wingless

Table of contents

The Wingless Pathway and Miscellaneous Wingless Targets

armadillo, disheveled and shaggy encode elements of a unique wingless signaling pathway used several times throughout development. WG signaling generates a hyperphosphorylated form of DSH, that is associated with a membrane fraction. Overexpressed dsh becomes hyperphosphorylated in the absence of extracellular WG and increases levels of the Armadillo protein, thereby mimicking the WG signal (Yanagawa, 1995).

Wingless acts through inactivation of the Shaggy/Zeste white 3 protein kinase to specify ventral cell fate in the leg. Ectopic expression of wg outside its normal ventral domain has been shown to reorganize the dorsal-ventral axis of the leg in a non-autonomous manner. Cells that lack Shaggy/Zeste-white 3 activity can influence the fate of neighboring cells to reorganize dorsal-ventral pattern in the leg, in the same manner as wg-expressing cells. Therefore, clones of cells that lack Shaggy/Zeste white 3 activity exhibit all of the organizer activity previously attributed to wg-expressing cells, but do so without expressing wg. The organizing activity of ventral cells depends upon the location of the clone along the dorsal-ventral axis (Diaz-Benjumea, 1994).

One enhancer element in the BXD region of Ultrabithorax, called 22186R is activated after the blastoderm stage, in contrast to other early enhancers. 22186R is not detectable until germ band extension is well under way. Ectodermal expression becomes visible in stage 9 embryos as a series of thin bands in PS6, 10, 11, 12, and 13, and shortly afterward staining appears in PS 7, 8 and 9, and then in the thoracic parasegments PS5, 4 and 3. The 2218R6 enhancer is unusual in directing expression in the central nervous system (CNS) at later embryonic stages. In wingless mutants expression from 22186R is absent (Poux, 1996).

The maintenance of gooseberry distal is controlled by the wg signal. A control element responsible for wg-dependent maintenance of gsb expression, gsb-late element, is separable from gsb-early element, which is required for the initial activation of gsb by pair-rule transcription factors (Li, 1993a).

In the dorsal epidermis and the terminal regions of the body, expression of wingless is independent of gooseberry but requires a wingless-lady bird regulatory feedback loop. Loss of lady bird function reduces the number of wingless-expressing cells in dorsal epidermis and leads to complete inactivation of wingless in the anal plate. Consequently, mutant lady bird embryos fail to develop anal plates and ubiquitous embryonic expression of either one or both lady bird genes leads to severe defects of the dorsal cuticle. Lack of late wingless expression and anal plate formation can be rescued with the use of a heat-shock-lady bird transgene (Jagla, 1997a).

Analysis of the expression of 18 wheeler in different mutant backgrounds shows that it is under control of segment polarity and homeotic genes. Initial accumulation of 18w is normal in wingless mutants. However, by full germband extension, the ventrolateral expression of 18wis narrower than in wild type. These changes appear well before cell death is seen in wg mutants. In patched mutants, the domains of wg and of 18w expand to include the expression domains of wingless and engrailed. These results suggest that wg and en positively regulate 18w expression within the ventromedial stripes (Eldon, 1994).

Members of the Hedgehog (Hh) and Wnt/Wingless (Wg) families of secreted proteins control many aspects of growth and patterning during animal development. Hh signal transduction leads to increased stability of the transcription factor Cubitus interruptus (Ci), whereas Wg signal transduction causes increased stability of Armadillo (Arm/beta-catenin), a possible co-factor for the transcriptional regulator Lef1/TCF. A new gene, slimb (for supernumerary limbs), is described which negatively regulates both of these signal transduction pathways. Loss of slimb function results in a cell-autonomous accumulation of high levels of both Ci and Arm, and the ectopic expression of both Hh- and Wg- responsive genes. Clones of slimb1 cells in the leg or wing disc ectopically express dpp or wg when they arise in the anterior (but not the posterior) compartments of these discs. Anterior clones reorganize normal limb pattern, creating supernumerary 'double-anterior' limbs. Slimb, like PKA, is a negative regulator that normally prevents activity of the Hh signal transduction pathway in the absence of ligand. slimb mutant cells that arise in the presumptive wing blade ectopically express Scute and differentiate ectopic sensory bristles instead of epidermal hairs on the surface of the wing blade. Both phenotypes are strictly autonomous to the mutant cells, as is the case when the Wg signal transduction pathway is constitutively activated, but not when Wg is ectopically expressed. The slimb gene encodes a conserved F-box/WD40-repeat protein related to Cdc4p, a protein in budding yeast that targets cell-cycle regulators for degradation by the ubiquitin/proteasome pathway. It is proposed that Slimb protein normally targets Ci and Arm for processing or degradation by the ubiquitin/proteasome pathway, and that Hh and Wg regulate gene expression, at least in part, by inducing changes in Ci and Arm, which protect both Ci and Arm from Slimb-mediated proteolysis (J. Jiang, 1998).

The tissue-specific regulation of Vn signaling was investigated by examining vn transcriptional control and Vn target gene activation in the embryo and the wing. The results show a complex temporal and spatial regulation of vn transcription involving multiple signaling pathways and tissue-specific activation of Vn target genes. In the embryo, vn is a target of Spi/Egfr signaling mediated by the ETS transcription factor PointedP1 (PntP1). This establishes a positive feedback loop in addition to the negative feedback loop involving Aos. The simultaneous production of Vn provides a mechanism for dampening Aos inhibition and thus fine-tunes signaling. In the larval wing pouch, vn is not a target of Spi/Egfr signaling but is expressed along the anterior-posterior boundary in response to Hedgehog (Hh) signaling. Repression by Wingless (Wg) signaling further refines the vn expression pattern by causing a discontinuity at the dorsal-ventral boundary (Wessells, 1999).

The gene proboscipedia (pb) is a member of the Antennapedia complex in Drosophila and is required for the proper specification of the adult mouthparts. In the embryo, pb expression serves no known function despite having an accumulation pattern in the mouthpart anlagen that is conserved across several insect orders. Several of the genes necessary to generate this embryonic pattern of expression have been identified. Nearly all the pair rule and segment polarity genes affect morphology of the gnathal segments and to varying degrees perturb pb accumulation. In general, mutations in the pair rule genes eliminate either the maxillary or labial lobe, as well as reduce the width of the respective segment. Despite these effects on morphology, pb expression can often be seen in the affected segments. Mutations in the segment polarity genes affect the morphology of both the maxillary and labial lobes. The overall effect is a reduction in the size of these lobes, resulting in a correspondingly reduced number of pb-expressing cells. Of the segment polarity genes tested, wingless has the strongest effect on pb expression. At early stages in wg^CX4 mutants, no pb expression is apparent in the presumptive labial lobe, though later, as head involution commences, some of these cells do begin to express pb (Rusch, 2000).

Chibby regulates Wingless targets

Inappropriate activation of downstream target genes by the oncoprotein ß-catenin is implicated in development of numerous human cancers. ß-catenin and its fruitfly counterpart Armadillo act as coactivators in the canonical Wnt/Wingless pathway by binding to Tcf/Lef transcription factors. A conserved nuclear protein, named Chibby, has been identified a screen for proteins that directly interact with the C-terminal region of ß-catenin. In mammalian cultured cells Chibby inhibits ß-catenin-mediated transcriptional activation by competing with Lef-1 to bind to ß-catenin. Inhibition of Drosophila Chibby by RNA interference results in segment polarity defects that mimick a wingless gain-of-function phenotype, and overexpression of the wingless target genes engrailed and Ultrabithorax. In addition, epistasis experiments indicate that chibby acts downstream of wingless and upstream of armadillo (Takemaru, 2003).

To investigate the role of Cby in Wnt signalling during development, Drosophila was examined because the fruitfly expresses a Cby ortholog that, like the human Cby described above, inhibits ß-catenin-dependent activation of reporter in mammalian cultured cells. cby double-stranded RNA was injected into wild-type embryos; this resulted in transformation of ventral denticles into naked cuticle. This phenotype is similar to embryos that overexpress wingless (wg) in all epidermal cells. In contrast, losing wg pathway activity results in transformation of naked cuticle into denticles. In addition, the head structures of cby(RNAi) embryos were missing, and the embryos were small. These phenotypes were also observed in embryos that express intron-spliced snapback RNA corresponding to cby. To confirm the specificity of the RNAi phenotype, a short interfering RNA was injected whose sequence did not overlap with either the dsRNA or snapback RNA used earlier. The siRNA duplex produced a highly similar phenotype, indicating that the phenotype is caused by specific reduction of cby activity. The cuticle phenotype suggests that cby functions as an antagonist of the wg pathway. To test this possibility further, expression of the gene engrailed (en), which is transcriptionally activated by wg signalling in the cuticle-secreting embryonic epidermis, was examined (Takemaru, 2003).

En is normally expressed in segmental stripes two cells wide, immediately behind cells expressing wg. In cby(RNAi) embryos, the En stripes expanded by another row of cells, a phenotype similar to one observed when embryos overexpress wg in the en domain. This indicates that cby normally represses expression of en. Because en is in turn required to maintain wg expression, wg activates its own expression through a paracrine feedback loop. RNAi depletion of cby results in expansion of wg messenger RNA and protein expression, further indicating that loss of cby function leads to wg pathway activation. The expanded width and intensity of the Wg stripes is greater than in embryos where wg signalling is maximally stimulated. This suggests that in addition to wg signalling, cby acts in a regulatory process at present unknown (Takemaru, 2003).

Experiments with mammalian Cby suggest that it inhibits Wnt signalling by binding to ß-catenin in the nucleus and blocking its interaction with Tcf/Lef transcription factors. If Drosophila Cby functions in a similar manner, then loss of cby should not affect the abundance or localization of Arm (ß-catenin). Arm localizes to nuclei in stripes of Wg-responding cells at stage 9 of embryogenesis (Takemaru, 2003).

Arm protein was examined in cby(RNAi) embryos; Arm abundance and localization were not detectably affected. If Cby were involved in transducing the Wg signal, then Wg would be predicted to lie upstream of Cby in an epistasis genetic pathway. To test for epistasis, the expression of a wg-responsive UbxB-lacZ reporter gene was examined in the embryonic midgut (Takemaru, 2003).

Expression is controlled by an enhancer that interacts with Drosophila Tcf and is activated in a manner dependent upon Wg signalling. Wg is expressed in parasegment (ps) 8, where it controls UbxB-lacZ expression in visceral mesoderm throughout ps7-9, as well as development of the second midgut constriction. In a wg null embryo, UbxB-lacZ expression is greatly reduced and the midgut constriction fails to form, indicating a strong dependence on wg function. In cby(RNAi) embryos, lacZ expression expanded into anterior and posterior parasegments, and the second but not the first midgut constriction formed; these phenotypes are reminiscent of moderate wg misexpression throughout the midgut. Thus, consistent with the observations of cby function in the epidermis, cby represses wg-dependent gene expression and development in visceral mesoderm. When cby was depleted by RNAi in wg null embryos, UbxB-lacZ expression was not blocked, and development of the second midgut constriction occurred. The wg;cby(RNAi) embryos resembled wild-type more than cby(RNAi) embryos; they sometimes exhibited more restricted lacZ expression and a first midgut constriction. Since embryonic RNAi usually leads to reduced levels of gene product, this implies that the residual cby product represses visceral mesoderm more effectively in the absence of wg. These results establish that wg function is mediated, at least in part, through cby (Takemaru, 2003).

It is concluded that Cby is a nuclear protein that is conserved throughout evolution. It antagonizes Wnt/Wg signalling by inhibiting ß-catenin/Arm function in mammalian cells and in Drosophila, raising the possibility that it may be a tumor suppressor gene. In this regard, Cby expression was found to be significantly downregulated in thyroid and metastatic uterine tumors, and nuclear Cby staining is missing or considerably weaker in thyroid tumors. Dysregulation of ß-catenin signalling has been reported in these types of cancers, so the decreased levels of Cby expression might be relevant to tumor formation (Takemaru, 2003).

Sunspot regulates wingless targets: Sunspot, a link between Wingless signaling and endoreplication

The Wingless (Wg)/Wnt signaling pathway is highly conserved throughout many multicellular organisms. It directs the development of diverse tissues and organs by regulating important processes such as proliferation, polarity and the specification of cell fates. Upon activation of the Wg/Wnt signaling pathway, Armadillo (Arm)/beta-catenin is stabilized and interacts with the TCF family of transcription factors, which in turn activate Wnt target genes. This study shows that Arm interacts with a novel BED (BEAF and Dref) finger protein that has been termed Sunspot (Ssp). Ssp transactivates Drosophila E2F-1 (dE2F-1) and PCNA expression, and positively regulates the proliferation of imaginal disc cells and the endoreplication of salivary gland cells. Wg negatively regulates the function of Ssp by changing its subcellular localization in the salivary gland. In addition, Ssp was found not to be involved in the signaling pathway mediated by Arm associated with dTCF. These findings indicate that Arm controls development in part by regulating the function of Ssp (Taniue, 2010).

Arm is composed of 12 imperfect protein interaction repeats (Armadillo repeat domain) flanked by unique N and C termini. In an attempt to identify novel Arm-binding proteins, a yeast two-hybrid screen of a Drosophila embryo cDNA library was performed using the Armadillo repeat domain of Arm as bait. Positive clones containing the same insert of a novel gene (CG17153) were isolated that were named sunspot (ssp; named after the phenotype of mutant flies). Sequence analysis of the full-length cDNA revealed that it encodes a protein of 368 amino acids. A region near its N terminus (amino acids 34 to 98) shows similarity to the BED (BEAF and Dref) finger domain, which is predicted to form a zinc finger and to bind DNA (Taniue, 2010).

To confirm the interaction between Ssp and Arm, whether Ssp produced by in vitro translation could interact with the Armadillo repeat domain of Arm fused to glutathione S-transferase (GST) was tested. Ssp specifically interacted with the Armadillo repeat domain of Arm (amino acids 140 to 713), but failed to interact with Pendulin (Pen), a Drosophila homolog of importin α, which also possesses the Armadillo repeat domain. Pull-down assays with a series of deletion fragments of Ssp showed that a fragment of Ssp containing amino acids 235 to 307 (termed the ABR, the Arm-binding region) binds to Arm in vitro. Also, it was found that Armadillo repeats 2-8 of Arm are responsible for binding to Ssp. Although TCF is known to bind to Armadillo repeats 3-10 of Arm, Ssp did not compete with TCF for binding to Arm (Taniue, 2010).

Next, whether Ssp is associated with Arm in living cells was examined. Drosophila Schneider-2 (S2) cells were transfected with Arm along with GFP-Ssp, GFP-SspδC (amino acids 1 to 217; a mutant lacking the ABR) or GFP-SspABR (amino acids 235 to 342; a fragment containing the ABR). GFP-fusion proteins were immunoprecipitated from S2 cell lysates and subjected to immunoblotting with anti-GFP and anti-Arm antibodies. For immunoprecipitation of GFP-fusion proteins, a 13-kDa GFP-binding fragment was used derived from a llama single chain antibody, which was covalently immobilized to magnetic beads (GFP-Trap-M), as the molecular weight of GFP-SspδC is the same as that of IgG. It was found that Arm is associated with GFP-Ssp and GFP-SspABR. By contrast, Arm barely co-immunoprecipitated with GFP-SspδC. In addition, pull-down assays were also performed with a mixture of lysates of S2 cells transfected with Arm alone and GFP-Ssp alone, respectively. It was found that Ssp and Arm co-precipitate only when both proteins are co-expressed in S2 cells, excluding the possibility that Ssp and Arm associate after cells are lysed. Taken together, these results suggest that Ssp interacts via its ABR with Arm not only in vitro but also in vivo (Taniue, 2010).

One lethal P-element insertion line, l(3)j2D3^j2D3, was found in which a P-element had been inserted into the gene adjacent to ssp, CG6801, which is located about 250 bp upstream of the 5' end of ssp. RT-PCR analysis revealed that the expression level and size of the CG6801 transcript were not changed compared with in wild-type larvae, which is consistent with the P-element being inserted into an intron in CG6801. To generate mutants that have a deletion in ssp but have intact CG6801, a local hop and imprecise excision approach was used. l(3)j2D3^j2D3 was used in a local hop to generate a P-element insertion line, sunspot^P, that completely complemented the lethality of l(3)j2D3^j2D3. Then ssp mutants were generated by imprecise excision of the P-element from sunspot^P. One allele was found that has a deletion of about 600 bp, and this was designated as ssp⁵⁹⁸. Sequence analysis showed that the deletion extends from a position 60 bp downstream of the presumptive ssp transcription start site to the ssp gene locus. Because this deletion removes the start codon and the BED finger domain of ssp, it is presumed that ssp⁵⁹⁸ represents a null allele for ssp. RT-PCR analysis revealed that ssp⁵⁹⁸ generates a truncated transcript. The truncated transcript encodes a peptide consisting of 13 amino acids, which is unrelated to Ssp. By contrast, RT-PCR analysis revealed that the intact CG6801 transcript is expressed in ssp⁵⁹⁸ mutant larvae, and that the expression level of CG6801 is unchanged in ssp⁵⁹⁸ mutant larvae compared with that in wild-type larvae. Furthermore, ssp⁵⁹⁸ fully complemented the phenotype of l(3)j2D3^j2D3, indicating that this mutant contains intact CG6801 (Taniue, 2010).

The imaginal discs, salivary glands and central nervous system of larvae homozygous for ssp⁵⁹⁸ were smaller than those of their normal counterparts. ssp⁵⁹⁸ homozygotes reached the third instar stage, but failed to reach the pupal stage and died between 10 and 20 days after egg laying (AEL). Furthermore, melanotic pseudotumors were formed in ssp⁵⁹⁸ mutant larvae. Melanotic pseudotumors are groups of cells within the larvae that are recognized by the immune system and encapsulated within a melanized cuticle. One or more small melanotic pseudotumors first appeared in the ssp mutants at 6 days AEL, and the number and size of these melanotic pseudotumors increased during the development of the larvae. Similar phenotypes were observed with hemizygotes for ssp⁵⁹⁸ and Df(3L)BK9, which has a deletion larger than that of ssp⁵⁹⁸ and lacks ssp. In situ hybridization analysis of imaginal discs using the coding region of the ssp cDNA as a probe revealed that ssp transcripts are expressed ubiquitously. Therefore whether ubiquitous expression of ssp restores the phenotypes of ssp⁵⁹⁸ homozygous animals was examined. It was found that ubiquitous expression of the full-length ssp cDNA with the Gal4-UAS system rescued the lethality and other phenotypes associated with ssp⁵⁹⁸ homozygous animals. Taken together, these results suggest that the phenotypes of ssp⁵⁹⁸ homozygotes are caused by the loss of ssp function, and that ssp is required for cell proliferation and morphogenesis of the imaginal disc and central nervous system (Taniue, 2010).

Arm is a key transducer of Wg signaling and many of the Arm-binding proteins are known to function as a component of the Wg signal transduction pathway. To explore the possibility that Ssp is related to the Wg signal transduction pathway, the effect of Wg on the distribution of GFP-Ssp was examined. Because imaginal disc cells are too small for detailed study, focus for this analysis was placed on the third instar salivary glands, and whether the subcellular localization of GFP-Ssp is linked to Wg signaling was studied. The larval salivary gland mainly consists of secretory gland cells and imaginal ring cells. Gland cells are large polyploid epithelial cells. Small imaginal ring cells reside at the proximal end of the secretory gland. Immunostaining with anti-Wg antibody revealed that Wg is expressed in imaginal ring cells. Furthermore, Drosophila frizzled 3 (dfz3)-lacZ, a target gene of Wg signaling, was found to be expressed in imaginal ring cells and proximal gland cells, which reside within several cell diameters of the Wg-expressing cells. These results suggest that Wg signaling is active in the proximal region in the third instar salivary gland. When GFP-Ssp was expressed ubiquitously under the control of dpp-Gal4 in the larval salivary gland, GFP-Ssp was found to be localized predominantly at the nuclear envelope in proximal gland cells. In addition, GFP-Ssp was detected as aggregates in the nucleus in the distal region of the salivary gland. To examine whether this region-specific subcellular localization of GFP-Ssp is related to Wg signaling, Wg or Axin, a negative regulator of Wg signaling, was overexpressed in the salivary gland under the control of dpp-Gal4. It was found that expression of Wg along with GFP-Ssp resulted in the accumulation of a certain population of GFP-Ssp at the nuclear envelope in both the distal and proximal regions. Again, a significant amount of GFP-Ssp was localized in nuclear aggregates in both distal and proximal cells, suggesting that ectopic expression of Wg can also change the subnuclear localization of Ssp in proximal cells, from the nuclear periphery to nuclear aggregates. This result also suggests that ectopic expression of Wg in distal cells is not sufficient to change the subnuclear localization of all GFP-Ssp protein, from nuclear foci to the nuclear periphery. By contrast, when Axin was expressed along with GFP-Ssp, GFP-Ssp was detected as nuclear aggregates, not only in the distal region but also in the proximal region, but was no longer detected at the nuclear envelope. These results suggest that the subcellular localization of Ssp is regulated at least in part by Wg signaling in the third instar salivary gland (Taniue, 2010).

To examine whether the effect of Wg signaling on Ssp localization is mediated by the direct interaction between Arm and Ssp, Ssp localization was studied in larvae expressing an RNAi targeting Arm. It was found that Ssp was localized in nuclear aggregates and that Wg overexpression did not alter its localization when the expression of Arm was suppressed by RNAi. Thus, Arm is required for Wg-induced Ssp relocalization. Ssp localization was also examined in cells expressing δArm, a mutant of Arm that localizes at the plasma membrane. It was found that overexpression of δArm under the control of dpp-Gal4 results in the localization of GFP-Ssp at the plasma membrane throughout the salivary gland. Next the subcellular localization of SspδC, a mutant that lacks the ABR and is unable to interact with Arm, was examined. When GFP-SspδC was expressed ubiquitously, it was found to localize homogenously in the nucleus of both distal and proximal cells. This result indicates that the localization of Ssp to nuclear aggregates requires the ABR and suggests that Ssp requires a direct interaction with Arm to localize to its target sites in the nucleus. Furthermore, it was found that the localization of GFP-SspδC was not changed by coexpression with Wg, or δArm. Taken together, these results suggest that the direct interaction between Arm and Ssp is required for the regulation of Ssp localization by Wg signaling (Taniue, 2010).

The N-terminal region of Ssp contains a BED finger domain. This presumptive DNA-binding domain is known to be contained in several Drosophila proteins, such as Dref and BEAF-32. Dref regulates the transcription of genes involved in DNA replication and cell proliferation, including dE2F-1 and PCNA, the promoters of which contain BED finger-binding elements (BBEs). To clarify whether Ssp regulates the transcription of these genes, the expression levels of dE2F-1 and PCNA were examined. For this purpose, the P-element (lacZ) insertion lines E2F⁰⁷¹⁷² and PCNA⁰²²⁴⁸ were used. dE2F-1-lacZ and PCNA-lacZ expression were found to be high in distal cells compared with proximal cells in the larval salivary gland. When ssp was ectopically expressed in the salivary gland, dE2F-1-lacZ expression was markedly elevated in distal cells, whereas it was only slightly elevated in proximal cells. However, PCNA-lacZ expression was markedly elevated throughout the salivary gland. By contrast, dE2F-1-lacZ and PCNA-lacZ expression were not elevated in distal cells of ssp mutant salivary glands compared with in wild-type salivary glands, and dfz3-lacZ expression in ssp mutant and Ssp-overexpressing salivary glands was not changed compared with in wild-type salivary glands, suggesting that Ssp is not involved in Arm-dTCF-mediated transactivation of Wg target genes. In addition, overexpression of Wg resulted in a decrease in the expression levels of dE2F-1-lacZ and PCNA-lacZ in distal cells. Thus, Ssp is active in the distal region where Wg signaling is not active, and Ssp is aggregated in the nucleus. Conversely, Ssp is not very active in the proximal region where Wg signaling is active, and Ssp is accumulated in the nuclear envelope (Taniue, 2010).

The expression of dE2F-1, PCNA and dfz3 was examined in the wing disc. Clones of cells lacking Ssp function were generated by FLP/FRT-mediated somatic recombination. Clones of ssp mutant cells underwent only a few divisions after they were generated in the presumptive wing blade: the mutant cells proliferated slowly and either died or were actively eliminated from the disc epithelium. Therefore, a Minute mutation, M(3)65F, was used to confer a growth advantage upon cells homozygous for ssp. When mitotic recombination was induced in a M(3)65F background using enhancer trap lines, ssp mutant cells exhibited reduced levels of dE2F-1-lacZ and PCNA-lacZ expression but did not show any change in the levels of dfz3-lacZ and Arm expression. These results suggest that ssp regulates the expression of dE2F-1 and PCNA, but is not involved in Arm-dTCF-mediated Wg signaling (Taniue, 2010).

To confirm these results, endogenous expression of dE2F-1 and PCNA was examined by quantitative real-time RT-PCR analysis using RNA from late third instar larvae. Flies carrying heat-shock-inducible Gal4 (hs-Gal4) were crossed with transgenic flies carrying UAS-GFP, UAS-ssp or UAS-wg. Consistent with the above results, overexpression of ssp resulted in elevated steady state levels of dE2F-1 and PCNA transcripts. Furthermore, overexpression of Wg induced decreases in the numbers of dE2F-1 and PCNA transcripts. These results suggest that dE2F-1 and PCNA expression is regulated positively by Ssp and negatively by Wg (Taniue, 2010).

Also whether Ssp regulates the expression of dE2F-1 by binding directly to its promoter region was examined. Electrophoretic mobility-shift assays (EMSA) showed that GST-Ssp, but not GST, bound to a 40-mer oligonucleotide corresponding to a region in the dE2F-1 promoter that contains three BBEs. By contrast, GST-Ssp barely bound to a mutated probe in which CG in each BBE had been replaced with AA. Binding of Ssp to the wild-type probe was inhibited in the presence of an excess amount of unlabeled wild-type probe, whereas the mutated probe did not inhibit the interaction significantly. When anti-Ssp antibody was included in the reaction mixture, the Ssp band was not detected. Furthermore, it was found that GST-SspδBFD, a mutant Ssp lacking the BED finger domain, did not bind to the wild-type probe. These results suggest that Ssp regulates dE2F-1 expression by binding directly to the BBEs in the dE2F-1 promoter region via its BED finger domain (Taniue, 2010).

To further elucidate the function of Ssp and Wg, the third instar salivary glands of ssp and wg mutants were examined. In the third instar salivary gland, the distal region undergoes greater endoreplication than does the proximal region. As a result, the nuclear size of distal gland cells is markedly larger than that of proximal gland cells. However, the nuclear size of ssp mutant distal cells was found to be smaller than that of wild-type distal cells. By contrast, the nuclear size of wg mutant proximal cells was larger than that of wild-type proximal cells. Thus, the difference in nuclear size between proximal and distal cells was also small in the salivary glands of wg mutants (Taniue, 2010).

To confirm these results, the effects were examined of Ssp and/or Wg overexpression on the nuclear size of salivary gland cells. When Ssp was overexpressed, the nuclear size of both proximal and distal cells was heterogenous. Overexpression of Wg decreased the nuclear size of distal cells: the difference in nuclear size between Wg-overexpressing proximal and distal cells was small. However, when Wg was overexpressed along with Ssp, the effect of Ssp was suppressed and the heterogeneity of nuclear size was not observed. Furthermore, to confirm that Ssp and Wg play important roles in the regulation of endoreplication, δArm-expressing clones were generated using the flip-out technique. It was found that the nuclear size of δArm-expressing cells is much smaller than that of surrounding cells. This result suggests that δArm mislocalizes Ssp to the plasma membrane, thereby negatively regulating Ssp activity for endoreplication (Taniue, 2010).

To directly show that ssp mutant cells undergo fewer endoreplications than do wild-type cells, BrdU-labeling experiments were performed. When wild-type salivary glands were labeled with BrdU, distal cells efficiently incorporated BrdU, indicating that they underwent at least one round of DNA replication during the labeling period. By contrast, very few nuclei of ssp mutant cells and Wg-overexpressing cells were labeled with BrdU (Taniue, 2010).

dMyc has also been reported to be required for the endoreplication of salivary gland cells. It is therefore interesting to examine the relationship between dMyc, Wg and Ssp in endoreplication. It was found that dMyc expression was unchanged in both ssp mutant and Ssp-overexpressing salivary glands. Thus, Ssp might not be involved in the regulation of dMyc (Taniue, 2010).

Taken together, these results suggest that Ssp and Wg play important roles in the regulation of endoreplication in the third instar salivary gland, and that Wg might exert its effect by negatively regulating the function of Ssp. It is interesting to speculate that Ssp plays a general role for endoreplication in all larval endocycling tissues (Taniue, 2010).

It is believed that Wg/Wnt target genes are transactivated by Arm/β-catenin associated with TCF. However, expression of some human genes is transactivated by β-catenin that is associated with proteins other than TCF. For example, β-catenin interacts with the androgen receptor in an androgen-dependent manner and enhances androgen-mediated transactivation. In the present study, it was shown that Arm interacts with Ssp and negatively regulates its function. Ssp transactivates dE2F-1 and PCNA expression, and positively regulates the endoreplication of salivary gland cells. Furthermore, the Wg signal represses the function of Ssp by altering the subcellular localization of Ssp in the salivary gland: the Wg signal induces the accumulation of Ssp at the nuclear envelope. Interestingly, recent studies indicate that the nuclear membrane provides a platform for sequestering transcription factors away from their target genes. For example, it has been shown that the tethering of transcription factors such as c-Fos and R-Smads to the nuclear envelope prevents transcription of their target genes. The results appear to be consistent with these findings. Although the precise mechanism remains to be investigated, the interaction between Arm and Ssp appears to be required for the regulation of Ssp localization by Wg signaling. It remains to be investigated whether the mechanisms identified in the salivary gland are applicable to other tissues (Taniue, 2010).

Wingless in the formation of imaginal discs

wingless and dpp are required for allocation of cells to the thoracic imaginal primordia in the germ band extended embryo (corresponding to phase three of dpp expression). Narrow horiontal stripes of DPP intercept vertical stripes of WG secreting cells to form a ladder-like pattern in the ectoderm. It is at the points where WG and DPP stripes intersect that wing and leg imaginal discs are specified and Distal-less is induced (Cohen, 1993).

A third signaling molecule, Hedgehog, is also required for Distal-less induction. HH is secreted from the posterior compartments of imagial discs. Taken together, these three secreted signaling molecules, HH, WG and DPP specify the distal-axis of imaginal discs; each is required for distal-less induction (Diaz-Benjumea, 1994).

Wingless function in imaginal discs

Two thoracic limbs of Drosophila, the leg and the wing, originate from a common cluster of cells that include the source of two secreted signaling molecules, Decapentaplegic and Wingless. Wingless, but not Decapentaplegic, is responsible for the initial distal identity specification of the limb primordia. Proximal limb precursors expressing escargot encircle the Distal-less expressing distal primordium. Dll expressing cells show a dynamic cell migration in the early stage of limb formation, migrating basally during stage 12. Cells that have just started to express Dll also express thickveins. This suggests a requirement for regulated Dpp signaling at the level of receptor expression. Limb formation is restricted to the lateral position of the embryo through exertion of negative control by Decapentaplegic and the EGF receptor, both of which determine the global dorsoventral pattern. dpp specifies proximal cell identities. In the absence of dpp Escargot and Snail are lost. A late function of Decapentaplegic locally determines additional cell identities in a dosage dependent manner. Loss of Decapentaplegic activity results in a deletion of the proximal structures of the limb, in contrast to the deletion of distal structures when decapentaplegic mutations affect the imaginal disc. The limb pattern elements appear to be added in a distal to proximal direction in the embryo, which is just the opposite of what is happening in the growing imaginal disc. It is proposed that Wingless and Decapentaplegic act sequentially to initiate the proximodistal axis. This model is contrary to that of Cohen (1993) who argues that Dpp and Wingless are both required to induce the limb. Since Dll expression persists and expands dorsally in the absence of Dpp, it is clear that Dpp plays no role in inducing initial Dll expression but that the dorsoventral limit of Dll expression is defined by repression as a result of Dpp expression. Similarly, EGF-R is required to repress Dll expression in the ventral ectoderm (Goto, 1997).

Limb development requires the formation of a proximal-distal axis perpendicular to the main anterior-posterior and dorsal-ventral body axes. The secreted signaling proteins Decapentaplegic and Wingless act in a concentration-dependent manner to organize the proximal-distal axis. Discrete domains of proximal-distal gene expression are defined by different thresholds of Decapentaplegic and Wingless activities. distal-less is expressed in a central domain that corresponds to the presumptive tarsal segments and the distal tibia. The dachshund gene is required for development of the femur and tibia. Dac is expressed in a ring corresponding to the presumptive femur, tibia and first tarsal segment, but is absent from the more distal tarsal segments of the leg disc. Although there is little or no overlap between Dll and Dac domains at early stages, by mid third instar the combination of Dac and Dll expression defines three regions along the P-D axis. Dll and Dac are expressed in circular domains centered on the point at which the ventral Wg domain and the dorsal Dpp domain meet. Dll expression in the center of the disc depends on the combined activities of wg and dpp. Wg and Dpp act directly to induce Dll, as analysis of constitutively active Thick-veins clones has shown (Tkv is the receptor for Dpp); analysis of shaggy/zeste white 3 clones (Sgg is required for transduction of the Wingless signal) reveals that both Wg and Dpp transduction pathways are activated cell autonomously. Continuous signalling is not required to maintain Dll or Dac expression. The spatial domains of Dac and Dll expression are defined by different threshold levels of both Wg and Dpp activities. Both Dpp and Wg act to directly repress Dac in the center of the disc. Dac repression is actively maintained by Wg and Dpp signaling long after Dac and Dll have been induced and are stably expressed in the absence of further signaling. Subsequent modulation of the relative sizes of these domains by growth of the leg is required to form the mature pattern (Lecuit, 1997).

aristaless is involved in the allocation of cells to the most distal elements of appendages. Like distal-less, aristaless is expressed at the intersection of Wingless and DPP stripes. Ectopic expression of aristaless is induced by ectopic wingless in regions expressing dpp. One or two cells expressing aristaless then invaginate with the formation of imaginal disc primordia, and are allocated to the distal-most element of appendages (Campbell, 1993).

In the leg disc, HH is secreted by posterior cells and acts at short range to induce dorsal anterior cells to secrete DPP and ventral anterior cells to secrete WG. Complementary patterns of decapentaplegic and wg expression are maintained by mutual repression. DPP signaling blocks wg transcription and WG signaling attenuates dpp transcription. This repression is essential for normal axial patterning because it ensures that the dorsalizing and ventralizing activities of DPP and WG are restricted to opposite sides of the leg primordium and meet only at the center of the primordium to distalize the appendage. A similar dorsoventral bias in the choice of dpp or wg expression is revealed by eliminating the activity of protein kinase A, an experimental intervention that mimics the reception of the HH signal. Constitutive activation of the WG signal transduction pathway by loss of Zeste white (Shaggy) kinase mimics the reception of WG signal, and is sufficient to bias dorsal cells to express wg rather than dpp (Jiang, 1996).

Different thresholds of Wg activity in the wing imaginal disc elicit different outcomes, which are mediated by regulation of decapentaplegic expression and result in alterations in the expression of homeotic genes. A high level of Wg activity leads to loss of all dorsal pattern elements and the formation of a complete complement of ventral pattern elements on the dorsal side of legs, and is correlated with repression of dpp expression. wg expression in dorsal cells of each disc also leads to dose-dependent transdetermination in those cells in homologous discs such as the labial, antennal and leg, but not in cells of dorsally located discs. When dpp expression is repressed by high levels of Wg, transdetermination does not occur, confirming that dpp participates with wg to induce transdetermination. These and other experiments suggest that dorsal expression of wg alters disc patterning and disc cell determination by modulating the expression of dpp. The dose-dependent effects of wg on dpp expression, ventralization of dorsal cells and transdetermination support a model in which wg functions as a morphogen in imaginal discs (Johnston, 1996).

Notch activation at the midline plays an essential role both in promoting the growth of the eye primordia and in regulating eye patterning. Specialized cells are established along the dorsal-ventral midline of the developing eye by Notch-mediated signaling between dorsal and ventral cells. D-V signaling in the eye shares many similarites with D-V signaling in the wing. In both cases an initial asymmetry is set up by Wingless expression. Both Eye and wing cells then go through a distinct intermediate step: in the wing, Wingless represses the expression of Apterous, a positive regulator of fringe (fng) expression; in the eye, Wingless promotes the expression of mirror (mrr), which encodes a negative regulator of fringe (unpublished observations of McNeill, Chasen, Papayannopoulos, Irvine, and Simon, cited by Papayannopoulos, 1998). Both wing and eye cells share a Fng-Ser-Dl-Notch signaling cassette to effect signaling between dorsal and ventral cells and establish Notch activation along the D-V midline. Local activation of Notch leads to production of diffusible, long-range signals that direct growth and patterning, which in the wing include Wingless, but in the eye remain unknown. At least one downstream target of D-V midline signaling, four jointed (fj), is also conserved. four jointed is also expressed in the wing and its expression there is indirectly influenced by Notch (Papayannopoulos, 1998 and references).

During early eye development, fringe is expressed by ventral cells. This expression appears to be complementary to that of the dorsally expressed gene mrr. During early to mid-third instar, additional expression of fng appears in the posterior of the eye disc. This line of posterior fng expression is just in front of the morphogenetic furrow and moves across the eye ahead of the furrow. In the wing disc, Dl and Ser induce each other's expression, and become up-regulated along the D-V border where they can productively signal. Dl and Ser are also preferentially expressed along the D-V midline during eye development. Ser expression, like fng expression, is complementary to that of mrr, whereas Dl expression partially overlaps that of mrr. The spatial relations among fng, Ser, and Dl expression in the eye are thus similar to those in the wing, although in the wing, their expressions are inverted with respect to the D-V axis (Papayannopoulos, 1998).

The Bar homeobox genes function as latitudinal prepattern genes in the developing Drosophila notum. In Drosophila notum, the expression of achaete-scute proneural genes and bristle formation have been shown to be regulated by putative prepattern genes expressed longitudinally. The two Bar locus genes may belong to a different class of prepattern genes expressed latitudinally: it is suggested that the developing notum consists of checker-square- like subdomains, each governed by a different combination of prepattern genes. BarH1 and BarH2 are coexpressed in the anterior-most notal region and regulate the formation of microchaetae within the region of BarH1/BarH2 expression through activating achaete-scute. Presutural macrochaetae formation also requires Bar gene activity. Bar gene expression is restricted in dorsal and posterior regions by Decapentaplegic signaling, while the ventral limit of the expression domain of Bar genes is determined by wingless, whose expression is under the control of Decapentaplegic signaling (Sato, 1999).

The Drosophila notum is considered genetically divided into several longitudinal, side by side, domains whose boundaries are determined by pannier, wingless and iroquois expression (listed respectively from medial to lateral). To further clarify relative locations of pnr, wg and iro expression areas, third-instar larval and pupal future notum were stained with various combinations of molecular markers. In larval and pupal future notum, pnr-Gal4 is expressed medially and iro-lacZ laterally. pnr-Gal4 and iro-lacZ domains partially overlap one another, and wg-lacZ (or Wg) expression is noted in the pnr-iro overlapping region and its immediate neighbors. Bar homeobox genes may belong to an additional class of notal subdivision genes. Staining for BarH1 indicates that BarH1 is expressed latitudinally (anterior vs. posterior) in the anterior-most region of future notum and postnotum. BarH1 expression begins at early to mid third instar. Anti-Ac antibody staining and neur-lacZ expression indicates PS macrochaetae are situated in the vicinity of posterior-ventral corners of the anterior BarH1 expression domain. BarH1 and BarH2 are referred to as Bar collectively and the anterior portion of the prescutum or its precursor expressing Bar is referred to as Bar prescutum. The Bar expression domain overlaps that of pnr, wg and iro. Bar expression similar to that in wing discs is observed in haltere discs (Sato, 1999).

It is concluded that a checker-board-like subdivision of future notum is regulated by putative prepattern gene expression. Future notum may be divided into square subdomains in a checker-board-like manner, each with its own unique combinations of prepattern gene expression. Putative prepattern genes, iro and pnr, form longitudinal domains. Bar homeobox genes form the anterior-most domain. This is the first demonstration of the presence of latitudinal, front to back, prepattern genes in the notum. Bristle formation in each subdomain may be positively regulated by a region-specific combination of prepattern genes. Consistent with this, microchaetae formation in the anterolateral prescutum (the lateral Bar prescutum), where Bar and iro are coexpressed, requires the concerted action of Bar and iro (Sato, 1999).

The Tbx20 homologs midline and H15 specify ventral fate in the Drosophila melanogaster leg

Regional fates in the developing limbs of Drosophila melanogaster are controlled by selector gene transcription factors. Ventral fate in the fly leg is specified by the expression of the ligand Wingless. Evidence is presented that midline and H15, members of the Tbx20 class of T-box transcription factors, are key mediators of the Wingless signal in the formation of the ventral region of the fly leg. midline and H15 are restricted to identical ventral domains of expression through activation by Wingless and repression by the dorsal signal Decapentaplegic. midline and H15 function redundantly and cell autonomously in the formation of ventral-specific structures. Conversely, midline is sufficient to induce ventral fate. Finally, the induction of ectopic ventral fate by mid is compromised when Wingless signaling is attenuated, suggesting that Wingless acts both upstream and in parallel with midline/H15 to specify ventral fate. Based on these results, it is proposed that midline and H15 may be considered as the selector genes for ventral leg fate (Svendsen, 2009).

The Wg-dependent domain is best delineated in the second leg tarsus, where eight rows of bristles are organized around the circumference and run the length of all five tarsal segments. Wg is secreted from a stripe of cells between the primordia of the two ventral-most rows of bristles (1 and 8), which are distinct from more dorsal rows because they are peg-shaped instead of rapier-shaped. The Wg morphogen diffuses to pattern a wedge of the imaginal disc that is broader than and centered on the wg expression domain. In wg hypomorphic mutants, rows 1 and 8 are replaced with a mirror image duplication of dorsal rows 3 through to 6, resulting in a leg with double dorsal symmetry. Similar transformations are observed in clones of cells blocked for Wg signaling, where the row 1/8 bristles are transformed to rapier-shape. Other prominent Wg-dependent ventral structures include the apical bristle (AB) of the distal ventral tibia in the second leg and the ventral transverse rows (TRs) and sex combs (SCs) of the first leg (Svendsen, 2009).

The Tbx20 homologs mid and H15 are essential for the proper development of the Wg-dependent structures in the leg. In the imaginal discs, mid and H15 are expressed in identical ventral domains that are broader than and centered on the Wg domain. In the tarsus, the mid H15 domain is similar to the Wg-dependent domain, encompassing row 1 and 8 bristles and extending to, but not including, rows 2 and 7, as determined by co-staining with an antibody to Achaete, a bristle row marker. Both mid and H15 are activated in ventral cells by Wg and restricted from dorsal cells by the dorsal morphogen Dpp, but neither H15 nor mid alone is essential for leg development. However, loss of both mid and H15 in marked clones caused the autonomous transformation of the Wg-dependent peg-shaped row 1/8 bristles into lateral or dorsal rapier-like bristles. In one sample, 54 out of 56 clones transformed bristles in row 1 or 8. Similar cell-autonomous transformations were observed in the second leg tibia, in which the ventral AB was lost in mid H15 clones that span the distal tibia of the second leg. In 24 out of 26 such clones, a large bristle similar to the dorsally located pre-apical bristle (PAB) developed in place of the AB. The AB is associated with a cluster of peg-shaped bristles called spur bristles (SBs), which, like the row 1/8 bristles, were autonomously transformed to dorsal-like rapier-shaped bristles in mid H15 clones. The SCs and TRs of the first leg were also deleted in mid H15 clones. Other ventral structures were either lost or disorganized within mid H15 clones. Clones located outside the mid H15 expression domain were normal and the few ventral clones with no phenotype were small and located in structures that have no obvious D/V differences (Svendsen, 2009).

The effects of wg mutants and clones of cells unable to detect the Wg signal differ from the effects of mid H15 clones, because they also cause non-autonomous effects such as axis bifurcation or ectopic bristle rows. The axis bifurcation caused by loss of Wg function is due to ectopic dpp expression. However, neither dpp-lacZ nor the dorsal marker omb-lacZ were increased in mid H15 clones located in ventral anterior cells. The ventral-to-dorsal transformation in mid H15 clones is also not a result of a decrease in the expression of Wg, which was unchanged in ventral mid H15 clones. The homeotic gene Sex combs reduced (Scr), which is required for the development of sex combs and TRs, is expressed at high levels in the anterior tibia and basitarsus segments. mid H15 mutant clones in ventral, but not lateral or dorsal, positions downregulate Scr to background anterior levels. Taken together, these results indicate that mid and H15 are required for the specification of ventral fate downstream of Wg and for some ventral gene expression. However, mid and H15 are not required to repress dorsal gene expression (Svendsen, 2009).

Ectopic expression of mid is sufficient to induce ectopic Wg-dependent ventral structures. Since flies with H15 deleted have normal ventral patterning, mid can mediate the function of both genes. Expression of mid in the dorsal omb (bi - FlyBase) domain results in ectopic SCs and TRs in the dorsal basitarsus and distal tibia of all male first legs. This was accompanied by the ectopic expression of Scr in the omb domain, which was appropriately restricted in the P/D axis to the basitarsus and tibia. In the second leg, ectopic expression of mid in the dorsal tibia under the control of the omb-GAL4 or in small clones of mid-expressing cells result in ectopic bristles similar to the AB and SBs. Small clones of mid-expressing cells either in or adjacent to rows 2/7 and 3/6 induce ventral row 1/8 bristles cell autonomously. Similar results were seen using other GAL4 drivers expressed in the tarsus (Svendsen, 2009).

The regions of the leg where mid induces ectopic ventral structures are within the range of the ventral Wg signal, which reaches many dorsal and lateral cells to induce P/D genes such as Dll. This leaves open the possibility that Wg might act both upstream of and in parallel with mid to specify ventral fate. To test the requirement for Wg, clones of cells were generated that were compromised for Wg signaling. Mouse Lef1, which acts as a dominant negative in Wg signaling in Drosophila, was expressed, and its effects on ventral development with and without the expression of ectopic mid were compared. The clones were induced in third instar larvae, at 84 to 108 hours, when the P/D axis is independent of Wg but Wg signaling is still necessary for specifying ventral fate. mid-expressing clones induced at earlier stages can cause more extensive repatterning, with the occasional repression of dpp and non-autonomous induction of wg. Lef1 clones were distributed evenly in the dorsal, ventrolateral, dorsolateral and ventral regions of the tarsus. As expected, dorsal clones were normal and clones in the ventral-most rows often showed transformation towards more dorsal fates (9/26). Clones expressing mid were recovered much more frequently in ventral regions, suggesting that dorsal mid-expressing clones either sort to more ventral positions or they are lost. Ventrolateral or dorsolateral mid-expressing clones are often transformed to ventral character. By contrast, clones expressing both mid and Lef1 are recovered more often in lateral and dorsal cells, indicating that the sorting behavior of mid-expressing clones depends on the transduction of the Wg signal. Dorsolateral clones expressing both mid and Lef1 do not transform towards ventral fate, whereas ventrolateral clones are still sometimes transformed to ventral fate. This is consistent with a requirement for Wg in ectopic ventral development, since the dorsolateral row 3/6 bristles are further from the source of Wg signal and would be expected to be more sensitive to the effects of Lef1. A similar effect on Scr was observed, where UAS-Lef1 blocked Scr expression; this was not rescued by the simultaneous expression of UAS-mid. These results suggest that mid regulates ventral fate and Scr expression in conjunction with Wg (Svendsen, 2009).

These results suggest that the ventral expression of mid and H15 represents a major function downstream of Wg and Dpp in the D/V fate decision. The cell-autonomous requirement for mid and H15 and the ability of ectopic mid expression to induce ventral fate and gene expression in dorsal cells mean that mid and H15 meet the criteria to be defined as selector genes. In the absence of mid and H15, ventral structures may assume a dorsal fate due to the low levels of Dpp signaling found in the ventral leg. However, it is not likely that dorsal is the default fate in the leg, as lateral structures prevail when the expression of both wg and dpp is greatly reduced. Ventral fate also requires Wg signaling, suggesting that mid and H15 act to provide a molecular context for the upstream Wg morphogen to direct ventral-specific patterns of gene expression, as has been observed for other selector genes. The ventral-specific expression of mid, H15 and wg is conserved throughout several arthropod orders, suggesting that it represents a fundamental mechanism in limb patterning (Svendsen, 2009).

Genes acting downstream of Wingless in disc regeneration

Regeneration is a vital process to maintain and repair tissues. Despite the importance of regeneration, the genes responsible for regenerative growth remain largely unknown. In Drosophila, imaginal disc regeneration can be induced either by fragmentation and in vivo culture or in situ by ubiquitous expression of wingless (wg/wnt1). Imaginal discs, like appendages in lower vertebrates, initiate regeneration by wound healing and proliferation at the wound site, forming a regeneration blastema. Most blastema cells maintain their disc-specific identity during regeneration; a few cells however, exhibit stem-cell like properties and switch to a different fate, in a phenomenon known as transdetermination. This study identified three genes, regeneration (rgn), augmenter of liver regeneration (alr) and Matrix metalloproteinase-1 (Mmp1) expressed specifically in blastema cells during disc regeneration. Mutations in these genes affect both fragmentation- and wg-induced regeneration by either delaying, reducing or positioning the regeneration blastema. In addition to the modifications of blastema homeostasis, mutations in the three genes alter the rate of regeneration-induced transdetermination. It is proposed that these genes function in regenerative proliferation, growth and regulate cellular plasticity (McClure, 2008b).

All three identified genes share homology with mammalian genes that have a characterized function in regeneration. Alr, for example, an enzyme with sulfhydryl oxidase activity, has been shown to function as a growth factor in mammalian liver and pancreatic regeneration. Based on the results of in vitro experiments, it is thought that Alr promotes liver regeneration by activating the MAPK cascade via the EGFR, one of the few signals that stimulates hepatocyte proliferation. However, other studies indicate that Alr supports hepatocyte proliferation during liver regeneration by suppressing the activity of liver-resident natural killer cells immediately after injury. Thus, how Alr promotes regeneration in vertebrates is still unclear and the situation remains unresolved by the lack of loss of function studies. Members of the Regenerating gene (Reg) protein family also act as growth factors in pancreatic and stomach regeneration. In vitro experiments indicate that Reg stimulates cell-cycle progression during pancreatic β-cell regeneration by activating the cyclin D1 gene. Consistent with these results, Reg-I knockout mice exhibit slower proliferation and migration of intestinal stem cells, while transgenic mice overexpressing Reg show increased proliferation of gastric progenitor cells. How these Reg-I transgenic mice respond to injury and regenerate has not yet been assessed. Finally, numerous studies have documented the expression of several MMPs in regenerating amphibian and fish appendages. Functional analyses suggest that remodeling of the extracellular matrix by MMPs is important for proper wound healing and blastema formation. Vertebrates have over twenty MMPs whereas flies have only two, Mmp1 and Mmp2. Drosophila Mmp1 is distinct from Mmp2 in that it lacks a predicted membrane anchor. In previous studies (Klebes, 2005) it was found that both Mmp1 and Mmp2 are upregulated during imaginal disc regeneration (McClure, 2008b).

This study used Drosophila leg imaginal discs to investigate how mutations in rgn, alr and Mmp1 affect regeneration, specifically the process of blastema formation and regeneration-induced cellular plasticity. It was found that rgn is involved in the initiation of regeneration by affecting the timing of blastema formation; alr functions in blastema cell proliferation and the extent of regeneration. Finally, Mmp1 delimits regeneration by regulating the cell-cycle arrest of non-blastema cells. The results demonstrate that each of these genes functions early in the regeneration process and this has direct effects on cellular plasticity (McClure, 2008b).

These three genes, rgn, alr and Mmp1, are activated (either directly or indirectly) by Wg signaling during disc regeneration. rgn and Mmp1 are not expressed in the prothoracic leg discs during development, and therefore, these genes are not necessary for the development of these discs. However, during leg disc regeneration, both genes are activated in the blastema and mutations in these two genes affect blastema formation and transdetermination. alr, in contrast, is expressed at cellular blastoderm stage, when disc cells become determined, but is not expressed in the imaginal discs during larval development, when disc cells grow and divide. In addition, alr mutant larvae lack imaginal discs. Together, these observations suggest that alr is required only during the earliest stages of disc development, possibly for their determination, but not for the growth and proliferation of the disc primordia. Activation of alr in the leg disc blastema, and in transdetermining cells, could indicate that this gene product is necessary to establish a new primordium, as in early development. A previous study reported that members of the Polycomb and trithorax Group genes are also involved in regeneration, specifically, several of these genes are up or down regulated during leg disc regeneration (Klebes, 2005). These genes are abundantly expressed throughout development and are necessary to maintain the on or off state of homeotic selector gene expression. Thus, at least three different classes of genes have been found to act during regeneration: (1) genes, like alr, that are re-activated. This observation supports the notion that regeneration reactivates genes that function in early disc development. (2) Genes that are only expressed in the leg disc during regeneration, like rgn and Mmp1, and thus constitute the group of true regeneration genes, and (3) genes that modify their expression level, such as members of the Polycomb and trithorax group genes (McClure, 2008b).

This study identified genes that modify when, where and how much of a blastema is formed. Since fragmentation of wild-type discs produces the same blastema phenotype as wg overexpression, the phenotypes observed are not an artifact of overexpression. Such phenotypes provide valuable information to dissect the process of blastema formation into different steps of gene function. For the understanding of such steps, multiple genes have to be identified that generate the same phenotype. Indeed, previously a failure was found of cell-cycle arrest in non-blastema cells not only in Mmp1^Q112/+ animals, but also jing/+ animals (McClure, 2008a). In addition, a delayed blastema phenotype and reduced transdetermination frequency was reported, similar to the one observed in rgn¹/+ animals, when blastema formation was induced in combgap (cg) and Aly heterozygous animals (McClure, 2008a). The observation that loss of rgn, cg and Aly gene function leads to a delay in blastema formation indicates that these and further studies will provide genetic tools to unravel the mystery of how gene products control the timing of blastema formation (McClure, 2008b).

Why do mutations of these genes have a dominant phenotype that is revealed only in regeneration? Many studies in Drosophila indicate that genetic buffering breaks down under environmentally-induced stress. For example, heat stress uncovers a dominant mutant phenotype in heterozygous recessive multiple wing hairs. More relevant to this study is that heat shock can phenocopy the bithorax mutant phenotype. Furthermore, human epidemiological studies most often correlate increased susceptibility to common diseases, such as cancer, with genetic heterozygosity. It is proposed that regeneration generates a stress situation, resulting in a dominant phenotype. For example, while cell-cycle dynamics are extremely well-buffered during normal development, these studies clearly indicate that genetic heterozygosity plays a noteworthy role in modulating the timing, extent and position of cell division in the blastema (McClure, 2008b).

Overexpression of wg in rgn¹/+ animals causes a delay in blastema formation and decreases regeneration-induced transdetermination. It is speculated for several reasons that these phenotypes may be caused by elevated Wg signaling in rgn mutant animals. First, ubiquitous expression of wg in rgn¹/+ animals leads to the maintenance of cell replication throughout the leg disc instead of blastema formation in the proximodorsal region. This phenotype is in contrast to the wg-expressing controls, where a proximodorsal blastema forms in the leg disc within 48h. However, if Wg signaling activity is further elevated then the maintenance of cell replication, not blastema formation, is observed. Second, previous studies have shown that elevated Wg signaling decreases leg-to-wing transdetermination, which is the phenotype observed in rgn¹/+ animals with wg overexpression. Thus, Rgn may normally attenuate wg signaling, and the loss of one gene copy of rgn could increase Wg activity, leading to a delay in blastema formation and suppression of transdetermination. Preliminary experiments in the wing imaginal disc indicate that overexpression of rgn causes an abnormal accumulation of Wg protein in wg-expressing cells, yet diminishes the extracellular gradient of Wg leading to the loss of wg target genes. As a result, the adult wing blades of these animals are significantly smaller in size. These experiments establish a tentative molecular link between rgn and the Wg signaling pathway (McClure, 2008b).

It is interesting to note that the Drosophila Rgn protein contains a CTLD at its N-terminus. Such domains have been reported to bind a variety of ligands, principally carbohydrates, but also glycoproteins, lipids, receptors and inorganic surfaces. Wg, as a glycoprotein, could potentially bind Rgn. In addition, a receptor for the mammalian Reg protein has been identified and it shares significant sequence homology to the Exostoses-like-3 (EXTL3) gene, which in Drosophila is brother of tout-velu (botv). Drosophila Botv functions as a glycosyltransferase required for the biosynthesis of heparan sulfate (HS) glycosaminoglycan (GAG) chains and for Wg signaling and distribution. Taken together, these observations support the notion that rgn may play a role in Wg signaling (McClure, 2008b).

Blastema formation in Mmp1^Q112/+ animals differs from controls in that cell-cycle exit in the ventral aspect of the disc is incomplete. It is hypothesized that continued proliferation of cells at the ventral site may lead to ectopic Vg expression, and thus transdetermination, in the ventral half of Mmp1^Q112/+ discs. Together, these phenotypes suggest that normally Mmp1 limits regeneration as well as cellular plasticity. What is the possible cause for Vg expression at ventral sites of Mmp1 leg discs? Ectopic Vg expression in both dorsal and ventral regions of leg discs is a phenotype previously observed when ectopic wg and dpp are coexpressed in the leg disc. Therefore, it is possible that during leg disc regeneration dpp expression and/or signaling is altered, and possibly elevated, by loss of one gene copy of Mmp1. Indeed, flies with hypomorphic mutations of Mmp1 exhibit clefts along the midline of the notum, a phenotype also observed by mutations in the dpp gene or interfering with dpp signaling. These studies indicate that loss of Mmp1 interferes with dpp expression and/or signaling, while the results suggest that during disc regeneration loss of Mmp1 may elevate dpp expression and/or signaling. Thus, it appears that the interaction between Mmp1 and Dpp signaling may be complex and different during regenerative processes (McClure, 2008b).

Drosophila has two Mmps, Mmp1 and Mmp2, and by microarray analysis, both are upregulated during imaginal disc regeneration (Klebes, 2005). However, only Mmp1 functions in the regeneration and transdetermination process, while Mmp2 has no detectable effect. Mmp1 differs from Mmp2 by lacking a membrane anchor; these observations may lend insight into the function of the broad vertebrate Mmp gene family in regeneration (McClure, 2008b).

Overexpression of wg in animals heterozygous for alr presents the most striking phenotype: an obvious suppression of wg-induced disc overgrowth and blastema size as well as an enhancement of transdetermination frequency. In yeast these enzymes localize to the mitochondrial intermembrane space and are involved in respiration, vegetative growth and biogenesis of mitochondrial intermembrane space proteins. In vertebrates, Alr is ubiquitously expressed and found in different subcellular localizations, including the cytosol. During liver regeneration, Alr is secreted from hepatocytes and functions as a co-mitogen with other growth-promoting factors such as HGF, EGF and TGFα. Thus, Alr proteins have a diverse array of functions. In this study, heterozygosity for alr effectively interferes with blastema growth as well as the growth-promoting effects of ectopic wg expression. This is the first time that loss of alr function and its consequences on regeneration have been examined. In addition, the study points to a new mechanism for how Alr functions in regeneration, by acting (directly or indirectly) downstream of Wg signaling (McClure, 2008b).

These results indicate that rgn, alr and Mmp1 function in the timing, extent and position of the regeneration blastema, as well as in the regulation of regenerative plasticity. rgn, alr and Mmp1 homologs have all been implicated in the process of regeneration in vertebrates, and now, invertebrates, which suggests that they play a broadly conserved role in organ and tissue regeneration. Such findings indicate that the function of Wg signaling in epimorphic regeneration can be investigated using Drosophila imaginal discs. Without doubt, the genome-wide access in Drosophila will identify additional regeneration genes, and thus broaden molecular and genetic understanding of the regenerative process. Ultimately, this information will aid in the development of therapies to replace or re-grow tissues lost to disease or injury (McClure, 2008b).

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.