The Abdominal-B (Abd-B) gene, a member of the bithorax complex (BX-C), specifies the identities of parasegments (PS) 10-14 in Drosophila. Abd-B codes for two structurally related homeodomain proteins, ABD-B m and ABD-B r, that are expressed in PS10-13 and PS14-15, respectively. Although ABD-B m and r proteins have distinct developmental functions, ectopic expression of either protein during embryogenesis induces the development of filzkörper and associated spiracular hairs, structures normally located in PS13, at ectopic sites in the larval thorax and abdomen. These results suggest that other parasegmental differences contribute to the phenotype specified by ABD-B r activity in PS14.

Both ABD-B m and r repress the expression of other homeotic genes, such as Ubx and abd-A, in PS10-14. However, the importance of these and other cross-regulatory interactions among homeotic genes has been questioned. Since ectopic UBX protein apparently failed to transform abdominal segments, González-Reyes et al. (González-Reyes, A., Urquía, N., Gehring, W. J., Struhl, G. and Morata, G. (1990). Nature 344, 78-80) proposed a functional hierarchy in which ABD-A and ABD-B activities override UBX activity. We tested this model by expressing UBX and ABD-B m proteins ectopically in wild-type and BX-C-deficient embryos. Ectopic ABD-B m does not prevent transformations induced by ectopic UBX. Instead, ectopic UBX and ABD-B m proteins compete for the specification of segmental identities in a dose-dependent fashion. Our results support a quantitative competition among the homeotic proteins rather than the existence of a strict functional hierarchy. Therefore, we suggest that cross-regulatory interactions are not irrelevant but are important for repressing the expression of competing homeotic proteins.

To explain the apparent failure of ectopic UBX to transform the abdominal segments, we expressed UBX at different times during embryonic development. Our results show that ectopic UBX affects abdominal cuticular identities if expressed during early stages of embryogenesis. In later embryonic stages, abdominal segments become resistant to transformation by ectopic UBX while thoracic segments remain susceptible. Head segments also show a similar stage-dependent susceptibility to transformation by ectopic UBX in early embryogenesis but become resistant in later stages. These results suggest that abdominal and head identities are determined earlier than are thoracic identities.

Different segmental identities in Drosophila are established by the action of the homeotic genes. The homeotic genes of the bithorax complex (BX-C), Ultrabithorax (Ubx), abdominal-A (abd-A) and Abdominal-B (Abd-B), determine the identities of the third thoracic (T3) and first through ninth abdominal (A1-A9) segments (Lewis, 1978), corresponding to parasegments (PS) 5 through 14. Ubx is required for the proper development of T3 and A1 (PS5, 6); Ubx and abd-A specify A2-A4 (PS7-9) identities; and Ubx, abd-A and Abd-B determine A5-A8 (PS10-13) identities (reviewed in Duncan, 1987; Morata et al., 1990). The domains of BX-C gene function correspond to domains of BX-C protein expression: UBX is expressed in PS5-13 (White and Wilcox, 1985; Beachy et al., 1985), ABD-A in PS7-13 (Karch et al., 1990; Macías et al., 1990) and ABD-B in PS10-15 (Celniker et al., 1989; DeLorenzi and Bienz, 1990; Boulet et al., 1991) (Fig. 1A).

Fig. 1.

(A) Summary of Ultrabithorax (Ubx), abdominal-A (abd-A) and Abdominal-B (Abd-B) m and r expression patterns in the ectoderm at germband extension. Black boxes represent high levels of expression. Unshaded hatched boxes represent low levels of expression. The light to dark shading of the hatched boxes for Abd-B m represents its graded expression. (B) Structure of the hs-m and hs-r fusion constructs used to express ABD-B m and r proteins ectopically. Unshaded boxes represent non-coding Abd-B sequences. Translational start and stop codons delimit the open reading frames (ORF). The lightly stippled box represents amino-terminal sequences unique to the Abd-B m ORF. Darkly stippled boxes represent sequences shared by Abd-B m and r ORFs. Black boxes denote the identical homeoboxes (HB). The thin lines flanking the 5′ and 3′ Abd-B non-coding regions respectively represent hsp70 promoter and 3′ untranslated sequences which include the polyadenylation signal. Arrows indicate transcriptional start sites. (C) Photomicrograph showing a lateral view of the tail region of a first instar larva (dorsal surface at top). The normal locations of two structures (fz, filzkörper; sph, spiracular hairs) scored in this study are indicated. (D) Schematic drawing of the tail region of a first instar larva adapted from Sato and Denell (1986) and Jürgens (1987). Thin dashed lines indicate segmental boundaries. Thick dashed lines indicate parasegment (PS) 13 boundaries. Rows of black squares denote ventral denticle belts.

Fig. 1.

(A) Summary of Ultrabithorax (Ubx), abdominal-A (abd-A) and Abdominal-B (Abd-B) m and r expression patterns in the ectoderm at germband extension. Black boxes represent high levels of expression. Unshaded hatched boxes represent low levels of expression. The light to dark shading of the hatched boxes for Abd-B m represents its graded expression. (B) Structure of the hs-m and hs-r fusion constructs used to express ABD-B m and r proteins ectopically. Unshaded boxes represent non-coding Abd-B sequences. Translational start and stop codons delimit the open reading frames (ORF). The lightly stippled box represents amino-terminal sequences unique to the Abd-B m ORF. Darkly stippled boxes represent sequences shared by Abd-B m and r ORFs. Black boxes denote the identical homeoboxes (HB). The thin lines flanking the 5′ and 3′ Abd-B non-coding regions respectively represent hsp70 promoter and 3′ untranslated sequences which include the polyadenylation signal. Arrows indicate transcriptional start sites. (C) Photomicrograph showing a lateral view of the tail region of a first instar larva (dorsal surface at top). The normal locations of two structures (fz, filzkörper; sph, spiracular hairs) scored in this study are indicated. (D) Schematic drawing of the tail region of a first instar larva adapted from Sato and Denell (1986) and Jürgens (1987). Thin dashed lines indicate segmental boundaries. Thick dashed lines indicate parasegment (PS) 13 boundaries. Rows of black squares denote ventral denticle belts.

Two genetic functions have been distinguished for Abd-B (Casanova et al., 1986): a morphogenetic (m) function is required in A5-A8 (PS10-13) and a regulatory (r) function is necessary in A9 (PS14). A genetic function for Abd-B r in A10 (PS15) has not been described. In Abd-B m r+ larvae, filzkörper, spiracular hairs and posterior spiracles, normally located in A8 (Fig. 1C,D), are absent (Karch et al., 1985; Casanova et al., 1986; Sato and Denell, 1986; Whittle et al., 1986). In Abd-B m+r larvae, the posterior spiracles are reduced in size, but the filzkörper and spiracular hairs are present (Karch et al., 1985; Casanova et al., 1986). These cuticular phenotypes suggest that Abd-B m function, but not Abd-B r function, is required for the formation of filzkörper and spiracular hairs in A8. Similarly, Abd-B m function, but not Abd-B r function, alters the denticle, or setal, belt in A8: the A8 denticle belt in Abd-B m r+ larvae is transformed to an anterior abdominal identity but is unaffected in Abd-B m+r larvae. In Abd-B m+r larvae, a narrow region of naked cuticle and a rudimentary denticle belt are formed posterior to the A8 denticle belt (Casanova et al., 1986). This phenotype requires Abd-B m function (Casanova et al., 1986). Additionally, Abd-B m expression is derepressed in PS14 of Abd-B m+r larvae (Boulet et al., 1991). Therefore, the primary function of Abd-B r appears to be the repression of Abd-B m in A9. However, Abd-B r function is not exclusively regulatory since weak morphogenetic activity can be detected when Abd-B r is derepressed in the absence of all other BX-C functions (Casanova et al., 1986). When both m and r functions are eliminated (Abd-B m r ), larvae resemble Abd-B m r+ larvae except that sclerotic plates develop in A9 (Karch et al., 1985; Casanova et al., 1986; Whittle et al., 1986).

The domains of Abd-B m and r genetic functions correlate with the expression patterns of two different ABD-B protein products. The ABD-B m (55×103Mr) and r (30×103Mr) proteins are translated from mRNAs that are transcribed from different promoters (Kuziora and McGinnis, 1988; Sánchez-Herrero and Crosby, 1988; Celniker et al., 1989; Zavortink and Sakonju, 1989; Celniker et al., 1990; Boulet et al., 1991). ABD-B m and r proteins share a carboxy-terminal domain that contains the ABD-B homeodomain, but only the ABD-B m protein has a glutamine-rich amino-terminal domain. An antibody specific for ABD-B m detects protein in PS10-13 (Boulet et al., 1991) while antibodies that recognize both ABD-B m and r proteins detect protein in PS10-15 (Celniker et al., 1989; DeLorenzi and Bienz, 1990; Boulet et al., 1991). Analyses of ABD-B protein distribution in Abd-B mutants show a correlation of the morphogenetic function with ABD-B m protein distribution and of the regulatory function with the ABD-B r protein distribution (Celniker et al., 1990; DeLorenzi and Bienz, 1990; Boulet et al., 1991; Sánchez-Herrero, 1991).

ABD-B m and r proteins participate in a network of interactions that regulates the expression of other homeotic genes. ABD-B r represses Abd-B m expression in PS14 and 15 while high levels of ABD-B m repress the expression of Ubx and abd-A in PS13 (Struhl and White, 1985; Casanova et al., 1986; Casanova and White, 1987; Karch et al., 1990; Macías et al., 1990; M. Lamka, unpublished observations). In PS7-12, ABD-A down-regulates Ubx expression in posterior compartments (Struhl and White, 1985; Karch et al., 1990; Macías et al., 1990). Therefore, although BX-C expression patterns overlap in PS7-13, cross-regulatory interactions establish distinct, reciprocal patterns of BX-C gene expression in each parasegment (Fig. 1A).

Although cross-regulatory interactions among many homeotic genes have been demonstrated (reviewed in Levine and Harding, 1989), their functional significance has recently been questioned (González-Reyes et al., 1990). Ectopic, embryonic expression of UBX protein, which disrupted normal, reciprocal patterns of BX-C expression, transformed larval cephalic and thoracic segments toward A1, a UBX-specified identity, but had no apparent phenotypic effect on larval abdominal segments (González-Reyes et al., 1990; González-Reyes and Morata, 1990; Mann and Hogness, 1990). Therefore, with respect to cuticular morphology, high levels of UBX seemed to be inconsequential in regions of the abdomen where cross-regulatory interactions keep UBX levels low. Since endogenous ABD-A and ABD-B apparently suppressed the phenotypic effects of ectopic UBX, this phenomenon was termed ‘phenotypic suppression’ (González-Reyes and Morata, 1990). As an explanation for phenotypic suppression, González-Reyes et al. (1990) proposed a hierarchy of homeotic protein functions, independent of transcriptional regulation, in which ABD-A and ABD-B protein activities override UBX protein activity.

In this paper, we have addressed two questions concerning ABD-B protein function in vivo: (1) do ABD-B m and r proteins have similar or different intrinsic activities? And (2) is ABD-B m protein functionally dominant to UBX protein? To analyze the activities of ABD-B m and r proteins, the effects of ectopic ABD-B m or r expression on larval segmental identities were examined. Our results indicate that ABD-B m and r proteins possess similar activities and suggest that the presence or absence of another factor(s) in PS14 and 15 affects ABD-B r activity. To test the homeotic protein hierarchical model, we ectopically expressed ABD-B m alone or in combination with UBX. Our results suggest that, contrary to a strict functional hierachy, ectopic UBX and ABD-B m proteins compete for the specification of segmental identities. Furthermore, we show, in contrast to previous reports (González-Reyes et al., 1990; González-Reyes and Morata, 1990; Mann and Hogness, 1990), that ectopic UBX is able to affect abdominal as well as head identities if expressed early in embryogenesis and suggest that abdominal and head segments are determined earlier in development than are thoracic segments.

Construction of hs-m and hs-r transformant lines

To construct the hs-m fusion gene (referred to as hsp70-M in Boulet et al., 1991), a 1.9 kb SspI-BglII fragment from the Class A cDNA clone B3 (Zavortink and Sakonju, 1989) was cloned into the pCaSpeR-hs P element vector (Thummel and Pirotta, 1992) cut at the BglII and HpaI sites. The polylinker in the pCaSpeR-hs vector lies between the hsp70 promoter and sequences from the 3′ end of the hsp70 gene which include the polyadenylation site. The hs-r fusion gene (referred to as hsp70-R in Boulet et al., 1991) was similarly constructed by cloning a 1.4 kb NruI-BglII fragment from the Class B cDNA clone E19 (Zavortink and Sakonju, 1989) into the pCaSpeR-hs P element vector cut at the BglII and HpaI sites. The hs-m and hs-r DNAs were coinjected with pπ25.7wc (Rubin and Spradling, 1982) into Df(1)w67c23y embryos. Six hs-m lines and one hs-r line were recovered. Four of the hs-m lines were homozygous viable. The hs-m(F1.2) and hs-m(M2) insertions were located on the first chromosome, the hs-m(M4) insertion on the second chromosome and the hs-m(F4) insertion on the third chromosome. The two homozygous lethal insertions hs-m(F1.1) and hs-m(M11), located on the second chromosome, were maintained as CyO-balanced stocks. The single hs-r line, designated hs-r(M2), carried a homozygous viable insertion on the third chromosome. Using a genomic source of transposase (Robertson et al., 1988), an additional hs-r line was obtained by mobilizing the original hs-r P element in the hs-r(M2) transformant. This homozygous viable hs-r(7A) line carried a new insertion on the second chromosome.

Drosophila strains and genetic crosses

We constructed flies of the following genotypes to analyze the effects of ectopic ABD-B m or r protein in BX-C embryos: w1118; hs-m(M4)/CyO; Dp(3;3)P5, Sb/Df(3R)P9 and w1118; hs-r(7A)/hs-r(7A); Dp(3;3)P5, Sb/Df(3R)P9. Df(3R)P9 is deficient for the entire BX-C (Lewis, 1978) whereas Dp(3;3)P5 contains a tandem duplication for the BX-C (Duncan and Lewis, 1982). hs-m, BX-C or hs-r, BX-C larvae were readily identified by their BX-C phenotypes (transformations of T3 and A1-A8 toward T2) and the development of filzkörper, which are normally absent from BX-C larvae.

The hs-Ubx-1a-22 line, a gift from Richard Mann and David Hogness, was used to express UBX ectopically (Mann and Hog-ness, 1990). This heterozygous line carries a third chromosome insertion of a Ubx-1a cDNA fused to the hsp70 promoter. To express UBX and ABD-B m proteins together throughout BX-C embryos, we constructed a hs-Ubx, DfP9 recombinant (maintained over Dp(3;3)P146 + McRS1-10, a multiply inverted third chromosome carrying a duplication of the BX-C, obtained from Ian Duncan) and examined heat-shocked progeny from the following crosses: w1118/Y; hs-m(M4)/hs-m(M4); Dp(3;3)P5, Sb/Df(3R)P9 × Dp(3;3)P146 + McRS1-10/Df(3R)P9, hs-Ubx (for 1 copy of hs-m with or without 1 copy of hs-Ubx) and +/Y; hs-m(M4)/+; Dp(3;3)P5, Sb/Df(3R)P9, hs-Ubx × w1118; hs-m(M4)/hs-m(M4); Dp(3;3)P5, Sb/Df(3R)P9 (for 1 or 2 copies of hs-m with or without 1 copy of hs-Ubx). BX-C larvae carrying 1 copy of hs-Ubx and at least 1 copy of hs-m were identified by the transformations of head and thoracic segments toward A1 (González-Reyes et al., 1990; González-Reyes and Morata, 1990; Mann and Hogness, 1990).

We generated chromosomes carrying two copies of hs-m by recombining second chromosome insertions: hs-m(M4) with hs-m(F1.1) or hs-m(M11). These flies were maintained as CyO-balanced stocks.

To analyze the phenotypes of larvae carrying 1-4 copies of hs-m with or without 1 copy of hs-Ubx, we heat shocked progeny from the following crosses: (1) w1118/Y; hs-Ubx/+ × w1118; hs-m(F4)/hs-m(F4) (for 1 copy of hs-m with or without 1 copy of hs-Ubx), (2) w1118/Y; hs-m(M4)/+; hs-Ubx/+ × w1118; hs-m(F4)/hs-m(F4) (for 1 or 2 copies of hs-m with or without 1 copy of hs-Ubx), (3) w1118/Y; hs-m(M4)/+; hs-Ubx/+ × w1118; hs-m(M11) hs-m(M4) or hs-m(F1.1) hs-m(M4)/CyO (for 1, 2 or 3 copies of hs-m with or without 1 copy of hs-Ubx) and (4) w1118/Y; hs-m(M11) hs-m(M4) or hs-m(F1.1) hs-m(M4)/+; hs-Ubx/+ × w1118; hs-m(M11) hs-m(M4) or hs-m(F1.1) hs-m(M4)/CyO (for 2 or 4 copies of hs-m with or without 1 copy of hs-Ubx). Progeny from these crosses were genotypically classified on the basis of phenotypic transformations. Larvae carrying hs-Ubx were identified by the consistent head and thoracic transformations toward an A1 identity. Larvae carrying hs-m were scored by the appearance of ventral pits in abdominal segments and/or non-involuted heads. Crosses yielded progeny with different copy numbers of hs-m, which we grouped together for analysis. In all cases, the observed frequencies of the phenotypic classes were in agreement with the predicted frequencies of the corresponding genotypic classes.

Heat-shock treatment of embryos

To express UBX and ABD-B proteins ectopically at specific times in development, embryos were staged by examination under Voltalef 3S oil. Stage 6 (2:50-3 hours) embryos (Campos- ortega and Hartenstein, 1985) were selected, transferred to coverslips and aged in a humidified 25°C incubator. Heat shocks were administered for the specified times described below by placing the coverslips on a glass plate in a humid chamber of a gravity convection incubator. Embryos were returned to the humidified 25°C incubator until larvae had hatched or embryos had aged for greater than 24 hours. The chorion and vitelline membranes were removed from unhatched larvae using a tungsten needle. Larvae were fixed in acetic acid:glycerin (4:1) for 15-30 minutes at 60°C and then left at room temperature for at least 24 hours. Larvae were mounted in CMCP-10 Mounting Media (Master’s Chemical Company):lactic acid (3:1) and placed on a 45°C slide warmer to clear overnight. Cuticles were analyzed using phase-contrast and darkfield microscopy.

For most of the experiments, embryos were heat shocked at 36°C for 30 minute periods at both 4 hours and 6 hours of embryogenesis. This method produced more severe ABD-B-induced phenotypic transformations than single 30 minute (36°C) heat shocks at either 4 or 6 hours or than two 30 minute (36°C) heat shocks at 5 and 7 hours. The double-pulse heat treatment at 4 and 6 hours did not significantly affect w1118 control embryos: 95% of w1118 embryos hatched as first instar larvae with cuticles indistinguishable from non-heat-shocked wild-type larvae. Under the same conditions, heat-shocked hs-m and hs-r embryos laid down larval cuticle but did not hatch.

To test for stage-related effects of ectopic UBX on the abdominal segments, we heat shocked embryos from w1118 control and w1118; hs-Ubx-1a-22/TM3 stocks at 4 and 6 hours, 6 and 8 hours, or 8 and 10 hours of embryogenesis. Each heat shock was 30 minutes at 36°C using the heat-shock protocol described above.

To determine the phenotypic consequences of ectopic expression of ABD-B m or ABD-B r protein, we created several transformant lines carrying Abd-B m or r cDNAs fused to the heat-inducible hsp70 promoter (Fig. 1B; Materials and methods). These lines were designated hs-m and hs-r. Anti-ABD-B antibodies detected the ABD-B m and ABD-B r protein forms on western blots of extracts from heat-shocked hs-m and hs-r embryos (Boulet et al., 1991). Antibody staining of embryos indicated that most cells expressed ABD-B m or r protein at a level similar to that of endogenous ABD-B in PS13 and 14 after heat shock of hs-m or hs-r embryos (data not shown). Antibody staining detected ectopic ABD-B protein in hs-m embryos at 1 hour and 2 hours, but not 3 hours, following a 30 minute heat shock at 36°C (data not shown). We induced ectopic ABD-B m or r protein in at least two independent hs-m or hs-r transformant lines with 30 minute, 36°C heat-shock pulses at 4 and 6 hours of embryogenesis, unless otherwise indicated. Under these heat-shock conditions, the cuticles of untransformed control larvae were indistinguishable from wild-type larval cuticles. Although many heat-shocked hs-m and hs-r embryos failed to undergo head involution and complete germ band retraction, in the subsequent text we only describe the phenotypic transformations that can be scored by the appearance or disappearance of larval cuticle markers.

Ectopic expression of ABD-B m protein induces A8 structures in thoracic and abdominal segments

Heat-shock induction of ABD-B m protein in embryos carrying one to four copies of hs-m produced major alterations in larval morphology. The most distinctive phenotype was the presence of filzkörper pairs, structures normally derived only from A8, in the thoracic and abdominal segments and occasionally as far anterior as the head (Figs 2B,C, 7). Spiracular hairs, which normally surround the openings of the posterior spiracles, were also found in thoracic and abdominal segments (Fig. 2C inset). Therefore, ABD-B m protein is necessary, as predicted by Abd-B mutant analyses and also sufficient to promote the development of filzkörper and spiracular hairs.

Fig. 2.

Dark-field photomicrographs showing the ventral cuticles of hs-m first instar larvae heat shocked at 4 and 6 hours of embryogenesis for 30 minute intervals. (A) Non-heat-shocked wild-type larva. Unique to T1 is the prothoracic beard (b). Pairs of Keilin’s organs (arrows) and ventral pits (asterisks) are present only in thoracic segments (T1-T3). Note the normal location of the filzk örper (fz). (B) Wild-type larva carrying one copy of hs-m. Ectopic filzk örper in thoracic and abdominal segments are slightly out of the plane of focus. Arrows point to the most anterior and posterior pairs of additional filzk örper in T2 and A7, respectively. The extra filzk örper are located on the dorsal side, opposite from the denticle belts, although compression of the cuticle distorted the spatial relationships in this photograph. (C) Wild-type larva carrying either two or four copies of hs-m. This larva shows a reduction in the size of the abdominal denticle belts. Note that the shape of the A1 belt resembles that of the T3 belt. Ectopic denticles (d) are seen in the head. A few of the ventral pits present in the abdominal segments are denoted by asterisks. Ectopic filzk örper and spiracular hairs are out of the ventral plane of focus. Only the most anterior and posterior extra filzk örper are marked (arrows). The inset shows an enlarged view of the region marked by the open arrow to illustrate the ectopic filzk örper (arrow) and spiracular hairs (arrowheads). (D) BX-C − larva carrying one or two copies of hs-m. The germ band failed to retract in this larva. Denticle belts resemble those of a BX-C− larva. Arrows point to the most anterior and posterior ectopic filzkörper. Arrowhead indicates the spiracular hairs in association with the filzk örper.

Fig. 2.

Dark-field photomicrographs showing the ventral cuticles of hs-m first instar larvae heat shocked at 4 and 6 hours of embryogenesis for 30 minute intervals. (A) Non-heat-shocked wild-type larva. Unique to T1 is the prothoracic beard (b). Pairs of Keilin’s organs (arrows) and ventral pits (asterisks) are present only in thoracic segments (T1-T3). Note the normal location of the filzk örper (fz). (B) Wild-type larva carrying one copy of hs-m. Ectopic filzk örper in thoracic and abdominal segments are slightly out of the plane of focus. Arrows point to the most anterior and posterior pairs of additional filzk örper in T2 and A7, respectively. The extra filzk örper are located on the dorsal side, opposite from the denticle belts, although compression of the cuticle distorted the spatial relationships in this photograph. (C) Wild-type larva carrying either two or four copies of hs-m. This larva shows a reduction in the size of the abdominal denticle belts. Note that the shape of the A1 belt resembles that of the T3 belt. Ectopic denticles (d) are seen in the head. A few of the ventral pits present in the abdominal segments are denoted by asterisks. Ectopic filzk örper and spiracular hairs are out of the ventral plane of focus. Only the most anterior and posterior extra filzk örper are marked (arrows). The inset shows an enlarged view of the region marked by the open arrow to illustrate the ectopic filzk örper (arrow) and spiracular hairs (arrowheads). (D) BX-C − larva carrying one or two copies of hs-m. The germ band failed to retract in this larva. Denticle belts resemble those of a BX-C− larva. Arrows point to the most anterior and posterior ectopic filzkörper. Arrowhead indicates the spiracular hairs in association with the filzk örper.

The number and thickness of the ectopic filzkörper depended on the level of ABD-B m protein. Generally, a single copy of hs-m induced fewer additional filzkörper than were induced by two or four copies of hs-m (Table 1). The ectopic filzkörper produced by a single copy of hs-m appeared thinner than those formed in the normal A8 position (Fig. 2B) and than those in larvae with two or four copies of hs-m (Fig. 2C).

Table 1.

Quantitation of filzkörper induced by increasing doses of hs-m with or without one copy of hs-Ubx

Quantitation of filzkörper induced by increasing doses of hs-m with or without one copy of hs-Ubx
Quantitation of filzkörper induced by increasing doses of hs-m with or without one copy of hs-Ubx

Several other phenotypes were produced by ectopic expression of ABD-B m. First, ventral pits, normally confined to the three thoracic segments (T1-T3), appeared in the abdomen (Fig. 2C). Because ventral pits develop in the abdominal segments of Ubx mutants, the formation of these structures is thought to be suppressed by UBX in wild-type larvae (Lewis, 1978; Mann and Hogness, 1990). Therefore, ABD-B m protein may interfere with either the expression or function of UBX protein in cells forming the ventral pits.

Second, Keilin’s organs, also normally in T1-T3, were absent from hs-m larvae. Finally, ectopic ABD-B m protein affected thoracic and abdominal denticle belts. The size of the T1 belt was reduced and the T1 beard, a small patch of denticles below the T1 belt, was absent (Fig. 2B,C). Since these alterations in T1 resembled the transformation of T1 toward T2 in Sex combs reduced (Scr) mutant larvae (Wakimoto et al., 1984; Sato et al., 1985; Pattatucci et al., 1991), the transformation of T1 in hs-m larvae may result from the repression of the endogenous Scr gene. Alternatively, endogenous Antennapedia (Antp) could be ectopically activated since T1 transforms toward T2 when ANTP is expressed throughout embryos (Gibson and Gehring, 1988). Individual thick denticles were observed in the thoracic belts and the head region (Fig. 2C). In larvae carrying two to four copies of hs-m, the abdominal denticle belts of A2-A8 were reduced in size, with varying numbers of missing denticles, suggesting a transformation toward A1 as in abd-A mutants (Fig. 2C; Sánchez-Herrero et al., 1985). The clearest effect of ABD-B m on the denticle belts was the transformation of the denticle belt in A1 toward a T3 denticle belt (Fig. 2C). This transformation of A1 is reminiscent of the Ubx mutant phenotype in which T3 and A1 are transformed toward T2 (Lewis, 1978). The similarity of abdominal transformations induced by ectopic ABD-B m to abdominal transformations in Ubx and abd-A mutants suggests that Ubx and/or abd-A mediate ABD-B m-induced effects on abdominal denticle belts. This interpretation is supported by the results of ectopic ABD-B m expression in the embryos deficient for the entire BX-C (see below).

Ectopic expression of ABD-B r protein also induces structures derived from segment A8

The primary function previously attributed to ABD-B r was the repression of Abd-B m expression in PS14 (Casanova et al., 1986). Surprisingly, heat induction of ABD-B r protein in homozygous hs-r embryos produced larval phenotypes very similar to those seen in hs-m larvae. Segmentally repeated filzkörper pairs and spiracular hairs were present in the thorax and abdomen, and ventral pits appeared occasionally in abdominal segments A1-A7 (Figs 3A,B, 7). Keilin’s organs were suppressed in hs-r larvae. Like ectopic ABD-B m, ectopic ABD-B r protein affected the denticle belts of the thoracic and abdominal segments (compare Figs 2C and 3A). Thoracic denticle belts maintained thoracic shapes, but T1 transformed toward T2, and some large denticles appeared in the thoracic belts and head region (Fig. 3A,B). The A1 belt weakly transformed toward a T3 denticle belt (Fig. 3A,B). The number of denticle rows in the A2-A8 belts of hs-r larvae were reduced, suggesting a transformation toward an A1 morphology (Fig. 3A). This effect is consistent with the repression of endogenous BX-C gene expression by ABD-B r in denticle precursor cells of the abdomen as described above. The similar phenotypes of hs-r and hs-m larvae indicate that ABD-B r protein is able to carry out the same activities as ABD-B m protein.

Fig. 3.

hs-r first instar larvae heat shocked for 30 minutes at 4 and 6 hours of development. Cuticles shown in A and B are from wild-type larvae carrying two copies of hs-r. (A) Ventral view showing the reduced size of the abdominal belts. Ectopic filzkörper (arrows) are out of the ventral plane of focus. Asterisks mark a few of the ventral pits present in the abdominal segments. (B) Lateral view showing the ectopic filzkörper. Arrows indicate six of the extra filzkörper. Arrowheads point to the ectopic spiracular hairs associated with the filzkörper. Ectopic denticles (d) are seen in the head. (C) Lateral view of a BX-C larvae carrying two copies of hs-r. Arrows mark ectopic filzkörper. Ectopic denticles (d) are seen in the head. The germ bands of the larvae shown in B and C failed to retract.

Fig. 3.

hs-r first instar larvae heat shocked for 30 minutes at 4 and 6 hours of development. Cuticles shown in A and B are from wild-type larvae carrying two copies of hs-r. (A) Ventral view showing the reduced size of the abdominal belts. Ectopic filzkörper (arrows) are out of the ventral plane of focus. Asterisks mark a few of the ventral pits present in the abdominal segments. (B) Lateral view showing the ectopic filzkörper. Arrows indicate six of the extra filzkörper. Arrowheads point to the ectopic spiracular hairs associated with the filzkörper. Ectopic denticles (d) are seen in the head. (C) Lateral view of a BX-C larvae carrying two copies of hs-r. Arrows mark ectopic filzkörper. Ectopic denticles (d) are seen in the head. The germ bands of the larvae shown in B and C failed to retract.

Induction of filzkörper and spiracular hairs by ectopic ABD-B m or r does not require endogenous BX-C genes

The ability of ectopically expressed ABD-B m and r proteins to induce filzkörper and spiracular hairs in segments as far anterior as the head, outside the domain specified by the BX-C genes, suggests that Ubx, abd-A and endogenous Abd-B genes are not required for the formation of these structures. To rule out the possibility that the phenotypic changes induced by ectopic ABD-B m or r are due to their effects on endogenous BX-C gene expression, hs-m or hs-r was induced in embryos deficient for the entire BX-C (BX-C ). In BX-C larvae, segments T3 and A1-A8 (PS5-13) all assume T2 (PS4)-like identities, showing the characteristic narrow denticle belts and ventral pits (Lewis, 1978). Expression of ABD-B m throughout BX-C embryos promoted the development of filzkörper and spiracular hairs in the larval thorax and abdomen (Figs 2D, 7), similar to ectopic structures induced in larvae of a wild-type genetic background (Figs 2B,C, 7). ABD-B r also induced the formation of filzkörper and spiracular hairs in BX-C larvae (Figs 3C, 7), but the transformations were much weaker than those induced by ectopic ABD-B r in a wild-type genetic background (Fig. 3A,B) or than those induced by ectopic ABD-B m in a BX-C genetic background (Fig. 2D). Like BX-C larvae, ventral pits were present in all thoracic and abdominal segments of hs-m or hs-r larvae of a BX-C genetic background. The denticle belts resembled those of BX-C larvae, but T1 transformed toward T2, and some thick denticles appeared in the denticle belts and the head region. These results indicate that ABD-B m or r protein is the only BX-C protein necessary to commit cells to filzkör-per and spiracular hair fates. Although ABD-B proteins affected the thickness of individual denticles, they did not transform the identities of denticle belts in T2, T3 or A1-A8 of BX-C larvae. These results support the conclusion that ectopic ABD-B proteins in wild-type embryos modify abdominal denticle belt shapes by affecting endogenous Ubx and/or abd-A expression or function.

ABD-B and UBX proteins compete for the specification of segmental identity when coexpressed throughout embryos

Ectopic expression of Ubx from a heat-inducible promoter transformed larval head and thoracic segments toward A1, but had no reported effect on larval abdominal segments (Figs 4A, 7; González-Reyes et al., 1990; González-Reyes and Morata, 1990; Mann and Hogness, 1990). A hierarchy of homeotic protein functions was proposed to account for the failure of ectopic UBX protein to transform the abdomen (González-Reyes et al., 1990). This model predicts that ABD-B function will override UBX function when both proteins are ectopically expressed. To test this hypothesis, UBX and ABD-B m proteins were simultaneously expressed throughout wild-type embryos. Wild-type larvae carrying single copies of hs-Ubx and hs-m (designated hs-Ubx + hs-m) exhibited the characteristic Ubx transformation of larval head and thoracic segments toward A1 identities (data not shown). Only 2% of hs-Ubx + hs-m larvae developed ectopic filzkörper compared to 19% of larvae carrying a single copy of hs-m without hs-Ubx (Table 1). Furthermore, ventral pits were not seen in the thorax or abdomen of these larvae, indicating that UBX was, in fact, active in these segments. Therefore, contrary to the expectations of the functional hierarchy model, the observed phenotypes suggest that ABD-B m does not prevent UBX from acting in the thorax and abdomen. Rather, UBX protein appears to interfere with the ability of ABD-B m to form ectopic A8 structures.

Fig. 4.

Cuticles from first instar larvae in which hs-Ubx or hs-Ubx + hs-m were ectopically induced by 30 minute heat shocks at 4 and 6 hours of embryonic development. (A) Lateral view of a hs-Ubx larva showing the transformation of the cephalic and thoracic segments toward A1 identities, as described by González-Reyes et al. (1990) and Mann and Hogness (1990). (B) Lateral view of a wild-type larva carrying one copy of hs-Ubx and two or four copies of hs-m. Note the transformation of the head and thoracic segments toward A1 identities and the additional filzkörper. Arrows indicate the most anterior and posterior ectopic filzkörper. Filzkörper are not always limited to segments A5-A7 as shown in this figure, but can be seen as far anterior as the thoracic segments. (C) Lateral view of a BX-C larva carrying one copy of hs-Ubx and one copy of hs-m. Note that all segments are transformed toward A1 and filzkörper are absent. The same phenotype was observed when an additional copy of hs-m was present.

Fig. 4.

Cuticles from first instar larvae in which hs-Ubx or hs-Ubx + hs-m were ectopically induced by 30 minute heat shocks at 4 and 6 hours of embryonic development. (A) Lateral view of a hs-Ubx larva showing the transformation of the cephalic and thoracic segments toward A1 identities, as described by González-Reyes et al. (1990) and Mann and Hogness (1990). (B) Lateral view of a wild-type larva carrying one copy of hs-Ubx and two or four copies of hs-m. Note the transformation of the head and thoracic segments toward A1 identities and the additional filzkörper. Arrows indicate the most anterior and posterior ectopic filzkörper. Filzkörper are not always limited to segments A5-A7 as shown in this figure, but can be seen as far anterior as the thoracic segments. (C) Lateral view of a BX-C larva carrying one copy of hs-Ubx and one copy of hs-m. Note that all segments are transformed toward A1 and filzkörper are absent. The same phenotype was observed when an additional copy of hs-m was present.

To test if the interference of UBX can be overcome by increasing the relative level of ABD-B m protein, we heat shocked groups of embryos, each with increasing copies of hs-m but only a single copy of hs-Ubx. When a single copy of hs-m was present, only 2% of larvae showed ectopic filzkörper (Table 1). A second copy of hs-m increased the percentage of larvae with ectopic filzkörper to 19% (Table 1). In the group of larvae with one to three copies of hs-m, the percentage of transformed larvae was even greater (82%) and, finally, 91% of larvae with two or four copies of hs-m showed ectopic filzkörper (Table 1; Figs 4B, 7). Since additional copies of hs-m overcome the interference of UBX, our results indicate that UBX and ABD-B m compete for their respective developmental programs in a dose-dependent manner.

To eliminate possible contributions of the endogenous BX-C genes to the phenotypes observed in larvae carrying a single copy of hs-Ubx and one to four copies of hs-m, UBX and ABD-B m proteins were coexpressed throughout embryos deficient for the BX-C. In homozygous BX-C larvae carrying one copy of hs-Ubx and either one or two copies of hs-m, the three cephalic, three thoracic and nine abdominal segments showed a transformation toward A1 (Figs 4C, 7). No filzkörper were seen in these larvae, even at the normal A8 position. Therefore, in the absence of endogenous BX-C genes, the identities of all segments were specified by ectopic UBX. The presence of extra filzkörper in wild-type larvae with a single copy of hs-Ubx and one or two copies of hs-m, but not in BX-C larvae with a single copy of hs-Ubx and one or two copies of hs-m, suggests that endogenous Abd-B contributed to the formation of filzkörper. In A5-A8 (PS10-13) of wild-type embryos, endogenous ABD-B m protein could supplement ectopic ABD-B m protein so that ABD-B m levels are sufficiently high to overcome suppression by ectopic UBX. Additionally, ectopic ABD-B m protein could activate endogenous Abd-B expression as was observed when a DEFORMED (DFD)/ABD-B chimeric protein was ectopically expressed (Kuziora and McGinnis, 1991). Taken together, these results argue that specification of segmental identity by ectopic UBX and ABD-B m is determined by a quantitative competition between UBX and ABD-B m proteins for the regulation of downstream genes, but do not support the presence of a hierarchy in which ABD-B activity dominates.

Ectopic expression of UBX protein at early embryonic stages transforms identities of abdominal segments

The ability of ectopic UBX to compete with ectopic ABD-B m appears to contradict the inability of UBX to transform abdominal segments of hs-Ubx embryos. This paradox could be explained if abdominal segment identities are determined earlier in development than thoracic segment identities. At the time of heat shock in previous experiments (González-Reyes et al., 1990; González-Reyes and Morata, 1990; Mann and Hogness, 1990), proper identities of the abdominal segments, but not of the thoracic segments, may have already been specified. To test this hypothesis, UBX protein was induced in accurately staged embryos by two 30 minute heat-shock pulses, with a 90 minute recovery, beginning at either 4, 6 or 8 hours of embryonic development.

Heat induction of hs-Ubx beginning at 4 hours or 6 hours of development weakly transformed abdominal segments toward A1 (Fig. 5B). We quantified the degree of abdominal transformation toward A1 using the number of setae in the first row of the denticle belts in A2-A8. This row derives from the posterior compartments of A1-A7 (Szabad et al., 1979; Jürgens, 1987), the region corresponding to high levels of ABD-A in wild-type embryos (Karch et al., 1990; Macías et al., 1990). This row of setae is absent from the A1 belt where ABD-A is not expressed (Fig. 5C). In abd-A mutants, the first denticle row is missing from the belts in A2-A8 (Sánchez-Herrero et al., 1985), further substantiating its use as a marker for the segmental transformations of A2-A8 toward A1. As a control, the setae in the second row of the denticle belts in A2-A8 were counted since this row derives from the anterior compartments of A2-A8 (Szabad et al., 1979; Jürgens, 1987), the region which corresponds to high levels of UBX in wild-type embryos (Fig. 5C; White and Wilcox, 1985). When embryos were heat shocked beginning at 4 hours of development, the number of setae in the first row of the denticle belts in A2-A8 was reduced by as much as 70% in hs-Ubx larvae relative to control larvae (Fig. 6A,E). Induction of hs-Ubx starting at 6 hours of development reduced the number of denticles in the first row of A2-A8 belts by 10% to 35% (Fig. 6B,F). However, the number of setae in the first row of the A2-A8 denticle belts was similar in hs-Ubx and control embryos when heat shocks were initiated at 8 hours of development (Fig. 6C,G).

Fig. 5.

(A-B) Bright-field photomicrographs showing the anterior halves of cuticles from w1118 control and hs-Ubx larvae heat shocked for 30 minutes at 4 and 6 hours of embryogenesis. (A) Cuticle from a w1118 control larva. Arrowheads point to the first row of setae in A2-A4 denticle belts. (B) Cuticle from a hs-Ubx larva. Note that the first row of setae are absent from denticle belts of A2 and A3 (arrows), and a reduced number of setae are seen in the first row of the A4 denticle belt (arrowhead). (C) Schematic diagram illustrating the segmental compartments. Note that a high level of UBX (stippled shading) is expressed in the anterior compartments (a) of segments A1-A7, corresponding to parasegments (PS) 6 through 12, while a high level of ABD-A (hatched shading) is expressed in the posterior compartments (p) of segments A1-A7, corresponding to PS7-12. Arrowheads represent denticles. Arrowhead direction indicates the polarity of denticles. The denticle row number is adjacent to each denticle row. Solid thick lines represent parasegmental boundaries. Dashed thick lines denote segmental boundaries.

Fig. 5.

(A-B) Bright-field photomicrographs showing the anterior halves of cuticles from w1118 control and hs-Ubx larvae heat shocked for 30 minutes at 4 and 6 hours of embryogenesis. (A) Cuticle from a w1118 control larva. Arrowheads point to the first row of setae in A2-A4 denticle belts. (B) Cuticle from a hs-Ubx larva. Note that the first row of setae are absent from denticle belts of A2 and A3 (arrows), and a reduced number of setae are seen in the first row of the A4 denticle belt (arrowhead). (C) Schematic diagram illustrating the segmental compartments. Note that a high level of UBX (stippled shading) is expressed in the anterior compartments (a) of segments A1-A7, corresponding to parasegments (PS) 6 through 12, while a high level of ABD-A (hatched shading) is expressed in the posterior compartments (p) of segments A1-A7, corresponding to PS7-12. Arrowheads represent denticles. Arrowhead direction indicates the polarity of denticles. The denticle row number is adjacent to each denticle row. Solid thick lines represent parasegmental boundaries. Dashed thick lines denote segmental boundaries.

Fig. 6.

(A-C) Quantitation of setae in the first and second rows of the denticle belts in the second though eighth abdominal segments (A2-A8) of heat-shocked w118 control and hs-Ubx larvae. Thirty minute heat shocks were given at (A) 4 and 6 hours of embryogenesis, (B) 6 and 8 hours of embryogenesis or (Q 8 and 10 hours of embryogenesis. The number of denticle belts (n) examined is represented below the bars. The plotted values represent the average number of setae from n denticle belts. Error bars indicate the standard deviation. Denticle belts in which setal rows 1 or 2 were difficult to identify were not included in the analysis. (D-G) Schematic summary of the phenotypic transformations induced by ectopic UBX: 4 and 6 hours of embryogenesis (E), 6 and 8 hours of embryogenesis (E) or 8 and 10 hours of embryogenesis (G). Different abdominal designations were used in E and F to indicate that the reduction in the number of row 1 setae in A2-A8 denticle belts, indicative of a weak transformation to A1, is more extreme when hs-Ubx is induced at 4 and 6 hours (as shown in A) than at 6 and 8 hours (as shown in B).

Fig. 6.

(A-C) Quantitation of setae in the first and second rows of the denticle belts in the second though eighth abdominal segments (A2-A8) of heat-shocked w118 control and hs-Ubx larvae. Thirty minute heat shocks were given at (A) 4 and 6 hours of embryogenesis, (B) 6 and 8 hours of embryogenesis or (Q 8 and 10 hours of embryogenesis. The number of denticle belts (n) examined is represented below the bars. The plotted values represent the average number of setae from n denticle belts. Error bars indicate the standard deviation. Denticle belts in which setal rows 1 or 2 were difficult to identify were not included in the analysis. (D-G) Schematic summary of the phenotypic transformations induced by ectopic UBX: 4 and 6 hours of embryogenesis (E), 6 and 8 hours of embryogenesis (E) or 8 and 10 hours of embryogenesis (G). Different abdominal designations were used in E and F to indicate that the reduction in the number of row 1 setae in A2-A8 denticle belts, indicative of a weak transformation to A1, is more extreme when hs-Ubx is induced at 4 and 6 hours (as shown in A) than at 6 and 8 hours (as shown in B).

The cephalic segments also showed a stage-dependent susceptibility to transformation by ectopic UBX as previously described (González-Reyes and Morata, 1991). When hs-Ubx embryos were heat shocked at 4 hours of embryogenesis, three cephalic segments (C1, C2, C3) developed A1-type denticle belts (Figs 5B, 6E). However, heat shocks begun at 6 hours of development transformed only two cephalic segments, C1 and C2, toward A1 (data not shown; Fig. 6F) and heat shocks initiated at 8 hours of development did not transform the cephalic segments except for the development of a few denticles in the head region (Fig. 6G). In contrast, heat induction of hs-Ubx beginning at 4, 6 or 8 hours transformed the thoracic denticle belts toward A1 denticle belts (Fig. 6E-G). These results suggest that the identities of the head segments as well as the abdominal segments are determined earlier in development than those of the thoracic segments. We suggest that the previous failure to transform the abdominal segments after heat shock of hs-Ubx embryos (González-Reyes et al., 1990; GonzálezReyes and Morata, 1990; Mann and Hogness, 1990) was due to differences in the timing or the level of ectopic UBX expression.

Both ABD-B m and r proteins possess morphogenetic functions

Genetic and molecular analyses of Abd-B have shown that it is a complex gene. Four overlapping Abd-B transcripts encode two homeodomain proteins with independent functions (Casanova et al., 1986; DeLorenzi et al., 1988; Kuziora and McGinnis, 1988; Sanchez-Herrero and Crosby, 1988; Celniker et al., 1989; Zavortink and Sakonju, 1989; Boulet et al., 1991). The m function, provided by a 55×103Mr protein, specifies the proper identities of PS10-13 and the r function, contributed by a 30×103Mr protein, is required for PS14 identity (Celniker et al., 1990; Boulet et al., 1991). The ABD-B r protein corresponds to a truncated ABD-B m protein which lacks the additional 25×103Mr amino-terminal domain (Celniker et al., 1989; DeLorenzi et al., 1988; Zavortink and Sakonju, 1989). By expressing ABD-B m or r protein ectopically in embryos, we have shown that the two proteins possess similar abilities to specify segmental identities. Ectopic ABD-B m and r proteins can both induce the formation of segmentally repeated pairs of filzkörper and associated spiracular hairs, structures thought to derive from PS13 (Jürgens, 1987), in the thorax and abdomen of larvae. A summary of these phenotypes is shown in Fig. 7.

Fig. 7.

Schematic summary of larval phenotypes resulting from ectopic ABD-B and/or UBX proteins in wild-type or BX-C genetic backgrounds. Abbreviations and symbols: fz, filzkörper; ps, posterior spiracles; sph, spiracular hairs; ⊥, ectopic filzkörper and spiracular hairs; ▬ , abdominal denticles; ▬, thoracic denticles; ◓ , Keilin’s organ; ⋆, ventral pit.

Fig. 7.

Schematic summary of larval phenotypes resulting from ectopic ABD-B and/or UBX proteins in wild-type or BX-C genetic backgrounds. Abbreviations and symbols: fz, filzkörper; ps, posterior spiracles; sph, spiracular hairs; ⊥, ectopic filzkörper and spiracular hairs; ▬ , abdominal denticles; ▬, thoracic denticles; ◓ , Keilin’s organ; ⋆, ventral pit.

Although induction of filzkörper and spiracular hairs by ectopic ABD-B m protein is consistent with the absence of these structures from Abd-B m r+ larvae (Karch et al., 1985; Casanova et al., 1986; Sato and Denell, 1986; Whittle et al., 1986), induction of PS13-derived structures by ectopic ABD-B r was unexpected. ABD-B r is normally expressed only in PS14 and 15, and the mutant phenotype of Abd-B m+r larvae suggests that r function is not required for specifying filzkörper and spiracular hairs (Casanova et al., 1986). It is possible, however, that filzkörper derive from cells of both PS13 and 14 and that ABD-B r function also contributes to their specification: Abd-B m is derepressed in PS14 of Abd-B m+r mutants (Casanova et al., 1986; Boulet et al., 1991) and may substitute for any r function required for the normal development of filzkörper.

The similarity of phenotypes induced by ectopic ABD-B m and ABD-B r proteins implies that the ability to promote the development of filzkörper and spiracular hairs can be primarily attributed to the domain common to both proteins. An important feature of the shared domain is the homeodomain. Homeodomain swap experiments suggest that most of the specificity of homeotic proteins lies in the homeodomain (Kuziora and McGinnis, 1989, 1991; Gibson et al., 1990; Mann and Hogness, 1990). For example, a DFD/ABD-B chimeric protein, in which the ABD-B homeodomain was substituted for DFD homeodomain, induced filzkörper and other PS13-derived structures in larval thoracic and abdominal segments, and also activated the ectopic expression of endogenous Ubx, abd-A and Abd-B (Kuziora and McGinnis, 1991). These experiments, however, did not distinguish whether the development of PS13-derived structures resulted from direct activation of down-stream genes or from ectopic activation of endogenous Abd-B and/or other BX-C genes which in turn activated their target genes. Our results show that ABD-B proteins are sufficient to induce filzkörper and spiracular hairs since these PS13-derived structures developed when ectopic ABD-B m or r was expressed in BX-C larvae. Because ectopic ABD-B m and ABD-B r induced similar structures, most, if not all, of the specificity for inducing filzkörper and spiracular hairs lies in the domain shared by both proteins. Therefore, ABD-B m and r proteins may regulate common downstream genes. The glutamine-rich amino-terminal domain of ABD-B m, not present in ABD-B r, may contribute to the ability of ectopic ABD-B m protein to induce stronger transformations toward PS13 than ectopic ABD-B r. Amino-terminal sequences of the ANTP protein have also been implicated in stronger transformations of the larval head and prothorax by ectopic ANTP (Gibson et al., 1990).

Although ABD-B r protein can direct the development of filzkörper and spiracular hairs, these structures do not appear in PS14 and 15. Our results suggest that parasegmental differences contribute to the phenotype specified by ABD-B r in its normal domain. A factor(s) found in PS14 and 15 may affect the activity of ABD-B r protein. A candidate for such a factor is the spalt (sal) gene product. The sal expression domain (Frei et al., 1988; Wagner-Bernholz et al., 1991) overlaps the Abd-B r expression domain (Kuziora and McGinnis, 1988; Sánchez-Herrero and Crosby, 1988; DiLorenzi et al., 1988; Boulet et al., 1991) and sal mutants exhibit ectopic pairs of filzkörper and spiracular hairs in PS14 and 15 (Jürgens, 1988). Since derepression of ABD-B m in PS14 of Abd-B m+r larvae does not result in extra filzkörper and spiracular hairs (Casanova et al., 1986), the SAL protein must also be capable of suppressing Abd-B m function. In fact, ectopic pairs of filzkör-per and spiracular hairs develop when ABD-B m is derepressed in PS14 and 15 of sal mutants (Jürgens, 1987). Therefore, SAL may modify ABD-B r activity (and can modify ABD-B m activity) in PS14 and 15. Alternatively, SAL activity may affect the fate of cells in PS14 and 15 prior to determination by ABD-B r. For example, SAL may act as a general repressor by altering the accessiblity of binding sites in downstream target genes to ABD-B r protein (Casanova, 1989).

A functional hierarchy does not explain the failure of ectopic UBX to transform the abdomen

In previous studies, ectopic expression of UBX transformed cephalic and thoracic segments but did not seem to affect segments A2-A9 (González-Reyes et al., 1990; GonzálezReyes and Morata, 1990; Mann and Hogness, 1990). The phenotypic effects of ectopic UBX activity were apparently suppressed in the abdomen by endogenous ABD-A and ABD-B functions (González-Reyes et al., 1990). When UBX was ectopically expressed in embryos lacking all BX-C functions except abd-A (Ubx abd-A+Abd-B ), the identities of A2-A4 were not transformed (González-Reyes and Morata, 1990). Similarly, in Ubx abd-A Abd-B m r+ embryos, ABD-B r function prevented ectopic UBX from inducing the development of an extra A1-type denticle belt in A9 (González-Reyes and Morata, 1990). Finally, ABD-B m was able to suppress the phenotypic effects of ectopic UBX in A8 of larvae deficient for Ubx and abd-A (Mann and Hogness, 1990).

In the hierarchical model proposed by González-Reyes et al. (1990), ABD-A and ABD-B activities override ectopic UBX activity. According to the model, a functionally dominant ABD-B m protein should suppress UBX-induced transformations toward A1. Our data do not support the existence of a hierarchy of homeotic protein functions (González-Reyes et al., 1990). Instead, concurrent expression of UBX and ABD-B m proteins throughout wild-type embryos resulted in UBX-induced transformations of cephalic and thoracic segments toward A1. Moreover, UBX suppressed the development of ABD-B m-induced filzkörper in the thorax and the abdomen. The relative levels of ectopic UBX and ectopic ABD-B m determine the resultant phenotype: an increase in the dose of hs-m from one to four copies decreases the ability of UBX to suppress filzkörper.

The sensitivity of ABD-B m-induced transformations to the relative levels of UBX and ABD-B m proteins suggests that UBX and ABD-B m compete for the regulation of common targets. Such a competition among homeotic proteins could explain the previously observed phenomenon of phenotypic suppression (González-Reyes and Morata, 1990). The failure of these experiments to show competition among homeotic proteins may have been due to the relative levels or stabilities of ectopic proteins. UBX and ABD-B m could compete for common binding sites. In vitro binding studies have indicated that homeodomains recognize similar consensus sequences (reviewed in Hayashi and Scott, 1990; Laughon, 1991). In vivo, common binding sites are used by different homeodomain proteins to activate or repress the Antp promoter 2 (Appel and Sakonju, 1992). Alternatively, UBX and ABD-B m could compete for an accessory factor required for the expression of downstream genes. Another possibility is that UBX and ABD-B m direct segments into mutually exclusive developmental pathways through the regulation of unique downstream genes.

Cephalic and abdominal segments may be determined earlier than thoracic segments

Previous reports concluded that ectopic UBX had no effect on abdominal cuticular identities (González-Reyes et al., 1990; González-Reyes and Morata, 1990; Mann and Hogness, 1990). However, in our experiments, high levels of ectopic UBX induced at 4 hours or 6 hours of development weakly transformed the denticle belts of A2-A8 toward A1 denticle belts. Ectopic UBX also transformed the cephalic and thoracic segments as previously described (GonzálezReyes et al., 1990; González-Reyes and Morata, 1990, 1991; Mann and Hogness, 1990). By 8 hours of development, the cephalic and abdominal segments appeared to be resistant to transformation by ectopic UBX, but the thoracic segments were still transformed toward A1.

The temporal difference in the determination of segmental identities may be a function of the sizes of homeotic gene transcription units (Gubb, 1986; Kornfeld et al., 1989; Thummel, 1992). The transcription units of Dfd, Scr, abd-A and Abd-B, homeotic genes responsible for the identities of head and abdominal segments, are shorter than the transcription units of Antp and Ubx which are required for proper thoracic patterning. Consequently, DFD and SCR may determine the identities of the cephalic segments, and ABD-A and ABD-B the identities of the abdominal segments, before ANTP and UBX determine the identities of the thoracic segments. If one postulates that the activation of the three BX-C genes occurs at the same time during development, transcription of the shorter abd-A (23 kb) and Abd-B m (8 kb) genes will be completed prior to the transcription of the 77 kb Ubx gene (Kornfeld et al., 1989; Zavortink and Sakonju, 1989; M. Lamka and S. Sakonju, unpublished observations). Assuming RNA polymerase transcribes at a rate of 1.4 kb/minute (Thummel et al., 1990; Shermoen and O’Farrell, 1991), it would take 16 minutes and 6 minutes to transcribe abd-A and Abd-B m, respectively, while 55 minutes would be required for Ubx transcription. Thus, the difference in transcription times alone might allow ABD-A and ABD-B m proteins to appear 40-50 minutes earlier in development than UBX protein. Similarly, transcription units of the Dfd (11 kb) and Scr (20 kb) genes are shorter than Antp (P1, 105 kb; P2, 40 kb) and Ubx (Kaufman et al., 1990). Therefore, Dfd and Scr would require shorter transcription times than Antp and Ubx. Although a number of studies report that UBX, ABD-A and ABD-B m proteins are detectable with antibodies at approximately 4 hours of embryogenesis (White and Lehmann, 1986; Celniker et al., 1989; DeLorenzi and Bienz, 1990; Karch et al., 1990; Macías et al., 1990; Boulet et al., 1991; Irvine et al., 1991), the relative sensitivities of these antibodies are unknown and staging by different groups may be variable. It is also difficult to assess the appearance of DFD and SCR relative to ANTP and UBX. In wild-type embryos, ABD-A and ABD-B m proteins may be present before UBX protein, and repression of Ubx by ABD-A and ABD-B m in posterior parasegments would ensure that UBX does not compete for segmental identities. We suggest, therefore, that transcriptional cross-regulatory interactions among the homeotic genes are important for normal development.

We are grateful to Richard Mann and David Hogness for the hs-Ubx flies, to Ian Duncan and Ed Lewis for other fly strains used in this study and to Joe Dickinson for the cuticle preparation protocol. We also thank Bruce Appel, Ken Burtis, Dan Cimbora, Barry Condron, Jill Hendrickson and Anthea Letsou for valuable comments on the manuscript and members of the Sakonju lab for discussion during the progress of this project. This work was supported by Howard Hughes Medical Institute. M. L. L. was supported, in part, by an NIH Genetics Predoctoral Training Grant to the University of Utah.

Appel
,
B.
and
Sakonju
,
S.
(
1992
).
Cell-type specific mechanisms of transcriptional repression by the homeotic gene products UBX and ABD-A in Drosophila embryos
.
EMBO J
.
(in press)
.
Beachy
,
P. A.
,
Helfand
,
S. L.
and
Hogness
,
D. S.
(
1985
).
Segmental distribution of bithorax complex proteins during Drosophila development
.
Nature
313
,
545
551
.
Boulet
,
A. M.
,
Lloyd
,
A.
and
Sakonju
,
S.
(
1991
).
Molecular definition of the morphogenetic and regulatory functions and the cis-regulatory elements of the Drosophila Abd-B homeotic gene
.
Development
111
,
393
405
.
Campos-Ortega
,
J. A.
and
Hartenstein
,
V.
(
1985
).
The Embryonic Development of Drosophila melanogaster
.
Berlin
:
Springer-Verlag
.
Casanova
,
J.
(
1989
).
Mutations in the spalt gene of Drosophila cause ectopic expression of Ultrabithorax and Sex combs reduced
.
Roux’s Arch. Dev. Biol
.
198
,
137
140
.
Casanova
,
J.
,
Sánchez-Herrero
,
E.
and
Morata
,
G.
(
1986
).
Identification and characterization of a parasegment specific regulatory element of the Abdominal-B gene of Drosophila
.
Cell
47
,
627
636
.
Casanova
,
J.
and
White
,
R. A. H.
(
1987
).
Trans-regulatory functions in the Abdominal-B gene of the bithorax complex
.
Development
101
,
117
122
.
Celniker
,
S. E.
,
Keelan
,
D. J.
and
Lewis
,
E. B.
(
1989
).
The molecular genetics of the bithorax complex of Drosophila: characterization of the products of the Abdominal-B domain
.
Genes Dev
.
3
,
1424
1436
.
Celniker
,
S. E.
,
Sharma
,
S.
,
Keelan
,
D. J.
and
Lewis
,
E. B.
(
1990
).
The molecular genetics of the bithorax complex of Drosophila: cis-regulation in the Abdominal-B domain
.
EMBO J
.
9
,
4277
4286
.
DeLorenzi
,
M.
,
Ali
,
N.
,
Saari
,
G.
,
Henry
,
C.
,
Wilcox
,
M.
and
Bienz
,
M.
(
1988
).
Evidence that the Abdominal-B r element function is conferred by a trans-regulatory homeoprotein
.
EMBO J
.
7
,
3223
3231
.
DeLorenzi
,
M.
and
Bienz
,
M.
(
1990
).
Expression of Abdominal-B homeoproteins in Drosophila embryos
.
Development
108
,
323
329
.
Duncan
,
I.
(
1987
).
The bithorax complex
.
Ann. Rev. Genet
.
21
,
285
319
.
Duncan
,
I.
and
Lewis
,
E. B.
(
1982
).
Genetic control of body segment differentiation in Drosophila
.
In Developmental Order: Its Origin and Regulation
. pp.
533
554
.
New York
:
Alan R. Liss, Inc
.
Frei
,
E.
,
Schuh
,
R.
,
Baumgartner
,
S.
,
Burri
,
M.
,
Noll
,
M.
,
Jürgens
,
G.
,
Seifert
,
E.
,
Nauber
,
U.
and
Jäckle
,
H.
(
1988
).
Molecular characterization of spalt, a homeotic gene required for head and tail development in the Drosophila embryo
.
EMBO J
.
7
,
197
204
.
Gibson
,
G.
and
Gehring
,
W. J.
(
1988
).
Head and thoracic transformation caused by ectopic expression of Antennapedia during Drosophila development
.
Development
102
,
657
675
.
Gibson
,
G.
,
Schier
,
A.
,
LeMotte
,
P.
and
Gehring
,
W. J.
(
1990
).
The specificities of Sex combs reduced and Antennapedia are defined by a distinct portion of each protein that includes the homeodomain
.
Cell
62
,
1087
1103
.
González-Reyes
,
A.
and
Morata
,
G.
(
1990
).
The developmental effect of overexpressing a Ubx product in Drosophila embryos is dependent on its interactions with other homeotic products
.
Cell
61
,
515
522
.
González-Reyes
,
A.
and
Morata
,
G.
(
1991
).
Organization of the Drosophila head as revealed by the ectopic expression of the Ultrabithorax product
.
Development
113
,
1459
1471
.
González-Reyes
,
A.
,
Urquía
,
N.
,
Gehring
,
W. J.
,
Struhl
,
G.
and
Morata
,
G.
(
1990
).
Are cross-regulatory interactions between homeotic genes functionally significant?
Nature
344
,
78
80
.
Gubb
,
D.
(
1986
).
Intron-delay and the precision of expression of homeotic gene products in Drosophila
.
Dev. Gene
.
7
,
119
131
.
Hayashi
,
S.
and
Scott
,
M. P.
(
1990
).
What determines the specificity of action of Drosophila homeodomain proteins?
Cell
63
,
883
894
.
Irvine
,
K. D.
,
Helfand
,
S. L.
and
Hogness
,
D. S.
(
1991
).
The large upstream control region of the Drosophila homeotic gene Ultrabithorax
.
Development
111
,
407
424
,
Jürgens
,
G.
(
1987
).
Segmental organization of the tail region in the embryo of Drosophila melanogaster. A blastoderm fate map of the cuticle structures of the larval tail region
.
Roux’s Arch. Dev. Biol
.
196
,
141
157
.
Jürgens
,
G.
(
1988
).
Head and tail development of the Drosophila embryo involves spalt, a novel homeotic gene
.
EMBO J
.
7
,
189
196
.
Karch
,
F.
,
Bender
,
W.
and
Weiffenbach
,
B.
(
1990
).
abd-A expression in Drosophila embryos
.
Genes Dev
.
4
,
1573
1587
.
Karch
,
F.
,
Weiffenbach
,
B.
,
Peifer
,
M.
,
Bender
,
W.
,
Duncan
,
I.
,
Celniker
,
S.
,
Crosby
,
M.
and
Lewis
,
E. B.
(
1985
).
The abdominal region of the bithorax complex
.
Cell
43
,
81
96
.
Kaufman
,
T. C.
,
Seeger
,
M. A.
and
Olsen
,
G.
(
1990
).
Molecular and genetic organization of the Antennapedia gene complex of Drosophila melanogaster
.
Adv. Genet
.
27
,
309
362
.
Kornfeld
,
K.
,
Saint
,
R. B.
,
Beachy
,
P. A.
,
Harte
,
P. J.
,
Peattie
,
D. A.
and
Hogness
,
D. S.
(
1989
).
Structure and expression of a family of Ultrabithorax mRNAs generated by alternative splicing and polyadenylation in Drosophila
.
Genes Dev
.
3
,
243
258
.
Kuziora
,
M. A.
and
McGinnis
,
W.
(
1988
).
Different transcripts of the Drosophila Abd-B gene correlate with distinct genetic sub-functions
.
EMBO J
.
7
,
3233
3244
.
Kuziora
,
M. A.
and
McGinnis
,
W.
(
1989
).
A homeodomain substitution changes the regulatory specificity of the Deformed protein in Drosophila embryos
.
Cell
59
,
563
571
.
Kuziora
,
M. A.
and
McGinnis
,
W.
(
1991
).
Altering the regulatory targets of the Deformed protein in Drosophila embryos by substituting the Abdominal-B homeodomain
.
Mech. Dev
.
33
,
83
94
.
Laughon
,
A.
(
1991
).
DNA binding specificity of homeodomains
.
Biochemistry
30
,
11357
11367
.
Levine
,
M. S.
and
Harding
,
K. W.
(
1989
).
Drosophila: the zygotic contribution
.
In Genes and Embryos
(eds.
D. M.
Glover
and
B. D.
Hames
), pp.
39
94
.
Oxford
:
IRL Press
.
Lewis
,
E. B.
(
1978
).
A gene complex controlling segmentation in Drosophila
.
Nature
276
,
565
570
.
Macías
,
A.
,
Casanova
,
J.
and
Morata
,
G.
(
1990
).
Expression and regulation of the abd-A gene of Drosophila
.
Development
110
,
1197
1207
.
Mann
,
R. S.
and
Hogness
,
D. S.
(
1990
).
Functional dissection of Ultrabithorax proteins in D. melanogaster
.
Cell
60
,
597
610
.
Morata
,
G.
,
Macías
,
A.
,
Uriquía
,
N.
and
González-Reyes
,
A.
(
1990
).
Homeotic genes
.
Sem. Cell Biol
.
1
,
219
227
.
Pattatucci
,
A. M.
,
Otteson
,
D. C.
and
Kaufman
,
T. C.
(
1991
).
A functional and structural analysis of the Sex combs reduced (Scr) locus of Drosophila melanogaster
.
Genetics
129
,
423
441
.
Robertson
,
H. M.
,
Preston
,
C. R.
,
Phillis
,
R. W.
,
Johnson-Schlitz
,
D. M.
,
Benz
,
W. K.
and
Engels
,
W. R.
(
1988
).
A stable genomic source of P-element transposase in Drosophila melanogaster
.
Genetics
118
,
461
470
.
Rubin
,
G. M.
and
Spradling
,
A. C.
(
1982
).
Genetic transformation of Drosophila with transposable element vectors
.
Science
218
,
348
353
.
Sánchez-Herrero
,
E.
(
1991
).
Control of the expression of the bithorax complex genes abdominal-A and Abdominal-B by cis-regulatory regions in Drosophila embryos
.
Development
111
,
437
449
.
Sánchez-Herrero
,
E.
and
Crosby
,
M. A.
(
1988
).
The Abdominal-B gene of Drosophila melanogaster : overlapping transcripts exhibit two different spatial distributions
.
EMBO J
.
7
,
2163
2173
.
Sánchez-Herrero
,
E.
,
Vernós
,
I.
,
Marco
,
R.
and
Morata
,
G.
(
1985
).
Genetic organization of Drosophila bithorax complex
.
Nature
313
,
108
113
.
Sato
,
T.
and
Denell
,
R. E.
(
1986
).
Segmental identity of caudal cuticular features of Drosophila melanogaster larvae and its control by the bithorax complex
.
Dev. Biol
.
116
,
78
91
.
Sato
,
T.
,
Hayes
,
P. H.
and
Denell
,
R. E.
(
1985
).
Homeosis in Drosophila: roles and spatial patterns of expression of the Antennapedia and Sex combs reduced loci in embryogenesis
.
Dev. Biol
.
111
,
171
192
.
Shermoen
,
A. W.
and
O’Farrell
,
P. H.
(
1991
).
Progression of the cell cycle through mitosis leads to abortion of nascent transcripts
.
Cell
67
,
303
310
.
Struhl
,
G.
and
White
,
R. A. H.
(
1985
).
Regulation of the Ultrabithorax gene of Drosophila by other bithorax complex genes
.
Cell
43
,
507
519
.
Szabad
,
J.
,
Schüpbach
,
T.
and
Wieschaus
,
E.
(
1979
).
Cell lineage and development in the larval epidermis of Drosophila melanogaster
.
Dev. Biol
.
73
,
256
271
.
Thummel
,
C. S.
(
1992
).
Mechanisms of transcriptional timing in Drosophila
.
Science
255
,
39
40
.
Thummel
,
C. S.
,
Burtis
,
K. C.
and
Hogness
,
D. S.
(
1990
).
Spatial and temporal patterns of E74 transcription during Drosophila development
.
Cell
61
,
101
111
.
Thummel
,
C. S.
and
Pirotta
,
V.
(
1992
).
New pCaSpeR P element vectors
.
Dros. Inf. Serv
.
71
,
150
.
Wagner-Bernholz
,
J. T.
,
Wilson
,
C.
,
Gibson
,
G.
,
Schuh
,
R.
and
Gehring
,
W. J.
(
1991
).
Identification of target genes of the homeotic gene Antennapedia by enhancer detection
.
Genes Dev
.
5
,
2467
2480
.
Wakimoto
,
B. T.
,
Turner
,
F. R.
and
Kaufman
,
T. C.
(
1984
).
Defects in embryogenesis in mutants associated with the Antennapedia gene complex of Drosophila melanogaster
.
Dev. Biol
.
102
,
147
172
.
White
,
R. A. H.
and
Lehmann
,
R.
(
1986
).
A gap gene, hunchback, regulates the spatial expression of Ultrabithorax
.
Cell
47
,
311
321
.
White
,
R. A. H.
and
Wilcox
,
M.
(
1985
).
Distribution of Ultrabithorax proteins in Drosophila
.
EMBO J
.
4
,
2035
2043
.
Whittle
,
J. R. S.
,
Tiong
,
S. Y. K.
and
Sunkel
,
C. E.
(
1986
).
The effect of lethal mutations and deletions within the bithorax complex upon the identity of caudal metameres in the Drosophila embryo
.
J. Embryol. Exp. Morph
.
93
,
153
166
.
Zavortink
,
M.
and
Sakonju
,
S.
(
1989
).
The morphogenetic and regulatory functions of the Drosophila Abdominal-B gene are encoded in overlapping RNAs transcribed from separate promoters
.
Genes Dev
.
3
,
1969
1981
.