Injection experiments examine the role of Easter and the protease cascade in the establishment of D/V asymmetry

Embryonic dorsal-ventral polarity is defined within the perivitelline compartment surrounding the embryo by the ventral formation of a ligand for the Toll receptor. Here (as demonstrated by the transplantation of perivitelline fluid) are found three separate activities present in the perivitelline fluid that can restore dorsal-ventral polarity to mutant easter, snake, and spatzle embryos, respectively. These activities are not capable of defining the polarity of the dorsal-ventral axis; instead they restore structures according to the intrinsic dorsal-ventral polarity of the mutant embryos. They appear to be involved in the ventral formation of a ligand for the Toll protein. This process requires serine proteolytic activity; the injection of serine protease inhibitors into the perivitelline space of wild-type embryos results in the formation of dorsalized embryos (Stein, 1992).

Injection experiments involving the use of dominant active Easter (Chasan, 1992) and Snake, as well as injection of perivitelline (PV) fluid from dorsal mutant embryos into gd mutant embryos, have lead to production of ventral elements at the site of injection, rather than in the normal ventral region (Stein, 1992). These data suggest that D/V polarity is established by asymmetric presentation of the Toll ligand to the oocyte. PV fluid from dorsal mutant embryos (thought to be depleted of Spätzle ligand because of the presence of the Toll receptor) can rescue D/V polarity in snake and easter mutant embryos. This same PV fluid cannot restore normal ventral structures to either gd or nudel embryos. In contrast, injection of PV fluid from Toll mutant embryos (thought to contain active Spätzle ligand) into gd embryos produces ventral structures at the site of injection (Stein, 1992). The same fluid injected into snake or easter embryos produces embryos with normal polarity, independent of the site of injection (Stein, 1992). Because the temperature-sensitive period for both Nudel and Gd action includes a period before fertilization when the initial D/V asymmetry is known to be established, it is possible that these two gene products cooperate to form a localized anchor for a Spätzle activating complex (Konrad, 1998 and references therein).

Analysis of EaD protease activity

The easter gene, required for the development of all lateral and ventral pattern elements in the Drosophila embryo, appears to encode an extracellular serine protease. Dominant easter alleles increase the number of cells that give rise to lateral and ventral structures. Nine dominant and four recessive mutations are caused by single amino acid substitutions at conserved sites in the putative serine protease catalytic domain. The activity of dominant products was assayed by injecting in vitro synthesized transcripts from the dominant alleles into young embryos. The results suggest that the dominant easter products cleave the normal substrate, but fail to respond to a spatially asymmetric regulator (Jin, 1990).

The product of the Drosophila easter gene, a member of the trypsin family of serine proteases, needs to be more active ventrally than dorsally in order to promote normal embryonic polarity. The majority of the Easter protein in the embryo is present in the unprocessed zymogen form and appears to be evenly distributed in the extracellular space, indicating that the asymmetric activity of wild-type Easter must arise post-translationally. A dominant mutant form of easter that does not require cleavage of the zymogen for activity (ea delta N) is active both dorsally and ventrally. The ea delta N mutant bypasses the requirement for five other maternal effect genes, indicating that these five genes exert their effects on dorsal-ventral patterning solely by controlling the activation of the Easter zymogen. It is proposed that dorsal-ventral asymmetry is initiated by a ventrally-localized molecule in the vitelline membrane that nucleates an Easter zymogen activation complex, leading to the production of ventrally active Easter enzyme (Chasen, 1992).

Proteolytic activation of Spatzle requires the sequential action of four different members of thetrypsin family. The first protease in this pathway is encoded bynudel, which is expressed in the somatic follicle cells of the ovary (which secrete the eggshell), whereas the other three proteases are expressed by the germline cells. gastrulation defective (gd) is closely related to the trypsin family, but the protein encoded by gd lacks a number of amino acid residues crucial for protease activity, so its biochemical function is not clear. Downstream of gd is the protease encoded by snake, which acts upstream of Easter, the final protease known in this pathway. The Easter protease is likely to be the direct proteolytic activator of Spatzle (Misra, 1998 and references).

The sequential activities of these four members of the trypsin family of extracellular serine proteases are required for the production of the ventrally localized ligand that organizes the dorsal-ventral pattern of the Drosophila embryo. The last protease in this sequence, encoded by easter, is a candidate to activate proteolytically the ligand encoded by spatzle. Processed Easter is present in extremely low amounts in the early embryo making it difficult to detect. Nevertheless, proteolytic processing of a catalytically inactivated Easter can be detected in vivo. Thus proteolytically activated Easter can be detected only in mutants of easter that lack catalytic activity and not in wild type. Activated Easter is rapidly converted into a high molecular mass complex, which may contain a protease inhibitor. The complex has a molecular mass of 80 to 85 kDa, larger than the 50 kDa Easter zymogen. The high molecular mass band (which is called Ea-X because the nature of the protein complexed with Easter is unknown) is not present in embryos that lack easter mRNA, so it appears to be a novel form of the Easter protein. The Ea-X complex is very stable; it is not disrupted by boiling in SDS for 30 minutes (Misra, 1998).

The Ea-X complex is shown to contain the proteolytically processed Easter protein. Based on the distribution of cysteines in the wild-type Easter protein, the N-terminal pro-domain is expected to remain disulfide bonded to the C-terminal catalytic domain after zymogen activation. If the activated Easter in Ea-X is a two-chain protease with the pro-and catalytic domains associated by a disulfide bond, then wild-type Ea-X should migrate more slowly on non-reducing gels (where the pro-domain would remain associated with the complex), than on reducing gels, which would contain only the Easter catalytic domain and X. Under non-reducing conditions, Ea-X migrates more slowly than the mutant EaN-X (coded for by an N-terminal deletion mutant of easter in which the pro-domain is deleted and the signal sequence is fused directly to the catalytic domain), confirming that the wild-type Ea in Ea-X has been proteolytically cleaved at the zymogen activation site anddemonstrating that the pro-domain of the wild-type enzyme remains disulfide-bonded to the catalytic domain after zymogen cleavage (Misra, 1998).

Easter zymogen activation is controlled by a negative feedback loop from Dorsal, the transcription factor at the end of the signaling pathway. Mutations that block the intracellular signaling pathway leading to the activation of Dorsal do not block the formation of Ea-X. In fact, all these downstream mutants contain 3- to 5-fold more Ea-X than wild-type embryos. Even mutant embryos that lack spatzle mRNA and protein show increased levels of Ea-X, indicating that Ea-X is not a stable complex of Easter, with its putative substrate, Spatzle. In ventralized cactus mutant embryos, the amount of Ea-X is less than half the amount present in wild-type embryos. Since the amount of Ea-X reflects the amount of activated Easter, these results suggest that a feedback loop regulated by nuclear Dorsal acts back across the plasma membrane of the syncytial embryo to regulate activation of the Easter zymogen. The accumulation of Ea-X in downstream mutants during the early zygotic phase occurs at the same time that signaling through the pathway occurs. While the zymogen (unprocessed) form of Easter is present at fertilization, Ea-X does not appear until 1 hour after fertilization, at approximately the time that the Easter protein is active in the embryo and increased during the next 2 hours of development. As soon as Ea-X is detectable, there is more Ea-X in Toll mutants than there is in wild-type embryos. Thus, well before the final gradient of Dorsal is achieved, at about 2.5 hours after fertilization, the level of Easter processing appears to be modified by this feedback loop (Misra, 1998).

The most likely molecules to bind the active site are substrates and inhibitors. Because Ea-X is extremely stable and because Ea-X does not include Spatzle, its likely substrate, it is hypothesized that Ea-X is an inhibited complex and that the other component of Ea-X is an inhibitor. A serine protease inhibitor could play an important role in patterning by preventing the spread of Easter activity away from its initial site of activation. The Ea-X complex has the properties expected for a complex with a protease inhibitor of the serpin family. The serpins are a family of proteins that act as suicide substrates. Serpin suicide substrates are inhibitors that are cleaved by their target enzymes. Serpins interact with the serine that constitutes the activesite of serine proteases, thereby irreversibly inactivating the protease. Most serpins are 40-60 kDa in size, which would be consistent with the size difference between the predicted masses of the catalytic domains of Easter and Ea-X. As with other serpin-protease complexes, Ea-X formation requires the active site serine and the protease-bindingpocket. Also, like other serpin complexes, the Ea-X complex is extremely stable. Only a single cloned Drosophila serpin-encoding gene has been well characterized; this serpin is transferred from the male to the female during copulation and its function is not known. Additional serpins have been identified in other insects, making it very likely that Drosophila will have a number of different serpin-encoding genes. The reverse genetic approach of purifying the high molecular mass complex, microsequencing and gene cloning will ultimately make it possible to test whether the other component of Ea-X is a serpin or another protease inhibitor (Misra, 1998).

In order to address the mechanism underlying changes in the Dorsal gradient in eaD embryo, it was asked if the eaD mutations affect the production or inhibition of active Easter. Easter zymogen activation requires cleavage at a site between an N-terminal pro-domain and a C-terminal catalytic domain (Chasan, 1992). However, the cleaved protease domain is never detected in wild-type embryos. Instead, activated Easter is found in a stable complex called Ea-X that migrates as a 80-85 kDa band (Misra, 1998; Chang, 2002).

Extracts were prepared from embryos produced by ea83l/ea- and ea5.13/ea- females, as representative of the ventralizing and lateralizing alleles. In order to generate a size marker for the C-terminal catalytic domain, extracts were prepared from embryos laid by eaDeltaN/+ and ea8/ea- females. In the N-terminal deletion mutant eaDeltaN, the pro-domain is deleted and the signal sequence is fused directly to the catalytic domain (Chasan, 1992). The embryos laid by females carrying a P[eaDeltaN] transgene are weakly ventralized. In embryos carrying the recessive ea8 allele, production of the Easter catalytic domain is observed rather than the Ea-X complex (Misra, 1998; Chang, 2002).

Easter forms were detected on an immunoblot probed with anti-Easter antibodies. The amount of Easter zymogen correlates with the dose of easter; a higher level is observed in wild type compared with embryos laid by +/ea- females. As expected, the Ea-X band is present in the wild-type extract, appears more prominent in the eaDeltaN/+ embryo extract and is absent in the ea8/ea- embryo extract. In embryos produced by the eaD mutations, the Ea-X band is reduced in the ea83l/ea- extract and virtually absent in the ea5.13/ea- extract. Significantly, there is a corresponding increase in the level of C-terminal catalytic domain. The amount of the catalytic domain in the ea5.13/ea- extract appears comparable with the level in the ea8/ea- extract. These findings suggest that in the eaD mutations, zymogen activation produces a catalytic domain that fails to be or is only partially inactivated by the inhibitor X (Chang, 2002).

The production of the C-terminal catalytic domain has been observed only with catalytically inactive easter mutants, exemplified by the ea8 allele, a missense mutation located in the presumptive substrate binding pocket, and the eaS338A allele, in which the active site serine-338 was replaced with alanine (Misra, 1998). It was hypothesized that the C-terminal catalytic domain in the eaD alleles, in contrast to the recessive mutations, retain protease activity after zymogen activation. In order to test this model, EaD protease activity was assessed in two different experiments (Chang, 2002).

(1) The level of processed Spätzle was determined in eaD embryo extracts, because Easter protease activity generates processed Spätzle. No processed Spätzle is produced in ea- extracts; a higher level of processed Spätzle is observed in a transformant line carrying the eaDeltaN mutation. Extracts were prepared from embryos laid by eaD/+ and eaD/ea- females. Precursor and processed forms of Spätzle were detected on an immunoblot probed with antibodies specific to the Spätzle C-terminal domain. As expected, a lower level of processed Spätzle is observed in the +/ea- embryo extract compared with the wild-type embryo extract. Similarly, the level of processed Spätzle is lower in the eaD/ea- embryo extract compared with the corresponding eaD/+ embryo extract. In general, the amount of processed Spätzle in each of the eaD/ea- lanes appears roughly comparable with the level in the +/ea- lane. The exception to this generalization is observed with the lateralizing ea5.13 allele. The amount of processed Spätzle observed in extracts prepared from embryos laid by ea5.13/+ and ea5.13/ea- females is reproducibly higher than in wild type (Chang, 2002).

(2) An injection assay was used to ask whether wild-type and eaD activated Easter could be detected in the embryo. By transferring perivitelline fluid from one embryo to another, Stein (1991) characterized a ventralizing activity exhibiting properties expected of the Toll ligand. When this 'polarizing activity', later identified to be processed Spätzle, was injected into the perivitelline space of recipient embryos laid by pipe- females, the site of injection defined the ventral pole of a new axis (Stein, 1991; Stein, 1992; Schneider, 1994). It was reasoned that by using donor embryos laid by spz- females (and thereby removing the presumptive Toll ligand), it might be possible to detect the activity responsible for generating processed Spätzle, i.e. active Easter (Chang, 2002).

Perivitelline fluid transfer experiments were carried out using gastrulating donor embryos laid by different mutant females; perivitelline fluid was transferred into stage 4 recipient embryos laid by pipe- females. No axis-inducing activity was detected from donor embryos laid by either spz- or spz- Toll- females, suggesting that active Easter is either rapidly inactivated or sequestered in a non-transplantable complex. Furthermore, no activity was detected from donor embryos laid by ea83l spz- females. By contrast, axis-inducing activity was observed from donor embryos laid by ea5.13 spz- females. This observation not only provides functional evidence for Ea5.13 protease activity, but also demonstrates that the activity remains stable until gastrulation, many hours after zymogen activation (Chang, 2002).

In order to study the Spätzle cleavage reaction, precursor Spätzle and wild-type EaDeltaN proteins were co-expressed in cultured Drosophila S2 cells. Production of processed Spätzle was observed in the conditioned medium, with the level of cleavage dependent on the amount of EaDeltaN expressed. No formation of the inhibited EaDeltaN-X form was observed in these transfected S2 cells (Chang, 2002).

The effect of the ea83l and ea5.13 mutations on the cleavage reaction was analyzed. The level of Spätzle processing carried out by Ea83lDeltaN was indistinguishable from wild-type EaDeltaN, suggesting that the ea83l mutation does not affect Easter catalytic activity. By comparison, Ea5.13DeltaN exhibits weaker protease activity, although it appears to be expressed at higher levels. This result suggests that the lateralized phenotype produced by the ea5.13 mutation arises from two separate effects: (1) lack of proper inhibition following zymogen activation leads to loss of the dorsal zen domain, and (2) reduced Easter catalytic activity prevents formation of the ventral twist domain (Chang, 2002).

Protein Interactions

At least three of the dorsal group genes (snake, easter and gastrulation defective) encode secreted serine proteinases which probably function during early development in the perivitelline compartment of the embryo. The Easter proteinase is homologous in its light chain sequence to the hemocyte proclotting enzyme (PCE) of the Japanese horseshoe crab Tachypleus tridentatus. PCE is the terminal member of a proteolytic cascade activated in response to microbial polysaccharides and acts to cleave coagulogen, an invertebrate equivalent of fibrinogen. On the basis of this homology the overall primary structure of the easter proteinase, its mode of activation and its substrate specificity can be predicted. The result also suggests that Easter functions zygotically in hemocytes in a Drosophila defensive response analogous to that found in Tachypleus. The Toll receptor protein is absent in early cleavage embryos but accumulates rapidly at the syncytial blastoderm stage, the developmental stage at which its function is required. This finding suggests that translation of Toll mRNA is regulated in response to fertilization and egg deposition. These two observations are consistent with a model of dorso-ventral pattern formation in which a proteolytic cascade is activated uniformly in the perivitelline compartment of the embryo and causes the release of ventrally localised ligands of the Toll receptor. A possible alternative model in which a proteolytic cascade is activated in response to a ventrally restricted signal is also discussed (Gay, 1992).

Mutant alleles of snake were cloned and sequenced, revealing two types of lesions: point mutations that alter the amino acid sequence (snk073 and snkrm4) and point mutations that alter the splicing (snk229 or snk233) of intron 1 of the mRNA from the normal 3' end of the intron to a cryptic site. snake mutant embryos derived from homozygous mothers can be fully rescued by injection of RNA transcripts of the wild-type snake cDNA. RNA phenotypic rescue and site-directed mutagenesis experiments indicate that Snake requires the serine, histidine and aspartic acid of the catalytic triad for normal activity. Deletion experiments show that an acidic proenzyme domain is required for Snake rescue activity to be uniformly distributed throughout the embryo. A second proenzyme domain, called the disulfide knot, appears to be essential for normal regulation of Snake. Transcripts encoding only the proenzyme polypeptides of either Snake or Easter can dorsalize wild type embryos. A model is proposed in which the proenzyme determinants of both the Snake and Easter enzymes mediate interaction between the serine proteases and other components of the dorsal-ventral patterning system (Smith, 1994).

Easter and Snake both share significant structural homology with extracellular trypsin-like serine proteases. Experiments using dominant active forms of Snake and Easter show that Snake activates Easter, which in turn activates the Spätzle ligand. In combination with the somatically expressed genes, windbeutel, pipe, and nudel, Gd activates Snake in a location-dependent manner that marks the future ventral cells. The recent cloning of nudel indicates that Ndl is a large (350 kDa) extracellular glyco-protein with motifs suggesting that it might bind to extracellular matrix as well as to other proteins. Nudel also contains a serine protease catalytic domain. The occurrence of several protease-like proteins both upstream (Nudel) and downstream (Snake, Easter) of Gd may be responsible for the multiply processed Gd peptides observed on Western blots. The physical location of the Nudel protein (including whether it is incorporated as a component of the vitelline membrane) is not yet known, although the fragility of the ndl embryos suggests that Ndl may be required for stability of the vitelline membrane. Gd mRNA is expressed in follicle cells beginning at about stage 10 and may be graded in a ventral to dorsal manner, although it is expressed uniformly in the nurse cell/oocyte complex. The location of Gd remains to be determined. Four of the five currently identified genes needed for asymmetric activation of the Spz ligand encode secreted members of the serine protease superfamily. It is proposed that Gd functions as part of an anchored complex that triggers a proteolytic activation hierarchy involving Nudel, Snake, and Easter, resulting in localized activation of Spz ligand and asymmetry along the D/V axis (Konrad, 1998 and references).

To test whether Gastrulation defective (Gd), Snake, and Easter function in a proteolytic cascade, as suggested by genetic studies, their biochemical activities were examined by coexpression in Drosophila S2 cells. In many experiments, inactive forms of Easter and Snake were used in which the catalytic serine had been mutated to an alanine residue. This strategy was adopted because the active forms seemed to be unstable after zymogen cleavage and thus difficult to detect. To test first whether Gd activates Snake, the processing of catalytically inactive Snake (SNKS-A) was examined by Western blotting after it was expressed with Gd. Because previous studies had suggested that Gd activity depends on proteolytic processing, the activities of two truncated Gd forms were initially examined. GdDeltaN1 lacks amino acids 1 through 211 after signal sequence cleavage, where Lys-211 is the nearest basic residue N-terminal to the conserved catalytic domain and thus might serve as a cleavage site for Gd activation. GdDeltaN2 lacks amino acids 1 through 253, corresponding to a potential cleavage site at the beginning of the conserved catalytic-domain sequence. Expression of SNKS-A with either of these Gd forms resulted in the appearance of a 40-kDa C-terminal Snake polypeptide, approximately the size predicted for the Snake catalytic domain of 29 kDa plus a Myc-based C-terminal epitope tag of 9.5 kDa. In contrast, it was found that Snake cleavage does not occur when the Snake substrate contains a mutation in the zymogen-activation site. These results suggest that Gd cleaves Snake at its zymogen-activation site. This result suggests that Snake is cleaved by Easter within the prodomain, which could either enhance or inhibit further Snake activation by Gd as part of a positive or negative feedback loop (LeMosy, 2001).

Surprisingly, the Gd zymogen is also able to induce the cleavage of SNKS-A to generate the 40-kDa Snake catalytic domain, although at lower levels than those seen with the truncated Gd forms. This observation led to a test of whether coexpressing the zymogen forms of all three proteases could result in the activation of Easter. It was found that a catalytically inactive Easter substrate (EAS-A) is cleaved when expressed with both Gd and Snake but not when expressed with either Gd or Snake alone. EAS-A is cleaved to a 35-kDa form representing a C-terminal fragment, as judged by its reactivity with an antibody against a C-terminal FLAG-epitope tag. This C-terminal fragment migrates similarly to EADeltaN. In addition, mutation of the zymogen-activation site in EAS-A eliminates production of the 35-kDa polypeptide when the mutant protein is expressed with Gd and Snake. These results suggest that, when expressed with Gd and Snake, Easter can be cleaved at its zymogen-activation site and therefore activated (LeMosy, 2001).

To determine whether Snake is responsible for the cleavage of EAS-A in this experiment, EAS-A was expressed with either preactivated Snake (SNKDeltaN), or preactivated Gd (GDDeltaN1). SNKDeltaN promotes efficient conversion of EAS-A to the 35-kDa C-terminal form, whereas GdDeltaN1 is unable to promote cleavage of this substrate, supporting the conclusion that Snake is the protease that activates Easter. In these experiments, it was not possible to directly demonstrate that Easter becomes an active protease because the catalytically inactive EAS-A substrate was used. However, similar experiments using wild-type Easter provided indirect evidence that a proteolytically active Easter is generated. When the wild-type Easter zymogen was expressed with both Gd and Snake, but not either alone, a fragment of 32 kDa lacking the C-terminal FLAG tag was the most abundantly cleaved form of Easter detected, although the 35-kDa form could still be detected, albeit weakly. These results suggest that the active Easter protease is unstable and may undergo further processing in a reaction that requires its own proteolytic activity. In further support of the conclusion that both Snake and Easter become activated when expressed with the Gd zymogen, it was found that both the 40-kDa Snake catalytic-domain fragment generated by Gd and the 50-kDa Snake product generated by Easter could be detected in cells expressing all three wild-type zymogens (LeMosy, 2001).

These results suggest that the Gd zymogen has some activity against Snake, but that the processing of Gd gives rise to more active forms. To determine whether the Gd zymogen undergoes proteolytic processing when it promotes the activation of Snake and Easter, the processing of Gd was examined by Western blotting. When expressed alone, Gd exists predominantly as the full-length zymogen, although minor processed forms can be detected. However, when expressed with Snake or both Snake and Easter, two cleavage products at 46 kDa and 50 kDa are seen prominently, whereas little cleavage is seen if Gd is expressed with Easter. Both processed forms react with an antibody to a C-terminal HA tag present in the protein, indicating that they represent C-terminal fragments of Gd, and the 46-kDa cleavage product is similar in size to GdDeltaN1 (LeMosy, 2001).

The cleavage of Gd in these reactions depends on Gd's own catalytic activity as well as that of Snake. This result suggests that Gd can promote its own activation, perhaps through self-cleavage or positive feedback regulation involving the downstream proteases. To test whether Gd is cleaved by itself, Snake, or Easter, the catalytically inactive Gd substrate (GDS-A) was coexpressed with either preactivated GdDeltaN2, SNKDeltaN, or EADeltaN. Whether Gd is cleaved by Nudel's protease domain, NDL-PD was examined. All of the proteases appear capable of cleaving Gd. However, SNKDeltaN generates both 46- and 50-kDa forms of cleaved Gd, previously seen when the Gd, Snake, and Easter zymogens are coexpressed, whereas NDL-PD and GdDeltaN2 primarily produces the 46-kDa form and EADeltaN primarily produces the 50-kDa form (LeMosy, 2001).

Three serine protease zymogens -- Gastrulation defective (Gd), Snake (Snk) and Easter (Ea) -- and a nerve growth factor-like growth factor ligand precursor, Spätzle, are required for specification of dorsal-ventral cell fate during Drosophila embryogenesis. The proteases have been proposed to function in a sequential activation cascade within the extracellular compartment called the perivitelline space. Biochemical interactions between these four proteins have been examined using a heterologous co-expression system. The results indicate that the three proteases do function in a sequential activation cascade, that Gd becomes active and initiates the cascade and that interaction between Gd and Snk is sufficient for Gd to cleave itself autoproteolytically. The proteolytically active form of Ea cleaves Gd at a different position, revealing biochemical feedback in the pathway. Both Gd and Snk bind to heparin-Sepharose, providing a link between the pipe-defined ventral prepattern and the protease cascade. These results suggest a model of the cascade in which initiation is by relief from inhibition, and spatial regulation of activity is due to interaction with sulfated proteoglycans (Dissing, 2001).

nudel, a somatically required dorsal group gene, encodes a 320 kDa mosaic protein with a centrally located serine protease catalytic domain. It has been suggested that nudel might activate a dorsal-ventral proteolytic cascade. This possibility was investigated using the assay system by co-expressing either the full-length form of Nudel, or a constitutively active Nudel serine protease catalytic chain (ndlDeltan) in different combinations with the four proteins described here. In all experiments conducted thus far, no reproducible effect of Nudel or nudDeltan upon any of the other proteins nor an effect of any of the other proteins upon Nudel-specific polypeptides was observed. Consequently there is no evidence for direct biochemical interaction of Nudel with Gd, Snk, Ea or Spz and therefore no direct role for Nudel within the protease cascade can be ascribed. Nudel apparently does not directly activate the cascade but rather is required earlier for proper establishment or maintenance of the ventral prepattern (Dissing, 2001).

Gd appears to play a critical role in the proteolytic cascade since it can initiate the cascade yet does not appear to require classical zymogen activation in order to do so. The data suggest that exposure to the zymogen form of Snk is sufficient for Gd to become active and activate Snk, triggering the cascade, and for Gd to generate lower molecular weight polypeptides. With respect to a mechanism of activation, it is interesting to note that Gd bears some similarity to mammalian complement factors C2 and B. These proteases have novel activation mechanisms requiring complex formation and a conformational change as a prerequisite to activation. An alternative explanation for how Gd functions is that it has some intrinsic activity as a zymogen. Upon binding to and activating Snk, it then proteolytically processes itself to generate lower molecular weight inactive forms (Dissing, 2001).

Activated Ea can proteolytically process Gd, suggesting that a second form of feedback occurs within the cascade. Ea cleaves Gd at a novel position to generate a Gd polypeptide that is slightly larger than the predominant band generated by Gd itself. The significance of processing by Ea is not altogether clear from the data. However, it is reasonable to assume that cleavage by Ea is a way to modify Gd's biochemical properties. It is proposed that Ea may feed back negatively on the precursor form and/or the active form of Gd. This would provide a means of down-regulating the protease cascade to prevent amplification from 'running away', resulting in overproduction of the ventralizing signal (Dissing, 2001).

Since Gd can autoactivate the cascade, a requirement for an upstream protease to activate the cascade need not be postulated. Rather, the data might argue that a mechanism exists to prevent Gd from becoming activated too early in embryogenesis before the Toll receptor is completely expressed on the plasma membrane. This idea has been incorporated into a biochemical model of the cascade. The existence of an inhibitory factor that prevents Gd from activating Snk is proposed. The inhibitory factor must itself be inactivated in a spatially or temporally regulated way for proper activation of the cascade. However, the net result must be to permit the activation of Gd within the ventral perivitelline space at the proper time for accurate elaboration of the ventralizing signal. This region of the perivitelline space may correspond to the ventral stripe prepattern described from gd mRNA injection experiments (Dissing, 2001).

The ability of both Gd and Snk to bind to heparin-Sepharose suggests that their activities may be regulated in vivo by sulfated proteoglycans. Since pipe expression in somatic follicle cells is ventrally restricted and the gene encodes a heparan sulfate 2-O-sulfotransferase, sulfate modification of an as yet unknown proteoglycan may provide the ventral cue in the egg. Gd and Snk may interact directly with this sulfated proteoglycan and this interaction may provide the ventral restriction to activation of the cascade (Dissing, 2001).

It has been suggested that the protease cascade may enable an initial asymmetry in the form of a ventral stripe prepattern to be converted into a graded distribution of processed Spz ligand. The potential for both (1) amplification with subsequent steps and (2) feedback after activation could enable the cascade to self-regulate the shape of the Dorsal protein gradient. This property could provide plasticity in the patterning process and a means of compensating for minor variation in the size and shape of individual embryos. Such a mechanism would also be sufficiently adaptable that it could be conserved evolutionarily. The data suggest some remarkable similarities between the dorsal-ventral protease cascade and the classical complement and blood coagulation pathways (Dissing, 2001).

A serpin, apparently targeting Easter, regulates dorsal-ventral axis formation in the Drosophila embryo

Extracellular serine protease cascades have evolved in vertebrates and invertebrates to mediate rapid, local reactions to physiological or pathological cues. The serine protease cascade that triggers the Toll signaling pathway in Drosophila embryogenesis shares several organizational characteristics with those involved in mammalian complement and blood clotting. One of the hallmarks of such cascades is their regulation by serine protease inhibitors (serpins). Serpins act as suicide substrates and are cleaved by their target protease, forming an essentially irreversible 1:1 complex. The biological importance of serpins is highlighted by serpin dysfunction diseases, such as thrombosis caused by a deficiency in antithrombin. This study describes how a serpin controls the serine protease cascade, leading to Toll pathway activation. Female flies deficient in Serpin-27A produce embryos that lack dorsal-ventral polarity and show uniform high levels of Toll signaling. Since this serpin restrains an immune reaction in the blood of Drosophila, this study demonstrates that proteolysis can be regulated by the same serpin in different biological contexts (Ligoxygakis, 2003).

Dorsal-ventral (d-v) axis formation in Drosophila is initiated by a signal restricted to the ventral side of the embryo, which is conveyed through the action of three extracellular serine proteases to the transmembrane receptor Toll. The earliest acting protease is Gastrulation defective (GD), followed by Snake (Snk), and finally Easter (Ea), which appears to process the Toll ligand, the cytokine-like polypeptide Spaetzle (Spz), to its active form. In addition, a fourth serine protease Nudel (Ndl), provided by the follicle cells during oogenesis, is required to trigger the cascade. Through activated Spz, the signal is transduced via the Toll receptor to the Cactus/Dorsal complex, which corresponds to the IκB/NF-κB complex of vertebrates. Cactus is targeted for degradation and the NF-κB transcription factor Dorsal is free to enter the nucleus. This regulated nuclear transport leads to the formation of a broad Dorsal gradient, with highest nuclear concentrations in a narrow ventral region and rapidly declining concentrations in lateral positions. Dorsal target genes respond to different levels of the gradient and implement cell specification and morphogenesis across the d-v axis (Ligoxygakis, 2003).

The spatially restricted activation of the protease cascade apparently depends on the modification of an as yet unknown somatically expressed heparan sulfate proteoglycan (hspg). It is presumed that the gene pipe, which encodes a heparan sulfate 2-O-sulphotransferase, mediates this modification. The pipe gene is the only one in the Toll pathway that has a ventrally restricted pattern of expression, suggesting that it provides the initial spatial cues for the embryonic d-v axis. Females that lack the activity of any of the protease genes (spz, Toll, pipe, or dorsal) produce dorsalized embryos in which cells at all positions give rise to dorsal ectoderm (Ligoxygakis, 2003).

It has been hypothesized that the extracellular protease cascade leading to the formation of the Dorsal gradient is kept in control of excessive or ectopic activation by a serpin (Misra, 1998). Flies lacking the secreted Drosophila serpin, Serpin-27A (Spn27A), have been generated by P element excision mutagenesis. Embryos derived from such females are severely ventralized, as expected from a spatially unrestricted activation of the Toll pathway. Dorsal is uniformly nuclear in all embryonic cells, and as a consequence, mutant embryos lack d-v polarity. As seen from the uniform expression of snail and the absence of rhomboid expression, all cells adopt the ventralmost cell fate, the presumptive mesoderm. Since the embryos lack ectodermal cell fates, they do not secrete cuticle and produce empty egg cases. A hypomorphic P allele of Spn27A reveals how the Dorsal gradient is affected by reduced Spn27A levels. The region of peak levels of nuclear Dorsal expands, while the lateral regions of declining concentration are shifted toward the dorsal side. Low levels of Dorsal are maintained in dorsalmost positions. Accordingly, the snail domain expands to cover 60% of the embryonic circumference, while the rhomboid domains are shifted toward the dorsal side. Thus, reduced Spn27A levels are not linked to pathway activation in dorsal positions but, rather, of an expansion of the ventral region with peak activity (Ligoxygakis, 2003).

Injection of Spn27A RNA into embryos lacking Spn27A restores polarity and differentiation as evidenced by the observation that approximately 60% of the embryos developed a normal cuticle pattern. This result confirms that the Spn27A phenotypes are due to the lack of Spn27A. One fifth of the injected embryos were weakly dorsalized, indicating that an excess of Spn27A is able to counteract the process of axis induction. However, a mutant form of the serpin with the putative protease cleavage site mutated lacks rescue activity. This provides further evidence that Spn27A indeed interacts with a protease and, thus, fulfils the predicted biochemical function as an inhibitor of the cascade (Ligoxygakis, 2003).

Since loss of Spn27A function results in apolar embryos, it was wondered whether the site of injection of the Spn27A RNA determined the polarity of the future d-v axis. Depositing the RNA to either the dorsal or the ventral side of the embryo led to the same degree of rescue. The polarity of gastrulation was always identical to that of wild-type embryos. This indicates that asymmetries in the initial distribution of Spn27A cannot reorient the d-v axis and that triggering of the cascade remains strictly dependent on the ventral cues provided by pipe expression (Ligoxygakis, 2003).

To further establish the role of Spn27A in the Toll pathway, the epistatic relation to different genes involved in d-v axis formation was explored. Double mutants with three proteases (GD, Snk, and Ea) as well as with spz were produced. If Spn27A is an inhibitor of the cascade, all loss-of-function mutants mentioned above should repress the ventralized phenotype because of a lack of ventral signal. This is indeed the case with one notable exception. The gd; Spn27A double mutant leads to a lateralized phenotype. The embryos express ventral-lateral or dorsal-lateral cell types around their entire circumference, indicating a uniform intermediate or low level of pathway activation. This suggests that the cascade has a basal activity in the absence of Gd, which is uncovered by removing the serpin. Alternatively, the proteins produced by the two mutant gd alleles tested could have residual function. However, this is unlikely to be the case for gd9, which carries a stop codon producing a truncated nonfunctional protein variant (Ligoxygakis, 2003).

It is known that complement and coagulation pathways can be negatively regulated by serpins bound to heparan sulfate proteoglycans (hspgs). Spn27A might require binding to an hspg for its activity and the ventral Pipe-dependent modification of this hspg might prevent Spn27A binding. This, in turn, might allow the cascade to be active at the ventral side. However, if pipe would act only via negative control of the serpin, the Spn27A; pipe double mutant should be ventralized like the Spn27A mutant alone. This is not the case. Spn27A; pipe females produce dorsalized embryos identical to those produced by pipe mutants. This indicates that Pipe-dependent hspg modifications do not act via Spn27A but, rather, have a positive input on the protease cascade. This is in line with the injection data showing that the asymmetry provided by pipe cannot be overridden by an artificially produced asymmetry in the Spn27A distribution. Together, these results indicate that the serpin acts as an inhibitory 'sink' that prevents the spreading of cascade activation to lateral and dorsal regions of the embryo (Ligoxygakis, 2003).

Earlier studies have presented indirect biochemical evidence for a serpin regulating the Ea protease. Since Spn27A is maternally expressed and secreted, as is Ea, the ability of Spn27A to inhibit Ea was tested in vivo using the UAS/GAL4 system (Ligoxygakis, 2003).

In addition to d-v axis formation, the Toll signaling pathway controls Drosophila host defense against gram-positive bacteria and fungi. Again, Spz is proteolytically cleaved to an active form by the serine protease Persephone, and signal transduction culminates in the induction of the antifungal peptide gene Drosomycin by way of NF-κB-dependent transcription in the fat body, the Drosophila analog of the mammalian liver. Although Ea is not involved in the cascade relaying the infection signal, overexpression of an activated form of Ea (ΔEa) results in infection-independent constitutive expression of drosomycin. When Spn27A is coexpressed with ΔEa, induction of drosomycin in adult fat body cells is blocked. The most plausible explanation for the above result is that of a direct interaction of the two proteins (Ligoxygakis, 2003).

Therefore, a serine protease inhibitor (Spn27A) provides an important control element in the proteolytic cascade leading to d-v axis establishment in early Drosophila embryogenesis. The following model accommodates these findings. The triggering of the d-v patterning cascade occurs in a limited area at the ventral side of the embryo in the same manner as the mammalian blood-clotting cascade normally generates a blood clot only at the site of the injury. Spn27A is required for this process of spatial restriction. Since Spn27A is uniformly distributed throughout the surface of the oocyte, it could function in preventing any misfiring of the cascade in inappropriate places. This is particularly relevant with regard to the observation that the downstream proteases have a basal activity in the absence of GD, which (as evident from the gd; Spn27A double mutant embryos) has to be kept in check by the serpin. The fact that the Spn27A; pipe double mutant females produce dorsalized embryos indicates that the Pipe-modified proteoglycan provides a positive input on the protease cascade, which locally enhances its activity. The target of this input is still unknown. However, it might be neither Ndl nor Gd, but one of the downstream proteases. The data indicate that Spn27A exercises its control at the level of Ea. The Pipe-dependent enhancement of the protease cascade may generate sufficient amounts of active Ea to target the serpin and thus overcome inhibition. This in turn may initiate a positive feedback loop between Ea and Gd and thereby trigger the full activity of the cascade. The existence of feedback loops within the cascade from Gd to Ea has recently been described (Ligoxygakis, 2003).

It is interesting to note that a reduction in Spn27A levels does not lead to uniform low-level triggering of the cascade but, rather, to an expansion of the ventral peak levels of activation. The cascade remains inhibited in the dorsalmost regions and a steep decrease of nuclear Dorsal can still be observed. This indicates that normal Spn27A levels are critical in restricting cascade activation to the pipe domain, but not crucial for determining the slope of the Dorsal gradient. Dorsal gradient formation apparently requires another inhibitory process directly linked to the activity of the cascade. Such a process has recently been described. The cleavage of Spz provides not only the ligand for the receptor Toll, but also a proteolytic fragment that apparently inhibits the cascade. It will be interesting to analyze at the biochemical level how the interplay between the serpin- and Spz-mediated inhibitory processes restrict and shape Dorsal gradient formation (Ligoxygakis, 2003).

Spn27A has been shown to control localized melanization of pathogen surfaces. This study has presented evidence of Spn27A involvement in the control of a proteolytic cascade in a developmental context. This is the first case that such a dual role is documented for a serpin. The paradigm of Spn27A could serve as a model to study the control of proteolysis in different biological contexts (Ligoxygakis, 2003).

An extracellular serine protease cascade generates the ligand that activates the Toll signaling pathway to establish dorsoventral polarity in the Drosophila embryo. This cascade is regulated by a serpin-type serine protease inhibitor, which plays an essential role in confining Toll signaling to the ventral side of the embryo. This role is strikingly analogous to the function of the mammalian serpin antithrombin in localizing the blood-clotting cascade, suggesting that serpin inhibition of protease activity may be a general mechanism for achieving spatial control in diverse biological processes (Hashimoto, 2003).

In order to explicitly test the hypothesis that a serpin is involved in spatially regulating the Easter protease, by analogy to the role of antithrombin in blood clotting, the Drosophila genome was searched for candidate serpins. A serpin that inhibits the extracellular Easter protease should have, in addition to the C-terminal reactive center loop sequence characteristically found in known inhibitory serpins, a basic residue at the predicted reactive site to match the amino acid specificity of Easter and an N-terminal signal sequence for secretion. Eight serpins were identified that fulfilled all three criteria, out of about 25 encoded in the genome. The predicted reactive sites of two serpins, Spn1 and Spn27A, additionally showed provocative sequence similarity to the cleavage site of Spätzle, the Easter substrate (Hashimoto, 2003).

To determine whether any of the eight candidate serpins could inhibit Easter, their inhibitory activity was assessed in a cultured cell assay involving coexpression of the Easter catalytic domain and Spätzle. Both Spn1 and Spn27A efficiently blocked Easter cleavage of Spätzle, while the other six candidates had no appreciable effect. Based on these results, Spn1 and Spn27A emerged as the best candidates for a natural inhibitor of Easter (Hashimoto, 2003).

To investigate the role of Spn1 and Spn27A in regulating Easter in vivo, the genetic consequences of removing maternal serpin activity were examined. For Spn27A, an apparently protein null mutation has been generated to assess the zygotic role of Spn27A in regulating the melanization reaction during the immune response. Therefore, this mutation was used to remove maternal spn27A function and the resulting phenotype was characterized by scoring embryos for the expression of dorsoventral zygotic genes. In the wild-type embryo at the blastoderm stage, the zen gene is expressed in a dorsal domain, the rho gene in two ventrolateral stripes, and the twi gene in a ventral domain. By contrast, embryos lacking maternal spn27A function show a striking expansion of twi expression across the entire dorsoventral axis, with a corresponding loss of rho and zen transcription. In addition, the mutant embryos failed to differentiate a cuticle at the end of embryogenesis, consistent with the interpretation that all cells had adopted the ventral-most mesodermal fate, as dictated by uniform twi expression. Finally, the ventralized phenotype was completely rescued by injection of embryos with in vitro synthesized spn27A RNA. This result demonstrates that the mutant phenotype was caused by the loss of spn27A function, and is consistent with a requirement for spn27A in germline rather than somatic tissue. The genetic characterization and rescue experiments together demonstrate that the serpin Spn27A is essential for establishing embryonic dorsoventral polarity (Hashimoto, 2003).

The strongly ventralized phenotype produced by the loss of spn27A function requires wild-type easter activity, consistent with the interpretation that the Spn27A protein acts to regulate Toll signaling rather than another pathway important for establishing embryonic dorsoventral polarity. Females lacking spn27A and either easter or spätzle function produce dorsalized embryos lacking all ventral and lateral structures, indistinguishable from the phenotype produced by the easter or spätzle mutation alone (Hashimoto, 2003).

Although it has not yet been possible to examine the role of Spn1 in embryonic dorsoventral patterning, the nearly complete ventralization caused by loss of Spn27A suggests that Spn1 is not functionally redundant with Spn27A. Its ability to inhibit Easter activity in vitro may therefore indicate that the natural target of Spn1 is a protease sharing substrate specificity with Easter (Hashimoto, 2003).

The ventralized phenotype observed with the loss of maternal spn27A function implies that regulation of Easter following zymogen activation is required for maintaining embryonic polarity. If activated Easter were capable of diffusion, Spn27A may primarily act to maintain the initial asymmetry of zymogen activation, by inhibiting Easter before its diffusion to the dorsal side of the embryo. Alternatively, if Spn27A were itself localized to the dorsal side, it could be providing an opposing gradient of a signaling antagonist. In fact, the following experiments support the first model. First, embryos lacking Spn27A can be completely rescued by injection of cultured cell medium containing Spn27A protein into the perivitelline space surrounding the embryo, irrespective of whether injection occurred on the dorsal or the ventral side. This result demonstrates that Spn27A acts in the same extracellular compartment where Easter functions, and suggests that there is no requirement for Spn27A to be prelocalized to a specific region along the dorsoventral axis. Second, Spn27A was detected in perivitelline fluid extracted from embryos, thus providing evidence that Spn27A is a soluble and diffusible protein. Finally, by immunostaining, Spn27A was detected across the entire dorsoventral axis of the embryo. Thus, Spn27A appears to be a circulating component of the perivitelline space surrounding the embryo (Hashimoto, 2003).

In conclusion, these experiments demonstrate that the serpin Spn27A is essential for spatially regulating the signal that defines embryonic dorsoventral polarity in Drosophila. This role for Spn27A reveals another link between development and immunity. The Toll signaling pathway was discovered for its role in Drosophila embryonic development, but is now also appreciated as a key defense mechanism against pathogens in the innate immune systems of both insects and mammals, for example, activating the production of antifungal peptides in Drosophila. Spn27A was first described to have a zygotic role in regulating activation of the melanization reaction during the immune response, and now has been discovered to have a maternal role in regulating activation of the Toll signaling pathway during embryonic patterning. In the melanization reaction, Spn27A presumably regulates the protease that activates phenol oxidase, a key enzyme in melanin biosynthesis. This protease may be distinct from Easter, since easter mutant flies do not show any gross defect in their ability to mount a melanization reaction at the site of tissue injury. Interestingly, it appears that in development and in immunity the same ligand, processed Spätzle, activates the Toll signaling pathway, yet distinct serine protease cascades and serpins regulate the processing of Spätzle. The data suggest that the role of Spn27A in establishing embryonic dorsoventral polarity is to control the spatial distribution of Toll signaling. Although its target, the Easter protease, is apparently only activated on the ventral side of the embryo, this initial asymmetry is not sufficient to establish axial polarity. As a circulating component of the extracellular fluid surrounding the embryo, Spn27A acts either by controlling the level of active Easter on the ventral side or by preventing diffusion of active Easter toward the dorsal side, thereby ensuring that the Toll ligand is ventrally restricted. In the absence of Spn27A, excess Toll ligand not bound to receptor or active Easter itself spreads toward the dorsal side, ultimately resulting in nearly uniform activation of Toll signaling that ventralizes the embryo. Conversely, increased Spn27A inhibits activated Easter before it cleaves enough Spätzle precursor, leading to insufficient Toll signaling that dorsalizes the embryo. These studies reveal that an active mechanism for preventing Toll activation on the dorsal side of the embryo is required to establish embryonic dorsoventral polarity and depends on a critical level of Spn27A. More generally, the role of Spn27A in localizing a serine protease cascade that generates a developmental signal is very analogous to the role of the mammalian plasma serpin antithrombin in confining the blood-clotting cascade to the site of vascular damage. This striking parallel demonstrates how serine protease cascades and serpins are used to exert spatial control in two distinct biological processes that both rely on posttranslational mechanisms (Hashimoto, 2003).

Spatially dependent activation of Easter

The dorsoventral axis of the Drosophila embryo is established by the activating cleavage of a signaling ligand by a serine protease, Easter, only on the ventral side of the embryo. Easter is the final protease in a serine protease cascade in which initial reaction steps appear not to be ventrally restricted, but where Easter activity is promoted ventrally through the action of a spatial cue at an unknown step in the pathway. In this study, biochemical studies demonstrate that this spatial control occurs at or above the level of Easter zymogen activation, rather than through direct promotion of Easter's catalytic activity against the signaling ligand (LeMosy, 2006).

The results establish that spatial control by pipe occurs at or above the level of Easter zymogen activation, rather than through direct promotion of Easter's catalytic activity against Spätzle. While Easter zymogen activation could be the first spatially controlled reaction, a complex interaction between GD and Snake has been shown to present a particularly attractive target for regulatory control. For example, an engineered auto-activating allele of Snake still requires function of the GD pro-domain for signaling, and a ventral region of the perivitelline space appears to be more sensitive to levels of GD introduced by microinjection than is the dorsal side. GD and Snake appear to interact in a way that stimulates GD catalytic activity in vitro, so it is possible that a ventral cofactor enriching the local concentrations of these components could greatly enhance the cascade. Further biochemical studies of GD and Snake processing in vivo should aid in defining the mechanism of spatial control in this developmentally important protease cascade (LeMosy, 2006).

easter: Biological Overview | Developmental Biology | Effects of Mutation | References

Home page: The Interactive Fly © 1997 Thomas B. Brody, Ph.D.

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