The terminal portions of the Drosophila body pattern are specified by the localized activity of the receptor tyrosine kinase Torso (Tor) at each pole of the early embryo. Tor activity elicits the transcription of two ‘gap’ genes, tailless (tll) and huckebein (hkb), in overlapping but distinct domains by stimulating the Ras signal transduction pathway. Here, we show that quantitative variations in the level of Ras activity can specify qualitatively distinct transcriptional and morphological responses. Low levels of Ras activity at the posterior pole direct tll but not hkb transcription; higher levels drive transcription of both genes. Correspondingly, low levels of Ras activity specify a limited subset of posterior terminal structures, whereas higher levels specify a larger subset. However, we also show that the response to Ras activity is not uniform along the body. Instead, levels of Ras activity which suffice to drive tll and hkb transcription at the posterior pole fail to drive their expression in more central portions of the body, apparently due to repression by other gap gene products. We conclude that tll and hkb transcription, as well as the terminal structures, are specified by two inputs: a gradient of Ras activity which emanates from the pole, and the opposing influence of more centrally deployed gap genes which repress the response to Ras.

Receptor tyrosine kinases are involved in transducing a variety of extracellular signals that elicit diverse cellular responses. Yet, transduction of these signals generally depends on a common intracellular pathway involving the small G-protein Ras, and the serine-threonine kinases Raf, MEK and MAPK (Marshall, 1994; McCormick, 1994; Marshall, 1995). Moreover, activity of this pathway alone is often sufficient to transduce the signal and lead to the expected cellular response (Perrimon, 1994; Kayne and Sternberg, 1995; Marshall, 1995). Thus, a central question is how activation of the Ras pathway can elicit so many distinct outputs, either in different cell types or through the exposure of equivalent cells to different kinds or amounts of ligand.

Specificity could be achieved in several ways. For example, cells of different type might express different constellations of transcription factors and hence be predisposed to respond to Ras activity in different ways (Brunner et al., 1994; Cowley et al., 1994). Alternatively, cells of the same type might express more than one receptor tryosine kinase and the activation of each of these receptors could produce a different output by engaging additional signal transduction pathways (Kazlauskas, 1994). Here, we consider a third way, namely that Ras activity may be continuously varied like a rheostat in response to different levels of a ligand and in this way provide a series of thresholds that elicit distinct outputs. Such a mechanism is of particular interest in developing tissues because some receptor tyrosine kinases are thought to mediate the patterned responses of cells to polarized or graded distributions of a given ligand (Casanova and Struhl, 1989; Sprenger and Nüsslein-Volhard, 1992; Casanova and Struhl, 1993; Katz et al., 1995; Schweitzer et al., 1995).

One instance of such a patterning phenomenon is the control of terminal body patterning in Drosophila embryos, which depends on the receptor tyrosine kinase Torso (Tor) (Schüpbach and Wieschaus, 1986; Nüsslein-Volhard et al., 1987; Klingler et al., 1988; Casanova and Struhl, 1989; Sprenger et al., 1989). Although Tor is expressed uniformly along the surface of the early embryo, it is activated only at the poles, apparently in response to the localized activity of an extracellular ligand, possibly the protein Trunk (Casanova and Struhl, 1989; Sprenger and Nüsslein-Volhard, 1992; Casanova and Struhl, 1993; Casanova et al., 1995). Activation of the Tor receptor then initiates a localized cascade of sequential Ras, Raf, MEK and MAPK activity within the embryo leading to the zygotic expression of two genes, tailless (tll) and huckebein (hkb), in overlapping but distinct domains at the poles (Pignoni et al., 1992; Lu et al., 1993a; Tsuda et al., 1993; Brunner et al., 1994; Casanova et al., 1994; reviewed by Lu et al., 1993b; Duffy and Perrimon, 1994). At least at the posterior pole, most if not all aspects of terminal patterning appear to be specified by the Tll and Hkb proteins which function as transcription factors to regulate the expression of a number of other target genes (Pignoni et al., 1990; Weigel et al., 1990; Bronner and Jâckle, 1991). Experiments in which the level of Tor activity is controlled by temperature, rather than by the distribution of its ligand, have shown that high levels of Tor activity specify the more terminal structures (such as the anal pad and anal tuft at the posterior end), whereas lower levels specify less terminal structures (such as derivatives of abdominal segments A7 and A8; Casanova and Struhl, 1989). Thus, the localized activity of the Tor receptor might organize tll and hkb expression and thereby specify terminal body pattern by generating a Ras activity gradient (Casanova and Struhl, 1989; Sprenger et al., 1989; Furriols et al., 1996).

In this paper, we demonstrate that different levels of constitutive Ras activity can elicit distinct molecular outputs - the transcription of either tll alone, or tll and hkb together - as well as the formation of distinct portions of terminal body pattern. However, we also provide evidence that the response to a given level of Ras activity is not absolute but depends on cellular context, in particular on the presence or absence of other gap gene products. These findings indicate that cell pattern can be organized by mechanisms which depend on variations in both the level of Ras activity and in the responsiveness of cells to this activity.

Composition of transgenes for ubiquitous expression of RasV12 in early embryos

To generate early embryos in which the constitutively active RasV12 protein is expressed ubiquitously, we constructed two transgenes in which the RasV12 coding sequence was placed under the control of the promoter of either the ribosomal protein 49 (rp49) gene (O’Connell and Rosbash, 1984; Kongsuwan et al., 1985) or the Tubulinal (Tubα1) gene (Theurkauf et al., 1986). The rp49 promoter fragment is an approx. 2 kb segment of DNA begining at a Pst site at the 5’ end and extending to the ATG at the 3′ end which has been mutated to GGTACC (a Kpnl site; D. Kalderon, personal communication). The Tuba l fragment has been described previously (Basler and Struhl, 1994). The rp49 and Tuba l promoters direct low or moderate levels of transcription respectively, in most cells, as determined by examining the expression of rp49-lacZ and Tubα1-lacZ transgenes (G.S., unpublished findings). Because even low level constitutive expression of the RasV12 coding sequence might be expected to be lethal, these transgenes were rendered conditional by inserting a Flp-out cassette between the promoter and coding sequence to terminate transcription (as in Struhl and Basler, 1993).

rp49>w+>rasV12 transgene

A genomic fragment containing the rp49 promoter was placed upstream of the rasV12 coding sequence (Fortini et al., 1992) followed by the 3′ UTR of the hsp70 gene (Struhl and Basler, 1993), and the resulting rp49-rasV12-hsp70 3’UTR gene inserted into a derivative of the Carnegie 20 transformation vector (Rubin and Spradling, 1982) lacking the ry+ rescuing marker. A >w+> Flp-out cassette was then constructed by introducing the w+ minigene derived from the Casper transformation vector (Pirrotta, 1988) between two minimal FRTs in the J33 plasmid (Struhl and Basler, 1993) and the cassette inserted between the rp49 and rasV12 sequences to create the final rp49>w+>rasV12 transgene.

Tubα1>w+>rasV12

A similar strategy was followed to that described for the rp49>w+>rasV12 transgene except that a genomic fragment containing the Tubα1 promoter (Basler and Struhl, 1994) was used in place of the rp49 promoter fragment.

Genetics

Generating 1X and 2X rp49>rasV12 embryos

A balanced stock of the following genotype was generated by standard crosses: y w; torRXhsp70-flp.2/CyO; rp49>w+>rasV12/TM2. Late larvae and early pupae from this stock were heat shocked at 37°C for 60 minutes to excise the >w+>Flp-out cassette from most of the resident transgenes (Struhl and Basler, 1993) and adult torRX/torRX females carrying either one or two copies of the transgene were selected based on the presence or absence of the TM2 balancer chromosome. These females were crossed to wild-type males and allowed to lay eggs for 3-4 days before embryos were collected for analysis (to ensure that the mutant embryos derive exclusively from mutant female germ cells, rather than from mosaic nurse cell/oocyte complexes).

Generating 1X rp49>rasV12bcd osk tsl embryos

Females of the genotype y w hsp70-flp.1; bcdE1osk166tsl691/TM2 were crossed to males of the genotype y w; rp49>w+>rasV12; bcdE1osk166tsl691 to generate y w/y w hsp70-flp.1; rp49>w+>rasV12/+; bcdE1osk166tsl691/ bcdE1osk166tsl691 females which were treated as described above. To obtain bcd osk or bcd osk tsl females carrying a single copy of the Tuba1>rasV12 transgene, y w hsp70-flp.1; bcdE1osk166tsl691, Tuba1>w+>rasV12/TM2 females were crossed to either bcdE1osk166 or bcdE1osk166tsl691 males.

In situ analysis

RNA in situs were performed as described by Jiang et al. (1991) with the following modifications. For double labelling experiments, one RNA probe was labelled with dioxigenenin-conjugated UTP while the other probe was labelled with flourescein-conjugated UTP. Both probes were hybridized at the same time and then detected sequentially using blue and red alkaline phosphate substrate reactions (as described by Strahle et al., 1994). For single labelling experiments, probes were labelled with dioxigenenin-conjugated UTP. Plasmids containing tll (Pignoni et al., 1990) and hkb (Bronner and Jâckle, 1991) cDNAs were a gift from J. Casanova (Casanova et al., 1994). Genomic PCR using the primers CGGAATTCGATCGAA-CATCCAGGG and CGGCGCTAAGCTATTCC was performed to obtain a 483 bp DNA fragment from the 3’ region of the byn transcript (Kispert et al., 1994; Singer et al., 1996). This fragment was subcloned into bluescript (Stratagene) and used for the production of RNA probes.

Cuticular analysis

Embryos were allowed to develop for 24 hours and then mounted in a 1:1 mixture of Hoyers mountant and lactic acid (Struhl, 1984).

Molecular and morphological responses to localized Tor activity at the posterior pole

The Tor receptor, as well as Ras, Raf, MEK and MAPK, are required for the normal patterns of transcription of the gap genes tll and hkb at each end of the early embryo. The products of these gap genes then, in turn, specify the pattern of the most anterior and most posterior portions of the body (see Introduction; Fig. 1). At the posterior pole, the requirement for Tor and the Ras/Raf signal transduction pathway is absolute - neither tll nor hkb are expressed in their absence, and no terminal structures form. In contrast, both tll and hkb are expressed at the anterior pole in the absence of terminal signaling, albeit in abbreviated domains, under the control of the anterior determinant Bicoid (Pignoni et al., 1992; Liaw and Lengyel, 1993). In the present study, we are concerned with how localized Tor activity organizes different patterns of gene expression and cuticular differentiation; hence, to simplify the analysis, we focus on these responses in the posterior half of the body.

Fig. 1.

Qualitatively distinct responses in embryos containing different levels of Ras activity. Embryos in which Ras activity derives solely from 0 (A–C), 1 (D–F), or 2 (G–I) copies of the rp49>rasVI2 transgene or from the wild-type ras+ gene (J-L) are shown stained for tll (red) and hkb (blue) expression in A,D,G and J, and for byn expression in B,E,H and K; the posterior terminal structures formed by these embryos are shown in C,E,I and L (see text for detailed description of genotypes). In wild-type embryos (ras+), tll and hkb are expressed in overlapping domains at the posterior pole (J) and activation by tll and repression by hkb results in a stripe of byn expression (K). tll and hkb also direct the expression of other downstream genes whose combined activities specify the posterior terminalia (L) such as the eighth abdominal denticle belt (A8), the anal pad and tuft, the posterior spiracles (sp), as well as internal structures, such as the hindgut. In 0X rp49>rasVI2 embryos, tll, hkb and byn are not expressed at the posterior pole (A,B) and no posterior terminal structures are formed (C). In contrast, the low level of Ras activity present in IX rp49>rasVI2 embryos is sufficient to induce tll and byn expression, but not hkb expression, in a narrow domain at the pole (D,E) and to specify the formation of those terminal structures, such as the A8 ventral denticle belt and the spiracles, that normally arise farthest from the posterior pole (F). Both tll and hkb are expressed in 2X rp49>rasVI2 embryos (G), with tll expressed in both a broader domain and higher levels than in IX rp49>rasVI2 embryos (D). The domain of byn expression also expands and appears to be expressed at higher levels (H). Finally, the levels of Ras activity in 2X rp49>rasVI2 embryos are sufficient to specify most or all of the exterior terminal structures, including the anal pad and anal tuft (I).

Fig. 1.

Qualitatively distinct responses in embryos containing different levels of Ras activity. Embryos in which Ras activity derives solely from 0 (A–C), 1 (D–F), or 2 (G–I) copies of the rp49>rasVI2 transgene or from the wild-type ras+ gene (J-L) are shown stained for tll (red) and hkb (blue) expression in A,D,G and J, and for byn expression in B,E,H and K; the posterior terminal structures formed by these embryos are shown in C,E,I and L (see text for detailed description of genotypes). In wild-type embryos (ras+), tll and hkb are expressed in overlapping domains at the posterior pole (J) and activation by tll and repression by hkb results in a stripe of byn expression (K). tll and hkb also direct the expression of other downstream genes whose combined activities specify the posterior terminalia (L) such as the eighth abdominal denticle belt (A8), the anal pad and tuft, the posterior spiracles (sp), as well as internal structures, such as the hindgut. In 0X rp49>rasVI2 embryos, tll, hkb and byn are not expressed at the posterior pole (A,B) and no posterior terminal structures are formed (C). In contrast, the low level of Ras activity present in IX rp49>rasVI2 embryos is sufficient to induce tll and byn expression, but not hkb expression, in a narrow domain at the pole (D,E) and to specify the formation of those terminal structures, such as the A8 ventral denticle belt and the spiracles, that normally arise farthest from the posterior pole (F). Both tll and hkb are expressed in 2X rp49>rasVI2 embryos (G), with tll expressed in both a broader domain and higher levels than in IX rp49>rasVI2 embryos (D). The domain of byn expression also expands and appears to be expressed at higher levels (H). Finally, the levels of Ras activity in 2X rp49>rasVI2 embryos are sufficient to specify most or all of the exterior terminal structures, including the anal pad and anal tuft (I).

As shown previously (Pignoni et al., 1990; Bronner and Jâckle, 1991) and in Fig. 1, hkb and tll are expressed in overlapping domains which extend approximately 8% and 15% egg’s length (EL) from the posterior pole in syncytial blastoderm embryos (late stage 4 and early stage 5; staging as in Campos-Ortega and Hartenstein (1985)). The combined activities of tll and hkb then specify the more complex patterns of expression of subordinate transcription factors such as forkhead, hunchback, AbdominalB (AbdB), cad, and brachyenteron (byn) (Casanova, 1990; Weigel et al., 1990; Kispert et al., 1994; Singer et al., 1996; reviewed by Jürgens and Hartenstein, 1993). For example, tll activates byn transcription, while hkb represses it, resulting in a stripe of byn expression (Kispert et al., 1994; see Fig. 1). All of these gene functions appear to play significant roles in directing the formation of the posterior terminalia, including the ventral dentical belt of the eighth abdominal segment (A8), the posterior spiracles, the anal pad and tuft, and internal structures such as the hindgut, posterior midgut, and Malpighian tubules (Casanova, 1990; Jürgens and Hartenstein, 1993).

Different levels of Ras activity specify distinct transcriptional responses.

To test whether different levels of Ras activity can suffice to specify distinct transcriptional responses at the posterior pole, we sought to generate embryos that lack Tor, but express different levels of a constitutively active form of Ras, RasV12 (Trahey and McCormick, 1987). This was accomplished by creating females with the following genetic properties. First, they were homozygous for a null allele of the tor gene: embryos developing from such mutant females lack Tor-dependent activity of endogenous Ras. Second, they carried one or two copies of a Flp-out transgene, rp49>w+>rasVI2, composed of the promoter from the ubiquitously expressed ribosomal protein 49 (rp49) gene (O’Connell and Rosbash, 1984), a w+ Flp-out cassette which blocks transcription (>w+>), and the rasV12 coding sequence (Materials and Methods). Finally, they carried an hsp70-flp transgene in which the coding sequence for the yeast recombinase Flp is placed under the control of the Drosophila hsp70 heat shock promoter. Late third instar larvae and early pupae of this genotype were heat shocked to catalyze the excision of the Flp-out cassette from most of the resident rp49>w+>rasV12 transgenes (see Materials and Methods). We refer to embryos derived from such tor- female germ cells as 1X rp49>rasV12 and 2X rp49>rasV12 embryos depending on whether the females carried one or two copies of the transgene. By the same convention, we refer to embryos derived simply from tor- or wildtype females as 0X rp49>rasV12 or ras+ respectively.

As shown in Fig. 1, neither tll nor hkb transcripts are expressed posteriorly in 0Xrp49>rasV12 embryos. However tll is expressed in a narrow domain at the posterior of IX rp49>rasV12 embryos and in a broader domain in 2X rp49>rasV12 embryos. We also found that tll appears to be expressed at higher level at the posterior pole of 2X rp49>rasV12 embryos compared to the posterior pole of 1X rp49>rasV12 embryos (data not shown). In contrast, we failed to detect posterior hkb expression in 1Xrp49>rasV12 embryos, but could readily detect hkb transcription at the posterior pole of 2X rp49>rasV12 embryos. hkb is expressed at the posterior pole of these 2X rp49>rasV12 embryos in a domain somewhat narrower, and at a level somewhat lower, than that seen in ras+ embryos.

We draw three conclusions from these results. First, the different levels of constitutive RasV12 activity in 1X and 2X rp49>rasV12 embryos appear to distinguish between two qualitatively distinct transcriptional outputs: tll alone, and tll plus hkb. Second, the level of RasV12 activity is also related quantitatively to target gene transcription, as the higher level of RasV12 activity in 2X rp49>rasV12 embryos appears to generate a higher level of posterior tll expression than that generated in IX rp49>rasV12 embryos. Third, even though we would anticipate that RasV12 is active uniformly throughout these embryos, we find that tll and hkb are only transcribed in tightly restricted domains at the poles, with the boundary of the tll domain depending on the different levels of RasV12 activity generated in 1X and 2X rp49>rasV12 embryos. These last results suggest the existence of a local differential in the ability of nuclei to respond to a constant level of RasV12 activity.

We also examined the transcription of the byn gene, in 1X and 2X rp49>rasV12 embryos (Fig. 1). In 1X rp49>rasV12 embryos we observe that byn transcripts are expressed in a small cap (approx. 9% EL; egg length), whereas the domain of expression in 2X rp49>rasV12 embryos is significantly broader (approx. 15% EL). We note that byn expression extends back to the posterior end of 2Xrp49>rasV12 embryos, unlike in ras+ embryos in which byn expression is normally repressed at the posterior pole. Because both the activation and repression of terminal byn expression are known to depend, respectively, on tll and hkb, we surmise that higher levels of Ras activity are required at the posterior of wild-type (ras’) embryos to drive sufficiently high levels of Hkb expression to repress byn expression. These results reinforce those obtained for tll and hkb transcription and further suggest that both these genes are regulated quantitatively as well as qualitatively by the level of Ras activity.

Different levels of Ras activity specify distinct terminal structures

We next examined the effects of different levels of Ras activity on the differentiation of terminal structures. As shown in Fig. 1, IX rp49>rasV12 embryos show a modest restoration of posterior terminal structures which are absent in the 0X rp49>rasV12 embryos. In particular, these embryos form the least terminal of the posterior terminal structures: the eighth abdominal dentical band and the posterior spiracles. The extent of restoration is considerably greater in 2X rp49>rasV12 embryos: these form additional terminal structures such as the anal tuft and anal pads. As observed for the patterns of tll and hkb transcription in 2X rp49>rasV12 embryos, these structures appear in the normal spatial order, and hence the external cuticular pattern of these embryos is similar to that of wildtype (ras’) embryos. Thus, we conclude that different levels of Ras can elicit the formation of distinct subsets of terminal pattern elements, just as they distinguish between different transcriptional outputs.

Different responses to a constant level of Ras activity

As noted above, the localized expression of tll and hkb at the posterior poles of IX and 2X rp49>rasVI2 embryos indicates that nuclei located at different positions along the anteroposterior axis of late syncytial embryos do not respond in the same way to a constant level of RasV12. The gap genes hunchback (hb), Krüppel (Kr), knirps (kni), and giant (gt) are expressed in overlapping central domains of the body, organized in large part by the anterior and posterior determinants Bicoid (Bcd) and Oskar (Osk) (reviewed by St. Johnston and Nüsslein-Volhard, 1992). All four of these gap genes encode transcription factors, and at least some are capable of acting as repressors (reviewed by Gray and Levine, 1996). Hence, it is possible that the localized expression of one or more of these factors blocks the response of nuclei in more central portions of the body to Ras activity.

To test this possibility, we have examined the consequences of generating uniform RasV12 activity in embryos lacking the anterior and posterior determinant systems as well as Tor receptor activity. This was accomplished by creating females that (i) were homozygous for recessive loss-of-function mutations of the bcd, osk, and torso-like (tsl) genes (the tsl gene is required during oogenesis for the activity of the Tor receptor during embryogenesis; Stevens et al., 1990; Savant-Bhonsale and Montell, 1993; Martin et al., 1994), (ii) carried the hsp70-flp transgene and (iii) carried one copy of either the rp49>w+>rasVI2 transgene or a Tuba1>w+>rasV12 transgene which has a stronger ubiquitous promoter derived from the Tubulinod (Tubod) gene (Materials and Methods).

In the absence of exogenous RasV12 activity, embryos derived from bcd-osk-tsl- females express Kr uniformly and at high level, but lack detectable hb, kni, gt, tll and hkb transcription (Struhl et al., 1992; data not shown). These embryos also express moderate levels of Hb protein derived from ubiquitous, maternally derived hb transcripts. When bcd-osk-tsl- females carrying a single copy of either the rp49>w+>rasVI2 or Tuba I>w+>rasVI2 transgene were heat shocked as late larvae or early pupae, they gave rise to embryos that expressed tll uniformly throughout the body but failed to express hkb (Fig. 2B and data not shown). Thus, all of the nuclei in these embryos appear to respond similarly to low to moderate levels of constitutive RasV12 activity by transcribing tll, but not hkb.

Fig. 2.

Opposing roles of Ras and centrally expressed gap genes in controlling tll and hkb transcription. The diagrams adjacent to each embryo represent, in simplified form, early gap gene expression and Ras activity in the embryo. In embryos derived from tor females carrying one copy of the Tuba1>rasV12 transgene (A), as in embryos derived from tor females carrying two copies of the rp49>rasV12 transgene (Fig. 1G–I), moderate levels of uniform RasV12 activity induce tll and hkb expression at the poles. In embryos derived from bcdosktsl females (B), Hb and Kr are expressed uniformly and Kni and Gt expression are repressed by Hb (Struhl et al., 1992); under these conditions, moderate levels of uniform RasV12 activity derived from the Tubα1>rasV12 transgene drive tll, but not hkb, transcription throughout the embryo. hkb expression is, however, restored at the posterior pole of embryos derived from bcdosk females carrying one copy of the wild-type tsl allele and one copy of the Tuba1>rasV12 transgene (C), presumably due to high levels of endogenous Ras activity generated by the Tor receptor. As shown in (D), this endogenous Ras activity is sufficient to specify essentially normal domains of expression of both tll and hkb in embryos derived from bcdosk-−females that lack RasV12 activity, even though the remaining gap gene products are initially either absent or uniformly expressed.

Fig. 2.

Opposing roles of Ras and centrally expressed gap genes in controlling tll and hkb transcription. The diagrams adjacent to each embryo represent, in simplified form, early gap gene expression and Ras activity in the embryo. In embryos derived from tor females carrying one copy of the Tuba1>rasV12 transgene (A), as in embryos derived from tor females carrying two copies of the rp49>rasV12 transgene (Fig. 1G–I), moderate levels of uniform RasV12 activity induce tll and hkb expression at the poles. In embryos derived from bcdosktsl females (B), Hb and Kr are expressed uniformly and Kni and Gt expression are repressed by Hb (Struhl et al., 1992); under these conditions, moderate levels of uniform RasV12 activity derived from the Tubα1>rasV12 transgene drive tll, but not hkb, transcription throughout the embryo. hkb expression is, however, restored at the posterior pole of embryos derived from bcdosk females carrying one copy of the wild-type tsl allele and one copy of the Tuba1>rasV12 transgene (C), presumably due to high levels of endogenous Ras activity generated by the Tor receptor. As shown in (D), this endogenous Ras activity is sufficient to specify essentially normal domains of expression of both tll and hkb in embryos derived from bcdosk-−females that lack RasV12 activity, even though the remaining gap gene products are initially either absent or uniformly expressed.

As shown in Fig. 1G, doubling the doseage of the rp49>rasVI2 transgene in tor- females generates embryos that have sufficient RasV12 activity to direct hkb expression at the posterior pole, and a similar result is obtained using a single copy of the TubaI>rasVI2 transgene in place of two copies of the rp49>rasVI2 transgene (Fig. 2A). Nevertheless, the same level of constitutive RasV12 activity is not sufficient to drive hkb transcription in embryos lacking all three determinant systems, even at the posterior pole (Fig. 2B). We note that the particular constellation of gap gene proteins uniformly expressed in embryos from bcdosktsl females (High Kr, moderate Hb, no Kni and no Gt; Struhl et al., 1992) is not found at the posterior pole of embryos derived from either wild-type or tor- females. Thus, the abnormal presence of Hb and/or Kr throughout these embryos might be responsible for their failure to express hkb in response to moderate RasV12 activity.

To ask whether a higher level of Ras activity can elicit hkb transcription in embryos lacking both the anterior and posterior determinant systems, we examined tll and hkb transcription in embryos derived from bcd-−osk females carrying a single copy of the Tubα1>rasV12 transgene. These embryos retain wild-type tsl function and hence give rise to embryos in which ubiquitous RasV12 activity is supplemented at the poles by normal levels of endogenous Ras activity. As shown in Fig. 2C, they showed localized transcription of hkb at both poles, presumably in response to normal, localized activity of endogenous Ras. Thus, we infer that higher levels of Ras activity generated under these conditions can suffice to drive hkb transcription.

Finally, it is informative to compare the patterns of tll and hkb transcription at the posterior poles of embryos derived from bcdosk females (Fig. 2D) with those derived from tor females which carry one copy of the Tuba1>rasV12 transgene (Fig. 2A). In the former, these patterns depend solely on the normal, localized activity of the Tor receptor, and hence on the polarized activity of wild-type Ras. In the latter, the only Ras activity is that of the uniformly expressed RasV12 protein; consequently, the restricted expression of tll and hkb presumably reflects the ability of the remaining gap gene products to repress Ras-dependent transcription of tll and hkb. Nevertheless, each input can generate a correctly ordered pattern of tll and hkb expression, suggesting that they function in a cooperative fashion to organize tll and hkb expression during normal development.

During early embryogenesis, terminal patterning is organized by a mechanism involving local activation of a transmembrane receptor tyrosine kinase, Tor, and transduction by the Ras signal transduction pathway (reviewed by Perrimon, 1993). Here, we show that different levels of a constitutively activated form of Ras, RasV12, are able to generate different transcriptional responses and specify the formation of distinct structures at the posterior end of the body. Low levels of RasV12 suffice only to direct tll transcription and the formation of posterior structures such as the A8 ventral denticle band and the posterior spiracles, whereas higher levels induce both tll and hkb transcription and specify additional, more terminal structures such as the anal pads and anal tuft. Thus, our results establish that quantitative variation in the level of the Ras activity can lead to qualitatively distinct outcomes, and hence suggest that localized activition of the Tor receptor organizes terminal body pattern by creating a gradient of activity of the Ras signal transduction pathway. Our results also draw attention to the presence of another significant influence on terminal body pattern, namely the localized deployment of other gap gene products in more central portions of the body. At least one of these gap gene products appears to dampen or block the responsiveness to activated Ras, counteracting the influence of the Ras activity gradient emanating from the pole.

Generation and interpretation of a Ras activity gradient

The receptor Tor is thought to be activated by an extracellular ligand that is generated locally at each pole and then adsorbed and sequestered by binding to the receptor (Stevens et al., 1990; Sprenger and Nüsslein-Volhard, 1992; Casanova and Struhl, 1993). Diffusion of the ligand before binding might suffice to generate a graded distribution of activated Tor and hence of activated Ras. It is also possible that formation of the Ras activity gradient depends at least in part on subsequent movement of activated Tor, Ras or other downstream signaling components within the syncytial embryo.

The most likely intracellular targets of the Ras transduction pathway are factors which regulate tll and hkb transcription. In the case of tll, analysis of cis-acting regulatory sequences upstream of the promoter suggests that activity of the Ras pathway normally activates tll transcription by antagonizing the action of a ubiquitously expressed transcriptional repressor which binds to these sequences (Liaw et al., 1995). One candidate for this repressor is the product of grainyhead gene, a homologue of the vertebrate NFT-1 gene. grainyhead function is apparently required to prevent general transcription of the endogenous tll gene (Liaw et al., 1995). Moreover, MAPK is capable of phosphorylating NTF-1 in vitro (Liaw et al., 1995). Hence, a Ras activity gradient might generate a graded distribution of phosphorylated Grainyhead protein, relieving repression of tll transcription when a sufficient fraction of the protein is phosphorylated and hence inactivated.

A similar mechanism could also apply to the localized transcription of hkb. For example, if Grainyhead were to act directly as a repressor of both tll and hkb, different levels of the active form of the protein might suffice to repress each gene, and hence, different levels of Ras activity would be required to release them from repression. Alternatively, other transcription factors that have a different sensitivity to the Ras signal transduction pathway may be responsible for regulating hkb transcription.

Whatever the mechanism, our findings indicate that the tll and hkb genes respond in qualitatively distinct ways to different levels of Ras activity and suggest that a relatively small difference in Ras activity – that resulting from a two-fold increase in the maternal dose of the rp49>rasV12 transgene – can suffice to distinguish between the ‘off’ and ‘on’ states of transcription. Thus, the threshold sensitivity of tll and hkb to differences in Ras activity may be similar to that exhibited by other gap genes in response to the Bicoid and Hunchback gradient morphogens (Struhl et al., 1989, 1992).

Although most, or all, aspects of posterior terminal pattern are governed by the actions of Tll and Hkb protein, the resulting patterns of gene expression and cuticular differentiation are complex, involving overlapping patterns of transcription of many target genes such as byn, fkh, caudal and abdB, and the formation of diverse morphological structures. Previous studies (Strecker et al., 1988; Casanova, 1990; Diaz et al., 1996) have provided evidence that this complexity arises in part from the formation of local gradients of Tll and Hkb expression within the domains in which the two genes are transcribed. We have observed that the levels of both tll and hkb transcription appear to be sensitive to the level of Ras activity. For example, tll is expressed at higher level at the posterior end of 2X rp49>rasV12 embryos compared to 1X rp49>rasV12 embryos, and the level of posterior hkb transcription appears similarly dependent when 2X rp49>rasV12 and ras+ embryos are compared. As illustrated by the transcription of byn in these embryos, quantitative differences in tll and hkb transcription appear to correlate with corresponding differences in the transcription of further downstream target genes. Thus, the activity gradient of Ras may be translated into local gradients of tll and hkb which in turn have instructive roles in organizing subordinate gene expression and differentiation.

A complicating factor in interpreting the organizing influence of the Ras activity gradient is the role played by other gap genes, which appear to modulate the response to Ras activity. Our results establish that nuclei at different positions along the anteroposterior axis are predisposed to respond in distinct ways to the same level of constitutive RasV12 activity. Moreover, they indicate that this predisposition results from the localized activities of one or more of the remaining gap genes in more central portions of the body. In the posterior half of the body, the gap proteins Hunchback (Hb), Krüppel (Kr), Knirps (Kni) and Giant (Gt) are expressed in an ordered series of overlapping domains, and these are organized principally in response to the anterior and posterior determinants Bicoid (Bcd) and Osk (Osk). Hence, irrespective of the activity of the terminal determinant system, nuclei at different positions from the posterior pole will be exposed to different constellations of these proteins, all of which are DNA binding transcription factors. As shown in Figure 2, localized expression of these ‘central’ gap gene proteins can suffice to organize overlapping posterior domains of tll and hkb transcription in 1X Tuba1>rasV12 embryos in which all nuclei are exposed to the same level of Ras activity.

Thus, nuclei in the posterior half of the body are subject to two, independent inputs which control the expression of tll and hkb: a Ras activity gradient spreading from the posterior pole which drives their expression, and an opposing gradient of one or more central gap gene proteins which repress their expression. Under the appropriate circumstances, either input can generate a correctly ordered pattern of tll and hkb transcription. In the wild-type condition, the two inputs may function in a mutually reinforcing fashion to generate the normal patterns of transcription of both target genes. This situation is analogous to the opposing roles of Bcd and Osk in organizing the expression of the central gap genes themselves (Hulskamp et al., 1989; Irish et al., 1989; Struhl, 1989; Wharton and Struhl, 1991; Struhl et al., 1992), and may reflect a common strategy used in pattern formation. As is the case for Bcd and Osk, the opposing activities of the Ras gradient and the central gap genes also appear to be mutually exclusive. Conditions which force high levels of ectopic gap gene expression at the posterior pole block terminal development, probably through the repression of tll and hkb transcription (Gaul and Jâckle, 1989; Struhl, 1989a). Conversely, conditions which force high levels of ectopic Tor or Ras activity in the central portion of the body can block transcription of the central gap genes (Klingler et al., 1988; Casanova and Struhl, 1989; Steingrimsson et al., 1991). A similar phenomenon of mutual exclusion and collaboration between opposing signals is observed in the developing adult legs, where Wg and Dpp restrict each other’s expression defining the dorso-ventral axis, but operate in conjunction to establish the proximo-distal axis (Brook and Cohen, 1996; Jiang and Struhl, 1996; Penton and Hoffmann, 1996).

A rheostat function for Ras: implications for other systems

As noted in the Introduction, receptor tyrosine kinases are capable of transducing a variety of signals through the Ras pathway, in some cases generating distinct outputs to different ligands or to different concentrations of the same ligand (reviewed by Marshall, 1995). Our present demonstration that different levels of a constitutively active form of Ras can elicit distinct transcriptional outputs indicates one mechanism by which this specificity can be achieved. In essence, Ras can function as a rheostat in which quantitative variations in activity provide one or more thresholds that differentially regulate gene expression. This mechanism may be particularly important in developmental contexts in which the graded or localized distribution of ligand appears to organize cell pattern. In addition to terminal body patterning, receptor tyrosine kinases have been implicated in a number of patterning phenomena of this kind in Drosophila, including organization of the follicular epithelium during oogenesis and control of dorso-ventral epidermal pattern in the embryo (Clifford and Schüpbach, 1989; Price et al., 1989; Brand and Perrimon, 1994; Schweitzer et al., 1995). A rheostat mechanism might also be involved in situations in which cells of the same type can respond in distinct ways to different ligands, each received by a different receptor tyrosine kinase (Marshall, 1995). A possible example of this phenomenon in Drosophila is the R7 cell in the Drosophila eye, which requires the activity of both Sevenless and the EGF receptor during development (Campos-Ortega et al., 1979; Tomlinson and Ready, 1987; Xu and Rubin, 1993; Freeman, 1996).

We thank Atsuko Adachi for generating transformant lines of the rp49>w+>rasV12 and Tubα1>w+>rasV12 transgenes, Mark Fortini for the rasV12 coding sequence, Jordi Casanova for plasmids for making the tll and hkb probes, Dan Kalderon for the rp49 promoter, and Andrew Tomlinson for advice. S. G. was supported by the College of Physicians and Surgeons, Columbia University. G. S. is an Investigator of the Howard Hughes Medical Institute.

Basler
,
K.
and
Struhl
,
G.
(
1994
).
Compartment boundaries and the control of Drosophila limb pattern by hedgehog protein
.
Nature
368
,
208
214
.
Brand
,
A. H.
and
Perrimon
,
N.
(
1994
).
Raf acts downstream of the EGF receptor to determine dorsoventral polarity during Drosophila oogenesis
.
Genes Dev
.
8
,
629
639
.
Bronner
,
G.
and
Jâckle
,
H.
(
1991
).
Control and function of terminal gap gene activity in the posterior pole region of the Drosophila embyro
.
Mech. Dev
.
35
,
205
211
.
Brook
,
W. J.
and
Cohen
,
S. M.
(
1996
).
Antagonistic interactions between wingless and decapentaplegic responsible for dorsal-ventral pattern in the Drosophila Leg
.
Science
273
,
1373
1377
.
Brunner
,
D.
,
Oellers
,
N.
,
Szabad
,
J.
,
Biggs
,
W. H.
,
Zipursky
,
S. L.
and
Hafen
,
E.
(
1994
).
A gain-of-function mutation in Drosophila MAP kinase activates multiple receptor tyrosine kinase signaling pathways
.
Cell
76
,
875888
.
Campos-Ortega
,
J.
and
Hartenstein
,
V.
(
1985
).
The Embryonic Development of Drosophila melanogaster
.
Berlin
:
Springer-Verlag
.
Campos-Ortega
,
J. A.
,
Jürgens
,
G.
and
Hofbauer
,
A.
(
1979
).
Cell clones and pattern formation: Studies on sevenless a mutant of Drosophila melanogaster
.
Wilhelm Roux’s Archiv. Dev. Biol
.
186
,
27
50
.
Casanova
,
J.
(
1990
).
Pattern formation under the control of the terminal system in the Drosophila embryo
.
Development
110
,
621
628
.
Casanova
,
J.
,
Furriols
,
M.
,
McCormick
,
C. A.
and
Struhl
,
G.
(
1995
).
Similarities between trunk and spatzle, putative extracellular ligands specifying body pattern in Drosophila
.
Genes Dev
.
9
,
2539
2544
.
Casanova
,
J.
,
Llimargas
,
M.
,
Greenwood
,
S.
and
Struhl
,
G.
(
1994
).
An oncogenic form of human raf can specify terminal body pattern in Drosophila
.
Mech. Dev
.
48
,
59
64
.
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
2938
.
Casanova
,
J.
and
Struhl
,
G.
(
1993
).
The torso receptor localizes as well as transduces the spatial signal specifying terminal body pattern in Drosophila
.
Nature
362
,
152
155
.
Clifford
,
R. J.
and
Schüpbach
,
T.
(
1989
).
Coordinately and differentially mutable activities of torpedo, the Drosophila melanogaster homolog of the vertebrate EGF receptor gene
.
Genetics
123
,
771
787
.
Cowley
,
S.
,
Paterson
,
H.
,
Kemp
,
P.
and
Marshall
,
C. J.
(
1994
).
Activation of MAP kinase kinase is necessary and sufficient for PC12 differentiation and for transformation of NIH 3T3 cells
.
Cell
77
,
841
852
.
Diaz
,
R.
,
Harbecke
,
R.
,
Singer
,
J.
,
Pignoni
,
F.
,
Janning
,
W.
and
Lengyel
,
J.
(
1996
).
Graded effect of tailless on posterior gut development: molecular basis of an alleleic series of a nuclear receptor gene
.
Mech Dev
.
54
,
119
139
.
Duffy
,
J. B.
and
Perrimon
,
N.
(
1994
).
The torso pathway in Drosophila: lessons on receptor tyrosine kinase signaling and pattern formation
.
Dev. Biol
.
166
,
380
395
.
Fortini
,
M. E.
,
Simon
,
M. A.
and
Rubin
,
G. M.
(
1992
).
Signalling by the sevenless protein tyrosine kinase is mimicked by Ras1 activation
.
Nature
355
,
559
561
.
Freeman
,
M.
(
1996
).
Reiterative use of the EGF receptor triggers differentiation of all cell types in the Drosophila eye
.
Cell
87
,
651
660
.
Furriols
,
M.
,
Sprenger
,
F.
and
Casanova
,
J.
(
1996
).
Variation in the number of activated torso receptors correlates with differential gene expression
.
Development
122
,
2313
2317
.
Gaul
,
U.
and
Jâckle
,
H.
(
1989
).
Analysis of maternal effect mutant combinations elucidates regulation and function of the overlap of hunchback and Kruppel gene expression in the Drosophila blastoderm embryo
.
Development
107
,
651
662
.
Gray
,
S.
and
Levine
,
M.
(
1996
).
Transcriptional repression in development
.
Curr.Opin. Cell Biol
.
8
,
358
364
.
Hulskamp
,
M.
,
Schroder
,
C.
,
Pfeifle
,
C.
,
Jâckle
,
H.
and
Tautz
,
D.
(
1989
).
Posterior segmentation of the Drosophila embryo in the absence of a maternal posterior organizer gene
.
Nature
338
,
629
632
.
Irish
,
V.
,
Lehmann
,
R.
and
Akam
,
M.
(
1989
).
The Drosophila posterior-group gene nanos functions by repressing hunchback activity
.
Nature
338
,
646
648
.
Jiang
,
J.
,
Kosman
,
D.
,
Ip
,
Y. T.
and
Levine
,
M.
(
1991
).
The dorsal morphogen gradient regulates the mesoderm determinant twist in early Drosophila embryos
.
Genes Dev
.
5
,
1881
1891
.
Jiang
,
J.
and
Struhl
,
G.
(
1996
).
Complementary and mutually exclusive activities of decapentaplegic and wingless organize axial patterning during Drosophila leg development
.
Cell
86
,
401
409
.
Jürgens
,
G.
and
Hartenstein
,
V.
(
1993
).
The terminal regions of the body pattern
. In
The Development of Drosophila melanogaster
. (ed.
M.
Bate
and
A.
Martinez Arias
), pp.
687
746
.
Cold Spring Harbor Laboratory Press
.
Katz
,
W. S.
,
Hill
,
R. J.
,
Clandinin
,
T. R.
and
Sternberg
,
P. W.
(
1995
).
Different levels of the C. elegans growth factor LIN-3 promote distinct vulval precursor fates
.
Cell
82
,
297
307
.
Kayne
,
P. S.
and
Sternberg
,
P. W.
(
1995
).
Ras pathways in Caenorhabditis elegans
.
Curr Opin. Genet. Dev
.
5
,
38
43
.
Kazlauskas
,
A.
(
1994
).
Receptor tyrosine kinases and their targets
.
Curr. Opin. Genet. Dev
.
4
,
5
14
.
Kispert
,
A.
,
Herrmann
,
B. G.
,
Leptin
,
M.
and
Reuter
,
R.
(
1994
).
Homologs of the mouse Brachyury gene are involved in the specification of posterior terminal structures in Drosophila, Tribolium, and Locusta
.
Genes Dev
.
8
,
2137
2150
.
Klingler
,
M.
,
Erdelyi
,
M.
,
Szabad
,
J.
and
Nüsslein-Volhard
,
C.
(
1988
).
Function of torso in determining the terminal anlagen of the Drosophila embryo
.
Nature
335
,
275
277
.
Kongsuwan
,
K.
,
Yu
,
Q.
,
Vincent
,
A.
,
Frisardi
,
M. C.
,
Rosbash
,
M.
,
Lengyel
,
J. A.
and
Merriam
,
J.
(
1985
).
A Drosophila Minute gene encodes a ribosomal protein
.
Nature
317
,
555
558
.
Liaw
,
G. J.
and
Lengyel
,
J. A.
(
1992
).
Control of tailless expression by bicoid, dorsal and synergistically interacting terminal system regulatory elements
.
Mech. Dev
.
40
,
47
61
.
Liaw
,
G. J.
,
Rudolph
,
K. M.
,
Huang
,
J. D.
,
Dubnicoff
,
T.
,
Courey
,
A. J.
and
Lengyel
,
J. A.
(
1995
).
The torso response element binds GAGA and NTF-1/Elf-1, and regulates tailless by relief of repression
.
Genes Dev
.
9
,
31633176
.
Lu
,
X.
,
Chou
,
T. B.
,
Williams
,
N. G.
,
Roberts
,
T.
and
Perrimon
,
N.
(
1993a
).
Control of cell fate determination by p21ras/Ras1, an essential component of torso signaling in Drosophila
.
Genes Dev
.
7
,
621
32
.
Lu
,
X.
,
Perkins
,
L. A.
and
Perrimon
,
N.
(
1993b
).
The torso pathway in Drosophila: a model system to study receptor tyrosine kinase signal transduction
.
Development Supplement 47-56
.
Marshall
,
C. J.
(
1994
).
MAP kinase kinase kinase, MAP kinase kinase and MAP kinase
.
Curr. Opin. Genet. Dev
.
4
,
82
89
.
Marshall
,
C. J.
(
1995
).
Specificity of receptor tyrosine kinase signaling: transient versus sustained extracellular signal-regulated kinase activation
.
Cell
80
,
179
185
.
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
.
McCormick
,
F.
(
1994
).
Activators and effectors of ras p21 proteins
.
Curr Opin. Genet. Dev
.
4
,
71
76
.
Nüsslein-Volhard
,
C.
,
Frohnhofer
,
H. G.
and
Lehmann
,
R.
(
1987
).
Determination of anteroposterior polarity in Drosophila
.
Science
238
,
16751681
.
O’Connell
,
P. O.
and
Rosbash
,
M.
(
1984
).
Sequence, structure, and codon preference of the Drosophila ribosomal protein 49 gene
.
Nucl Acids Res
.
12
,
5495
5513
.
Penton
,
A.
and
Hoffmann
,
F. M.
(
1996
).
Decapentaplegic restricts the domain of wingless during Drosophila limb patterning
.
Nature
382
,
162
164
.
Perrimon
,
N.
(
1993
).
The torso receptor protein-tyrosine kinase signaling pathway: an endless story
.
Cell
74
,
219
222
.
Perrimon
,
N.
(
1994
).
Signalling pathways initiated by receptor protein tyrosine kinases in Drosophila
.
Curr Opin. Cell Biol
.
6
,
260
266
.
Pignoni
,
F.
,
Baldarelli
,
R. M.
,
Steingrimsson
,
E.
,
Diaz
,
R. J.
,
Patapoutian
,
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
.
Pirrotta
,
V.
(
1988
).
Vectors for P-mediated transformations in Drosophila
. In
Vectors: A survey of Molecular Cloning Vectors and Their Uses
. (ed.
R. L.
Rodriguez
and
D. T.
Reinhardt
), pp.
437
456
.
Boston
:
Butterworths
.
Price
,
J. V.
,
Clifford
,
R. J.
and
Schüpbach
,
T.
(
1989
).
The maternal ventralizing locus torpedo is allelic to faint little ball, an embryonic lethal, and encodes the Drosophila EGF receptor homolog
.
Cell
56
,
1085
1092
.
Rubin
,
G. M.
and
Spradling
,
A. C.
(
1982
).
Genetic transformation of Drosophila with transposable element vectors
.
Science
218
,
348
353
.
Savant-Bhonsale
,
S.
and
Montell
,
D. J.
(
1993
).
torso-like encodes the localized determinant of Drosophila terminal pattern formation
.
Genes Dev
.
7
,
2548
2555
.
Schüpbach
,
T.
and
Wieschaus
,
E.
(
1986
).
Maternal effect mutations altering the anterior-posterior pattern of the Drosophila embryo
.
Roux’s Arch Dev Biol
.
195
,
302
317
.
Schweitzer
,
R.
,
Shaharabany
,
M.
,
Seger
,
R.
and
Shilo
,
B. Z.
(
1995
).
Secreted Spitz triggers the DER signaling pathway and is a limiting component in embryonic ventral ectoderm determination
.
Genes Dev
.
9
,
1518
1529
.
Singer
,
J.B.
,
Harbecke
,
R.
,
Kusch
,
T.
Reuter
,
R.
, and
Lengyel
,
J.A.
(
1996
).
Drosophila brachyenteron regulates gene activity and morphogenesis in the gut
.
Development
122
,
3707
3718
.
Sprenger
,
F.
and
Nüsslein-Volhard
,
C.
(
1992
).
Torso receptor activity is regulated by a diffusible ligand produced at the extracellular terminal regions of the Drosophila egg
.
Cell
71
,
987
1001
.
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
,
D.
and
Nüsslein-Volhard
,
C.
(
1992
).
The origin of pattern and polarity in the Drosophila embryo
.
Cell
68
,
201
219
.
Steingrimsson
,
E.
,
Pignoni
,
F.
,
Liaw
,
G. J.
and
Lengyel
,
J. A.
(
1991
).
Dual role of the Drosophila pattern gene tailless in embryonic termini
.
Science
254
,
418
421
.
Stevens
,
L. M.
,
Frohnhofer
,
H. G.
,
Klingler
,
M.
and
Nüsslein-Volhard
,
C.
(
1990
).
Localized requirement for torso-like expression in follicle cells for development of terminal anlagen of the Drosophila embryo
.
Nature
346
,
660663
.
Strahle
,
U.
,
Blader
,
P.
,
Adam
,
J.
and
Ingham
,
P. W.
(
1994
).
A simple and efficient procedure for non-isotopic in situ hybridization to sectioned material
.
Trends Genet
.
10
,
75
76
.
Strecker
,
T. R.
,
Merriam
,
J. R.
and
Lengyel
,
J. A.
(
1988
).
Graded requirement for the zygotic terminal gene, tailless, in the brain and tail region of the Drosophila embryo
.
Development
102
,
721
734
.
Struhl
,
G.
(
1984
).
Spiltting the bithorax complex in Drosophila
.
Nature
308
,
454
457
.
Struhl
,
G.
(
1989
).
Differing strategies for organizing anterior and posterior body pattern in Drosophila embryos
.
Nature
338
,
741
744
.
Struhl
,
G.
and
Basler
,
K.
(
1993
).
Organizing activity of wingless protein in Drosophila
.
Cell
72
,
527
540
.
Struhl
,
G.
,
Johnston
,
P.
and
Lawrence
,
P. A.
(
1992
).
Control of Drosophila body pattern by the hunchback morphogen gradient
.
Cell
69
,
237
249
.
Struhl
,
G.
,
Struhl
,
K.
and
Macdonald
,
P. M.
(
1989
).
The gradient morphogen bicoid is a concentration-dependent transcriptional activator
.
Cell
57
,
12591273
.
Theurkauf
,
W. E.
,
Baum
,
H.
,
Bo
,
J.
and
Wensink
,
P. C.
(
1986
).
Tissuespecific and constitutive alpha-tubulin genes of Drosophila melanogaster code for structurally distinct proteins
.
Proceedings of the National Academy of Sciences of the United States of America
83
,
8477
8481
.
Tomlinson
,
A.
and
Ready
,
D. F.
(
1987
).
Cell fate in the Drosophila ommatidium
.
D ev. Biol
.
120
,
366
376
.
Trahey
,
M.
and
McCormick
,
F.
(
1987
).
A cytoplasmic protein stimulates normal N-ras p21 GTPase, but does not affect oncogenic mutants
.
Science
238
,
542
545
.
Tsuda
,
L.
,
Inoue
,
Y. H.
, et al. 
. (
1993
).
A protein kinase similar to MAP kinase activator acts downstream of the raf kinase in Drosophila
.
Cell
72
,
407
414
.
Weigel
,
D.
,
Jürgens
,
G.
,
Klingler
,
M.
and
Jackle
,
H.
(
1990
).
Two gap genes mediate maternal terminal pattern information in Drosophila
.
Science
248
,
495
498
.
Wharton
,
R. P.
and
Struhl
,
G.
(
1991
).
RNA regulatory elements mediate control of Drosophila body pattern by the posterior morphogen nanos
.
Cell
67
,
955
967
.
Xu
,
T.
and
Rubin
,
G. M.
(
1993
).
Analysis of genetic mosaics in developing and adult Drosophila tissues
.
Development
117
,
1223
1237
.