Single-cell RNA sequencing of mid-to-late stage spider embryos: new insights into spider development.

BACKGROUND: The common house spider Parasteatoda tepidariorum represents an emerging new model organism of arthropod evolutionary and developmental (EvoDevo) studies. Recent technical advances have resulted in the first single-cell sequencing (SCS) data on this species allowing deeper insights to be gained into its early development, but mid-to-late stage embryos were not included in these pioneering studies. RESULTS: Therefore, we performed SCS on mid-to-late stage embryos of Parasteatoda and characterized resulting cell clusters by means of in-silico analysis (comparison of key markers of each cluster with previously published information on these genes). In-silico prediction of the nature of each cluster was then tested/verified by means of additional in-situ hybridization experiments with additional markers of each cluster. CONCLUSIONS: Our data show that SCS data reliably group cells with similar genetic fingerprints into more or less distinct clusters, and thus allows identification of developing cell types on a broader level, such as the distinction of ectodermal, mesodermal and endodermal cell lineages, as well as the identification of distinct developing tissues such as subtypes of nervous tissue cells, the developing heart, or the ventral sulcus (VS). In comparison with recent other SCS studies on the same species, our data represent later developmental stages, and thus provide insights into different stages of developing cell types and tissues such as differentiating neurons and the VS that are only present at these later stages.

Evolution of the Spider Homeobox Gene Repertoire by Tandem and Whole Genome Duplication.

Gene duplication generates new genetic material that can contribute to the evolution of gene regulatory networks and phenotypes. Duplicated genes can undergo subfunctionalization to partition ancestral functions and/or neofunctionalization to assume a new function. We previously found there had been a whole genome duplication (WGD) in an ancestor of arachnopulmonates, the lineage including spiders and scorpions but excluding other arachnids like mites, ticks, and harvestmen. This WGD was evidenced by many duplicated homeobox genes, including two Hox clusters, in spiders. However, it was unclear which homeobox paralogues originated by WGD versus smaller-scale events such as tandem duplications. Understanding this is a key to determining the contribution of the WGD to arachnopulmonate genome evolution. Here we characterized the distribution of duplicated homeobox genes across eight chromosome-level spider genomes. We found that most duplicated homeobox genes in spiders are consistent with an origin by WGD. We also found two copies of conserved homeobox gene clusters, including the Hox, NK, HRO, Irx, and SINE clusters, in all eight species. Consistently, we observed one copy of each cluster was degenerated in terms of gene content and organization while the other remained more intact. Focussing on the NK cluster, we found evidence for regulatory subfunctionalization between the duplicated NK genes in the spider Parasteatoda tepidariorum compared to their single-copy orthologues in the harvestman Phalangium opilio. Our study provides new insights into the relative contributions of multiple modes of duplication to the homeobox gene repertoire during the evolution of spiders and the function of NK genes.

New insights into mesoderm and endoderm development, and the nature of the onychophoran blastopore.

BACKGROUND: Early during onychophoran development and prior to the formation of the germ band, a posterior tissue thickening forms the posterior pit. Anterior to this thickening forms a groove, the embryonic slit, that marks the anterior-posterior orientation of the developing embryo. This slit is by some authors considered the blastopore, and thus the origin of the endoderm, while others argue that the posterior pit represents the blastopore. This controversy is of evolutionary significance because if the slit represents the blastopore, then this would support the amphistomy hypothesis that suggests that a slit-like blastopore in the bilaterian ancestor evolved into protostomy and deuterostomy. RESULTS: In this paper, we summarize our current knowledge about endoderm and mesoderm development in onychophorans and provide additional data on early endoderm- and mesoderm-determining marker genes such as Blimp, Mox, and the T-box genes. CONCLUSION: We come to the conclusion that the endoderm of onychophorans forms prior to the development of the embryonic slit, and thus that the slit is not the primary origin of the endoderm. It is thus unlikely that the embryonic slit represents the blastopore. We suggest instead that the posterior pit indeed represents the lips of the blastopore, and that the embryonic slit (and surrounding tissue) represents a morphologically superficial archenteron-like structure. We conclude further that both endoderm and mesoderm development are under control of conserved gene regulatory networks, and that many of the features found in arthropods including the model Drosophila melanogaster are likely derived.

Expression of netrin and its receptors uncoordinated-5 and frazzled in arthropods and onychophorans suggests conserved and diverged functions in neuronal pathfinding and synaptogenesis.

BACKGROUND: Development of the nervous system and the correct connection of nerve cells require coordinated axonal pathfinding through an extracellular matrix. Outgrowing axons exhibit directional growth toward or away from external guidance cues such as Netrin. Guidance cues can be detected by growth cones that are located at the end of growing axons through membrane-bound receptors such as Uncoordianted-5 and Frazzled. Binding of Netrin causes reformation of the cytoskeleton and growth of the axon toward (or away from) the source of Netrin production. RESULTS: Here, we investigate the embryonic mRNA expression patterns of netrin genes and their potential receptors, uncoordinated-5 and frazzled in arthropod species that cover all main branches of Arthropoda, that is, Pancrustacea, Myriapoda, and Chelicerata. We also studied the expression patterns in a closely related outgroup species, the onychophoran Euperipatoides kanangrensis, and provide data on expression profiles of these genes in larval tissues of the fly Drosophila melanogaster including the brain and the imaginal disks. CONCLUSION: Our data reveal conserved and diverged aspects of neuronal guidance in Drosophila with respect to the other investigated species and suggest a conserved function in nervous system patterning of the developing appendages.

Developmental and genomic insight into the origin of the tardigrade body plan.

Tardigrada is an ancient lineage of miniaturized animals. As an outgroup of the well-studied Arthropoda and Onychophora, studies of tardigrades hold the potential to reveal important insights into body plan evolution in Panarthropoda. Previous studies have revealed interesting facets of tardigrade development and genomics that suggest that a highly compact body plan is a derived condition of this lineage, rather than it representing an ancestral state of Panarthropoda. This conclusion was based on studies of several species from Eutardigrada. We review these studies and expand on them by analyzing the publicly available genome and transcriptome assemblies of Echiniscus testudo, a representative of Heterotardigrada. These new analyses allow us to phylogenetically reconstruct important features of genome evolution in Tardigrada. We use available data from tardigrades to interrogate several recent models of body plan evolution in Panarthropoda. Although anterior segments of panarthropods are highly diverse in terms of anatomy and development, both within individuals and between species, we conclude that a simple one-to-one alignment of anterior segments across Panarthropoda is the best available model of segmental homology. In addition to providing important insight into body plan diversification within Panarthropoda, we speculate that studies of tardigrades may reveal generalizable pathways to miniaturization.

Expression of posterior Hox genes and opisthosomal appendage development in a mygalomorph spider.

Spiders represent an evolutionary successful group of chelicerate arthropods. The body of spiders is subdivided into two regions (tagmata). The anterior tagma, the prosoma, bears the head appendages and four pairs of walking legs. The segments of the posterior tagma, the opisthosoma, either lost their appendages during the course of evolution or their appendages were substantially modified to fulfill new tasks such as reproduction, gas exchange, and silk production. Previous work has shown that the homeotic Hox genes are involved in shaping the posterior appendages of spiders. In this paper, we investigate the expression of the posterior Hox genes in a tarantula that possesses some key differences of posterior appendages compared to true spiders, such as the lack of the anterior pair of spinnerets and a second set of book lungs instead of trachea. Based on the observed differences in posterior Hox gene expression in true spiders and tarantulas, we argue that subtle changes in the Hox gene expression of the Hox genes abdA and AbdB are possibly responsible for at least some of the morphological differences seen in true spiders versus tarantulas.

Lack of evidence for conserved parasegmental grooves in arthropods.

In the arthropod model species Drosophila melanogaster, a dipteran fly, segmentation of the anterior-posterior body axis is under control of a hierarchic gene cascade. Segmental boundaries that form morphological grooves are established posteriorly within the segmental expression domain of the segment-polarity gene (SPG) engrailed (en). More important for the development of the fly, however, are the parasegmental boundaries that are established at the interface of en expressing cells and anteriorly adjacent wingless (wg) expressing cells. In Drosophila, both segmental and transient parasegmental grooves form. The latter are positioned anterior to the expression of en. Although the function of the SPGs in establishing and maintaining segmental and parasegmental boundaries is highly conserved among arthropods, parasegmental grooves have only been reported for Drosophila, and a spider (Cupiennius salei). Here, we present new data on en expression, and re-evaluate published data, from four distantly related spiders, including Cupiennius, and a distantly related chelicerate, the harvestman Phalangium opilio. Gene expression analysis of en genes in these animals does not corroborate the presence of parasegmental grooves. Consequently, our data question the general presence of parasegmental grooves in arthropods.

Phylogenetic analysis of forkhead transcription factors in the Panarthropoda.

Fox genes encode transcription factors that contain a DNA binding domain, the forkhead domain, and are known from diverse animal species. The exact homology of the Fox genes of different species is debated and this makes inferences about the evolution of the Fox genes, and their duplications and losses difficult. We have performed phylogenetic analyses of the Fox gene complements of 32 panarthropod species. Our results confirm an ancestral complement of FoxA, FoxB, FoxC, FoxD, FoxF, FoxG, FoxJ1, FoxJ2/3, FoxK, FoxL1, FoxL2, FoxN1/4, FoxN2/3, FoxO, FoxP, and FoxQ2 in the Arthropoda, and additionally FoxH and FoxQ1 in the Panarthropoda (including tardigrades and onychophorans). We identify a novel Fox gene sub-family, that we designate as FoxT that includes two genes in Drosophila melanogaster, Circadianly Regulated Gene (Crg-1) and forkhead domain 3F (fd3F). In a very recent paper, the same new Fox gene sub-family was identified in insects (Lin et al. 2021). Our analysis confirms the presence of FoxT and shows that its members are present throughout Panarthropoda. We show that the hitherto unclassified gene CG32006 from the fly Drosophila melanogaster belongs to FoxJ1. We also detect gene losses: FoxE and FoxM were lost already in the panarthropod ancestor, whereas the loss of FoxH occurred in the arthropod ancestor. Finally, we find an ortholog of FoxQ1 in the bark scorpion Centruroides sculpturatus, confirmed not only by phylogenetic analysis, but also by forming an evolutionarily conserved gene cluster with FoxF, FoxC, and FoxL1. This suggests that FoxQ1 belongs to the ancestral Fox gene complement in panarthropods and also in chelicerates, but has been lost at the base of the mandibulate arthropods.

The embryonic expression pattern of a second, hitherto unrecognized, paralog of the pair-rule gene sloppy-paired in the beetle Tribolium castaneum.

In the fly Drosophila melanogaster, a hierarchic segmentation gene cascade patterns the anterior-posterior body axis of the developing embryo. Within this cascade, the pair-rule genes (PRGs) transform the more uniform patterning of the higher-level genes into a metameric pattern that first represents double-segmental units, and then, in a second step, represents a true segmental pattern. Within the PRG network, primary PRGs regulate secondary PRGs that are directly involved in the regulation of the next lower level, the segment-polarity genes (SPGs). While the complement of primary PRGs is different in Drosophila and the beetle Tribolium, another arthropod model organism, both paired (prd) and sloppy-paired (slp), acts as secondary PRGs. In earlier studies, the interaction of PRGs and the role of the single slp ortholog in Tribolium have been investigated in some detail revealing conserved and diverged aspects of PRG function. In this study, I present the identification and the analysis of embryonic expression patterns of a second slp gene (called slp2) in Tribolium. While the previously identified gene, slp, is expressed in a typical PRG pattern, expression of slp2 is more similar to that of the downstream-acting SPGs, and shows expression similarities to slp2 in Drosophila. The previously reported differences between the function of slp in Drosophila and Tribolium may partially account for the function of the newly identified second slp paralog in Tribolium, and it may therefore be advised to conduct further studies on PRG function in the beetle.

The forkhead box containing transcription factor FoxB is a potential component of dorsal-ventral body axis formation in the spider Parasteatoda tepidariorum.

In the spider, determination of the dorsal-ventral body (DV) axis depends on the interplay of the dorsal morphogen encoding gene decapentaplegic (Dpp) and its antagonist, short gastrulation (sog), a gene that is involved in the correct establishment of ventral tissues. Recent work demonstrated that the forkhead domain encoding gene FoxB is involved in dorsal-ventral axis formation in spider limbs. Here, Dpp likely acts as a dorsal morphogen, and FoxB is likely in control of ventral tissues as RNAi-mediated knockdown of FoxB causes dorsalization of the limbs. In this study, we present phenotypes of FoxB knockdown that demonstrate a function in the establishment of the DV body axis. Knockdown of FoxB function leads to embryos with partially duplicated median germ bands (Duplicitas media) that are possibly the result of ectopic activation of Dpp signalling. Another class of phenotypes is characterized by unnaturally slim (dorsal-ventrally compressed) germ bands in which ventral tissue is either not formed, or is specified incorrectly, likely a result of Dpp over-activity. These results suggest that FoxB functions as an antagonist of Dpp signalling during body axis patterning, similarly as it is the case in limb development. FoxB thus represents a general player in the establishment of dorsal-ventral structures during spider ontogeny.

Expression of the zinc finger transcription factor Sp6-9 in the velvet worm Euperipatoides kanangrensis suggests a conserved role in appendage development in Panarthropoda.

The Sp-family genes encode important transcription factors in animal development. Here we investigate the embryonic expression patterns of the complete set of Sp-genes in the velvet worm Euperipatoides kanangrensis (Onychophora), with a special focus on the Sp6-9 ortholog. In arthropods, Sp6-9, the ortholog of the Drosophila melanogaster D-Sp1 gene plays a conserved role in appendage development. Our data show that the expression of Sp6-9 during the development of the velvet worm is conserved, suggesting that the key function of the Sp6-9 gene dates back to at least the last common ancestor of arthropods and onychophorans and thus likely the last common ancestor of Panarthropoda.

Embryonic expression of priapulid Wnt genes.

Posterior elongation of the developing embryo is a common feature of animal development. One group of genes that is involved in posterior elongation is represented by the Wnt genes, secreted glycoprotein ligands that signal to specific receptors on neighbouring cells and thereby establish cell-to-cell communication. In segmented animals such as annelids and arthropods, Wnt signalling is also likely involved in segment border formation and regionalisation of the segments. Priapulids represent unsegmented worms that are distantly related to arthropods. Despite their interesting phylogenetic position and their importance for the understanding of ecdysozoan evolution, priapulids still represent a highly underinvestigated group of animals. Here, we study the embryonic expression patterns of the complete sets of Wnt genes in the priapulids Priapulus caudatus and Halicryptus spinulosus. We find that both priapulids possess a complete set of 12 Wnt genes. At least in Priapulus, most of these genes are expressed in and around the posterior-located blastopore and thus likely play a role in posterior elongation. Together with previous work on the expression of other genetic factors such as caudal and even-skipped, this suggests that posterior elongation in priapulids is under control of the same (or very similar) conserved gene regulatory network as in arthropods.

Phylogenetic analysis and embryonic expression of panarthropod Dmrt genes.

BACKGROUND: One set of the developmentally important Doublesex and Male-abnormal-3 Related Transcription factors (Dmrt) is subject of intense research, because of their role in sex-determination and sexual differentiation. This likely non-monophyletic group of Dmrt genes is represented by the Drosophila melanogaster gene Doublesex (Dsx), the Caenorhabditis elegans Male-abnormal-3 (Mab-3) gene, and vertebrate Dmrt1 genes. However, other members of the Dmrt family are much less well studied, and in arthropods, including the model organism Drosophila melanogaster, data on these genes are virtually absent with respect to their embryonic expression and function. RESULTS: Here we investigate the complete set of Dmrt genes in members of all main groups of Arthropoda and a member of Onychophora, extending our data to Panarthropoda as a whole. We confirm the presence of at least four families of Dmrt genes (including Dsx-like genes) in Panarthropoda and study their expression profiles during embryogenesis. Our work shows that the expression patterns of Dmrt11E, Dmrt93B, and Dmrt99B orthologs are highly conserved among panarthropods. Embryonic expression of Dsx-like genes, however, is more derived, likely as a result of neo-functionalization after duplication. CONCLUSIONS: Our data suggest deep homology of most of the panarthropod Dmrt genes with respect to their function that likely dates back to their last common ancestor. The function of Dsx and Dsx-like genes which are critical for sexual differentiation in animals, however, appears to be much less conserved.

Investigation of endoderm marker-genes during gastrulation and gut-development in the velvet worm Euperipatoides kanangrensis.

The ancestral state of animal gastrulation and its bearing for our understanding of bilaterian evolution still is one of the most controversially discussed topics in the field of evolutionary and developmental biology. One hypothesis, the so-called amphistomy scenario, suggests the presence of a slit-like blastopore in the last common ancestor of Bilateria. Onychophoran ontogeny at least superficially appears to support this scenario since a ventral groove clearly forms during gastrulation. The origin and nature of this groove, however, is another matter of ongoing controversy; i.e. the question of whether this structure actually represents the blastopore, or at least part of it. Recent research using genetic markers argued against the furrow representing a blastoporal structure. Here we investigate the origin of endoderm, which usually originates from the blastopore. We find conserved expression patterns of the endoderm- and gut-marker genes GATA456, GATA123, Hnf4 and fkh during gut development, and discuss the formation of the onychophoran gut in comparison with that in a range of arthropods. Despite expression of endodermal markers in and around the furrow we do not find convincing evidence that the furrow may be part of the blastopore, and thus we suggest that onychophoran development does not yield support for the amphistomy scenario.

Embryonic expression patterns and phylogenetic analysis of panarthropod sox genes: insight into nervous system development, segmentation and gonadogenesis.

BACKGROUND: Sox (Sry-related high-mobility-group box) genes represent important factors in animal development. Relatively little, however, is known about the embryonic expression patterns and thus possible function(s) of Sox genes during ontogenesis in panarthropods (Arthropoda+Tardigrada+Onychophora). To date, studies have been restricted exclusively to higher insects, including the model system Drosophila melanogaster, with no comprehensive data available for any other arthropod group, or any tardigrade or onychophoran. RESULTS: This study provides a phylogenetic analysis of panarthropod Sox genes and presents the first comprehensive analysis of embryonic expression patterns in the flour beetle Tribolium castaneum (Hexapoda), the pill millipede Glomeris marginata (Myriapoda), and the velvet worm, Euperipatoides kanangrensis (Onychophora). 24 Sox genes were identified and investigated: 7 in Euperipatoides, 8 in Glomeris, and 9 in Tribolium. Each species possesses at least one ortholog of each of the five expected Sox gene families, B, C, D, E, and F, many of which are differentially expressed during ontogenesis. CONCLUSION: Sox gene expression (and potentially function) is highly conserved in arthropods and their closest relatives, the onychophorans. Sox B, C and D class genes appear to be crucial for nervous system development, while the Sox B genes Dichaete (D) and Sox21b likely play an additional conserved role in panarthropod segmentation. The Sox B gene Sox21a likely has a conserved function in foregut and Malpighian tubule development, at least in Hexapoda. The data further suggest that Sox D and E genes are involved in mesoderm differentiation, and that Sox E genes are involved in gonadal development. The new data expand our knowledge about the expression and implied function of Sox genes to Mandibulata (Myriapoda+Pancrustacea) and Panarthropoda (Arthropoda+Onychophora).

Embryonic expression of a Long Toll (Loto) gene in the onychophorans Euperipatoides kanangrensis and Cephalofovea clandestina.

Recent research has shown that Toll genes, and in particular a newly defined class of Toll genes, the so-called Long Toll Genes (Loto genes), are crucial factors in embryogenesis. In arthropods, they are involved in axis formation via a process called convergent extension (CE). A hallmark of Loto genes is their relatively (compared to other Toll genes) high number of leucine-rich repeat elements (LRRs) coupled with the fact that they are expressed in transverse stripes in all segments, or a subset of segments, patterns that are reminiscent of classical segmentation genes such as the pair-rule genes. Onychophorans represent a close outgroup to the arthropods; however, their embryonic development differs substantially. It is unclear if convergent extension contributes to onychophoran germ band formation and, if so, whether Loto genes are involved in governing this process. This study identifies a single onychophoran Toll gene from a sequenced embryonic transcriptome in two onychophoran species. The identified gene shows sequence and expression pattern characteristics of Loto genes. However, its expression pattern also comprises some general differences to arthropod Loto genes that are involved in CE.

The house spider genome reveals an ancient whole-genome duplication during arachnid evolution.

BACKGROUND: The duplication of genes can occur through various mechanisms and is thought to make a major contribution to the evolutionary diversification of organisms. There is increasing evidence for a large-scale duplication of genes in some chelicerate lineages including two rounds of whole genome duplication (WGD) in horseshoe crabs. To investigate this further, we sequenced and analyzed the genome of the common house spider Parasteatoda tepidariorum. RESULTS: We found pervasive duplication of both coding and non-coding genes in this spider, including two clusters of Hox genes. Analysis of synteny conservation across the P. tepidariorum genome suggests that there has been an ancient WGD in spiders. Comparison with the genomes of other chelicerates, including that of the newly sequenced bark scorpion Centruroides sculpturatus, suggests that this event occurred in the common ancestor of spiders and scorpions, and is probably independent of the WGDs in horseshoe crabs. Furthermore, characterization of the sequence and expression of the Hox paralogs in P. tepidariorum suggests that many have been subject to neo-functionalization and/or sub-functionalization since their duplication. CONCLUSIONS: Our results reveal that spiders and scorpions are likely the descendants of a polyploid ancestor that lived more than 450 MYA. Given the extensive morphological diversity and ecological adaptations found among these animals, rivaling those of vertebrates, our study of the ancient WGD event in Arachnopulmonata provides a new comparative platform to explore common and divergent evolutionary outcomes of polyploidization events across eukaryotes.

Gene expression reveals evidence for EGFR-dependent proximal-distal limb patterning in a myriapod.

Evolution of segmented limbs is one of the key innovations of Arthropoda, allowing development of functionally specific specialized head and trunk appendages, a major factor behind their unmatched evolutionary success. Proximodistal limb patterning is controlled by two regulatory networks in the vinegar fly Drosophila melanogaster, and other insects. The first is represented by the function of the morphogens Wingless (Wg) and Decapentaplegic (Dpp); the second by the EGFR-signaling cascade. While the role of Wg and Dpp has been studied in a wide range of arthropods representing all main branches, that is, Pancrustacea (= Hexapoda + Crustacea), Myriapoda and Chelicerata, investigation of the potential role of EGFR-signaling is restricted to insects (Hexapoda). Gene expression analysis of Egfr, its potential ligands, and putative downstream factors in the pill millipede Glomeris marginata (Myriapoda: Diplopoda), reveals that-in at least mandibulate arthropods-EGFR-signaling is likely a conserved regulatory mechanism in proximodistal limb patterning.

Gene expression analysis reveals that Delta/Notch signalling is not involved in onychophoran segmentation.

Delta/Notch (Dl/N) signalling is involved in the gene regulatory network underlying the segmentation process in vertebrates and possibly also in annelids and arthropods, leading to the hypothesis that segmentation may have evolved in the last common ancestor of bilaterian animals. Because of seemingly contradicting results within the well-studied arthropods, however, the role and origin of Dl/N signalling in segmentation generally is still unclear. In this study, we investigate core components of Dl/N signalling by means of gene expression analysis in the onychophoran Euperipatoides kanangrensis, a close relative to the arthropods. We find that neither Delta or Notch nor any other investigated components of its signalling pathway are likely to be involved in segment addition in onychophorans. We instead suggest that Dl/N signalling may be involved in posterior elongation, another conserved function of these genes. We suggest further that the posterior elongation network, rather than classic Dl/N signalling, may be in the control of the highly conserved segment polarity gene network and the lower-level pair-rule gene network in onychophorans. Consequently, we believe that the pair-rule gene network and its interaction with Dl/N signalling may have evolved within the arthropod lineage and that Dl/N signalling has thus likely been recruited independently for segment addition in different phyla.

Aspects of dorso-ventral and proximo-distal limb patterning in onychophorans.

Onychophorans (velvet worms) are closely related to the arthropods, but their limb morphology represents a stage before arthropodization (i.e., the segmentation of the limbs). We investigated the expression of onychophoran homologs of genes that are involved in dorso-ventral (DV) and proximo-distal (PD) limb patterning in arthropods. We find that the two onychophoran optomotor-blind (omb) genes, omb-1 and omb-2, are both expressed in conserved patterns in the dorsal ectoderm of the limbs, including the onychophoran antennae (the frontal appendages). Surprisingly, the expression of decapentaplegic (dpp), which acts upstream of omb in Drosophila, is partially reversed in onychophoran limbs compared to its expression in arthropods. A conserved feature of dpp expression in arthropods and onychophorans, however, is the prominent expression of dpp in the tips of developing limbs, which, therefore, may represent the ancestral pattern. The expression patterns of wingless (wg) and H15 are very diverged in onychophorans. The wg gene is only expressed in the limb tips and the single H15 gene is expressed in a few dorsal limb cells, but not on the ventral side. The expression of wg and dpp at the limb tips is one of the three possible alternatives predicted by the topology model of arthropod limb patterning and is, thus, compatible with a conserved function of wg and dpp in the patterning of the PD axis. On the other hand, DV limb gene expression is less conserved, and the specification of ventral fate appears to involve neither wg nor H15 expression.