wingless

Table of contents

Wingless, mesoderm and myogenesis

How does Drosophila mesoderm become subdivided? The process may be illustrated by Bagpipe expression, which is restricted to metameric clusters of cells in the dorsal mesoderm. Under the control of bap, cell clusters develop into midgut visceral mesoderm, whereas cells in segmental portions lacking bap form other mesodermal derivatives. Among the segment polarity genes, both hedgehog and engrailed are required for full activation of bap. These results suggest that ectodermal hh and en participate in the establishment of the mesodermal posterior (P) domains opposite the posterior domains of the ectoderm. Ectodermal Wingless is synthesized adjacent to the anterior (A) domains. Wingless appears to act negatively on bap and serpent, because bap and srp expression is expanded in wg mutant embryos. Thus it appears that ectodermal WG and HH have opposing roles in establishing mesodermal A and P domains (Azpiazu, 1996).

Sloppy paired (Slp) and Even-skipped are involved in cell fate determination and segmentation in the Drosophila mesoderm. The primordia for heart, fat body, and visceral and somatic muscles arise in specific areas of each segment in the Drosophila mesoderm. The primordium of the somatic muscles, which expresses high levels of twist, a crucial factor of somatic muscle determination, is lost in sloppy-paired mutants. The effect of slp on Twist levels is probably partly, but not completely mediated by wg. wg mutant embryos show a premature and ectopic decay of Twist, but not to the same degree as seen in slp embryos. Whereas patches of cells expressing high levels of Twist are initially established in wg mutant embryos, no Twist is seen in the trunk region of slp mutant embryos after stage 11. At the same time that twist expression is lost in slp mutants, the primordium of the visceral muscles is expanded (Riechmann, 1997).

bagpipe and serpent expressing mesodermal domains corresponding to the ectodermal even-skipped domains, alternate with the sloppy-paired expressing high-twist mesodermal domains. Ectodermal even-skipped is thought to act through engrailed and subsequently hedgehog to promote bagpipe expression in cardiac and dorsal muscle and serpent in the fat body (Azpiazu, 1996). Ectodermal Dpp is required for the maintenance of mesodermal tinman, which in turn activates bap expression in the eve domain. The visceral muscle and fat body primordia require even-skipped for their development and the mesoderm is thought to be unsegmented in even-skipped mutants. However, it has been found that even-skipped mutants retain the segmental modulation of the expression of twist. Both the domain of even-skipped function and the level of twist expression are regulated by sloppy-paired, and eve serves reciprocally to regulate the slp domain. sloppy-paired thus controls segmental allocation of mesodermal cells to different fates (Riechmann, 1997).

The Drosophila visceral mesoderm (VM) is a favorite system for studying the regulation of target genes by Hox proteins. The VM is formed by cells from only the anterior subdivision of each mesodermal parasegment (PS). The VM itself acquires modular anterior-posterior subdivisions similar to those found in the ectoderm. Mesodemal cells located just under the engrailed-expressing cells in the posterior ectodermal compartment have been called the mesodermal "P domain." The dorsal-most cells of the mesodermal P domain in each PS express the homeobox gene bagpipe (bap); they detach from the mesodermal fold and move inward toward the center of the embryo. These bap-expressing cells form the VM progenitor groups. The VM cells initiate expression of Fasciclin III (FasIII) as they migrate to join each other and form a continuous band of VM running along each side of the embryo. Thus all the VM derive from the posterior parts of the initial mesoderm metameres. As VM progenitors merge to form a continuous band running anterior to posterior along the embryo, expression of connectin (con) occurs in 11 metameric patches within the VM, revealing VM subdivisions analogous to ectodermal compartments (Bilder, 1998b).

The VM subdivisions, and the metameric expression of con, form in response to ectodermal production of secreted signals encoded by the segment polarity genes hedgehog (hh) and wingless (wg) and are independent of Hox gene activity. A cascade of induction from ectoderm to mesoderm to endoderm thus subdivides the gut tissues along the A-P axis. Induction of VM subdivisions may converge with Hox-mediated information to refine spatial patterning in the VM. Con patches align with ectodermal engrailed stripes, so the VM subdivisions correspond to PS 2-12 boundaries in the VM. The PS boundaries demarcated by Con in the VM can be used to map expression domains of Hox genes and their targets with high resolution. The resultant map suggests a model for the origins of VM-specific Hox expression in which Hox domains clonally inherited from blastoderm ancestors are modified by diffusible signals acting on VM-specific enhancers (Bilder, 1998b).

Since Con expression marks the imprint of ectodermal PS boundaries on the VM, Con patches can be used to precisely map the domains of Hox gene transcription in relation to Con patches. teashirt is expressed in two domains. The anterior midgut domain extends from visceral mesoderm segment (VS) 4 to mid-VS 6, where it shares a posterior boundary with Antennapedia; the central midgut domain extends several cells to either side of the VS 8 boundary. dpp is also expressed in two domains: at the gastric caeca, it is found in the A domain of VS2 and the P domain of VS 3, while in the central midgut it extends from the A domain of VS 6 to terminate just anterior to the VS 8 boundary. wg is expressed just anterior to the VS 8 boundary, with some cells after stage 12 lying in VS 8. pnt is expressed throughout VS 8, although expression is not seen until early stage 13. At stage 13, the two domains of odd paired (opa) expression extend from the P domain of VS 4 to the VS 6 boundary and from VS 9 through VS 11 (Bilder, 1998b).

Several Hox targets appear to respect the PS subdivision organization of the VM. The initial VM expression of opa is seen only adjacent to Con patches, in A domains of VS 3-5 and 8-11. Similarly, wg is limited to a subset of abdA-expressing cells: those at the border of VS 8. wg is activated by abdA and dpp. Ectopic expression of abdA leads to induction of wg in a single posterior patch. Strikingly, the sites of ectopic wg induction in both genotypes align with the VS boundaries: in cells just anterior to VS 3, 5, and 6 in ectopic AbdA embryos and anterior to VS 9 in ectopic Dpp embryos. these results suggest that metameric subdivisions in the VM limit Hox gene activation of VM targets (such as wg) to restricted areas. It is suggested that divergent Hox expression in the VM has its basis in tissue-specific regulation of Hox expression in the VM and this expression is governed by unknown regulators that control VM-specific Hox enhancers (Bilder, 1998b).

The wg gene is required for the formation of a subset of muscle founder cells during embryogenesis. The wingless signal moves from ectoderm to provide an inductive signal for the initiation of the development of mesodermal derived muscle (Baylies, 1995).

A population of nautilus expressing cells, present in a medial position in ventral mesoderm, is dependent on wingless, expressed either in ectoderm or in mesoderm. Transgenic expression of wg solely in the ectoderm of wg mutants is sufficient to rescue nautilus expression in medial cells. Thus wingless function is required during and after gastrulation for the formation of nau-expressing medial muscle precursor cell clusters (Ranganayakulu, 1996).

even-skipped is expressed in heart precursor cells in the mesoderm, and is involved in the process of mesodermal segmentation. Expression of eve depends on Wingless, supplied either endogenously from mesodermal cells, or exogenously, from overlying ectodermal cells. even-skipped is expressed in clusters even when wingless is uniformly expressed, suggesting that Wingless is acting here in a permissive and not in an instructive role (Lawrence, 1995).

The Drosophila mef2 gene encodes a MADS domain transcription factor required for the differentiation of cardiac, somatic, and visceral muscles during embryogenesis and the patterning of adult indirect flight muscles assembled during metamorphosis. A prerequisite for Mef-2 function in myogenesis is its precise expression in multiple cell types. Novel enhancers for Mef-2 transcription in cardiac and adult muscle precursor cells have been identified and their regulation by the Tinman and Twist myogenic factors have been demonstrated. However, these results suggest the existence of additional regulators and provide limited information on the specification of progenitor cells for different muscle lineages. The heart enhancer has been further characterized and shown to be part of a complex regulatory region controlling the activation and repression of Mef-2 transcription in several cell types. The presence of two Tinman binding sites is necessary but not sufficient for enhancer function; additional sequences are required for cardial cell expression. The mutation of a GATA sequence in the enhancer changes its specificity from cardial to pericardial cells. Also, the addition of flanking sequences to the heart enhancer results in the expression of Mef-2 in a new cell type: the founder cells for a subset of body wall muscles. Since tinman function is required for Mef-2 expression in both the cardial and founder cells, these results define a shared regulatory DNA that functions in distinct lineages due to the combinatorial activity of Tinman and other factors that work through adjacent sequences. The forced mesodermal expression of Twist causes a repression of the enhancer element in founder cells while allowing normal function in cardial cells. The analysis of Mef-2-lacZ fusion genes in mutant embryos reveals that the specification of the muscle precursor cells involves the wingless gene. Wg is required both in the formation of specific founder cells and in the specification of the progenitors of the cardial cells. Ectodermal cells must have a ventral identity for the formation of founder cells. These results demonstrate that the cell fate status of ectodermal cells adjacent to the domain of ventral founder cell specification is crucial for the proper formation of these cells. Mesodermal cell fate also depends on the activation of a receptor tyrosine kinase signaling pathway. Ectopic expression of an activated form of Ras1 throughout the mesoderm results in a substantial overproduction of the ventral founder cells as compared to control embryos. This signal may be transduced through the mesodermally-active EGF or FGF receptor tyrosine kinases (Gajewski, 1998).

Expression of ladybird genes in the subset of cardioblast and pericardial cell precursors is critically dependent on mesodermal tinman function, epidermal Wingless signaling and the coordinate action of neurogenic genes. lb-expressing heart progenitors contribute to the increased number of cardiac precursor cells in Notch, Delta, Enhancer of split, mastermind, big brain and neuralized mutants. Negative regulation by hedgehog is required to restrict ladybird expression to two out of six cardioblasts in each hemisegment. Overexpression of ladybird causes a hyperplasia of heart precursors and alters the identity of even-skipped-positive pericardial cells. Surprisingly, the number of eve-expressing pericardial cells is strongly reduced in overexpressors. These lb expressing cells are transformed into l-paracardial cells. Loss of ladybird function leads to the opposite transformation, suggesting that ladybird participates in the determination of heart lineages and is required to specify the identities of subpopulations of heart cells. Both early Wingless signaling and ladybird-dependent late Wingless signaling are required for proper heart formation. Thus, it is proposed that ladybird plays a dual role in cardiogenesis: (1) during the early phase, it is involved in specification of a segmental subset of heart precursors as a component of the cardiogenic tinman-cascade and (2) during the late phase, it is needed for maintaining wingless activity and thereby sustaining the heart pattern process. These events result in a diversification of heart cell identities within each segment. Since tinman, bagpipe, S59 and ladybird genes are all part of the same homeobox gene cluster, it is likely that their association has to do with the orchestrated diversification of mesoderm (Jagla, 1997b).

Subsets of differentiating muscles in the Drosophila embryo express putative transcription factors, such as NK1/S59 and vestigial. These genes may control the development of specific muscle properties. Myogenesis in embryos mutant for wingless is grossly deranged. Mesodermal expression of S59 is lost, whereas some vestigial-expressing muscles do develop. wingless dependence and independence of specific muscle subsets correlates with an early derangement of twist expression in wingless mutants (Bate, 1993).

bagpipe expression in mesodermal tissue overlying foregut and hindgut, both considered to be ectodermal derivatives, is regulated by wingless and hedgehog activities in the underlying gut epithelium. The mesodermal layer of the fore- and hindgut is gradually assembled around the invaginating stomodeal and proctodeal tubes. bagpipe is strongly expressed in mesodermal cells on top of the proctodeum that will give rise later to the muscles of the hindgut. The expression domain then splits to give rise to two subdomains, one around the future small intestine and the other around the future rectum of the hindgut. Later bagpipe expression appears in a continuous expression domain. In both wg and hh mutants, bap expression is reduced or absent in the visceral mesoderm primordia of the developing hindgut. Similar results were obtained for the foregut (Hoch, 1996).

Ectodermal and mesodermal muscle segment homeobox expression depend on wingless and hedgehog. The intricate pattern of msh expression in segmentally arranged clusters during stages 10 and 11, is altered in segment polarity mutants. Mutation of hh affects the intermediate column of msh expressing clusters. In hh mutant embryos, ectodermal msh expression is absent at these positions and the mesodermal expression in fat body precursors is strongly reduced. In contrast, in mutants for wingless the intermediate clusters of msh are normal, whereas the dorsal clusters, both from ectoderm and the mesoderm are completely absent. As a consequence, later stage embryos lack msh expression both in dorsal muscles and around chordotonal organs (D'Alessio, 1996).

Heartless is a mesoderm-specific fibroblast growth factor receptor in Drosophila. A protein-null mutant of heartless and heartless expression was examined using anti-Heartless antibody. After invagination, mesodermal cells expressing heartless undergo proliferation and spread out dorsally to form a monolayer beneath the ectoderm. In mutant embryos, however, the mesoderm is not capable of extending to the normal dorsal limit and consequently mesodermal cells fail to receive ectodermal signals and are thus rendered incapable of differentiating into primordia for the heart, visceral and somatic muscles. Thus there appears to be a specific defect in cell migration in heartless mutants. In heartless mutants the intensity of tinman expression decreases inappropriately and is no longer expressed in presumptive heart cells, although visceral mesodermal expression in maintained. In a large fraction of presumptive somatic muscle precursors, Htl expression is wingless dependent. wingless is expressed in the mesoderm for only a short period after gastrulation, suggesting that the second phase of Htl expression may be controlled by Wingless signals from ectodermal cells (Shishido, 1997).

Inactivation of either the secreted protein Wingless (Wg) or the forkhead domain transcription factor Sloppy Paired (Slp) has been shown to produce similar effects in the developing Drosophila embryo. In the ectoderm, both gene products are required for the formation of the segmental portions marked by naked cuticle. In the mesoderm, Wg and Slp activities are crucial for the suppression of bagpipe (bap), and hence visceral mesoderm formation, and the promotion of somatic muscle and heart formation within the anterior portion of each parasegment. During these developmental processes, wg and slp act in a common pathway in which slp serves as a direct target of Wg signals that mediates Wg effects in both germ layers. Evidence has been found that the induction of slp by Wg involves binding of the Wg effector Pangolin (Drosophila Lef-1/TCF) to multiple binding sites within a Wg-responsive enhancer, located in 5' flanking regions of the slp1 gene. Based upon genetic and molecular analysis, it is concluded that Wg signaling induces striped expression of Slp in the mesoderm. Mesodermal Slp is then sufficient to abrogate the induction of bagpipe by Dpp/Tinman, which explains the periodic arrangement of trunk visceral mesoderm primordia in wild type embryos. Conversely, mesodermal Slp is positively required, although not sufficient, for the specification of somatic muscle and heart progenitors. It is proposed that Wg-induced slp provides striped mesodermal domains with the competence to respond to subsequent slp-independent Wg signals that induce somatic muscle and heart progenitors. It is also proposed that in wg-expressing ectodermal cells, slp is an integral component in an autocrine feedback loop of Wg signaling (Lee, 2000).

Since the progenitors of pericardial cells, cardioblasts and somatic muscles are largely derived from the segmental areas between the bap domains, and their formation requires wg and slp as positive activities, it was asked whether in this regulatory pathway slp also acts strictly downstream of wg. In wild-type embryos, eve-expressing progenitors of pericardial cells and dorsal muscles are located underneath the ectodermal wg stripes and within the domains of mesodermal slp expression. In wg mutant embryos, these progenitors are not formed. Interestingly, while ectopic slp expression in the mesoderm of wg mutants rescues bap repression, it fails to rescue the formation of heart and dorsal muscle progenitors. Identical effects are observed in slp mutants and in slp mutant embryos with ectopic mesodermal expression of wg; neither produces any eve-expressing heart and muscle progenitors. The results of ectopic expression of slp or wg in wild-type backgrounds are also consistent with these observations. Although ectopic expression of slp in the mesoderm of wild-type embryos causes complete repression of bap, it does not produce a significant increase of eve-expressing cells under the same conditions. Similarly, ectopic expression of wg in the mesoderm (or ectoderm) of wild-type embryos with the GAL4/UAS system fails to produce a major increase of eve-expressing cells. By contrast, when wg and slp are coexpressed in the mesoderm, there is a dramatic increase in eve-expressing cells, which become distributed all along the dorsal margin of the mesoderm. Analogous experiments have shown that the formation of slouch-expressing ventrolateral muscle progenitors and tin-expressing heart progenitors also requires the joint activities of wg and slp. These combined data show that neither wg nor slp alone is sufficient to allow the formation of various heart and muscle progenitors. Instead, it appears that wg and slp cooperate in parallel or consecutive pathways during the formation of these cell types (Lee, 2000).

It was next asked whether one of these pathways involves the regulation of twist expression by mesodermal Slp. Previous reports have indicated that wg and slp are required to generate segmentally elevated levels of twist expression during stage 11 and that these higher levels of twist are necessary for normal somatic muscle specification. The stripes of elevated Twist coincide with the mesodermal domains of Slp during stage 11. Moreover, uniform expression of slp in the mesoderm produces uniform expression of high Twist levels, even in the absence of Wg activity. Thus, it appears that one route of wg function in myogenesis involves the upregulation of twist by Wg-induced slp in the mesoderm. Additional experiments confirm that, analogous to bap repression, the mesodermal component of slp is essential for somatic muscle and heart specification. In these experiments, mesodermal slp expression is specifically removed by ectopically expressing en in the mesoderm of wild-type embryos, as described above. Embryos of this genotype show a complete absence of eve-expressing muscle and pericardial progenitors, and an almost complete absence of Mef2-labeled somatic mesoderm. The expression of nautilus in muscle founders is also missing in embryos of this genotype. In addition, tin-expressing heart progenitors fail to be formed and only a few somatic muscle fibers are present in late-stage embryos. In contrast to the loss of derivatives of the mesodermal A (slp) domains, derivatives of the P (eve) domains, including trunk visceral mesoderm and fat body, are present and even expanded in these embryos. In the aggregate, the evidence suggests that Wg-induced slp activity in the mesoderm is required but not sufficient to generate somatic muscle and heart progenitors. It appears that, in addition to this route, wg is required independent of slp for the formation of these cell types (Lee, 2000).

Based upon the breakpoints and phenotypes of several large deletions in the slp locus, it has been concluded that the 5' flanking regions of slp1 contain essential regulatory elements that are shared between the slp1 and slp2 genes. Indeed, reporter gene analysis has identified three separate enhancer elements within ~5.5 kb of upstream sequences, each of which reproduces one particular aspect of endogenous slp expression. The two distal elements drive head-specific expression during blastoderm stages and striped expression in the ventral ectoderm following stage 11, respectively. The third, more proximal element is active in both the mesoderm and ectoderm between stages 10 and 11, coinciding with the period when mesodermal genes such as bap and eve are activated. Its spatial and temporal profile, as well its wg dependency, are fully consistent with the notion that this enhancer is targeted by the Wg signaling cascade during mesoderm induction and autocrine signaling in the ectoderm. Because it is not active during the period when the pair-rule genes are expressed and is turned on in every parasegment, without displaying any transient pair-rule pattern, this enhancer appears to be exclusively responsive to Wg in terms of its spatial regulation. Hence, there must be separate pair-rule enhancer(s), which were not uncovered by this dissection. The observation that this Wg response element becomes rapidly inactive at late stage 11, although wg continues to be expressed, indicates that it requires at least one additional, temporally restricted regulator in conjunction with the Wg signal (Lee, 2000).

Ubx is one of the few genes known to be a direct target of Wingless, although there are indications that en may be another example. An enhancer element of Ubx, which is activated by Wg after the formation of the midgut visceral mesoderm, contains at least two functionally important Pan-binding sites. In the present study, it has been determined that a Wg-responsive enhancer of slp contains nine in vitro Pan-binding sites. It is not known whether all of these sites are functional because in vivo tests of mutations of individual sites or various combinations of them have not yet been carried out. However, the current data from two rounds of mutagenesis show that some of them are indeed functional in vivo. This analysis also indicates that the activities of different functional binding sites are additive and therefore partially redundant (Lee, 2000).

Although most of the known wg downstream genes are upregulated by Wg signals, there are several examples of negatively regulated genes, in addition to bap. For instance, the Ubx midgut enhancer is downregulated by high levels of Wg signaling through a mechanism that involves Smad-binding sites. It has been speculated that Wg signals may induce the transcription of an unknown repressor that competes with Mad/Medea binding to these sites. By contrast, the repression of shavenbaby (ovo) in the epidermis of late stage embryos by Wg signaling has been proposed to involve the direct binding of hypothetical Pangolin/Armadillo/corepressor complexes to ovo regulatory elements. The present study provides strong evidence for an indirect mode of bap repression by Wg that is more comparable to the one proposed for Ubx. It appears that during this pathway, Wg signals induce the transcription of slp, whose encoded protein product probably acts as a repressor that interferes with Dpp/Tinman mediated activation of bap transcription (Lee, 2000).

lame duck expression was examined in wg and N mutant embryos to determine the relative position of lmd within the genetic hierarchy that controls somatic muscle specification. In wg^cx4 mutant embryos, lmd RNA expression is not detectable in dorsolateral and lateral somatic mesodermal cells although there is residual expression in cells located in the ventral region. Thus, activation of lmd expression is mediated via wg-dependent and wg-independent pathways. Significantly, lmd expression in the somatic mesoderm is completely abolished in N⁵⁴¹⁹ mutant embryos, indicating that activation of lmd expression in presumed fusion-competent myoblasts requires active Notch signaling. By contrast, founder cell formation is promoted in the absence of Notch function (Duan, 2001).

The flight muscles of Drosophila derive from myoblasts found on the third instar disc. These myoblasts already show distinctive properties: how this diversity is generated was examined. In the late larva, Vestigial and low levels of Cut are expressed in myoblasts that will contribute to the indirect flight muscles. Other myoblasts, which express high levels of Cut but no Vestigial, are required for the formation of the direct flight muscles. Vestigial and Cut expression are stabilized by a mutually repressive feedback loop. Vestigial expression begins in the embryo in a subset of adult myoblasts, and Wingless signaling is required later to maintain this expression. Thus, myoblasts are divided into identifiable populations, consistent with their allocation to different muscles, and ectodermal signals act to maintain these differences (Sudarsan, 2001).

If the specification of the adult myoblasts is influenced by the disc epithelium, signals must be able to diffuse between germ layers to stimulate the myoblasts. A striking feature of the presumptive notum of the wing disc is an ectodermal stripe of wingless (wg) expression, which is initiated in the third instar. When conventional fixation procedures are used, Wg is detectable at the apical surface of the disc epithelium, whereas the protein gradient forms on the basolateral domain, adjacent to the myoblast layer. To assess whether Wg secreted from epidermal cells could signal to the associated myoblasts, a truncated, nonfunctional, GPI-linked form of the Wingless receptor, DFrizzled2, was expressed in all the adult myoblasts. Whether Wg protein was detectable in the mesodermal layer was then examined. It was found that Wg can cross between the germ layers and bind to the myoblasts in a graded fashion, covering a domain that is wider than the Wg stripe (Sudarsan, 2001).

It was asked whether myoblasts were receptive to inductive signaling from the epidermis and whether wg is required for the divergence of the two myoblast populations in wild-type third instar discs. Using a temperature-sensitive allele, wg^IL114, Wingless function was removed from the second instar onward. This results in a strong reduction of Vg expression in the myoblasts. The importance of signaling through the canonical Wg pathway was tested by expressing a dominant-negative form of the transcriptional effector, TCF, in all the myoblasts of the notum. A complete loss of Vg expression was found, and, in the adult, the IFMs were much reduced in size. In contrast, when the Wg pathway was activated by expressing arm^S10, a constitutively active form of Armadillo, throughout the notum myoblasts, a small expansion of Vg expression into the most distal population of myoblasts was found. In the adult, some DFMs are lost, while those that remain are very reduced in size. These results show that spatially organized Wg signaling from the ectoderm is required to maintain Vg expression in the adjoining population of myoblasts (Sudarsan, 2001).

In order to compare the levels of Cut in wild-type and genetically manipulated cells within single populations of myoblasts, mitotic clones were generated of myoblasts carrying a mutation in dishevelled (dsh) and were therefore unable to transduce Wg signaling. As expected, dsh clones lose Vg expression, but they also show an increase in the level of Cut expression relative to neighboring wild-type myoblasts. This indicates that, in addition to its role in maintaining Vg expression, Wg signaling is also required, directly or indirectly, to modulate myoblast expression of Cut. Interestingly, changes in gene expression are confined to the mutant cells, indicating a cell-autonomous response; the induction of secondary signaling by Wg-activated cells is not involved (Sudarsan, 2001).

Wingless in the genital disc

The integration of multiple developmental cues is crucial to the combinatorial strategies for cell specification that underlie metazoan development. In the Drosophila genital imaginal disc, which gives rise to the sexually dimorphic genitalia and analia, sexual identity must be integrated with positional cues, in order to direct the appropriate sexually dimorphic developmental program. Sex determination in Drosophila is controlled by a hierarchy of regulatory genes. The last known gene in the somatic branch of this hierarchy is the transcription factor doublesex (dsx); however, targets of the hierarchy that play a role in sexually dimorphic development have remained elusive. The gene dachshund (dac) is differentially expressed in the male and female genital discs, and plays sex-specific roles in the development of the genitalia. Furthermore, the sex determination hierarchy mediates this sex-specific deployment of dac by modulating the regulation of dac by the pattern formation genes wingless (wg) and decapentaplegic (dpp). The sex determination pathway acts cell-autonomously to determine whether dac is activated by wg signaling, as in females, or by dpp signaling, as in males (Keisman, 2001a).

A number of obstacles make it difficult to demonstrate that the sex determination pathway is responsible for the sex-specific regulation of a gene in the genital disc. These obstacles stem from the fact that the male and female primordia, which are the primary constituents of their respective discs, differ in their segmental origin. This raises the possibility that 'sex-specific' gene regulation is really just segment-specific gene regulation, made to look sex specific by the fact that only one primordium develops in each sex. Attempts were made to address this concern by creating clones of the opposite genetic sex in chromosomally male and female genital discs. Thus, for example, dac regulation could be examined in the male (A9) primordium, in both male and female cells. By varying the genetic sex of cells in a context where segmental identity is uniform, it was hoped that the contributions of sex and segmental identity to dac regulation could be disentangled (Keisman, 2001a).

In the male primordium of both male and female discs, the regulation of dac varies according to the genetic sex of the cell. Genetically female clones in the male (A9 derived) primordium of the male genital disc are unable to express dac in the lateral male (dpp-dependent) domain, but are able to express dac when they extended medially, towards the source of Wg. Conversely, in the female genital disc, genetically male clones in the repressed male primordium (A9) lose their ability to express dac in the medial, wg-dependent domain, and begin to express dac laterally, presumably in response to Dpp. Finally, dac expression is abnormal in intersexual genital discs from dsx mutant larvae: the male primordium of dsx genital discs expresses dac in both the endogenous, lateral male domains, and in a slightly weaker medial domain that corresponds roughly to the region where tra + clones are able to activate dac. Thus, it is concluded that in the male primordium, the sex determination pathway determines how a cell will regulate dac (Keisman, 2001a).

In the female primordium the results fail to show a role for the sex determination pathway in dac regulation. If such a role exists, it would be expected that genetically male clones in the female primordia of a female genital disc would activate dac laterally, like their counterparts in the male primordia. They do not, even when they take up much of the presumptive dpp-expressing domain. It would also be expected that such clones would repress dac medially. Only a few clones were observed to extend into the medial wg-expressing domain, and as expected these appear to repress dac. Interpretation of these results is complicated by the fact that changing the genetic sex of a cell in the genital disc can cause it to enter the 'repressed' state. Thus, for example, if a genetically male clone represses dac when it intersects the medial dac domain in the female primordia, it can be concluded either that the sex determination pathway regulates dac expression or that the cells, which are now male, have adopted a repressed state and are generally unresponsive. A similar caveat prevents interpreting the failure of tra2IR clones to activate dac ectopically in the female primordium. That tra + clones in the male primordium of male genital discs enter such a generally non-responsive state was not of concern, because these clones both repress and activate dac expression. The expression pattern of dac in the female primordium of a dsx mutant genital disc is also difficult to interpret. dac is not activated ectopically in the lateral domains of the dsx female primordium, which is consistent with the failure of tra2IR clones to cause such activation. However, even the medial, wg-dependent dac domain is frequently absent or severely reduced in the dsx female primordium, and thus the authors are reluctant to draw any conclusions from the absence of ectopic dac laterally (Keisman, 2001a).

A model is proposed for dac regulation in the male primordium, in which the different isoforms of Dsx protein modulate dac regulation by wg and dpp. In the absence of dsx, both wg and dpp can activate dac, producing the two domains of dac expression observed in the male primordium of a dsx disc. In the female, Dsx^f modulates dpp activity so that dpp becomes a repressor of dac; Dsx^f may also potentiate the activation of dac by wg. In the male, Dsx^m modulates wg activity so that it becomes a repressor of dac, leaving dpp alone to activate dac. In support of this model, it is noted that the Dsx proteins act in a similar manner to positively or negatively modulate the effect of tissue-specific regulators on the yp genes (Keisman, 2001a and references therein).

The behavior of tra + and tra2IR clones provides insight into the mechanism of repression in the undeveloped genital primordium. It was anticipated that such clones would be difficult to recover when they occurred in the male and female primordium, respectively, because they should adopt the repressed state. Instead, large tra + (female) clones were recovered in the male primordium of a male disc, and large tra2IR (male) clones were recovered in the female primordium of a female disc. Some of these clones constitute a substantial fraction of the primordium in question. Though tra + or tra2IR clones were not scored in adults, previous studies strongly suggest that such clones would fail to differentiate adult genital structures (Keisman, 2001a).

It has been shown that tra - (male) clones cause large deletions in the female genitalia, indicating that genetically male cells like those in a tra2IR clone divide but cannot differentiate female genital structures. Further, it has also been shown that male structures are deleted when the mosaic border passes through the male genitalia, suggesting that female tissue cannot differentiate male structures. To reconcile these data, it is proposed that repression of the inappropriate genital primordium involves two separable processes: repression of growth and the prevention of differentiation. Thus, clones of cells of the inappropriate genetic sex cannot differentiate, but they can grow and contribute to a morphologically normal genital primordium. This poses yet another question. Cells in a tra + clone in the male primordium of a male genital disc are analogous to the cells in the repressed male primordium of a wild-type female genital disc: both are genetically female, and both have A9 segmental identity. Why do tra + clones in the male primordium grow, while the repressed male primordium in a female disc does not? One possibility is that the decision of the male primordium to grow in a male disc is made before tra + clones were induced and cannot be over-ridden by a later switch of genetic sex. However, temperature-shift experiments with tra-2 ts alleles suggest that the decision of a genital primordium to develop can be reversed later in development. Furthermore, occasional, large tra + clones can cause severe reductions in male genital discs. This observation leads to the suggestion of a model in which growth in the genital disc is regulated from within organizing zones, such as the domains of wg and dpp expression. According to this model, the sex of the cells in the organizing regions would determine how the disc grows, while cells in other regions would respond accordingly, regardless of their sex. The tra + clones that cause reduction could result when such a clone intersects with one of the postulated organizing centers within the disc. The implication is that the sex determination pathway acts in yet undiscovered ways to modulate the function of the genes that establish pattern in the genital disc. One such interaction was found in the regulation of dac; further study is needed to determine if others exist, and what role they play in producing the sexual dimorphism of the genital disc and its derivatives (Keisman, 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, 2001).

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, 2001).

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, 2001).

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, 2001).

Wingless signaling initiates mitosis of primordial germ cells during development in Drosophila

The germline cells of Drosophila are derived from pole cells, which form at the posterior pole of the blastoderm and become primordial germ cells (PGCs). To elucidate the signal transduction pathways for the development of embryonic PGCs, the effects of various growth factors on the proliferation of PGCs were examined. Up- and down-regulation of Wingless (Wg) in both of soma and PGCs caused an increase and a decrease in the number of PGCs, respectively. The Wg/beta-catenin signaling pathway began to occur in PGCs at the same time as the PGCs began to divide during the embryonic stage in both sexes. In addition, PGCs were found to produce wg mRNA as they begin to divide. Thus, Wg functions as an autocrine factor to initiate mitosis in embryonic PGCs. Decapentaplegic affected the growth of PGCs from the end of the embryonic stage. The results indicate that these growth factors regulate the division of embryonic PGCs in a stage-specific manner (Sato, 2008).

The present study clearly shows that the Wg signal has an essential role in the initiation of mitosis in embryonic PGCs. Wg functions as an autocrine growth factor for the reentry of PGCs into mitosis. This is much different from the situation with regard to adult GSCs. The maintenance and division of GSCs have been shown to be regulated by niche cells that produce Dpp and Gbb. Wg regulates the maintenance of somatic stem cells but not GSCs. Wg had no effect on the growth of bam GSCs in culture (Niki, 2006). The simultaneous occurrence of the Wg signaling pathway and reentry of PGCs into mitosis in both sexes indicates that the sexual dimorphism of the mitotic property of PGCs during embryonic development would depend on the action of the Wg signaling pathway. At present, it is an open question what upstream factor(s) regulates the timing of Wg signaling in PGCs during embryonic development in each sex. One of the factors might be the JAK/STAT pathway, which was shown to be required for male-specific germ cell behavior during early embryonic development (Wawersik, 2005). When Upd was overexpressed in female embryos, extraordinary mitosis was induced at the mid-embryonic stage (Wawersik, 2005). It is important to elucidate the factor(s) that regulates the Wg signaling pathway for the initiation of the reentry into mitosis of embryonic PGCs in the two sexes (Sato, 2008).

Wnt is a family of secreted proteins that regulate several physiological and pathological processes in the development and maintenance of various tissues. In Drosophila, Wg signaling promotes cell cycle progression in the embryonic Malpighian tubules and in the early larval imaginal wing disc. In mammals, Wnt/β-catenin signaling regulates the maintenance of various stem cell systems, including intestinal epithelial, follicular, hematopoietic, and embryonic stem cells. Unexpectedly, the present results showing that the role of Wg signal in initiating the mitosis of embryonic PGCs are contrary to the role of Wnt signaling in PGC development in mouse. In mouse PGCs, nuclear-localized β-catenin gradually disappears after E13.5 (Kimura, 2006). The suppression of Wnt/β-catenin signaling is a prerequisite for normal cell cycle progression in PGCs. The cell cycle is arrested by overexpression of β-catenin, resulting in germ cell deficiency followed by apoptosis. At present, the reason for Wg to have different roles in PGCs of fly and mouse are not known. Further work will be required to elucidate the function of Wg in PGCs of diverse animals (Sato, 2008).

Considering that in the present study the percentage of mitotic PGCs reached the same level in the later embryonic stage even when Wg was down-regulated constitutively with a tissue-specific driver, the lack of Wg signaling does not affect the function of Dpp, which becomes effective from the late embryonic stage. During larval development, only Dpp is effective in promoting the growth of PGCs (Sato, unpublished results). Dpp may function as a main growth factor for the division of embryonic larval development. Further analyses of the growth factors that regulate the division of PGCs after the embryonic stages are needed (Sato, 2008).

The increase in the number of PGCs was always less than 2-fold compared with controls. This may be due to a limitation in the time schedule and space available for gonads during development. Recently, Gilboa (2006) showed that the proliferation of PGCs and survival of the intermingled somatic cells that contacted them were coordinated by means of a feedback mechanism composed of a positive signal and a negative signal during gonad development. They proposed that homeostasis and coordination of growth between soma and germ line in the larval ovary were achieved using a sensor of PGC numbers (epidermal growth factor-mediated survival of intermingled cells) coupled to a correction mechanism inhibiting PGC proliferation. Thus, it is reasonable to consider that the number of PGCs is controlled by this mechanism (Sato, 2008).

Table of contents

Home page: The Interactive Fly © 1995, 1996 Thomas B. Brody, Ph.D.

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