short gastrulation

Promoter

Genome-wide analysis of clustered Dorsal binding sites was used to examine the distribution of Dorsal recognition sequences in the Drosophila genome. The homeobox gene zerknullt (zen) is repressed directly by Dorsal, and this repression is mediated by a 600-bp silencer, the ventral repression element (VRE), which contains four optimal Dorsal binding sites. The arrangement and sequence of the Dorsal recognition sequences in the VRE were used to develop a computational algorithm to search the Drosophila genome for clusters of optimal Dorsal binding sites. There are 15 regions in the genome that contain three or more optimal sites within a span of 400 bp or less. Three of these regions are associated with known Dorsal target genes: sog, zen, and Brinker. The Dorsal binding cluster in sog is shown to mediate lateral stripes of gene expression in response to low levels of the Dorsal gradient. Two of the remaining 12 clusters associated with genes that exhibit asymmetric patterns of expression across the dorsoventral axis. These results suggest that bioinformatics can be used to identify novel target genes and associated regulatory DNAs in a gene network (Markstein, 2002).

zen is an immediate target of the maternal Dl gradient. The gene is activated initially at nuclear cleavage cycle 11-12 within 1 h after the Dl gradient is formed. zen initially exhibits a broad pattern of expression in the presumptive dorsal ectoderm and at the termini. High and low levels of the Dl gradient keep zen off in ventral and lateral regions. sog exhibits a complementary pattern of expression because it is activated by Dl, whereas zen is repressed. As seen for zen, sog expression is detected shortly after the formation of the Dl gradient (Markstein, 2002).

The zen VRE contains four optimal Dl recognition sequences within a span of 400 bp. Three of the four Dl binding sites contained within the zen VRE conform to the following consensus sequence for high-affinity Dl binding sites: GGG(W)_nCCM (where W = A or T, M = C or A, and n corresponds to either four or five W residues). The fourth recognition sequence (binding site 3 within the VRE) contains a G residue in the AT-rich central region and is represented by the optimal consensus sequence GGGWDWWWCCM (where D = A, T, or G). To determine whether a similar density of optimal Dl sites might account for the regulation of sog, the entire Drosophila genome was scanned for clusters of any of the 208 unique Dl sequences that conform (either directly or by reverse complement) to two degenerate sequences: GGG(W)₄CCM and GGGWDWWWCCM (Markstein, 2002).

The genome was scanned for clusters of Dl binding sites in windows of 400 bp, the interval within which the sites are clustered in the zen VRE, and also for clustering in windows of 1,000 bp because the operational size of enhancers can generally be thought of as about 1,000 bp. Although the genome-wide occurrence of 676 clusters of two or more optimal Dl sites in 1,000 bp is not statistically significant, the occurrence of 55 clusters with at least three sites and of eight clusters containing four sites is enriched beyond what one would expect from random chance. However, none of the clusters within 1,000 bp identified known Dl targets that were missed by the more stringent screen for clustering within 400 bp. Therefore, this study focussed on the results from the more stringent screen (Markstein, 2002).

As expected, the occurrence of 400-bp windows containing at least two sites (327 clusters) is much greater than the occurrence of 400-bp windows containing at least three sites (15 clusters) or four sites (3 clusters). However, the statistical significance of the clusters increases with their rarity. For example, the occurrence of 15 clusters with three or more Dl sites is 6 standard deviations from expected, making the probability of finding 15 clusters by random chance less than one in a million. The probability of finding three 400-bp clusters with at least four Dl sites is less than 10^-49. Remarkably, two of the clusters in this rarest class are associated with the sog and zen genes, which exhibit the most sensitive response to the Dl gradient. Of the remaining 13 clusters containing three or more Dl sites, one is associated with the Brinker gene, which is expressed in lateral stripes and probably is a direct target of the Dl gradient. The Brinker site is located ~10 kb 5' of the transcription start site. Brinker probably is a direct target of the Dl gradient in that it exhibits lateral stripes of expression that are similar to those observed for rhomboid. The other remaining 12 clusters were found to neighbor genes that were not known previously to be involved in dorsoventral patterning (Markstein, 2002).

sog is expressed initially in broad lateral stripes that encompass the entire presumptive neurogenic ectoderm. Staining persists in these lateral stripes during cellularization and the onset of gastrulation but quickly refines within the mesectoderm at the ventral midline of elongating embryos. There are four optimal Dl binding sites located within a 263-bp region of sog intron 1. Three different DNA fragments that encompass this region of the sog gene were placed 5' of a lacZ reporter gene and expressed in transgenic embryos. The largest fragment is 6 kb in length and includes two-thirds of intron 1, whereas the smallest is just 393 bp and centered around the cluster of Dl binding sites. A 1.5-kb fragment that extends into the 5'-flanking region was also tested. All three fragments direct lateral stripes of lacZ expression that are similar to those seen for the endogenous gene. These broad stripes persist until gastrulation and then refine within the mesectoderm. Midline staining becomes weak and erratic in older embryos. The 393-bp sog fragment directs essentially the same staining pattern as those obtained with the 6-kb DNA fragment as well as the 1.5-kb fragment. These results suggest that the 393-bp fragment (hereafter called the sog lateral stripe enhancer) contains most of the cis elements responsible for regulating the early sog pattern (Markstein, 2002).

The sog lateral stripe enhancer shares a number of similarities with the previously characterized rhomboid NEE, which also mediates gene expression in the neurogenic ectoderm. However, the NEE stripes are narrower than those generated by the sog enhancer, suggesting that the sog enhancer responds to lower levels of the Dl gradient than does the NEE. This difference might be due, at least in part, to the quality or organization of the Dl binding sites in the two enhancers. For example, only two of the four Dl binding sites contained in the NEE are optimal sites, whereas all four sites are optimal in the sog enhancer. The NEE contains four binding sites for the zinc finger snail repressor, which is expressed selectively in the ventral mesoderm and thereby restricts rhomboid expression to lateral regions. The sog lateral stripe enhancer contains two potential snail repressor sites (CACCT) that might be responsible for attenuated staining in ventral regions (Markstein, 2002).

Transcriptional Regulation

Decapentaplegic, in addition to its role as a morphogen in structuring gene expression and positioning of veins in the larval wing disc, is expressed in vein primordia during pupal wing development and functions to promote vein formation. In contrast, sog is expressed in complementary intervein cells and suppresses vein formation. sog and dpp function during the same phenocritical periods (i.e. 16-28 hours after pupariation) to influence the vein versus intervein cell fate choice. The conflicting activities of dpp and sog are also revealed by antagonistic dosage-sensitive interactions between these two genes during vein development. Analysis of vein and intervein marker expression in dpp and sog mutant wings suggests that dpp promotes vein fates indirectly by activating the vein gene rhomboid (rho), and that sog functions by blocking an autoactivating DPP feedback loop. Ectopic expression of dpp activates rho and suppresses sog expression. It is thought that the dpp suppression of sog is indirect, acting through rhomboid. A network of gene interactions promote vein fates as EGF-R ligands Vein and Spitz are also involved in intervein and vein fates respectively. These data support the view that SOG is a dedicated DPP antagonist (Yu, 1997).

Differential activation of the Toll receptor leads to the formation of a broad Dorsal nuclear gradient that specifies at least three patterning thresholds of gene activity along the dorsoventral axis of precellular embryos. The activities of the Pelle kinase and Twist basic helix-loop-helix (bHLH) transcription factor in transducing Toll signaling have been investigated. Pelle functions downstream of Toll to release Dorsal from the Cactus inhibitor. Twist is an immediate-early gene that is activated upon entry of Dorsal into nuclei. Transgenes misexpressing Pelle and Twist were introduced into different mutant backgrounds and the patterning activities were visualized using various target genes that respond to different thresholds of Toll-Dorsal signaling. These studies suggest that an anteroposterior gradient of Pelle kinase activity is sufficient to generate all known Toll-Dorsal patterning thresholds and that Twist can function as a gradient morphogen to establish at least two distinct dorsoventral patterning thresholds. How the Dorsal gradient system can be modified during metazoan evolution is discussed and it is concluded that Dorsal-Twist interactions are distinct from the interplay between Bicoid and Hunchback, which pattern the anteroposterior axis (Stathopoulos, 2002).

The snail, sim, vnd and sog expression patterns represent four different Toll-Dorsal signaling thresholds. snail is activated only by peak levels of the Dorsal gradient; sim and vnd are activated by intermediate levels, and sog is activated by the lowest levels of the gradient. These expression patterns were visualized in mutant and transgenic embryos via in situ hybridization using digoxigenin-labeled antisense RNA probes (Stathopoulos, 2002).

Dorsal target genes are essentially silent in mutant embryos that lack an endogenous dorsoventral Dorsal nuclear gradient. Mutant embryos were collected from females that are homozygous for a null mutation in the gastrulation defective (gd) gene, which blocks the processing of the Spätzle ligand and the activation of the Toll receptor. These mutants permit the analysis of ectopic, anteroposterior Dorsal and Twist gradients in 'apolar' embryos that lack dorsoventral polarity. snail, vnd, and sog are sequentially expressed along the anteroposterior axis of mutant embryos that contain a constitutively activated form of the Toll receptor (Toll^10b) misexpressed at the anterior pole using the bicoid (bcd) promoter and 3' UTR. These expression patterns depend on an ectopic anteroposterior Dorsal nuclear gradient. The repression of the vnd and sog patterns at the anterior pole is probably mediated by Snail, which normally excludes expression of these genes in the ventral mesoderm of wild-type embryos (Stathopoulos, 2002).

The activated Pelle-Tor⁴⁰²¹ kinase also directs sequential anteroposterior patterns of snail, vnd, and sog expression in gd^–/gd^– mutant embryos. As in the case of Toll^10b, the activated Pelle kinase was misexpressed at the pole using the bcd 3' UTR. The snail, vnd and sog expression patterns are similar to those obtained with the Toll^10b transgene. The vnd and sog expression patterns are probably repressed at the anterior pole by Snail. These results suggest that the levels of Pelle kinase activity are sufficient to determine different Dorsal transcription thresholds (Stathopoulos, 2002).

sog is normally activated throughout the neurogenic ectoderm by the lowest levels of the Dorsal gradient. The low levels of Dorsal present in Toll^rm9/Toll^rm10 mutant embryos are sufficient to activate sog everywhere except the extreme termini. The twist-bcd transgene leads to the loss of sog expression in anterior regions, probably because of repression by Snail. Snail also appears to repress vnd and sog expression in anterior regions of transgenic embryos that contain the Toll^10b or Pelle-Tor⁴⁰²¹ transgenes (Stathopoulos, 2002).

The twist-bcd transgene was introduced into mutant embryos that completely lack Dorsal. Without the transgene these mutants do not express twist, snail, sim, vnd or sog. Introduction of the twist-bcd transgene causes intense expression of twist in the anterior 40% of the embryo. This broad Twist gradient fails to activate snail, but succeeds in inducing weak expression of sim and somewhat stronger staining of vnd at the anterior pole. The activation of vnd in mutant embryos is comparable with the expression seen in wild-type and Toll^rm9/Toll^rm10 embryos. However, in both wild-type and mutant embryos the vnd pattern is transient, and lost after the completion of cellularization. These results indicate that Twist can activate dorsoventral patterning genes in the absence of Dorsal (Stathopoulos, 2002).

Targets of Activity

SOG acts non-cell autonomously to antagonize the activity of dorsally active genes, such as decapentaplegic and tolloid (Francois, 1994).

SOG counteracts the antineurogenic effects of Decapentaplegic. To assess the role of dpp in regulating neurogenesis, the effect of this signaling pathway on neurogenesis in dorsal cells was examined. The dorsal cuticle of embryos lacking dpp activity appears to be ventralized. However, this inference is tenuous, as the number of differentiated neurons is reduced, not expanded, in late dpp mutant embryos. In contrast, for early gastrulating mutants, dorsal expression is observed of neuroectodermal markers such as thick veins and lethal of scute. Similarly, neuroblasts visualized with markers such as scratch and snail subsequently form ectopically in the dorsal region of dpp mutants. Consistent with dpp acting early to suppress initiation of neurogenesis, ectopic expression of the proneural gene lethal of scute is first detectable in dorsal cells of late blastoderm stage dpp mutants. Paradoxically, the increased number of neuroblasts in dpp mutants does not generate a hypertrophied differentiated nervous system. Thus, dpp mutants may lack a late positive role for dpp in neuronal maturation or may hyperactivate pathways functioning to inhibit subsequent steps in neurogenesis (Biehs, 1996).

Short gastrulation prevents DPP from suppressing neurogenesis laterally in the blastoderm embryo. It is possible to exacerbate defects in sog mutants by increasing the level of DPP. The earliest neuroectodermal marker affected in sog mutants with a double dose of dpp is rhomboid, which is normally expressed in lateral stripes 8-10 cells wide in wild-type embryos but rapidly narrows to stripes 4-6 cells across in sog mutants with elevated DPP. Similarly l'sc expression is reduced in sog mutants with elevated DPP. Surprisingly, dpp itself is induced throughout the neuroectoderm in this genetic combination. This provides the first evidence that dpp is capable of autoactivating its own expression during early embryogenesis. Ubiquitous dpp expression results in zerknüllt expression throughout the entire trunk neuroectoderm and mesoderm (Biehs, 1996).

A striking feature of the effects of DPP on neural suppression and dorsalization is that neuronal suppression is induced by a lower threshold of DPP activity than is dorsalization. Much less DPP is required to suppress expression of neuroectodermal genes than is required to activate dorsal markers. For example, brief submaximal heat induction of heat shock dpp in a wild type sog background leads to nearly maximal suppression of lethal of scute, scratch and snail expression during germ band extension, but there is no detectable ectopic expression of zerknüllt in the neuroectoderm (Biehs, 1996).

Gene dosage experiments are consistent with SOG diffusing dorsally 12 to 15 cell diameters from the lateral source of SOG mRNA to determine the limit of dorsal rhomboid expression. Thus SOG diffuses from the neuroectoderm into the presumptive mesoderm to interfere with DPP signaling. Since the effect of SOG is highly dosage dependent, it is likely that there is a gradient of SOG activity in both dorsal and ventral regions of the embryo creating a reciprocal gradient of DPP activity in the dorsal region of the embryo. In this respect, SOG displays many features of a classic morphogen (Biehs, 1996).

The short gastrulation (sog) and decapentaplegic (dpp) genes function antagonistically in the early Drosophila zygote to pattern the dorsoventral (DV) axis of the embryo. This interplay between sog and dpp determines the extent of the neuroectoderm and subdivides the dorsal ectoderm into two territories. Evidence exists that sog and dpp also play opposing roles during oogenesis in patterning the DV axis of the embryo. Maternally produced Dpp increases levels of the IkappaB-related protein Cactus and reduces the magnitude of the nuclear concentration gradient of the NFkappaB-related Dorsal protein, and Sog limits this effect. Evidence is presented suggesting that Dpp signaling increases Cactus levels by reducing a signal-independent component of Cactus degradation. Epistasis experiments reveal that sog and dpp act downstream of, or in parallel to, the Toll receptor to reduce translocation of Dorsal protein into the nucleus. These results broaden the role previously defined for sog and dpp in establishing the embryonic DV axis and reveal a novel form of crossregulation between the NFkappaB and TGFbeta signaling pathways in pattern formation (Araujo, 2000).

In aggregate, the results support models in which Sog and Dpp proteins are produced by the follicle cells and then are delivered to the embryo. These proteins could be deposited in the vitelline membrane or in the oocyte plasma membrane, or might be sequestered in the perivitelline space and remain there protected until early embryogenesis. The fact that sog and dpp are expressed in follicle cells of stage 10 egg chambers, around the time that follicle cells are secreting major structural proteins of the vitelline envelope, is consistent with their products being delivered to the vitelline membrane or perivitelline space. Since sog and dpp are secreted proteins, they could be exported like components of the vitelline membrane to the extracellular compartment between the follicle cells and the oocyte. After stage 13, the vitelline membrane is thought to be an impermeant barrier separating the oocyte from follicle cells making it unlikely that sog and dpp products are transferred after this time. A similar model has been proposed to explain the functions of the dorsal group gene nudel and of the maternal terminal system gene torsolike (tsl). Both of these genes are expressed during midoogenesis, long before their activity is required during early embryogenesis. According to this model, the Sog and Dpp proteins would remain in the perivitelline space until early embryogenesis, when the Tl pathway is activated by Spatzle. In the early embryo, maternal Dpp would decrease the level of Tl-mediated nuclear translocation of Dorsal by decreasing Cactus signal-independent degradation through a pathway acting in parallel to Tl. Presumably, Sog antagonizes the action of Dpp, resulting in maximal nuclear Dorsal translocation (Araujo, 2000).

Consistent with the view that Sog and Dpp proteins are made early (e.g. midoogenesis), but act later in the early embryo, induction of sog expression during midoogenesis by use of a heat-shock sog construct increases levels of a Sog fragment in the early embryo detected by a specific anti-Sog antibody. Thus, Sog protein produced during midoogenesis can be stably stored for a protracted period until the onset of embryogenesis. In contrast to Sog protein, SOG mRNA does not perdure at detectable levels in early pre-blastoderm embryos in these experiments. The fragment of Sog generated in these experiments is the same size (60 kDa) as one that may have activity during pupal development (Araujo, 2000).

There are several unanswered questions regarding how maternal Dpp signaling contributes to embryonic DV patterning. An important remaining question is how maternal Dpp signaling contributes to defining discrete zones of gene expression along the DV axis? Two leading possibilities, which are not necessarily mutually exclusive are: (1) sog and dpp function to determine the relative proportions and positions of the different primary DV domains, and (2) Dpp signaling is necessary to sharpen borders between embryonic DV territories. There is good evidence in support of the first possibility, since the extents of DV expression domains can be altered by increasing maternal Dpp activity. As mentioned above, maternally produced Dpp results in a ventral shift of all DV domains, presumably by lowering the amount of nuclear Dorsal in cells along the entire DV axis. These results also support a role for maternal sog and dpp in refining the normally sharp borders between different territories, since altering the maternal dose of sog or dpp generates overlapping expression of mesodermal and neuroectodermal genes (Araujo, 2000).

Another question is by what mechanism does maternal Sog oppose Dpp in patterning the embryo? Perhaps Sog is necessary to inhibit Dpp signaling through a specific receptor subtype such as the Sax receptor or to restrict Dpp signaling to a specific type of Dpp receptor (e.g. mediated only by Tkv). Alternatively, Sog could be involved in antagonizing another BMP molecule in addition to Dpp, which also functions in embryonic DV patterning. In summary, the results presented in this study indicate that maternal components of Dpp signaling modify elements that converge with signaling downstream of the Tl receptor by regulating Cactus levels and nuclear translocation of Dorsal. This analysis suggests that maternal sog and dpp function to define the relative proportions of embryonic DV domains and may play a role in creating sharp borders between these domains. Further experiments will be necessary to determine the mechanism by which maternal sog and dpp function and how interactions between the Tl and Dpp pathways collaborate to pattern the DV axis of the Drosophila embryo (Araujo. 2000).

Protein Interactions

Although a direct physical interaction between DPP and SOG has not been established, such an interaction is implied by the known physical interaction of BMP4 and Chordin, DPP and SOG homologs in Xenopus (Piccolo, 1996).

Noggin, a protein expressed in the Spemann organizer region of the Xenopus embryo, promotes dorsal cell fate within the mesoderm and neural development within overlying ectoderm. noggin, expressed in Drosophila, promotes ventral development, specifying ventral ectoderm and CNS in the absence of all endogenous ventral-specific zygotic gene expression. Noggin blocks DPP signaling upstream of DPP receptor activation. It is proposed that, whole most or all of the DPP produced in the dorsal-most region binds to its receptors, DPP produced more laterally has an increased probability of being bound by ventrally produced Short gastrulation, and that DPP can be released from this diffusible complex by the action of a third dorsal-specific gene, perhaps tolloid (Holley, 1996).

Extracellular gradients of signaling molecules can specify different thresholds of gene activity in development. A gradient of Decapentaplegic (Dpp) activity subdivides the dorsal ectoderm of the Drosophila embryo into amnioserosa and dorsal epidermis. The proteins Short gastrulation (Sog) and Tolloid (Tld) are required to shape this gradient. Sog has been proposed to form an inhibitory complex with either Dpp or the related ligand Screw, and is subsequently processed by the protease Tld. Paradoxically, Sog appears to be required for amnioserosa formation, which is specified by peak Dpp signaling. Sog appears to be required for peak Dpp/Screw activity, since sog mutants lack amnioserosa. SOG transcripts are detected in two ventrolateral stripes within the presumptive neurogenic ectoderm. Several amnioserosa marker genes, including Kruppel, rhomboid and hindsight exhibit broadened patterns of expression that gradually diminish in older embryos. In contrast, the Race (Related to angiotensin converting enzyme) pattern is not transiently expanded in sog mutants; instead, by the onset of gastrulation, expression is nearly lost in central regions. Race may represent a more definitive marker for the presumptive amnioserosa than the genes used in previous studies (Ashe, 1999).

The misexpression of sog using the even-skipped stripe-2 enhancer redistributes Dpp signalling in a mutant background in which dpp is expressed throughout the embryo. Dpp activity is diminished near the Sog stripe and peak Dpp signaling is detected far from this stripe. However, a tethered form of Sog suppresses local Dpp activity without augmenting Dpp activity at a distance, indicating that diffusion of Sog may be required for enhanced Dpp activity and consequent amnioserosa formation. The long-distance stimulation of Dpp activity by Sog requires Tld, whereas Sog-mediated inhibition of Dpp does not. The heterologous Dpp inhibitor Noggin inhibits Dpp signaling but fails to augment Dpp activity. These results suggest an unusual strategy for generating a gradient threshold of growth-factor activity, whereby Sog and its protease specify peak Dpp signaling far from a localized source of Sog. Different models have been proposed to explain the requirement of Sog in generating peak Dpp activity. One invokes the diffusion of Sog-Dpp or Sog-Screw complexes away from the ventrolateral Sog stripes, thereby focusing Dpp and/or Screw at the dorsal midline. An alternative model suggests that a product resulting from the cleavage of Sog directly signals formation of the amnioserosa, possibly by augmenting the binding of Dpp or Screw to the receptors Thick veins and Saxophone (Ashe, 1999).

Processing of the Drosophila Sog protein creates a novel BMP inhibitory activity

Structurally unrelated neural inducers in vertebrate and invertebrate embryos have been proposed to function by binding to BMP4 or Dpp, respectively, and preventing these homologous signals from activating their receptor(s). The functions of various forms of the Drosophila Sog protein were examined using the discriminating assay of Drosophila wing development. Misexpression of Drosophila Sog, or its vertebrate counterpart Chordin, generates a very limited vein-loss phenotype. This sog misexpression phenotype is very similar to that of viable mutants of glass-bottom boat (gbb), which encodes a BMP family member. Consistent with Sog selectively interfering with Gbb signaling, Sog can block the effect of misexpressing Gbb, but not Dpp in the wing. In contrast to the limited BMP inhibitory activity of Sog, carboxy-truncated forms of Sog, referred to as Supersog, have been identified which when misexpressed cause a broad range of dpp minus mutant phenotypes (Yu, 2000).

The predicted Sog protein is 1038 amino acids in length and contains four cysteine-rich (CR) domains in the extracellular domain. The metalloprotease Tld cleaves Sog at three major sites. Supersog1 is an N-terminal fragment of Sog including CR1 plus another 114 amino acids, and contains an additional 33 amino acids derived from vector sequences at its C terminus. Supersog2, which contains the same amino acids as Supersog1 but terminates abruptly at the end of Sog sequences, also generates Supersog phenotypes, albeit slightly weaker than those observed with Supersog1. Supersog4 is an N-terminal fragment of Sog ending 80 amino acids before CR2 and includes 130 sog 3' UTR derived amino acids (Yu, 2000).

In line with its phenotypic effects, Supersog can block the effects of both misexpressing Dpp and Gbb in the wing. Vertebrate Noggin, in contrast, acts as a general inhibitor of Dpp signaling, which can interfere with the effect of overexpressing Dpp, but not Gbb. Evidence suggests that Sog processing occurs in vivo and is biologically relevant. Overexpression of intact Sog in embryos and adult wing primordia leads to the developmentally regulated processing of Sog. This in vivo processing of Sog can be duplicated in vitro by treating Sog with a combination of the metalloprotease Tolloid (Tld) plus Twisted Gastrulation (Tsg), another extracellular factor involved in Dpp signaling. In accord with this result, coexpression of intact Sog and Tsg in developing wings generates a phenotype very similar to that of Supersog. Evidence is provided that tsg functions in the embryo to generate a Supersog-like activity, since Supersog can partially rescue tsg minus mutants. Consistent with this finding, sog minus and tsg minus mutants exhibit similar dorsal patterning defects during early gastrulation. These results indicate that differential processing of Sog generates a novel BMP inhibitory activity during development and, more generally, that BMP antagonists play distinct roles in regulating the quality as well as the magnitude of BMP signaling (Yu, 2000).

To determine whether Sog might be processed in vivo to generate Supersog-like molecules, an anti-Sog antibody directed against an amino fragment of Sog was used on immunoblots to analyze protein extracts from different stages and tissues of developing Drosophila. This antibody recognizes an epitope present in the stem portion of Supersog. An examination was made of the nature of Sog products produced both in wild-type individuals as well as in flies overexpressing Sog. This analysis reveals that Sog is processed in vivo, and that this processing is developmentally regulated. For example, in heat shocked early embryos carrying eight copies of an HS-sog construct, a 76 kDa band, a doublet of bands migrating at 42/40 kDa, and a 28 kDa band were observed, in addition to a 120 kDa band corresponding to full-length Sog. These bands are likely to represent various forms of Sog since they are strongly induced only in heat shocked HS-sog blastoderm stage embryos. Heat induction of HS-sog pupae results in the elevated production of prominent Sog fragments migrating at 76, 60, 50 and 42 kDa. In pupal wings, the same pattern of Sog fragments is present in overloaded extracts of wild-type pupal wings as observed in heat induced HS-sog wings, albeit at lower levels. This significant level of endogenous processing is not surprising given that wild-type pupal wings express high levels of Sog throughout intervein regions, which account for approximately 90% of cells in the wing (Yu, 2000).

Processing of exogenously provided Sog is developmentally regulated. During embryonic and pupal stages, when Sog is expressed in a significant fraction of cells and plays important developmental roles, distinct patterns of Sog fragments are produced. For example, during pupal development, 60, 50 and 42 kDa fragments are induced in heat shocked HS-sog wings, while in early embryos, a pair of induced bands migrating at 42/40 kDa is most prominent. In contrast, during late embryonic or third larval instar stages, only the full length Sog band is observed upon induction of HS-sog larvae. During these latter stages of development, sog is expressed in only a small percentage of cells and is not known to have any significant developmental function. Thus, Sog is processed in vivo at developmentally relevant times and in different patterns to generate fragments that are likely to have distinct activities from Sog in addition to being degraded into inert products (Yu, 2000).

The fact that pulses of Supersog1 expression delivered during the late blastoderm stage of development can partially rescue the tsg minus mutant embryos suggests that a Supersog-like activity might mediate part of tsg function in vivo. In addition, late blastoderm stage tsg minus mutant embryos display defects similar to those of sog mutants, suggesting that tsg is involved in a late function of Sog. Consistent with the view that tsg acts during early gastrulation, tsg minus mutants can not be rescued by driving expression of a tsg transgene under the control of the tld promoter, which is expressed only early during the blastoderm stage. In contrast, it is possible to rescue tsg minus mutants by driving tsg expression with promoters that continue to be expressed into early gastrulation. Several possible ways in which Supersog-like activities could contribute to this stage of development can be imagined, given that they have different ligand specificities from intact Sog and are stable to further proteolysis by Tld. Since Sog has been proposed to block the activity of Scw in embryos, it is likely that some other BMP is the preferred target of Supersog molecules. In addition, since Scw is only expressed transiently during the blastoderm stage of development, intact Sog would have no obvious target to inhibit beyond this stage. Perhaps a stable broad-spectrum BMP antagonist such as Supersog could inhibit the action of other BMPs expressed in the dorsal ectoderm during early stages of gastrulation (possibly Dpp itself) and thereby provide a form of molecular memory, which helps maintain the distinction between neural and non-neural ectoderm (Yu, 2000).

The observation that Supersog is less effective than Sog in blocking BMP signaling in the early embryo is consistent with the view that Supersog is not just a higher affinity version of Sog and suggests that Supersog is actually less effective than Sog at blocking the effect of Scw. The fact that Supersog does not inhibit Dpp itself during early blastoderm stages is likely to be the result of insufficient levels of Supersog being expressed by the heat shock vector. It is possible, however, that an endogenously produced Supersog activity (e.g. generated upon Tsg binding to Sog) has a higher affinity for Dpp than the artificially created Supersog1 construct. In any case, it is proposed that Supersog acts in the late blastoderm embryo or during early gastrulation stages rather than in the early blastoderm embryo, and that during this latter period, it is able to block the activity of a BMP (e.g. Dpp?) not recognized by Sog. It is tempting to consider a two step temporal model for the action of Sog and Supersog during embryonic dorsal-ventral patterning to account for the fact that sog mutants display a dorsal-ventral phenotype earlier than tsg minus mutants. According to one such scenario, the labile Tld-sensitive form of full-length Sog is produced from a localized source (i.e. the neuroectoderm) and diffuses dorsally to be degraded by Tld. Tld acts as a sink to create a transiently stable gradient of Sog, which creates a reciprocal gradient of Dpp activity. The Sog gradient created by this classic source/sink configuration would only be short-lived, however, since cells begin migrating when gastrulation begins. At this stage, the embryo elongates and the Dorsal gradient collapses, leading to loss of gene expression in early zygotic D/V domains. Following the establishment of the short-lived hypothetical Sog gradient, tsg expression is initiated in dorsal cells and leads to the production of stable Supersog-like molecules by switching the activity of Tld from degrading to activating Sog. Supersog-like molecules then could provide a stable record of high versus low BMP signaling domains during a subsequent step of development (Yu, 2000).

Creation of a Sog morphogen gradient in the Drosophila embryo

A variety of genetic evidence suggests that a gradient of Decapentaplegic (Dpp) activity determines distinct cell fates in the dorsal region of the Drosophila embryo, and that this gradient may be generated indirectly by an inverse gradient of the BMP antagonist Short gastrulation (Sog). It has been proposed that Sog diffuses dorsally from the lateral neuroectoderm where it is produced, and is cleaved and degraded dorsally by the metalloprotease Tolloid (Tld). This study shows directly that Sog is distributed in a graded fashion in dorsal cells and that Tld degradation limits the levels of Sog dorsally. In addition, Dynamin-dependent retrieval of Sog acts in parallel with degradation by Tld as a dorsal sink for active Sog (Srinivasan, 2002).

As a first step in determining whether Sog diffusion contributes to formation of a Dpp activity gradient, the distribution of Sog protein was directly examined in wild-type Drosophila embryos using two anti-Sog antibodies raised to different regions of the protein. In wild-type cellular blastoderm stage embryos, high levels of Sog protein are recognized by both the 8B and 8A anti-Sog antibodies, which colocalize with SOG RNA in broad ventrolateral stripes corresponding to the neuroectoderm. In addition to the high levels of Sog protein present in the lateral neuroectoderm, lower levels of staining are also observed in dorsal epidermal and ventral mesodermal cells far from the source of Sog. It is notable that the staining observed with either anti-Sog antiserum is consistently stronger in the ventral mesoderm than in the dorsal ectoderm, consistent with there being a mechanism(s) to limit accumulation of Sog dorsally. In the dorsal ectoderm, Sog staining appears graded, with the highest levels present immediately adjacent to the neuroectoderm and progressively lower levels observed dorsally. The dorsal gradient of Sog protein is best revealed by the 8A anti-Sog antibody and becomes most pronounced in late blastoderm stage embryos. Sog immunolabeling is present in two parallel tracks of punctate staining in dorsal ectodermal and ventral mesodermal cells. These observations indicate that an early gradient of Sog is present in wild-type embryos at a time when Sog is known to function in a dose-dependent fashion as a Dpp antagonist (Srinivasan, 2002).

The finding that Tld collaborates with the Tolkin (Tok) protease to limit Sog diffusion indicates that these two closely related proteases are likely to share at least this one important substrate. Consistent with this possibility, in vitro studies indicate that Tok can cleave Sog in vitro, but with significantly reduced activity relative to Tld. Since Tld cleaves Sog in only a limited number of specific sites in vitro, it is likely that another class of extracellular protease degrades the products of Tld/Tok cleavage to peptide fragments, which may be too small to be recognized by either the 8A or 8B Sog antibodies. It is noteworthy that Tld degradation of Sog occurs on a much more rapid time scale in vivo (e.g., 30 min) than in vitro (e.g., several hours). This finding is consistent with the developmental timescale of Tld activity and suggests that additional factors present in vivo accelerate the action of Tld (Srinivasan, 2002).

The observation that Sog degradation fails to take place in dorsal cells of dpp^- mutants is consistent with in vitro experiments in which Dpp is required as a cofactor for Tld-dependent cleavage of Sog. In contrast to in vitro studies in which either Dpp or Scw can act as cofactors, only Dpp serves as a critical cofactor function for in vivo degradation of Sog. An interesting difference between the ectopic Sog observed in dpp^- versus Df(tld) embryos is that the staining is uniform in dpp^- mutants but retains some degree of gradation in Df(tld) mutants. It is possible that another yet uncharacterized metalloprotease collaborates with Tld and Tok to degrade Sog in the early embryo. Alternatively, Sog might bind to a complex containing Dpp that is still present in Df(tld) mutant and limits Sog diffusion dorsally. The formation of this complex, or the ability of Sog to bind to it, may be strictly dependent on Dpp, so that in its absence, there is no restraint on Sog diffusion dorsally (Srinivasan, 2002).

An additional aspect of this study is the finding that Dynamin (shi) functions in parallel with Tld/Tok to limit active Sog levels in dorsal cells, which is required to generate a peak response to BMP signaling in dorsal-most cells. The fact that shi was not picked up previously as a D/V mutant in systematic screens for embryonic patterning mutants presumably reflects the pleiotropic requirement for Dynamin function, which is also required for Hh, Wg, Notch, and EGF-R signaling as well as various other cell biological processes involving membrane trafficking. While Dynamin function is not required for diffusion of Sog dorsally, it does appear to be required for the maintenance of the Sog gradient by removing Sog from the extracellular space. It is also possible that Dynamin plays other roles in promoting BMP signaling and that removing Sog from shi^ts; sog RNAi embryos compensates for this reduced function. One argument against this latter possibility is that elimination of Dynamin function prior to the production and secretion of Sog does not compromise BMP signaling at that earlier stage. In any case, it is clear that an active form of Sog mediates the reduction of BMP signaling associated with loss of Dynamin function (Srinivasan, 2002).

In addition to inhibiting the activity of Scw and thereby reducing BMP signaling, there is evidence that Sog can exert other activities. For example, in the presence of the secreted protein Twisted gastrulation, Sog is cleaved in a different pattern by Tld in vitro to generate a truncated form of Sog consisting of CR1 and part of the stem. This truncated molecule, called Supersog, can inhibit Dpp as well as the auxiliary BMPs Scw and Glass bottom boat (Gbb). A major function of Tsg is to generate a Supersog-like activity in vivo, since expression of Supersog, but not intact Sog, can partially rescue tsg^- mutant embryos. Supersog may play a persistent role in inhibiting BMP signaling following the transient expression of Scw, since it is refractory to degradation by Tld. There is also indirect genetic evidence that Sog acts at a long range to promote BMP signaling as judged by activation of the target gene RACE. Since Tld plays a dose-dependent role in generating this putative positive Sog activity, it too may be a processed form of Sog. It has also been proposed that some form of Sog might carry Dpp to the dorsal midline and thereby concentrate BMP along the dorsal midline. One line of evidence supporting this model is that the pattern of phosphorylation and activation of Mad observed in situ by staining with an anti-pMAD antibody reveals a narrow dorsal band of peak BMP activity with little evidence for a gradient diminishing ventrally. However, there is also a wealth of indirect genetic evidence that there are several intermediate levels of BMP activity that activate several dorsally expressed BMP target genes at different levels (Srinivasan, 2002).

The findings in this study, which reveal a continuous Sog gradient that diminishes progressively in dorsal cells with antibodies recognizing two different portions of Sog, do not provide any direct support for the existence of processed forms of Sog accumulating dorsally in a pattern that presages the profile of pMAD staining. These differences in assessing the shape of the BMP activity gradient may reflect nonlinear properties of the anti-pMAD antibody or the existence of distinct forms of pMAD, only one of which is efficiently recognized by the currently used anti-pMAD reagent. Alternatively, an intrinsically nonlinear pMAD transducing mechanism may be integrated with other spatially regulated information to create graded and distributed activation of various BMP target genes (Srinivasan, 2002)

Although the final BMP activity gradients generated in the early embryo and wing imaginal discs have strikingly similar shapes, the mechanisms for creating them are very different. In the wing disc, the spread of Dpp from its narrow localized source in the center of the disc is limited by sequestration by means of the BMP receptor Tkv. In contrast, in the embryo, a BMP activity gradient forms within a broad domain of uniform dpp expression in response to the creation of an inverse gradient of the BMP antagonist Sog, which diffuses into the dorsal domain from the adjacent neuroectoderm. This Sog gradient in dorsal cells is created by a combination of specific proteolytic degradation by Tld and Tok and Dynamin-mediated retrieval of Sog. It is noteworthy that Dynamin exerts opposite effects in the early embryo and wing disc. In the precellular embryo, where extracellular molecules may diffuse in a passive ink-in-water fashion in the surrounding perivitelline fluid, a Dynamin-dependent mechanism limits extracellular accumulation of Sog dorsally. In contrast, in the cellularized context of the wing imaginal disc, an active Dynamin-dependent transport process is required for Dpp movement between cells. The fact that completely different mechanisms can ultimately create similarly shaped BMP activity gradients highlights the flexibility of evolutionary processes, which can arrive at more than one type of solution to the same basic problem (Srinivasan, 2002).

Robustness of the BMP morphogen gradient in Drosophila embryonic patterning

Developmental patterning relies on morphogen gradients, which generally involve feedback loops to buffer against perturbations caused by fluctuations in gene dosage and expression. Although many gene components involved in such feedback loops have been identified, how they work together to generate a robust pattern remains unclear. The network of extracellular proteins that patterns the dorsal region of the Drosophila embryo by establishing a graded activation of the bone morphogenic protein (BMP) pathway has been studied. The BMP activation gradient itself is robust to changes in gene dosage. Computational search for networks that support robustness shows that transport of the BMP class ligands (Scw and Dpp) into the dorsal midline by the BMP inhibitor Sog is the key event in this patterning process. The mechanism underlying robustness relies on the ability to store an excess of signaling molecules in a restricted spatial domain where Sog is largely absent. It requires extensive diffusion of the BMP-Sog complexes, coupled with restricted diffusion of the free ligands. Dpp is shown experimentally to be widely diffusible in the presence of Sog but tightly localized in its absence, thus validating a central prediction of a theoretical study (Eldar, 2002).

Graded activation of the BMP pathway subdivides the dorsal region of Drosophila embryos into several distinct domains of gene expression. This graded activation is determined by a well-characterized network of extracellular proteins, which may diffuse in the perivitelline fluid that surrounds the embryo. The patterning network is composed of two BMP class ligands (Scw and Dpp), a BMP inhibitor (Sog), a protease that cleaves Sog (Tld) and an accessory protein (Tsg), all of which are highly conserved in evolution and are used also for patterning the dorso-ventral axis of vertebrate embryos. Previous studies have suggested that patterning of the dorsal region is robust to changes in the concentrations of most of the crucial network components. For example, embryos that contain only one functional allele of scw, sog, tld or tsg are viable and do not show any apparent phenotype. Misexpression of scw or of tsg also renders the corresponding null mutants viable (Eldar, 2002).

To check whether robustness is achieved at the initial activation gradient, signaling was monitored directly by using antibodies that recognize specifically an activated, phosphorylated intermediate of the BMP pathway (pMad). Prominent graded activation in the dorsal-most eight cell rows was observed for about 1h, starting roughly 2h after fertilization at 25°C. This activation gradient was quantified in heterozygous mutants that were compromised for one of three of the crucial components of the patterning network, Scw, Sog or Tld. Whereas homozygous null mutants that completely lack the normal gene product have a deleterious effect on signaling, the heterozygotes, which should produce half the amount of the gene product, were indistinguishable from wild type. Similarly, overexpression of the Tld protein uniformly in the embryo did not alter the activation profile. The activation profile at 18°C is the same as that at 25°C. This robustness to temperature variations is marked, considering the wide array of temperature dependencies that are observed in this temperature span. By contrast, the profile of pMad is sensitive to the concentration of Dpp. The dosage sensitivity of Dpp is exceptional among morphogens and is singled out as being haploid-insufficient (Eldar, 2002).

No apparent transcriptional feedback, which might account for the robustness of dorsal patterning, has been identified so far. Robustness should thus be reflected in the design of interactions in the patterning network. To identify the mechanism underlying robustness, a general mathematical model of the dorsal patterning network was formulated. For simplicity, initial analysis was restricted to a single BMP class ligand (Scw or Dpp), a BMP inhibitor (Sog) and the protease (Tld). The general model accounted for the formation of the BMP-Sog complex, allowed for the diffusion of Sog, BMP and BMP-Sog, and allowed for the cleavage of Sog by Tld, both when Sog is free and when Sog is associated with BMP. Each reaction was characterized by a different rate constant (Eldar, 2002).

Extensive simulations were carried out to identify robust networks. At each simulation, a set of parameters (rate constants and protein concentrations) was chosen at random and the steady-state activation profile was calculated by solving three equations numerically. A set of three perturbed networks representing heterozygous situations was then generated by reducing the gene dosages of sog, tld or the BMP class ligand by a factor of two. The steady-state activation profiles defined by those networks were solved numerically and compared with the initial, nonperturbed network. A threshold was defined as a given BMP value (corresponding to the value at a third of the dorsal ectoderm in the nonperturbed network). The extent of network robustness was quantified by measuring the shift in the threshold for all three perturbed networks. Over 66,000 simulations were carried out, with each of the nine parameters allowed to vary over four orders of magnitude (Eldar, 2002).

As expected, in most cases (97.5%) the threshold position in the perturbed networks was shifted by a large extent (>50%). In most of those nonrobust cases, the BMP concentration was roughly uniform throughout the dorsal region. By contrast, Sog was distributed in a concentration gradient with its minimum in the dorsal midline, defining a reciprocal gradient of BMP activation. Thus, the key event in this nonrobust patterning mechanism is the establishment of a concentration gradient of Sog, which was governed by diffusion of Sog from its domain of expression outside the dorsal region, coupled with its cleavage by Tld inside the dorsal region. Although such a gradient has been observed, it is also compatible with other models (Eldar, 2002).

A small class of networks (198 networks, 0.3%) was identified in which a twofold reduction in the amounts of all three genes resulted in a change of less than 10% in the threshold position. Notably, in all of these robust cases, BMP was redistributed in a sharp concentration gradient that peaked in the dorsal midline. In addition, this concentration gradient decreases as a power-low distribution with an exponent n = 2, which indicates the uniqueness of the robust solution. In these cases, Sog was also distributed in a graded manner in the dorsal region. Analysis of the reaction rate constants of the robust networks showed a wide range of possibilities for most parameters. But two restrictions were apparent and defined the robust network design: (1) in the robust networks the cleavage of Sog by Tld was facilitated by the formation of the complex Sog-BMP; (2) the complex BMP-Sog was broadly diffusible, whereas free BMP was restricted (Eldar, 2002).

To identify how robustness is achieved, an idealized network was considered by assuming that free Sog is not cleaved and that free BMP does not diffuse. The steady-state activation profile defined by this network can be solved analytically; the solution reveals two aspects that are crucial for ensuring robustness. First, the BMP-Sog complex has a central role, by coupling the two processes that establish the activation gradient: BMP diffusion and Sog degradation. This coupling leads to a quantitative buffering of perturbations in gene dosage. Second, restricted diffusion of free BMP enables the system to store excess BMP in a confined spatial domain where Sog is largely absent. Changes in the concentration of BMP alter the BMP profile close to the dorsal midline but do not change its distribution in most of the dorsal region (Eldar, 2002).

The complete system, comprising Sog, Tld, Tsg, both Scw and Dpp, and their associated receptors was examined next. Two additional molecular assumptions are required to ensure the robustness of patterning. First, Sog can bind and capture the BMP class ligands even when the latter are associated with their receptors. Second, Dpp can bind Sog only when the latter is bound to Tsg. Indeed, it has been shown that, whereas Sog is sufficient for inhibiting Scw, both Tsg and Sog are required for inhibiting Dpp. This last assumption implies that Tsg functions to decouple the formation of the Scw gradient from the parallel generation of the Dpp gradient, ensuring that Scw and Dpp are transported to the dorsal midline independently by two distinct molecular entities (Eldar, 2002).

The complete model was solved numerically for different choices of rate constants. In particular, the effect of twofold changes in gene dosage was assessed. The steady-state activation profiles can be superimposed, indicating the robustness of the system. In addition, with the exception of Dpp, the expression of all other crucial network components can be altered by at least an order of magnitude before an effect on the position of a given threshold is observed. In the model, the lack of robustness to Dpp stems from its insufficient dosage. Note that the time taken to reach steady state is sensitive to these concentrations of protein. For the wide range of parameters that were used, however, the adjustment time does not exceed the patterning time. Flexible adjustment time thus facilitates the buffering of quantitative perturbations (Eldar, 2002).

This analysis has identified two principle molecular features that are essential for robust network design: first, free Sog is not cleaved efficiently -- an assumption that is supported by the in vitro finding that Sog cleavage by Tld requires BMP; second, the diffusion of free BMP is restricted. This is the central prediction of the theoretical study, namely, that Scw diffusion requires Sog, whereas Dpp diffusion requires both Sog and Tsg. Although several reports suggest that in wild-type embryos both Dpp and Scw are widely diffusible, their ability to diffuse in a sog or tsg mutant background has not been examined as yet (Eldar, 2002).

To monitor the diffusion of Scw or Dpp, the even-skipped (eve) stripe-2 enhancer (st2) was used to misexpress Dpp or Scw in a narrow stripe perpendicular to the normal BMP gradient. In transgenic embryos, dpp or scw RNA was detected in a stripe just posterior to the cephalic furrow. Initially the stripe was about 12 cells wide at early cleavage cycle 14, but refined rapidly to about 6 cells by late cycle 14. The st2-dpp and st2-scw embryos were viable, despite the high expression of these proteins as compared with their endogenous counterparts (Eldar, 2002).

The activation of the BMP pathway was monitored either by staining for pMad or by following dorsal expression of the target gene race, which requires high activation. Scw is a less potent ligand than is Dpp. This experimental setup could not be used to study Scw diffusion properties because expressing st2-scw did not alter the pattern of pMad or race expression in wild-type or sog^-/- embryos. By contrast, expression of st2-dpp led to an expansion of both markers in a region that extends far from the st2 expression domain, indicating a wide diffusion of Dpp in a wild-type background. Conversely, on expression of st2-dpp in sog^-/- or in tsg^-/- embryos, both markers were confined to a narrow stripe in the st2 domain. The width of this stripe was comparable to that of st2-dpp expression, ranging from 6 to 12 cells, indicating that Dpp does not diffuse from its domain of expression in the absence of Sog or Tsg. Taken together, these results show that both Sog and Tsg are required for Dpp diffusion, as predicted by the theoretical analysis (Eldar, 2002).

The computation ability of biochemical networks is striking when one considers that they function in a biological environment where the amounts of the network components fluctuate, the kinetics is stochastic, and sensitive interactions between different computation modules are required. Studies have examined the effect of these properties on cellular computation mechanisms, and robustness has been proposed to be a 'design principle' of biochemical networks. The applicability of this principle to morphogen gradient patterning has been shown during early development. Quantitative analysis can be used to assess rigorously the robustness of different patterning models and to exclude incompatible ones. The remaining, most plausible model points to crucial biological assumptions and serves to postulate the central feedback mechanisms. Applying the same modelling principles to other systems might identify additional 'design principles' that underlie robust patterning by morphogen gradients in development (Eldar, 2002).

Physical properties of Tld, Sog, Tsg and Dpp protein interactions are predicted to help create a sharp boundary in Bmp signals during dorsoventral patterning of the Drosophila embryo

Dorsal cell fate in Drosophila embryos is specified by an activity gradient of Decapentaplegic. Genetic and biochemical studies have revealed that the Sog, Tsg and Tld proteins modify Dpp activity at the post-transcriptional level. The predominant view is that Sog and Tsg form a strong ternary complex with Dpp that prevents it from binding to its cognate receptors in lateral regions of the embryo, while in the dorsalmost cells Tld is proposed to process Sog and thereby liberate Dpp for signaling. In this model, it is not readily apparent how Tld activity is restricted to the dorsal-most cells, since it is expressed throughout the entire dorsal domain. In this study, additional genetic and biochemical assays were developed to further probe the relationships between the Sog, Tsg, Tld and Dpp proteins. Using cell based assays, it has been found that the dynamic range over which Dpp functions for signaling is the same range in which Dpp stimulates the cleavage of Sog by Tld. In addition, the data support a role for Tsg in sensitizing the patterning mechanism to low levels of Dpp. It is proposed that the strong Dpp concentration dependence exhibited by the processing reaction, together with movement of Dpp by Sog and Tsg protein can help explain how Tld activity is confined to the dorsal-most region of the embryo through formation of a spatially dependent positive and negative reinforcement loop. Such a mechanism also explains how a sharp rather than smooth signaling boundary is formed (Shimmi, 2003).

According to the prevailing view, Sog, Tsg and Tld act to create a transport mechanism that helps promote Dpp diffusion from lateral regions of the embryos towards the dorsal side. According to this model, Sog would diffuse into the dorsal domain from its ventral lateral site of synthesis and capture Dpp, thereby preventing Dpp from binding to receptor. Net flux of Sog towards the dorsal side is envisioned to help transport Dpp and thereby increase its concentration in the dorsalmost tissue, which is destined to become the amnioserosa. Tld acts to liberate Dpp by cleaving Sog, and Dpp once released, will either be recaptured by another Sog molecule or bound to its receptors (Shimmi, 2003).

In order for the transport model to produce a Dpp concentration peak, the proper balance between binding affinities, diffusion rates and proteolytic processing is needed. Tsg has been suggested to have several activities that could influence this balance. In one model, Tsg would act to slow down the intrinsic rate of Sog cleavage by Tld. In this case, loss of Tsg is predicted to result in elevated processing of Sog. This should produce a sog loss-of-function phenotype, as is observed when molecular markers are examined. That data argues strongly against this possibility. First, it has been demonstrated that Tsg function is epistatic to Tld. If the tsg mutant phenotype is caused by excess Tld activity, then eliminating Tld should produce a tld loss-of-function phenotype. However, a tsg-like phenotype is observed where there is a general lowering and flattening of the Dpp activity gradient, as assayed by marker gene expression. In addition, biochemical studies reveal that Tsg actually enhances the ability of Tld to cleave Sog. Taken together, it is concluded that Tsg does not function during DV patterning to retard Tld proteolytic activity (Shimmi, 2003).

A second property has been attributed to Tsg: it alters the selection of Tld cleavage sites in Sog thereby producing novel Sog fragments with unique properties. In particular, a Sog fragment termed Supersog containing the first CR domain and a region of the spacer between CR1 and CR2 appears to be produced in vitro by the action of Tsg and Tld. Although the production of Supersog-like fragments are seen under the present reaction conditions described in this study, no enhancement in their production is seen upon Tsg addition. This may reflect loss of an unidentified component during purification or differences in the sensitivities of the CR1 antibodies used in the two studies. These issues are presently under examination. Whether Supersog-type molecules contribute to DV patterning in vivo is unclear. The fact that overexpression of Supersog can partially rescue tsg mutant embryos suggests that they could be important. A full resolution of the role of Supersog will need to await the results of in vivo rescue experiments employing mutants of the different Sog cleavage sites, especially those that lead to the production of Supersog-like fragments (Shimmi, 2003).

One of the primary findings in this report is that the rate of Sog cleavage is very sensitive to the level of the Dpp protein and varies substantially over a 10-fold range. Interestingly, this is the same Dpp concentration range within which low to maximal signaling occurs in S2 cell culture. Tsg sensitizes the system such that both the binding of Dpp to Sog as well as the rate of cleavage of Sog by Tld is stimulated by Tsg protein. Because in the invertebrate system, the binding of ligand to Sog is required for efficient processing of Sog, it is not surprising that the rate of Sog processing goes up in the presence of Tsg. This follows because, at a given concentration of Sog and Dpp, more complex will be formed in the presence of Tsg leading to a higher substrate concentration for the Tld protease. It is speculated that this system evolved in part to enable the embryo to produce a patterning mechanism that functions within the context of a very short developmental window. In Drosophila, the time between initial transcription of dpp during the early blastoderm stage and assignment of fate required for proper gastrulation is only about 40 minutes. In this short time-window, Dpp concentration must reach an effective signaling level. However, using a genomic Dpp-HA construct, it has been possible to visualize Dpp in the early embryo and it is present at much lower levels than in other tissues, such as the epidermis, at later stages of embryogenesis. It is proposed that under these conditions of low Dpp concentration, the presence of Tsg is required to enable Sog to bind to Dpp and to stimulate Sog cleavage in order to create a cyclic binding and release process that enables Dpp to be carried towards the dorsal midline. Furthermore, it is proposed that the intrinsic sensitivity of the cleavage reaction to the Dpp concentration is crucial for formation of a sharp signaling boundary. Thus, as the Dpp concentration drops in the lateral regions as a consequence of Dpp movement towards the dorsal side, the rate of Sog cleavage drops, allowing more Sog to enter this region and further reducing signaling in lateral regions. The movement of Dpp will simultaneously raise Dpp concentration in the dorsal region, further stimulating cleavage and clearance of Sog and thereby reinforcing Dpp signaling at the dorsal midline. This built-in positive and negative reinforcement mechanism should help establish sharp signaling boundaries by formation of steep ligand gradients, instead of the more gradual gradients that would form if Sog cleavage was not sensitive to the Dpp concentration (Shimmi, 2003).

In some vertebrate systems, DV patterning mechanisms have been conserved with respect to the molecules employed, but the polarity of axis over which they act has been inverted. Thus, in both amphibians and zebrafish, Bmp ligands specify ventral cell fates, whereas Bmp inhibitors, such as Chordin, are secreted from dorsal cells. In each of these systems, Tsg- and Tld-like proteins also contribute to axis formation, but the biochemical details of their associations appear different from those found in Drosophila. Two distinctions are most apparent and these probably have biological significance with respect to the patterning mechanism employed by these organisms. In Xenopus, the affinity of chordin for Bmps is significantly higher than Sog for Dpp; Bmps can be coimmunoprecipitated by chordin alone whereas this is not the case for the Drosophila components. In addition, once cleaved by Xolloid, at least some of the CR1 containing fragments of chordin continue to have significant affinity for the Bmp ligand preventing it from signaling (Shimmi, 2003 and references therein).

The second major difference between the Drosophila and Xenopus systems is that the Drosophila processing of Sog is dependant on prior binding of Sog to Dpp, while in Xenopus this is not the case. Rather, Chordin cleavage by Xolloid appears to be constitutive and is not enhanced by any tested ligand. Without ligand dependent cleavage, net movement of Bmps by Chordin diffusion may not readily occur nor would there be a mechanism to both positively and negatively reinforce the processing reaction. Indeed, recent studies have demonstrated that in the Drosophila embryo, Chordin does not have the ability to promote Dpp signaling at a distance, whereas Sog does. As a result, spatially enhanced Bmp concentrations and sharp signaling boundaries that result from net ligand movement by the activities of the Chordin, Xolloid and Tsg proteins may not occur in Xenopus. In fact there is no evidence in Xenopus that loss of Chordin activity actually results in a reduction in Bmp signaling in select regions of the embryo as occurs in Drosophila (Shimmi, 2003).

Despite these differences, Tsg may, nevertheless, play both positive and negative roles in modulating Bmp signaling; however, its mechanism is somewhat different. As processed fragments of Chordin still have reasonable affinity for ligand, they may need to be dislodged to allow for signaling. Tsg binding to Bmps appears to help promote this dislodgment and their ultimate degradation. In Drosophila, since Sog binds poorly to ligand in the absence of Tsg there is no need for Tsg to help promote dissociation of Sog fragments. Rather, it is its ability to help promote association of Sog with Dpp that is key to understanding its function. Tsg appears also to alter the rate of chordin proteolysis. Thus, at a high Tsg-to-chordin ratio, Chordin may be degraded and in this way Tsg might help promote signaling. It is possible that some combination of these properties is used in other vertebrates. For example, in zebrafish it has recently been shown that loss of chordin can enhance a phenotype that results from haplo-insufficiency for swirl, a gene that encodes Bmp2b. This paradoxical observation, that loss of an inhibitor exacerbates a phenotype resulting from loss of a ligand, is exactly analogous to the case of amnioserosa development in Drosophila where loss of Sog (an inhibitor) leads to less Dpp signaling in the dorsal domain. Detailed studies examining the ligand dependence of Chordin cleavage in zebrafish by minifin, the gene encoding a Tld homolog, have not been reported. It is possible therefore, that like Drosophila, this system may also employ a transport mechanism involving Tsg, Chordin and Tld that acts to boost Bmp signaling in specific tissues. It is interesting to note that the mouse homologs of Tsg, Chordin and Tld also exhibit their own distinct biochemical properties. Thus, a new Tld processing site in Chordin is induced by the presence of Tsg but this is not seen when the Xenopus components are used. Thus, it seems probable that the inherent complexity of this multi-component regulatory mechanism has provided numerous targets for evolutionary change. It is speculated that these changes account for the remarkable diversity that this mechanism exhibits with respect to the actual details by which it regulates Bmp signaling in different organisms (Shimmi, 2003).

A model for formation of the BMP activity gradient in the Drosophila embryo

The dorsoventral axis of the Drosophila embryo is patterned by a gradient of bone morphogenetic protein (BMP) ligands. In a process requiring at least three additional extracellular proteins, a broad domain of weak signaling forms and then it abruptly sharpens into a narrow dorsal midline peak. Using experimental and computational approaches, how the interactions of a multiprotein network create the unusual shape and dynamics of formation of this gradient was investigated. Starting from observations suggesting that receptor-mediated BMP degradation is an important driving force in gradient dynamics, a general model is developed that is capable of capturing both subtle aspects of gradient behavior and a level of robustness that agrees with in vivo results (Mizutani, 2005).

This study began by showing that robustness with respect to variations in the expression of single genes is not a characteristic of this system. This is an important observation, given that considerable attention has been focused lately on the robustness of morphogen-patterning systems, as well as biological signaling in general. The fact that sog^-/+ embryos eventually develop normally underscores the ability of embryos to compensate at later stages for early errors. It is not clear why marked effects of sog heterozygosity were not seen previously in previous experiments (Mizutani, 2005).

The diffusibility of BMPs in the embryo in the presence and absence of Sog was examined. By examining embryos in which Dpp is ectopically expressed, it was observed that the range of Dpp action is reduced in the absence of Sog, but still substantial, and consistent with unhindered diffusion. By observing the rate at which continuous ectopic Dpp expression gives rise to an unchanging response profile, it was also possible to infer that Dpp must undergo rapid degradation, presumably through receptor-dependent means. In these experiments, levels of expression of ectopic Dpp were not high (2.5-fold above normal when two copies of st2-dpp were present; presumably only slightly above normal when one copy was present (Mizutani, 2005).

The above observations were used to produce a simplified model of gradient formation. The goal was not necessarily to reproduce all aspects of the in vivo gradient, but rather to begin with a minimum number of elements -- and as few assumptions as possible -- and then ask which of the behaviors of the in vivo gradient could be captured. Interestingly, a great many of those behaviors emerge from a model in which a single ligand (e.g., Dpp or a Dpp/Scw heterodimer) diffuses freely, is degraded by receptors, forms a complex with Sog and Tsg, and is released from that complex when Tld cleaves Sog. These behaviors include rapid dynamics, formation of a broad domain of weak dorsal signaling that abruptly refines to a sharp midline peak, and peak narrowing or broadening when sog dosage is either increased or decreased, respectively. These behaviors depend upon the combined presence of Sog, Tsg, and Tld and are also highly sensitive to dpp dosage. Interestingly, highly localized expression of Tld and an absolute dependence of Sog cleavage on Dpp are not essential. Also not critical is the order of assembly of Dpp-Sog-Tld complexes (Mizutani, 2005).

Although the ability of the model to form a midline peak of BMP activity exemplifies the Sog/Tld-dependent 'shuttling', that mechanism does not give a complete picture of events for two reasons: (1) the abrupt onset of midline peak growth after a substantial plateau phase reflects a BMP-catalyzed chain reaction of Sog destruction that is independent of BMP transport per se; (2) calculations show that any soluble inhibitor has the ability to expand the range of action of a morphogen simply by protecting it from receptor-mediated destruction. Indeed, this effect alone could underlie some of the greater range of action of ectopically expressed Dpp in wild-type versus sog^- embryos (Mizutani, 2005).

At least one feature of the model that does not match in vivo observations, even when investigated over a wide range of parameter values, is the magnitude of the effect of sog heterozygosity on PMad peak width. The results suggest a near doubling of peak width, whereas calculations predict a more modest increase. Even accounting for the nonlinearity of immunohistochemistry and the fact that PMad may not be an instantaneous read-out of BMP receptor occupancy, the data suggest that other processes, not captured in the simple model, regulate the shape of PMad peaks. For example, it might be necessary to include the effects of a novel truncated form of Sog that promotes, rather than inhibits, BMP signaling (Mizutani, 2005).

One process that seems especially likely to shape PMad peaks is a BMP-driven, transcription-dependent feedback loop that has very recently been shown to markedly amplify high and depress low levels of BMP signaling in the Drosophila embryo. Such feedback could not only modify the shapes of PMad peaks, but also potentially explain another peculiarity of the model, which is that its peak heights and widths best fit mutant data when they are looked at up to the 30-45 min period, but not much later (i.e., not in the mathematical steady state). Since positive-feedback regulation of BMP signaling can be expected to both sharpen and maintain patterns that might otherwise have continued to evolve, it is perhaps not surprising that, at long enough times, in vivo behavior diverges from predictions of the model. Put another way, this issue serves as a reminder that, unlike BMP gradients at larval stages of Drosophila development (e.g., in the imaginal discs), the embryonic BMP gradient forms and acts so rapidly that there is little justification for assuming that steady-state calculations should reproduce in vivo observations. Indeed, it is only by considering the dynamics of gradient formation that the model presented here is able to explain the seemingly paradoxical result that decreased dorsal midline PMad staining in dpp^-/+ embryos can be rescued by lowered sog dosage, when loss of sog function, by itself, is associated with decreased dorsal midline PMad staining (Mizutani, 2005).

In summary, the results presented here indicate that known properties of the molecules required for formation of the Drosophila embryonic BMP gradient are sufficient to account for many aspects of gradient dynamics, shape, and robustness, with no need for assumptions such as lack of diffusion of free BMP, transient BMP synthesis, removal of BMP from its receptors by Sog, or attainment of a steady state. Although computational data indicate that a Sog/Tld-dependent shuttling mechanism plays a key role in shaping and timing this BMP gradient, other dynamic processes appear to participate as well (Mizutani, 2005).

Tolloid-related processes Sog in order to help specify the posterior crossvein in the Drosophila wing

Tolloid (Tld) and Tolloid related (Tlr) belong to a family of developmentally important proteases that includes Bone Morphogenetic Protein 1 (Bmp1). Tld is required early in Drosophila development for proper patterning of dorsal embryonic structures, whereas Tlr is required later during larval and pupal stages of development. The major function of Tld is to augment the activity of Decapentaplegic (Dpp) and Screw (Scw), two members of the Bmp subgroup of the Tgfßü superfamily, by cleaving the Bmp inhibitor Short gastrulation (Sog). Evidence is presented that Tlr also contributes to Sog processing. Tlr cleaves Sog in vitro in a Bmp-dependent manner at the same three major sites as does Tld. However, Tlr shows different site selection preferences and cleaves Sog with slower kinetics. To test whether these differences are important in vivo, the role of Tlr and Tld during development of the posterior crossvein (PCV) in the pupal wing was investigated. tlr mutants lack the PCV as a result of too little Bmp signaling. This is probably caused by excess Sog activity, since the phenotype can be suppressed by lowering Sog levels. However, Tld cannot substitute for Tlr in the PCV; in fact, misexpressed Tld can cause loss of the PCV. Reducing levels of Sog can also cause loss of the PCV, indicating that Sog has not only an inhibitory but also a positive effect on signaling in the PCV. It is proposed that the specific catalytic properties of Tlr and Tld have evolved to achieve the proper balance between the inhibitory and positive activities of Sog in the PCV and early embryo, respectively. It is further suggested that, as in the embryo, the positive effect of Sog upon Bmp signaling probably stems from its role in a ligand transport process (Serpe, 2005).

The major distinction between the two Drosophila proteases in terms of their Sog processing function is the time and tissue in which they act. Tld activity is primarily confined to the early embryo, while Tlr is required during pupal wing development. To some extent, the functional differences between them can be attributed simply to differences in expression pattern. In the pupal wing Tlr is far more abundantly expressed than Tld, and this alone might account for the lack of redundancy. In the early embryo, however, the situation is more complex. Both enzymes are expressed with similar profiles, but Tlr does not seem to be capable of providing sufficient Sog processing activity, even when several extra copies are provided as transgenes (Serpe, 2005).

It has been speculated that this difference in activity might reflect differences in activation of the proteases at the level of pro-peptide removal. Like all the members of the Bmp1 family, Tld and Tlr are secreted as pro-enzymes; the processing of the pro-peptide is necessary for the activation of proteolytic activity, since the N-terminal end of the astacin motif is buried inside the catalytic domain forming an internal salt bridge. Mutations at the processing site render the enzymes inactive, whereas removal of the pro-peptides produce activated forms of Tld and Tlr. Tlr has a much longer pro-peptide that could either aid or inhibit activation in a tissue-specific manner. However, the inability of Tlr to rescue Tld mutants does not appear to result from an inefficient activation step. Tld activation, both in the embryo and in S2 cells, is very inefficient with most of the protein found in the pro-enzyme state. By contrast, pro-peptide removal from Tlr is very efficient in S2 cells, and the same is true when Tlr is ectopically expressed in the embryo (Serpe, 2005).

Instead, it seems likely that the difference in kinetics of Sog processing by Tlr is the reason behind the inability of Tlr to rescue tld mutants. Tlr is much less efficient at cleaving Sog in vitro than Tld. Given the rapid developmental time of early embryogenesis, where patterning by Bmps during cellularization occurs within approximately a 30 minute time window, the slower kinetics of Sog processing by Tlr may not support proper patterning. Indeed, computational work has shown that a three to fourfold reduction in kinetic properties of Tld will completely disrupt the patterning process (Serpe, 2005).

Although the slow processing kinetics of Tlr towards Sog may prevent it from functioning effectively in early embryonic patterning, this property may be exactly what is required to achieve proper formation of the PCV. Unlike patterning in the early embryo, formation of the PCV, as assessed by profile of pMad accumulation, occurs over at least a 7 hour time frame. The slower processing rate of Tlr towards Sog may be required to achieve the appropriate balance of Sog destruction and diffusion that is necessary for proper patterning to occur. Consistent with this view, overexpression of various UAS-tld lines under the control of the A9-Gal4 driver in a tlr mutant background does not rescue PCV formation. In fact, in many cases overexpression of an activated Tld, or co-expression of wild type tld and tlr together produce loss of the crossvein tissue in a wild-type background. It is envisioned that, under these conditions, the increased level of enzymatic activity results in over-digestion of Sog, a situation that would phenocopy sog hypomorphs. Consistent with this view, hypomorphic sog allelic combinations also result in the loss of the PCV. In addition, large sog null clones can also result in loss of the PCV (Serpe, 2005).

In a Xenopus assay, it was found that Tlr is only slightly less efficient than Tld at reverting secondary axis induction caused by Sog. Although it is not known how well each enzyme is activated in this animal, it should be noted that the developmental time period over which the patterning process functions in Xenopus is long compared with early Drosophila development. The longer time frame may enable the less efficient protease to produce a similar biological response. Protease domain swap experiments suggest that the reduced processing rate does not involve evolution of intrinsic differences in the catalytic abilities of the protease domain itself, but rather changes occur in the way that the Sog substrate initially interacts with the enzyme. In summary, it is proposed that during evolution there was selection for particular biophysical properties of these two enzymes to properly match the developmental time frame over which the patterning mechanisms operate. It cannot be exclude however, that other differences besides kinetic activity might also play a role in providing functional specificity. For example, it is possible that variation in cleavage site selection might also contribute to the different biological activities of Tlr and Tld. It is worth noting in this regard that different fragments of Sog have been shown to have both positive and negative effects when overexpressed in the wing. However the in vivo roles of endogenous Sog fragments have not been defined (Serpe, 2005).

The results suggest that a proper balance of Sog and protease activity is necessary to pattern the PCV. Interestingly, the same situation holds true in the early embryo. In this case, Sog plays both positive and negative roles in patterning the dorsal domain. It is required in the dorsolateral regions to block Bmp signaling, but it also acts as an agonist to achieve peak levels of Bmp signals at the dorsal midline. Two types of models have been proposed to account for these dual activities. In one model, the different cleavage fragments of Sog are thought to provide either agonist or antagonist function, but the details of the mechanism are unclear. In the other model, both functions are proposed to come about as a result of Sog providing a transport mechanism that spatially redistributes Bmp ligands from the lateral region to the dorsal most cells. This transport mechanism also requires the activity of Tsg, a small cysteine-rich secreted protein which has been shown to form a tripartite complex with Sog and Dpp. The prevailing view is that as Sog diffuses into the dorsal domain it forms a high affinity complex with Tsg and Dpp. This complex is unable to bind to receptors and is responsible for the antagonistic activity of Sog. At the same time, the complex protects Dpp from degradation and receptor binding allowing it to diffuse and accumulate dorsally where it is released by Tld processing. The ability of Sog to redistribute the Bmp ligands accounts for the agonist function of Sog. Computational analyses have provided additional support for this model (Serpe, 2005).

It is proposed that the same type of mechanism may be responsible for patterning the PCV. Recent analysis has provided evidence that the longitudinal veins act as the source of Dpp for PCV specification. Dpp is thought to diffuse from these veins into a PCV competent zone. The exact mechanism by which the competent zone is specified is not clear, but low levels of Sog expression are required. tlr is expressed within the PCV competent zone during the initial stages of crossvein development, suggesting the Tlr:Sog ratio will be higher in this region. Furthermore, because processing of the Sog/Dpp complex by Tld-like enzymes is dependent on the Dpp concentration, the complex will be most efficiently processed in the center of the competent zone (Serpe, 2005).

According to this model, there is limited processing of Sog and therefore limited release of Dpp from its inhibitor in tlr mutants. Conversely, Sog also supplies a positive function for PCV formation, probably by providing a transport mechanism for Dpp, accounting for the partial loss of the PCV in hypomorphic sog mutants and complete loss of the PCV in large sog-null clones. The partial reversion of the tlr mutant phenotype by introduction of hypomorphic sog alleles is also consistent with the view that it is the balance between these two factors that is crucial for proper patterning. Interestingly, this is the way in which Sog was originally identified as an inhibitor of Dpp signaling in the embryo: weak sog alleles were isolated as partial suppressors of tld mutations. One difference is that, in the case of this partial reversion, lowering Sog levels is able to revert a null mutation instead of a hypomorphic condition, as was the case in the embryo. There are at least two possibilities that can explain this suppression effect. First, although these animals may be null for tlr, there could be some low level tld expression in the pupal wing. If so, then these wings would not be devoid of all Sog-processing activity and therefore lowering Sog levels might enable the limited amount of Tld to provide the proper production-destruction balance. Alternatively, neither Sog nor Tlr may be absolutely required for PCV formation. Instead, their functions may be simply to ensure that the patterning occurs reproducibly. Thus, in the absence of both Sog and Tlr, partial PCV formation may occur as a result of some Bmp ligand accumulating in the correct position. However, under these circumstances, the patterning mechanism would be unreliable and would produce different results on case-by-case basis. To prevent this from occurring, it is posited that evolution has selected for supplementary regulatory controls involving Sog and Tlr to ensure that the PCV always forms completely and reliably at the correct position (Serpe, 2005).

Two additional observations make the comparison between formation of the PCV and establishing the high point in embryonic Bmp signaling even more compelling. (1) In the embryo, Tsg is required to enable Sog to bind to Dpp and Scw to achieve peak levels of Bmp signaling. Although Tsg and Scw are not transcribed in the pupal wing, a Tsg-related gene, encoded by the crossveinless (cv) locus, is expressed in the pupal wing. Since cv mutants exhibit a crossveinless phenotype, it seems likely that Cv functionally substitutes for Tsg during PCV formation. (2) Gbb, another Bmp-like ligand, may functionally replace Scw, since gbb hypomorphic mutations lack the PCV and associated Bmp signaling (Serpe, 2005).

A major distinction between embryonic amnioserosa development and PCV formation is that PCV specification also requires the activity of Cv2, a protein that contains cysteine-rich repeats similar to those found in Sog, while amnioserosa specification does not, despite the expression of Cv2 in those cells. Vertebrate homologs of Cv2 can bind Bmps, and act variously as agonists or antagonists of Bmp signaling in different assays. It is not clear by what mechanism Cv2 promotes Bmp signaling during PCV formation. It is also not clear why Cv2 is not required in the early embryo, even though it is expressed in dorsal blastoderm cells (Serpe, 2005).

Finally, Tlr plays additional roles during development, besides processing Sog for specification of the PCV. In contrast to cv and cv2 null mutations, which result in homozygous viable and fertile flies, most tlr mutant animals die during larval stages when there is no known requirement for Sog. In addition, although reducing Sog levels does suppress the PCV defect observed in the tlr mutant escaper flies, it does not increase the frequency of eclosing animals. Therefore Tlr may be required for processing of some other essential component(s) during Drosophila development (Serpe, 2005).

Facilitated transport of a Dpp/Scw heterodimer by Sog/Tsg leads to robust patterning of the Drosophila blastoderm embryo

Patterning the dorsal surface of the Drosophila blastoderm embryo requires Decapentaplegic (Dpp) and Screw (Scw), two BMP family members. Signaling by these ligands is regulated at the extracellular level by the BMP binding proteins Sog and Tsg. Tsg and Sog play essential roles in transporting Dpp to the dorsal-most cells. Furthermore, biochemical and genetic evidence is presented that a heterodimer of Dpp and Scw, but not the Dpp homodimer, is the primary transported ligand and that the heterodimer signals synergistically through the two type I BMP receptors Tkv and Sax. It is proposed that the use of broadly distributed Dpp homodimers and spatially restricted Dpp/Scw heterodimers produces the biphasic signal that is responsible for specifying the two dorsal tissue types. Finally, it is demonstrated mathematically that heterodimer levels can be less sensitive to changes in gene dosage than homodimers, thereby providing further selective advantage for using heterodimers as morphogens (Shimmi, 2005).

The suggestion that the facilitated transport of a BMP signaling molecule might be the primary mechanism that generates pattern within the dorsal domain of the Drosophila blastoderm embryo (Holley, 1996) was a conceptual breakthrough, since it could account for the paradoxical abilities of Sog and Tsg to have both positive and negative effects on patterning. However, there was no direct evidence that either Dpp or Scw actually concentrated to the midline. In addition, it did not explain the roles of Dpp and Scw in producing the restricted high-level signaling output at the midline, as measured by p-Mad accumulation, nor did it explain how a lower level of signal was maintained in the more lateral regions to help fate the future dorsal ectoderm. Lastly, it was not apparent how the system achieves resiliency to changes in gene dosages of certain components. The experimental and computational observations described in this study have addressed these issues (Shimmi, 2005).

One of the primary findings is that Dpp and Scw form heterodimers both in tissue culture and in vivo and that these heterodimers are able to synergistically stimulate phosphorylation of Mad in cell culture. Since the Dpp/Scw heterodimers have highest affinity for Sog and Tsg, it is inferred that the heterodimer is the primary ligand transported dorsally by Sog and Tsg, resulting in high levels of p-Mad accumulation at the dorsal midline just prior to gastrulation. Consistent with this view, it was found that Dpp localization to the midline depends on Scw (Shimmi, 2005).

In addition to heterodimers being the preferred translocated species, the heterodimer model also explains the mechanism by which Scw contributes to dorsal patterning. This issue has been enigmatic since scw and its receptor, Sax, are expressed ubiquitously in the early embryo, yet signal output is limited to dorsal cells. In addition, misexpression of Scw or activated Sax produces very limited effects in most tissues, while misexpression of Dpp or activated Tkv results in very dramatic consequences. A partial resolution to this issue was suggested by the finding that coexpression of activated Sax and activated Tkv in embryos or imaginal discs produces a synergistic signal, implying that both the Sax and Tkv signals are necessary for a robust output. However, it has remained unclear whether endogenous, nonactivated receptors can produce a synergistic signal in response to ligands. As described in this study, the formation of a heterodimer between Dpp and Scw resolves these issues. In tissue culture assays, Scw homodimers produce very limited signal, while Dpp homodimers produce a moderate signal requiring only the Tkv receptor. The differential signaling ability of each homodimer explains their nonequivalence in producing patterning abnormalities when misexpressed during development. In contrast, the Dpp/Scw heterodimer is able to produce a synergistic phosphorylation of Mad that requires both the Tkv and Sax receptors; simply mixing homodimers is not sufficient. These observations demonstrate that synergistic signaling occurs at the level of receptor-mediated Mad phosphorylation and not through integration of separate signals at downstream targets. The molecular mechanism by which the Tkv and Sax receptors produce a synergistic output remains unclear (Shimmi, 2005).

Although the original role for Scw in dorsal patterning invoked formation of a heterodimer as the primary signaling species, this model fell into disfavor because ventral injection of scw mRNA or ventral expression of scw from the twist promoter can partially rescue amnioserosa formation. Since disulfide-linked heterodimer formation of TGF-β type ligands is known to occur in the Golgi during the secretion process, ventral expression of Scw without Dpp should preclude formation of heterodimers, and, therefore, any rescuing activity should be brought about by homodimers. Although some rescue was observed in these experiments, it is important to note that even multiple copies of ventrally expressed Scw do not lead to viability. In contrast, a single copy of Scw expressed in the dorsal domain using the tld promoter gives complete viability and fertility. In addition, these experiments assume that there is no internalization within the dorsal domain of Scw homodimers followed by isomerization with Dpp and resecretion. This possibility is mechanistically very similar to models in which Dpp is proposed to undergo transcytosis. Therefore, while ventral overexpression of Scw homodimers may have some ability to compensate for loss of Scw dorsally, normal patterning is most efficiently achieved when Scw is expressed in a domain in which heterodimers can form (Shimmi, 2005).

BMP-directed patterning of dorsal blastoderm cells ultimately results in the specification of two tissues, amnioserosa and dorsal ectoderm. In general, these tissues derive from cells receiving high and low BMP signal, respectively. Whether there are additional cell fate subdivisions specified within the steep signaling transition zone is not clear, although cells can discriminate subtle signaling differences as evidenced by the slightly wider expression pattern of the BMP target genes rho and usp compared to zen and race. Although both Dpp and Scw are required to establish the high point of signaling necessary to specify amnioserosa, only Dpp is needed to specify dorsal ectoderm. This is consistent with observations that the Dpp/Scw heterodimer will be preferentially concentrated at the midline because of its high affinity for Sog and Tsg. In contrast, Dpp and Scw homodimers will be more broadly distributed because of their lower affinities for Sog and Tsg. Although the different species cannot be directly distinguished in vivo, analysis of downstream target genes in a scw mutant embryo revealed that there is sufficient BMP activity to activate pnr transcription, but its pattern is very wide, consistent with the observed broad distribution of Dpp homodimers. In the wild-type case, Dpp and Scw homodimers, together with a small number of heterodimers that escape from Sog and Tsg, may contribute to signaling in the lateral ectoderm, since the pnr signal is stronger in wild-type than in scw mutants. These homodimers also likely signal in a repressive manner to prevent ectopic transcription of neurogenic genes within the dorsal domain. Thus, patterning of dorsal tissue appears to take advantage of the differing properties of homo- and hetero-dimers to establish a biphasic signaling state. Specifically, selective transport of the heterodimer and synergistic receptor signaling produce a restricted high point and amnioserosa cell fate, while Dpp and Scw homodimers generate a broad low level of signal that help fate the future dorsal ectoderm and restrict neurogenic activity to more lateral regions. It is likely that the full specification of dorsal ectoderm does not occur until a second round of dpp transcription takes place after germ band extension. It is also likely that additional components help reinforce the formation of the biphasic state, since recent genetic data indicate that tight localization of Dpp to the midline requires an initial phase of low-level Dpp signal reception (E.L. Ferguson, personal communication). The suggestion is that this initial low-level Dpp signal induces expression of an additional component that participates in the localization process. The identity of this component remains elusive (Shimmi, 2005).

Lastly, it is noted that employment of heterodimers in early embryonic patterning may be a common theme. In zebrafish, both BMP2b and BMP7 are required for dorsal-ventral patterning, and loss-of-function mutations in each gene exhibit identical severely dorsalized phenotypes. Since this phenotype is not enhanced in double mutants and overexpression of these two gene products reveals synergy in the ventralization of wild-type embryos, it has been suggested that BMP2a/BMP7 heterodimers are the primary molecules that specify ventral cell fates in this organism. These observations further highlight the overall similarity in the molecular components used to pattern the early zebrafish and Drosophila (Shimmi, 2005).

Use of the Dpp/Scw heterodimer provides the patterning system with an effective buffer at a very early step in dorsal cell fate specification. Buffering for reductions of Scw or Dpp is predominantly determined by the relative monomer production rates, and if Scw is in slight excess with respect to Dpp, reductions in the levels of Scw will have little effect on the output Dpp/Scw heterodimer, regardless of the specific choices of parameters (Shimmi, 2005).

Patterning is also resilient to reductions of Sog and Tsg. Sog and Tsg have synergistic BMP binding activity and the concentration of Sog/Tsg in the PV space is governed by the interaction of reaction and diffusion. The Sog/Tsg ratio can be computed as described for Dpp/Scw to determine the compensation in this subsystem, and the results are different from those for Dpp/Scw. Now there are two distinct solution regions, one for small β (β is the ratio of the wild-type production rates for monomer Dpp to monomer Scw) where many choices of parameters provide significant compensation for reductions of gene dosage, and one for large β where there is virtually no compensation. Because the behavior for large β and small β is very different, this analysis can explain the compensation for reductions in either Sog or Tsg but not both. This suggests that other mechanisms must be involved to explain the experimentally observed resilience in both sog and tsg heterozygous embryos. These could include the following: (1) the spatial separation of Sog and Tsg expression, (2) downstream kinetic mechanisms that compensate after Sog/Tsg formation, or (3) both. Both may contribute, but the following focuses on the possible effects of compensation in downstream kinetic interactions (Shimmi, 2005).

After Sog/Tsg formation, the next step downstream is the binding of the inhibitor Sog/Tsg to Dpp/Scw. Experimentally, it is observed that Tolloid cleavage of Sog is greatly enhanced when bound to Dpp/Scw and is enhanced in the presence of Tsg. In addition, a previous mathematical model of BMP patterning suggested that cleavage of Sog (only when bound in the complex Sog/BMP) is a requirement for the system to exhibit resilience to changes in gene dose of sog, tsg, or scw. These data support the idea that Dpp/Scw transported from the broad dorsal region must be released from the Sog/Tsg/Dpp/Scw complex. Interestingly, the local dynamics of Sog/Tsg + Dpp/Scw complex formation are completely analogous to the local dynamics for Sog + Tsg complex formation. This suggests that, if the level of Dpp/Scw or Sog/Tsg is decreased from the original wt levels, the output complex Sog/Tsg/Dpp/Scw would be less affected. Taken together, the Sog/Tsg and Sog/Tsg/Dpp/Scw steps lead to a cascade in which the compensation in the first step is enhanced in the second step. In effect, the output from one complex formation stage becomes the input substrate for the next stage. Of course, the level of buffering achieved depends on the system parameters. The output suggests that patterning would be most compensated for reduction of Scw, followed by Tsg, then Sog, and lastly Dpp. Of course, other downstream steps may also contribute to compensation (Shimmi, 2005).

In reality, patterning involves diffusive transport as well, but the analysis shows how a cascade of stages can produce compensation in the kinetic steps. When the full BMP patterning model that incorporates transport is compared to a previous model mediated by homodimers and monomers, there are approximately 100 times more 'robust' hits when scw^+/−, sog^+/−, tsg^+/−, and tld^+/− cases are considered. In principle, the binding cascade analysis extends to other systems and can be used to explore other changes of input, including overexpression of a protein (Shimmi, 2005).

Post-transcriptional Regulation

Lack of abstrakt function does not compromise the ability of cells to undergo their differentiation program or cellular behaviour typical of their differentiated state. Thus, neurite outgrowth still occurs, and pole cells are able to migrate over large distances, but both types of cells appear to fail to recognize or react to the cues that direct their morphogenetic behavior in the spatially appropriate manner. One reason for this defect might be a loss of subcellular order or polarity. Various aspects of cell polarity were indeed affected in abs mutants. The first abnormality detected was in the localization of mRNAs in the blastoderm. In the wild-type blastoderm, the mRNAs of several genes are distributed unevenly in the cell. For example, the transcripts of short gastrulation (sog) and crumbs (crb) are tightly apposed to the apical cell surface. In abs^14B mutants, CRB mRNA is not seen apically but predominantly at the level of the nuclei, while SOG mRNA forms a gradient from the apical towards the basal side of the cell. In the case of crb, the mislocalization of RNA is also seen at later stages and in other tissues. For example, the tight apical localization in the hindgut epithelium is disturbed in abs mutants. Remarkably, the Crumbs protein is synthesized properly and targeted to the apical cell surface, showing that apical-basal cell polarity in the epithelial cells is correctly established and maintained in abs mutants (Irion, 1999).

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

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