In contrast to the segmentation of the embryonic trunk region which has been extensively studied, relatively little is known about the development and segmentation of the Drosophila head. Proper development of the cephalic region requires the informational input of three of the four maternal coordinate systems. Head-specific gene expression is set up in response to a complex interaction between the maternally provided gene products and zygotically expressed genes. Several zygotic genes involved in head development have recently been characterized. A genetic analysis suggests that the segmentation of the head may use a mechanism different from the one acting in the trunk. The two genes of the sloppy paired locus (slp1 and slp2) are also expressed in the embryonic head. slp1 plays a predominant role in head formation while slp2 is largely dispensible. A detailed analysis of the slp head phenotype suggests that slp is important for the development of the mandibular segment as well as two adjacent pregnathal segments (antennal and ocular). Our analysis of regulatory interactions of slp with maternal and zygotic genes suggests that it behaves like a gap gene. Thus, phenotype and regulation of slp support the view that slp acts as a head-specific gap gene in addition to its function as a pair-rule and segment polarity gene in the trunk. We show that all three maternal systems active in the cephalic region are required for proper slp expression and that the different systems cooperate in the patterning of the head. The terminal and anterior patterning system appear to be closely linked. This cooperation is likely to involve a direct interaction between the bcd morphogen and the terminal system. Low levels of terminal system activity seem to potentiate bcd as an activator of slp, whereas high levels down-regulate bcd rendering it inactive. Our analysis suggests that dorsal, the morphogen of the dorsoventral system, and the head-specific gap gene empty spiracles act as repressor and corepressor in the regulation of slp. We discuss how positional information established independently along two axes can act in concert to control gene regulation in two dimensions.

Extensive studies on the development of the Drosophila embryo have provided deep insight into the mechanisms governing embryonic pattern formation. These studies have focused primarily on the trunk region of the embryo and led to the discovery of an intricate cascade of hierarchical, cross- and autoregulatory gene interactions that progressively subdivide the embryo into increasingly smaller functional units (for review see e.g. Ingham, 1988; Pankratz and Jäckle, 1990). In contrast to the well understood segmentation of this central region, relatively little is known about the mechanisms involved in the patterning of the embryonic head. Recently, however, several genes have been identified that are involved in the segmentation of the topologically complex cephalic region (for review see Finkelstein and Perrimon, 1991; Cohen and Jürgens, 1991). Genetic analysis suggests that the mechanisms controlling head development may be different from the ones governing the segmentation of the trunk. To further our understanding of pattern formation in the head it is of utmost importance to unravel the genetic controls specifying the spatial expression of such head specific genes.

In the Drosophila embryo the primary determination of cell fates along the anteroposterior and dorsoventral axes depends on the complex interaction of maternally expressed genes and zygotic effector genes (Nüsslein-Volhard et al., 1987; Manseau and Schüpbach, 1989). Three distinct groups of maternal coordinate genes are involved in the establishment of positions along the anteroposterior axis and one along the dorsoventral axis (see reviews by Nüsslein-Volhard, 1991; St. Johnston and Nüsslein-Volhard, 1992). The activity of these systems results in the spatially restricted transcriptional activation of zygotic genes that subdivide the embryo into different functional domains. Three maternal systems are required for proper development of the embryonic head: the anterior system determining the head and thorax region, the terminal system, responsible for the formation of the unsegmented regions of head (acron) and tail (telson) and the dorsoventral system defining positional information along the dorsoventral axis.

The morphogen of the anterior system, bicoid (bcd), is distributed in a monotonic gradient with its highest concentration at the anterior pole (Driever and Nüsslein-Volhard, 1988a,b). Different concentration levels of bcd, a homeodomain protein, lead to the activation of different zygotic genes (Tautz, 1988; Driever and Nüsslein-Volhard, 1989; Struhl et al., 1989; Driever et al., 1989; Gaul and Jäckle, 1989; Hülskamp et al., 1990; Eldon and Pirrotta, 1991). The best studied example is the zygotic expression of the gap gene hunchback (hb), which is directly activated by the binding of bcd to enhancer elements in the hb promoter (Driever and Nüsslein-Volhard, 1989; Struhl et al., 1989). Since the hb phenotype does not account for the absence of all head structures in bcd embryos (Lehmann and Nüsslein-Volhard, 1987), Driever et al. (1989) proposed that at least one other bcd-activated gene (termed gene X) controls the development of head regions anterior to the functional domain of hb. Three genes that may play such a role, empty spiracles (ems), orthodenticle (otd) and buttonhead (btd) have been cloned (Dalton et al., 1989; Finkelstein et al., 1990; Walldorf and Gehring, 1992; Wimmer et al., 1993). The head-specific expression of ems and otd is dependent on bcd activity and responds to altered bcd concentration gradients (Dalton et al., 1989; Finkelstein and Perrimon, 1990; Walldorf and Gehring, 1992).

The genes of the terminal system also establish graded positional information (Casanova and Struhl, 1989). They encode components of a signal transduction cascade (Perrimon, 1993). The uniformly distributed torso (tor) receptor tyrosine kinase (Sprenger et al., 1989) is locally activated at the poles by the torso-like (tsl) gene product (Stevens et al., 1990; Savant-Bhonsale and Montell, 1993; Martin et al., 1994). Ras1, encoding a guanine nucleotide-binding protein, relays the tor signal to pole hole (phl), the Drosophila homolog of the Raf-1 serine/threonine kinase (Nishida et al., 1988; Ambrosio et al., 1989; Lu et al., 1993). The transmission of the signal further involves the Drosophila homologs of the tyrosine/threonine kinase Mek (Dsor1; Tsuda et al., 1993) and the mitogen-activated protein (MAP) kinase (rolled; Brunner et al., 1994). The terminal signalling cascade eventually leads to the transcriptional activation of the zygotic gap genes huckebein (hkb) and tailless (tll) by an unknown transcription factor(s) (Strecker et al., 1986; Klingler et al., 1988; Strecker et al., 1989; Casanova, 1990; Weigel et al., 1990; Pignoni et al., 1990 and 1992; Brönner and Jäckle, 1991).

The dorsoventral system also encodes members of a signal transduction pathway (for review see Stein and Stevens, 1991; St. Johnson and Nüsslein-Volhard, 1992). A localized signal, possibly the spätzle gene product (Morisato and Anderson, 1994), is produced only ventrally through the action of several other dorsal group genes including easter (ea). Activation of the Toll (Tl) transmembrane receptor and subsequent intracellular propagation of the signal leads to dissociation of a complex between dorsal (dl) a transcription factor homologous to NF-κB (Steward, 1987; Ghosh et al., 1990), and cactus (cact), a I-κB homolog (Kidd, 1992; Geisler et al., 1992). dl, originally trapped by cact in the cytoplasm, can now enter the nucleus and regulate gene expression. The asymmetric activation of Tl thus leads to a gradient of nuclear dl protein with its highest levels on the ventral side. dl acts both as an activator (Roth et al., 1989; Jiang et al., 1991; Pan et al., 1991; Thisse et al., 1991; Ip et al., 1992a,b; Jiang and Levine, 1993) and a repressor of zygotic genes (Rushlow et al., 1987; St. Johnston and Gelbart, 1987; Ip et al., 1991). Whereas dl appears to have an inherent property to act as a transcriptional activator, in order to function as a repressor it interacts with another factor(s) or co-repressor(s) (Jiang et al., 1993, Kirow et al., 1993).

Phenotypic characterization of maternal effect mutants showed that the four maternal systems are largely independent of each other (Nüsslein-Volhard et al., 1987). An exception to this generalization is provided by the development of the anterior-most part of the embryo. Both, the anterior and terminal systems are required for proper development of the acron (Frohnhöfer and Nüsslein-Volhard, 1986). In fact, bcd appears to be directly regulated by the tor pathway (Ronchi et al., 1993). The positional information provided by the maternal coordinate systems is interpreted by zygotic target genes. These often show complex expression patterns that are modulated along both the anteroposterior and dorsoventral axes suggesting that the establishment of region-specific gene expression in the cephalic region depends on the concerted action of the maternal systems determining both embryonic axes. Analysis of the genetic controls specifying such expression domains is required to understand how the independently established positional cues are integrated to ensure proper development of the embryo.

The slp locus encodes two transcription units, slp1 and slp2, both of which contain a fork head DNA-binding domain (Grossniklaus et al., 1992; Weigel and Jäckle, 1990; Clark et al., 1993) which is found in organisms from yeast to mammals (Häcker et al., 1992; Clevidence et al., 1993). The two putative transcription factors are expressed in a highly dynamic pattern showing characteristics of all three segmentation gene classes, the gap, pair-rule and segment polarity genes (Grossniklaus et al., 1992). The spatial expression patterns of the two genes are identical with the exception of the early expression in the embryonic head (see below; Grossniklaus et al., 1992). Amorphic slp1 and slp2 double mutants show strong segmentation defects and a failure of head involution (Grossniklaus et al., 1992).

Here we concentrate on the control and function of the early gap gene-like expression of slp in the embryonic head. We analyzed the early head-specific expression of slp1 and slp2 in wild-type and mutant embryos. In all mutants analyzed we obtained the same results for slp1 and slp2 supporting our previous hypothesis that the two genes are regulated by a common enhancer region upstream of slp1 (Grossniklaus et al., 1992). Phenotypic analysis of the slp head phenotype indicates that slp1 plays a predominant role in head development whereas slp2 is largely dispensible. slp expression depends on bcd and responds to alterations of the bcd concentration gradient suggesting that the slp genes are two further candidates for gene X. Head-specific expression of slp is generated in response to complex interactions between the anterior, terminal and dorsoventral systems. Our analysis provides evidence that the three maternal systems acting in the head cooperate in the regulation of slp. Our findings are in agreement with a recent report by Ronchi et al. (1993) suggesting that bcd itself may be a target of the terminal system.

Analysis of the modulation of the slp expression domain along the dorsoventral axis suggests that the head-specific gap gene ems may function as a co-repressor of dl thereby linking the anteroposterior and dorsoventral systems at the molecular level.

Fly stocks

The wild-type strain used was Oregon-R. The slp mutations CyO, △46G P[lArB]A208.1M2 and CyO, △34B P[lArB]A208.1M2 (referred to as △46G and △34B) have been described earlier (Gross-niklaus et al., 1992). The strain carrying four copies of bcd has been described by Driever and Nüsslein-Volhard (1988b). A detailed description of other mutants used can be found in Lindsley and Zimm (1992) and references therein. Here we refer to the allele names used in the new Red Book but also give the original allele names in brackets. Alleles of maternal effect mutants were bcd6 (bcdE1), spz2 (spz197), Tlr2 (Tlr444) and ndl3 (ndl111). Since many of the chromosomes carrying maternal effect mutations have accumulated lethal mutations or show very low fertility when homozygous, we used trans-heterozygous combinations to increase the yield in embryo collections. Terminal mutants: tor1/tor4 (torWK/torPM), tsl3/tsl4 (tsl146/tsl691) and the dominant alleles tor13D (tor4021) and rlSem (Brunner et al., 1994). Dorsalizing mutants: dl1/Df(2L)TW119, ea3/ea4, pll315/Df(3R)IR16, tub3/Df(3R)XM3, pip1/pip2 (pip386/pip664), snk1/snk2 (snk073/snk229), Tlrv4/Df(3R)Tl-X (TlRXH/Df(3R)9QRX1) and Tlrv18/Df(3R)Tl-X (Tl5BRXV/Df(3R)9QRX1). Ventralizing mutants: cact7/cact12 (cactA2/cactO11) and dominant alleles of Tl and ea : Tl2 (Tl5B), Tl3 (Tl9Q), Tl8 (Tl10b), eaD3 (ea83l) and eaD6 (ea125.3). The embryos of cact 7/cact12 mothers display a phenotype very similar to the one described for embryos of cact7 homozygous mothers [phenotypic class V2 (Roth et al., 1991), the strongest class for viable cact alleles]. Zygotic lethal alleles used in this study were tll1 (tllL10), hkb132, sna1 (snaHG), twi1 (twi1D96), ems2 (ems9H16) and ems3 (ems9Q64), oc9 (otdYH), btdIIIA, hb12 (hb14F), gtX11, Kr9.

Germline phl7 (l(1)EA75) mutant females were generated by site-specific mitotic recombination using the yeast FLP recombinase in combination with the dominant female sterile technique (Chou and Perrimon, 1992). The stocks were described in Chou and Perrimon (1992). Males carrying an X-chromosome with a FLP recombinase target sequence (FRT) close to the centromer as well as the dominant female sterile mutation ovoD1 were crossed to females carrying a X-chromosome with the same FRT and the phl7 allele. The males also carried the FLP recombinase under heat shock control on both second chromosomes. Progeny from this cross were grown at 25°C and heat shocked at puparium formation for 2 hours at 37°C. Females of the desired genotype were used for embryo collections. Embryos displayed a strong terminal cuticular phenotype (not shown).

Germline clones of the lethal cact alleles cact2 (cactD12) and cact3 (cactD13) were generated by the dominant female sterile technique (Perrimon, 1984; Perrimon et al., 1986). A second chromosome with two copies of a P-element insertion carrying the dominant female sterile allele ovoD1 at the cytological positions 28A and 30D was kindly provided by Tze-Bin Chou and Norbert Perrimon before publication. This stock is described in Chou et al. (1993). Males carrying this chromosome were crossed to females carrying one copy of the lethal cact allele. Progeny from this cross were irradiated at 1000 rad during the first instar stage and allowed to develop at 25°C. Females having both the cact as well as the ovoD1 chromosomes were selected and used for embryo collections. The embryos derived from these mothers were a mixture of wild-type (mitotic recombination event between cact and the P-element carrying ovoD1) and cact mutants (mitotic recombination event between cact and the centromer). These cact mutants display a very strong ventralized phenotype showing small patches of poorly differentiated ventral cuticle [phenotypic class V1 (Roth et al., 1991)]. Rarely, ventral denticle belts can be seen around most of the circumference of the embryo, especially close to the poles.

RNA and protein detection

In situ hybridization to whole-mount embryos was performed essentially according to the procedure developed by Tautz and Pfeifle (1989) with modifications (a detailed lab protocol is available upon request). Embryos were viewed with Nomarski optics and photographed with a Kodak Ektrachrome 64 slide film at a setting of 25 ASA or Kodak Ektar 100 colour print film at 50 ASA.

Full-length slp1 and slp2 cDNAs were used as probes. A comparison with results obtained using specific probes derived from the C terminus of slp1 and slp2 (Grossniklaus et al., 1992) showed that no cross-hybridization between the conserved regions of slp1 and slp2 is detected under these conditions. The ems and en probes were cW13/7 (Walldorf and Gehring, 1992) and pF7036 (Fjose et al., 1985). wg, pBS-CV (Rijsewijk et al., 1987) was a generous gift of M. van den Heuvel and R. Nusse (Stanford University, California) and an otd probe (Finkelstein et al., 1990) was obtained from R. Finkelstein and N. Perrimon (Harvard Medical School, Massachusetts).

The protocol for antibody stainings combined with in situ hybridization is essentially as described by Manoukian and Krause (1992). First, antibody staining was performed as described by Frasch et al. (1987) and Grossniklaus et al. (1992) with the modification that PBTH [PBS (500 mM NaCl), 0.1% Tween 20, 50 μg/ml heparin, 250 μg/ml tRNA, 0.05 U/ml RNAsin] was used instead of PBT [PBS (500 mM NaCl), 0.1% Tween 20]. The primary rat α slp1 antibody (Cadigan et al., 1994b) was used at a dilution of 1:300, the secondary biotinylated goat α rat antibody (Vectastain Elite Kit) at 1:200. After antibody detection in situ hybridization was performed as described above.

Data analysis

To analyze slp expression in zygotic mutants we analyzed 20-30 embryos of the appropriate stage and orientation. Pictures were taken under Nomarski optics using a Panasonic CL110 video camera and a Mitsubishi video copy processor (Model P68E). The prints on Sony high density printing paper UPP-110HD were used for measuring the extent of slp expression in each embryo and converted to % EL. Means and standard deviations were calculated for the position of the expression boundaries. If the standard deviation was more than 2%, the data points were plotted to see whether they could be resolved into two peaks segregating 1:3. A change of less than 2% EL was considered as not significant. Where the slp expression was obviously altered or in maternal mutants, between 10 and 15 mutant embryos were analyzed. In zygotic mutants approx. 25% of the embryos at the appropriate stage displayed the altered pattern. In addition, the gap mutants hb, gt and Kr display an obvious phenotype of the slp pairrule expression and the final head expression (head cap and head stripe) could be analyzed unambiguosly in embryos of this stage. tll and hkb embryos were unambigously identified by the altered ftz expression pattern in these mutants.

Scanning electron microscopy

Wild-type, △46G and △34B embryos were collected for 4 hours and aged for another 8 hours to isolate embryos in the developmental time window of 8-12 hours of embryogenesis. They were dechorionated, fixed and devitellinized as for in situ hybridization. All solutions used were filtered through a 0.2 μm filter unit to avoid disturbing dirt particles. They were critical point dried using acetone as the transition phase in an automatic CPD-apparatus. Finally, they were mounted on double sided sticky tape and coated with a thin layer of gold. Observation was with a Hitachi S800 field emission scanning electron microscope. Pictures were taken using Ilford FP4 plus 125 ASA rollfilm.

Head-specific expression of slp in wild-type embryos

As a reference point for our analysis of slp expression in mutants we briefly describe the early expression patterns of the two genes. We can distinguish four phases of slp expression in the head: initiation, anterior repression, ventral repression and splitting. slp1 starts to be expressed in the anterior 30% of the embryo during syncytial blastoderm at nuclear division cycle 9 (Fig. 1A). Subsequently, slp1 gets repressed at the anterior pole (Fig. 1B). Progressive anterior repression (Fig. 1C) leads to the establishment of a circumferential ring ranging ventrally from about 71% EL to approx. 87% EL (EL = egg length, 0% being the posterior pole) and dorsally from 81-70% EL (Fig. 1D). During the early stages of cellularization the anterior half of the head ring starts to get repressed ventrally (Fig. 1E). Then the splitting of the head expression into an anterior cap and a narrow posterior stripe occurs (Fig. 1F). slp2 is not initiated as early as slp1 (Fig. 1G) and expression is only detectable at the very end of syncytial blastoderm (Fig. 1H). Unlike slp1, slp2 appears immediately as a ring (Fig. 1I). The subsequent phases of slp2 expression are identical to the ones described for slp1, both at the spatial and temporal level (Fig. 1J-L).

Fig. 1.

slp1 and slp2 head-specific expression in wild-type embryos. slp1 (A-F) and slp2 (G-L) mRNA was detected by whole-mount in situ hybridization. In all figures embryos are oriented with their anterior pole to the left. Dorsal is uppermost except where stated otherwise. Stages in all figures are according to Campos-Ortega and Hartenstein (1985). Subdivisions of stage 5 [5(1), 5(2) and 5(3)] are according to Lawrence and Johnston (1989) and can be distinguished by the amount the membranes have grown in during cellularization. (A) Stage 4 embryo at nuclear division cycle (NDC) 9. Very faint slp1 expression is detectable in the head region of the embryo (anterior to the arrowheads). (B) Stage 4 embryo (approx. NDC 12). Repression at the anterior pole (arrowhead). (C) Early stage 5(1) embryo where nuclei have elongated. Anterior repression is not yet completed. (D) Stage 5(1) embryo. Membrane ingrowth has started and anterior repression is completed. (E) Stage 5(1) embryo. Ventral repression occurs (arrowhead). Due to beginning repression at the posterior end of the expression domain the posterior border is less sharply defined than in C and D. (F) Stage 5(2) embryo. Splitting of head expression (arrowhead) into a head stripe (HS) and an anterior cap. Eventually, the cap and stripe will be separated by about approx. 3 cells and the stripe will have a width of approx. 2 cells. Note that the developing head stripe is slightly broader ventrally. Repression in the 2 to 3 posterior-most cells does not occur in the ventral region where the stripe will retain a width of 4 to 5 cells (this is more visible in Fig. 6). (G) Stage 4 embryo (approx. NDC 12). No slp2 expression detectable. (H) Stage 4 embryo (NDC 13). Initiation of slp2 expression as an immediate ring (arrowhead). (I) Stage 5(1) embryo. Staining in the head ring is fully established. (J) Stage 5(1) embryo. Ventral repression (arrowhead). (K) Stage 5(1)/5(2) embryo. Splitting of the head expression starts (arrowhead). Note the larger width of the developing head stripe on the ventral side. (L) Stage 5(2) embryo. Head expression is completely separated (arrowhead) into a cap and a head stripe (HS).

Fig. 1.

slp1 and slp2 head-specific expression in wild-type embryos. slp1 (A-F) and slp2 (G-L) mRNA was detected by whole-mount in situ hybridization. In all figures embryos are oriented with their anterior pole to the left. Dorsal is uppermost except where stated otherwise. Stages in all figures are according to Campos-Ortega and Hartenstein (1985). Subdivisions of stage 5 [5(1), 5(2) and 5(3)] are according to Lawrence and Johnston (1989) and can be distinguished by the amount the membranes have grown in during cellularization. (A) Stage 4 embryo at nuclear division cycle (NDC) 9. Very faint slp1 expression is detectable in the head region of the embryo (anterior to the arrowheads). (B) Stage 4 embryo (approx. NDC 12). Repression at the anterior pole (arrowhead). (C) Early stage 5(1) embryo where nuclei have elongated. Anterior repression is not yet completed. (D) Stage 5(1) embryo. Membrane ingrowth has started and anterior repression is completed. (E) Stage 5(1) embryo. Ventral repression occurs (arrowhead). Due to beginning repression at the posterior end of the expression domain the posterior border is less sharply defined than in C and D. (F) Stage 5(2) embryo. Splitting of head expression (arrowhead) into a head stripe (HS) and an anterior cap. Eventually, the cap and stripe will be separated by about approx. 3 cells and the stripe will have a width of approx. 2 cells. Note that the developing head stripe is slightly broader ventrally. Repression in the 2 to 3 posterior-most cells does not occur in the ventral region where the stripe will retain a width of 4 to 5 cells (this is more visible in Fig. 6). (G) Stage 4 embryo (approx. NDC 12). No slp2 expression detectable. (H) Stage 4 embryo (NDC 13). Initiation of slp2 expression as an immediate ring (arrowhead). (I) Stage 5(1) embryo. Staining in the head ring is fully established. (J) Stage 5(1) embryo. Ventral repression (arrowhead). (K) Stage 5(1)/5(2) embryo. Splitting of the head expression starts (arrowhead). Note the larger width of the developing head stripe on the ventral side. (L) Stage 5(2) embryo. Head expression is completely separated (arrowhead) into a cap and a head stripe (HS).

Head phenotype in slp mutants

During stages 13 and 14 of Drosophila embryogenesis the process of head involution takes place: a complex rearrangement of the six head lobes, which migrate inwardly through the stomodeal opening (Turner and Mahowald, 1979). After involution the ectodermal cells of the head lobes form the cuticular and sensory structures of the head. Their normal morphogenesis depends on proper juxtaposition and fusion of the involuted lobes. We have previously described how structures derived from the mandibular and labial segments are missing in terminally developed slp embryos. This finding was based on the analysis of hypomorphic alleles, since in amorphic alleles of slp1 (△46G) and slp1 and slp2 (△34B) head involution is disrupted and primary defects cannot be distinguished from secondary ones that result from a failure of head involution. The lack of structures derived from the labial segment is qualitatively different from the defects associated with the mandibular segment: whereas the mandibular lobe appears not to form at all, the labial lobe deteriorates together with the mesothoracic and odd-numbered abdominal segments (corresponding to the pair-rule phenotype of these mutants; Gross-niklaus et al., 1992). We now report on an analysis of embryos homozygous for the amorphic slp alleles △46G and △34B by scanning electron microscopy (SEM) and by using wingless (wg) and engrailed (en) as molecular markers for head segments.

At full germband extension (stage 11) the parasegmental grooves and the three gnathal lobes, mandibular, maxillary and labial, become visible (Fig. 2A). In △46G homozygous embryos maxillary and labial lobes appear to be normal but the mandibular lobe is completely missing (Fig. 2B) The same is true for △34B homozygous embryos (Fig. 2C). The labial segment is still present in slp mutants at stage 11. However, during germband retraction (stage 13) the labial lobe deteriorates leaving slp mutants with only the maxillary lobe (not shown). Since both △46G and △34B homozygous embryos show the same head phenotype we conclude that slp1 is predominantly involved in head development. Indeed, slp2 null mutants (Cadigan et al., 1994a) show normal head involution and only very slight head defects in terminally differentiated embryos: all elements of the cephalopharyngeal skeleton are present with the only defect being the length of the Lateral-gräten which is often slightly reduced (not shown). The Later-algräten are derived from the mandibular segment (Jürgens et al., 1986). Head defects were also analyzed using molecular markers, which are expressed in all head segments in wild-type embryos (Fig. 2D,F, for a detailed description see Schmidt-Ott and Technau, 1992). In slp1 null mutants (△46G) the mandibular wg (Fig. 2E,G) and en stripes (not shown) are clearly absent confirming our morphological observations. In addition, the wg antennal stripe is not detectable and the head blob staining is partially missing (Fig. 2D,E; see also Cadigan et al., 1994a). The en antennal stripe is fading in △34B mutants and the head spot is absent (not shown; see also Cadigan et al., 1994b). Whereas the mandibular wg and en stripes appear not to be established at all, the other pattern elements decay during the development of the cephalic region. Thus, slp1 is required for the establishment of the mandibular segment primordium and for proper development of the antennal as well as ocular segment, which is characterized by wg head blob and en head spot expression.

Fig. 2.

Head defects in slp mutants. Scanning electron micrographs (A-C) and whole-mount embryos hybridized in situ with a wg probe (D-G) are shown. F and G show ventral views. The slp allele shown is indicated in the lower right corner of each panel. (A) Wild-type embryo. Mandibular (MD), maxillary (MX) and labial (LI) lobes are indicated. (B,C) △46G and △34B mutants miss the mandibular lobe (arrowheads). (D)and (F) Wild-type embryos. wg labels all gnathal (MD, MX, LI), thoracic and abdominal segments. Antennal stripe (AN) and head blob (HB) in the procephalon are indicated. (E,G) △46G mutants lack the mandibular segment (arrowhead). The antennal stripe is absent and the head blob staining, which is characteristic for the ocular segment, partially deleted (star in E). The intercalary spot ventral to the antennal stripe is present (not readily visible in this picture).

Fig. 2.

Head defects in slp mutants. Scanning electron micrographs (A-C) and whole-mount embryos hybridized in situ with a wg probe (D-G) are shown. F and G show ventral views. The slp allele shown is indicated in the lower right corner of each panel. (A) Wild-type embryo. Mandibular (MD), maxillary (MX) and labial (LI) lobes are indicated. (B,C) △46G and △34B mutants miss the mandibular lobe (arrowheads). (D)and (F) Wild-type embryos. wg labels all gnathal (MD, MX, LI), thoracic and abdominal segments. Antennal stripe (AN) and head blob (HB) in the procephalon are indicated. (E,G) △46G mutants lack the mandibular segment (arrowhead). The antennal stripe is absent and the head blob staining, which is characteristic for the ocular segment, partially deleted (star in E). The intercalary spot ventral to the antennal stripe is present (not readily visible in this picture).

Spatial relationship between the expression pattern of slp, ems and otd

Three genes that are required for proper head development, otd, ems and btd, are expressed in distinct regions at the anterior pole of the embryo (Finkelstein and Perrimon, 1990; Dalton et al., 1989; Walldorf and Gehring, 1992; Wimmer et al., 1993). In mutants with defects in these genes several adjacent segments in the head are missing (Finkelstein and Perrimon, 1990; Cohen and Jürgens, 1990). The functional domains defined by their phenotypes are shifted by one segment with respect to each other. However, their expression patterns are not fully corresponding to the region that is affected in the mutants and their expression domains are largely overlapping. To define how slp is expressed in relation to otd and ems we performed double labelling experiments (Fig. 3).

Fig. 3.

slp expression relative to other head-specific expression patterns. otd (A-F) and ems (G-L) mRNA was detected by whole-mount in situ hybridization (blue staining) either individually (A,G) or combined with an anti-slp1 immunostaining (B-F; H-L; orange staining). Overlap between blue (otd, ems) and orange (slp1) staining results in black colour. The overlapping domains are indicated by bars. All panels show sagittal optical sections except for D and J, which are focused on the surface of the embryo. (A) otd expression pattern in a cellularizing embryo extends roughly from 70% EL to 90% EL laterally. (B,E) Early stage 5 embryos before ventral repression of slp1. The inset in E shows a similar embryo focused on the surface. The domain of overlap spans approx. 6-7 nuclei ventrally and approx. 8 dorsally. Posterior to this domain slp1 protein is expressed in a ring extending from approx. 3 nuclei on the dorsal side to approx. 7 nuclei on the ventral side. Anteriorly, otd is expressed in a trapezoidal domain covering 7-8 nuclei dorsally and 4-6 ventrally. (C,F) Stage 5 embryo, ventral repression of slp1 completed. The arrowheads mark the region that is free of staining. (D) Stage 5 embryo, primary slp1 stripes fully established. Dorsolaterally, the slp1 head stripe is adjacent to the otd domain (not in focus), in more lateral regions there is a gap of about one cell width between the otd domain and the head stripe (HS). (G) ems expression pattern in a cellularizing embryo covering 4-5 cells from 70% EL to 75% EL dorsally and approx. 9 cells from 74% EL to 88% EL ventrally. (H,K) Embryos of the same stage as in G. Posterior to the domain of overlap slp1 is expressed in a region covering approx. 2 cells on the dorsal side and 3-4 ventrally. Anteriorly, the domain expressing exclusively slp1 spans approx. 5 cells dorsally and 1 or 2 on the ventral side. (I,L) Stage 5 embryos, ventral repression of slp1 completed. Note that the blue ems staining occupies a large part of the ventral repression domain. Ventrally, ems and slp overlap by approx. 2 cells. (J) Stage 5 embryo after full establishment of the primary slp1 stripes. The arrowhead indicates the 2 cell overlap of ems and slp1 head cap expression.

Fig. 3.

slp expression relative to other head-specific expression patterns. otd (A-F) and ems (G-L) mRNA was detected by whole-mount in situ hybridization (blue staining) either individually (A,G) or combined with an anti-slp1 immunostaining (B-F; H-L; orange staining). Overlap between blue (otd, ems) and orange (slp1) staining results in black colour. The overlapping domains are indicated by bars. All panels show sagittal optical sections except for D and J, which are focused on the surface of the embryo. (A) otd expression pattern in a cellularizing embryo extends roughly from 70% EL to 90% EL laterally. (B,E) Early stage 5 embryos before ventral repression of slp1. The inset in E shows a similar embryo focused on the surface. The domain of overlap spans approx. 6-7 nuclei ventrally and approx. 8 dorsally. Posterior to this domain slp1 protein is expressed in a ring extending from approx. 3 nuclei on the dorsal side to approx. 7 nuclei on the ventral side. Anteriorly, otd is expressed in a trapezoidal domain covering 7-8 nuclei dorsally and 4-6 ventrally. (C,F) Stage 5 embryo, ventral repression of slp1 completed. The arrowheads mark the region that is free of staining. (D) Stage 5 embryo, primary slp1 stripes fully established. Dorsolaterally, the slp1 head stripe is adjacent to the otd domain (not in focus), in more lateral regions there is a gap of about one cell width between the otd domain and the head stripe (HS). (G) ems expression pattern in a cellularizing embryo covering 4-5 cells from 70% EL to 75% EL dorsally and approx. 9 cells from 74% EL to 88% EL ventrally. (H,K) Embryos of the same stage as in G. Posterior to the domain of overlap slp1 is expressed in a region covering approx. 2 cells on the dorsal side and 3-4 ventrally. Anteriorly, the domain expressing exclusively slp1 spans approx. 5 cells dorsally and 1 or 2 on the ventral side. (I,L) Stage 5 embryos, ventral repression of slp1 completed. Note that the blue ems staining occupies a large part of the ventral repression domain. Ventrally, ems and slp overlap by approx. 2 cells. (J) Stage 5 embryo after full establishment of the primary slp1 stripes. The arrowhead indicates the 2 cell overlap of ems and slp1 head cap expression.

otd and slp1 are expressed in very similar patterns: both genes are initiated in the anterior 30% of the embryo and, subsequently, show anterior repression. The otd ring narrows considerably from dorsal to ventral (Fig. 3A). slp1 protein is detectable earlier than otd mRNA, which comes up towards the end of syncytial blastoderm. Initially, otd mRNA and slp1 protein are expressed in roughly the same region (not shown) but by the time they are expressed in the typical ring they overlap only in a central domain (Fig. 3B,E; for details see figure legend). By the time the splitting of slp1 head expression starts, a gap of about three to four cells that is free of either otd or slp1 is present ventrally (Fig. 3C,F). At cellular blastoderm when the slp1 pair rule stripes are fully established the slp1 head cap is entirely within the otd expression domain whereas the head stripe does not overlap with otd (Fig. 3D).

ems is expressed at syncytial blastoderm in a ring which is thinner dorsally and wider ventrally (Fig. 3G). ems mRNA appears considerably later than slp1 protein. By the time slp1 is fully repressed anteriorly, ems expression falls completely within the slp1 head ring (Fig. 3H,K). After ventral repression of slp1 has occurred, ems is expressed in a large part of the repression domain (Fig. 3I,L). Once the slp1 primary stripes have been established, ems expression overlaps with the slp1 head cap by about two cells and is exactly adjacent to the slp1 head stripe (Fig. 3J). The results are summarized in Fig. 4 and show that slp, otd and ems together define five distinct domains in the cephalic region.

Fig. 4.

Expression domains in the embryonic head. Data obtained from the experiments shown in Fig. 3B,E,H and K have been combined and are schematically represented. The extent of overlapping and individual expression domains was analyzed in 11 embryos and the most frequently counted width of each domain was taken for the schematic representation. The expression patterns of the genes slp1, otd and ems specify five distinct domains in the head each expressing a particular combination of head-specific genes. Three of them can be allocated to segment primordia (see discussion): The [otd, slp], [otd, slp, ems] and [slp] domains (or parts thereof) correspond to the ocular, antennal and mandibular segment.

Fig. 4.

Expression domains in the embryonic head. Data obtained from the experiments shown in Fig. 3B,E,H and K have been combined and are schematically represented. The extent of overlapping and individual expression domains was analyzed in 11 embryos and the most frequently counted width of each domain was taken for the schematic representation. The expression patterns of the genes slp1, otd and ems specify five distinct domains in the head each expressing a particular combination of head-specific genes. Three of them can be allocated to segment primordia (see discussion): The [otd, slp], [otd, slp, ems] and [slp] domains (or parts thereof) correspond to the ocular, antennal and mandibular segment.

Activation of slp by the anterior system

Proper development of head and thorax depends on the anterior system of maternal coordinate genes. The bcd morphogen is required for the activation of several genes expressed in the anterior region of the embryo, among them hb, ems, otd and, together with the terminal system, tll (Driever et al., 1989; Struhl et al., 1989; Dalton et al., 1989; Finkelstein and Perrimon, 1990; Pignoni et al., 1992). Since the slp genes are expressed in the head region of the embryo it seemed likely that bcd might be activating the genes of the slp locus. Indeed, no slp1 and slp2 head-specific expression can be detected in bcd embryos (embryos from homozygous maternal effect mutant mothers are referred to by the genotype of the mother; Fig. 5A). We conclude that bcd is required for activation of slp expression in the head. Since the first slp1 expression is detectable as early as nuclear division cycle 9, even before zygotic transcripts of hb and Kr have been reported (Knipple et al., 1985; Tautz et al., 1987; Tautz and Pfeifle, 1989), it is unlikely that the activating effect of bcd is mediated through other gap genes. Rather, the early slp1 expression suggests a direct regulation by bcd.

Fig. 5.

bcd determines the position of the slp1 and slp2 expression domains. slp1 mRNA has been detected by whole-mount in situ hybridization. All embryos shown are at the early cellular blastoderm stage (5(1)), when ventral repression occurs. (A) bcd6/bcd6 embryo that has been strongly overstained (-> background) to make sure that the head-specific expression is indeed missing. (B) bcd 1+ embryo (bcd6/+). (C) Wild-type embryo. (D) bcd 4+ embryo (P[bcd+]/P[bcd+]; +/+).

Fig. 5.

bcd determines the position of the slp1 and slp2 expression domains. slp1 mRNA has been detected by whole-mount in situ hybridization. All embryos shown are at the early cellular blastoderm stage (5(1)), when ventral repression occurs. (A) bcd6/bcd6 embryo that has been strongly overstained (-> background) to make sure that the head-specific expression is indeed missing. (B) bcd 1+ embryo (bcd6/+). (C) Wild-type embryo. (D) bcd 4+ embryo (P[bcd+]/P[bcd+]; +/+).

To test whether the position of the slp1 and slp2 rings in wild-type embryos is under the control of the bcd morphogen, we analyzed slp expression in embryos with an altered bcd concentration gradient. The bcd gradient can be modified by changing the number of bcd copies in the mother (Driever and Nüsslein-Volhard, 1988b). The slp1 (Fig. 5C) and slp2 (not shown) head rings are narrower and shifted anteriorly in bcd 1+ embryos (Fig. 5B) and broader and located more posteriorly in bcd 4+ embryos (Fig. 5D). These results suggest that the concentration of bcd is critical for the correct positioning of the head-specific expression of the slp genes. We determined the position of the anterior and posterior borders of slp expression as well as the border defined by the ventral repression domain (Table 1). The results show that all three borders are dependent on bcd concentration, suggesting that bcd is not only involved in determining the domain of activation but also in defining the extent of anterior and ventral repression domains.

Table 1.

slp responds to the bcd concentration gradient and terminal system activity

slp responds to the bcd concentration gradient and terminal system activity
slp responds to the bcd concentration gradient and terminal system activity

Repression at the anterior pole by the terminal system

After the initial establishment of slp1 expression in the entire head region the expression domain starts to retract from the anterior pole. The zygotic expression patterns of hb, otd and tll show a similar anterior repression, which depends on terminal system activity (Tautz, 1988; Driever et al., 1989; Finkelstein and Perrimon, 1990; Pignoni et al., 1992). To test whether the terminal system is responsible for anterior repression of slp, we analyzed slp expression in several mutants affecting this pathway. In tor (Fig. 6A-D) and phl embryos (not shown) slp1 is initially expressed as in wild type (Fig. 6A). Subsequently, repression of slp1 at the anterior pole starts but does not proceed to form the typical head ring (Fig. 6B). Rather, ventral repression (Fig. 6C) and splitting of the extended head expression into a head stripe and a very broad anterior cap occurs (Fig. 6D). In tsl embryos the extent of repression is intermediate if compared to tor and wild-type embryos (not shown), suggesting that the tsl alleles (tsl3/tsl4) used are not amorphic. No obvious deviation from the wild-type expression pattern could be found in tll (Fig. 6F) and hkb embryos (not shown), indicating that they are not involved in anterior repression. Taken together, these results show that anterior repression is primarily caused by the terminal system. There is residual repressing activity at the tip of the embryo, which we detect in both tor and phl mutants. This activity may be independent of the tor pathway or, alternatively, the mutants analyzed may express very low levels of tor activity.

Fig. 6.

Control of slp expression by the terminal system. Whole-mount in situ hybridization to slp1 mRNA. (A-D) Development of slp1 expression in tor1/tor4 embryos: initiation (A), residual anterior repression at the very tip of the embryo (B), ventral repression (C) and splitting (D). Arrowheads mark the anterior border of slp1 expression. (E) tor13D embryo during cellularization. The slp expression domain is expanded and shifted more posteriorly. Ventral repression occurs in a triangular domain. (F) Wild-type embryo stained for slp1 and fushi tarazu (ftz) mRNA. The arrowhead marks the sixth ftz stripe. (G) tll1/tll1 embryo stained for slp1 and ftz mRNA. The last ftz stripe is missing and the sixth stripe is shifted posteriorly (arrowhead). slp expression (cap and head stripe) is normal. (H) vas, exu embryo during cellularization. slp is expressed at both termini, in the normal domain anteriorly and an ectopic smaller domain at the posterior pole. Repression occurs at both poles.(I) vas, exu embryo during gastrulation. The slp pair rule stripes are expressed as a single block in the center of the embryo. Note that the terminal expression has split into two stripes.

Fig. 6.

Control of slp expression by the terminal system. Whole-mount in situ hybridization to slp1 mRNA. (A-D) Development of slp1 expression in tor1/tor4 embryos: initiation (A), residual anterior repression at the very tip of the embryo (B), ventral repression (C) and splitting (D). Arrowheads mark the anterior border of slp1 expression. (E) tor13D embryo during cellularization. The slp expression domain is expanded and shifted more posteriorly. Ventral repression occurs in a triangular domain. (F) Wild-type embryo stained for slp1 and fushi tarazu (ftz) mRNA. The arrowhead marks the sixth ftz stripe. (G) tll1/tll1 embryo stained for slp1 and ftz mRNA. The last ftz stripe is missing and the sixth stripe is shifted posteriorly (arrowhead). slp expression (cap and head stripe) is normal. (H) vas, exu embryo during cellularization. slp is expressed at both termini, in the normal domain anteriorly and an ectopic smaller domain at the posterior pole. Repression occurs at both poles.(I) vas, exu embryo during gastrulation. The slp pair rule stripes are expressed as a single block in the center of the embryo. Note that the terminal expression has split into two stripes.

To confirm the repressing activity of the terminal system we analyzed slp expression in a dominant gain-of-function tor allele exhibiting a phenotype opposite to the one observed in embryos lacking tor activity: the terminal regions of the embryo are expanded at the expense of the central trunk region (Strecker et al., 1989). In such mutants the tor tyrosine receptor kinase is constitutively activated throughout the embryo and thus we expected a repression of slp in regions further away from the termini. To our surprise, slp expression is not weaker but rather the head ring is broader and shifted posteriorly (Fig. 6G). The slp expression pattern in tor13D resembles the pattern we observed in embryos from by mothers carrying four copies of bcd (Table 1). A qualitatively similar result was observed in embryos carrying a dominant allele of the MAP kinase homolog rolled (rlSem; Brunner et al., 1994), a kinase acting at the bottom of the signalling cascade. However, the extent of the posterior shift is weaker (Table 1) reflecting the less severe phenotype of this mutant. In summary, the terminal system appears not only to be involved in repressing slp at the anterior pole but also in its activation. Higher levels of tor activity allow the expression of slp in regions where bcd levels are low suggesting that the anterior and terminal system also interact in determining the posterior border of the slp expression domain.

Anterior repression depends primarily on the terminal system but is not mediated by tll and hkb. The mechanism of repression may involve cooperation with the morphogen of the anterior system. In embryos with altered bcd gradients both the anterior and posterior borders of slp expression shift (Fig. 5). Thus, bcd is involved in the determination of both the domains of repression as well as activation. Since a manipulation of the activity of the terminal system also leads to altered repression and activation domains it is likely that the anterior and terminal system act in concert to regulate the expression of slp. This hypothesis is strongly supported by an analysis of slp expression in vasa exuperantia (vas,exu) double mutants (Fig. 6H,I; Schüpbach and Wieschaus, 1986). These mutants display a homogeneously low level of bcd protein along the antero-posterior axis (Driever and Nüsslein-Volhard, 1988b). If activation of slp depended only on bcd we would expect either no early slp expression (if bcd concentration is below the ‘activation’ threshold) or a uniform slp expression (if bcd level is above the threshold). We observe slp expression at both poles in vas, exu double mutants strongly suggesting that in addition to sufficiently high levels of bcd terminal system activity is involved in slp activation. After an initial expression of slp in the entire pole regions (not shown), repression occurs at both termini (Fig. 6H) and the split of the head-specific expression occurs in both the normal and ectopic expression domain (Fig. 6I). Thus, as observed for bcd, the terminal system is required for both proper activation and repression of slp.

The role of the dorsoventral system in ventral repression

The modulation of slp expression along the dorsoventral axis suggests a role of the dorsoventral patterning system in regulating slp transcription. This system encodes components of a signal transduction pathway that leads to the graded nuclear uptake of a transcription factor, the dorsal (dl) protein (Steward, 1987, 1989; Rushlow et al., 1989; Roth et al., 1989). The gradient of nuclear dl protein controls patterning by the differential regulation of a set of target genes along the dorsoventral axis. For example, high levels of dl are required for the activation of the zygotic target genes twist (twi) and snail (sna) (Thisse et al., 1987; Leptin and Grunewald, 1990). Loss-of-function alleles of the dorsal group genes lead to a dorsalized phenotype where dl is cytoplasmic along the entire dorsoventral axis. Loss-of-function alleles of cact (Schüpbach and Wieschaus, 1989) and dominant alleles of Tl and ea, on the other hand, show ventralized phenotypes accompanied by an expansion of the nuclear dl gradient towards the dorsal side (Steward, 1989; Roth et al., 1989).

Examination of slp expression in dl embryos shows that the dl gene product is required for ventral repression. Initially, the slp expression pattern in these embryos and all other dorsal group mutants tested (see Materials and Methods) develops as in wild-type embryos (Fig. 7A,B) but ventral repression does not occur (Fig. 7B). The head ring then splits into two circumferential rings (Fig. 7C). The posterior one (corresponding to the head stripe) starts to fade (Fig. 7D) and disappears completely (Fig. 7E). Subsequently the posterior half of the remaining stripe also fades and ectopic slp expression appears at the anterior tip of the embryo (Fig. 7F). These observations suggest that dl, or one of its target genes, is repressing slp in the ventral repression domain and that the dorsoventral system is involved in maintaining the head stripe and part of the head cap, two pattern elements whose expression and position depends mainly on the anterior system. Since no obvious deviation from the wild-type pattern could be detected in twi and sna embryos (not shown), dl itself may be responsible for ventral repression.

Fig. 7.

Control of slp expression by the dorsoventral system. slp1 mRNA was detected by whole-mount in situ hybridization. (A-F) Development of slp1 expression in dl1/Df(2L)TW119 embryos. No ventral repression occurs (arrowhead in B). After the splitting of the head expression the posterior stripe fades (arrowheads in C and D). Note the derepression of slp1 at the anterior tip after cellularization is completed (arrowhead in F). (G and H) Tl8/+ embryos. Only the posterior part of the pattern is detectable. The anterior half of the broad stripe in G fades (arrowhead) such that only the head stripe remains (H). (I) Embryo derived from a cactD13/cactD13 germline clone (a strong lethal cact allele).

Fig. 7.

Control of slp expression by the dorsoventral system. slp1 mRNA was detected by whole-mount in situ hybridization. (A-F) Development of slp1 expression in dl1/Df(2L)TW119 embryos. No ventral repression occurs (arrowhead in B). After the splitting of the head expression the posterior stripe fades (arrowheads in C and D). Note the derepression of slp1 at the anterior tip after cellularization is completed (arrowhead in F). (G and H) Tl8/+ embryos. Only the posterior part of the pattern is detectable. The anterior half of the broad stripe in G fades (arrowhead) such that only the head stripe remains (H). (I) Embryo derived from a cactD13/cactD13 germline clone (a strong lethal cact allele).

To characterize the role of dl in ventral repression in more detail we analyzed slp expression in mutants that show a ven-tralized phenotype. The presence of nuclear dl protein in more dorsal regions should lead to an expansion of the ventral repression domain in a dorsal direction in these mutants. We could not detect any obvious deviation from the wild-type pattern in viable, hypomorphic cact mutants and the dominant mutations Tl3, eaD3 and eaD6 whereas the ventral repression domain extends about twice as much in the dorsal direction in Tl2 embryos (not shown). In Tl8 embryos showing a complete ven-tralization and uniformly high levels of nuclear dl protein only the head stripe remains. In such embryos slp is initially expressed normally (not shown) but then the entire anterior half of the head ring disappears leaving only the circumferential part of the early slp expression pattern (Fig. 7G). Subsequently, the anterior half of the remaining stripe gets repressed as in wild type, such that only the head stripe expression remains (Fig. 7H). The same is observed in embryos derived from germline clones of lethal and presumably amorphic cact alleles (Fig. 7I). Thus, very high levels of nuclear dl are required for ventral repression. Indeed, the same observation has been reported for the activation of sna and twi by dl, which are expressed all over in Tl8 embryos but show no deviation from the wild-type pattern in Tl3 mutants (Ray et al., 1991). In the hypomorphic dorsal group mutants easter (ea3/ea4) and Tlr2 ventral repression occurs in a much smaller domain than in wild type (not shown). The graded shift of the dorsoventral extent of the ventral repression domain strongly supports the hypothesis that the dl gradient is positioning this border of repression.

Cross-regulatory interactions of slp with other gap genes

slp expression shows several characteristics that are typical of a gap gene. First, the slp genes are expressed very early during embryogenesis (especially slp1), as early (or earlier) as other gap genes start to be expressed. Second, the slp genes are controlled by three of the four maternal systems. The timing of slp expression and our analysis in mutant embryos suggest that some of these interactions may be direct rather than indirect. Thus, the slp genes would be directly involved in the interpretation of the positional information provided by the maternal coordinate genes. Since many of the gap genes show cross-regulatory interactions (e.g. Jäckle et al., 1986; Hülskamp et al., 1990; Pankratz et al., 1989) we tested the influence of zygotic gap genes active in the anterior part of the embryo on slp expression. No significant deviation from the wild-type pattern was found in embryos mutant for the gap genes giant (gt) or hb (not shown) supporting our model that no gap gene intermediaries ar involved in the activation of slp by bcd. Since only zygotic hb activity was removed, it is possible that maternally provided activity masked a possible effect of hb on slp expression. However, this seems unlikely because first, slp expression is completely absent in bcd embryos where maternal hb expression is as in wild type and second slp is expressed normally in nanos mutants (not shown) that show ectopic expression of maternal hb in the posterior half of the embryo (Hülskamp et al., 1989; Irish et al., 1989; Struhl, 1989). In Krüppel (Kr) embryos, the initiation of slp1 expression is normal (Fig. 8A). As development proceeds the slp1 expression domain broadens posteriorly (Fig. 8B,C). Thus, Kr is required for the maintenance of the posterior border of slp expression. However, Kr protein can only be detected between 33% and 60% EL (Gaul and Jäckle, 1989) and, thus, is not abutting the slp expression domain. Since other gap genes also exert their function at protein concentrations below present levels of detection (for review see Hülskamp and Tautz, 1991) and no other known zygotic mutant affects the position of the posterior slp expression boundary, it is likely that Kr is regulating slp directly.

Fig. 8.

Regulatory interaction between Krüppel and slp. slp1 expression was analyzed by whole-mount in situ hybridization. The embryo shown in A could not be unambiguously identified as a Kr embryo since all embryos at this stage look alike. (A-C) Progressive development of slp1 expression in Kr9 embryos. The dramatic derepression is observed in about 25% of the embryos of the corresponding stage.

Fig. 8.

Regulatory interaction between Krüppel and slp. slp1 expression was analyzed by whole-mount in situ hybridization. The embryo shown in A could not be unambiguously identified as a Kr embryo since all embryos at this stage look alike. (A-C) Progressive development of slp1 expression in Kr9 embryos. The dramatic derepression is observed in about 25% of the embryos of the corresponding stage.

Regulatory interactions with other head-specific genes

The head-specific gap genes otd, ems and btd play an important role in head development (Dalton et al., 1989; Finkelstein and Perrimon, 1990; Cohen and Jürgens, 1990; Walldorf and Gehring, 1992). To test whether these genes influence slp transcription we analyzed the head-specific expression of slp in otd, btd and ems mutants. No obvious deviation from the early wild-type pattern could be found in otd or btd embryos (not shown). The same has been reported for ems expression, which is not altered in either btd or otd mutants (Cohen and Jürgens, 1990) nor in slp1 slp2 double mutants (not shown).

In ems embryos the early phase of the slp expression pattern is not altered, suggesting that there are no cross-regulatory interactions of the gap gene type among these head-specific genes. The later phase, however, is strongly affected in ems mutants. Ventral repression occurs only in a small ventrolateral region and splitting does not take place (Fig. 9B). However, we detect repression at the posterior border of the head stripe (Fig. 9B,D). The gap between the head stripe and the first pair-rule stripe appears somewhat broader in ems embryos (Fig. 9B). The positioning of the head-specific expression is not altered, rather the border of repression in the head stripe domain is shifted more anteriorly by one to two cells in ems mutants. Whereas ventral repression does not occur initially in these mutants (Fig. 9D) it does later during development (Fig. 9E). In fact, ems itself is repressed ventrally at this stage (Walldorf and Gehring, 1992) in a dl dependent way (not shown) and it is likely that some other factor is taking over the function previously played by ems.

Fig. 9.

ems represses slp ventrally and is involved in slpitting. slp1 expression was analyzed by whole-mount in situ hybridization. (C,E) Ventral and ventrolateral views, respectively. (A,C) Wild-type embryos. Head staining has resolved into an anterior cap and a head stripe (half-filled triangle). The region free of transcript (star) between the head stripe and first primary stripe is about 5-6 cells wide. The head stripe spans 3-4 cells ventrally. (B) ems embryo. Ventrally repression occurs only in a small ventrolateral domain (arrowhead). The half-filled triangle marks the same position as in A. The gap between head stripe and first pair-rule stripe (star) is expanded by approx. 2 cells but the overall width between anterior expression border and first primary stripe is unchanged (16-17 cells). (D,E) ems embryos. Whereas ventral repression does not occur initially (arrow in D) it does so later when ems itself is getting repressed on the ventral side (half-filled triangle in E). The arrowhead in D marks the ventrolateral repression domain. Ventrally, the head stripe is 3-4 cells broader than in wild-type due to a lack of repression in the circumferential splitting domain.

Fig. 9.

ems represses slp ventrally and is involved in slpitting. slp1 expression was analyzed by whole-mount in situ hybridization. (C,E) Ventral and ventrolateral views, respectively. (A,C) Wild-type embryos. Head staining has resolved into an anterior cap and a head stripe (half-filled triangle). The region free of transcript (star) between the head stripe and first primary stripe is about 5-6 cells wide. The head stripe spans 3-4 cells ventrally. (B) ems embryo. Ventrally repression occurs only in a small ventrolateral domain (arrowhead). The half-filled triangle marks the same position as in A. The gap between head stripe and first pair-rule stripe (star) is expanded by approx. 2 cells but the overall width between anterior expression border and first primary stripe is unchanged (16-17 cells). (D,E) ems embryos. Whereas ventral repression does not occur initially (arrow in D) it does so later when ems itself is getting repressed on the ventral side (half-filled triangle in E). The arrowhead in D marks the ventrolateral repression domain. Ventrally, the head stripe is 3-4 cells broader than in wild-type due to a lack of repression in the circumferential splitting domain.

slp functions as a head-specific gap gene

The segmentation of the trunk region in the Drosophila embryo is governed by zygotic segmentation genes. They have been grouped into three phenotypic classes. Gap gene mutations result in deletions of several adjacent segments; pair-rule mutants lack alternate segments, and segment polarity mutants lack portions of each segment (Nüsslein-Volhard and Wieschaus, 1980). The slp locus plays an exceptional role in the segmentation hierarchy in that it shows an expression pattern reminiscent of all three classes. We have previously shown that slp functions at the pair-rule as well as the segment polarity level in the patterning of the trunk region (Grossniklaus et al., 1992; Cadigan et al., 1994b).

The head-specific expression of slp shows features of a gap gene with respect to both its spatial and temporal aspects, especially slp1, which is predominantly responsible for the head phenotype. This is supported by our analysis of the regulatory control of the locus. slp appears to respond directly to the positional cues of the anterior, terminal and dorsoventral maternal coordinate systems, independent of other target genes of these systems. slp is regulating pair-rule and segment polarity genes at the blastoderm stage. The first two paired stripes, which lie in the gnathal segment primordia, are wider than normal (Baumgartner and Noll, 1991). The maxillary stripe of en is also expanded (DiNardo and O’Farrell, 1987) and the mandibular en and wg stripes are not established. Thus, slp appears to be directly involved in the interpreta-tion of maternal positional information and its transmission to lower levels of the segmentation hierarchy supporting its status as a gap gene.

The head phenotype of slp mutants shows that slp1 is not only required for the development of the mandibular segment as reported earlier (Grossniklaus et al., 1992) but also for the development of the ocular and antennal segments. The intercalary segment, which lies inbetween, appears to be normal in slp1 mutants but en expression in this segment is dramatically expanded in slp1 slp2 double mutants (Cadigan et al., 1994a,b). Our previous analysis of terminally differentiated embryos, which relied on hypomorphic slp1 alleles, showed that the antennal segment is present in these mutants while the mandibular segment primordium is affected even in the weakest mutants (Grossniklaus et al., 1992). The absence of the antennal segment in a slp1 null mutant suggests that it is less sensitive to a reduced level of slp1 activity than the mandibular segment. In summary, our analysis of the regulation of the slp locus and the phenotypic characterization of the head defects in slp embryos strongly suggests that slp functions as a head-specific gap gene.

Segmentation of the head: a combinatorial model

The segmentation process in the trunk is relatively well under-stood and the hierarchical interactions that lead to a progressive subdivision of the embryo into smaller and smaller units have been studied extensively. Combinatorial inputs from the gap genes establish the domains of pair-rule gene expression (for reviews see Small and Levine, 1991; Pankratz and Jäckle, 1990), which in turn define the metameric patterns of segment polarity gene expression (reviewed by Akam, 1987; Ingham, 1988). Several of the segment polarity genes, for instance wg, en, gooseberry and slp are expressed in the head segments (Baker, 1987; Mahaffey et al., 1989; Baumgartner et al., 1987; Grossniklaus et al., 1992). However, their expression in the head is independent of the pair-rule genes. It has been proposed that, unlike in the trunk, gap genes in the head are directly controlling segment polarity genes in a combinatorial fashion and that the pair-rule level is omitted (Cohen and Jürgens, 1990; Finkelstein and Perrimon, 1990).

The spatial relationship between the expression patterns of head-specific gap genes has not been investigated in great detail and therefore the relationships between expression patterns and functional domains are unclear. Our analysis of ems, otd and slp1 expression by immunostaining combined with in situ hybridization (Fig. 3) provides further evidence for a combinatorial model. Taken together, our results demonstrate that five distinct domains from anterior to posterior specified by [otd], [otd and slp], [otd, slp and ems], [slp and ems,] and [slp] can be distinguished (Fig. 4), spanning together a total of five pattern elements corresponding to four head segments: the ocular (characterized by the wg head blob and en head spot), antennal, intercalary and mandibular segments. The striking similarity between functional and combinatorial domains defined by the expression patterns of slp, otd and ems supports the combinatorial model as an attractive idea for head development. A comparison between the combinatorial and functional domains (for review see Finkelstein and Perrimon, 1991; Cohen and Jürgens, 1991) allows the allocation of some of the combinatorial domains to specific head segments. Proper development of the mandibular segment requires neither ems nor otd. Therefore, the domain expressing only slp can be allocated to the mandibular segment primordium. The development of the antennal segment requires the activities of all three genes analyzed, suggesting that the domain expressing ems, otd and slp together, or a part of it, corresponds to the antennal segment primordium. Finally, wg head blob expression depends on both otd and slp activity while en head spot expression requires all three activities, indicating that the ocular segment is composed of parts of the [otd and slp] and [otd, slp and ems] combinatorial domains. We cannot correlate all combinatorial domains with actual functional requirements at this moment. However, including other head-specific genes like btd into such an analysis is likely to complete the current picture of combinatorial domains in the head. btd’s expression domain has recently been reported to range from 65 to 77% EL (Wimmer et al., 1993). Thus, the anterior expression boundary is expected to lie within the [otd, slp and ems] combinatorial domain and the posterior boundary several cells posterior to the posterior slp border. This would define a total of seven combinatorial domains, [otd], [otd and slp], [otd, slp and ems], [otd, slp, ems and btd], [slp, ems and btd], [slp and btd] and [btd], although the extent of some of these domains is not known at cellular resolution. The functional requirements for these genes suggests that the domains solely expressing otd or btd do not correspond to any head segment. The five combinatorial domains inbetween (or parts thereof) correspond to the ocular ([otd and slp] plus [otd, slp and ems]), antennal ([otd, slp, ems and btd]), intercalary ([slp, ems and btd], ems and btd are required for the development of the intercalary segment and slp1 slp2 double mutants show derepression of en in this segment; Cadigan et al, 1994a,b) and mandibular segment ([slp and btd]).

The anterior and terminal systems act in concert to regulate slp transcription

Our analysis of slp expression in mutants of the terminal and anterior system have shown that these two systems are cooperating both in the repression as well as activation of slp in the embryonic head. These findings confirm the genetic data suggesting that the proper development of the anterior part of the embryo requires both these systems, whereas all other regions are patterned solely by one of the three maternal systems determining the anteroposterior axis. slp expression in the head region shows an absolute requirement for bcd. In the absence of bcd activity the slp genes are not expressed (Fig. 5A). The dependence of the position of the slp head ring on the bcd gradient (Fig. 5) shows that the slp transcription in this domain is activated by specific concentrations of the bcd protein. Thus, like otd and ems (Dalton et al., 1989; Finkelstein and Perrimon, 1990; Walldorf and Gehring, 1992) the slp genes are two new candidates for gene X. Since both the anterior and posterior borders of slp expression shift if the bcd concentration gradient is modified, bcd is also involved in specifying the anterior domain of repression. On the other hand, we have shown that the main determinant for anterior repression is the terminal system (Fig. 6). However, if it would act independently of bcd, the anterior border of slp expression should not change when the bcd concentration is altered. Thus, bcd has not only a function as an activator of slp transcription, but also, in conjunction with the terminal system, determines the extent of the anterior repression domain. Models in which the bcd protein both activates and represses have also been proposed for the control of the gap genes Kr and gt (Gaul and Jäckle, 1989; Hülskamp et al., 1990; Eldon and Pirrotta, 1991; Kraut and Levine, 1991). In the case of slp this ‘repressor’ function is not just dependent on the local bcd concentration, i.e. medium levels of bcd acting as an activator and high levels of bcd as a repressor. In mutants for the terminal class of maternal genes bcd is unable to ‘repress’ (with the exception of the slight repression at the very tip of the embryo) although its distribution is not altered in such mutants. Thus, bcd ‘repressor’ function depends on the terminal system.

Ronchi and co-workers (1993) have suggested a model in which the terminal signal transduction system acts directly on the bcd protein. Phosphorylation of bcd is proposed to render it inactive as a transcriptional activator. These authors have shown that the appearance of phosphorylated forms of bcd takes time to develop. They suggested that this is likely to be correlated with the time the terminal signal transduction system needs to activate the signalling pathway and the effector kinases, which presumably phosporylate bcd. Consistent with this hypothesis the head-specific genes can be subdivided into two classes: the genes that are activated in the entire pole region and get repressed anteriorly (otd, hb, slp1) and the ones appearing as a distinct stripe (ems, slp2, btd). The genes of the first class are all expressed earlier than the ones of the second class. Our analysis of slp1 and slp2 expression suggests that the terminal system has not yet acted on bcd by the time slp1 appears, but the putative phosphorylation event has taken place when slp2 starts to be expressed.

Our results have shown that the terminal system and bcd act together to determine the extent of slp repression at the anterior tip. A similar observation has been reported for the transcriptional regulation of tll (Pignoni et al., 1992). In the case of tll the two systems are also cooperating to activate tll. Although slp is activated in terminal mutants we have strong evidence suggesting that the terminal system is involved in the activation of slp. Whereas the requirement for bcd is absolute, the terminal system appears to have a potentiating role. That slp is expressed at both poles in vas,exu mutants suggests that relatively low levels of bcd are sufficient for slp activation. The bcd concentration level in vas, exu embryos is roughly the same as found at the posterior border of slp expression around 70% EL or somewhat lower. Since the level of bcd is homogenous in these embryos (Driever and Nüsslein-Volhard, 1988b), slp should be expressed along the entire length of the embryo or not at all, if bcd was the only determinant of the slp transcriptional domain. We only see two domains of slp expression at the termini of the embryo suggesting that, in addition to bcd, a second determinant is involved in slp activation that is provided by the terminal system. Although bcd is absolutely required for slp expression it seems that activation of slp by bcd is only possible at the termini of the embryo. Further evidence for a role of the terminal pathway in slp activation comes from an analysis of slp expression in a dominant tor allele. In these embryos the anterior repression domain is expanded more posteriorly. The posterior border, which was thought to be defined by a minimal bcd level required for slp activation is shifted, too, although the bcd concentration gradient should be unchanged. Thus, low levels of bcd concentration, which are normally not sufficient for slp activation, can activate slp if there is sufficient activity from the terminal system.

We propose that the terminal system can turn bcd into a more potent activator if present at low levels but renders it inactive as an activator in regions with high levels of terminal activity. Thus, activation of slp by bcd would depend both on the concentration of bcd itself and the activity of the terminal system. Consistent with this model the positioning of the slp ring can be altered by changing bcd concentration levels or terminal system activity. A prediction of the model is that slp gets activated in regions of low bcd concentration that without the potentiation of the terminal system would not be sufficient for activation. Indeed, we find that the posterior border of slp expression shifts more anteriorly by an average of 2.5% EL in tor mutants. That initial slp expression is unchanged in Kr mutants but extends posteriorly once terminal activity is established gives further support for this model. The activation of slp at both poles in vas,exu mutants can be explained by the same mechanism. This proposed potentiation of bcd and its inactivation at the poles may occur by phosphorylation of bcd. Various phosphorylated forms of bcd have been detected (Driever and Nüsslein-Volhard, 1988a, 1989). We propose that a low level of phosphorylation potentiates bcd whereas additional phophorylation inactivates it (see also Ronchi et al., 1993).

Integration of positional information along the dorsoventral and anteroposterior axis: ems as a corepressor of dl

Positional cues along the dorsoventral axis are determined by the nuclear concentration of dl protein. Our analysis of slp expression in mutants of the dorsoventral system (Fig. 7) suggest that the dl gradient is determining the dorsoventral extent of the ventral repression domain. Since this repression does not occur along the complete width of the head ring but dl is nuclear and presumably active along the entire length of the embryo ventrally, there must be a second determinant that defines the anteroposterior extent of the ventral repression domain. Since the anteroposterior border of this repression domain depends on bcd concentration (Fig. 5), it is likely that bcd or one of its targets plays this role. Ventral repression is incomplete in ems mutant embryos suggesting that ems is this second factor. There is compelling evidence that dl has the inherent property to act as a transcriptional activator. However, repression by dl seems to require an additional factor or corepressor (Jiang et al., 1992, 1993; Pan and Courey, 1992; Kirow et al., 1993). Our results suggest that ems may function as a co-repressor of dl for the following reasons. First, slp is not repressed ventrally in both dl and ems mutants. Second, although the ems stripe does not show any dorsoventral modulation in dl embryos and has a uniform width of about four cells, ventral activation of ems does not depend on dl at this stage (not shown). Therefore, it is unlikely that ems acts downstream of dl. Third, repression in ems embryos fails to occur only in the domain of overlap between ems and dl. ems mRNA can be detected in a large portion of the ventral repression domain at this stage (Fig. 3I and L) but not in the small ventrolateral regions that still show repression in ems mutants (Fig. 9B and D). It appears that in the absence of ems, dl cannot function as a repressor of slp or, in other words, dl requires ems as a co-repressor in the domain of overlap. Fourth, in Tl8 embryos ems is expressed in a broad ring spanning about nine cells which shows no modulation along the dorsoventral axis (not shown). The ventral repression domain extends around the entire circumference in these embryos where nuclear dl and ems are expressed together. Finally, ventral repression of slp only starts at cellularization coinciding with the first detection of ems protein (Dalton et al., 1989) although dl protein is already nuclearly localized at nuclear division cycle 10 (Steward, 1989; Rushlow et al., 1989; Roth et al., 1989). This temporal coincidence of ventral repression and ems expression suggests that the interaction between dl and ems as repressor and co-repressor may be direct.

ems has also been identified as an activity involved in splitting of the head expression. However, slp does not get repressed in all cells that also express ems. The ems stripe is broader than the repression domain and there is a two to three cell overlap between ems and the slp head cap at this stage while the posterior border of ems expression is adjacent to slp (Fig. 3J). Therefore, it seems likely that ems requires another co-factor in this additional repression domain. We predict that the anterior border of expression (or better activity) of this additional factor will abut the slp cap expression. The existence of a second factor acting together with ems to repress slp in this circumferential repression domain is supported by the development of the slp expression pattern in wild-type embryos. First, only the anterior half of the head ring gets repressed by the combined action of ems and dl corresponding only to a part of their overlapping expression domains (Fig. 10C). In a subsequent phase, repression of the head expression occurs around the entire circumference of the embryo (splitting, Fig. 10D). The second factor cooperating with ems, which shows no modulation along the dorsoventral axis has presumably only become active after repression by ems/dl has occurred. The third effect observed in ems mutants, i.e. extended repression at the posterior border of the head stripe, must be an indirect one since ems is not expressed in these cells. We propose that a repressor of slp is derepressed in anterior direction in ems mutants by one to two cells.

Fig. 10.

Summary of the genetic control of slp head expression. Schematic representation of the head region of a Drosophila embryo. The development of slp expression and the regulating activities are illustrated. Activating [inline] and repressing [inline] activities are indicated. Formation of slp head-specific expression requires the input of three maternal systems. (A) Activation of slp shows an absolute requirement for bcd but is modulated by the tor pathway. (B) Anterior repression requires a high level of terminal system activity but also high levels of bcd. Kr is required to maintain the posterior border of slp expression. (C) Ventral repression depends on dl in conjunction with ems acting as a putative co-repressor. (D) ems is also involved in the splitting process.

Fig. 10.

Summary of the genetic control of slp head expression. Schematic representation of the head region of a Drosophila embryo. The development of slp expression and the regulating activities are illustrated. Activating [inline] and repressing [inline] activities are indicated. Formation of slp head-specific expression requires the input of three maternal systems. (A) Activation of slp shows an absolute requirement for bcd but is modulated by the tor pathway. (B) Anterior repression requires a high level of terminal system activity but also high levels of bcd. Kr is required to maintain the posterior border of slp expression. (C) Ventral repression depends on dl in conjunction with ems acting as a putative co-repressor. (D) ems is also involved in the splitting process.

The four maternal systems act largely independently of each other (Nüsslein-Volhard et al., 1987). An exception to this generalization is provided by the anterior and terminal systems that are both required to establish a normal acron (Frohnhöfer and Nüsslein-Volhard, 1986). The analysis of the slp head expression has shown that the terminal and anterior sytem cooperate in the repression as well as activation of slp. The two systems may be linked directly in that bcd itself is likely to be a target of the terminal signalling cascade. Although the anteroposterior and dorsoventral axes are initially determined independently it is clear that they have to act in concert to ensure complex gene regulation in two dimensions. A model implying two factors required together for either activation or repression can explain how positional information along the two embryonic axes can be integrated to lead to complex expression patterns showing modulations along both axes. The level of nuclear dl defines the domain of slp repression only in the dorsoventral dimension. The position of the ems head stripe however is set up by the anterior system, i.e. specific levels of bcd protein concentration. If both of these components are required together, for example as repressor and co-repressor, then the repression domain will be defined in both dimensions. A molecular dissection of the slp control region will be required to dissect further the mechanisms of cooperation between the maternal systems and their complex interplay with zygotically expressed genes in the regulation of the slp segmentation locus.

We are indebted to Christiane Nüsslein-Volhard and Kathryn V. Anderson for providing a large part of the collection of mutants which made this study possible. Thanks go also to Claude Desplan, Damian Brunner and Ernst Hafen, Trudi Schüpbach, Steve Wasserman and Mike Levine for sending us mutant fly stocks. Without the generous help of Norbert Perrimon, who provided fly stocks before publication, the germline clone experiments would not have been possible. We are particularly grateful to Uwe Walldorf, who made additional in situs on snail embryos, and to Georg Halder for his valuable help during the preparation of this manuscript. We also thank the members of the Hafen laboratory for their hospitality while using their X-ray facilities. Many thanks go to Andreas Hefti for his help with the SEM pictures and to Margrit Jäggi, Verena Grieder, Lieselotte Müller and Jim Duffy for artwork. Anette Preiss and Georg Halder made helpful suggestions and comments on the manuscript. Our thanks go also to the members of the TIFR Center in Bangalore, India, for providing microscope facilities during the final stage of this work. U. G. was a recipient of a fellowship by the ‘Stipendienfonds der Basler Chemis-chen Industrie’ and K. M. C. was a American Cancer Society and Swiss National Science Foundation fellow. This research was supported by the Swiss National Science Foundation and the Kantons Basel-Stadt and Basel-Landschaft.

Akam
,
M.
(
1987
).
The molecular basis for metameric pattern in the Drosophila embryo
.
Development
101
,
1
22
.
Ambrosio
,
L.
,
Mahowald
,
A. P.
and
Perrimon
N.
(
1989
).
Requirement of the Drosophila raf homolog for torso function
.
Nature
342
,
288
291
.
Baker
,
N. E.
(
1987
).
Molecular cloning of sequences from wingless, a segment polarity gene in Drosophila: the spatial distribution of a transcript in embryos
.
EMBO J
.
6
,
1765
1774
.
Baumgartner
,
S.
, Bopp,D.,
Burri
,
M.
and
Noll
,
M.
(
1987
).
Structure of two genes at the gooseberry locus related to the paired gene and their spatial expression during Drosophila embryogenesis
.
Genes Dev
.
1
,
1247
1267
.
Baumgartner
,
S.
and
Noll
,
M.
(
1991
).
Network of interactions among pair-rule genes regulating paired expression during primordial segmentation of Drosophila
.
Mech. Dev
.
33
,
1
18
.
Brönner
,
G.
and
Jäckle
,
H.
(
1991
).
Control and function of terminal gap gene activity in the posterior pole of Drosophila
.
Mech. Dev
.
35
,
205
212
.
Brunner
,
D.
,
Oellers
,
N.
,
Szabad
,
J.
,
Biggs III
,
W.
H.,
Zipursky
,
S. L.
and
Hafen
,
E.
(
1994
).
A gain-of-function mutation in Drosophila MAP kinase activates multiple receptor tyrosine kinase signalling pathways
.
Cell
76
,
875
888
.
Cadigan
,
K. M.
,
Grossniklaus
,
U.
and
Gehring
,
W. J.
(
1994a
).
Functional redundancy: the respective roles of the sloppy paired genes in Drosophila segmentation
.
Proc. Natl. Acad. Sci. USA
91
,
6324
6328
.
Cadigan
,
K. M.
,
Grossniklaus
,
U.
and
Gehring
,
W. J.
(
1994b
).
Localized distribution of the sloppy paired protein is required for the asymmetric expression of wingless and engrailed in Drosophila segments
.
Genes Dev
.
8
,
899
913
.
Campos-Ortega
,
J. A.
and
Hartenstein
,
V.
(
1985
).
The Embryonic Development of Drosophila melanogaster
.
Berlin: Springer Verlag
.
Casanova
,
J.
and
Struhl
,
G.
(
1989
).
Localized surface activity of torso, a receptor tyrosine kinase, specifies terminal body pattern in Drosophila
.
Genes Dev
.
3
,
2025
2038
.
Casanova
,
J.
(
1990
).
Pattern formation under the control of the terminal system in the Drosophila embryo
.
Development
110
,
621
628
.
Chou
,
T.-B.
and
Perrimon
,
N.
(
1992
).
Use of a yeast site-specific recombinase to produce female germline chimeras in Drosophila
.
Genetics
131
,
643
653
.
Chou
,
T.-B.
,
Noll
,
E.
and
Perrimon
,
N.
(
1993
).
Autosomal P[ovoD1] dominant female sterile insertions in Drosophila and their use in generating germline chimeras
.
Development
119
,
1359
1369
.
Clark
,
K. L.
,
Halay
,
E. D.
,
Lai
,
E.
and
Burley
,
S. K.
(
1993
).
Co-crystal structure of the HNF-3/fork head DNA-binding motif resembles histone H5
.
Nature
364
,
412
420
.
Clevidence
,
D. E.
,
Overdier
,
D. G.
,
Tao
,
W.
,
Qian
,
X.
,
Pani
,
L.
,
Lai
,
E.
and
Costa
,
R. H.
(
1993
).
Identification of nine tissue-specific transcription factors of the hepatocyte nuclear factor 3 / forkhead DNA-binding domain family
.
Proc. Natl. Acad. Sci. USA
90
,
3948
3952
.
Cohen
,
S.
and
Jürgens
,
G.
(
1990
).
Mediation of Drosophila head development by gap like segmentation genes
.
Nature
346
,
482
485
.
Cohen
,
S.
and
Jürgens
,
G.
(
1991
).
Drosophila headlines. Trends Genet
.
7
,
267
272
.
Dalton
,
D.
,
Chadwick
,
R.
and
McGinnis
,
W.
(
1989
).
Expression and embryonic function of empty spiracles: a Drosophila homoeobox gene with two patterning functions on the anterior posterior axis of the embryo
.
Genes Dev
.
3
,
1940
1956
.
DiNardo
,
S.
and
O’Farrell
,
P. H.
(
1987
).
Establishment and refinement of segmental pattern in the Drosophila embryo: spatial control of engrailed expression by pair-rule genes
.
Genes Dev
.
1
,
1212
1225
.
Driever
,
W.
and
Nüsslein-Volhard
,
C.
(
1988a
).
A gradient of bicoid protein in Drosophila embryos
.
Cell
54
,
83
93
.
Driever
,
W.
and
Nüsslein-Volhard
,
C.
(
1988b
).
The bicoid protein determines position in the Drosophila embryo in a concentration-dependent manner
.
Cell
54
,
95
104
.
Driever
,
W.
and
Nüsslein-Volhard
,
C.
(
1989
).
The bicoid protein is a positive regulator of hunchback transcription in the early Drosophila embryo
.
Nature
337
,
138
143
.
Driever
,
W.
,
Thoma
,
G.
and
Nüsslein-Volhard
,
C.
(
1989
).
Determination of spatial domains of zygotic gene expression in the Drosophila embryo by the affinity of binding sites for the bicoid morphogen
.
Nature
340
,
363
367
.
Eldon
,
E. D.
and
Pirrotta
,
V.
(
1991
).
Interactions of the Drosophila gap gene giant with maternal and zygotic pattern forming genes
.
Development
111
,
367
378
.
Finkelstein
,
R.
and
Perrimon
,
N.
(
1990
).
The orthodenticle gene is regulated by bicoid and torso and specifies Drosophila head development
.
Nature
346
,
485
488
.
Finkelstein
,
R.
,
Smouse
,
D.
,
Capaci
,
T. M.
,
Spradling
,
A. C.
and
Perrimon
,
N.
(
1990
).
The orthodenticle gene encodes a novel homeodomain protein involved in the development of the Drosophila nervous system and ocellar visual structures
.
Genes Dev
.
4
,
1516
1527
.
Finkelstein
,
R.
and
Perrimon
,
N.
(
1991
).
The molecular genetics of head development in Drosophila melanogaster
.
Development
112
,
899
912
.
Fjose
,
A.
,
McGinnis
,
W.
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
.
Frasch
,
M.
and
Levine
,
M.
(
1987
).
Complementary patterns of even-skipped and fushi tarazu expression involve their differential regulation by a common set of segmentation genes in Drosophila
.
Genes Dev
.
1
,
981
995
.
Frohnhöfer
,
H. G.
and
Nüsslein-Volhard
,
C.
(
1986
).
Organization of anterior pattern in the Drosophila embryo by the maternal gene bicoid
.
Nature
324
,
120
125
.
Gaul
,
U.
and
Jäckle
,
H.
(
1989
).
Analysis of maternal effect mutant combinations elucidates regulation and function of the overlap of hunchback and Krüppel gene expression in the Drosophila blastoderm embryo
.
Development
107
,
651
662
.
Geisler
,
R.
,
Bergmann
,
A.
,
Hiromi
,
Y.
and
Nüsslein-Volhard
,
C.
(
1992
).
cactus, a gene involved in dorso-ventral pattern formation of Drosophila, is related to the I kappa B gene family of vertebrates
.
Cell
71
,
613
621
.
Ghosh
,
S.
,
Gifford
,
A. M.
,
Riviere
,
L. R.
,
Tempst
,
P.
,
Nolan
,
G. P.
and
Baltimore
,
D.
(
1990
).
Cloning of the p50 DNA binding subunit of NF-κB: homology to rel and dorsal
.
Cell
62
,
1019
1029
.
Grossniklaus
,
U.
,
Pearson
,
R. K.
and
Gehring
,
W. J.
(
1992
).
The Drosophila sloppy paired locus encodes two proteins involved in segmentation that show homology to mammalian transcription factors
.
Genes Dev
.
6
,
1030
1051
.
Häcker
,
U.
,
Grossniklaus
,
U.
,
Gehring
,
W. J.
and
Jäckle
,
H.
(
1992
).
Developmentally regulated Drosophila gene family encoding the fork head domain
.
Proc. Natl. Acad. Sci. USA
89
,
8754
8758
.
Hülskamp
,
M.
,
Schröder
,
C.
,
Pfeifle
,
C.
,
Jäckle
,
H.
and
Tautz
,
T.
(
1989
).
Posterior segmentation of the Drosophila embryo in the absence of a maternal posterior organizer gene
.
Nature
338
,
629
635
.
Hülskamp
,
M.
,
Pfeifle
,
C.
and
Tautz
,
D.
(
1990
).
A morphogenetic gradient of hunchback protein organizes the expression of the gap genes Krüppel and knirps in the early Drosophila embryo
.
Nature
346
,
577
580
.
Hülskamp
,
M.
and
Tautz
,
D.
(
1991
).
Gap genes and gradients - the logic behind the gaps
.
Bioessays
13
,
261
268
.
Ingham
,
P. W.
(
1988
).
The molecular genetics of embryonic pattern formation in Drosophila
.
Nature
335
,
25
34
.
Ip
,
Y. T.
,
Kraut
,
R.
,
Levine
,
M.
and
Rushlow
,
C.
(
1991
).
The dorsal morphogen is a sequence-specific DNA binding protein that interacts with a long-range repression element in Drosophila
.
Cell
64
,
603
613
.
Ip
,
Y. T.
,
Park
,
R. E.
,
Kosman
,
D.
,
Yazdanbakhsh
,
K.
and
Levine
,
M.
(
1992a
).
dorsal-twist interactions establish snail expression in the presumptive mesoderm of the Drosophila embryo
.
Genes Dev
.
6
,
1518
1530
.
Ip
,
Y. T.
,
Park
,
R. E.
,
Kosman
,
D.
,
Bier
,
E.
and
Levine
,
M.
(
1992b
).
The dorsal gradient morphogen regulates stripes of rhomboid expression in the presumptive neuroectoderm of the Drosophila embryo
.
Genes Dev
.
6
,
1728
1739
.
Irish
,
V.
,
Lehmann
,
R.
and
Akam
,
M.
(
1989
).
The Drosophila posterior-group gene nanos functions by repressing hunchback activity
.
Nature
338
,
646
648
.
Jäckle
,
H.
,
Tautz
,
D.
,
Schuh
,
R.
,
Seifert
,
E.
and
Lehmann
,
R.
(
1986
).
Cross-regulatory interactions among the gap genes of Drosophila
.
Nature
324
,
668
670
.
Jiang
,
J.
,
Hoey
,
T.
and
Levine
,
M.
(
1991
).
Autoregulation of a segmentation gene in Drosophila, Combinatorial interaction of the even-skipped homeo box protein with a distal enhancer element
.
Genes Dev
.
5
,
265
277
.
Jiang
,
J.
,
Cai
,
H.
,
Zhou
,
Q.
and
Levine
,
M.
(
1993
).
Conversion of a dorsal-dependent silencer into an enhancer, evidence for dorsal corepressors
.
EMBO J
.
12
,
3201
3209
.
Jiang
,
J.
and
Levine
,
M.
(
1993
).
Binding affinities and cooperative interactions with bHLH activators delimit threshold responses to the dorsal gradient morphogen
.
Cell
72
,
741
752
.
Jiang
,
J.
,
Rushlow
,
C. A.
,
Zhou
,
Q.
,
Small
,
S.
and
Levine
,
M.
(
1992
).
Individual dorsal morphogen binding sites mediate activation and repression in the Drosophila embryo
.
EMBO J
.
11
,
3147
3154
.
Jürgens
,
G.
,
Lehmann
,
R.
,
Schardin
,
M.
and
Nüsslein-Volhard
,
C.
(
1986
).
Segmental organisation of the head in the embryo of Drosophila melanogaster
.
Wilhelm Roux’s Arch. Dev. Biol
.
195
,
359
377
.
Kidd
,
S.
(
1992
).
Characterization of the Drosophila cactus locus and analysis of interactions between cactus and dorsal proteins
.
Cell
71
,
623
635
.
Kirow
,
N.
,
Zhelnin
,
L.
,
Shah
,
J.
and
Rushlow
,
C.
(
1993
).
Conversion of a silencer into an enhancer, evidence for a co-repressor in dorsal-mediated repression in Drosophila
.
EMBO J
.
12
,
3193
3199
.
Klingler
,
M.
,
Erdélyi
,
M.
,
Szabad
,
J.
and
Nüsslein-Volhard
,
C.
(
1988
).
Function of torso in determining the terminal anlagen of the Drosophila embryo
.
Nature
335
,
275
277
.
Knipple
,
D. C.
,
Seifert
,
E.
, Rosenberg,U. B.,
Preiss
,
A.
and
Jäckle
,
H.
(
1985
).
Spatial and temporal patterns of Krüppel gene expression in early Drosophila embryos
.
Nature
317
,
40
44
.
Kraut
,
R.
and
Levine
,
M.
(
1991
).
Spatial regulation of the gap gene giant during Drosophila development
.
Development
111
,
601
609
.
Lawrence
,
P. A.
and
Johnston
,
P.
(
1989
).
Pattern formation in the Drosophila embryo: allocation of cells to parasegments by even-skipped and fushi tarazu
.
Development
105
,
761
767
.
Lehmann
,
R.
and
Nüsslein-Volhard
,
C.
(
1987
).
hunchback, a gene required for segmentation of an anterior and posterior region of the Drosophila embryo
.
Dev. Biol
.
119
,
402
417
.
Leptin
,
M.
and
Grunewald
,
B.
(
1990
).
Cell shape changes during gastrulation in Drosophila
.
Development
110
,
73
84
.
Lindsley
D. L.
and
Zimm
,
G. G.
(
1992
).
The genome of Drosophila melanogaster
.
San Diego, California, USA
:
Academic Press Inc
.
Lu
,
X.
,
Chou
,
T.-B.
,
Williams
,
N. G.
,
Roberts
,
T.
and
Perrimon
,
N.
(
1993
).
Control of cell fate determination by p21ras/Ras1, an essential component of torso signalling in Drosophila
.
Genes Dev
.
7
,
621
632
.
Mahaffey
,
J. W.
,
Diederich
,
R. J.
and
Kaufman
,
T. C.
(
1989
).
Novel patterns of homeotic protein accumulation in the head of the Drosophila embryo
.
Development
105
,
167
174
.
Manoukian
,
A. S.
and
Krause
,
H. M.
(
1992
).
Concentration-depedent activities of the even-skipped protein in Drosophila embryos
.
Genes Dev
.
6
,
1740
1751
.
Manseau
,
L. J.
and
Schüpbach
,
T.
(
1989
).
cappuccino and spire, two unique maternal-effect loci required for both the anteroposterior and dorsoventral patterns of the Drosophila embryo
.
Genes Dev
.
3
,
1437
1452
.
Martin
,
J.-R.
,
Raibaud
,
A.
and
Ollo
,
R.
(
1994
).
Terminal pattern elements in Drosophila embryo induced by the torso-like protein
.
Nature
367
,
741
745
.
Morisato
,
D.
and
Anderson
,
K. V.
(
1994
).
The spatzle gene encodes a component of the extracellular signalling pathway establishing the dorsal-ventral pattern of the Drosophila embryo
.
Cell
76
,
677
688
.
Nishida
,
Y.
,
Hata
,
M.
,
Ayaki
,
T.
,
Ryo
,
H.
,
Yamagata
,
M.
,
Shimizu
,
K.
and
Nishizuka
,
Y.
(
1988
).
Proliferation of both somatic and germ cells is affected in the Drosophila mutants of raf proto-oncogene
.
EMBO J
.
7
,
775
781
.
Nüsslein-Volhard
,
C.
and
Wieschaus
,
E.
(
1980
).
Mutations affecting segment number and polarity in Drosophila
.
Nature
287
,
795
801
.
Nüsslein-Volhard
,
C.
,
Frohnhöfer
,
H. G.
and
Lehmann
,
R.
(
1987
).
Determination of anteroposterior polarity in Drosophila
.
Science
238
,
1675
1681
.
Nüsslein-Volhard
,
C.
(
1991
).
Determination of the embryonic axes of Drosophila
.
Development
Supplement
1
,
1
10
.
Pan
,
D.
,
Huang
,
J.-D.
and
Courey
,
A. J.
(
1991
).
Functional analysis of the Drosophila twist promoter reveals a dorsal-binding ventral activator region
.
Genes Dev
.
5
,
1892
1901
.
Pan
,
D.
and
Courey
,
A. J.
(
1992
).
The same dorsal binding site mediates both activation and repression in a context-dependent manner
.
EMBO J
.
11
,
1837
1842
.
Pankratz
,
M.
,
Hoch
,
M.
,
Seifert
,
E.
and
Jäckle
,
H.
(
1989
).
Krüppel requirement for knirps enhancement reflects overlapping gap gene activities in the Drosophila embryo
.
Nature
341
,
337
340
.
Pankratz
,
M. J.
and
Jäckle
,
H.
(
1990
).
Making stripes in the Drosophila embryo
.
Trends Genet
.
6
,
287
292
.
Perrimon
,
N.
(
1984
).
Clonal analysis of dominant female sterile, germline-dependent mutations in Drosophila melanogaster
.
Genetics
108
,
927
939
.
Perrimon
,
N.
,
Mohler
,
J. D.
,
Engstrom
,
L.
and
Mahowald
,
A. P.
(
1986
).
X-linked female sterile loci in Drosophila melanogaster
.
Genetics
113
,
695
712
.
Perrimon
,
N.
(
1993
).
The torso receptor protein-tyrosine kinase signaling pathway: An endless story
.
Cell
74
,
219
222
.
Pignoni
,
F.
,
Baldarelli
,
R. M.
,
Steingrimsson
,
E.
,
Diaz
,
R. J.
,
Patapuotian
,
A.
,
Merriam
,
J. R.
and
Lengyel
,
J. A.
(
1990
).
The Drosophila gene tailless is expressed at the embryonic termini and is a member of the steroid receptor superfamily
.
Cell
62
,
151
163
.
Pignoni
,
F.
,
Steingrimsson
,
E.
and
Lengyel
,
J. A.
(
1992
).
bicoid and the terminal system activate tailless expression in the early Drosophila embryo
.
Development
115
,
239
251
.
Ray
,
P. R.
,
Arora
,
K.
,
Nüsslein-Vollhard
,
C.
and
Gelbart
,
W. M.
(
1991
).
The control of cell fate along the dorsal-ventral axis of the Drosophila embryo
.
Development
113
,
35
54
.
Rijsewijk
,
F.
,
Schueremann
,
M.
,
Wagenaar
,
E.
,
Parren
,
P.
,
Weigel
,
D.
and
Nusse
,
R.
(
1987
).
The Drosophila homolog of the mouse mammary oncogene int-1 is identical to the segment polarity gene wingless
.
Cell
50
,
649
657
.
Ronchi
,
E.
,
Treisman
,
J.
,
Dostani
,
N.
,
Struhl
,
G.
and
Desplan
,
C.
(
1993
).
Down-regulation of the Drosophila morphogen bicoid by the torso receptor-mediated signal transduction cascade
.
Cell
74
,
1
9
.
Roth
,
S.
,
Hiromi
,
Y.
,
Godt
,
T.
and
Nusslein-Volhard
,
C.
(
1991
).
cactus, a maternal gene required for proper formation of the dorsoventral morphogen gradient in Drosophila embryos
.
Development
112
,
317
388
.
Roth
,
S.
,
Stein
,
D.
and
Nüsslein-Volhard
,
C.
(
1989
).
A gradient of nuclear localization of the dorsal protein determines dorsoventral pattern in the Drosophila embryo
.
Cell
59
,
1189
1202
.
Rushlow
,
C.
,
Frasch
,
M.
,
Doyle
,
H.
and
Levine
,
M.
(
1987
).
Maternal regulation of zernküllt, a homeobox gene controlling differentiation of dorsal tissues in Drosophila
.
Nature
330
,
583
586
.
Rushlow
,
C. A.
,
Han
,
K.
,
Manley
,
J. L.
and
Levine
,
M.
(
1989
).
The graded distribution of the dorsal morphogen is initiated by selective nuclear transport in Drosophila
.
Cell
59
,
1165
1177
.
Savant-Bhonsale
,
S.
and
Montell
,
D. J.
(
1993
).
torso-like encodes the localized determinant of Drosophila terminal pattern formation
.
Genes Dev
.
7
,
2548
2555
.
Schmitt-Ott
,
U.
and
Technau
,
G. M.
(
1992
).
Expression of en and wg in the embryonic head and brain of Drosophila indicates a refolded band of seven segment remnants
.
Development
116
,
111
125
.
Schüpbach
,
T.
and
Wieschaus
,
E.
(
1986
).
Maternal-effect mutations altering the anterior-posterior pattern of the Drosophila embryo
.
Wilhelm Roux’s Arch. Dev. Biol
.
195
,
302
317
.
Schüpbach
,
T.
and
Wieschaus
,
E.
(
1989
).
Female sterile mutations on the second chromosome of Drosophila melanogaster. I. Maternal effect mutations
.
Genetics
121
,
101
117
.
Small
,
S.
and
Levine
,
M.
(
1991
).
The initiation of pair-rule stripes in the Drosophila blastoderm
.
Curr. Op. Genet. Dev
.
1
,
255
260
.
Sprenger
,
F.
,
Stevens
,
L. M.
and
Nüsslein-Volhard
,
C.
(
1989
).
The Drosophila gene torso encodes a putative receptor tyrosine kinase
.
Nature
338
,
478
483
.
St.
Johnston
, R. D. and
Gelbart
,
W. M.
(
1987
).
decapentaplegic transcripts are localized along the dorso-ventral axis of the Drosphila embryo
.
EMBO J
.
6
,
2785
2791
.
St.
Johnston
, D. and
Nüsslein-Volhard
,
C.
(
1992
).
The origin of pattern and polarity in the Drosophila embryo
.
Cell
68
,
201
219
.
Stein
,
D.
and
Stevens
,
L. M.
(
1991
).
Establishment of dorsal-ventral and terminal pattern in the Drosophila embryo
.
Curr. Op. Genet. Dev
.
1
,
247
254
.
Stevens
,
L. M.
,
Frohnhöffer
,
H. G.
,
Klingler
,
M.
and
Nüsslein-Volhard
,
C.
(
1990
).
Localized requirement for torso-like expression in follicle cells for development of the terminal anlagen of the Drosophila embryo
.
Nature
346
,
660
663
.
Steward
,
R.
(
1987
).
Dorsal, an embryonic polarity gene in Drosophila, is homologous to the vertebrate proto-oncogene, c-rel
.
Science
238
,
692
694
.
Steward
,
R.
(
1989
).
Relocalization of the dorsal protein from the cytoplasm to the nucleus correlates with its function
.
Cell
59
,
1179
1188
.
Strecker
,
T. R.
,
Halsell
,
S. R.
,
Fischer
,
W. W.
and
Lipshitz
,
H. D.
(
1989
).
Reciprocal effect of hyper- and hypoactivity mutations in the Drosophila pattern gene torso
.
Science
243
,
1062
1066
.
Strecker
,
T. R.
,
Kongsuwan
,
K.
,
Lengyel
,
J. A.
and
Merriam
,
J. R.
(
1986
).
The zygotic mutant tailless affects the anterior and posterior ectodermal regions of the Drosophila embryo
.
Dev. Biol
.
11
,
64
76
.
Struhl
,
G.
,
Struhl
,
K.
and
Macdonald
,
P. M.
(
1989
).
The gradient morphogen bicoid is a concentration-dependent transcriptional activator
.
Cell
57
,
1259
1273
.
Struhl
,
G.
(
1989
).
Differing strategies for organizing anterior and posterior body pattern in Drosophila embryos
.
Nature
338
,
741
744
.
Tautz
,
D.
,
Lehmann
,
R.
,
Schnürch
,
H.
,
Schuh
,
R.
,
Seifert
,
E.
,
Kienlin
,
A.
,
Jones
,
K.
and
Jäckle
,
H.
(
1987
).
Finger protein of novel structure encoded by hunchback, a second member of the gap class of Drosophila segmentation genes
.
Nature
327
,
383
389
.
Tautz
,
D.
(
1988
).
Regulation of the Drosophila segmentation gene hunchback by two maternal morphogenetic centres
.
Nature
332
,
281
284
.
Tautz
,
D.
and
Pfeifle
,
C.
(
1989
).
A non-radioactive in situ hybridization method for the localization of specific RNAs in Drosophila embryos reveals translational control of the segmentation gene hunchback
.
Chromosoma
98
,
81
85
.
Thisse
,
B.
,
Stoetzel
,
C.
,
El Messal
,
M.
and
Perrin-Schmidt
,
F.
(
1987
).
Genes of the Drosophila maternal dorsal group control the specific expression of the zygotic gene twist in presumptive mesodermal cells
.
Genes Dev
.
1
,
709
715
.
Thisse
,
C.
,
Perrin-Schmitt
,
F.
,
Stoetzel
,
C.
and
Thisse
,
B.
(
1991
).
Sequence-specific transactivation of the Drosophila twist gene by the dorsal gene product
.
Cell
65
,
1191
1201
.
Tsuda
,
L.
,
Inoue
,
Y. H.
,
Yoo
,
M. A.
,
Mizuno
,
M.
,
Hata
,
M.
,
Lim
,
Y. M.
,
Adachi
,
Y. T.
,
Ryo
,
H.
,
Masamune
,
Y.
and
Nishida
,
Y.
(
1993
).
A protein kinase similar to MAP kinase activator acts downstream of the raf kinase in Drosophila
.
Cell
72
,
407
414
.
Turner
,
F. R.
and
Mahowald
,
A. P.
(
1979
).
Scanning electron microscopy of Drosophila melanogaster embryogenesis. III. Formation of the head and caudal segments
.
Dev. Biol
.
86
,
96
109
.
Walldorf
,
U.
and
Gehring
,
W. J.
(
1992
).
Empty spiracles, a gap gene containing a homeobox involved in Drosophila head development. EMBO J
.
11
,
2247
2259
.
Weigel
,
D.
and
Jäckle
,
H.
(
1990
).
Fork head, a new eukaryotic DNA binding motif?
Cell
63
,
455
456
.
Wimmer
,
E. A.
,
Jäckle
,
H.
,
Pfeifle
,
C.
and
Cohen
,
S. M.
(
1993
).
A Drosophila homologue of human Sp1 is a head-specific segmentation gene
.
Nature
366
,
690
694
.