DISCUSSION

Comparative effects of trx¹ on homeotic phenotypes: The P&E profiles of trx¹ genotypes suggest different quantitative requirements for TRX at the four homeotic loci examined. Three factors may contribute to the different sensitivities of target genes to decreased levels of TRX: (1) Each target gene accumulates a unique amount of TRX at its PRE(s). Genes that normally accumulate less TRX may be more sensitive to decreased levels of TRX. (2) Target genes with normally equal TRX accumulation may have different threshold TRX levels below which they no longer function for proper structural determination. (3) Different tissues undergo different numbers of cell divisions, which may lead to differential loss of limiting quantities of TRXs. These three factors are further described below.

1: Polytene chromosomal analyses show that different target genes do accumulate different amounts of TRX at their PREs (Kuzinet al. 1994; Chinwallaet al. 1995). The level of accumulation of TRX proportionally decreases at all target genes in trx¹ mutants, which shows that target genes that normally accumulate more TRX recruit limiting amounts of TRX more efficiently. Even if there is not a linear relationship between the amount of TRX accumulated at a target gene and the amount needed for function, target genes that normally accumulate less TRX should be more susceptible to reduced nuclear TRX concentration than those that normally accumulate more. Due to low resolution, polytene chromosomal analysis did not determine if Scr in the ANT-C and Ubx, abd-A, and Abd-B in the BX-C accumulate different amounts of TRXs (Chinwallaet al. 1995).

2: Some trx target genes in specific tissues may require relatively small amounts of TRX to be transcribed sufficiently for normal development, though they may normally be supplied with abundant TRX. Only very low TRX levels would elicit a mutant phenotype in such tissues. Reduced trx function would produce low P&E of phenotypes associated with such a relatively TRX-insensitive target gene.

3: trx¹ mutants have reduced maternal and/or zygotic production of TRX. Thus, target genes in imaginal precursor cells of trx¹ mutants initially accumulate less TRX than those in wild-type cells. During subsequent cell divisions, target genes are increasingly susceptible to insufficient TRX accumulation caused by continually impaired trx transcription or translation. Different imaginal tissues begin with different numbers of precursor cells and proliferate to different extents (Cohen 1993). Target genes expressed in imaginal tissues with more precursor cells and greater proliferation would be more likely to have TRX accumulation fall below threshold levels than those in tissues with fewer precursors and less proliferation. This scenario requires that all imaginal precursor cells initially have similarly reduced levels of TRX and all proliferating imaginal cells have the same reduced transcription or translation of trx.

Comparative effects of other hypomorphic alleles on homeotic phenotypes: Hypomorphic trx alleles encode proteins that can assemble and provide some function at target genes. They function in combination with one wild-type maternal dose of trx for seemingly normal embryogenesis, which is consistent with their complementation of trx¹ during imaginal development. However, a maternal dose of trx⁺ is not sufficient to complement hypomorphic mutant function in imaginal precursors whose progeny cells produce only mutant protein.

P&E qualitative profiles produced by different hypomorphic genotypes, excluding trx¹ and trx^E3, are proportionally similar. This suggests that the mutant proteins equivalently impair TRX function at the homeotic genes examined. However, instances of complementation among hypomorphic alleles suggest different mutations alter different functional domains. These observations infer, not surprisingly, that different hypomorphic mutant proteins inefficiently interact with different factors present at many, if not all, target genes.

trx^Z16 and trx^Z11 are associated with point mutations in the PHD finger and SET domains, respectively (Stassenet al. 1995). As mentioned previously, the SET domain of HRX at least mediates protein-protein interactions used in signal transduction and maturation (Cuiet al. 1998; De Vivoet al. 1998). PHD fingers may also mediate protein-protein interactions (Aaslandet al. 1995). trx^M17, trx^Z32, and trx^M18 are point or pseudopoint mutations (Breen and Harte 1991) that impair different functional interactions as determined by complementation. Disruption of any TRX protein-protein interaction would reduce the ability of a mutant protein to function at any target gene. Results of this study show that Ubx transcription in T3 leg discs is most easily affected by any of several small changes in different regions of TRX. The observed different sensitivities of trx target genes to a variety of mutant TRX infers that each target gene employs TRX uniquely, though similar factors are present at each.

Proteins encoded in hypomorphic alleles have greater function at abd-A than reduced levels of TRX in trx¹ mutants. There does not appear to be any quantitative mechanism that could produce this difference. It is unlikely that maternally supplied, wild-type TRX could endure to complement hypomorphic mutant proteins through tergite development. It seems more likely that hypomorphic mutations do not compromise the ability of their mutant proteins to function at abd-A as much as they do at other homeotic target genes. This is consistent with the previous comment that TRX has unique structural interactions at each of its target genes.

In hypomorphic trx mutants, there is higher penetrance of A2 and A3 tergite transformations to A1 than of more posterior tergite transformations. Thus, hypomorphic mutant trx proteins are less efficient in stimulating abd-A transcription in A2 and A3 than in more posterior tergites. Reduced levels of wild-type TRX in trx¹ mutants also have a greater effect on A2 and A3 than on more posterior tergites. In wild-type flies, abd-A is expressed at lower levels in PS7 and PS8 than in more posterior parasegments (Karchet al. 1990). Thus, moderately impaired TRX may lower abd-A expression slightly but still below a threshold needed to ensure proper A2 and A3 development. Reduced levels of TRX in trx¹ mutants may lead to a greater decrease in abd-A transcription sufficient to affect development of more posterior tergites.

trx^JY16 encodes wild-type TRXI and a TRXII fusion protein that has at least 172 N-terminal residues replaced by fusion partner residues (Figure 1). These proteins must be present in sufficient quantity with sufficient function for successful embryogenesis. However, they do not support larval growth. trx^JY16 does supply significant function to complement trx¹, and it can form functionally impaired complexes with some of the other hypomorphic proteins. Normally, large mRNAs that encode TRXII are the only forms expressed during larval and pupal stages (Breen and Harte 1991; Sedkovet al. 1994); however, they also encode TRXI. It may be that TRXII supplies trx function for larval growth and imaginal development, though a role for TRXI cannot presently be excluded. If the larval and imaginal role for TRXII is correct, it is likely that the fusion TRXII encoded in trx^JY16 cannot supply that function. This implies that the N terminus of TRXII is necessary for proper stimulation of target gene transcription in imaginal cells.

In combination with other strong hypomorphic alleles, trx^JY16 partial complementation of Ubx function in haltere discs, but not T3 leg discs, demonstrates tissue-specific use of TRX at the same target gene. This is consistent with the observation that trx has tissue-specific effects on expression of Ubx and other homeotic target genes during embryogenesis (Breen and Harte 1993). Abnormal tracheal development (Vincentet al. 1997; Wappneret al. 1997) and lack of larval growth in trx^JY16 mutants may indicate that TRX regulates target gene expression as an element of DPP and/or EGFR signaling pathways (see below).

The protein encoded in trx^JY21 interferes with residual function of the other hypomorphic alleles at Scr, Ubx, and Abd-B, suggesting that it is antimorphic, but its interference with trx function at Scr is most noticeable. This is particularly evident when comparing trx¹/amorph to trx¹/trx^JY21 P&E profiles. These results allow that trx^JY21 protein may inhibit wild-type and hypomorphic TRX from forming into complexes or interfere with their function in complexes. Regardless, its differential effect on Scr expression again illustrates the differential use of TRX in the context of other factors at different target genes. In combination with strong hypomorphic alleles, trx^JY21 may prove particularly useful in that it appears to almost completely remove trx function at Scr in T1 leg discs and Ubx in T3 leg discs.

Male genitalia to leg phenotype: trx^Z11, trx^Z16, and trx^M18 in combination with trx^JY21 often produce pharate adult males with genitalia transformed to T2 leg. This may occur because reduced Abd-B r expression in A9 primordia of male genital discs allows Antp P2 expression, which contributes significantly to leg development (Schneuwly and Gehring 1985; Abbott and Kaufman 1986), to have increased influence on the specification of these cells. Casares et al. (1997) showed that Abd-B r is expressed in male genital disc cells that derive from A9 and produce male genitalia (Freeland and Kuhn 1996). It is not known if Antp P2 is expressed in male genital discs, but it is expressed in A9 during embryogenesis (Berminghamet al. 1990). In A9 of trx mutant embryos, Abd-B r expression is reduced (Duncan and Lewis 1982; Breen and Harte 1993) and Antp P2 expression appears nearly wild type (T. R. Breen, unpublished results). Thus, it is possible that male genital disc precursors from A9 have relatively enhanced ANTP expression from Antp P2 directing T2 leg development. This situation is unique to male genital discs. Though female genital disc precursors would experience the same relative levels of Abd-B r and Antp P2, cells that express Abd-B r in female genital discs do not produce female genitalia (Freeland and Kuhn 1996; Casareset al. 1997).

Model of trx function: TRX is recruited to PREs of target genes (Chanet al. 1994; Kuzinet al. 1994; Chinwallaet al. 1995; Orlandoet al. 1998). Once assembled, it acts with other trxG proteins to stimulate target gene transcription through chromatin remodeling as inferred by the SET domain it shares with other proteins known to alter chromatin (Jones and Gelbart 1993; Tschierschet al. 1994). Pirrotta (1998) offers the interesting possibility that TRX influences the level of target gene histone acetylation. It is not clear if TRX participates in the initial transcription of its target genes. In specific cells, it is necessary for detectable levels of target gene transcription. In others, it is needed only for enhanced target gene transcription (Breen and Harte 1993; Kuzinet al. 1994; Sedkovet al. 1994; Breenet al. 1995). Functional TRX assemblies are inherited by progeny cells (Ingham 1981, 1985a) so that they will have levels of target gene transcription similar to their parent cells. From this information, it is conceivable that TRX assembles in a lineage-dependent manner to act as a constitutive facilitator of other transcription factors. If transcription of a target gene is not initiated in a cell, PcG proteins bound to the gene’s PRE chromatin form a silencing structure that supersedes colocalized TRX (Pouxet al. 1996). A gene’s PcG protein-silencing structure is then inherited by progeny cells.

Phenotypic and gene expression analyses of two trxG genes, ash2 (Adamson and Shearn 1996) and mor (Brizuela and Kennison 1997), suggest that their proteins participate in downstream functions of developmental signaling pathways. Phenotypic results of this study allow that TRX activity may be modulated downstream of cell signaling to attain cell-specific levels of target gene transcription. This role of TRX is supported by findings that propose a similar role for HRX. The interaction of a dual-specificity phosphatase inhibitor with the SET domain of HRX suggests it is activated through signal transduction and later deactivated to promote differentiation (Cuiet al. 1998; De Vivoet al. 1998). In this light, I present a model in which TRX acts as a downstream mediator in multiple signal transduction pathways, including those signaled by morphogens, to elicit ligand concentration-dependent responses at target genes. Here, TRX provides a dynamic response capacity to a variety of cell-signaling events.

• Download figure
• Open in new tab
• Download powerpoint

Figure 7.

—Model of TRX activation through signal transduction. The model is based on experimental findings on UBX and DPP regulatory interactions in the visceral mesoderm and DPP receptor functions in imaginal discs (see discussion). UBX with other factors (Sunet al. 1995) initiates dpp transcription. DPP may signal the cell from which it came (autocrine, shown in model) or nearby cells (exocrine). DPP signals through receptor heterodimers. Each receptor heterodimer consists of one TGF-β type I subunit and one TGF-β type II subunit. In Drosophila, thickvein and saxophone proteins (TKV and SAX) are type I receptors and punt protein (PUNT) is a type II receptor. DPP signaling through undetermined intermediates may lead to phosphorylation (^*) of CREB bound to a promoter proximal CRE (Ereshet al. 1997). Phosphorylated CREB recruits CBP (Chriviaet al. 1993) to the promoter proximal region where it may participate in histone acetylation that may form transcriptionally permissive chromatin (Bannister and Kouzarides 1996; Ogryzkoet al. 1996). DPP signaling may additionally act through chromosomal trxG proteins including TRX to boost Ubx expression. Upstream elements of this signaling cascade may include the known DPP-signaling intermediate, MAD, and perhaps nonchromosomal trxG proteins. Signaling may affect any combination of chromosomal trxG proteins that would subsequently interact in an unknown way to augment Ubx transcription. Other known chromosomal trxG proteins include ASH1 (Tripoulaset al. 1996), GAGA factor (Farkaset al. 1994), and perhaps BRM (Dingwallet al. 1995), which is at least nuclear. These may work cooperatively or in parallel. They may be components with TRX of a signal response pathway, or they may function independently to prepare Ubx for elevated transcription. TRX transcriptional augmentation appears to operate in cells where Ubx was previously activated. PcG proteins silence Ubx transcription in cells where there is no Ubx transcription when PcG proteins are activated during germ band elongation. Therefore, TRX mediates DPP-signaled elevation of Ubx transcription in cells where Ubx is previously activated and its previous activation prevents PcG silencing. Other target genes in other cells may similarly use TRX to respond to DPP and other signal transduction pathways. The TRX-mediated signaling response of a cell may be quantitatively controlled by the concentration of a signaling ligand within its gradient distribution and by the number of TRXs recruited to a target gene’s regulatory elements.

The model shown in Figure 7 is based on activities in PS7 of the visceral mesoderm (Immerglücket al. 1990; Panganibanet al. 1990; Reuteret al. 1990; Hurshet al. 1993; Thüringer and Bienz 1993; Staehling-Hampton and Hoffmann 1994) where trx is needed for normal levels of Ubx expression (Breen and Harte 1993). However, other ligands and their receptors may be substituted to account for the effects of trx mutations on other cell types. In the model, TRX is modified as a downstream substrate of signaling pathways, whose ligands may include the TGF-β homologue (Padgettet al. 1987), decapentaplegic protein (DPP), the WNT-1 homologue (Rijsewijket al. 1987) wingless protein (WG), hedgehog protein (Mohler and Vani 1992; Tabataet al. 1992; Ingham and Hidalgo 1993; Tashiroet al. 1993), and ligands of the Drosophila EGFR (Livnehet al. 1985; Thompsonet al. 1985; Wadsworthet al. 1985) such as spitz protein (Rutledgeet al. 1992). Signaling intermediates may include factors such as Mothers against dpp (Mad) and schnurri (shn) proteins in the DPP pathway (Griederet al. 1995; Staehling-Hamptonet al. 1995; Newfeld et al. 1996, 1997). They may also include known chromosomal trxG proteins and trxG proteins that may be proven to be nonchromosomal (Kennison and Tamkun 1988; Farkaset al. 1994; Kennison 1995; Adamson and Shearn 1996; Tripoulaset al. 1996; Rozenblatt-Rosenet al. 1998).

It is also possible that TRX is required for normal levels of expression of signaling pathway genes. Thus, trx mutants would develop hypomorphic signal transduction phenotypes. The expression of dpp in trx mutants does not support this possibility (T. R. Breen, unpublished results), though the expression of many other signaling element genes in trx mutants needs to be examined.

Below, I interpret aspects of trx mutant phenotypes that are consistent if TRX is modulated by signaling intermediates.

Higher frequency of anterior transformations: As noted by Ingham and Whittle (1980), trx¹ genotypes produce adults with a higher frequency of transformed patches that include only anterior metameric structures compared to transformed patches that include more posterior structures. I observed that successively stronger hypomorphic genotypes produce successively larger transformed patches that encompass increasingly more posterior regions of affected adult segmental structures. It is important to note that these transformed posterior segmental structures are not limited to posterior compartment derivatives, but include structures derived from more posterior regions of anterior compartments. Thus, precursor cells that give rise to structures closer to the anterior margin of an adult segment are more susceptible to reduced trx function than precursor cells that give rise to more posterior structures, including those of posterior compartment origin.

Successively more severe trx hypomorphic genotypes reveal an anterior-to-posterior gradient of TRX activity. This activity gradient may be due, in whole or in part, to an anterior-to-posterior gradient of TRX in imaginal tissues, as is seen in thoracic discs (Kuzinet al. 1994). It is also possible, as suggested in the model presented above, that TRX activity is dependent not only on TRX levels at homeotic and other target genes, but may be modulated in response to gradients of DPP, WG, hedgehog protein (HH), or spitz protein (SPI) that are generated near segmental anterior/posterior (A/P) borders. Threshold levels of TRX may exist below which signal input is insufficient to effect or enhance target gene transcription. Morphogen concentrations diminish at increasing distances from their sources. Weak trx hypomorphs may produce insufficient TRX to interpret low morphogen concentrations only in anterior cells that are most distant from morphogen sources. Stronger trx hypomorphs may not have sufficient TRX to interpret morphogens even in cells at a morphogen source during imaginal proliferation. Consistent with either possible cause of the graded trx mutant phenotype is the observation that Ubx expression is diminished in anterior cells of haltere and T3 leg discs in moderate trx hypomorphs (Cabreraet al. 1985; Ingham 1985b). How hypomorphic trx mutations affect expression of other target genes in other imaginal tissues remains to be seen.

Uab phenotype: Many trx hypomorphic genotypes produce a Uab phenotype, particularly the development of dark pigmentation and large bristles at the posterior of A1 tergites. The Uab phenotype was originally proposed to be caused by abd-A dominant gain-of-function alleles producing ectopic abd-A protein (ABD-A) in A1 tergite precursors (Lewis 1978). Most Uab mutations do not noticeably alter ABD-A expression patterns (Karchet al. 1990) including those associated with the recessive abd-A phenotype examined in this study. Similarly, ABD-A is not ectopically expressed in A1 tergite precursors in trx mutants (Breen and Harte 1993).

The posterior of a tergite is the posterior of a parasegment corresponding to the posterior of an anterior compartment (Kornberg 1981; Hamaet al. 1990). Tergites develop from anterior dorsal histoblasts whose precursors arise in anterior abdominal compartments during embryogenesis. Struhl et al. (1997) showed that dark pigmentation and large bristles develop in A2–A6 tergites when anterior dorsal histoblasts receive HH signals from adjacent posterior dorsal histoblasts. This process should also occur at the A/P border of the A1 tergite, yet its posterior border is unpigmented and develops small bristles typical of the rest of the tergite.

A difference present at the A1 A/P border not present at the A/P borders of more posterior segments is the relative level of expression of UBX and ABD-A. UBX is expressed in reiterated gradients in PS7-12 with high levels at the posterior and lower levels toward the anterior of each parasegment (White and Lehmann 1986). UBX is expressed at very high levels throughout PS6. ABD-A is expressed in reiterated gradients in PS7-12 with high levels at the anterior and lower levels toward the middle of each parasegment (Karchet al. 1990). ABD-A gradients in PS7 and PS8 are weaker than in more posterior parasegments. Thus, there are very high levels of UBX on the anterior side of the A1 A/P border juxtaposed to moderate levels of ABD-A on the posterior side of the border. More posterior segments have lower levels of UBX juxtaposed to higher levels of ABD-A. Additionally, there is no ABD-A on the anterior side of the A1 A/P border, whereas there are low levels of ABD-A on the anterior sides of A/P borders of more posterior segments.

It is possible ABD-A contributes to HH signaling from posterior dorsal histoblasts and UBX contributes to its interpretation in anterior dorsal histoblasts. In trx and Uab mutants, relative levels of UBX and ABD-A at A1 A/P borders may establish and interpret HH signaling as at more posterior abdominal A/P borders. Weatherbee et al. (1998) showed UBX contributes to regulation of genes downstream of HH signaling in imaginal discs, which may be thought of as an interpretation function. Again, it will be interesting to determine if this downstream regulation by Ubx is parallel to signaling regulation or more directly controlled by HH signaling through TRX.

Other possible modulation of TRX activity by signal transduction pathways: trx^Z11 and trx^E3 genotypes frequently cause ocelliless phenotypes similar to those caused by decreased oc and Egfr function (Clifford and Schupbach 1989; Finkelsteinet al. 1990). EGFR may function through the pointed (pnt) transcription factor (PNT) to regulate oc transcription in eye imaginal discs as it does in embryogenesis (Gabayet al. 1996). It is noteworthy that the cytological location of pnt (94F) is a site of TRX localization (Chinwallaet al. 1995). A possible scheme is that EGFR activation of TRX boosts pnt expression leading to proper levels of oc expression. Loss of function of any of these genes would result in ocelliless phenotypes. It is also possible that TRX may mediate HH signaling for proper oc expression as HH also affects dorsal head development (Royet and Finkelstein 1996).

A possible target of the oc transcription factor (often called by its alternative name, orthodenticle protein, or OTD) in eye discs is en (Royet and Finkelstein 1995). en is a target of trx function, particularly in later developing cells (Breenet al. 1995). It is possible that reduced en function in trx and oc mutants contributes to the ocelliless phenotype. However, it has yet to be demonstrated that loss of en activity produces an ocelliless phenotype (Lawrence and Struhl 1982), and, unfortunately, dorsal head structures were not examined in en; trx double mutants (Breenet al. 1995). Regardless, it appears that en contributes only to ocellar development (Royet and Finkelstein 1995). Hence, reduced en function alone cannot account for the full range of ocelliless phenotypes seen in trx mutants. Perhaps TRX stimulates the transcription of multiple genes affected by a signaling pathway, such as pnt and en, creating positive impetus toward a particular level of cell fate determination.

trx^Z11, trx^M18, and trx^E3 genotypes frequently cause growth deficiencies manifested as small or incomplete head development, small anterior thorax, and incomplete chitinization. These phenotypes suggest incomplete development of imaginal tissues that may be caused by abnormal cell death or impaired cell proliferation. DPP and EGFR pathways are necessary for normal growth and differentiation of imaginal tissues (Spenceret al. 1982; Clifford and Schupbach 1989; Baker and Rubin 1992; Burke and Basler 1996). Perhaps TRX mediates growth control by these signaling pathways through homeotic and uncharacterized target genes. Phenotypic analyses of trxG members ash1 and ash2 suggest they similarly function downstream in cell proliferation and differentiation pathways and their proteins may act in concert with TRXs (Shearnet al. 1987; Shearn 1989; Adamson and Shearn 1996).

trx^Z11 genotypes produce growth deficiency phenotypes seen in dpp and proximal ANT-C mutants. Abnormal antennal outgrowths similar to those caused by dpp disc III mutations may indicate that TRX interprets DPP signals in antennal discs. trx^Z11 mutants occasionally have reduced or missing maxillary palps seen in Dfd, lb, and pb hypomorphs. It is apparent that Dfd, lb, and pb contribute to normal growth of maxillary palps (Merrill et al. 1987, 1989; Pultzet al. 1988), but it is not known how these genes respond to developmental signals during imaginal proliferation. During embryogenesis, trx function is needed for elevated levels of Dfd expression in the anterior of its domain (Breen and Harte 1993). It is not known if trx is similarly required for normal Dfd, lb, and pb expression during imaginal proliferation. It will be interesting to determine the role of trx and homeotic genes in regulating imaginal proliferation as a component of developmental signaling responses.

trx^Z11 is associated with the change of a conserved glycine to a serine in the SET domain (Stassenet al. 1995). Cui et al. (1998) showed that Sbf1 binds with the SET domain of HRX and may oppose maturation and differentiation promoted by the interaction of the SET domain with dual-specificity phosphatases such as myotubularin. It is possible that in imaginal cells the trx^Z11 mutation prevents protection by an Sbf1 homologue, allowing premature growth repression and differentiation.

It is difficult to guess how trx^M18 and trx^E3 may interfere with cell proliferation and differentiation. trx^M18 is uncharacterized and may map to the SET domain or a region of TRX that interacts with it. It may also interfere with normal phosphorylation, thus inhibiting the activation of TRX needed to promote growth. The region of TRX missing due to the trx^E3 deletion may also be necessary for SET domain function or signal reception. In imaginal cells, trx^E3 appears to affect ANT-C transcription primarily as it does in embryogenesis (Sedkovet al. 1994), yet its growth deficiencies appear to affect all imaginal tissues. It is possible that the deletion reduces the ability of the protein to assemble at ANT-C PREs or receive developmental signal input unique to the ANT-C in head and thoracic imaginal cells. At the same time, it may reduce the protein’s ability to assemble into structures or receive signal input at target genes involved with cell proliferation in many tissues.

trx^M17/amorph genotypes occasionally cause labial palp transformations, unilateral eye duplications, posterior wing abnormalities, and anterior wing duplications. Labial palp transformations suggest that trx^M17 proteins have reduced function at some combination of pb and Scr in labial discs (Percival-Smithet al. 1997). Mirror image eye duplications are reminiscent of phenotypes associated with disruption of a signal transduction pathway, yet this phenotype is not associated with a described signaling mutation. The posterior wing compartment effects are similar to those seen in en; trx double hypomorph mutants (Breenet al. 1995), posterior compartment en mutant clones (Lawrence and Struhl 1982), or clones mutant for the DPP receptor SAX (Singeret al. 1997). The anterior compartment duplications are similar to those produced by clones mutant for the DPP receptor PUNT (Penton and Hoffmann 1996). trx^M17 proteins may have reduced activity at en. Reduced en function in the wing disc may impair HH signaling that could reduce the level of DPP expressed at the A/P border (Zeccaet al. 1995). Reduced DPP levels would produce a weaker gradient of this morphogen in the wing disc, leading to the dpp deficient phenotypes. These phenotypes are unique to trx^M17, indicating that the mutation disrupts some specific target gene interactions, but the mutation also has a low P&E effect on other homeotic genes.

Differences between the phenotypes discussed above for hypomorphic genotypes and those observed by Ingham (1985a) in trx^- somatic clones are probably due to differences in alleles used and timing of trx loss of function. In this study, the source of wild-type TRX was from a single maternal dose of trx⁺ from heterozygous mothers or reduced amounts supplied by trx¹. Otherwise, only mutant gene products were available throughout development. trx^- clones derived from progenitors that had one maternally contributed dose of wild-type trx mRNA and one wild-type zygotic dose until the first or second larval instar time of clone induction. Thereafter, progeny cells had no source of functional TRX except reserves of wild-type products that diminish through each round of cell division.