Using sequence homology to Drosophila homeobox-containing genes, we have cloned a homologue of abdominal-A from the locust Schistocerca gregaria. The Schistocerca clone encodes a stretch of 78 amino acids including the homeodomain and its flanking regions identical to the corresponding region of abdominal-A.

We have shown by in situ hybridization that this gene is transcribed and have used an antibody raised against its protein product to examine the expression of abdominal-A during early Schistocerca embryogenesis. Schistocerca is a short germ insect. Although the segmented body plan is very similar to that of Drosophila, the segments are generated sequentially by a process of growth, not simultaneously by subdivision of a syncytial blastoderm. In both organisms, abdominal-A is expressed throughout the abdomen from a sharp anterior boundary located within the first abdominal segment (Al). The initial activation of the genes in the two species differs. Schistocerca initiates expression in a small group of cells in the anterior of A2, shortly after this segment is defined by the appearance of engrailed protein. This contrasts with the appearance of abdominal-A expression in Drosophila, which appears simultaneously throughout the entire abdomen.

Insects are characterized by a segmented body subdivided into a head, a set of mouthparts, a thorax, an abdomen and a terminal region (Snodgrass, 1935; Anderson, 1973). This body plan is obscured in the postembryonic stages of some insect groups, but it is clearly displayed in the embryonic germ band of all insects. Such a conserved body plan suggests an underlying conservation in the genes that define the metameric pattern, and in those that impose regional identity on the different segments.

The array of segments in the germ band can be formed in one of three general ways (Sander, 1976; Krause, 1939): (i) by subdivision of an epithelial sheet that occupies the whole of the egg (long germ band insects); (ii) by growth from a primordium that occupies a very small portion of the egg (short germ band insects), or (iii) by some combination of these processes, whereby the anterior segments are generated by subdivision, but the posterior segments are generated by growth (intermediate germ band insects). These different modes by which the germ band may be generated suggest variation in the mechanisms that establish the patterns of gene activity in the germ band (Sander, 1983, 1988). By comparing aspects of gene expression during segmentation of the locust Schistocerca with the equivalent stages of development in Drosophila, we hope to learn something of the evolutionary changes that distinguish pattern generation in short and long germ insects (Sander, 1988; Tear et al. 1988).

In Drosophila localized maternal cues co-ordinate the partitioning of the blastoderm into metameric units through the concerted action of gap and pair-rule gene products. These establish the initial metameric activity of some of the segment polarity genes which then interact to maintain and elaborate the repeating segment pattern (Ingham, 1988; Martinez-Arias, 1989). At the same time the gap and pair-rule genes direct the localized activation of the homeotic genes, thereby establishing differences that will endow segments with unique identities. This patterning process occurs almost simultaneously throughout the entire body axis, after the rapid cycles of nuclear cleavage that generate the syncytial blastoderm, but before the completion of cellularization and the onset of gastrulation (Howard, 1988; Akam, 1987).

This situation contrasts with that of the short germ band insects, e.g. Schistocerca gregaria, which generate segments after cellularization, during a phase of polarized growth. In these insects, the embryonic primordium is restricted to a small group of cells that form as a disc near the posterior tip of the egg. Gastrulation begins during the disc stage and the gastrulation furrow extends into the posterior regions of the embryo as it elongates. The first visible signs of segmentation appear in the thorax, three days after fertilization. During the next day, the gnathal segments and then the abdominal segments appear, the latter forming one at a time in an anterior-to-posterior progression (Roonwal, 1936; Krause, 1938).

Cell ablation studies during the elongating disc stage have demonstrated that, in the early embryo, different regions of the adult are represented in relative proportions that differ markedly to those existing in the mature germ band (Sander, 1976; Krause, 1953). In the mature germ band, the abdomen occupies a much larger region than the thorax, whereas the relative sizes of these primordia appear to be reversed at the earlier blastoderm stage. Understanding this transition requires molecular markers for the processes of generation and specification of metameres in wild-type and experimental embryos.

In Drosophila, the posterior of every segment is defined by the continuous expression of the gene engrailed (Kornberg et al. 1985; Dinardo et al. 1985). A homologue of the Drosophila engrailed gene has been identified in the locust (Patel et al. 1989a), where its expression is also restricted to the posterior part of every segment. The expression of engrailed in the emerging abdominal segments follows an anteroposterior sequence which precedes visible segmentation (Patel et al. 1989b). This suggests that at least one of the genes that define the segment pattern in Drosophila has a very similar role in the locust. It also confirms the conclusion from earlier experiments of UV and X-ray ablations (Sander, 1976), ligations and heat shocks (Mee and French, 1986; Mee, 1986) that segment patterning takes place sequentially, and only shortly before the first overt signs of segmentation are visible.

In Drosophila, the activity of the homeotic genes in restricted spatial domains defines the identity of different regions of the body (Lewis, 1978; Akam, 1987). Thus these genes provide molecular markers to determine when, in relation to segmentation, regional identity is specified. In this paper, we report the isolation of the locust homologue for one of the Drosophila homeotic genes, abdominal-A. Genetic and molecular analysis of this gene in Drosophila suggest that it is largely responsible for the identity of the abdominal segments (Lewis, 1978; Sánchez-Herrero et al. 1985; Tiong et al. 1985; Karch et al. 1990). Using an antibody raised against the coding region of the locust gene, we show that its product is expressed in the embryo in a wide domain comprising the whole of the abdominal region, comparable to that defined by the Drosophila abdominal-A gene. The generation of this pattern follows an anteroposterior sequence, compatible with the known features of locust embryogenesis but which highlights the different developmental strategies evolved by long and short germ insects.

Animals

Schistocerca gregaria (Forskål) eggs were laid in moist sand and collected over defined periods at 26°C. Those required for experiments were removed from the sand and cultured on moist filter paper in a Petri dish at 26°C.

Construction and screening of library

The Schistocerca gregaria genomic library was constructed by size fractionating a Sau3A partial digest of adult testis DNA on a sucrose gradient and ligating the 9–23 kb fraction into the BamHl site of λEMBL3. The library contained 2×106 recombinant plaque forming units (pfu). The Drosophila probes used were: engrailed, a 400bp BamHI-Aval fragment of pS799-7 (Fjöse et al. 1985); even-skipped, a 580 bp Hindlll-Accl fragment of p48-X1.4 (Macdonald et al. 1986); Antennapedia, a 400bp BamHI-EcoRl fragment of pID203a (gift of I. Dawson). These were labelled with α-32P-dATP by nick-translation to a specific activity of 108–109 disintsmin-1μg−1. All hybridizations were carried out at reduced stringency in 5xSSPE, 0.5% SDS, 5×Denhardts, 43% formamide and 100 μgmU1 sheared single-stranded salmon testis DNA at 37°C followed by washing at 50°C in 2×SSPE, 0.1% SDS. (McGinnis et al. 1984).

Sequencing

Regions of interest were sequenced by the dideoxy chain termination method (Sanger et al. 1977) from double-stranded templates with SequenaseTM enzyme (USB Corporation, Ohio) using forward and reverse primers. Sequence analysis was performed using ANALYSED (Staden, 1984), FASTN (Lipman and Pearson, 1985) and the University of Wisconsin Genetics Computing Group package (Devereux et al. 1984).

Generation of antibody

A 790 bp ApaI-PstI fragment from the Schistocerca abdominal-A clone λG610 (Fig. 1) was subcloned into Bluescript (Stratagene, La Jolla. California) to generate pAP3. Using the KpnIand SacI sites from the polylinker the fragment was cloned into Bluescribe (Stratagene, La Jolla, California). BamHI digestion of this subclone released an 810 bp fragment containing the entire open reading frame of the genomic exon plus 14 nucleotides from the polylinker at the 5′ end. This fragment was subcloned into pGEX2T (Amrad Corporation, Melbourne, Australia) to yield the plasmid, pFUSl.

Fig. 1.

Restriction map of the abdominal-A genomic clone from Schistocerca, λG610, showing position of homeobox homology and the region used as a probe for in situ hybridization.

Fig. 1.

Restriction map of the abdominal-A genomic clone from Schistocerca, λG610, showing position of homeobox homology and the region used as a probe for in situ hybridization.

A second fusion was constructed by subcloning the 640 bp BglII-PstI fragment from λG610 into the BamHI-PstI sites of pUR292 (gift from R.Weinzierl, Department of Anatomy, Cambridge) to yield the resulting plasmid, pF4.

The region of fusion of the coding regions was sequenced to check that the open reading frames were continuous and in the correct frame.

E. coll strain JM101 carrying pFUSl or pF4 were grown to late log phase and then induced with 1 mM IPTG for 2–3 h. The cultures were harvested and resuspended in 2×SDS gel loading buffer (100mM Tris-HCl, pH 6.8, 4% SDS, 20% glycerol, 10% β-mercaptoethanol, 0.1 % bromophenol blue). Samples were then loaded onto SDS-polyacrylamide slab gels as described by Hames (1981) for analysis.

Protein bands were visualized on 10% preparative SDS-polyacrylamide gels by incubation in 0.25 M KC1 at 0°C for 20 min. The band containing the fusion protein was excised and the protein eluted in SDS running buffer (0.1% SDS, 25 mM Tris, 192 mM glycine) using a Biotrap elution chamber (Schleicher and Schuell); electrophoresis was at 200 V for 2–4 h. The eluted protein was dialyzed against PBS (130 mM NaCl, 7mM Na2HPO4, 3mM NaH2PO4) overnight, snap frozen in liquid nitrogen and stored at −70°C.

Rabbits were immunized subcutaneously at multiple sites with an initial inoculum of 200 μg of fusion protein expressed from the pFUSl construct, in complete Freund’s adjuvant, followed by boosts of 100–250 μg in incomplete Freund’s adjuvant every two to four weeks. Activity against the ABDOMINAL-A (ABD-A) protein was first observed after ten weeks. All immunizations were carried out commercially by ABC Ltd, Cambridge.

Large quantities of the fusion protein F4 were purified for use on affinity-purification columns using APTG affinity. chromatography exactly as described by Carroll and Laughon (1987).

Purification of antisera

To remove any antibodies to bacterial proteins or glutathione transferase, the serum was pretreated with an acetone powder (Harlow and Lane, 1988) prepared from E. coli JM101 overproducing glutathione transferase from pGEX2T for 1 h at room temperature. The supernatant was passed over an Affigel 10 (Biorad) column bound with 200–400 μg of purified fusion protein. Bound antibodies were eluted with 0.1M glycine-HCl, pH2.5 and neutralized with 0.1 volume 2 M Tris-HCl, pH 7.4. Bovine serum albumin (BSA) was added to 1 % and the serum was dialysed against PBS. Affinity-purified serum was stored at 4 °C, in the presence of 0.2 % sodium azide.

Immunochemistry

Schistocerca gregaria embryos were dissected from their egg cases under locust embryo saline (8.76gl-1 NaCl, 0.22gl-1 KC1, 0.29gI-1 CaCl2, 0.25gl-1 MgSO4, 1.15g l-1 TES). They were then fixed in 3.7% formaldehyde in PBS at 0°C for 40–60min. The embryos were washed in PTX (PBS+0.3% Triton X-100) and then blocked in PBTX (PTX+0.1% BSA and 10 mM NaN3) at 4°C for 2–4h. Affinity-purified primary antibody was diluted 1:500 in PBTX and incubated with the embryos at 4 °C overnight. The embryos were then washed in PBTX for 3h with 4–6 changes. The embryos were subsequently incubated with secondary antibody, biotinylated goat anti-rabbit IgG (Vector Laboratories, Peterborough, England), diluted 1:400 in PBTX for 90min at 4°C then washed in PBS+0.1% Tween 20 (PT) for 1 h with 4–6 changes at room temperature. The embryos were then incubated with avidin-biotin complex (ABC Elite, Vector Laboratories) for 30 min at room temperature and then washed through six changes of PT in 30 min and stained in the following solution: 50 μl 6% nickel ammonium sulphate, 6.25 μl 10mgml-1 diaminobenzidine (Sigma), 1 μl commercial hydrogen peroxide solution (6%; 20 vols available oxygen, Boots, Nottingham) and PT to 500 μl.

For double antibody stainings, the embryos were first stained with Mab 4D9 (Patel et al. 1989a) under the conditions described above using biotinylated horse anti-mouse IgG (Vector Laboratories) as secondary antibody. After visualization of the first antibody, the embryos were washed extensively with PBTX overnight and then treated exactly as above except that nickel ammonium sulphate enhancement (Adams, 1981) was not used in the ensuing staining reaction.

Drosophila embryos were dechorionated with commercial bleach then washed with water and fixed in heptane saturated with 0.1M Pipes, pH6.95, ImM MgSO4, 2mM EGTA, 4% paraformaldehyde for 40min. The aqueous phase was removed and replaced with methanol. This mixture was shaken vigorously to remove the embryonic vitelline membrane. The devitellinized embryos were washed with methanol, then PBTX and then treated as the Schistocerca embryos except that the primary antibody (anti-Drosophila abdominal-A, supplied by J. Casonova) was used at a dilution of 1:5000 and biotinylated rabbit anti-rat (Vector Laboratories) was used as the secondary antibody.

All embryos were dehydrated, cleared in xylene and mounted in DPX-Permount (BDH Ltd. Poole, England).

In situ hybridization to embryo sections

In situ hybridization was performed essentially as described for Drosophda (Akam and Martinez-Arias, 1985; Sánchez-Herrero and Crosby, 1988). A 1.4 kb EcoRl-Xbal fragment from λG610 (Fig. 1) was random primed (Feinberg and Vogelstein, 1983) in the presence of 35S-dATP and used as probe.

Cloning and identification of an abdominal-A homologue

We constructed a genomic library from Schistocerca gregaria containing 2×106 recombinant phage particles with an average insert size of 16 kb. Assuming a haploid genome size of 60×l08 bp (Wilmore and Brown, 1975) this represents 4 genome equivalents. A total of 5×l05 phage were screened at reduced stringency with Drosophila homeobox probes derived from the genes engrailed, even-skipped and Antennapedia (see Materials and methods). Plaques that hybridized with more than one probe were selected for further characterization. One of these phages, λG610, hybridized with the three probes and was analyzed further. The region of homology, located within a 790 bp fragment of the clone (Fig. 1), was subcloned and sequenced on both strands.

This clone contained a complete homeobox sequence, 82% identical to the homeobox from abdominal-A of Drosophila (Fig. 2). The conceptual translation of this region generates a peptide that is identical to the Drosophila ABD-A protein throughout the homeodomain and for six aminoacids 5′ and twelve aminoacids 3′ to the homeodomain (Karch et al. 1990). This extensive conserved region serves to identify the Schistocerca fragment as a specific homologue of the Drosophila abdominal-A gene.

Fig. 2.

Comparison of the Schistocerca gregaria genomic sequence from λG610 with cDNA sequence of Drosophila abdominal-A from Karch et al. (1990). Positions of nucleic acid identity are marked by a vertical bar and amino acid identity by a shaded box. The homeobox is underlined and the positions of introns in the Drosophila sequence are indicated bv arrows.

Fig. 2.

Comparison of the Schistocerca gregaria genomic sequence from λG610 with cDNA sequence of Drosophila abdominal-A from Karch et al. (1990). Positions of nucleic acid identity are marked by a vertical bar and amino acid identity by a shaded box. The homeobox is underlined and the positions of introns in the Drosophila sequence are indicated bv arrows.

5′ to the homeobox, similarity between the Schistocerca and Drosophila sequences stops abruptly at a position where the Drosophila abdominal-A gene has an intron. At this position, the Schistocerca DNA has a strong match to a consensus splice acceptor sequence (Py4TTGCAG/G, H. D. Lipshitz, personal communication, Breathnach and Chambon, 1981). It too is likely to contain an intron at this position. 27 bases 3′ to the homeobox, the Drosophila abdominal-A gene has a 70bp intron (Karch et al. 1990; G. Tear, unpublished data). This is not present in the Schistocerca clone. The genomic fragment includes sequences homologous to the Drosophila cDNA sequence on either side of the intron (Fig. 2). This 3′ intron is also absent in an abdominal-A homologue from Manduca (L. Nagy and L. Riddiford personal communication) suggesting that Drosophila has gained an intron after its separation from these insects.

Further 3′ to the homeobox the similarity between Drosophila abdominal-A and λG610 is reduced. The locust open reading frame continues to a stop codon ninety-two amino acids beyond the end of the homeobox. The only similarities in this region are a long polyglutamine repeat, and a short stretch of amino acids located near the carboxy termini of both proteins.

The genomic fragment containing this abdominal-A sequence was used to probe Schistocerca DNA on Southern blots. At high stringency, the clone hybridizes to only a single fragment in the Schistocerca genome (data not shown). To test whether this genomic fragment is transcribed, we used it as a probe for homologous transcripts in sections of locust embryos. Using embryos that had completed formation of the germ band (45% of development), we found that the probe hybridized to transcripts, but only within the abdominal region of the locust embryos, in a domain between abdominal segments 1 and 10 (A1–A10) (Fig. 3). The abdominal-A gene in Drosophila is expressed in a similar region (Rowe, 1987; E. Sánchez-Herrero, unpublished data) suggesting that the locust fragment derives from a functional homologue of the Drosophila abdominal-A gene.

Fig. 3.

In situ hybridization using the 1.4 kb Abol-EcoRl fragment indicated in Fig. 1 to a section of a Schistocerca embryo at approximately 45% development. Anterior is to the left, the asterisk indicates the third thoracic segment (T3) and an arrowhead marks the anterior extent of expression. Scale bar: 250 μm.

Fig. 3.

In situ hybridization using the 1.4 kb Abol-EcoRl fragment indicated in Fig. 1 to a section of a Schistocerca embryo at approximately 45% development. Anterior is to the left, the asterisk indicates the third thoracic segment (T3) and an arrowhead marks the anterior extent of expression. Scale bar: 250 μm.

Production of antibodies against Schistocerca ABD-A protein

To examine in more detail the expression of the Schistocerca abdominal-A gene, we raised antibodies against a fragment of the Schistocerca protein. Two fusion proteins were generated. In the first (FUS1), 158 aminoacids of Schistocerca ABD-A were fused to the Schistosoma japonicum glutathione transferase (Smith and Johnson, 1988). In the second (F4), 106 aminoacids of Schistocerca ABD-A were fused to E. coli β- galactosidase (Carroll and Laughon, 1987). An antiserum was raised against FUS1 and affinity purified against F4 to ensure that the purified serum contained antibodies specific for the Schistocerca portion of the fusions.

The affinity-purified serum preferentially recognizes one new band that is not seen using preimmune serum on protein extracts from Schistocerca embryos immobilized on a nitrocellulose membrane. The size of this protein, 42–45×103Mr, is similar to that seen for the Drosophila abdominal-A product on Westerns (R.Weinzierl personal communication).

In both species, abdominal-A is expressed within the abdominal segments of the embryo. In Schistocerca, the expression of abdominal-A extends in the epidermis from the posterior of the first abdominal segment to the tenth abdominal segment (Fig. 4E) whereas in Drosophila, abdominal-A is expressed between the posterior of Al and the seventh abdominal segment (Fig. 4F,G,H). The fact that our serum recognizes an epitope in locust embryos that is expressed solely in the abdomen and not in more anterior regions suggests that our antiserum is specific for the Schistocerca homologue of ABD-A.

Fig. 4.

Appearance of ABD-A protein during early development in Schistocerca and Drosophila embryos stained with anti-ABD-A antibodies. All Schistocerca embryos are shown at the same magnification, anterior is up in (A,B,C,D,E and H) and to the left in (F and G), the third thoracic segment is indicated by an asterisk. Photographs of abdominal-A staining in Schistocerca at (A) 22%, (B) 23%, (C) 26%, (D) 29%, (E) 31% of development and in Drosophila at (F) early stage 9, (G) late stage 9 and (H) after germ band shortening. Enlargements of parts of embryos A and B are shown in Fig. 6. Embryonic staging is as described in Bentley et al. (1979) or Campos-Ortega & Hartenstein, (1985). Expression of abdominal-A is first seen in Schistocerca in a group of cells, indicated by an arrow, at about 22% of development (A), whereas in Drosophila the first expression occurs in cells in each of the abdominal segments (F). In Schistocerca expression extends from the initially expressing cells anteriorly and posteriorly (B) and forms a defined anterior boundary (C). Expression concentrates at the lateral edges of the forming segments (D) and eventually extends to the tenth abdominal segment (E). After the initial expression of abdominal-A in stripes in Drosophila, expression is activated throughout the expression domain from parasegment 7 to the anterior of parasegment 13, marked in F and G, which is maintained after germ band shortening (H). Scale bar: 500 μm (A,B,C,D and E) and 204 μm (F,G and H).

Fig. 4.

Appearance of ABD-A protein during early development in Schistocerca and Drosophila embryos stained with anti-ABD-A antibodies. All Schistocerca embryos are shown at the same magnification, anterior is up in (A,B,C,D,E and H) and to the left in (F and G), the third thoracic segment is indicated by an asterisk. Photographs of abdominal-A staining in Schistocerca at (A) 22%, (B) 23%, (C) 26%, (D) 29%, (E) 31% of development and in Drosophila at (F) early stage 9, (G) late stage 9 and (H) after germ band shortening. Enlargements of parts of embryos A and B are shown in Fig. 6. Embryonic staging is as described in Bentley et al. (1979) or Campos-Ortega & Hartenstein, (1985). Expression of abdominal-A is first seen in Schistocerca in a group of cells, indicated by an arrow, at about 22% of development (A), whereas in Drosophila the first expression occurs in cells in each of the abdominal segments (F). In Schistocerca expression extends from the initially expressing cells anteriorly and posteriorly (B) and forms a defined anterior boundary (C). Expression concentrates at the lateral edges of the forming segments (D) and eventually extends to the tenth abdominal segment (E). After the initial expression of abdominal-A in stripes in Drosophila, expression is activated throughout the expression domain from parasegment 7 to the anterior of parasegment 13, marked in F and G, which is maintained after germ band shortening (H). Scale bar: 500 μm (A,B,C,D and E) and 204 μm (F,G and H).

Expression of abdominal-A in Schistocerca embryos

In Schistocerca embryos, abdominal-A expression is first observed at 20% of development, when the thoracic and gnathal segments are already visible as bulges in the germ band (Fig. 4A). In embryos at this stage, there are no morphological signs of segmentation in the abdomen. Initially, only a few cells express abdominal-A. These cells are located laterally, a short distance posterior to the metathoracic segment. Often they appear asymmetrically, a small patch of cells staining on one side, but only a few cells on the other (Fig. 4A, 6A). Other similar embryos, presumably slightly older, show larger laterally paired elongated patches of staining with cells near the centre of each patch showing strongest expression (Fig. 4B, 6B). The position of these strongly expressing cells relative to T3, corresponds approximately to that of the first expressing cells visible in earlier embryos. Thus it seems that expression is initiated in a small group of cells and then extends to both anterior and posterior neighbours.

To confirm the early expression in Schistocerca, and to locate the site of abdominal-A expression with respect to the forming segments, we have doublestained embryos with our antibody and with Mab4D9, which recognises the locust Engrailed protein (Patel et al. 1989a). Fig. 5 shows the pattern of expression of abdominal-A and engrailed in the Schistocerca embryo at 22 % of development. Abdominal-A is first expressed in cells that lie just anterior to, and sometimes overlap, the engrailed stripe that defines posterior A2 (A2p). At this stage, the A2p stripe of engrailed is well defined, and the A3p stripe is just appearing (Fig. 5A). Slightly later the expression of abdominal-A extends anteriorly to the engrailed stripe of Al and posteriorly beyond the A2 engrailed stripe (Fig. 5B,C). This pattern is very different from that of abdominal-A in Drosophila embryos where ABD-A protein is first observed in the posterior cells of each abdominal segment during stage 9 (Campos-Ortega and Hartenstein, 1985) from pAl to pA7; shortly afterwards, it is expressed throughout this domain with a strong intrasegmental modulation (Fig. 4G).

Fig. 5.

Expression of abdominal-A and engrailed in Schistocerca embryos at approximately 22% of development. Cells expressing abdominal-A are stained brown and cells expressing engrailed are stained dark violet. (A) Initial expression of abdominal-A in the anterior of A2, which then extends both anteriorly and posteriorly (B). The final anterior boundary of abdominal-A expression (C) corresponds with the anterior boundary of engrailed in Al. Scale bar: 50 μm.

Fig. 5.

Expression of abdominal-A and engrailed in Schistocerca embryos at approximately 22% of development. Cells expressing abdominal-A are stained brown and cells expressing engrailed are stained dark violet. (A) Initial expression of abdominal-A in the anterior of A2, which then extends both anteriorly and posteriorly (B). The final anterior boundary of abdominal-A expression (C) corresponds with the anterior boundary of engrailed in Al. Scale bar: 50 μm.

Initially the anterior margin of abdominal-A expression in Schistocerca embryos is uneven. When expression has reached the posterior of Al, but before any visible signs of abdominal segmentation, the anterior boundary straightens (Fig. 6). We do not know whether this involves cell movement or changes in expression. A similar sharpening of the anterior border of the engrailed stripe also occurs and abdominal-A expression may be following the same cues. Once it is clearly defined, the anterior limit of abdominal-A expression coincides with the anterior margin of the engrailed stripe of Al. This boundary corresponds to the anterior limit of parasegment 7 (Martinez-Arias and Lawrence, 1985), and is therefore identical to the anterior limit of abdominal-A expression in Drosophila. Occasionally, ectopic cells expressing abdominal-A can be seen anterior to a well-formed boundary (Fig. 6E), but such isolated cells are not seen in later stages, after abdominal segmentation is visible. Similar ectopically expressing cells can also be seen with Mab 4D9 where expression of engrailed anterior to the anterior limit of the engrailed stripe is observed (Patel et al. 1989b).

Fig. 6.

Detail of early stages of embryogenesis showing anterior extension of abdominal-A staining and maturation of the posterior Al boundary. Photograph A shows the same embryo as in Fig. 4A, revealing the bilateral asymmetry in the activation of expression of abdominal-A, with fewer cells staining for expression on the right side. In B, expression extends anterior and posterior to the initial cells. C shows the same embryo as Fig. 4B. In D, the anterior boundary has sharpened. Note the first appearance of staining in a line of cells near the ventral midline of the embryo. Photograph E shows an ectopic abdominal-A expressing cell within Al, which is indicated by the arrowhead. Scale bar: 100 μm (A,B,C & D) and 80 μm (E).

Fig. 6.

Detail of early stages of embryogenesis showing anterior extension of abdominal-A staining and maturation of the posterior Al boundary. Photograph A shows the same embryo as in Fig. 4A, revealing the bilateral asymmetry in the activation of expression of abdominal-A, with fewer cells staining for expression on the right side. In B, expression extends anterior and posterior to the initial cells. C shows the same embryo as Fig. 4B. In D, the anterior boundary has sharpened. Note the first appearance of staining in a line of cells near the ventral midline of the embryo. Photograph E shows an ectopic abdominal-A expressing cell within Al, which is indicated by the arrowhead. Scale bar: 100 μm (A,B,C & D) and 80 μm (E).

At the time that ABD-A protein is first expressed, the unsegmented posterior region of the embryo is growing rapidly. As it grows, detectable levels of ABDA protein appear in more posteriorly located cells. The distance between these cells and the posterior tip of the growing germ band remains approximately constant, and the appearance of ABD-A protein follows about one segment behind the appearance of engrailed stripes. Thus, in cells throughout the abdomen, there is a slight delay between the first expression of engrailed and of abdominal-A.

When the germ band is just completed, (30% of development) expression of abdominal-A extends as far as the posterior of the tenth abdominal segment (Fig. 7A). This posterior limit appears to be segmental, not parasegmental, as it coincides with the groove between A10 and All. For a short while, the level of protein in A10 is comparable with that in the more anterior segments, but soon after the germ band is complete, it begins to fall. Segments Alp to A9 retain high levels of ABD-A protein, but in A10 ABD-A protein falls to barely detectable levels by 40–42 % of development (Fig. 7B).

Fig. 7.

Photographs of Schistocerca embryos stained for abdominal-A expression at (A) 32% development and (B) 42% development. Expression extends to the posterior of A10 in A but only as far as the posterior of A9 in B. P, proctodeum. Scale bar-200 μm.

Fig. 7.

Photographs of Schistocerca embryos stained for abdominal-A expression at (A) 32% development and (B) 42% development. Expression extends to the posterior of A10 in A but only as far as the posterior of A9 in B. P, proctodeum. Scale bar-200 μm.

From its first appearance, the ABD-A protein is differentially expressed around the future dorsalventral axis of the embryo. In each segment, expression is first seen in cells in the mediolateral portion of the embryo, underlying the mesoderm. Protein then appears in more laterally located cells, just before constrictions in the mesoderm demarcate the forming segments (Roonwal, 1936; Krause, 1938). Shortly after this the lateral edges of the ectoderm thicken to form the characteristic epidermal bulges of the segments. Just prior to this thickening, the lateral edges of the epidermis begin to stain strongly for ABD-A protein. Highest levels of expression persist in these lateral buds as the segments arise. Later expression spreads dorsally, and then extends completely around the ventral epidermis.

By 31 % of development, the expression of abdominal-A in the locust shows an intrasegmental modulation similar to that of Drosophila with expression in the posterior of the segment stronger than that in the anterior (Fig. 4E).

The conserved morphology of the mature germ band led Seidel (1960) and later Sander (1983) to identify it as a phylotypic stage in insect development - a stage showing markedly less variation between different insect orders than either earlier or later phases of development. This conservation suggests that the genes that define this stage may show conserved patterns of activity. However, the diversity in developmental mechanisms used to generate the germ band indicates that the processes that establish these patterns may differ considerably between species.

The similarity of the mature germ band in Drosophila and in the locust is now documented by two molecular markers, the engrailed gene product, (Patel et al. 1989b) and the ABD-A protein, examined here. For both of these genes, the pattern of expression in the mature germ band of Schistocerca is similar to that seen in Drosophila, but the patterns of activation of the genes differ. In Drosophila their activation is dependent on the spatial distributions of maternal and zygotic gap and pair-rule gene products during blastoderm formation. In Schistocerca it appears that their activation is dependent on cellular interactions that follow a temporal pattern in the postblastoderm embryo. Our cloning of Schistocerca abdominal-A allows us to compare the process of acquisition of segment identity in short and long germ insects.

Schistocerca gregaria possesses a close homologue of Drosophila abdominal-A

The conservation between the locust and Drosophila homologues of abdominal-A is 100% within and flanking the homeodomain. Such an extensive level of similarity between a homeotic gene of Drosophila and that of a different species has only been observed within the insects (Akam et al. 1988; Fleig et al. 1988) and as such does indeed imply that those genes active in the germ band are strongly conserved. This is confirmed by the fact that the patterns of expression of homeotic genes in the germ band of insects are very similar, e.g. Deformed in Apis (Fleig et al. 1988) and abdominal-A and Sex combs reduced in Schistocerca (this report and Dawson, Tear, Martinez-Arias and Akam in preparation). This conservation raises the question of how such highly conserved genes can give rise to speciesspecific segment identity.

Experiments involving the switching of homeoboxes between homeotic genes (Kuziora and McGinnis, 1989) have revealed that many of their regulatory properties are conferred by the homeodomain itself. Single amino acid changes in the homeodomain can change its DNA-binding specificity (Treisman et al. 1989). The high conservation of the homeodomain between Schistocerca and Drosophila abdominal-A means that this domain will be able to acquire the same detailed spatial conformation in both proteins. It is interesting that although both proteins are expressed in comparable domains in the germ band, each species develops unique segment specializations. This suggests that the ABD-A protein must be using the same DNA-binding activity to activate similar sets of downstream genes in these animals, but that during evolution these gene products have changed allowing the development of specific species differences. However, the ABD-A proteins must also regulate those genes common to both animals that promote the differentiation of abdomen rather than thorax or gnathos; this constraint might be an important element in the conservation of the homeodomain.

The activation of abdominal-A expression in Schistocerca

In Schistocerca, as in Drosophila, the expression of abdominal-A characterizes the entire abdomen, from the parasegment border in the middle of the first abdominal segment to the posterior of the overtly segmented region (A7 in Drosophila, A10 in the locust). In Drosophila, the development of the abdominal segments occurs synchronously throughout the whole abdomen (as evidenced by the pattern of the first postblasdoderm mitoses in the region (Foe, 1989)) and ABD-A protein appears at the same time in each segment from Al to A7 (Karch et al. 1990, see Fig. 4F). By contrast, in the locust embryo the abdominal segments develop in sequence, with each segment from A2 to A10 reaching the same apparent stage of development a few hours (1 % of development) later than its anterior neighbour. The expression of abdominal-A first appears in the more anterior part of the abdomen, and spreads backward, appearing in each segment as it reaches approximately the same developmental stage, following the sequence of visible development.

The first expression of abdominal-A appears in cells within the anterior of A2, before any visible signs of segmentation in the abdomen, but shortly after the most anterior abdominal segments have been defined by the appearance of engrailed stripes. It does not appear that the initial expression of abdominal-A is tightly linked to the process of segmentation. The first cells to accumulate detectable levels of ABD-A protein are not those that will abut the final anterior boundary of expression. Neither do they appear in a precise relationship to the nearest stripe of engrailed expressing cells; rather there appears to be some variability in the behaviour of individual cells, as evidenced by the irregular shape and bilateral asymmetry frequently seen in the early patches of cells expressing abdominal-A.

Clearly some property of the developing abdomen becomes permissive for abdominal-A expression, and this happens first in the anterior region of A2. That decision seems to be made on a cell-by-cell basis, during growth. There is no evidence that the abdominal region of the embryo is ‘predetermined’ before cellularization as it is in Drosophila. At present, we do not know what signal might activate gene expression. In Drosophila, the choice between activating thoracic and abdominal homeotic genes is defined by a spatial gradient in the concentration of the gap protein Hunchback (Irish et al. 1989; Hülskamp et al. 1989; Struhl, 1989). We could invoke a similar mechanism in Schistocerca, by assuming that Hunchback protein levels are high in the cells of the embryo during thoracic and gnathal segmentation, but fall rapidly in the cells of the developing abdominal primordium. In addition, since in Drosophila the expression of homeotic genes is modulated by the activity of segment polarity genes (Martinez-Arias et al. 1988), whose products are involved in patterning within cellular fields, it is likely they play an important role in the activation of abdominal-A, once the cells have become permissive for its expression.

Even though the segmentation machinery may not have a role in the initial activation of abdominal-A in Schistocerca, it must have an important role in setting the anterior limit of abdominal-A expression, for this becomes precisely ‘parasegmental’ respecting the equivalent of the A/P compartment boundary in Drosophila. The domain of abdominal-A expression expands anteriorly until it is coincident with the anterior boundary of an engrailed stripe. This anterior expansion may reflect an influence of cellular interactions on the ability of a cell to activate abdominal-A-, for once a cell expresses abdominal-A, its neighbours appear to follow suit soon after. At present we do not understand the mechanisms that act to limit this expansion of expression anteriorly.

The changing anterior boundary of abdominal-A expression in Schistocerca parallels in some respects the activation of the vertebrate Hox genes in the CNS. Hox 2.6 and 2.7 are first expressed in the posterior part of the neural tube, and expression then extends anteriorly until it respects a sharp boundary coincident with the limits of specific rhombomeres in the hindbrain (Wilkinson et al. 1989).

The limits of abdominal-A expression in Schistocerca

In both Drosophila and Schistocerca, abdominal-A is expressed in most abdominal segments, but is not expressed in the anterior compartment of the first abdominal segment. In Drosophila the unique development of the first abdominal segment is under the control of Ultrabithorax, the adjacent gene of the bithorax complex. Ultrabithorax and abdominal-A are the two most similar homeotic genes in Drosophila, both in terms of protein sequence, extensively overlapping domains of expression, and partially redundant functions (Akam et al. 1988). Ultrabithorax and abdominal-A are therefore candidates for the most recently diverged pair among the set of homeotic genes in the insects. The observation that the Schistocerca abdominal-A homologue is not expressed in anterior Al strongly suggests that the functional distinction between Ultrabithorax-Error! Hyperlink reference not valid. genes predates the divergence of the Neopteran insect orders, (ca 300My ago). Unique specialization of the first abdominal segment is clearly seen in early Schistocerca embryos. In common with many other lower insect orders (Anderson, 1973), but in contrast to Drosophila, a pair of limb buds develop in the first abdominal segment. These give rise to the pleuropodia, specialized embryonic organs, which secrete a hatching enzyme and are lost at hatching (Slifer, 1937).

In the Drosophila embryo, the posterior limit of abdominal-A expression extends to parasegment 13 (i.e. A7p/A8a). The tail of the embryo (parasegment 14 and the fused derivatives of more posterior segments (Jiirgens, 1987)) never express detectable ABD-A protein. Overt signs of segmentation are suppressed in this tail region both in the embryo and the adult. It gives rise only to the male genitalia, (A9 or PS14), to specialized sense organs of the posterior spiracle and to anal structures. In Schistocerca, the corresponding region comprises fully formed segments. A9 generates the male genitalia, as in Drosophila-, A10 is without appendages, and All gives rise to the anal cerci. Abdominal-A expression initially extends throughout A9 and A10, but excludes All. This suggests to us that the differential development of the posterior abdomen in Schistocerca and Drosophila may in part depend on the altered regulation of abdominal-A; in Drosophila when Abdominal-B is removed and abdominal-A is expressed in parasegment 14, reduction of the posterior segments is suppressed, and the denticle belt of a typical abdominal segment appears in A9 (Karch et al. 1990; Casanova et al. 1986).

We thank P. Lasko for his generous and patient assistance and advice concerning fusion proteins and the production of antisera, C. M. Bate and D. Shepherd for teaching us locust embryogenesis and together with I. Dawson and A. Bejsovec for many helpful discussions. G.T. thanks M. Metzstein for help in producing Fig. 2 and C. Fox for technical assistance. This work was supported by a Medical Research Council grant to M.A. and G.T and a Wellcome Trust senior fellowship to A.M.-A. The sequence data reported in this article will appear in the EMBL/Genbank nucleotide sequence database with accession number X54674.

Adams
,
J. C.
(
1981
).
Heavy metal intensification of DAB-based HRP reaction product
.
J. Histochem. Cytochem
.
29
,
775
.
Akam
,
M.
(
1987
).
The molecular basis for metameric pattern in the Drosophila embryo
.
Development
101
,
1
22
.
Akam
,
M.
,
Dawson
,
I.
and
Tear
,
G.
(
1988
).
Homeotic genes and the control of segment identity
.
Development
104
Supplement
123
133
.
Akam
,
M. E.
and
Martinez-Arias
.
A.
(
1985
).
The distribution of Ultrabithorax transcripts in Drosophila embryos
.
EMBO J
.
4
,
1689
1700
.
Anderson
,
D. T.
(
1973
).
Embryology and Phylogeny in Annelids and Arthropods
.
Oxford
:
Pergamon
.
Bentley
,
D.
,
Keshishian
,
H.
,
Shankland
,
M.
and
Toroian-Raymond
.
A.
(
1979
).
Quantitative staging of embryonic development of the grasshopper Schistocerca nitens
.
J. Embryol. exp. Morph
.
54
,
47
74
.
Breathnach
,
R.
and
Chambon
,
P.
(
1981
).
Organization and expression of eukaryotic split genes coding for proteins
.
A. Rev. Btochem
.
50
,
349
383
.
Campos-Ortega
,
J. A.
and
Hartenstein
,
V.
(
1985
).
The Embryonic Development of Drosophila melanogaster
.
Berlin
:
Springer-Verlang
.
Carroll
,
S. B.
and
Laughon
,
A.
(
1987
). Production and purification of polyclonal antibodies to the foreign segment of β galactosidase fusion proteins.
In DNA Cloning-A Practical Approach. Vol. 3
.
Oxford
:
IRL Press Limited
.
Casanova
,
J.
,
Sánchez-Herrero
,
E.
and
Morata
,
G.
(
1986
).
Identification and characterization of a parasegmental specific regulatory element of the Abdominal-B gene of Drosophila
.
Cell
47
,
627
636
.
Devereux
,
J.
,
Haeberli
,
P.
and
Smithies
,
O.
(
1984
).
A comprehensive set of sequence analysis programs for the VAX
.
Nucl. Acids Res
.
12
,
387
395
.
Dinardo
,
S.
,
Kuner
,
J. M.
,
Theis
,
J.
and
O’Farrell
,
P. H.
(
1985
).
Development of embryonic pattern in D. melanogaster as revealed by accumulation of the nuclear engrailed protein
.
Cell
43
,
59
69
.
Feinberg
,
A. P.
and
Vogelstein
,
B.
(
1983
).
A technique for radiolabeling DNA restriction endonuclease fragments to high specific activity
.
Anal. Btoch
.
132
,
6
13
.
Eiose
,
A.
,
Mcginnis
,
W. J.
and
Gehring
,
W. J.
(
1985
).
Isolation of a homeobox containing gene from the engrailed region of Drosophila and the spatial distribution of its transcripts
.
Nature
313
,
284
289
.
Fleig
,
R.
,
Walldorf
,
U.
,
Gehring
,
W. J.
and
Sander
,
K.
(
1988
).
In-situ localization of the transcripts of a homeobox gene in the honey bee Apis mellifera L
.
(Hymenoptera). Roux’s Arch, devl Biol
.
197
,
269
274
.
Foe
,
V. E.
(
1989
).
Mitotic domains reveal early commitment of cells in Drosophila embryos
.
Development
107
,
1
22
.
Hames
,
B. D.
(
1981
). An introduction to polyacrylamide gel electrophoresis.
In Gel Electrophoresis of Proteins: A Practical Approach
, (ed.
B. D.
Hames
and
D.
Rickwood
.)
Oxford, England
:
IRL Press
.
Harlow
,
E.
and
Lane
,
D.
(
1988
).
Antibodies: A Laboratory Manual
.
New York
:
Cold Spring Harbor Laboratory
.
Howard
,
K.
(
1988
).
The generation of periodic pattern during early Drosophila embryogenesis
.
Development
104
Supplement
35
50
.
Hülskamp
,
M.
,
Schroder
,
C.
,
Pfeifle
,
C.
,
Jâckle
,
H.
and
Tautz
,
D.
(
1989
).
Posterior segmentation of the Drosophila embryo in the absence of a maternal posterior organizer gene
.
Nature
338
,
629
632
.
Ingham
,
P. W.
(
1988
).
The molecular genetics of embryonic pattern formation in Drosophila
.
Nature
335
,
25
34
.
Irish
,
V. F.
,
Martinez-Arias
.
A.
and
Akam
,
M.
(
1989
).
Spatial regulation of the Antennapedta and Ultrabithorax homeotic genes during Drosophila early development
.
EMBO J
.
8
,
1527
1537
.
Jurgens
,
G.
(
1987
).
Segmental organisation 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, devl Biol
.
196
,
141
157
.
Karch
,
F.
,
Bender
,
W.
and
Weiffenbach
,
B.
(
1990
).
abd-A expression in Drosophila embryos, submitted to Genes and Dev
.
Kornberg
,
T.
,
Sidén
,
L
,
O’farrell
,
P.
and
Simon
,
M.
(
1985
).
The engrailed locus of Drosophila: in situ localization of transcripts reveals compartment-specific expression
.
Cell
40
,
45
53
.
Krause
,
G.
(
1938
).
Die Ausbuildung der Korpergrundgestalt im Ei der Gewachschausschrecke Tachycines asynamorus. Z
.
Morph, und Okol. der Tiere
34
,
499
564
.
Krause
,
G.
(
1939
).
Die Eitypen der Insekten
.
Biol. Zbl
.
59
,
495
536
.
Krause
,
G.
(
1953
).
Die Aktionsfolge zur Gestaltung des Keimstreifs von Tachycines (Saltatoria) insbesondere das morthogenetische Konstruktionsbild bei Duplicitas parallela
.
Wilhelm Roux Arch. EntwMech. Org
.
146
,
275
370
.
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
.
Lewis
,
E. B.
(
1978
).
A gene complex controlling segmentation in Drosophila
.
Nature
276
,
565
570
.
Lipman
,
D. J.
and
Pearson
,
W. R.
(
1985
).
Rapid and sensitive protein similarity searches
.
Science
227
,
1435
1441
.
Macdonald
,
P. M.
,
Ingham
,
P. W.
and
Struhl
,
G.
(
1986
).
Isolation, structure and expression of even-skipped: A second pair-rule gene of Drosophila containing a homeo-box
.
Cell
47
,
721
734
.
Martinez-Arias
,
A.
(
1989
).
A cellular basis for pattern formation in the insect epidermis
.
Trends Genet
.
5
.
262
267
.
Martinez-Arias
,
A.
,
Baker
,
N. E.
and
Ingham
,
P. W.
(
1988
).
Role of segment polarity genes in the definition and maintenance of cell states in the Drosophila embryo
.
Development
103
,
157
170
.
Martinez-Arias
,
A.
and
Lawrence
,
P. A.
(
1985
).
Parasegments and compartments in the Drosophila embryo
.
Nature
313
,
639
642
.
Mcginnis
,
W.
,
Levine
,
M. S.
,
Hafen
,
E.
,
Kuroiwa
,
A.
and
Gehring
,
W. J.
(
1984
).
A conserved DNA sequence in homeotic genes of the Drosophila Antennapedia and bithorax complexes
.
Nature
308
,
428
433
.
Mee
,
J.
(
1986
).
Pattern formation in fragmented eggs of the short germ insect, Schistocerca gregaria
.
Roux’s Arch, devl Biol
.
195
,
506
512
.
Mee
,
J.
and
French
,
V.
(
1986
).
Disruption of segmentation in a short germ insect embryo I. The localization of segmental abnormalities induced by heat shock
.
J. Embryol. exp. Morph
.
96
,
245
266
.
Patel
,
N. H.
,
Kornberg
,
T. B.
and
Goodman
,
C. S.
(
1989b
).
Expression of engrailed during segmentation in grasshopper and crayfish
.
Development
107
,
201
212
.
Patel
,
N. H.
,
Martin-Blanco
.
E.
,
Coleman
,
K. G.
,
Poole
,
S. J.
,
Ellis
,
M. C.
,
Kornberg
,
T. B.
and
Goodman
,
C. S.
(
1989a
).
Expression of engrailed proteins in arthropods, annelids and chordates
.
Cell
58
,
955
968
.
Roonwal
,
M. L.
(
1936
).
Studies on the embryology of the African migratory locust Locusta migratoria migratoroides. I. Early development with a new theory of multiphased gastrulation among insects
.
Phil. Trans. R. Soc. B
226
,
391
421
.
Rowe
,
A.
(
1987
).
Studies on the expression and regulation of the bithorax complex of Drosophila. D
.
Phil. Thesis University of Cambridge
.
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.
,
Vern
,
S. L
,
Marco
,
R.
and
Morata
,
G.
(
1985
).
Genetic organization of Drosophila bithorax complex
.
Nature
313
,
108
113
.
Sander
,
K.
(
1976
).
Specification of the basic body pattern in insect embryogenesis
.
Adv. Insect Physiol
.
12
,
125
238
.
Sander
,
K.
(
1983
). The evolution of patterning mechanisms: gleanings from insect embryogenesis and spermatogenesis.
In Development and Evolution
(ed.
B. C.
Goodwin
,
N.
Holder
and
C. C.
Wylie
) pp.
137
159
.
Cambridge University Press
.
Sander
,
K.
(
1988
).
Studies in insect segmentation: from teratology to phenogenetics
.
Development
104
Supplement
112
121
.
Sanger
,
F.
,
Nicklen
,
S.
and
Coulson
,
A. R.
(
1977
).
DNA sequencing with chain terminating inhibitors
.
Proc. natn. Acad. Sci. U.S.A
.
74
,
5463
5467
.
Seidel
,
F.
(
1960
).
Korpergrundgestalt und Keimstruktur. Eine Erorterungiiber die Grundlagen der vergleichenden und experimentellen Embryologie und deren Gultigkeit bei pylogenetischen Úberlegungen
.
Zool. Anz
.
164
,
245
305
.
Slifer
,
E. H.
(
1937
).
The origin and fate of the membranes surrounding the grasshopper egg: together with some experiments on the source of the hatching enzyme
.
Q. J. microsc. Sci
.
79
.
493
506
.
Smith
,
D. B.
and
Johnson
,
K. S.
(
1988
).
Single step purification of polypeptides expressed in Escerichia coli as fusions with glutathione transferase
.
Gene
67
,
31
40
.
Snodgrass
,
R. E.
(
1935
).
Principles of Insect Morphology
.
London and New York
:
McGraw-Hill
.
Staden
,
R.
(
1984
).
Graphic methods to determine the function of nucleic acid sequences
.
Nucl. Acids Res
.
12
.
521
538
.
Struhl
,
G.
(
1989
).
Differing strategies for organizing anterior and posterior body pattern in Drosophila embryos
.
Nature
338
,
741
744
.
Tear
,
G.
,
Bate
,
C. M.
and
Martinez-Arias
,
A.
(
1988
).
A phylogenetic interpretation of the patterns of gene expression in Drosophila embryos
.
Development 104 Supplement 135-145
.
Tiong
,
S.
,
Bone
,
L. M.
and
Whittle
,
J. R. S.
(
1985
).
Recessive lethal mutations within the bithorax-complex in Drosophila
.
Molec. gen. Genet
.
200
,
335
342
.
Treismann
,
J.
,
Gonczy
,
P.
,
Vashishtha
,
M.
,
Harris
,
E.
and
Desplan
,
C.
(
1989
).
A single amino acid can determine the DNA binding specificity of homeodomain proteins
.
Cell
59
,
553
562
.
Wilkinson
,
D. G.
,
Bhatt
,
S.
,
Cook
,
C.
,
Boncinelli
,
E.
and
Krumlauf
,
R.
(
1989
).
Segmental expression of Hox2 homeobox-containing genes in the chick hindbrain
Nature
341
405
409
.
Wilmore
,
P. J.
and
Brown
,
A. K.
(
1975
).
Molecular properties of Orthopteran DNA
.
Chromosoma (Berl.)
51
,
337
345
.