REGULATION (part 2/3)

Transcriptional Regulation

Midway through embryogenesis, decapentaplegic is expressed in the visceral mesoderm, and enhances the expression of labial in the underlying midgut endoderm (Staehling-Hampton, 1994).

Mutations affecting thick veins block the expression of two DPP-responsive genes, dpp and labial, in the embryonic midgut (Penton, 1994).

Decapentaplegic is an extracellular signal of the transforming growth factor-beta family with multiple functions during Drosophila development. For example, it plays a key role in the embryo during endoderm induction. During this process, Dpp stimulates transcription of the homeotic genes Ultrabithorax in the visceral mesoderm and labial in the subjacent endoderm. A cAMP response element (CRE) from an Ultrabithorax enhancer mediates Dpp-responsive transcription in the embryonic midgut, and endoderm expression from a labial enhancer depends on multiple CREs. The enhancer, called Ubx B confers Wingless- and Decapentaplegic-dependent expression in the visceral mesoderm. Staining mediated by Ubx B is in two stripes of cells in the visceral mesoderm, a wide prominent one in parasegments 6-9 and a narrow weak one in parasegment 3. The Drosophila CRE-binding protein dCREB-2 binds to the Ultrabithorax CRE. Binding is at a palindromic sequence TGGCGTCA that resembles a typical cAMP response element (CRE) (TGACGTCA). Mutation of this site results in the elimination of response to Dpp, but a maintenance of response to Wg. This residual expression is in parasegment 8 and 9 coinciding with the main source of wg expression in the middle midgut. The Ubx CRE can also mediate response to Dpp signaling in the endoderm. Other transcription factors act through the Ubx B enhancer to confer its tissue-specific response to Dpp in the visceral mesoderm. CRE needs to cooperate with a LEF-1 binding site to respond to the Dpp signal in the visceral mesoderm. Schnurri, a transcription factor implicated in Dpp signaling, fails to interact with Ubx B. Adjacent to the CRE is another palindromic sequence that antagonizes the activating effects of Dpp and Wg signaling on the Ubx B enhancer. Ubiquitous expression of a dominant-negative form of dCREB-2 suppresses CRE-mediated reporter gene expression and reduces labial expression in the endoderm. Therefore, a dCREB-2 protein may act as a nuclear target, or as a partner of a nuclear target, for Dpp signaling in the embryonic midgut (Eresh, 1997).

The fact that Transforming growth factor beta at 60A mutations are dominant enhancers of a sensitized dpp pathway implicates Tgfbeta-60A in potentiating dpp signaling. This is most obvious in the visceral mesoderm of the midgut where dpp signaling is required to regulate homeotic gene expression and to maintain its own expression through a positive feedback mechanism. Although dpp signaling in the visceral mesoderm appears intact in Tgfbeta-60A mutants, a requirement for Tgfbeta-60A is revealed in tkv 6 Tgfbeta-60A double mutants. When dpp signaling is attenuated through a mutant tkv receptor, eliminating Tgfbeta-60A function reduces the signaling to below threshold level. The derepression of Sex combs reduced in the anterior midgut and the loss of expression of dpp target genes (wingless, Ultrabithorax and dpp) in the visceral mesoderm and labial in the endoderm are consistent with inadequate dpp signaling. A similar requirement for Tgfbeta-60A is observed during dorsal closure of the embryonic ectoderm (Chen, 1998).

Drosophila wingless functions in the larval midgut, and acts at two different thresholds to pattern this tissue. Low wingless levels are required to promote the development of copper cells, highly differentiated cells of the larval midgut that are specified by the homeotic gene labial. In contrast, high wingless levels repress copper cell development and allow differentiation of an alternative cell type: large, flat cells. These two different developmental outcomes reflect labial expression, which is stimulated at low levels and repressed at high levels of wingless signaling. Thus, midgut cells respond differentially to distinct wingless thresholds in terms of both gene control and cellular differentiation (Hoppler, 1995).

Fos-related antigen plays a critical role during Drosophila endoderm induction. Fra is required downstream of Dpp signaling for the regulation of labial. Expression of labial in the midgut coincides with copper cells. Labial has a role not only in determination and differentiation of copper cells, but also in the maintenance of their differentiated state. A truncated version of Fra (Fbz)(Eresh, 1997) acts in a dominant negative fashion to interfere with Fra function. Fra acting through labial, by stimulating its expression, to promote copper cell development. The number of labial expressing cells is much reduced in Fbz-expressing larvae. It follows that Fbz must have an effect on labial induction as early as this induction can be detected. These results suggest an early function of Fra during labial induction. Although the Dpp response sequence of a labial midgut enhancer is a cyclic AMP response element (CRE), thought to be the target of Creb-17A, and no interference of Fbz can be detected with the CRE-mediated response, this does not rule out the possibility that Fra acts through the CREs of the labial enhancer, especially if Fra interacts with another DNA-binding protein to do so (Riese, 1997).

Dpp has a prime function during endoderm induction in Drosophila. Dpp is secreted from the outer cell layer of the embryonic midgut (the visceral mesoderm), where Dpp's main source of expression in parasegment ps7 depends directly on the homeotic gene Ultrabithorax (Ubx). In the same cell layer, Dpp stimulates expression of another extracellular signal, Wingless (Wg), in a neighboring parasegment (ps8), which in turn feeds back to ps7 to stimulate Ubx expression. Thus, Dpp is part of a "parautocrine" feedback loop of Ubx (i.e., an autocrine feedback loop based partly on paracrine action) that sustains its own expression through Dpp and Wg. Dpp also spreads to the inner layer of the embryonic midgut, the endoderm, where it synergizes with Wg to induce expression of the homeotic gene labial (lab). To achieve this, Dpp locally elevates the endodermal expression levels of Drosophila D-Fos with which it cooperates to induce lab. Differentiation of various cell types in the larval gut depends on these inductive effects of Dpp and Wg. A cAMP response element (CRE) from the Ubx midgut enhancer has been shown to be necessary and to some extent sufficient to mediate the Dpp response in the embryonic midgut (Eresh, 1997).

CREs are known to be signal-responsive elements, not only for cAMP signaling as described initially but also for other signals including ones acting through Ras. This prompted an investigation of whether any other signal may play a part in the Dpp response. This led to the discovery that the Drosophila Epidermal growth factor receptor (Egfr) has a critical function during endoderm induction. A secondary signal was discovered with a permissive role in this process, namely Vein, a neuregulin-like ligand that stimulates the epidermal growth factor receptor and Ras signaling. Dpp and Wg up-regulate vein expression in the midgut mesoderm in two regions overlapping the Dpp sources. This up-regulation depends on dpp and wg. Vein is thus a secondary signal of Dpp and Wg, and it stimulates homeotic gene expression in both cell layers of the midgut (Szuts, 1998).

Because loss-of-function mutants of the Drosophila Egfr are very abnormal and do not develop properly beyond the early embryonic stages, a temperature-sensitive allele of Egfr, flb1F26, was used to ask whether this receptor has any function in the embryonic midgut. flb1F26 embryos were stained with anti-Labial antibody after shifting the embryos from the permissive to the restrictive temperature at 6-8 hr of development (i.e., before midgut formation, but allowing normal germ-band retraction). The midguts of the homozygous flb1F26 embryos are severely abnormal, with none of the constrictions forming properly, and they show virtually no Lab staining in the midgut epithelium. These phenotypes indicate a critical function for Egfr in the embryonic midgut. Many endodermal cells were missing or seemingly unhealthy, especially in the middle midgut where lab is induced and in the anterior midgut near the gastric caeca. These two midgut regions correspond to the domains of Dpp expression. Similar effects of Egfr loss of function on cell health have been observed in earlier studies of the embryonic epidermis. Although this putative function of Egfr in cell survival may contribute to the observed loss of lab induction, it is believed that it does not account for all aspects of the gut phenotypes attributable to Egfr loss of function (Szuts, 1998).

The activity of the minimal Ubx midgut enhancer (Ubx B) was examined after mesodermal expression of dominant negative Egfr (DN-DER). Ubx B normally mediates strong lacZ staining in a region spanning the middle midgut constriction (ps6-ps9) and also some staining in the gastric caeca (ps3); the strongest staining in ps7/ps8 spans the main Dpp and Wg sources in the middle midgut, whereas the ps3 staining coincides with the anterior source of Dpp. Mesodermal expression of DN-DER almost completely eliminates staining in ps3 and strongly reduces staining in the ps6/ps7 region. These results lend strong support to the notion that Egfr functions in the visceral mesoderm; they indicate that Egfr positively regulates Ubx expression (Szuts, 1998).

Two ligands are known that activate Egfr in the somatic cells of Drosophila: Spitz, which apparently needs to be processed to an active form by the membrane-spanning protein Rhomboid, and Vein. spitz and rhomboid loss-of-function mutants were examined by staining embryos with Lab antibody, but these mutants appear to have only a minor effect on Lab expression: typically, Lab staining is found to be missing in just a few cells in the lab domain, and the midgut constrictions are normal in these mutants. However, vein mutant embryos show a drastic effect on Lab expression. The most extreme mutant conditions caused nearly complete loss of Lab staining in the midgut; none of the midgut constrictions form, nor do the gastric caeca elongate. Milder mutant conditions have only sporadic effects in the midgut, since only some cells in the lab domain lack Lab expression; the constrictions and the gastric caeca form normally under these conditions. These results implicate Vein as a critical ligand of Egfr in the embryonic midgut (Szuts, 1998).

EGFR expression is thought to be fairly ubiquitous in the embryo. However, vein transcripts are found in a highly restricted pattern, primarily in the embryonic mesoderm. In the midgut too, vein expression is spatially regulated, as follows: vein transcripts in the midgut are restricted to the visceral mesoderm. Initially, during stage 13, low levels of vein expression are seen at intervals throughout the midgut mesoderm. However, soon after the formation of the midgut epithelium, vein transcripts start to accumulate locally, and two main domains of prominent vein expression develop, one in the anterior and one in the middle midgut. Anteriorly, vein expression spans approximately ps2-ps4 and is strongest around the ps3/ps4 junction, that is, posterior to the gastric caeca. In the middle midgut, there is a fairly wide band of low vein expression spanning approximately ps6-ps10, with strongly up-regulated expression levels throughout ps7 (and trailing into anterior ps8). Posterior ps7 becomes the most prominent site of vein expression in the midgut. Finally, a narrow band with low levels of vein transcripts is seen at the posterior end of the midgut. The two main expression domains of vein overlap the two domains of Dpp expression in the visceral mesoderm (in ps3 and ps7), but each of them is considerably wider than the corresponding dpp domain. vein expression in the visceral mesoderm is severely diminished in dpps4 mutants. The prominent band of vein expression in ps7 is no longer seen, and expression in ps4 is reduced too. Instead, the strongest expression of vein in these mutants is seen at a novel location, at the ps5/ps6 junction around the incipient first midgut constriction (this ps5/ps6 expression is higher than in the wild type, and can be used to identify young dpp mutant embryos). It is concluded that dpp is required for the localized up-regulation of vein expression in the midgut (Szuts, 1998).

vein expression is also strongly diminished in wg mutants. vein expression can still be seen at moderate levels in the ps4 region, but vein expression is barely visible elsewhere in the midgut of these mutants. In particular, there are only traces of vein expression in the ps7/ps8 region, and expression at both midgut ends is almost undetectable. Clearly, wg plays an essential role as well in up-regulating vein expression. dpp and wg are sufficient to position the two domains of vein up-regulation. High mesodermal Wg causes very strong vein expression in ps2-ps7, significantly stronger than that caused in this region by mesodermal Dpp expression alone. This indicates that wg cooperates with dpp in positioning vein up-regulation. It is shown that neither Dpp for Egfr signaling is particularly effective in the absence of the other. Thus these two pathways are functionally interdependent and that they synergize with each other, revealing functional intertwining (Szuts, 1998).

The mutant analysis suggests strongly that Vein is the main, if not the only, ligand that stimulates Egfr in the embryonic midgut. This contrasts with other tissues, mainly of ectodermal origin, in which Spitz is the main Egfr ligand. Interestingly, Vein also has a major role during an inductive process between muscle and epidermis: Vein is secreted from muscle cells and triggers differentiation of the receiving epidermal cells into tendon cells. These functions of Vein during inductive processes between different cell layers suggest that the molecular properties of Vein are particularly suited to such processes that require the signal to cross basal membranes. Similarly, the extensive mesodermal expression of Vein may mean that this signal protein is particularly well-adapted to its production in this cell layer. Note that Vein is similar to mammalian neuregulins that appear to function in developmental contexts that involve communication between different cell layers (Szutz, 1998 and references).

The transcriptional response elements for the Dpp signal in midgut enhancers from homeotic target genes are bipartite, comprising CRE sites as well as binding sites for the Dpp signal-transducing protein Mad. Of these sites, the CRE seems to function primarily in the response to Ras, the secondary signal of Dpp. It is also shown that the Dpp response element in the labial enhancer comprises CREs and Mad binding sites. The results with the labial enhancer confirm the conclusions derived from the Ubx enhancer, namely that the response element to Dpp signaling is bipartite and contains Mad binding sites as well as CREs. The latter are critical in both cell layers for the signal response, whereas the former seem less criticial in the endoderm than in the visceral mesoderm. Perhaps this reflects the fact that lab is the ultimate target gene of the endoderm induction and that its enhancer clearly integrates a number of distinct positional inputs, some of which may be partially redundant (Szuts, 1998).

Why should there be this secondary signal whose role is entirely permissive, namely to assist the primary signal in implementing its tasks? Two kinds of answers are proposed. The first one is based on the observation that lack of Vein/Egfr signaling in the midgut appears to make cells sick and perhaps causes them to die. Therefore, Vein/Egfr signaling may serve as a "survival signal." Intriguingly, cell survival in embryos lacking vein or Egfr function appear to be affected preferentially near the two Dpp sources (where vein expression is up-regulated). Perhaps high levels of Dpp signaling can cause cell death; if so, vein signaling may be up-regulated to counteract a putative local deleterious effect of Dpp. A precedent for such a scenario may be found in the developing chick limb bud where the cell death-inducing properties of BMP (a TGF-beta-like signal) seem to be antagonized locally by a signal triggering the Ras pathway. However, although antagonistic effects between Egfr- and TGF-beta-type signaling have been observed, the evidence provided here suggests strongly that Vein/Egfr and Dpp both act positively in the embryonic midgut of Drosophila. Furthermore, they synergize with each other in the transcriptional stimulation of target genes. This observed synergy parallels cooperation between Ras and TGF-beta signaling during epithelial tumor progression. It is therefore thought unlikely that Vein functions in the midgut entirely as a survival signal near Dpp sources (Szuts, 1998).

The second kind of answer builds on the observations that indicate functional interdependence and synergy between the two signaling pathways in stimulating transcription of target genes. This could be beneficial for developmental systems in two ways: (1) if cells need to be costimulated by cooperating primary and secondary signals, this would serve to sharpen their signal response. This putative sharpening effect may be a contributory factor in sharp responses to signaling thresholds such as those observed in the Xenopus embryo.(2) The need for costimulation would safeguard against fortuitous and random stimulation of cells by any one signal, thus improving the reliability of their signal response. And although a requirement for the secondary signal is observed throughout the functional realm of the primary signal, it is envisaged that the role of the secondary signal is particularly critical in remote cells where the distribution of the primary signal becomes shallow, imprecise, and unreliable. Therefore, the secondary signal may provide primarily "remote stimulation." Whatever the case, it seems very likely that the use of a functionally coupled primary-secondary signal system results in a refinement and stabilization of positional information and in a degree of precision of this information that could not be conferred by one signal alone. Functional intertwining of a secondary and a primary signal may represent a mechanistic solution of how morphogens such as Dpp and activins work. Perhaps, signaling pathways do not function on their own in eliciting multiple different cellular responses, as envisaged by the purest version of the morphogen concept (Szuts, 1998).

The endoderm of Drosophila is patterned during embryogenesis by an inductive cascade emanating from the adhering mesoderm. An immediate-early endodermal target gene of this induction is Dfos whose expression is upregulated in the middle midgut by Dpp signaling. Previous evidence based on a dominant-negative Dfos construct has indicated that Dfos may cooperate with Dpp signaling to induce the HOX gene labial, the ultimate target gene of the inductive cascade. Here, kayak mutants that lack Dfos were examined to establish that Dfos is indeed required for labial induction. Evidence is provided that Dfos acts on labial through a CRE (cyclic AMP response element)-like sequence, previously identified to be a target for signaling by Dpp and by the Epidermal growth factor receptor (Egfr) in the embryonic midgut. Dfos expression is stimulated by Egfr signaling. Dominant negative Egfr-expressing embryos were stained with antibody against Dfos. These embryos never show high levels of Dfos protein in the endodermal cells of the parasegment (ps) 6/7 region in the midgut as normally seen in the wild type. Finally, Dfos function is found to be required for its own upregulation. Thus, endoderm induction is based on at least four tiers of positive autoregulatory feedback loops (Szuts, 2000).

These results show that Dfos acts through a CRE-like sequence. It is possible that Dfos binds directly to this sequence, but needs a partner protein to do so, explaining why efficient binding in vitro of Dfos to this sequence could not be detected. In support of this, Dfos is found to be a highly context-dependent transcriptional activator that requires additional DNA-binding partners to stimulate transcription. Of note, a CRE reporter without any CRE context sequence does not confer any expression in transgenic embryos, again indicating the need for DNA-binding partners. Alternatively, and perhaps less likely, Dfos may act indirectly through the CRE-like sequence, by upregulating locally the expression of an unknown CRE-binding protein (Szuts, 2000).

Dfos expression is upregulated locally only in the inner cell layer of the midgut, the endoderm, but not in the outer cell layer, the visceral mesoderm. Consistent with this, the current results indicate that Dfos is only required in the former but not in the latter. Nevertheless, CRE- like sequences are present in the midgut enhancers of both Ubx (expressed in the visceral mesoderm) and labial (expressed in the endoderm), and function in both enhancers in the response to Dpp. Therefore, the stimulatory effects of Dpp signaling may be mediated by Dfos only in the endoderm, whereas an unknown protein may substitute for this role of Dfos in the visceral mesoderm. This explains perhaps why the minimal CRE reporters function robustly only in the endoderm but not in the visceral mesoderm, despite the fact that they are derived from the mesodermally expressed Ubx gene. For these reporters to function in the visceral mesoderm, they need additional context, one of these being the TCF binding site (Szuts, 2000).

The CRE-like sequence appears to be a target for Egfr signaling. Furthermore, the embryonic gut phenotype of kay1 mutants in the endoderm is similar to that due to loss of Egfr signaling. In particular, both mutant conditions seem to cause some degree of cell death in the midgut epithelium. Although this cell death may contribute to, it does not account completely for, the mutant phenotypes observed. Finally, it has been shown that Dfos upregulation in the ps6/7 region of the endoderm depends on Egfr function. Taken together, these observations indicate that Dfos, or its DNA-binding partner(s), may be a critical target of Egfr signaling during endoderm induction, and that the effects of Egfr signaling in the endoderm may be partly if not largely mediated by Dfos (Szuts, 2000).

Dfos is a context-dependent transcriptional activator whose function in the embryo requires a dimerization partner, such as Djun, as well as combined Jun N-terminal kinase (JNK) and Dpp signaling. This dimerization partner rather than Dfos itself is likely to be the target factor that is directly modified and activated by JNK signaling. In some embryonic tissues, for example the dorsal epidermal cells, this dimerization partner is probably Djun, a transcription factor known to be targeted directly by JNK and Rolled MAP kinase signaling. However, Djun is not a good candidate for a MAP kinase-activated dimerization partner of Dfos in the midgut since Djun neither detectably affects labial nor CRE-mediated expression in the midgut. This putative signal-activated dimerization partner of Dfos in the midgut thus remains elusive (Szuts, 2000).

The data indicate that Dfos autostimulates its own upregulation in the middle midgut. This parallels labial induction which is autoregulatory in the same endodermal region. The autoinductive function of Labial is probably provided by the low levels of Labial protein expressed in the endodermal primordia. Similarly, the low basal levels of Dfos expression in the endoderm may provide the autoregulatory function of Dfos in this tissue. Thus, the process of endoderm induction in Drosophila involves at least four tiers of positive autoregulatory feedback loops, two in the visceral mesoderm (Ubx and dpp) and two in the subjacent endoderm (Dfos and labial). In each case, the autoregulatory capacity of the transcription factors involved depends on simultaneous stimulation by extracellular signals (Dpp, Wingless or Vein). Thus, each of these positive feedback loops is conditional on cell communication. This design may serve to safeguard against fluctuation in the genetic activity of individual cells and may ensure the co-ordinated pursuit of a given developmental pathway within a group of cells (Szuts, 2000).

teashirt inhibits labial expression in the thorax (Röder, 1992).

Ultrabithorax and labial are a target of wingless signaling in the midgut. dishevelled, shaggy/zeste-white 3 and armadillo are required for transmission of the wingless signal in the Drosophila epidermis. These genes act in the same epistatic order in the embryonic midgut to transmit the wingless signal. In addition to mediating transcriptional stimulation of the homeotic genes Ultrabithorax and labial, they are also required for transcriptional repression of labial by high levels of wingless . Efficient labial expression thus only occurs within a window of intermediate wingless pathway activity. The shaggy/zeste-white 3 mutants reveal that wingless signaling can stimulate decapentaplegic transcription in the absence of Ultrabithorax, identifying decapentaplegic as a target gene of wingless. Since decapentaplegic itself is required for wingless expression in the midgut, this represents a positive feed-back loop between two cell groups signaling to each other to stimulate one another's signal production (Yu, 1996).

Cross-regulation of Homeotic Complex (Hox) genes by ectopic Hox proteins during the embryonic development of Drosophila was examined using Gal4 directed transcriptional regulation. The expression patterns of the endogenous Hox genes were analyzed to identify cross-regulation while ectopic expression patterns and timing were altered using different Gal4 drivers. Evidence is provided for tissue specific interactions between various Hox genes and the induction of endodermal labial (lab) by ectopically expressed Ultrabithorax outside the visceral mesoderm (VMS). Similarly, activation and repression of Hox genes in the VMS from outside tissues seems to be mediated by decapentaplegic gene activation. Additionally, it has been found that proboscipedia (pb) is activated in the epidermis by ectopically driven Sex combs reduced (Scr) and Deformed (Dfd); however, mesodermal pb expression is repressed by ectopic Scr in this tissue. Mutant analyses demonstrate that Scr and Dfd regulate pb in their normal domains of expression during embryogenesis. Ectopic Ultrabithorax and Abdominal-A repress only lab and Scr in the central nervous system (CNS) in a timing dependent manner; otherwise, overlapping expression in the CNS in tolerated. A summary of Hox gene cross-regulation by ectopically driven Hox proteins is tabulated for embryogenesis (Miller, 2001a).

The expression of the lab gene is regulated in a time and tissue specific manner by ectopic Ubx accumulation. Lab accumulation in the CNS and epidermis are differentially affected by ectopic Ubx. lab repression in the CNS only occurs in prd=>Ubx (prdGal4;UASUbx driver=>responder) animals even though both the 31-1=>Ubx and 69B=>Ubx demonstrate expression in this tissue. However, 31-1=>Ubx animals do not show significant accumulation of Ubx in the CNS prior to stage 9 while 69B=>Ubx genotypes do. This suggests that timing may be the critical factor in setting up regulatory interactions between these genes. If timing is the difference between the 31-1=>Ubx and 69B=>Ubx repression of lab in the CNS, then after stage 9, the locus becomes immune to Ubx's negative influence. This could be due to the absence of an important cofactor, the masking of a lab cis-regulatory site used by Ubx, or changes in the signaling environment (Miller, 2001a).

Driver expression levels in the nearby VMS may be a factor since the prd=>LacZ confocal images shows significantly more LacZ accumulation there. Expanded endodermal lab induction is demonstrated in all three driver=>Ubx combinations. Normally, endodermal lab expression is activated by Ubx in the VMS through a signaling mechanism involving dpp. However, only the prd=>LacZ combination shows significant VMS accumulation which indicates that the VMS may not be the only tissue contributing to this process. It is possible that the observed 31-1 driver expression in the endoderm could autonomously activate the Ubx/Dpp/Lab cascade and provide the observed expansion of Lab accumulation (Miller, 2001a).

Additionally, the sparse VMS accumulation seen using the 69B driver could reflect sufficient ectopic Ubx accumulation to activate the Dpp signal and subsequent lab induction in the endoderm. Consistent with this latter possibility is the fact that the dpp gene demonstrates auto-catalytic regulation in this tissue and could amplify low level stimulation by ectopic Ubx. The activation of Ubx in the VMS by 31-1=>Ubx and 69B=>Ubx, suggests that this signaling pathway is involved. Reduced first midgut constrictions are found in a few 69B=>Ubx and 31-1=>Ubx embryos, likely due to high levels of Dpp protein and subsequent Antp repression. Typically, however there is not enough ectopic Ubx expression in the VMS by any driver=>Ubx combination to repress the first midgut constriction. Despite this, expansion of the Lab endodermal domain is seen in these animals indicating that either the threshold response for lab induction is lower than that for Antp repression or that there is a signaling source other than the VMS for the Ubx generated signal. This signal could originate from the CNS for 31-1=>Ubx or the CNS, somatic mesoderm and epidermis for 69B=>Ubx. Additionally, the amnioserosa normally exhibits dpp expression where 31-1=>LacZ and 69B=>LacZ expression is observed. Gal4 activation of the Ubx responder in the amnioserosa could contribute to dpp activation levels and hence the pool of the diffusible agonist through stage 14 (Miller, 2001a).

Epidermal Hox interactions seem to fit the posterior dominance model best. That is, the observed effects on resident Hox gene expression caused by ectopic Hox protein accumulation usually exhibit repression of the more anteriorly expressed gene. Antp represents the predominant exception to this hierarchy. Ectopic Antp protein represses lab expression in epidermal cells but has no significant effect on pb, Dfd or Scr expression in this tissue. However, Antp normally restricts the posterior domain of Scr in the VMS in a manner that appears to be mediated through short-range signaling. Since there is such a clear effect on the most anteriorly expressed Hox gene lab, while the three Hox genes (expressed more posterior to lab yet anterior to Antp's domain) appear to be refractory to Antp's negative control, it would appear these indifferent Hox genes or some other factor is negating Antp's influence. However, a resolution of the underlying cause of this observation awaits further experimentation (Miller, 2001a).

Homeosis and Homeotic Complex (Hox) regulatory hierarchies have been evaluated in the somatic and visceral mesoderm. Both Hox control of signal transduction and cell autonomous regulation are critical for establishing normal Hox expression patterns and the specification of segmental identity and morphology. Novel regulatory interactions have been identified associated with the segmental register shift in Hox expression domains between the epidermis/somatic mesoderm and visceral mesoderm. A proposed mechanism for the gap between the expression domains of Sex combs reduced (Scr) and Antennapedia (Antp) in the visceral mesoderm is provided. Previously, Hox gene interactions have been shown to occur on multiple levels: direct cross-regulation, competition for binding sites at downstream targets and through indirect feedback involving signal transduction. Extrinsic specification of cell fate by signaling can be overridden by Hox protein expression in mesodermal cells and the term autonomic dominance is proposed for this phenomenon. The endoderm was used to monitor target gene regulation by the Hox proteins (specifically wg, dpp and lab) through signal transduction (Miller, 2001b).

Ectopic Hox protein expression in the mesoderm can induce lab, lab-LacZ and dpp-LacZ expression in the midgut. Typically, the anterior ectopic endodermal lab expression parallels the observed expression pattern in the visceral mesoderm. The lack of ectopic lab expression posterior to ps7 is probably due to the unaltered high levels of wg expression, that repress lab. Normal lab induction in the endoderm requires wg, dpp and vein; however, dFos dependent (wg independent) lab transcription can be accomplished with high Dpp levels. Typically, lab induction by dpp, wg and vein is coordinated by sgg (GSK3) which may be responsible for the ps4 gap in lab, lab-LacZ, and expression patterns seen in experiments involving ectopic Antp visceral mesoderm expression. Moreover, the lack of expanded lab-LacZ expression (unlike native lab) by ectopic Antp indicates the existence of presently undefined cis-regulatory elements at the lab locus that are not contained in genomic fragments of the identified enhancers. Antp protein may be regulating other influential signaling pathways while the corresponding cis-acting elements are not located in the genomic lab enhancers tested. Antp expression is functionally linked to another TGF-beta agonist 60A (glass bottom boat), as well as the Wnt pathway agonist DWnt4 (Miller, 2001b).

It is concluded that Hox gene interactions in the mesoderm are not always consistent with previous governing hierarchies: posterior dominance and phenotypic suppression. In the visceral mesoderm it is found that posterior dominance (Hox direct cross-regulation) seems legitimate but may be mediated by signal transduction. Phenotypic suppression is violated by morphological changes and target gene regulation. In the somatic mesoderm, more anterior Hox genes alter the identity of the ventral T2 segment, but this tissue is largely extrinsically regulated in the absence of direct Hox expression. In light of this result, the notion of autonomic dominance is proposed: Hox genes cell-autonomously dominate tissues regulated by signal transduction (Miller, 2001b).

The predominant paradigm depends on whether cells are extrinsically or autonomously specified by Hox gene expression. It is argued that non-typical homeosis caused by ectopically expressed Hox proteins (i.e. not following the dictates of posterior prevalence) can be taken to indicate inductively specified tissues and hence, confer autonomic dominance. Interestingly, ectopic expression of the Hox proteins also exhibit non-typical homeosis in the chordotonal organs of the PNS and the thoracic cuticle, suggesting that inductive specification and autonomic dominance may not be restricted to the mesoderm. However, Hox regulatory hierarchies seem to be of limited value in other tissues as well. The mechanism responsible for autonomic dominance has not been determined in this study; only the correlation between autonomous Hox dominance over inductively specified tissue. Signal transduction pathway cross-talk could be the predominant cause of autonomic dominance phenotypes (homeosis) due to Hox regulation of signaling agonists. These agonists could then contribute to the signaling environment to alter the tissue, since these morphogens are potent factors in differentiation. Meanwhile, Hox genes cross-regulate each other cell autonomously and in nearby tissues through signal transduction. This occurs in a tissue specific manner that likely depends on both the signaling environment, transcriptional co-factors, and perhaps any of an estimated 100 target genes for a given Hox protein. The signaling environment of any given tissue is dictated primarily by Hox genes, which is critical for maintenance of Hox expression domains and subsequent differentiation, determination and morphogenesis. This complex set of intrinsic and extrinsic Hox controls are likely responsible for the means by which Hox genes were genetically identified for their abilities to dominate segmental identities as homeotic selector genes (Miller, 2001b).

Targets of Activity

ventral veins lacking is required for specification of the tritocerebrum in embryonic brain development of Drosophila: vvl acts genetically downstream of lab in the specification of the tritocerebral neuromere

The homeotic or Hox genes encode a network of conserved transcription factors which provide axial positional information and control segment morphology in development and evolution. During embryonic brain development of Drosophila, the Hox gene labial (lab) is essential for tritocerebral neuromere specification; lab loss of function results in tritocerebral cells that fail to adopt a neuronal identity, causing axonal pathfinding defects. Evidence is presented that the POU-homeodomain DNA-binding protein ventral veins lacking (vvl) acts genetically downstream of lab in the specification of the tritocerebral neuromere. In the embryonic brain, vvl expression is seen in all brain neuromeres, including the tritocerebral lab domain. Lab mutant analysis shows that vvl expression in the tritocerebrum is dependent on lab activity. Loss-of-function analysis focussed on the tritocerebrum reveals that inactivation of vvl results in patterning defects which are comparable to the brain phenotype caused by null mutation of lab. In the absence of vvl, mutant tritocerebral cells are generated and positioned correctly, but these cells fail to express neuronal markers. This indicates defects in neuronal differentiation. Moreover, longitudinal axon pathways in the tritocerebrum are severely reduced or absent and the tritocerebral commissure is missing in the vvl mutant brain. Genetic rescue experiments show that vvl is able to partially replace lab in the specification of the tritocerebral neuromere. These results indicate that vvl acts downstream of the Hox gene lab and regulates specific aspects of neuronal differentiation within the tritocerebral neuromere during embryonic brain development of Drosophila (Meier, 2006).

vvl is expressed in the embryonic brain from the extended germ band stage onwards. For an analysis of the protein distribution pattern of vvl in the embryonic brain, immunocytochemical experiments with a polyclonal antibody against the Vvl protein were carried out in combination with an anti-HRP antibody; anti-HRP immunoreactivity revealed the entire neural lineage of the developing CNS excluding the glial lineage. At late stage 11, vvl expression is first detected in few neuroblasts of the developing brain anlage, and by stage 13 becomes abundant in neuroblasts and their progeny within each brain neuromere. By stage 15, when neural progeny were generated and axonal projections are formed, Vvl protein is observed in specific cell clusters within all brain neuromeres. Accordingly, a prominent expression domain was also observed in the developing tritocerebrum (Meier, 2006).

Expression of vvl in the tritocerebrum suggested a possible overlap with the expression of the Hox gene lab. To investigate this, double-immunocytochemical experiments were carried out either on transgenic flies expressing a lab-lacZ reporter construct in which antibodies against Vvl were used together with anti-βgal antibodies, or on wildtype embryos in which antibodies against Vvl were used together with anti-Lab antibodies. These experiments revealed that the majority of cells expressing vvl in the tritocerebrum are located within the lab expression domain. Together with the observation that vvl appears to be differentially regulated by lab, the co-expression of lab and vvl in the tritocerebrum suggested that vvl activity might be lab-dependent in this neuromere. To study this further, whether mutational inactivation of lab affects vvl expression in the tritocerebrum was examined (Meier, 2006).

Mutational inactivation of lab results in regionalized axonal patterning defects which are due to both cell-autonomous and cell-nonautonomous effects. Thus, in the absence of lab, mutant cells are generated and positioned correctly in the brain, but these cells do not extend axons. Additionally, extending axons of neighboring wildtype neurons stop at the mutant domains or project ectopically, resulting in the disruption of the longitudinal connectives and a lack of the tritocerebral commissure. To characterize vvl expression in a lab-/- background, double-immunocytochemical experiments were carried out on homozygous lab null mutant embryos using antibodies against Vvl and HRP. These experiments revealed that vvl immunoreactivity is lacking in the tritocerebral lab mutant domain, in addition to the expected lack of anti-HRP immunoreactivity despite the persistence of cells in this region. This suggested that vvl expression in the posterior tritocerebrum is affected by loss of lab function during late stages of embryonic brain development, indicating that vvl expression in the tritocerebrum is lab-dependent (Meier, 2006).

To assess the functional role of vvl in tritocerebral neuromere formation, vvl null mutants were analyzed using immunocytochemical markers including anti-HRP, anti-ELAV, anti-REPO, and anti FASII, which label general neuronal (or glial) domains and tracts in the developing embryonic brain. In vvl loss-of-function mutants, a pronounced brain phenotype is observed in the late stage embryonic brain. Immunolabelling with neuron-specific anti-HRP and anti-FASII antibodies identified a gap separating the deutocerebral brain region from the neuromeres of the more posterior subesophageal ganglion. This dramatic phenotype is associated with severe axonal patterning defects in the embryonic brain. The longitudinal connectives that normally run from the deutocerebral and tritocerebral neuromeres to the subesophageal ganglion were severely reduced or missing and the tritocerebral commissure, which interconnects the brain hemispheres at the level of the tritocerebrum, was completely absent. Analysis of FASII immunoreactivity revealed that descending and ascending axons which in the wildtype normally project through the tritocerebrum in well formed fascicles, fail to project through this domain in vvl mutants. Moreover, a loss of anti-ELAV immunolabelling is observed in the tritocerebral domain, whereas glia-specific anti-REPO immunoreactivity revealed that glial cells are present in the vvl mutant but fail to be correctly localized in the affected region. In addition to the observed defects in the tritocerebral brain region, marked axonal patterning defects in the protocerebrum are also seen in vvl mutant embryos. Moreover the organization of the subesophageal ganglion and the VNC is affected in the vvl mutant. These latter two phenomena were not studied further (Meier, 2006).

At the gross histological level, the vvl mutant brain phenotype described above was, in part, reminiscent of the mutant brain phenotype observed for lab. Since lab and vvl show overlapping expression in the tritocerebral neuromere, and vvl expression is lacking in lab mutants, these findings suggest that either lab expression itself or lab expressing tritocerebral cells are affected in vvl mutant embryos. To investigate this, anti-Lab immunolabelling was carried out in late vvl loss-of-function mutant brains. Surprisingly, despite the expected lack of expression of neuronal differentiation markers, anti-Lab immunolabelling was detected in a wildtype-like pattern in the vvl mutant tritocerebral domain. This suggests that the expression of lab is not affected in the absence of vvl during late stages of embryonic brain development. Moreover, the lab expressing cells in the vvl mutant generally have the same relative position in the brain as does the normal lab expressing cells in the wildtype. Thus, despite the severe axonal patterning defects observed in this domain, mutant cells are generated and appear to be properly positioned in the developing tritocerebrum of the vvl null mutant. This, in turn, suggests that the pattern of proliferation in the tritocerebrum is initiated correctly in the absence of the vvl gene product, but that the cells that normally express vvl might become incorrectly specified in the vvl mutant leading to axogenesis defects. Moreover, the lack of anti-ELAV and anti-HRP immunolabelling together with the observed severe fasciculation defects in the vvl mutant tritocerebrum strongly suggest that mutational inactivation of vvl affects neuronal differentiation in the developing tritocerebrum (Meier, 2006).

These data imply that vvl might be involved in the specification of tritocerebral neuronal identity -- either by acting directly or indirectly downstream of tritocerebral lab activity. To further assess a possible lab-dependent vvl activity in the developing tritocerebrum, the potential of vvl to rescue the lab mutant brain phenotype was determined using the Gal4-UAS system. For this, a transgenic fly line carrying a Gal4 transcriptional activator under the control of the lab promoter together with CNS-specific upstream enhancer elements of the lab gene were used. By crossing this lab::Gal4 line to different UAS-responders it was possible to express the responder constructs in a pattern that corresponded to that of the endogenous lab gene. Using this approach, it has been shown that the lab mutant brain phenotype can be rescued by transgenic expression of the Lab protein in a lab null mutant background. To determine whether vvl might also be able, at least in part, to rescue the lab mutant brain defects, a transgenic UAS::vvl line was used in which the vvl coding sequence was placed under UAS control (Meier, 2006).

As a control, it was first determined whether lab::Gal4 driven misexpression of vvl in a lab+ background had any effects on the development and specification of the tritocerebral Lab domain. In none of these experiments were morphological abnormalities detected in the tritocerebrum or in any other part of the embryonic brain. The UAS::vvl responder was expressed under the control of the lab::Gal4 driver in the lab mutant domain. Remarkably, the Vvl protein was able to rescue specific lab mutant brain defects. Thus, the longitudinal pathways were restored and cells in the mutant domain showed wildtype-like anti-HRP immunolabelling. Moreover, FASII immunoreactivity revealed that descending and ascending axons from other parts of the brain again projected through the tritocerebral lab mutant domain. The vvl responder achieved a rescue efficiency (97.3%) which was comparable to the rescue efficiency of Lab, which was taken as 100%. In contrast, lab::Gal4-specific expression of vvl in the lab mutant domain did not rescue tritocerebral commissure formation, nor correct axonal projection of the frontal connective. This suggested that lab::Gal4 driven misexpression of vvl is sufficient to restore both neuronal marker gene expression like HRP and correct axonal patterning of longitudinal connectives in the lab mutant tritocerebrum. These findings together with the vvl mutant brain phenotype indicate that vvl acts genetically downstream of lab in the specification of the triocerebral neuromere (Meier, 2006).

Taken together, these findings demonstrate that vvl function is required for the specification of the developing tritocerebrum. The vvl gene is important for correct axon guidance and fasciculation of longitudinal connectives in the tritocerebral neuromere. In the absence of vvl, longitudinal and commissural axon pathways are severely affected. Comparable findings have been reported for the role of vvl in VNC development, where vvl mutant embryos exhibit aberrantly localized midline glia and axonal defects in that commissures are often fused and the longitudinal connectives are severly reduced or even disrupted. These findings suggest that vvl acts genetically downstream of the Hox gene lab in the control of regionalized neuronal identity and tritocerebral brain neuromere specification. In accordance with this notion is the successive time of lab and vvl expression in the tritocerebral neuromere. From stage nine onwards, lab expression commences in the intercalary segment and by early stage 11, lab is detected in all neuroblasts of the developing tritocerebrum. Accordingly, by late stage 11, vvl expression is first seen in the tritocerebral neuromere and thus succeeds initial lab expression. Interestingly, this time of initial vvl expression exactly coincides with the temporal requirement of lab for tritocerebral neuronal fate specification. Moreover, the results demonstrate that tritocerebral vvl expression is lab-dependent. In addition, in vvl mutants, lab is expressed normally in the tritocerebrum and yet cells in the affected tritocerebral domain phenocopy the lab mutant brain and do not express molecular markers characteristic of neuronal cells. Furthermore, the vvl gene can mediate neuronal specification and longitudinal connective formation in the absence of the Hox gene lab if expressed under appropriate spatiotemporal control. This indicates that vvl is required for the specification of neuronal identity in the tritocerebral lab domain and is sufficient to provide a permissive substrate for the migration of axons originating from outside this region. However, vvl cannot rescue tritocerebral commissure formation in the lab mutant brain. This indicates that lab exerts at least some of its effects on tritocerebral development through subordinate genes other than vvl (Meier, 2006).

Hox proteins coordinate peripodial decapentaplegic expression to direct adult head morphogenesis in Drosophila

The Drosophila BMP, decapentaplegic (dpp), controls morphogenesis of the ventral adult head through expression limited to the lateral peripodial epithelium (P e) of the eye-antennal disc by a 3.5 kb enhancer in the 5' end of the gene. A 15 bp deletion mutation within this enhancer was recovered that identified a homeotic (Hox) response element that is a direct target of labial and the homeotic cofactors homothorax and extradenticle. Expression of labial and homothorax are required for dpp expression in the peripodial epithelium, while the Hox gene Deformed represses labial in this location, thus limiting its expression and indirectly that of dpp to the lateral side of the disc. The expression of these homeodomain genes is in turn regulated by the dpp pathway, as dpp signalling is required for labial expression but represses homothorax. This Hox-BMP regulatory network is limited to the peripodial epithelium of the eye-antennal disc, yet is crucial to the morphogenesis of the head, which fate maps suggest arises primarily from the disc proper, not the peripodial epithelium. Thus Hox/BMP interactions in the peripodial epithelium of the eye-antennal disc contribute inductively to the shape of the external form of the adult Drosophila head (Stultz, 2012).

dpp expression in the lateral PE of the eye-antennal disc is necessary for correct morphogenesis of the adult Drosophila head. This study shows that dpp expression related to ventral head formation is part of a Hox/BMP genetic network restricted to the PE of the eye-antennal disc. The homeotic gene lab, and its cofactors hth and exd positively regulate PE dpp expression. This is supported by the observation that Lab, Exd and Hth bind in vitro to the dpphc enhancer and the consensus sites for these factors are required in vivo for expression. In addition, individually, lab and hth are both genetically necessary and together demonstrate sufficiency for expression from dpphc enhancer, as shown from both LOF and GOF clonal analyses. lab exerts positive control over Dfd expression, as indicated by loss of Dfd expression in lab LOF clones. In contrast, Lab is ectopically expressed in Dfd LOF clones, demonstrating that Dfd represses lab in domains of its expression. Dpp signalling is genetically required for the transcription of lab, as expression from a lab reporter construct is reduced in tkv LOF clones. Expression of a hth enhancer trap increases in tkv LOF clones, and is reduced when activated Tkv is ectopically expressed, indicating that hth transcription is negatively regulated by Dpp signalling. Finally, dpp directly autoregulates its own expression (Stultz, 2006), and may be spatially limited to domains of signalling by repression by brk, as demonstrated by the ability of ectopically expressed Brk to repress expression from the dpphc enhancer. Lab and Hth (acting with Exd) activate the expression of dpp. Lab also contributes to the activation of Dfd, which when expressed, represses lab, acting as a switch to limit the extent of lab expression. It is envisioned that during disc development, lab initiates both dpp and Dfd, and when Dfd reaches a certain threshold level, it turns off lab, establishing the boundary between the two Hox proteins. However, while loss of Dfd is capable of derepressing lab throughout the disc, it does not do so to dpp, so further negative regulation must exist. brk may provide this repression to further ensure the lateral boundary of PE dpp through a potential AE element in the enhancer. These inputs collaborate to define the sharp boundary of PE dpp expression. The level of dpp transcription is positively modulated by feedback between lab and dpp and autoregulation of dpp, presumably through Mad/Med binding to the AE element. Negative feedback between dpp and hth provides a brake on expression; others may exist. For example, the inhibitory Smad protein, daughters against dpp is a target of peripodial Dpp expression (Stultz, 2006). It is presumed these interactions activate dpp expression rapidly but shut it down when a certain expression level is reached (Stultz, 2012).

The Hox response region represents one of what will likely be many inputs into the expression of this 3.5 kb enhancer. Another input, opa, is homologous to the Zinc Finger Protein of the Cerebellum or Zic family of transcription factors, and was identified due to its genetic interaction with dpps-hc mutations. Other transcription factors and signalling pathways display genetic interactions, and their contribution to PE dpp expression is being actively investigated, although it is noteworthy that lab, Dfd, hth, and exd are not among them. It is expected that many transcription factors and signalling pathways impinge on the dpphc enhancer. In this regard, the dpphc enhancer may resemble the dpp visceral mesoderm enhancer, another identified Hox target, where direct Ubx, Abdominal A, Exd, and Hth homeodomain inputs collaborate with the Fox-F-related factor binou, as well as Dpp and Wingless signalling to control gene expression. Enhancers that respond to signalling pathways often demonstrate characteristic behaviours: 'activator insufficiency', 'cooperative activation', and 'default repression', and the dpphc enhancer conforms to this model. No single activator is able to induce expression over the entire disc, as shown by GOF experiments. Ectopically expressed Lab produced activation only in close proximity to the domain of endogenous dpp, while Hth activated only in the PE of the posterior eye disc. Addition of two inputs together (Lab and Hth or Lab and Dpp signalling, activated over a much broader area. Only Opa has broad ability to activate on its own over the PE but only in concert with Lab was it able to activate outside the PE. Thus each activator is insufficient individually; the enhancer requires simultaneous cooperative inputs of multiple factors to produce correct spatial expression. Brk would provide the default repression, preventing Lab and Hth individually from successfully activating in the middle of the disc, away from domains of dpp activity (Stultz, 2012).

Based on the transcriptional inputs so far identified, it is proposed that activation is controlled on the lateral side at a minimum by Lab, Hth, Exd, Mad, Med, and Opa. In the middle of the disc, the presence of only Hth and Exd is insufficient to activate the enhancer, particularly over resident default repression provided by Brk. On the medial (future dorsal) side of the disc, Dpp and phosphorylated Mad expression are observed, controlled by an unknown area of the dpp gene. Lab, Hth, Exd, and Opa are expressed there as well, so an additional repressor was hypothesised to be needed that limits expression driven by the dpphc enhancer to the lateral side. In this model, Lab is the activity required for peripodial specificity, with its cofactors Hth and Exd, while Mad/Med and Opa act as necessary collaborative activators of the enhancer (Stultz, 2012).

At the nucleotide level, the Hox response element in and adjacent to the dpps-hc1 deficiency bears sequence homology to previously identified Lab response elements: the mouse Hoxb1 autoregulatory enhancer (b1-ARE), which also generates a lab-like pattern, dependent on lab and exd activity, in Drosophila, and the lab autoregulatory enhancer. Both these enhancers have binding sites for Hox (Hoxb1, Lab), PBC (Pbx, Exd) and MEIS (Prep, Hth) proteins. The orientation of the bipartite Exd/Lab site relative to the MEIS site is the same in these elements as seen in the dpphc Hox response element, and the relative spacing between the PBC/Hox and MEIS components is very similar. However, the dpphc Hox response element has a cluster of three overlapping Hth sites, two residing on the opposite strand, and an additional functional Exd site downstream of the Hth sites, as determined by its requirement for expression in vivo). The expression of mutated reporter constructs in vivo, as well as LOF analyses of lab and hth, indicate that Hth/Exd plays a more critical role in enhancer activity than does Lab, as neither mutations in the Lab binding site nor Lab loss-of-function within somatic clones completely extinguished expression. This suggests that there may be multiple ways that homeodomain transcription factors activate the enhancer, depending on the cellular context. It is noted that the expression driven by the dpphc enhancer actually manifests as two separate lines {see also Stultz, 2006b). The level of Lab associated with each of these lines is not equivalent, therefore the control of expression may be specific to each line. This would be reminiscent of a situation seen within dpp itself, where the Ubx responsive visceral mesoderm enhancer is activated by Ubx/Exd/Hth in parasegment seven, but only requires Hth/Exd for activation in parasegment three. The in vitro EMSA data further support this, as Hth and Exd bind synergistically to more locations within the enhancer than Lab. The TALE family homeodomain proteins function independently of Hox proteins in many contexts. An additional explanation for the apparent primacy of hth may be because it plays both direct and indirect roles on enhancer expression. Hth acts with the transcription factor Yorkie (Yki) as part of the Hippo signalling pathway, and the nuclear activity of Yki and Hth are required to specify the PE of the eye-antennal disc. In the absence of hth, the PE is incorrectly fated. This may effect early gene expression upstream of the Hox/BMP interactions described in this study, magnifying the genetic affect of hth (Stultz, 2012).

The Hox/BMP network described in this study plays a prominent role in the external appearance of the adult head, yet is restricted completely to the PE of the eye-antennal disc. The terminal mutant phenotypes of dpps-hc, Dfd, and lab have similarities, but are sufficiently distinct that additional targets for each must exist, and for the cell autonomous Dfd and lab, these targets must reside in the PE. Other signalling proteins such as Wingless and Hedgehog, and the Notch pathway ligands Serrate and Delta, are expressed in the PE of the eye-antennal disc. While some adult structures derive from the PE, and PE cells likely contribute to other adult structures, it is likely that much of the effect of the PE on head morphogenesis is via inductive interactions with the DP, either through secreted signalling molecules, or targeted cell protrusions. Based on the cuticular alterations seen in dpps-hc, Dfd, and lab mutations, such interactions are capable of exerting structural modifications on the final head shape. Dipterans demonstrate great variety in the external morphology of their heads often with sexually dimorphic alterations within a species. Much of this variety involves changes in the relative proportions of eye and head capsule tissue. BMP expression has been implicated in shaping the jaws of cichlid fish and the beak shape of finches, while dpp expression itself is correlated with the growth of beetle horns, a specialized cuticular structure of the head. It is speculated that the PE specific Hox/BMP network described in this study could be a motor for such types of shape change in the Drosophila species (Stultz, 2012).

Protein interactions and phenotypic suppression

Continued: Regulation part 3/3 | back to part 1/3

labial: Biological Overview | Evolutionary Homologs | Developmental Biology | Effects of Mutation | References

Home page: The Interactive Fly © 1997 Thomas B. Brody, Ph.D.

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