circuits. Motoneurons lie at the heart of these circuits, receiving sensory input directly or indirectly via interneurons within the central nervous system (CNS) and relaying information to muscles in the periphery. During development neural precursors give rise to progeny that eventually adopt unique motoneuron subtype identities. Their axons each follow distinct trajectories into the periphery to innervate specific target muscles. Understanding of the molecular mechanisms that control the differentiation and respective connectivity of distinct neuronal subtypes is still limited (Oyallon, 2012).

The Olig family of basic Helix-Loop-Helix (bHLH) transcription factors in vertebrates includes the Oligodendrocyte lineage proteins Olig1-3, Bhlhb4 and Bhlhb5. All members play pivotal roles in regulating neural development. Olig2 controls the sequential generation of somatic motoneurons and one type of myelinating glia, the oligodendrocytes, from the pMN progenitor domain in the ventral neural tube. Olig2 mediates progenitor domain formation by cross-repressive transcriptional interactions and motoneuron differentiation upstream of the LIM-homeodomain containing transcription factors Lim3 (Lhx3) and Islet1/2 (Isl1/2). Downregulation of Olig2 enables Lim3 and Isl1/2 together with the proneural bHLH transcription factor Neurogenin2 (Neurog2) to activate the expression of Hb9, a homeodomain protein and postmitotic motoneuron determinant. In addition, Olig2 cooperates with the homeodomain protein Nkx2.2 to promote oligodendrocyte formation from uncommitted pMN progenitors. Olig1 mediates gliogenesis redundantly with Olig2, while Olig3 controls interneuron specification within dorsal neural tube progenitor domains. Recent studies uncovered important requirements of Bhlhb4 in retinal bipolar cell maturation, and Bhlhb5 in regulating the specification of retinal amacrine and bipolar cells, area-specific identity acquisition and axon targeting of cortical postmitotic neurons, as well as differentiation and survival of distinct interneuron subtypes in the spinal cord. In Drosophila, genome-wide data base searches identified one single family member, called Olig family (Oli)), and a recent study described Oli expression in the embryonic ventral nerve cord (VNC). However, despite the central roles of vertebrate Olig family members, the function of their Drosophila counterpart has not been investigated (Oyallon, 2012 and references therein).

In Drosophila, neurons are derived from stem cell-like neuroblasts (NBs). These divide asymmetrically to generate secondary precursor cells, the ganglion mother cells (GMCs), which divide once to produce two postmitotic neurons and/or glia. 15 of 30 embryonic NB lineages give rise to 36 motoneurons in addition to interneurons per abdominal hemisegment. Zfh1 regulates general motoneuron fate acquisition at the postmitotic level. The specification of ventrally projecting motoneuron subtypes is mediated by a combinatorial expression of five transcriptional regulators -- the fly orthologs of Isl, Lim3, Hb9 and Nkx6, as well as the POU protein Drifter (Dfr; Ventral veinless -- FlyBase). Many of these determinants are highly conserved, raising the question as to whether Oli functions as part of this genetic network that shapes motoneuron diversity. Although related molecules in vertebrates and invertebrates appear to mediate late aspects of glial function, factors that regulate early steps of gliogenesis and are molecularly and functionally conserved have so far not been identified. Olig2 is essential for oligodendrocyte development in vertebrates, and a recent study also implicated the C. elegans homolog Hlh-17 in regulating gliogenesis). Thus, Oli is also a potential candidate that could control early glial development in Drosophila (Oyallon, 2012).

This study provides insights into the so far unexplored function of the Oli bHLH transcription factor in the Drosophila nervous system. Oli is not required in glia; however, taking advantage of the well-defined embryonic motoneuron lineages and axonal projections, this study demonstrates that oli controls trajectory selection and muscle targeting of ventral motoneuron subtypes. Moreover, Oli is expressed in postembryonic lineages, which include glutamatergic leg-innervating motoneurons. Loss-of-function experiments revealed that oli is required for larval and adult locomotion. Chick Olig2 can partially rescue these defects in adults, highlighting at least one evolutionarily conserved role of Olig transcription factors in flies and vertebrates (Oyallon, 2012).

Oli protein is mainly expressed in postmitotic neurons, as well as in some GMCs during embryonic development. This is consistent with in situ hybridization labeling detecting high levels of oli mRNA in postmitotic progeny, in addition to transient expression in MP2 and 7.1 NBs. Oli is also expressed in postmitotic progeny of postembryonic lineages. By contrast, vertebrate Olig2 is required in progenitors to promote commitment to a general motoneuron identity. Also Olig1 and 3 largely function in progenitors. Interestingly, Bhlhb4 and Bhlhb5 are expressed and required in postmitotic progeny of the retina, brain and spinal cord. Thus, with respect to its primarily postmitotic expression, Drosophila Oli resembles more that of Bhlhb4 and Bhlhb5 than Olig1–3 in vertebrates (Oyallon, 2012).

The dynamic expression of Drosophila Oli is not consistent with that of a member of the temporal series of transcriptional regulators. With the latter, neurons largely maintain the determinant they expressed at the time of their birth. By contrast, Oli is widely expressed in newly born progeny, but subsequently levels decrease, and only some subtypes show high expression during late stages. Vertebrate Olig2 acts as a transcriptional repressor in homomeric and heteromeric complexes, and expression is downregulated in differentiating motoneurons to enable the activation of postmitotic determinants such as Hb9 by Lim3, Isl1/2 and Neurog2 . Strikingly in flies, Oli expression decreases in RP and lateral ISNb motoneurons during embryogenesis and prolonged high expression of Oli elicits muscle innervation defects, supporting the notion that Oli downregulation is critical for its function in some neurons. Oli could thus act in a dual mode to regulate the differentiation of neuronal subtypes. The first one may rely on downregulation and be a feature shared with vertebrate Olig2, the second one may require persistent activity, and possibly be a feature more in common with Bhlhb4 and Bhlhb5 family members (Oyallon, 2012).

The findings indicate that Drosophila Oli, unlike vertebrate Olig2, does not act as a general early somatic motoneuron determinant. It rather contributes to shaping ventral motoneuron subtype development as part of a postmitotic transcriptional regulatory network in concert with Drosophila Lim3, Isl, Hb9 and Dfr (Drifter/Vvl). This notion is supported by findings that (1) Oli is co-expressed in specific combinations with these determinants in differentiated ISNb and TN motoneuron subtypes; (2) similar to other ventral determinants, oli mutant embryos display distinct axonal pathfinding and muscle targeting defects; (3) oli does not act upstream of hb9, isl, lim3 or Dfr; and (4) oli and hb9 genetically interact, as loss of both enhances phenotypes in ISNb axons. Because of the proximity of oli, isl and lim3 genetic loci, it has so far not been possible to further extend these interaction assays. Some defects observed in oli mutants, such as failure to innervate the clefts of muscles 12/13 or aberrant contacts between ISNb and TN motoneurons are qualitatively similar to those observed in isl, lim3, hb9 and dfr, while the phenotype of isl-τ-myc-positive neurons abnormally exiting the VNC via the SN branch appears characteristic for oli. Moreover, the connectivity phenotypes observed in oli gain-of-function experiments were not reminiscent of trajectories of other motoneuron subtypes. This suggests that although Oli is a member of the combinatorial code, unlike for instance Dfr, it does not act as a simple switch between fates. It may rather act in concert or partially redundantly with these other determinants in regulating the stepwise process of axon guidance to ensure robustness of trajectory selection (Oyallon, 2012).

Individual transcription factors within an ensemble may regulate different biological properties to tightly coordinate the differentiation and synaptic connectivity of a given neuron subtype. As Oli does not act upstream of Isl, Lim3, Hb9 and Dfr, it may control the expression of other yet to be identified transcription factors, or - similar to dfr, Nkx6 and eve in Drosophila and Bhlhb5 in mice - axon guidance determinant or - as reported for Neurog2 - cytoskeletal regulators. Examining Fasciclin 3, N-Cadherin, PlexinA, and Frazzled, no obvious altered expression was observed in the absence of oli). Thus, future studies using approaches such as microarrays will be required to identify oli downstream targets that control subtype-specific axonal connectivity (Oyallon, 2012).

While the role of oli in controlling neuronal development linked to locomotion appears conserved in Drosophila and vertebrates, conservation does not extend to glia. Oli is neither expressed in glia during embryonic or postembryonic development, nor is it essential for basic glial formation in the embryonic VNC or required in glia for locomotion. This also applied to other parts of the nervous system, such as the 3rd instar larval visual system endowed with large glial diversity. hlh-17, the C. elegans Oli homolog, is expressed in cephalic sheath glia in the brain, and interestingly in some motoneurons in the larval CNS. However, as analysis of hlh-17 mutants could not pinpoint any requirement in glial generation and differentiation possibly due to redundancy with related factors, the precise role of the worm homolog remains elusive. Although ensheathing glia can be found in both invertebrates and vertebrates, myelinating glia have so far only been identified in vertebrates. This raises the possibility that the glial requirement of vertebrate Olig family members could be secondary, and Olig2 may have been recruited to collaborate with additional transcriptional regulators to promote the formation of myelinating glia. Indeed, Olig2 promotes motoneuron development together with Neurog2, and subsequently collaborates with Nkx2.2 to enable the generation of oligodendrocyte precursors and differentiating offspring from newly formed, uncommitted pMN progenitors. Interestingly in cell-based assays, Oli can physically interact with the Nkx2.2 homolog Ventral nervous system defective (Vnd). Together with the current observation that Oli is not essential for glial development, this suggests that the potential of these determinants to interact is evolutionarily conserved, while the steps depending on them diverged in flies and vertebrates (Oyallon, 2012).

The locomotion defects in oli mutant larvae are likely the consequence of embryonic wiring defects, whereas the adult phenotypes may be due to an additional or even sole postembryonic requirement. Unlike the so far identified widely expressed determinants Chinmo, Broad Complex or Castor in the postembryonic VNC, Oli expression is restricted to distinct lineages. That these include motoneurons is supported by observations that Oli is detected in postembryonic lineages 20-22 and 15, and expression overlaps with that of OK371-Gal4. Moreover, locomotion defects can be partially rescued by over-expressing oli in glutamatergic neurons with this driver. This initial characterization raises many new questions regarding the specific postembryonic role of Oli. Because of the expression in lineage 15, future experiments will need to specifically test, whether oli contributes to consolidating motoneuron subtype identity by regulating dendritic arbor-formation or leg muscle innervation with single cell resolution. The wider expression of Oli and the partial rescue with OK371-Gal4 further suggest a requirement of oli in lineages that are part of locomotion-mediating neural circuits beyond motoneurons. Because of the expression pattern and the severe walking defects of adult oli escapers, these observations open the door for future functional studies to unravel the mechanisms that shape neural circuits underlying adult locomotion (Oyallon, 2012).


UNC-4 represses CEH-12/HB9 to specify synaptic inputs to VA motor neurons in C. elegans

In C. elegans, VA and VB motor neurons arise as lineal sisters but synapse with different interneurons to regulate locomotion. VA-specific inputs are defined by the UNC-4 homeoprotein and its transcriptional corepressor, UNC-37/Groucho, which function in the VAs to block the creation of chemical synapses and gap junctions with interneurons normally reserved for VBs. To reveal downstream genes that control this choice, a cell-specific microarray strategy was used that has identified unc-4-regulated transcripts. One of these genes, ceh-12, a member of the HB9 family of homeoproteins, is normally restricted to VBs. Expression of CEH-12/HB9 in VA motor neurons in unc-4 mutants imposes VB-type inputs. Thus, this work reveals a developmental switch in which motor neuron input is defined by differential expression of transcription factors that select alternative presynaptic partners. The conservation of UNC-4, HB9, and Groucho expression in the vertebrate motor circuit argues that similar mechanisms may regulate synaptic specificity in the spinal cord (Von Stetina, 2007).

Transcription factor cascades define the structure of the vertebrate motor circuit by regulating the differentiation of specific neurons that contribute to this network. A striking feature of these pathways is the frequent use of negative gene regulation to produce distinct fates between neurons generated from adjacent progenitor domains. This study shows that a similar mechanism of repression involving conserved transcriptional components distinguishes the fates of C. elegans motor neurons born as sisters from a common mother cell. These results also offer a strikingly new finding, an explicit link between this biological strategy and the choice of presynaptic partners, a developmental decision of critical importance to motor neuron function. A model is presented of transcriptional regulation of synaptic specificity in C. elegans and the possibility that related schemes may also define wiring in the vertebrate spinal cord is discussed (Von Stetina, 2007).

C. elegans mutants in the unc-4 homeodomain gene display a strong backward movement defect that results from the miswiring of VA class motor neurons with inputs normally reserved for VB motor neurons. Intriguingly, other aspects of VA cell fate (i.e., axon trajectory and process placement) are unchanged, suggesting that UNC-4 functions to control only the synaptic fate of this cell type. This study shows that this change in synaptic specificity depends in part on misexpression of the VB-specific transcription factor, CEH-12/HB9, in VA motor neurons. Normally, UNC-4 functions with UNC-37/Groucho to block ceh-12/HB9 expression in the VAs. Because HB9 is also believed to function as a transcriptional repressor in other organisms, it is proposed that ectopic CEH-12/HB9 in unc-4 and unc-37 mutants triggers miswiring by turning off genes that specify VA inputs. It is possible that ectopic CEH-12/HB9 also activates VB genes that drive the creation of VB-type inputs. These results provide strong genetic evidence for at least one additional pathway downstream from UNC-4 that functions in parallel to CEH-12/HB9. The relative contributions of these pathways to VA input specificity are biased along the anterior-posterior (A/P) axis with ectopic CEH-12 selectively driving the creation of VB inputs to posterior VA motor neurons in unc-4 mutants and the presumptive parallel pathway imposing VB inputs to anterior VAs. Finally, a third set of VB genes, glr-4, del-1, and acr-5 are negatively regulated by unc-4 but have no detectable role in the VA miswiring defect. These cell surface proteins and ion channel components could be indicative of physiologically important differences in the excitability or signaling capacity of VA versus VB motor neurons. In the future, it will be interesting to determine if ectopic ceh-12 expression contributes to the observed depletion of synaptic vesicles in unc-4 mutant neurons (Von Stetina, 2007).

Although ceh-12 is required for the imposition of VB-type inputs to posterior VA motor neurons in unc-4 mutants, inputs to most VB motor neurons apparently do not depend on ceh-12 activity. Two lines of evidence support this conclusion: (1) ceh-12 knock-out mutants do not show an obvious forward movement defect as would be expected if VB motor neurons were miswired; (2) the elimination of ceh-12 activity in these mutants does not perturb the creation of gap junctions between most VBs and AVB command interneurons. These data are consistent with the proposal that ceh-12 functions in parallel to a redundant pathway in VB motor neurons that is sufficient to retain VB-type inputs (Von Stetina, 2007).

This work describes the use of a GFP-tagged UNC-7S marker protein for visualizing gap junctions between specific neuron pairs in the C. elegans motor circuit. This assay has provided an unprecedented opportunity to score gap junction specificity in the light microscope in multiple animals and in a variety of different mutant backgrounds. These experiments indicate that the innexin, UNC-7S, is expressed in AVB command interneurons for assembly into gap junctions with B-class motor neurons. Genetic and physiological data suggest that these gap junctions are likely to be heterotypic, and also include the innexin UNC-9. The ectopic gap junctions between AVB and A-class motor neurons that appear in unc-4 mutants may have a similar subunit composition, since unc-9 is the most abundant innexin transcript expressed in A-class motor neurons. It follows that UNC-9 is also a likely candidate for assembly into gap junctions between VA and AVA command interneurons in wild-type animals. Gap junctions with AVB tend to be located on the motor neuron soma, whereas gap junctions with AVA are more often distributed along the length of the motor neuron partner. Thus, unc-4 may orchestrate the assembly of UNC-9 into gap junctions at particular locations within A-class motor neurons and with selected presynaptic partners. Although gap junctions have been previously thought to provide a largely developmental role in the generation of neural networks in higher vertebrates, recent evidence suggests that these 'electrical' synapses are also important for neural function in adult nervous systems. This view is consistent with ultrastructural and immunochemical data showing that gap junctions are widely distributed in the mature mammalian brain and spinal cord. Since the mechanisms that control the specificity of gap junction assembly in the vertebrate CNS are unknown, the discovery of downstream genes that regulate gap junction placement in C. elegans could provide targets for molecular studies in more complex nervous systems. Moreover, the joint regulation by unc-4 (or ceh-12) of the specificity of chemical and electrical synapse formation is indicative of a common nexus for pathways controlling the assembly of both types of synapses (Von Stetina, 2007).

These findings indicate that ceh-12 conspires with at least one additional pathway in VA motor neurons to control input specificity. unc-4 regulation of ceh-12 is restricted to VA motor neurons in the posterior region of the ventral nerve cord. Because anterior VA motor neurons are also miswired in unc-4 mutants, it is proposed that the presumptive downstream pathway functioning in parallel to ceh-12 may be selectively derepressed in anterior VAs. Other unc-4-regulated genes should be represented in the microarray profile of unc-37 mutant VA motor neurons. One plausible candidate in this data set that could function in parallel to ceh-12 is cog-1, the C. elegans homolog of the homeodomain transcription factor, Nkx6. In Drosophila, dHB9 and Nkx6 act together in ventrally projecting motor neurons to repress dorsal motor neuron traits. COG-1 regulates a similar decision in the C. elegans nervous system by preventing ASER sensory neurons from adopting characteristics normally reserved for ASEL. Potential COG-1 interactions with CEH-12 are suggested by the observation that cog-1::GFP is also expressed in VA and VB motor neurons. cog-1 and other candidate unc-4 target genes in the microarray data set that function in parallel to ceh-12 may be revealed by RNA interference (RNAi) tests currently underway to detect genes that enhance ceh-12-dependent suppression of the Unc-4 phenotype (i.e., improved backward locomotion). Conversely, RNAi of transcripts that are depleted in the unc-37 microarray data set and therefore potentially repressed by ectopic ceh-12 should result in an Unc-4 like movement defect if these genes are required for specifying VA-type inputs (Von Stetina, 2007).

The results showing that ceh-12 preserves VB motor neuron fate by repressing VAB-7/Eve, parallels earlier observations that HB9 regulates motor neuron differentiation in flies, birds, and mammals. In Drosophila, dHB9 is expressed in a subset of ventrally projecting motor neurons where it represses the dorsal motor neuron determinant, Eve, and blocks the adoption of a dorsal axon trajectory. Eve, in turn, opposes ventral fates in dorsal motor neurons by reciprocally repressing dHB9 in a Groucho-dependent mechanism. Interestingly, HB9 is also restricted to ventrally projecting motor neurons in the vertebrate spinal cord where it acts to prevent expression of markers for interneurons arising from the adjacent V2 progenitor domain. In this case, ectopic expression of HB9 in V2 neuroblasts is sufficient to drive expression of motor neuron markers as well as impose motor neuron-like morphological characteristics (i.e., ventral axonal projections). This dual function of HB9 to block as well as activate expression of motor neuron-specific traits is similar to the finding that CEH-12 inhibits VA motor neuron differentiation while simultaneously promoting a specific VB trait. Together, these observations suggest that the key role of HB9 function in motor neuron differentiation is evolutionarily ancient. In this regard, it is noted that the UNC-4 homolog, UNCX4.1, is strongly expressed in the V3 neural progenitor domain immediately adjacent to the MN region in which HB9 resides. It will be interesting to determine if UNCX4.1 functions in the V3 domain to block HB9 expression (Von Stetina, 2007).

Other invertebrate Exex homologs

Characterization of a sea urchin (P. lividus) homeobox gene PIHbox 9 has shown that the homeodomain of PIHbox9 is 95% identical to the homeodomain of the human HB9 gene, indicating that the two genes are highly related. Temporal expression analysis during sea urchin embryogenesis showed an absence of transcripts at early cleavage stages. At late gastrula stage, transcripts were barely detectable and reached the highest abundance at prism/early pluteus stages. By whole mount in situ hybridization, a highly restricted expression was observed in a few cells of the ectoderm-endoderm boundary of embryos at the prism stage. At pluteus stages, expression of PIHbox 9 was confined around the anus (Bellomonte, 1998).

The HB9 homeobox gene has been cloned from several vertebrates and is implicated in motor neuron differentiation. In the chick, a related gene, MNR2, acts upstream of HB9 in this process. An amphioxus homolog of these genes is described; it diverged before the gene duplication yielding HB9 and MNR2. AmphiMnx RNA is detected in two irregular punctate stripes along the developing neural tube, comparable to the distribution of 'dorsal compartment' motor neurons, and also in dorsal endoderm and posterior mesoderm. A new homeobox class, Mnx, is proposed to include AmphiMnx, HB9, MNR2 and their Drosophila and echinoderm orthologs; it is suggested that vertebrate HB9 is renamed Mnx1 and MNR2 be renamed Mnx2 (Ferrier, 2001).

Exex homologs in frogs and birds

While the role of the notochord and floor plate in patterning the dorsal-ventral (D/V) axis of the neural tube is clearly established, relatively little is known about the earliest stages of D/V regionalization. In an effort to examine more closely the initial, preneural plate stages of regionalization along the prospective D/V neural axis, a series of explant experiments were performed employing xHB9, a novel marker of the motor neuron region in Xenopus. Using tissue recombinants and Keller explants it has been shown that direct mesodermal contact is both necessary and sufficient for the initial induction of xHB9 in the motor neuron region. Presumptive neural plate explants removed as early as midgastrulation and cultured in isolation are already specified to express xHB9 but do so in an inappropriate spatial pattern, while identical explants are specified to express the floor plate marker vhh-1 with correct spatial patterning. These data suggest that, in addition to floor plate signaling, continued interactions with the underlying mesoderm through neural tube stages are essential for proper spatial patterning of the motor neuron region (Saha, 1997).

The homeobox gene Hb9, like its close relative MNR2, is expressed selectively by motor neurons (MNs) in the developing vertebrate CNS. In embryonic chick spinal cord, the ectopic expression of MNR2 or Hb9 is sufficient to trigger MN differentiation and to repress the differentiation of an adjacent population of V2 interneurons. Genetic evidence is provided that Hb9 has an essential role in MN differentiation. In mice lacking Hb9 function, MNs are generated on schedule and in normal numbers but transiently acquire molecular features of V2 interneurons. The aberrant specification of MN identity is associated with defects in the migration of MNs, the emergence of the subtype identities of MNs, and the projection of motor axons. These findings show that HB9 has an essential function in consolidating the identity of postmitotic MNs (Arber, 1999).

A homeobox gene, HB9, has been isolated from the tarsometatarsal skin of 13-day-old chick embryos using a degenerate RT-PCR-based screening method. In situ hybridization analysis has revealed that, during development of chick embryonic skin, the HB9 gene is expressed in epidermal basal cells of the placodes, but not in those of interplacodes, and in the dermal cells under the placodes at 9 days before addition of an intermediate layer by proliferation of the basal cells in the placodes. With the onset of epidermal stratification, the direction of the basal cell mitosis changes, with the axis becoming vertical to the epidermal surface. Placodes and interplacodes form outer and inner scales, respectively, after they have elongated distally. During scale ridge elongation at 12-15 days, HB9 is strongly expressed in the epidermis of the outer scale face, where the cell proliferation is more active than in the epidermis of the inner scale face; hence, stratification of the outer scale face is more prominent than that of the inner scale face. After 16 days, when mitotic activity in the epidermal basal cells decreases and the thickness of the epidermis is maintained at a constant level, the HB9 expression decreases with the onset of epidermal keratinization. These results suggest that HB9 may be involved in the proliferation of the epidermal basal cells that accompany epidermal stratification (Kosaka, 2000a).

In situ hybridization and immunohistochemical analysis of HB9 homeobox gene mRNA and protein, respectively, were performed during chick feather development. HB9 mRNA is highly expressed in epidermal basal cells and dermal cells of the placodes and feather buds, but not in those of the interplacodes and interbud regions. HB9 protein is predominantly expressed in dermal cells of the symmetric short buds and decreases after the asymmetric bud stage when the feather bud had becomes elongated along the anterior-posterior (A-P) and proximal-distal (P-D) axis. These results suggest that HB9 gene is regulated in a spatiotemporal manner during feather development, and may be involved in early feather bud morphogenesis (Kosaka, 2000b).

Exex homologs: control of motor neuron differentiation

Sonic hedgehog signaling controls the differentiation of motor neurons in the ventral neural tube, but the intervening steps are poorly understood. A differential screen of a cDNA library derived from a single Shh-induced motor neuron has identified a novel homeobox gene, MNR2, expressed by motor neuron progenitors and transiently by postmitotic motor neurons. The ectopic expression of MNR2 in neural cells initiates a program of somatic motor neuron differentiation characterized by the expression of homeodomain proteins, by neurotransmitter phenotype, and by axonal trajectory. These results suggest that the Shh-mediated induction of a single transcription factor, MNR2, is sufficient to direct somatic motor neuron differentiation (Tanabe, 1998).

Sonic hedgehog (Shh) specifies the identity of both motor neurons (MNs) and interneurons with morphogen-like activity. Evidence is presented that the homeodomain factor HB9 is critical for distinguishing MN and interneuron identity in the mouse. Presumptive MN progenitors and postmitotic MNs express HB9, whereas interneurons never express this factor. This pattern resembles a composite of the avian homologs MNR2 and HB9. In mice lacking Hb9, the genetic profile of MNs is significantly altered, particularly by upregulation of Chx10, a gene normally restricted to a class of ventral interneurons. This aberrant gene expression is accompanied by topological disorganization of motor columns, loss of the phrenic and abducens nerves, and intercostal nerve pathfinding defects. Thus, MNs actively suppress interneuron genetic programs to establish their identity (Thaler, 1999).

In the developing spinal cord, motor neurons acquire columnar subtype identities that can be recognized by distinct profiles of homeodomain transcription factor expression. The mechanisms that direct the differentiation of motor neuron columnar subtype from an apparently uniform group of motor neuron progenitors remain poorly defined. In the chick embryo, the Mnx class homeodomain protein MNR2 is expressed selectively by motor neuron progenitors, and has been implicated in the specification of motor neuron fate. MNR2 expression persists in postmitotic motor neurons that populate the median motor column (MMC), whereas its expression is rapidly extinguished from lateral motor column (LMC) neurons and from preganglionic autonomic neurons of the Column of Terni (CT). The extinction of expression of MNR2, and the related Mnx protein HB9, from postmitotic motor neurons appears to be required for the generation of CT neurons but not for LMC generation. In addition, MNR2 and HB9 are likely to mediate the suppression of CT neuron generation that is induced by the LIM HD protein Lim3. Finally, MNR2 appears to regulate motor neuron identity by acting as a transcriptional repressor, providing further evidence for the key role of transcriptional repression in motor neuron specification (William, 2003).

Several lines of evidence suggest that MNR2, and its relative HB9, function as transcriptional repressors during the process of motor neuron specification: (1) the N-terminal domain of MNR2 essential for its activity in motor neuron specification can function as a potent transcriptional repressor in cell-based reporter assays; (2) the HD of MNR2, when fused to a known co-repressor recruitment domain, the E1a C-terminal domain, can mimic the activity of the wild-type MNR2 protein, both in motor neuron specification and in repression of CT subtype identity. These findings are complemented by genetic studies of HB9 function in mouse, in which HB9 has been shown to repress its own expression and to repress expression of V2 interneuron determinants in motor neurons (William, 2003).

The precise mechanism of MNR2- and HB9-mediated transcriptional repression remains unclear. MNR2, like many other HD proteins, possesses a well conserved eh1 motif that, in other contexts, can recruit Groucho class co-repressors. However, elimination of the eh1 motif in MNR2 does not abolish its ability to induce motor neuron generation. Moreover, fusion of the HD of MNR2 to a potent Groucho recruitment domain results in poor motor neuron-inducing activity in vivo. Thus, the repressor functions of MNR2, and by inference of HB9, may not simply reflect the recruitment of Groucho class co-repressors. The data show that the MNR2 HD-E1a C-terminal repressor domain fusion protein mimics the activity of the wild-type MNR2 protein, raising the possibility that MNR2 repressor activity involves the recruitment of Ctbp class co-repressors. However, additional experiments are necessary to resolve whether the repressor functions of MNR2 normally involve the recruitment of Ctbp class co-repressors. In addition, studies on co-repressor function in Drosophila raise the possibility of cooperative interactions between eh1 Groucho recruitment and Ctbp recruitment domains present within the same transcription factor (William, 2003).

Regardless of the precise co-repressors recruited by MNR2, the evidence supports the view that MNR2 function in vivo is likely to reflect its role as a transcriptional repressor. These findings therefore add to the emerging view that the logic of motor neuron fate specification is grounded in transcriptional repression. Many of the progenitor transcription factors involved in motor neuron specification at steps upstream of MNR2, e.g. Nkx6.1, Nkx6.2 and Olig2, also function as transcriptional repressors. Unlike the Nkx6 and Mnx proteins, Olig2 does not possess a clear eh1 motif, further supporting the idea that the transcriptional repressors that function in motor neuron specification recruit distinct classes of co-repressor protein. Finally, the similarities in sequence and activities of Mnx class HD proteins, and genetic studies of HB9 in mouse and Drosophila indicate that all Mnx class proteins may function as transcriptional repressors. Since HB9 expression in spinal cord is restricted largely to postmitotic motor neurons, these observations imply that the key role of transcriptional repression in motor neuron fate specification extends from progenitor cells into postmitotic neurons (William, 2003).

A LIM-complex directed regulatory network to segregate the identity of neuronal subtypes

Spinal motor neurons (MNs) and V2 interneurons (V2-INs) are specified by two related LIM-complexes, MN-hexamer and V2-tetramer, respectively. This study shows how multiple parallel and complementary feedback loops are integrated to assign these two cell fates accurately. While MN-hexamer response elements (REs) are specific to MN-hexamer, V2-tetramer-REs can bind both LIM-complexes. In embryonic MNs, however, two factors cooperatively suppress the aberrant activation of V2-tetramer-REs. First, LMO4 blocks V2-tetramer assembly. Second, MN-hexamer induces a repressor, Hb9, which binds V2-tetramer-REs and suppresses their activation. V2-INs use a similar approach; V2-tetramer induces a repressor, Chx10, which binds MN-hexamer-REs and blocks their activation. Thus, this study uncovers a regulatory network to segregate related cell fates, which involves reciprocal feedforward gene regulatory loops (Lee, 2008).

A SELEX study revealed that the Lhx3-binding sites deviate between HxRE and TeRE in sequence. As HxRE is recognized by MN-hexamer but not by V2-tetramer, the conformation of Lhx3 in MN-hexamer and V2-tetramer is likely different. This may involve an allosteric structural change in the DNA binding domain of Lhx3 in MN-hexamer, induced by the Isl1:Lhx3 interaction. As MN-hexamer is assembled only in MNs, HxRE-containing genes would be stimulated specifically in MNs but not in V2-INs. In contrast, TeRE is activated by both V2-tetramer and MN-hexamer in vitro, suggesting that TeRE-containing V2 genes could be inappropriately induced in MNs. However, TeRE is a V2-specific response element in the developing embryos due to a collaborative action of Hb9 and LMO4 to silence TeRE in MNs. Thus, these studies demonstrate that DNA-REs for specific transcription complexes are sufficient to confer gene expressions to proper cell types in developing embryos (Lee, 2008).

Although Lhx4 and their cofactor NLI (Ldb, CLIM, Chip) dimerization is dispensable for the DNA-binding activity of V2-tetramer and MN-hexamer, it is essential for their robust transactivation. Thus, reiterated TeRE1/2s and HxRE1/2s are necessary for functional TeREs and HxREs. Indeed, the MN-specific enhancer of Hb9 has two functional HxRE1/2s spaced ~150 nt apart. Similarly, three evolutionarily conserved TeRE1/2 sequences were found in the Chx10-TeRE region. Three possible advantages can be proposed for the NLI dimerization in V2-tetramer and MN-hexamer. First, the requirement for multiple repeats of TeRE1/2 and HxRE1/2 may impose higher stringency for functional target gene selection. Second, NLI dimerization bridges two NLI-interacting transcription factors bound to their DNA-binding sites separated by a relatively long spacer region in Drosophila (Heitzler, 2003; Morcillo, 1997). This raises the possibility that V2-tetramer and MN-hexamer may integrate the transcriptional activity of multiple TeRE1/2s or HxRE1/2s located within a single target gene or across multiple target genes. For instance, both Chx10 and Lin-52 have additional TeRE1/2s within their gene. Thus, it will be interesting to test whether the Chx10-TeRE region is required to regulate both Chx10 and Lin-52 by V2-tetramer and whether the multiple TeRE1/2s throughout these two genes enable V2-tetramer to temporally coordinate expression of these genes during development. Third, NLI dimerization may potentiate the transcriptional activity of LIM-complexes by stabilizing the LIM-complexes and/or facilitating recruitment of transcriptional coactivators and chromatin remodeling complexes. Indeed, single-stranded DNA-binding proteins have been found to interact with NLI and augment the transactivation of NLI-containing complexes (Lee, 2008).

DNA-REs affect the protein-protein interaction properties of their cognate transcription factors. Sox2 and Pou factor family members Oct1 or Oct4 dimerize onto different DNA-REs in distinct conformational arrangements, offering one molecular explanation for the wide spectrum of developmental functions for Sox/Pou factors. Thus, it will be interesting to interrogate the role of TeRE and HxRE on the spatial alignment of DNA-protein complex of Lhx3/TeRE1/2 and of Isl1:Lhx3/HxRE1/2 (Lee, 2008).

As p2 and pMN cells are exposed to relatively similar concentration of Shh, deregulation of the transcriptional events downstream of Shh often results in V2-MN fate conversion or hybrid phenotypes. The initial segregation of V2 and MN pathways appears to involve transcriptional crossrepression of progenitor factors Irx3 and Olig2 in neuroepithelial cells. However, additional mechanisms are likely to be needed, as both differentiating V2 and MN cells express Lhx3/4, which are necessary for their cell fates. The results reveal an efficient feedforward gene regulatory circuitry in which a cell-type specific LIM-complex triggers expression of a transcriptional repressor, which in turn binds and represses the DNA-RE of another related LIM-complex, thereby blocking the unwanted choice of alternative fates. This likely contributes to establishing a precise cell identity once a specific LIM-complex is assembled and activates the downstream target genes in neural precursors (Lee, 2008).

First, MN-hexamer upregulates Hb9, which in turn refines MN-gene expression by silencing V2-genes. Hb9 selectively binds TeREs, which prevents undesirable recruitment of MN-hexamer (and any V2-tetramer) to TeREs and represses their inappropriate activation in embryonic MNs. Indeed, V2 genes are upregulated in MNs lacking Hb9. Hb9, a transcriptional repressor, may suppress TeRE-containing V2 genes by recruiting corepressors to their TeREs. Mnr2, an Hb9 paralog, is also known to recruit Ctbp-like corepressor. Thus, TeRE-containing V2 genes are likely to be activated by V2-tetramer in V2-INs, while they are simultaneously silenced by Hb9 in MNs (Lee, 2008).

Second, V2-tetramer directly binds Chx10-TeRE and upregulates Chx10 in chick spinal cord, suggesting that Chx10 is a direct target gene of V2-tetramer. Although Chx10 regulates retinal development, neither its function in the developing spinal cord nor its in vivo target genes are known. Interestingly, Chx10 binds Hb9-HxRE through its homeodomain and represses both the basal and MN-hexamer-induced levels of transcriptional activity mediated by HxRE, consistent with a previous report that Chx10 primarily functions as a transcriptional repressor. In V2 cells, Chx10 could be necessary to completely shut off any leaky expression of HxRE-containing MN genes or to actively block erroneous activation of MN genes by other transcription factors shared between V2-INs and MNs via recruiting corepressors to HxREs. Overall, repression of HxRE by Chx10 is likely to contribute to further refining V2 identity by suppressing unwanted MN-gene expression in V2 cells. Analysis of V2 specification in Chx10 mutant embryos should help examine this possibility genetically (Lee, 2008).

Overall, these studies reveal a sequential regulatory cascade of gene expression that operates to ensure the high fidelity in gene regulation required to specify two closely related, but distinct neural subtypes during vertebrate CNS development. This cascade resembles feedforward loops described in other organisms such as E. Coli, yeast, C. elegans and Drosophila. In particular, this study highlights the key role for DNA-REs in feedfoward gene regulatory loop and combinatorial transcription code, which have been underappreciated previously. Both HxRE and TeRE function as binary switches in the developing spinal cord; i.e., TeRE is off-switch in MNs and on-switch in V2-INs, while HxRE is on-switch in MNs and off-switch in V2-INs. Importantly, these strategies should reinforce the distinct gene expression outcomes in MNs and V2-INs, as expression of HxRE- and TeRE-containing genes would be precisely coregulated to opposite directions depending on the cell context. Thus, cell-type-specific DNA-RE alone is capable of decoding all the cell-fate-specifying genetic programs installed in each cell type, sensing both transcriptional activation and repression machineries. This model also predicts that a set of genes with TeRE or HxRE would be synonymously regulated during cell fate specification. Thus, the defined consensus TeRE and HxRE sequences could be useful in bioinformatics approaches to find a group of genes, which are specifically expressed in V2-INs and MNs and direct V2 and MN differentiations and maturations. Together, these findings provide a prototypic gene regulatory network for cell-type specification in development, which involves feedforward gene regulatory loops (Lee, 2008).

Hb9-MNe of Hb9 gene consists of functional HxRE and E-box elements that recruit proneural basic helix-loop-helix (bHLH) factors. MN-hexamer transcriptionally synergizes with proneural bHLH factors Ngn2 and NeuroM to fully activate Hb9 gene and subsequently specify MNs in the developing spinal cord and P19 cells. This synergistic interaction of MN-hexamer and Ngn2/NeuroM requires DNA bindings of these transcription factors in proximity. Thus, full activation of Chx10-TeRE in the neural tube may also need other transcription factors bound elsewhere in Chx10. It will be interesting to test whether V2-tetramer indeed cooperates with other transcription factors involved in V2 specification, such as Mash1, GATA2, FoxN4, or SCL, to promote V2-IN fate (Lee, 2008).

LMO4 disrupts the assembly of V2-tetramer in newborn MNs by displacing Lhx3 from NLI and suppresses V2-IN development in chick embryos. Among LMOs, LMO4 is most highly expressed in differentiating MNs in chick and mouse embryos and it binds NLI with a 2-fold higher affinity than LMO2. Thus, LMO4 is a good candidate to regulate the formation of LIM-complexes in MNs. Under the condition of 1:1 interactions, the affinities of Lhx3 and Isl1 for NLI binding are comparable, suggesting that LMO4 inhibits similarly the formation of V2-tetramer and MN-hexamer through competition for NLI binding. However, the data indicate that LMO4 functions as a selective competitor to disrupt V2-tetramer over MN-hexamer assembly. Interestingly, the binding of NLI and Isl1 without Lhx3 is also sensitive to LMO4, suggesting that the resistance of NLI:Isl1-Lhx3 binding to LMO4 is not simply due to the differences in the binding affinities between NLI:Lhx3 interaction and NLI:Isl1 interaction. Rather, the differences in the complex architecture of MN-hexamer and V2-tetramer may contribute to the distinct sensitivity of the two complexes to LMO4. Relative to V2-tetramer, MN-hexamer is a higher-order multiprotein complex. Thus, it could be more stable through multiple protein-protein interactions and less sensitive to a competitor such as LMO4. In MNs, LMO4 should increase the population of MN-hexamer, as MN-hexamer is formed at the expense of V2-tetramer. Consistently, deletion of LMO4 results in progressive increase of V2-INs. Although V2-MN hybrid cells are consistently found in LMO4 mutants, the phenotype is relatively subtle in LMO4 single mutant and greatly enhanced in LMO4:Hb9 compound mutants. Thus, LMO4 may provide a fine-tuning mechanism to control the stoichiometry of LIM-complexes in the developing spinal cord by increasing MN-hexamer concentration in MNs (Lee, 2008).

The results demonstrate that Hb9 and LMO4 cooperate to silence V2 genes. Why is the cooperative action of Hb9 and LMO4 necessary to inhibit V2 genes in MNs? LMO4 seems to function as a modulator, rather than an active MN fate selector, to promote MN-hexamer formation over other possible LIM-complexes in MNs. Likewise, although Hb9 is dominant over MN-hexamer for TeRE binding and thus blocks the access of MN-hexamer to TeRE-containing V2 genes, Hb9 may have intrinsically modest affinity to TeRE (weaker than V2-tetramer). Thus, Hb9 alone could be inefficient in blocking binding of V2-tetramer to TeRE, and may not completely shut down V2-gene expression in MNs, unless LMO4 helps Hb9 to bind TeRE more readily by destabilizing V2-tetramer, Hb9's competitor to bind TeRE. The loss of LMO4 and Hb9 in LMO4:Hb9-DKO likely permits both V2-tetramer and MN-hexamer to bind TeREs and upregulate V2 genes. As a consequence, both TeRE-containing V2 genes and HxRE-containing MN genes are activated in MNs, thereby resulting in MN-V2 hybrid cells. Thus, the functional cooperation between Hb9 and LMO4 is expected to be a critical component in the overall strategy to suppress expression of V2 genes in MNs. Together, these findings underscore the importance of actively suppressing alternative fate choice to generate correct neuronal subtype (Lee, 2008).

The data demonstrate that transcriptionally active endogenous MN-hexamer is assembled in MMCm-MNs. Interestingly, LMO4 is maintained mainly in MMCm-MNs among motor columns. Thus, the HxRE may mediate not only MN specification but also postmitotic MN diversification to MMCm cells and LMO4 may also antagonize the unwanted formation of V2-tetramer in MMCm cells. V2-INs also undergo further diversification to excitatory Chx10+ V2a-INs and inhibitory GATA2/3+ V2b-INs. Analogous to MN development, Lhx3 is maintained in V2a-INs, but extinguished in V2b-INs. Thus, TeRE activity could be maintained only in V2a subtype in which V2-tetramer is assembled. In comparison, V2b-INs express GATA2/3, SCL, and LMO4, which could assemble a transactivating complex similar to a hematopoietic complex containing NLI, GATA1, SCL, and LMO2. This raises the possibility that LMO4 might control the V2 subtype segregation by acting as an activator in V2b and a repressor in V2a (Lee, 2008).

In summary, these studies have established that subtle differences in DNA-REs can direct segregation of lineage-specific transcription pathways in the developing nervous system by concertedly mobilizing the action of transcriptional activators and repressors. This regulatory network likely represents a prototypic genetic mechanism for segregating related but distinct cell fates during the nervous system development. Importantly, this knowledge should provide a rational strategy to direct stem/progenitor cells into MNs in vitro (Lee, 2008).

Transcriptional regulation of Hb9

The underlying transcriptional mechanisms that establish the proper spatial and temporal pattern of gene expression required for specifying neuronal fate are poorly defined. This study characterizes how the Hb9 gene is expressed in developing motoneurons in order to understand how transcription is directed to specific cells within the developing CNS. Non-specific general-activator proteins such as E2F and Sp1 are capable of driving widespread low level transcription of Hb9 in many cell types throughout the neural tube; however, their activity is modulated by specific repressor and activator complexes. The general-activators of Hb9 are suppressed from triggering inappropriate transcription by repressor proteins Irx3 and Nkx2.2. High level motoneuron expression is achieved by assembling an enhancesome on a compact evolutionarily-conserved segment of Hb9 located from -7096 to -6896. The ensemble of LIM-HD and bHLH proteins that interact with this enhancer change as motoneuron development progresses, facilitating both the activation and maintenance of Hb9 expression in developing and mature motoneurons. These findings provide direct support for the derepression model of gene regulation and cell fate specification in the neural tube, as well as establishing a role for enhancers in targeting gene expression to a single neuronal subtype in the spinal cord (Lee, 2004).

Developing motoneurons sequentially express several bHLH proteins, including Ngn2 in the progenitor cells followed by NeuroM in the early postmitotic motoneurons and NeuroD in the more mature cells. Ngn2 and NeuroM have been shown to contribute to the activation of Hb9 during the initial stages of motoneuron development, but it remained unclear whether NeuroD in the mature cells could also stimulate Hb9 expression. To compare the activity of these transcription factors, P19 cells were transfected with expression constructs encoding bHLH proteins together with a luciferase reporter containing seven E box elements. Under these conditions Ngn2 activated the reporter much more than either NeuroM or NeuroD. Despite this inherent difference in transactivation, Ngn2, NeuroM, and NeuroD each synergized in a similar way with the LIM factors Isl1 and Lhx3 to trigger Hb9 expression. Likewise, each bHLH factor dimerizes with E47 and binds to the M50 and M100 E box elements in a sequence-specific manner, and exhibits a similar ability to promote motoneuron differentiation from transfected P19 embryonic carcinoma cells when expressed with Isl1 and Lhx3. Taken together, these findings suggest that the initial activation of Hb9 expression is dependent on Ngn2 and NeuroM as motoneurons become postmitotic, and that NeuroD contributes to the maintenance of Hb9 expression in mature motoneurons (Lee, 2004).

Nkx2.2, Nkx6.1, Pax6 and Irx3 control progenitor cell fate by repressing transcription. Since the deletion analysis of Hb9 indicated that repressor proteins might interact with the 2.5 kb distal segment from -8129 to -5575, tests were performed to see whether constructs with this DNA segment were repressed by Nkx2.2, Nkx6.1, Pax6 and/or Irx3 using 293 cell transfections. The Hb9 promoter was repressed ~50-500 fold by Nkx2.2 and Irx3, whereas Pax6 and Nkx6.1 were significantly less active. These findings suggest that progenitor cell factors such as Nkx2.2 and Irx3 expressed by non-motoneuron cells suppress the expression of Hb9 (Lee, 2004).

Genetic studies have shown that Hb9 feeds back negatively to modulate its own expression. Whether Hb9 could suppress the activity of its enhancer when LIM and bHLH factors synergize to activate transcription was tested. The native Hb9 protein and the EnR-Hb9 repressor (Hb9 homeodomain linked to eh1 engrailed repressor domain) both inhibited transcription under these conditions, whereas the Hb9-HD and a fusion of Hb9 to the VP16 activation domain (VP16-Hb9) lacked this activity. Thus, in developing motoneurons where Hb9 transcription is synergistically activated, co-repressors such as those recruited by the engrailed fusion (EnR) appear to be involved in negative feedback regulation. Consistent with these findings, Hb9 protein binds in a sequence-specific manner to the ATTA motifs in the enhancer (Lee, 2004).

Exex homologs: expression in pancreas and mesoderm

A novel human homeobox gene, HB9, was isolated from a cDNA library prepared from in vitro stimulated human tonsil B lymphocytes and from a human genomic library. The HB9 gene is composed of 3 exons spread over 6 kilobases of DNA. An open reading frame of 1206 nucleotides is in frame with a diverged homeodomain. The predicted HB9 protein has a molecular mass of 41 kilodaltons and is enriched for alanine, glycine, and leucine. The HB9 homeodomain is most similar to that of the Drosophila melanogaster homeobox gene proboscipedia. Northern blot analysis of poly(A) RNA purified from the human B cell line RPMI 8226 and from activated T cells has revealed a major mRNA transcript of 2.2 kilobases. Similar analysis of poly(A) RNA from a variety of adult tissues has demonstrated HB9 transcripts in pancreas, small intestine, and colon. Reverse transcriptase-polymerase chain reaction was used to examine HB9 RNA transcripts in hematopoietic cell lines. HB9 RNA transcripts were most prevalent in several human B cell lines and K562 cells. In addition, transcripts were detected in RNA prepared from tonsil B cells and in situ hybridization studies have localized them in the germinal center region of adult tonsil. These findings suggest the involvement of HB9 in regulating gene transcription in lymphoid and pancreatic tissues (Harrison, 1994).

In most mammals the pancreas develops from the foregut endoderm as ventral and dorsal buds. These buds fuse and develop into a complex organ composed of endocrine, exocrine and ductal components. This developmental process depends upon an integrated network of transcription factors. Gene targeting experiments have revealed critical roles for Pdx1, Isl1, Pax4, Pax6 and Nkx2-2. The homeobox gene HLXB9 (encoding HB9) is prominently expressed in adult human pancreas, although its role in pancreas development and function is unknown. To facilitate its study, the mouse HLXB9 ortholog, Hlxb9, was isolated. During mouse development, the dorsal and ventral pancreatic buds and mature beta-cells in the islets of Langerhans express Hlxb9. In mice homologous for a null mutation of Hlxb9, the dorsal lobe of the pancreas fails to develop. The remnant Hlxb9-/- pancreas has small islets of Langerhans with reduced numbers of insulin-producing beta-cells. Hlxb9-/- beta-cells express low levels of the glucose transporter Glut2 and homeodomain factor Nkx 6-1. Thus, Hlxb9 is key to normal pancreas development and function (Harrison, 1999).

The initial stages of pancreatic development occur early during mammalian embryogenesis, but the genes governing this process remain largely unknown. The homeodomain protein Pdx1 is expressed in the developing pancreatic anlagen from the approximately 10-somite stage, and mutations in the gene Pdx1 prevent the development of the pancreas. The initial stages of pancreatic development, however, still occur in Pdx1-deficient mice. Hlxb9 is a homeobox gene that in humans has been linked to dominant inherited sacral agenesis and Hb9 is expressed at early stages of mouse pancreatic development and later in differentiated beta-cells. Hlxb9 has an essential function in the initial stages of pancreatic development. In the absence of Hlxb9 expression, the dorsal region of the gut epithelium fails to initiate a pancreatic differentiation program. In contrast, the ventral pancreatic endoderm develops but exhibits a later and more subtle perturbation in beta-cell differentiation and in islet cell organization. Thus, Hlxb9 is required dorsally for specifying the gut epithelium to a pancreatic fate and ventrally for ensuring proper endocrine cell differentiation (Li, 1999).

The homebox gene Hlxb9, encoding Hb9, exhibits a dual expression profile during pancreatic development. The early expression in the dorsal and ventral pancreatic epithelium is transient and spans from embryonic day 8 to 9 and 10, whereas the later expression is confined to differentiating beta-cells as they appear. Hlxb9 is critically required for the initiation of the dorsal, but not the ventral, pancreatic program. This study demonstrates the requirement for a stringent temporal regulation of Hlxb9 expression during early stages of pancreatic development. In transgenic mice, where Hlxb9 expression (under control of the Ipf1/Pdx1 promoter) is extended beyond e9-e10, the development of the pancreas is drastically perturbed. Morphological analyses show that the growth and morphogenesis of the pancreatic epithelium is impaired. Moreover, differentiation of pancreatic endocrine and exocrine cells is diminished; instead, the pancreatic epithelium with its adjacent mesenchyme adopts an intestinal-like differentiation program. Together, these data point to a need for a tight temporal regulation of Hlxb9 expression. Thus, a total loss of Hlxb9 expression results in a block of the initiation of the dorsal pancreatic program, while a temporally extended expression of Hlxb9 results in a complete impairment of pancreatic development (Li, 2001).

The vertebrate endocrine pancreas has the crucial function of maintaining blood sugar homeostasis. This role is dependent upon the development and maintenance of pancreatic islets comprising appropriate ratios of hormone-producing cells. In all vertebrate models studied, an initial precursor population of Pdx1-expressing endoderm cells gives rise to separate endocrine and exocrine cell lineages. Within the endocrine progenitor pool a variety of transcription factors influence cell fate decisions, such that hormone-producing differentiated cell types ultimately arise, including the insulin-producing beta cells and the antagonistically acting glucagon-producing alpha cells. In previous work, it was established that the development of all pancreatic lineages requires retinoic acid (RA) signaling. Zebrafish was used to uncover genes that function downstream of RA signaling, and this study identified mnx1 (hb9) as an RA-regulated endoderm transcription factor-encoding gene. By combining manipulation of gene function, cell transplantation approaches and transgenic reporter analysis it was established that Mnx1 functions downstream of RA within the endoderm to control cell fate decisions in the endocrine pancreas progenitor lineage. It was confirmed that Mnx1-deficient zebrafish lack beta cells, and, importantly, the novel observation was made that they concomitantly gain alpha cells. In Mnx1-deficient embryos, precursor cells that are normally destined to differentiate as beta cells instead take on an alpha cell fate. These findings suggest that Mnx1 functions to promote beta and suppress alpha cell fates (Dalgin, 2011).


Search PubMed for articles about Drosophila Hb9

Arber, S., Han, B., Mendelsohn, M., Smith, M., Jessell, T. M. and Sockanathan, S. (1999) Requirement for the homeobox gene Hb9 in the consolidation of motor neuron identity. Neuron 23: 659-674. 10482234

Bellomonte, D., et al. (1998). Highly restricted expression at the ectoderm-endoderm boundary of PIHbox 9, a sea urchin homeobox gene related to the human HB9 gene. Mech. Dev. 74(1-2): 185-8. 9651524

Broihier, H. T. and Skeath, J. B. (2002). Drosophila homeodomain protein dHb9 directs neuronal fate via crossrepressive and cell-nonautonomous mechanisms Neuron 35: 39-50. 12123607

Broihier, H. T., Kuzin, A., Zhu, Y., Odenwald, W. and Skeath, J. B. (2004). Drosophila homeodomain protein Nkx6 coordinates motoneuron subtype identity and axonogenesis. Development 131: 5233-5242. 15456721

Dalgin, G., et al. (2011). Zebrafish mnx1 controls cell fate choice in the developing endocrine pancreas. Development 138(21): 4597-608. PubMed Citation: 21989909

Evans, T. A., Bashaw, G. J. (2010) Functional diversity of Robo receptor immunoglobulin domains promotes distinct axon guidance decisions. Curr Biol 20: 567-572. PubMed ID: 20206526

Ferri, A. L., et al. (2007). Foxa1 and Foxa2 regulate multiple phases of midbrain dopaminergic neuron development in a dosage-dependent manner. Development 134: 2761-2769. PubMed Citation: 17596284

Ferrier, D. E., et al. (2001). The Mnx homeobox gene class defined by HB9, MNR2 and amphioxus AmphiMnx. Dev. Genes Evol. 211(2): 103-7. 11455421

Fujioka, M., et al. (2003). Even-skipped, acting as a repressor, regulates axonal projections in Drosophila. Development 130: 5385-5400. 13129849

Harrison, K. A., et al. (1994). A novel human homeobox gene distantly related to proboscipedia is expressed in lymphoid and pancreatic tissues. J. Biol. Chem. 269(31): 19968-75. 7914194

Harrison, K. A., et al. (1999). Pancreas dorsal lobe agenesis and abnormal islets of Langerhans in Hlxb9-deficient mice. Nat. Genet. 23(1): 71-5. 10471502

Heitzler, P., Vanolst, L., Biryukova, I. and Ramain, P. (2003). Enhancer-promoter communication mediated by Chip during Pannier-driven proneural patterning is regulated by Osa. Genes Dev. 17: 591-596. PubMed Citation: 12629041

Jacob, J., et al. (2007). Transcriptional repression coordinates the temporal switch from motor to serotonergic neurogenesis. Nat. Neurosci. 10: 1433-1439. PubMed Citation: 17922007

Kosaka, Y., et al. (2000a). Expression of the HB9 homeobox gene concomitant with proliferation accompanying epidermal stratification during development of chick embryonic tarsometatarsal skin. Histochem J. 32(5): 275-80. 10939514

Kosaka, Y., et al. (2000b). Localization of HB9 homeobox gene mRNA and protein during the early stages of chick feather development. Biochem. Biophys. Res. Commun. 276(3): 1112-7. 11027598

Kuzin, A., Brody, T., Moore, A. W. and Odenwald, W. F. (2005). Nerfin-1 is required for early axon guidance decisions in the developing Drosophila CNS. Dev. Biol. 277: 347-365. PubMed Citation: 15617679

Lacin, H., Zhu, Y., Wilson, B. A. and Skeath, J. B. (2009). dbx mediates neuronal specification and differentiation through cross-repressive, lineage-specific interactions with eve and hb9. Development 136: 3257-3266. PubMed Citation: 19710170

Lanuza, G. M., Gosgnach, S., Pierani, A., Jessell, T. M. and Goulding, M. (2004). Genetic identification of spinal interneurons that coordinate left-right locomotor activity necessary for walking movements. Neuron 42: 375-386. PubMed Citation: 15134635

Lee, S.-K., et al. (2004). Analysis of embryonic motoneuron gene regulation: derepression of general activators function in concert with enhancer factors. Development 131: 3295-3306. 15201216

Lee, S., et al. (2008). A regulatory network to segregate the identity of neuronal subtypes. Dev. Cell 14: 877-889. PubMed Citation: 18539116

Li, H., et al. (1999). Selective agenesis of the dorsal pancreas in mice lacking homeobox gene Hlxb9. Nat. Genet. 23(1): 67-70. 10471501

Li, H. and Edlund, H. (2001). Persistent expression of Hlxb9 in the pancreatic epithelium impairs pancreatic development. Dev. Biol. 240(1): 247-53. 11784060

Miguel-Aliaga, I., Thor, S. and Gould, A. P. (2008). Postmitotic specification of Drosophila insulinergic neurons from pioneer neurons. PLoS Biol. 6(3): e58. PubMed Citation: 18336071

Morcillo, P., et al. (1997). Chip, a widely expressed chromosomal protein required for segmentation and activity of a remote wing margin enhancer in Drosophila. Genes Dev. 11: 2729-2740. PubMed Citation: 9334334

Norton, W. H., et al. (2005). Monorail/Foxa2 regulates floorplate differentiation and specification of oligodendrocytes, serotonergic raphe neurones and cranial motoneurones. Development 132(4): 645-58. PubMed Citation: 15677724

Noyes, M. B., Christensen, R. G., Wakabayashi, A., Stormo, G. D., Brodsky, M. H. and Wolfe, S. A. (2008). Analysis of homeodomain specificities allows the family-wide prediction of preferred recognition sites. Cell 133: 1277-1289. PubMed Citation: 18585360

Odden, J. P., Holbrook, S. and Doe, C. Q. (2002). Drosophila HB9 Is expressed in a subset of motoneurons and interneurons, where it regulates gene expression and axon pathfinding. J. Neurosci.22(21): 9143-9149. 12417636

Oyallon, J., Apitz, H., Miguel-Aliaga, I., Timofeev, K., Ferreira, L. and Salecker, I. (2012). Regulation of locomotion and motoneuron trajectory selection and targeting by the Drosophila homolog of Olig family transcription factors. Dev Biol 369: 261-276. PubMed ID: 22796650

Saha, M. S., Miles, R. R. and Grainger, R. M. (1997). Dorsal-ventral patterning during neural induction in Xenopus: assessment of spinal cord regionalization with xHB9, a marker for the motor neuron region. Dev. Biol. 187(2): 209-23. 9242418

Santiago, C., Labrador, J. P., Bashaw, G. J. (2014) The Homeodomain Transcription Factor Hb9 Controls Axon Guidance in Drosophila through the Regulation of Robo Receptors. Cell Rep 7: 153-165. PubMed ID: 24685136

Sharma, K., et al. (1998). LIM homeodomain factors Lhx3 and Lhx4 assign subtype identities for motor neurons. Cell 95: 817-828. 9865699

Spitzweck, B., Brankatschk, M., Dickson, B. J. (2010) Distinct protein domains and expression patterns confer divergent axon guidance functions for Drosophila Robo receptors. Cell 140: 409-420. PubMed ID: 20144763

Sund, N. J., et al. (2001). Tissue-specific deletion of Foxa2 in pancreatic beta cells results in hyperinsulinemic hypoglycemia. Genes Dev. 15: 1706-1715. PubMed Citation: 11445544

Tanabe, Y., William, C. and Jessell, T. M. (1998). Specification of motor neuron identity by the MNR2 homeodomain protein. Cell 95: 67-80. 9778248

Thaler, J., et al. (1999). Active suppression of interneuron programs within developing motor neurons revealed by analysis of homeodomain factor HB9. Neuron 23: 675-687. 10482235

Thor, S. and Thomas, J. B. (1997). The Drosophila islet gene governs axon pathfinding and neurotransmitter identity. Neuron 18: 397-409. 9115734

Thor, S., Andersson, S. G., Tomlinson, A. and Thomas, J. B. (1999). A LIM-homeodomain combinatorial code for motor-neuron pathway selection. Nature, 397:76-80. 9892357

Von Stetina, S. E., et al. (2007). UNC-4 represses CEH-12/HB9 to specify synaptic inputs to VA motor neurons in C. elegans. Genes Dev. 21: 332-346. Medline abstract: 17289921

Wang, S., Tulina, N., Carlin, D. L. and Rulifson, E. J. (2007). The origin of islet-like cells in Drosophila identifies parallels to the vertebrate endocrine axis. Proc. Natl. Acad. Sci. 104: 19873-19878. PubMed Citation: 18056636

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 Citation: 16631157

William, C. M., Tanabe, Y. Jessell, T. M. (2003). Regulation of motor neuron subtype identity by repressor activity of Mnx class homeodomain proteins. Development 130: 1523-1536. 12620979

Biological Overview

date revised: 20 April 2014

Home page: The Interactive Fly © 2017 Thomas Brody, Ph.D.