BIOLOGICAL OVERVIEW

Torso-like is the localized determinant for activation of Torso, a growth-factor receptor that is synthesized ubiquitously over the surface membrane of early embryos. Activation of Torso, the pivotal receptor of the terminal system, leads through the ras pathway to the activation of genes (tailless and huckebein) coding for transcription factors capable of directing the terminal system of the embryo. Trunk (Trk), the proposed ligand for Tor, is secreted as an inactive precursor into the perivitelline fluid that lies between the embryonic membrane and the vitelline membrane (VM), the inner layer of the eggshell. The spatial regulation of Trk processing is thought to be mediated by the secreted product of the torsolike (tsl) gene, which is expressed during oogenesis by a specialized population of follicle cells present at the two ends of the oocyte. What must take place in order for the ubiquitously distributed Torso to become activated at the embryonic poles?

Early transplant experiments indicated that torso-like was provided by the ovary. To determine where normal expression of tsl occurs, and whether or not this site would in fact be in the somatic cells of the ovary, ovarian transplants were used. These were ovaries mutant for tsl but whose other somatic tissues were normal. The progeny derived from these females exhibited the tsl mutant phenotype. Satisfied that torso-like must be provided by the ovary, subsequent experiments tried to further localize the requirement for torso-like. Was tsl expression required in polar follicle cells? Mitotic recombinants were induced to create mosaic flies with ovaries that contained clones of tsl mutant follicle cells. The genetic marker fragile chorion was used, a mutation that deletes the filzkörper, a prominent telson structure. Among the 10,960 resulting embryos examined, seven showed mutant tsl posterior phenotypes, but had normal terminal structures in the anterior. These results show that the two ends of the embryo are independently affected and supports the idea of a localized supply of tsl at each of the embryo's two poles (Stevens, 1990).

Once the tsl gene had been cloned, localization of the tsl gene product could be examined using in situ hybridization. As expected, in stage 8 embryos TSL mRNA is detected in anterior and posterior follicle cells (Martin, 1994). Thus the ubiquitously expressed Torso receptor could receive signals made only at the anterior and posterior poles, not in the egg or embryo itself, but in follicle cells that surround the embryo.

The anterior expression of tsl is confine to border cells. These are specialized follicle cells that exhibit behaviors quite different from the vast majority of follicle cells. Instead of forming a cell layer around the oocyte as do other follicle cells, border cells remain at the very anterior tip of the egg chamber. These border cells will later migrate to the oocyte-nurse cell boundary, where they associate with other cells to form a hollow cone that allows sperm cell passage. In addition to expressing TSL, border cells express the gene slow border cells, the fly homolog of CCAAT/enhancer-binding protein (CEB/P) (Montell, 1992).

Border cells also express the breathless FGF receptor homolog required for border cell migration (Murphy, 1995). The special circumstance that results in tsl expression by posterior follicle cells has not yet been characterized.

The argument is made that Trunk (with its growth hormone sequence characteristics) is in fact the true ligand for Torso, and that TSL has a function in follicle cells analogous to wind, pip and ndl in the dorsoventral system. These accessory genes are active in follicle cells and are required for activation of Spätzle, thereby enabling Spätzle to bind to and activate the receptor Toll. By analogy, Trunk might be secreted everywhere from the egg, but activated locally through a protein cascade involving TSL. This argument might be valid, considering the fact that Trunk's sequence is that of a growth factor (Casanova, 1995).

The Torso tyrosine kinase receptor is distributed along the surface of the embryo but it is only activated at the poles by a diffusible extracellular ligand (generated at each pole), that is trapped by the receptor, thereby impeding further diffusion. It is not well understood how this signal is generated, although it is known to depend on the activity of many genes such as torso-like and trunk. To further investigate the mechanism involved in the local activation of the Tor receptor, the normal expression of the Tsl protein was altered by generating females in which the tsl gene is expressed in the oocyte under the control of the tor promoter, rather than in the ovarian follicle cells. Analysis of the phenotypes generated by this hybrid gene and its interactions with mutations in other genes in the pathway has enabled dissection of the mechanism of Tor receptor activation and a more precise definition of the role of the different genes acting in this process. Uniform expression of tsl in the oocyte is able to trigger unrestricted activation of the Tor receptor. Observed is an expansion of the terminal portions of the body at the expense of the middle body segments. Specifically, tailless expression expands in mutant embryos. Triggering of Tor activation by Tsl expressed in the germ-line requires the product of the trk gene. Triggering of Tor activation by Tsl in the germ-line requires the product of fs(1)ph and fs(1)Nas. These two gene products are thought to be involved in the process of transference of the Tsl product from the follicle cells to the germ-line. But these two gene products are still required for Tor receptor activation, even when Tsl is expressed in the germ-line (Furriols, 1998).

Trunk the proposed ligand for Tor, is secreted as an inactive precursor into the perivitelline fluid that lies between the embryonic membrane and the vitelline membrane (VM), the inner layer of the eggshell. Tsl protein is specifically localized to the polar regions of the VM in laid eggs. Although Tsl can associate with nonpolar regions of the VM, the activity of polar-localized Tsl is enhanced, suggesting the existence of another spatially restricted factor acting in this pathway. The incorporation of Tsl into the VM provides a mechanism for the transfer of spatial information from the follicle cells to the developing embryo. Tsl represents the first example of an embryonic patterning determinant that is a component of the eggshell (Stevens, 2003).

Although it has been reported that Tsl protein is present on the embryonic membrane, this result could not be replicated. Instead, using affinity-purified anti-Tsl antibodies, strong staining of Tsl protein was detected in 0- to 2-hr-old embryos at the anterior and posterior poles of the VM, the inner layer of the eggshell. These stainings were carried out by dechorionating embryos and attaching them, still contained within their VMs, onto glass microscope slides. A needle was then used to penetrate the VMs and release the embryos; the empty VMs were left attached to the slide and were fixed and then incubated with anti-Tsl antibodies. Tsl staining could only be detected if the antibodies had access to the inner face of the VM; there was no staining of intact VMs (with the embryo still inside). This indicates that Tsl protein is present on the inside of the VM and is thus capable of interacting with components of the perivitelline fluid. No signal was present on the VMs of embryos derived from females homozygous for tsl^PZRev32, a protein null allele of tsl ; this finding indicates that the staining is specific for Tsl protein. The binding of Tsl to the VM is quite stable, since the polar staining pattern of VMs from 7- to 9-hr-old embryos was equivalent in intensity to that of 0- to 2-hr-old embryos stained in parallel (Stevens, 2003).

The distribution of Tsl on the VM of embryos maternally mutant for tor or trk is identical to that observed for wild-type embryos, as would be expected given their proposed downstream functions as receptor and ligand, respectively. In addition to trk and tsl, two other maternally expressed genes, female sterile(1)Nasrat [fs(1)N] and female sterile(1)pole hole [fs(1)ph], are required for the activation of the Tor receptor. Although females homozygous for the hypomorphic alleles fs(1)ph¹⁹⁰¹ and fs(1)N²¹¹ produce embryos with the terminal class phenotype of head and tail defects, females mutant for stronger alleles of these two genes produce collapsed eggs, suggesting an additional role in VM formation. Consistent with this finding, stronger alleles of fs(1)ph and fs(1)N disrupt cross-linking of the VM, an important step in eggshell formation. Both fs(1)ph and fs(1)N encode large extracellular proteins that are expressed in the germline during oogenesis and coat the oocyte surface. The activities of fs(1)N and fs(1)ph have been reported to be required for the accumulation and stabilization at the oocyte surface of an epitope-tagged version of the Tsl protein expressed in its normal domain at the poles of the follicle. When anti-Tsl antibodies were used to stain the VMs of embryos from mothers mutant for fs(1)ph¹⁹⁰¹ and fs(1)N²¹¹, Tsl was still detected in a polar cap, but the staining was less intense and appeared to be distributed over a larger region than in wild-type VMs. Thus, in eggs from fs(1)ph¹⁹⁰¹ and fs(1)N²¹¹ mutant females, Tsl protein is retained on the VM, but it appears to spread from the poles and is not as highly concentrated at the ends as in wild-type eggs (Stevens, 2003).

tsl expressed in the germline can rescue the loss-of-function phenotype of embryos derived from tsl mutant mothers and, at higher levels, produces the segmentation defects characteristic of the gain-of-function phenotype caused by ectopic Tor activation. However, these effects are not seen in the embryos of fs(1)ph and fs(1)N mutant mothers expressing tsl in the germline. Although the concentration of polar Tsl protein is reduced on the VMs of fs(1)ph maternal mutants, uniformly high levels of Tsl protein were achieved on the VMs of these embryos by using the Gal4/UAS system to ectopically express tsl in the follicle cell layer of mutant mothers. In the progeny of otherwise wild-type females, high levels of Tsl protein on the VM produced a strong gain-of-function phenotype in the embryo, in which the head and tail structures were present but the segmented thoracic and abdominal regions of the embryo were disrupted. In contrast, despite high levels of Tsl on their VMs, the embryos from fs(1)ph/fs(1)ph homozygous mutant mothers ectopically expressing tsl exhibit a terminal loss-of-function phenotype indistinguishable from that of embryos from fs(1)ph mutant mothers not misexpressing tsl. In these embryos, the segmented region of the embryo is normal, but the head is disrupted and structures posterior to abdominal segment 8 are deleted. Similar results were obtained for fs(1)N maternal mutants. These results indicate that the loss-of-function phenotype seen in embryos from fs(1)ph¹⁹⁰¹ and fs(1)N²¹¹ mothers is not due simply to decreased amounts of Tsl protein on the VM, but rather reflects a specific requirement for Polehole and Nasrat activities in order for Tsl to exert its function (Stevens, 2003).

The ability of ectopically expressed Tsl to produce an embryonic phenotype similar to that of constitutively active Tor has been interpreted to mean that Tsl is capable of functioning ectopically, and that consequently, the restriction of tsl expression to the poles of the follicle is critical for the production of a localized Tor ligand. However, because the ligand for Tor is diffusible, the spatial parameters of Tor activation are determined by the concentration and distribution of Tor protein in the embryonic membrane relative to the amount of ligand processed at the poles. Evidence is presented that even when tsl is expressed ectopically, it is active only in the polar regions. Thus, the tsl gain-of-function phenotype is likely the result of diffusion of excess ligand from the poles. In these experiments, tsl was expressed at low levels in the female germline; these low levels resulted in the uniform distribution of Tsl protein in the VM of the embryonic progeny. tsl mutant females carrying this construct, termed CBBtsl, produced some embryos in which the terminal structures were completely restored. Despite the complete rescue of terminal cuticular structures, many of these embryos did not exhibit the segmentation defects associated with uniform Tor activation. These segmentation defects are caused by ectopic expression of tailless (tll), the terminal region gap gene, which at high levels can interfere with the expression of central gap genes such as Krüppel (Kr). Consistent with the relatively normal cuticular phenotypes of the embryonic progeny of tsl mutant mothers carrying CBBtsl, tll expression in these embryos was found to be restricted to the ends of the embryo; this finding suggests that Tor receptor activation is restricted to the polar regions (Stevens, 2003).

Spatial regulation of the expression of the gap genes, the first zygotic patterning genes to be expressed during embryogenesis, is determined by the activity of the three maternal pathways (anterior, posterior, and terminal) required for the development of the anterior-posterior axis of the embryo. In addition to localized maternal input, interactions between the gap gene products themselves led to further refinement of their expression domains. tll expression, for example, is specifically repressed in the segmented region of the embryo by central gap gene products such as Kr. Thus, depending on their relative levels of activity, Kr and Tll are both capable of suppressing one another's expression. This raises the possibility that centrally expressed Kr is responsible for the polar restriction of tll expression that is observed in the progeny of tsl mutant females expressing tsl from the germline. To address this question, the CBBtsl insertion was crossed into females that were mutant for all three anterior-posterior maternal pathways. Embryos produced by mothers triply mutant for bicoid (anterior), oskar (posterior), and tsl (terminal) lack all anterior-posterior patterning, express low levels of Kr uniformly along the anterior-posterior axis, and do not express tll at all. In contrast, the embryos produced by triply mutant females carrying CBBtsl did express tll in distinct polar domains, either at the anterior alone or at both poles. Consistent with this pattern of tll expression, Kr expression is specifically repressed in the corresponding polar domains. Further, these embryos also differentiate Filzkörper material, a posterior cuticular structure that requires terminal pathway activity, at one or both poles. Thus, although Tsl was distributed uniformly in their VMs, these embryos developed gap gene expression patterns and cuticular phenotypes consistent with polar activation of the Tor receptor. These findings suggest that the activity of Tsl is enhanced at, and perhaps restricted to, the polar regions of the VM; this finding implies that there is an as yet unidentified component of the terminal class pathway that is restricted to the poles and is required for the function of Tsl (Stevens, 2003).

The expression of tsl by the somatic follicle cells, and its role in patterning the embryo, can be thought of as an inductive event between the soma and the germline. However, the delay between the secretion of Tsl during oogenesis and the activation of the Tor receptor during embryogenesis necessitates stabilization of the localized signal. This is achieved by the incorporation of Tsl into the eggshell. The localization of Tsl on the inside of the VM allows it to be accessible to components of the perivitelline fluid, such as the Trk precursor, and the restriction of its activity to the poles of the VM limits the spatial parameters of Tor activation. This is the first demonstration of an embryonic patterning molecule associated with the eggshell (Stevens, 2003).

The development of the dorsal-ventral axis in Drosophila embryos also requires the transfer of patterning information from the soma to the germline, which leads to the asymmetric activation of a uniformly distributed receptor by a locally processed ligand that shares structural elements with the Trk ligand. It is intriguing to note that, like fs(1)ph and fs(1)N, one of the genes required for dorsal-ventral development, nudel, is required for the formation of the eggshell. Thus, these findings raise the possibility that spatial information for dorsal-ventral patterning may also be stored in the VM. However, none of the known genes in this pathway share homology with Tsl, which carries structural features, such as a membrane-attack complex/perforin domain, that may promote membrane interactions. Although the VM is not a classic lipid membrane, it is highly hydrophobic. These data indicate that even in embryos maternally mutant for fs(1)ph¹⁹⁰¹ or fs(1)N²¹¹, Tsl protein is still associated with the VM; this association demonstrates that Tsl has an affinity for the VM that is independent of its interaction with Polehole and Nasrat. An intriguing possibility is that Polehole and Nasrat are required to stabilize secreted Tsl such that it becomes incorporated into the VM in an active conformation (Stevens, 2003).

The existence of another localized factor in this pathway indicates that there are at least four levels of control that ensure the polar restriction of tll expression during embryonic development. (1) First to act is the restriction of tsl expression to a specific subpopulation of follicle cells present at the poles of the oocyte. (2) Next is the stabilization of secreted Tsl protein at the poles of the VM and its incorporation into the eggshell in an active form. (3)The facilitation of Tsl function follows, through its proposed interaction with another localized factor. (4)The final layer of control is the exclusion of tll expression from nonpolar regions through the inhibitory effects of centrally expressed gap genes. Although it has long been assumed that the spatial restriction of tsl expression was the uniquely localized element in the terminal pathway, data is presented implying the existence of another factor that enhances the activity of Tsl specifically at the poles. The function of Tsl itself is unknown, and there are currently no candidate genes encoding proteins with the enzymatic activity to bring about the proposed processing of the Trk precursor. It is likely, therefore, that the identification of this factor will greatly enhance understanding of the mechanism by which the Tor ligand is formed (Stevens, 2003).

GENE STRUCTURE

Bases in 5' UTR - 539

Exons - three

Bases in 3' UTR - 244

PROTEIN STRUCTURE

Amino Acids - 353

Structural Domains

tsl encodes a novel protein with a putative amino-terminal signal sequence (Savant-Bhonsale, 1993). The protein is basic, contains 11% leucine and 12% lysine and arginine residues. There is a unique cysteine residue that could serve for dimerization, and two C terminal leucine rich regions, that may serve to mediate protein-protein or protein-lipid interactions (Martin, 1994 and Savant-Bhonsale, 1993).

Evolutionary homologs

Maternal torso signaling controls body axis elongation in a short germ insect

In the long germ insect Drosophila, all body segments are determined almost simultaneously at the blastoderm stage under the control of the anterior, the posterior, and the terminal genetic system. Most other arthropods (and similarly also vertebrates) develop more slowly as short germ embryos, where only the anterior body segments are specified early in embryogenesis. The body axis extends later by the sequential addition of new segments from the growth zone or the tail bud. The mechanisms that initiate or maintain the elongation of the body axis (axial growth) are poorly understood. The terminal system in the short germ insect Tribolium was functionally analyzed. Unexpectedly, Torso signaling is required for setting up or maintaining a functional growth zone and at the anterior for the extraembryonic serosa. Thus, as in Drosophila, fates at both poles of the blastoderm embryo depend on terminal genes, but different tissues are patterned in Tribolium. Short germ development as seen in Tribolium likely represents the ancestral mode of how the primary body axis is set up during embryogenesis. It is therefore concluded that the ancient function of the terminal system mainly was to define a growth zone and that in phylogenetically derived insects like Drosophila, Torso signaling became restricted to the determination of terminal body structures (Schoppmeier, 2005).

In Drosophila, the anterior- and posterior-most terminal body regions of the embryo depend on the maternal terminal-group genes. One of them, the torso-like (tsl) gene is expressed in somatic follicle cells located at the anterior and posterior pole of the oocyte. In the embryo, tsl contributes to the local activation of the receptor tyrosine kinase Torso at the egg poles. The signal is transduced to the nucleus via a Ras-Raf-MAP-K/Erk phosphorylation cascade, and leads to the expression of the zygotic target genes tailless (tll) and huckebein (hkb) at the posterior terminus of the embryo. Failure to activate Torso signaling results in defects in the head skeleton and loss of all segments posterior to abdominal segment 7, in addition to loss of the hindgut and posterior midgut anlagen (Schoppmeier, 2005 and references therein).

Whether an anteriorly acting terminal system is a general feature of all insects has been challenged because under certain conditions, Torso function at the anterior is dispensable for head development in Drosophila. This hypothesis is supported by the expression of the Tribolium ortholog of tll at the posterior, but not at the anterior pole of blastoderm stage embryos. Thus, in Tribolium, posterior terminal cells appear to be determined before the onset of abdomen formation. It is unknown, however, whether these cells specify posterior fate after axis elongation and abdomen formation is completed or whether they also contribute to earlier steps of segmentation (Schoppmeier, 2005).

The orthologs of the key components of the Torso pathway have been isolated in the short germ beetle Tribolium torso (Tc-tor) and torso-like (Tc-tsl). As in Drosophila, Tc-torso mRNA is maternally inherited by the embryo and expressed ubiquitously in freshly laid eggs, and Tc-tsl is expressed during oogenesis anteriorly and posteriorly in the follicle cells of the oocyte (Schoppmeier, 2005).

Knocking down the function of Tc-torso or Tc-tsl using parental RNA interference leads to identical embryonic phenotypes. Whereas the head and the anterior thorax are unaffected, unexpectedly the most extreme Tc-torso^RNAi and Tc-tsl^RNAi embryos lack all structures that develop during postblastodermal abdominal growth. Thus, the head and thoracic segments that form in torso or tsl RNAi embryos likely represent the structures, which are determined already during the Tribolium blastoderm stage. Less strongly affected embryos fail to form the full number of abdominal segments (Schoppmeier, 2005).

To determine whether the Tc-torso RNAi phenotype does not reflect a late function of maintaining abdominal fate prior to cuticularization, the expression of Engrailed protein was examined in Tc-torso^RNAi embryos at a stage when abdominal segments should already have developed. Indeed, in strongly affected embryos, Engrailed stripes corresponding to the head and thorax, but not to abdominal segments, are present (Schoppmeier, 2005).

The emergence of segments was visualized in embryos with impaired Torso signaling by analyzing the Tc-even-skipped (Tc-eve) expression pattern. In wild-type embryos, Tc-eve is initially expressed in a double segmental pattern that later resolves into secondary segmental stripes. Tc-tsl ^RNAi does not interfere with the formation of the first two primary Tc-eve stripes that give rise to the gnathal and the first thoracic (T1) segments. However, although the third primary Tc-eve expression domain (Tc-eve stripe 3) forms normally, this domain does not resolve into segmental stripes, and no additional primary eve-stripes form. In the wild-type, Tc-eve stripe 3 covers the region where the second (T2) and third thoracic (T3) segment will develop. Although Tc-eve stripe 3 does not split in Tc-tsl ^RNAi embryos, this domain gives rise to the second thoracic segment. Thus, Torso signaling is required for the initiation of axial growth or maintaining the segmentation process (Schoppmeier, 2005).

As revealed by DAPI staining and by morphology, posterior invagination of cells is abolished in both Torso- and tsl ^RNAi embryos, and as a consequence, no posterior pit forms. To understand how Torso signaling is propagated at the posterior pole and to test whether downstream gene activity is affected in the growth zone in Tc-torso^RNAi embryos, the activity of the Map-kinase and the expression of Tc-wingless (Tc-wg), Tc-tailless (Tc-tll), Tc-caudal (Tc-cad), and Tc-forkhead (Tc-fkh) RNA was analyzed in early embryos (Schoppmeier, 2005).

The active state of the Torso receptor is transduced to the nucleus via the Ras-Raf signal transduction pathway and leads to the activation of zygotic target genes. The activity of this pathway can be visualized with an antibody that recognizes ErkPP but does not discriminate between the different pathways that involve ErkPP signaling. In nontreated embryos, ErkPP can be detected in a subpopulation of the serosa, a single row of cells at the border of the serosa and the embryonic anlage; at the rims of the mesoderm; and at the posterior pole. In torso^RNAi embryos posterior ErkPP expression is lost, further indicating that ErkPP is involved in propagating terminal signaling. ErkPP expression in the serosa is mildly affected whereas the other sites where ErkPP activity is detected in the wild-type are normal. ErkPP activity in the amnion appears not to be reduced; however, the amnion itself does also not form properly. Whether this is a direct or indirect consequence of Torso reduction is unclear (Schoppmeier, 2005).

In addition to segmental stripes, a terminal wingless (wg) expression domain first seen at the blastoderm stage is present throughout the phase of body elongation in the growth zone of the wild-type. In Tc-torso^RNAi embryos, the posterior terminal Tc-wg domain is missing at the blastoderm stage, as well as in older embryos corresponding in age to wild-type embryos undergoing body axis extension. Drosophila-torso mutant embryos also lack the posterior terminal wg expression domain, indicating, that the dependence of wg on torso is conserved. The segmental wg stripes that were built prior to the growth process, form close to the posterior end. This shows that a presegmented region (PSR) normally separating the last segment formed at the posterior end of the embryo is strongly reduced or absent in torso^RNAi embryos. The absence of Tc-cad and Tc-tll in torso^RNAi embryos establishes these genes as potential targets of terminal signaling also in Tribolium (Schoppmeier, 2005).

As judged from the lack of the posterior Tc-forkhead (Tc-fkh) expression domain, Tc-torso^RNAi embryos do not develop a hindgut. Tc-fkh itself, which is also expressed in the growth zone throughout most of the segmentation process, seems not to be required in the axis elongation process because Tc-fkh RNAi leads only to malformation of the hindgut (Schoppmeier, 2005).

The irregular expression of ErkPP in the serosa of embryos deficient for Torso signaling suggests a function for this pathway also in patterning this extraembryonic tissue. Indeed, in Tc-tsl^RNAi embryos, the serosa is severely reduced in size whereas the presumptive head region appears enlarged and extended toward the anterior. This finding is corroborated by the expanded expression domain of the head marker gene Tc-006A12 in Tc-tsl^RNAi embryos, which in wild-type embryos is expressed just posterior of the serosa in a wedge-shaped domain. Nevertheless, embryos develop with normal head structures, similarly as in embryos where serosa reduction results from reduction of zen-1 activity (Schoppmeier, 2005).

The anterior phenotype, however, differs among embryos depleted for either Tc-torso or Tc-tsl. Compared with the Tc-tsl RNAi, the serosa is less affected with Tc-torso RNAi, indicating that Tc-tsl has been more sufficiently downregulated via RNAi than Tc-torso (Schoppmeier, 2005).

Thus, in a short germ embryo Torso signaling is—as in Drosophila—required for patterning both the anterior and the posterior region of the embryo. Due to the difference in the anlagenplan of short and long germ embryos, however, different tissues and different processes depend on the Torso pathway (Schoppmeier, 2005).

This study has presented the first functional characterization of the terminal-class genes torso and torso-like outside the dipterans. At the anterior, Torso signaling is involved in the specification of the anterior-most structure in the Tribolium egg, the extraembryonic serosa. At the posterior, Tc-torso-like and Tc-torso are required for terminal fates, including the posterior gut primordial and for body axis growth (Schoppmeier, 2005).

These results show that the terminal system in Tribolium functions to establish the growth zone and to initiate rostrocaudal growth. In the absence of the terminal signal, caudal expression and pair-rule patterning is not maintained during later stages of development. It is not known how the loss of posterior patterning and the loss of posterior growth relate to each other. Since the growth zone specific expression domain of wg depends on torso one could speculate that the posterior domain of Tc-wg is required for both, regulation of cell proliferation and coordination of continued segmental patterning. The involvement of the Wg pathway in the axis elongation process has already been demonstrated for Gryllus (Schoppmeier, 2005).

Insects like Drosophila that develop as long germ embryos have short life cycles and are thus perfectly adapted to quickly changing environments like rotting fruits. During Drosophila embryogenesis, the posterior-most Engrailed stripe corresponding to abdominal segment 9 forms under the influence of Torso slightly later than the other stripes. This could be seen as a rudimentary elongation of the body axis in Drosophila and likely reflects the ancient function of Torso signaling (Schoppmeier, 2005).

In contrast to long germ embryos, the formation of the complete trunk occurs in a secondary growth process in short germ embryos of arthropods and vertebrates. It is proposed that the involvement of Torso signaling in body axis growth likely represents the ancient function of this gene in the development of short germ insects. Thus, during the evolution from short to long germ development, Torso signaling lost its major function in axial growth and was recruited as an additional gradient system to specify the position of posterior abdominal segmentation gene domains during the blastoderm stage. Whether Torso signaling is involved in body axis growth also in other short germ animals remains to be shown (Schoppmeier, 2005).

REGULATION

Promoter

Two distinct but convergent groups of cells trigger Torso receptor tyrosine kinase activation by independently expressing torso-like

Cell fate determination is often the outcome of specific interactions between adjacent cells. However, cells frequently change positions during development, and thus signaling molecules might be synthesized far from their final site of action. This study analyzed the regulation of the torso-like gene, which is required to trigger Torso receptor tyrosine kinase activation in the Drosophila embryo. Whereas torso is present in the oocyte, torso-like is expressed in the egg chamber, at the posterior follicle cells and in two separated groups of anterior cells, the border cells and the centripetal cells. JAK/STAT signaling regulates torso-like expression in the posterior follicle cells and border cells but not in the centripetal cells, where torso-like is regulated by a different enhancer. The border and centripetal cells, which are originally apart, converge at the anterior end of the oocyte, and both groups contribute to trigger Torso activation. These results illustrate how independently acquired expression of a signaling molecule can constitute a mechanism by which distinct groups of cells act together in the activation of a signaling pathway (Furriols, 2007; full text of article).

Although tsl is expressed in three different groups of follicle cells, these cells are not completely unrelated. Thus, for example, both the BCs and the CCs are derived from a common pool of anterior follicle cells and express and require some of the same genes for their development. Likewise, many similarities have also been recognized between the BCs and the PFCs. This raises the possibility that a common mechanism could single out these cells for tsl expression. Alternatively, each of these groups of follicle cells could be independently targeted to express tsl. As a first attempt to address how these distinct groups of follicle cells acquire the ability to express a common signaling factor, an analysis of the tsl promoter was undertaked (Furriols, 2007).

As a first indication of what constitutes the tsl regulatory region, the P-element insertion carrying the lacZ gene upstream of the 5'-UTR exon in the tsl⁰⁶¹⁷ mutant, thereafter tsl⁰⁶¹⁷-lacZ, was know to reproduced all of the features of tsl expression in the follicle cells, as judged by comparison with the tsl in situ hybridization pattern. By transformation of lacZ reporter constructs using different regions upstream of the coding sequences of tsl, it was found that a single fragment of ~1,500 bp upstream of the 5'-UTR exon (see Distinct Enhancers Regulate tsl Expression in Specific Groups of Follicle Cells) reproduces the tsl wild-type pattern. Further dissection allowed splitomg the tsl promoter into two nonoverlapping regions responsible for a different subset of the tsl expression pattern. In particular, it was found that a 604-bp sequence drives expression only in the CCs, hereafter referred to as the CC enhancer, whereas an adjacent 954-bp sequence drives expression in both the BCs and the PFCs. Comparison between the different constructs suggested that the enhancer for BCs and PFCs could be further refined to a region of 298 bp (fragment K). This assumption was confirmed by establishing that two copies of fragment K are sufficient to drive expression in BCs and PFCs, hereafter referred to as the BC/PFC enhancer. Thus, in summary, two different regions of the tsl promoter are responsible for distinct subsets of tsl expression. It is remarkable that a single promoter fragment (fragment K) drives tsl expression in two independent group of follicle cells (the BCs and PFCs), whereas separate enhancers (fragments K and F) are responsible for tsl expression in the BCs and CCs, which are derived from a common pool of anterior follicle cells (Furriols, 2007).

This study found that tsl expression is controlled by different cis-regulatory regions and different transactivating factors independently in different cell populations: a single promoter fragment responds to JAK/STAT signaling and activates tsl expression in both the BCs and PFCs, whereas another enhancer drives tsl expression in the CCs. Moreover, putative STAT binding sites (consensus TTCNNNGAA) were found in the identified BC/PFC enhancer. Mutations in those sites greatly reduce tsl-lacZ expression in the BCs and PFCs, pointing to a direct regulation by the JAK/STAT pathway. The fact that some reporter expression can occasionally be detected in those cells even when these sites are mutated could be attributed to regulation by other factors, which could also contribute to tsl expression in BCs and PFCs. In this regard, microarray analysis has shown that activity of the slbo transcription factor, which has been shown to function as a simple transcriptional activator and whose expression is also dependent on the JAK/STAT pathway, induces a 2-fold increase of tsl expression. In summary, these results show that the JAK/STAT pathway acts as a primary regulator of tsl expression in the BCs and PFCs (Furriols, 2007).

The JAK/STAT pathway is triggered in the Drosophila egg chamber by localized expression of its ligand, Upd, in two polar cells at each end of the chamber. Signaling from this pathway is responsible for the patterning of the follicle cells at both ends of the egg chamber, and the results show now that it is also responsible for tsl expression in the BCs and the PFCs. Thus, these results indicate that a common mechanism is responsible for initially patterning the egg chamber terminal epithelium and later triggering the mechanism that specifies the embryonic terminal regions (Furriols, 2007).

At the anterior end of the egg chamber, three populations of follicle cells can be distinguished: the BCs, the CCs, and the stretched cells in between. Among those, BCs and CCs, but not stretched cells, express tsl. Although the role of the JAK/STAT pathway in patterning the follicle cells at both ends of the egg chamber is well established, there are conflicting data about whether a gradient of its ligand, Upd, could indeed be responsible for patterning all of the anterior follicle cells. If that was the case, it might be expected that the JAK/STAT pathway could play a role in tsl expression in both the BCs and the CCs. In this scenario, absence of tsl expression in the stretched cells could be due to specific mechanisms of tsl gene repression in those cells. Conversely, the current results show that the JAK/STAT pathway does not have a specific role in the activation of tsl in the CCs. These results do not necessarily argue against a gradient of Upd. It could be argued, for example, that lower levels of JAK/STAT signaling in the CCs might not be sufficient to trigger activation of the BC/PFC enhancer. Alternatively, it could also be the case that a specific repressor element in this enhancer might inhibit its expression in the CCs. However, irrespective of a role of the Upd gradient in patterning the follicle cells, the results show that tsl expression in the CCs is independent of JAK/STAT. This result indicates that there are JAK/STAT-independent differences within the anterior epithelial cells of the egg chamber, as has been hypothesized (Furriols, 2007).

The results show that the two groups of anterior cells, the BCs and the CCs, contribute to trigger anterior Tor activation. Moreover, they indicate that this is accomplished by independent regulation of tsl in each of these cell populations. At first glance, either the BCs or the CCs appear to be sufficient to trigger Tor activation. Thus, GAL4-driven expression of tsl in either the BCs or the CCs is able to promote normal development of the terminal anterior structures in embryos derived from otherwise tsl mutant females. Additionally, RNAi-mediated inactivation of tsl in either the BCs or the CCs is not able to generate an anterior tsl phenotype, whereas inactivation in both the BCs and CCs produces embryos with anterior tsl mutant phenotypes. Thus, tsl expression in the BCs and CCs might be redundant. However, there are some caveats to those experiments that should be considered. First, GAL4-driven expression might generate higher tsl levels than the normal in the BCs or CCs. Second, in these experiments, RNAi-mediated inactivation does not completely impair tsl function; this is clearly observed because ~40% of the embryos develop anterior terminal structures even when UAStsldsRNA is expressed in both the BCs and the CCs using the C306 and 55B drivers (Furriols, 2007).

Given these results, it is proposed that an absolute level of tsl expression may be crucial to trigger Tor signaling. Therefore, it might not be so important whether tsl is supplied by the BCs or the CCs, provided it reaches an absolute amount. This would explain why overexpression of tsl in either the BCs or the CCs can rescue the anterior tsl mutant phenotype. It would also explain the additive effects of lowering tsl activity from the BCs and the CCs to generate an anterior tsl phenotype. Besides, it has to be considered that too much Tsl could also be damaging. In this regard, it has to be noted that tsl overexpression driven by the slboGAL4 driver produces head involution defects in many embryos. Taking this into account, expression of tsl from both the BCs and CCs could be a means to reach a minimum amount of Tsl product, but also not to exceed a certain limit (Furriols, 2007).

To understand how such a mechanism could have been established, the differences in ovary organization among insects should be considered. Although all insect ovaries consist of morphologically and physiologically discrete entities (the ovarioles), there are differences on how the oocyte is positioned in reference to the follicle cells. In more ancient insects, the oocyte is surrounded by a monolayer of somatic follicle cells. Conversely, in more evolved insects, a group of nurse cells are clustered at the anterior end of the oocyte and it is only later that the anterior side of the oocyte is separated from the nurse cells and contacts the follicle cells. In Tribolium, an insect with a more primitive ovary in which tsl expression has been examined, tsl is precisely expressed in the follicle cells overlying both edges of the oocyte. Therefore, Drosophila tsl expression in the BCs and PFCs may represent an adaptation, or the remnant, of a more ancient pattern of tsl expression. The difference in Drosophila is that the anterior tsl-expressing follicle cells, initially separated from the oocyte, have acquired the capacity to migrate through the nurse cells to reach the anterior end of the oocyte. Thus, two insects with different type of ovaries share a common pattern of tsl expression in two groups of follicle cells at both ends of the oocyte, although the mechanism to position these cells next to the oocyte differ in both insects. Conversely, tsl expression in the CCs of Drosophila appears to be a more recent acquisition. The CCs are a new particularly evolved set of follicular cells that migrate to separate the oocyte from the adjacent nurse cells. In this context, concomitant tsl expression in the CCs in Drosophila may have been independently attained by the acquisition of a new distinct enhancer in the tsl promoter (Furriols, 2007).

Therefore, the complex pattern of tsl expression could provide a means to ensure the full triggering and robustness of Tor receptor tyrosine kinase activation and illustrates a mechanism by which the full response of a receptor cell can be accomplished by the independent acquisition of signaling capacity in distinct cell populations and their combined action (Furriols, 2007).

Protein Interactions

Structural cell-surface and extracellular-matrix proteins modulate intercellular signaling events during development, but how this is achieved remains largely unknown. Identified here is a novel family of Drosophila proteins, Nasrat and Polehole, that coat the oocyte surface and play two roles: they mediate assembly of the eggshell, and act in the Torso RTK signaling pathway that specifies the terminal regions of the embryo. Nasrat and Polehole are essential for extracellular accumulation of Torso-like, a factor secreted during oogenesis that initiates Torso receptor activation. Stabilization of secreted factors by specialized pericellular proteins may be a general mechanism during signaling and developmental patterning (Jiménez, 2002).

fs(1)N has been mapped to chromosomal position 1E-1F. To begin the molecular identification of fs(1)N, new alleles of the gene were sought by mobilizing a collection of P-element insertions in the region. Three imprecise excisions of insertion EP(X)1336 were identified that behave as fs(1)N mutations, and one of them, designated fs(1)N¹⁴, was studied. fs(1)N¹⁴ behaves as a null allele that affects both eggshell integrity and terminal patterning: females homozygous for fs(1)N¹⁴ produce collapsed eggs that fail to develop, whereas fs(1)N¹⁴/fs(1)N¹² females produce embryos with the terminal phenotype associated with the fs(1)N¹² allele. Indeed, the latter embryos exhibit severely reduced tll and hkb expression at the blastoderm stage, similar to the effect of mutations in other components of the terminal pathway (Jiménez, 2002).

The EP(X)1336 insertion is located 200-bp upstream of a novel gene, CG11411, predicted by the Drosophila genome project. Molecular analyses showed that fs(1)N¹⁴ consists in a small deletion (<700 bp) of putative promoter and first exon sequences of CG11411. Also, a genomic construct of this gene rescues the phenotypes associated with fs(1)N mutations, demonstrating that CG11411 is fs(1)N (Jiménez, 2002).

A full-length fs(1)N cDNA clone was isolated by screening an early embryonic library. The sequence of this clone extends 65-bp upstream of the putative ATG-initiator codon and fits the predicted exon/intron structure of the gene. fs(1)N encodes a protein of 2118 residues rich in leucine residues (15%). Comparative analyses did not reveal significant similarities of Nasrat to known proteins. Nasrat contains a short hydrophobic stretch at the N terminus that probably acts as a signal peptide. No other putative transmembrane regions were detected, suggesting that Nasrat is secreted. In this regard, Nasrat contains a large number of potential glycosylation sites, as seen in many secreted and membrane-anchored proteins; there are 25 N-linked glycosylation motifs and five putative glycosaminoglycan (GAG) attachment sites in the protein. In addition, Nasrat has an ATP/GTP-binding site motif present in proteins with diverse functions (Jiménez, 2002).

The expression of fs(1)N in ovaries was examined by in situ hybridization. fs(1)N transcripts are detected in the nurse cells throughout most of oogenesis and in early blastoderm embryos, but not in the somatic follicle cells. This pattern of expression is consistent with a maternal function of fs(1)N in the germline as deduced from genetic analyses (Jiménez, 2002).

Next, Nasrat protein distribution was monitored in the ovary using a transgenic construct encoding a Flu-tagged (influenza hemagglutin YPYDVPDYA epitope) Nasrat derivative. This construct completely rescues the sterility of fs(1)N¹⁴ females, indicating that the Flu epitope does not interfere with Nasrat function. In early egg chambers (stage 5-6), Nasrat is found in the oocyte cytoplasm. In contrast, by stage 10 the protein localizes to the oocyte periphery. At high magnification, Nasrat is detected apically of cortical actin (visualized with rhodamine-phalloidin), suggesting that Nasrat lies on the external surface of the oocyte membrane. Moreover, Nasrat shows a finger-like pattern that probably reflects the microvilli formed by the plasma membrane of the oocyte. Actin staining was also seen of similar extensions from the plasma membrane of follicle cells, which appear to interdigitate with the oocyte microvilli. Finally, there is localized accumulation of Nasrat at the ring canals between follicle cells. This may result from undetectable fs(1)N expression in those cells, or internalization of Nasrat protein from the extracellular space. Because the primary site of fs(1)N function is the germline, the significance of this localization has not been studied further (Jiménez, 2002).

Significant similarities of Nasrat to the CG4790 gene product have been noted. The similarity is moderate but extends over a considerable length: 23% identity in a 700-amino acid overlap (residues 805-1495 of Nasrat and 430-1119 of CG4790 protein). Within this alignment there is a block of 200 amino acids with 25% identity. Because fs(1)N and fs(1)ph share many genetic features, it was considered that CG4790 might be fs(1)ph. Consistent with this hypothesis, CG4790 maps to chromosomal position 5C8-10 whereas fs(1)ph lies in the 5C5-D6 interval. It was confirmed that CG4790 corresponds to fs(1)ph using a CG4790 transgene that rescues the eggshell and terminal defects caused by fs(1)ph mutations (Jiménez, 2002).

A full-length fs(1)ph cDNA clone was obtained from an embryonic library. Its sequence is in agreement with the exon/intron annotations made for CG4790 by the Berkeley Genome Project, except for two differences in splicing sites that make the putative Polehole protein 43 amino acids longer than originally predicted. Like Nasrat, Polehole is rich in leucine residues (14%) and contains a putative signal peptide of 25 residues at the N terminus. Polehole also has a large number of potential glycosylation sites, including 26 N-linked glycosylation motifs and two GAG attachment sites. Database searches did not detect similarity of Polehole to proteins other than Nasrat, suggesting that these two proteins form a unique family (Jiménez, 2002).

The cellular localization of Polehole in the ovary was examined using an epitope-tag strategy. Flu-tagged Polehole construct was generated that rescued the sterility of females homozygous for fs(1)ph¹⁹⁰¹. As in the case of Nasrat, Polehole lies on the outer leaflet of the oocyte and outlines its microvilli. Thus, Nasrat and Polehole colocalize within a thin layer on the oocyte surface that establishes an intricate pattern of connections with the follicle cells (Jiménez, 2002).

To investigate this colocalization further, accumulation of Flu-tagged Nasrat and Flu-tagged Polehole was assayed in fs(1)ph^K646 and fs(1)N¹⁴ mutant ovaries, respectively. In both cases, the tagged derivatives are lost from the oocyte surface. Unlocalized Nasrat and Polehole proteins are not observed within the mutant oocytes, suggesting a defect in stability rather than transport. These results imply that Nasrat and Polehole are mutually required for their pericellular accumulation, which explains their similar mutant phenotypes. Whether this requirement is specific or reflects a general disorganization of the oocyte pericellular environment in the mutant ovaries was also examined. To this end, localization of Nudel, a protein required for dorsoventral patterning that associates to the oocyte surface, was also examined. Nudel shows a normal distribution in fs(1)N¹⁴ egg chambers, arguing that interdependent accumulation of Nasrat and Polehole is selective (Jiménez, 2002).

The above results suggest that Nasrat and Polehole function coordinately at the oocyte surface. To explore their role in eggshell assembly, the localization of Nasrat in relation to the nascent vitelline membrane was examined. Double staining for Nasrat and sV23 (Vitelline membrane 26Ab), a vitelline membrane component, showed early cytoplasmic accumulation of Nasrat and sV23 in the oocyte and the follicle cells, respectively. At stage 10, the two proteins marked complementary layers in the space between the oocyte and the follicle cells. These layers establish a remarkable pattern of connections that appear to reflect the interdigitating microvilli from both cell types. Weak sV23 staining on the oocyte surface was detected. The elaborate interrelation of Nasrat with the vitelline membrane is consistent with a role of this protein in eggshell biogenesis (Jiménez, 2002).

The distribution of sV23 protein in fs(1)N¹⁴ and fs(1)ph^K646 mutant ovaries, which in both cases simultaneously lack Nasrat and Polehole at the oocyte surface, was examined. sV23 remains unaffected in stage 10 egg chambers of both mutant backgrounds, indicating that sV23 is secreted and begins to accumulate independently of Nasrat and Polehole. This suggests that Nasrat and Polehole mediate subsequent steps of eggshell formation. Such steps include cross-linking modifications of vitelline membrane components that progressively render this membrane insoluble in reducing agents (e.g., DTT) and detergents. It has been shown that Nudel is required for cross-linking of sV23 and sV17 vitelline membrane proteins via nondisulfide bonds, which are insoluble in DTT. To test if Nasrat and Polehole also mediate these modifications, blastoderm embryos from either wild-type or fs(1)N¹⁴ females were extracted with 100 mM DTT and recovery of sV23 protein was assayed by Western blot. Whereas wild-type embryos do not contain soluble sV23 protein, a large amount of product is recovered from embryos laid by fs(1)N¹⁴ females. These results indicate that the combined activities of Nasrat and Polehole are required, directly or indirectly, for nondisulfide cross-linking of sV23 protein during oogenesis (Jiménez, 2002).

The fs(1)N¹² and fs(1)ph¹⁹⁰¹ alleles do not affect the eggshell but give rise to embryos that lack terminal structures. The molecular lesions associated to fs(1)N¹² and fs(1)ph¹⁹⁰¹ were identified: P830L and Y742N, respectively. To characterize the fs(1)N¹² mutation further, a Nasrat derivative was generated carrying a deletion of 161 amino acids encompassing P830 (Nasrat^Delta161). Nasrat^Delta161 failed to rescue both the collapsed egg and terminal phenotypes associated with different fs(1)N alleles, indicating that it lacks sequences required for both Nasrat functions. In contrast, a deletion of P830 and five adjacent residues (construct Nasrat^Delta6) disrupts only terminal patterning, suggesting that this motif specifically mediates terminal signaling (Jiménez, 2002).

What are the roles of Nasrat and Polehole in terminal signaling? Because Torso-like is the localized determinant for Torso receptor activation, the effects of loss of Nasrat and Polehole function on Torso-like distribution were investigated. Flu-tagged Torso-like protein is detected at stage 10 in the cytoplasm of posterior follicle cells. In addition, Torso-like accumulates in a posterior crescent between the follicle cells and the oocyte that probably corresponds to the secreted, active form of the protein. This signal forms a gradient that peaks at the posterior end, the site of Torso-like production. The effects on Torso-like distribution were examined of the fs(1)N¹² and fs(1)ph¹⁹⁰¹ mutations that specifically disrupt terminal cell signaling. In both cases, extracellular Torso-like protein is still present between the posterior follicle cells and the oocyte. However, the signal appears consistently weaker than in wild-type ovaries, suggesting that Nasrat and Polehole are required for efficient Torso-like accumulation. To test this idea, Torso-like was examined in fs(1)N¹⁴ ovaries, which lack both Nasrat and Polehole at the oocyte surface. These ovaries show barely any Torso-like protein between the posterior follicle cells and the oocyte; weak localized staining is still observed in some cases, but most egg chambers lack the extracellular signal. These results indicate that Nasrat and Polehole are essential for accumulation and/or stability of secreted Torso-like product (Jiménez, 2002).

Whether this requirement involves direct physical interactions between Nasrat, Polehole, and Torso-like was also examined. Although different interactions were observed, their specificity could not be demonstrated. For example, associations of Nasrat and Polehole with Torso-like were not prevented by the P830L and Y742N mutations. It may be difficult to prove relevant associations of Nasrat and/or Polehole with Torso-like if they require specific polysaccharide chains attached to these proteins, an attachment thought to occur during binding of cell surface proteoglycans to secreted proteins (Jiménez, 2002).

Nasrat and Polehole illustrate a role of cell surface molecules as common effectors of eggshell architecture and cell signaling events during development. To mediate these functions, Nasrat and Polehole promote their own accumulation at the oocyte surface, and also stabilize the Torso-like product deposited by follicle cells at each pole of the oocyte. It is still not known how these stabilizations occur molecularly, but one possibility is that Nasrat and Polehole function as protective molecules against unspecific degradation by extracellular proteases present between the oocyte and the follicle cells. Stabilization of secreted signals by cell surface molecules may be an important mechanism to ensure efficient activation of target receptors in other contexts. Indeed, recent studies have implicated cell surface proteoglycans in signaling by a variety of effectors such as the FGF, Hedgehog, TGF-ß, and Wnt proteins, raising the possibility that proteoglycans and related molecules promote accumulation of signaling products in these systems (Jiménez, 2002).

DEVELOPMENTAL BIOLOGY

Embryonic

See the embryonic expression pattern of tsl at the Berkeley Drosophila Genome Project Patterns of Gene Expression Site.

TSL mRNA first appears in the vitellarium at stage 8, with staining first observed at the anterior pole of the egg chamber in border cells, just prior to their migration, and at the posterior pole in the polar and flanking follicle cells. During development, tsl is expressed in the CNS and tracheal systems (Martin, 1994).

Effects of mutation or deletion

Genetic mosaic analysis has shown that tsl is required during oogenesis in follicle cells at each end of the oocyte. Ectopic expression of tsl produces embryos with a phenotype similar to that resulting from constitutively active tor alleles. (Savant-Bhonsale, 1993).

Loss-of-function mutations in tor and tsl cause an identical phenotype in which pattern elements from the anterior are missing (the acron or labrum), and the head skeleton is reduced in size. In the same mutants, posterior abdominal segment A8 and the telson have been deleted. Unrestricted expression of the TSL protein in tsl female mutants induces terminal pattern elements and suppresses the formation of abdomen in embryos (Stevens, 1990).

In Drosophila, the gradient of the Bicoid (Bcd) morphogen organizes the anteroposterior axis while the ends of the embryo are patterned by the maternal terminal system. At the posterior pole, expression of terminal gap genes is mediated by the local activation of the Torso receptor tyrosine kinase (Tor). At the anterior, terminal gap genes are also activated by the Tor pathway but Bcd contributes to their activation. Evidence is presented that Tor and Bcd act independently on common target genes in an additive manner. Furthermore, the terminal maternal system is shown not to be required for proper head development, since high levels of Bcd activity can functionally rescue the lack of terminal system activity at the anterior pole. This observation is consistent with a recent evolution of an anterior morphogenetic center consisting of Bcd and anterior Tor function (Schaeffer, 2000).

The terminal maternal system directly modifies Bcd by phosphorylation at several MAPK sites in a Ser/Thr (S/T)-rich region located between the homeodomain and the identified transcriptional activation domains. A deletion variant of Bcd that lacks all these activation domains but still contains the S/T-rich region (BcdDeltaQAC) is able to rescue to viability bcd loss-of-function mutants. Hence, it is conceivable that the ability of the tor pathway to create negative charges through phosphorylation of this region of Bcd might result in an acidic-rich transcriptional activation domain that compensates for the loss of all the other activation domains. If this were the case, then the transcriptional activity of the BcdDeltaQAC deletion variant should be highly dependent on tor function. To test this hypothesis, the ability of a BcdDeltaQAC transgene to rescue the bcd phenotype in embryos derived from bcd;tsl double mutant mothers was assayed. BcdDeltaQAC rescues the bcd phenotype of the bcd;tsl double mutant similarly to a wild-type bcd transgene, resulting in a tsl only phenotype. Since BcdDeltaQAC is functionally independent of the tor pathway, it is concluded that the terminal system is not responsible for BcdDeltaQAC's activation potential. This result is also consistent with the notion that, in transient transfection experiments and transgenic studies, Bcd transcriptional activity is not significantly modified by mutations of the putative MAPK consensus sites. Thus, the described direct modification of Bcd by the tor pathway does not appear to be necessary for Bcd's function (Schaeffer, 2000).

When a complete series of Bcd deletion variants was assayed for their ability to rescue the bcd loss-of-function phenotype in the absence of terminal system activity, one transgenic line was found that not only rescues the bcd phenotype but also the anterior part of the tsl phenotype (labrum and dorsal bridge), resulting in a posterior terminal mutant phenotype only. This particular transgenic line carries a bcd variant that deletes an alanine-rich domain (BcdDeltaA) and has been shown to activate the bcd target gene hb in a widely enlarged expression domain. Using Bcd immunostaining, it has been shown that this transgenic line exhibits levels of Bcd that are approximately 2- to 3-fold higher than wild type. Since other BcdDeltaA lines did not exhibit the same ability to rescue the tsl phenotype, it is concluded that the higher expression level of this particular line rather than the lack of a specific negative protein element (alanine-rich domain) is responsible for overcoming the requirement for the terminal pathway at the anterior (Schaeffer, 2000).

To further address whether high levels of bcd activity are sufficient to rescue the anterior terminal system phenotype or, if only a particular Bcd deletion variant is capable thereof, the ability of increased doses of wild-type bcd transgenes to rescue several terminal mutant backgrounds was tested. Since the previous experiments were performed with the tsl¹ allele, which might only represent a strong hypomorphic allele rather than a null, another tsl mutant, tsl⁴ , was included that is among the strongest in the allelic series, as well as null mutant alleles of the terminal genes trk and tor. To increase the Bcd expression level, flies containing an X chromosome or a third chromosome each carrying two wild-type bcd rescue constructs were used; these flies carry up to six copies of bcd. The phenotypes of all terminal mutants (tsl, trk or tor) are similar: lack of labrum and dorsal bridge in the anterior and deletion of all structures posterior to A7. Four copies of the bcd gene were able to rescue anterior structures including labrum and dorsal bridge in about 40% of all embryos derived from a tsl⁴ mutant background, while the posterior terminal phenotype is unaffected. Six copies of bcd are necessary to obtain the same anterior rescue in about 15% of all embryos derived from trk mutants and in about 5% of all embryos derived from tor mutants. However, not all embryos with rescued labrum and dorsal bridge had a perfectly aligned head skeleton. This might be due to incomplete rescue, but it could also be due to Bcd-mediated overexpression of hb at the anterior pole, which results in terminal-like phenotypes (Schaeffer, 2000).

Actually 50%, 70% or 85% of the head cuticles of tsl, trk or tor mutants, respectively, could not be analyzed for rescue due to severe anterior defects, which seemed more severe than normal terminal phenotypes. Nonetheless, some of the rescued embryos (less than 2%) were able to hatch and move around, which suggests complete anterior rescue. These probably represent embryos where just enough Bcd was present to overcome the lack of the terminal system but not too much to induce the phenotype due to high ectopic expression of hb. All larvae died within 2 hours, likely due to the posterior terminal defects. It should be noted that very few embryos exhibited the type of abdominal segment fusions that have been described for embryos derived from mothers carrying excess copies of the bcd gene. This might be due to the lack of terminal system function at the posterior pole in these experiments. Since no tail is made, there is probably more space for fate-map shifts towards the posterior, resulting in the correct establishment of abdominal segments A1 to A6. The rescue of the anterior terminal phenotype by high levels of bcd further indicates that the major role of the anterior terminal system is the potentiation of Bcd activity (Schaeffer, 2000).

In the posterior region of the embryo, the tor pathway activates the zygotic effectors tll and hkb, which are sufficient to specify the most posterior anlagen and the gut of the larva. At the anterior, the function of the terminal system is more difficult to interpret and, in tor mutants, hkb expression is only reduced. It actually requires bcd;tsl double mutants to lose all anterior hkb expression, which indicates additive functions of the anterior and terminal systems on this common target gene. hkb seems particularly interesting in this context, as its function is required for the formation of the labrum: reduction of hkb expression, as observed in terminal mutant background leads to the deletion of this particular structure (Schaeffer, 2000).

In Drosophila, the germline precursor cells, i.e. pole cells, are formed at the posterior of the embryo. As observed for newly formed germ cells in many other eukaryotes, the pole cells are distinguished from the soma by their transcriptional quiescence. To learn more about the mechanisms involved in establishing quiescence, a potent transcriptional activator, Bicoid (Bcd), was ectopically expressed in pole cells. Bcd overrides the machinery that downregulates transcription, and activates not only its target gene hunchback but also the normally female specific Sex-lethal promoter, Sxl-Pe, in the pole cells of both sexes. Unexpectedly, the terminal pathway gene torso-like is required for Bcd-dependent transcription. However, terminal signaling is known to be attenuated in pole cells, and this raises the question of how this is accomplished. Evidence is presented indicating that polar granule component (pgc) is required to downregulate terminal signaling in early pole cells. Consistently, pole cells compromised for pgc function exhibit elevated levels of activated MAP kinase and premature transcription of the target gene tailless (tll). Furthermore, pgc is required to establish a repressive chromatin architecture in pole cells (Deshpande, 2004).

That Bcd protein depends upon other ancillary factors to turn on transcription in pole cells is demonstrated by the requirement for tsl function in the activation of both the hb and Sxl-Pe promoters. tsl is a component of the maternal terminal signaling pathway that activates the zygotic genes, tll and huckebein (hkb), at the poles of the embryo. In addition, the terminal pathway has opposing effects on the expression of bcd-dependent gap genes. At the anterior pole, where terminal signaling activity is highest, Bcd targets such as hb and orthodenticle (otd) are repressed. At a distance from the anterior pole, where both the concentration of Bcd protein and the strength of the terminal signaling cascade is much lower, the terminal pathway has an opposite, positive effect on hb and otd expression. Two mechanisms are thought to account for the positive effects of the terminal pathway on bcd target genes: (1) Bcd is a direct target for phosphorylation by the terminal signaling cascade; (2) regulatory regions of bcd target genes have sites for other transcription factors whose activity can be directly modulated by the terminal system (Deshpande, 2004).

Spatially distinct downregulation of Capicua repression and Tailless activation by the Torso RTK pathway in the Drosophila embryo

Specification of the terminal regions of the Drosophila embryo depends on the Torso RTK pathway, which triggers expression of the zygotic genes tailless and huckebein at the embryonic poles. However, it has been shown that the Torso signalling pathway does not directly activate expression of these zygotic genes; rather, it induces their expression by inactivating, at the embryonic poles, a uniformly distributed repressor activity. In particular, it has been shown that Torso signalling regulates accumulation of the Capicua transcriptional repressor: as a consequence of Torso signalling Capicua is downregulated specifically at the poles of blastoderm stage embryos. Extending the current model, it is shown that activation of the Torso pathway can trigger tailless expression without eliminating Capicua. In addition, analysis of gene activation by the Torso pathway and downregulation of Capicua unveil differences between the terminal and the central embryonic regions that are independent of Torso signalling, hitherto thought to be the only system responsible for confering terminal specificities. These data provide new insights into the mode of action of the Torso signalling pathway and on the events patterning the early Drosophila embryo (de las Heras, 2006).

While the Tor pathway is normally activated only at the embryonic poles, tor constitutive mutations trigger its activation over the entire embryo in a ligand-independent manner. In these cases, expression of the tor target genes is expanded too much broader domains and embryos develop head and tail structures lacking most of the segmented trunk. According to the current model one would expect that tll domain expansion in these mutations would be accompanied by an expansion of the Cic downregulation domain (de las Heras, 2006).

Embryos from mutant females bearing the tor^D4021 constitutive mutation (a strong gain-of-function mutation that acts as a dominant female sterile) have been analyzed and instead it was found that Cic protein is still downregulated only at the poles, as in the wild-type embryos. Therefore, while in the wild-type the posterior tll domain is complementary to the domain of Cic accumulation, in embryos from tor^D4021/+females these domains overlap and tll is expressed in spite of the presence of nuclear Cic. This behaviour is not allele-specific since embryos from homozygous females for another tor constitutive mutation (tor^RL3) display the same kind of Cic distribution and tll expression (de las Heras, 2006).

It has been postulated that wild-type Tor receptors and Tor receptors activated by ligand-independent constitutive mutations could signal through distinct downstream effectors. Therefore, whether the persistent accumulation of Cic in embryos from tor constitutive mutant females could be due to a distinct property of these mutations was analyzed. Alternatively, the persistent Cic accumulation could reflect a difference in response between Tor activation in the middle versus the terminal embryonic regions. To test these possibilities, ligand-dependent activation of the Tor receptor was triggered over the entire embryo by general expression of the torso-like (tsl) gene. tsl is the only known gene in the Tor pathway whose expression is locally restricted. Indeed its restricted expression in a group of cells at each end of the developing oocyte is the determinant for the local activation of the Tor pathway, since its ectopic expression is sufficient to induce widespread activation of the Tor receptor. Accordingly, it was found that driving tsl expression with a tubGAL4 driver in the oocyte gives rise to an expansion of the tll expression domain and to the generation of embryos with a tor-gain-of-function phenotype, in that they develop head and tail structures and lack most of the segmented trunk. However, and similarly to what is described above for tor constitutive mutations, in these embryos Cic downregulation is not expanded to a broader domain, indicating that even ligand-induced activation of the Tor pathway is unable to inhibit Cic protein accumulation in the embryonic middle regions (de las Heras, 2006).

In the experiments described above, activation of the Tor pathway over the whole embryo did not result in an expansion of Cic downregulation. Paradoxically, activated Tor could trigger downstream targets in the middle region even though Cic was still present. These observations raise the question of whether under these circumstances Cic is still able to act as a transcriptional repressor. Alternatively, Tor signalling could impair cic activity without removing Cic protein from the nuclei. To address this issue, the contribution of cic function was analyzed in embryos from tor constitutive mutants (de las Heras, 2006).

The strong transformations associated with the ectopic activation of the Tor pathway due to tor^D4021 mutations and tubGAL4 driven expression of tsl make it difficult to assess the operational state of the Cic repressor under these circumstances. To overcome this difficulty use was made of the weaker tor^RL3 constitutive mutation and cuticular transformations, which are more sensitive to small changes in the expression of tor targets genes than what can be visualized by whole mount in situs, were scored. Besides, in the following experiments the tor^RL3 genotype was examined in a trunk (trk) background to eliminate ligand-induced activation. On its own, a single copy of tor^RL3 gives rise to a very mild phenotype, in which occasionally one abdominal segment is deleted. In contrast, removing just one copy of the cic gene does not affect the embryonic pattern. However, a single copy of the tor^RL3 mutation combined with the removal of just one copy of the cic gene gives rise to prominent transformations; embryos from such females display variable phenotypes but in every case they show major deletions of the embryonic segments. Accordingly, there is an expansion of the domain of tll expression, which also in that case overlaps with the domain where Cic accumulates. In this situation, whether nuclear Cic protein is still functional can be assessed by removing the remaining copy of the cic gene and comparing the two phenotypes. Indeed, embryos from trk tor^RL3/+; cic/cic have a much stronger phenotype that those from trk tor^RL3/+; cic/+. Therefore, the Cic protein present in trk tor^RL3/+; cic/+ embryos is still at least in part functional implying that the tor^RL3 mutation is able to trigger tll activation without eliminating all cic repression activity (de las Heras, 2006).

What mechanisms are activated by Tor signalling that could bypass the need for Cic downregulation to activate terminal target genes? It has been suggested that the Stat92E transcription factor plays a role as a mediator of Tor signalling elicited by a Tor constitutive mutant receptor, but not in Tor signalling promoted by ligand-dependent activation of the receptor at the poles. The role of Stat92E was assessed in the tor constitutive mutant background. A reduction was found in the transformations associated with the trk tor^RL3/+; cic/+ genotype by removing a single copy of the stat92E gene. Whether this could also apply in the case of ectopic activation of the Tor pathway through ligand binding was analyzed; also in this case it was found that there is a reduction of the strength of the phenotype. In this case, however, the reduction is smaller, which could be due to the fact that the original transformation generated by the tubGAL4/UAStsl combination is much stronger and/or to a weaker involvement of stat92E in ligand-induced Tor signalling. Regardless, the results suggest that there is no fundamental difference in the role of stat92E between ligand-induced or constitutive activation of the Tor receptor. In support of this conclusion there is the recent observation that Stat92E is specifically phosphorylated at the poles by ligand-induced Tor signalling. Therefore, similarly to what was observed in the embryonic middle regions, it is proposed that Tor could also induce tll activation in the poles, and this occurs by a Cic downregulation-independent mechanism via stat92E. Altogether these results suggest that Tor signalling could normally trigger tll expression at the poles of wild-type embryos by two kinds of regulatory mechanisms, relief of cic repression and positive activation of tll expression. The positive effect of Tor signalling on tll expression could have been obscured by the fact that there is also a still unidentified Tor-independent activator, since terminal fate is specified in embryos lacking both Tor signalling and Cic repression. Accordingly, it has to be noted that stat92E mutants suppress ectopic activation of tll in the middle embryonic regions but not tll activation at the poles, which suggests that the role of stat92E on Tor signalling could be somehow redundant at the poles but absolutely required when Tor signalling is triggered in the embryonic middle regions (de las Heras, 2006).

The following conclusions can be drawn from these results. First, while activation of the Tor pathway at the embryonic poles downregulates Cic, Tor signalling appears to be necessary but not sufficient to eliminate Cic protein, as it can do so only at the embryonic poles. In this regard, it has to be noted that recent results indicate that the posterior maternal system can also affect Cic downregulation. Second, impairment of Cic repressor function is not an absolute requirement for tll expression, since tll can be expressed in situations where Cic repressor is still functional. In this regard, tll expression appears to be the result of a balance between repressor and activator factors and Cic repression might be overcome provided that activation is enhanced. And finally, there are differences between the terminal and the central embryonic regions that are independent of Tor signalling, as judged by the spatially restricted capacity of the Tor pathway to inhibit Cic accumulation and by the apparently distinct regional redundancy of stat92E function in Tor-dependent patterning. These results suggest that the Tor signalling pathway is not the only system that establishes a difference between the terminal and the central regions of the Drosophila embryo (de las Heras, 2006).

REFERENCES

Casanova, J., et al. (1995). Similarities between trunk and spätzle, putative extracellular ligands specifying body pattern in Drosophila. Genes Dev 9: 2539-2544. 7590233

de las Heras, J. M. and Casanova, J. (2006). Spatially distinct downregulation of Capicua repression and Tailless activation by the Torso RTK pathway in the Drosophila embryo. Mech. Dev. 123(6): 481-6. 16753285

Deshpande, G., Calhoun, G. and Schedl, P. (2004). Overlapping mechanisms function to establish transcriptional quiescence in the embryonic Drosophila germline. Development 131: 1247-1257. 14960492

Furriols, M., Casali, A. and Casanova, J. (1998). Dissecting the mechanism of torso receptor activation. Mech. Dev. 70(1-2): 111-118. 9510028

Furriols, M., Ventura, G. and Casanova, J. (2007). Two distinct but convergent groups of cells trigger Torso receptor tyrosine kinase activation by independently expressing torso-like. Proc. Natl. Acad. Sci. 104(28): 11660-5. PubMed citation

Jiménez, G., González-Reyes, A. and Casanova, J. (2002). Cell surface proteins Nasrat and Polehole stabilize the Torso-like extracellular determinant in Drosophila oogenesis. Genes Dev. 16: 913-918. 11959840

Martin, J. R., Raibaud, A. Ollo, R. (1994). Terminal pattern elements in Drosophila embryo induced by the torso-like protein. Nature 367: 741-5

Montell, D.V., Rorth, P. and Spradling, A.C. (1992). slow border cells, a locus required for a developmentally regulated cell migration during oogenesis, encodes Drosophila C/EBP. Cell 71: 51-62. 1394432

Murphy, A.M., et al. (1995). The breathless FGF receptor homolog, a downstream target of Drosophila C/EBP in the develomental control of cell migration. Development 121: 2255-63. 7671793

Savant-Bhonsale, S. and Montell, D. J. (1993). torso-like encodes the localized determinant of Drosophila terminal pattern formation. Genes Dev 7: 2548-55

Schaeffer, V., et al. (2000). High Bicoid levels render the terminal system dispensable for Drosophila head development. Development 127: 3993-3999. 10952897

Schoppmeier, M. and Schr–der, R. (2005). Maternal torso signaling controls body axis elongation in a short germ insect. Current Biol. 15: 2131-2136. 16332539

Stevens, L.M. (1990). Localized requirement for torso-like expression in follicle cells for development of terminal anlagen of the Drosophila melanogaster embryo. Nature 346: 660-663. 2385293

Stevens, L. M., et al. (2003). The Drosophila embryonic patterning determinant Torsolike is a component of the eggshell. Curr. Biol. 13: 1058-1063. 12814553

date revised:  10 April 2008

The Interactive Fly resides on the
Society for Developmental Biology's Web server.