Comparative analysis of leg and antenna development in wild-type and homeotic<Emphasis Type="Italic"> Drosophila melanogaster</Emphasis>

The insect leg and antenna are thought to be homologous structures, evolved from a common ancestral appendage. The homeotic transformations of antenna to leg in Drosophila produced by mutation of the Hox gene Antennapedia are position-specific, such that every particular antenna structure is transformed into a specific leg counterpart. This has been taken to suggest that the developmental programmes of these two appendages are still similar. In particular, the mechanisms for the specification of a cell's position within the appendage would be identical, only their interpretation would be different and subject to homeotic gene control. Here we explore the degree of conservation between the developmental programmes of leg and antenna in Drosophila and other dipterans, in wild-type and homeotic conditions. Most of the appendage pattern-forming genes are active in both appendages, and their expression domains are partially conserved. However, the regulatory relationships and interactions between these genes are different, and in fact cells change their expression while undergoing homeotic transformation. Thus, the positional information, and the mechanisms which generate it, are not strictly conserved between leg and antenna; and homeotic genes alter the establishment of positional clues, not only their interpretation. The partial conservation of pattern-forming genes in both appendages ensures a predictable re-specification of positional clues, producing the observed positional specificity of homeotic transformations.     

The Hox gene Abdominal-B antagonizes appendage development in the genital disc of Drosophila

In Drosophila, the Hox gene Abdominal-B is required to specify the posterior abdomen and the genitalia. Homologues of Abdominal-B in other species are also needed to determine the posterior part of the body. We have studied the function of Abdominal-B in the formation of Drosophila genitalia, and show here that absence of Abdominal-B in the genital disc of Drosophila transforms male and female genitalia into leg or, less frequently, into antenna. These transformations are accompanied by the ectopic expression of genes such as Distal-less or dachshund, which are normally required in these appendages. The extent of wild-type and ectopic Distal-less expression depends on the antagonistic activities of the Abdominal-B gene, as a repressor, and of the decapentaplegic and wingless genes as activators. Absence of Abdominal-B also changes the expression of Homothorax, a Hox gene co-factor. Our results suggest that Abdominal-B forms genitalia by modifying an underlying positional information and repressing appendage development. We propose that the genital primordia should be subdivided into two regions, one of them competent to be transformed into an appendage in the absence of Abdominal-B.     

The origin and evolution of appendages

Current awareness of gene expression patterns and developmental mechanisms involved in the outgrowth and patterning of animal appendages contributes to our understanding of the origin and evolution of these body parts. Nevertheless, this vision needs to be complemented by a new adequate comparative framework, in the context of a factorial notion of homology. It may even be profitable to categorize as appendages also gut diverticula, body ingrowths and 'virtual appendages' such as the eye spots on butterfly wings. Another unwarranted framework is the Cartesian co-ordinate system onto which the appendages are currently described and where it is supposed that one patterning system exists for each separate Cartesian axis. It may be justified, instead, to look for correspondences between the appendages and the main body axis of the same animal, as the latter might be the source of the growth and patterning mechanisms which gave rise to the former. This hypothesis of axis paramorphisms is contrasted with the current hypothesis of gene co-option. Recapitulationism is a common fault in current Evo-Devo perspectives concerning the origin of the appendages, in that the evolutionary origin of appendages is often expected to be the same as one of the key mechanisms involved in the ontogenetic inception of appendage formation. This unwarranted perspective is also evident in the current debate on the nature of the default arthropod appendage. Most likely, a default arthropod appendage never did exist, as the first appendages probably developed along the trunk of an animal already patterned extensively along the antero-posterior body axis.     

昆虫腹肢发育相关基因的研究进展

昆虫躯干外着生有一系列附属器官,主要包括背侧附器和腹侧附器,其中腹肢的多样性表现尤为突出。腹肢的发育过程受到多种调控因子的作用。本文就腹肢发育相关基因的表达、功能及调控因子间的相互作用等方面进行简要的综述。一方面,腹肢作为整体受Hox基因和成形素基因（Dpp/Wg）的调控,Hox基因直接决定腹肢的有无,Dpp/Wg通过其表达产物形成浓度梯度调控整个腹肢的发育,两者在腹肢整体发育中的作用不可取代。另一方面,腹肢基部、中部及远端部位分别受到各自特异的调控因子的作用。其中hth,tsh及al等均主要调节腹肢基部的发育,dac通过与Dll和Dpp/Wg相互作用从而调节腹肢中部的发育,bab,Dll及Lim1等对腹肢远端发育发挥重要作用。关节的形成对腹肢分节的形成至关重要,Notch信号通路相关因子如配体基因Dl和Ser,修饰物基因fng及下游靶基因odd,sob,drm和bowl等调节该过程。因此,研究昆虫腹肢发育相关基因,对于深入揭示腹肢的发育及其在进化过程中多样性形成的分子机制具有至关重要的作用。     

Hox and wings

In many bilaterian phyla, appendages are morphological traits that characterise the identity of the various body parts. In pterygote insects, wings are dorsal appendages on the thorax. The famous "bithorax" fly created by Ed Lewis is the emblematic example of the role of Hox genes.1 Now, Tomoyasu et al.,2 using classical genetics, transgenesis and RNAi, have examined the function of thoracic Hox genes in the beetle Tribolium castaneum. Beetles have rigid elytra in place of the first pair of wings. Instead of the expected transformation of the elytron into a wing, loss of Hox genes' function leads to the homeotic transformation of the second pair of dorsal appendages, the wings, into elytra. This has important consequences for the way that we see the role of Hox genes in development and evolution.     

<Emphasis Type="Italic">Proboscipedia</Emphasis> represses distal signaling in the embryonic gnathal limb fields of<Emphasis Type="Italic"> Tribolium castaneum</Emphasis>

Orthologs of the Hox genes Sex combs reduced ( Scr) and proboscipedia ( pd) are active in the developing labial appendages of all insect species tested. The remarkable variation among insect gnathal structures, particularly in the distal podomeres, suggests two Hox genes may enhance the adaptive potential of gnathal appendage morphology. Functional studies in the fruitfly Drosophila melanogaster, the flour beetle Tribolium castaneum and the milkweed bug Oncopeltus fasciatus show that cooperation between Scr and pb has been generally conserved, but specific mechanisms have been altered during evolution. Cross-regulation of pb by Scr is evident in Drosophila and Tribolium, the more closely related of the three species, but not in Oncopeltus. In all three species, pb function is restricted to the distal podomeres, but details are only known for Drosophila and Oncopeltus, two species exhibiting specialized stylate-haustellate mouthparts. Drosophila pb is required for distal Scr expression, and to repress the appendage patterning genes dachshund and Distal-less ( Dll). Oncopeltus pb has the novel capacity to specify leg fates. Little is known about distal functions of Tribolium pb. Hypomorphic mutations of the Tribolium pb ortholog maxillopedia can be arranged in a graded phenotypic series of palp to leg transformations along both the proximodistal and dorsoventral axes. Mid-embryonic expression profiles of Tribolium pb, Scr, wingless ( wg) and Dll genes were examined in maxillopedia hypomorphic and null mutant backgrounds. Levels of pb and Scr are significantly reduced in the distal appendage field. Tribolium pb therefore positively regulates distal Scr expression, a role that it has in common with Drosophila pb. Tribolium wg is normally down-regulated in the distal domain of the embryonic gnathal appendage buds. It becomes activated distally in maxillopedia hypomorphs. Repression of wg by pb has not been reported in the labial imaginal discs of Drosophila. Alterations of Tribolium Scr and wg expression occur in Dll-expressing cells, however, unlike in Drosophila labial imaginal discs, Dll expression appears unaffected in pb hypomorphic backgrounds. We conclude that the Hox genes Sex combs reduced and proboscipedia control an appendage organizer and cell autonomous fate determination during embryonic labial palp development in Tribolium.     

Trilobite tagmosis and body patterning from morphological and developmental perspectives

The Trilobita were characterized by a cephalic region in whichthe biomineralized exoskeleton showed relatively high morphologicaldifferentiation among a taxonomically stable set of well definedsegments, and an ontogenetically and taxonomically dynamic trunkregion in which both exoskeletal segments and ventral appendageswere similar in overall form. Ventral appendages were homonomousbiramous limbs throughout both the cephalon and trunk, exceptfor the most anterior appendage pair that was antenniform, preoral,and uniramous, and a posteriormost pair of antenniform cerci,known only in one species. In some clades trunk exoskeletalsegments were divided into two batches. In some, but not all,of these clades the boundary between batches coincided withthe boundary between the thorax and the adult pygidium. Therepeated differentiation of the trunk into two batches of segmentsfrom the homonomous trunk condition indicates an evolutionarytrend in aspects of body patterning regulation that was achievedindependently in several trilobite clades. The phylogeneticplacement of trilobites and congruence of broad patterns oftagmosis with those seen among extant arthropods suggest thatthe expression domains of trilobite cephalic Hox genes may haveoverlapped in a manner similar to that seen among extant arachnates.This, coupled with the fact that trilobites likely possessedten Hox genes, presents one alternative to a recent model inwhich Hox gene distribution in trilobites was equated to eightputative divisions of the trilobite body plan.     

Parasegmental appendage allocation in annelids and arthropods and the homology of parapodia and arthropodia

The new animal phylogeny disrupts the traditional taxon Articulata (uniting arthropods and annelids) and thus calls into question the homology of the body segments and appendages in the two groups. Recent work in the annelid Platynereis dumerilii has shown that although the set of genes involved in body segmentation is similar in the two groups, the body units of annelids correspond to arthropod parasegments not segments. This challenges traditional ideas about the homology of "segmental" organs in annelids and arthropods, including their appendages. Here I use the expression of engrailed, wingless and Distal-less in the arthropod Artemia franciscana to identify the parasegment boundary and the appendage primordia. I show that the early body organization including the appendage primordia is parasegmental and thus identical to the annelid organization and by deriving the different adult appendages from a common ground plan I suggest that annelid and arthropod appendages are homologous structures despite their different positions in the adult animals. This also has implications for the new animal phylogeny, because it suggests that Urprotostomia was not only parasegmented but also had parasegmental appendages similar to extant annelids, and that limb-less forms in the Protostomia are derived from limb-bearing forms.     

Limbs and tail as evolutionarily diverging duplicates of the main body axis

SUMMARY Contrasting hypotheses have been proposed to explain the pervasive parallels in the patterning of arthropod and vertebrate appendages. These hypotheses either call for a common ancestor already provided with patterned appendages or body outgrowths, or for the recruitment in limb patterning of single genes or genetic cassettes originally used for purposes other than axis patterning. I suggest instead that body appendages such as arthropod and vertebrate limbs and chordate tails are evolutionarily divergent duplicates (paramorphs) of the main body axis, that is, its duplicates, albeit devoid of endodermal component. Thus, vertebrate limbs and arthropod limbs are not historical homologs, but homoplastic features only transitively related to real historical homologs. Thus, the main body axis and the axis of the appendages have distinct but not independent evolutionary histories and may be involved in processes of homeotic co-option producing effects of morphological assimilation. For instance, chordate segmentation may have originated in the posterior appendage (tail) and subsequently extended to the trunk.     

A conserved genetic mechanism specifies deutocerebral appendage identity in insects and arachnids

The segmental architecture of the arthropod head is one of the most controversial topics in the evolutionary developmental biology of arthropods. The deutocerebral (second) segment of the head is putatively homologous across Arthropoda, as inferred from the segmental distribution of the tripartite brain and the absence of Hox gene expression of this anterior-most, appendage-bearing segment. While this homology statement implies a putative common mechanism for differentiation of deutocerebral appendages across arthropods, experimental data for deutocerebral appendage fate specification are limited to winged insects. Mandibulates (hexapods, crustaceans and myriapods) bear a characteristic pair of antennae on the deutocerebral segment, whereas chelicerates (e.g. spiders, scorpions, harvestmen) bear the eponymous chelicerae. In such hexapods as the fruit fly, Drosophila melanogaster, and the cricket, Gryllus bimaculatus, cephalic appendages are differentiated from the thoracic appendages (legs) by the activity of the appendage patterning gene homothorax (hth). Here we show that embryonic RNA interference against hth in the harvestman Phalangium opilio results in homeonotic chelicera-to-leg transformations, and also in some cases pedipalp-to-leg transformations. In more strongly affected embryos, adjacent appendages undergo fusion and/or truncation, and legs display proximal defects, suggesting conservation of additional functions of hth in patterning the antero-posterior and proximo-distal appendage axes. Expression signal of anterior Hox genes labial, proboscipedia and Deformed is diminished, but not absent, in hth RNAi embryos, consistent with results previously obtained with the insect G. bimaculatus. Our results substantiate a deep homology across arthropods of the mechanism whereby cephalic appendages are differentiated from locomotory appendages.     

Development and evolution of the mammalian limb: adaptive diversification of nails, hooves, and claws

SUMMARY Paleontological evidence indicates that the evolutionary diversification of mammals early in the Cenozoic era was characterized by an adaptive radiation of distal limb structures. Likewise, neontological data show that morphological variation in distal limb integumentary appendages (e.g., nails, hooves, and claws) can be observed not only among distantly related mammalian taxa but also among closely related species within the same clade. Comparative analysis of nail, claw, and hoof morphogenesis reveals relatively subtle differences in mesenchymal and epithelial patterning underlying these adult differences in distal limb appendage morphology. Furthermore, studies of regulatory gene expression during vertebrate claw development demonstrate that many of the signaling molecules involved in patterning ectodermal derivatives such as teeth, hair, and feathers are also involved in organizing mammalian distal limb appendages. For example, Bmp4 signaling plays an important role during the recruitment of mesenchymal cells into the condensations forming the terminal phalanges, whereas Msx2 affects the length of nails and claws by suppressing proliferation of germinal epidermal cells. Evolutionary changes in the form of distal integumentary appendages may therefore result from changes in gene expression during formation of mesenchymal condensations ( Bmp4 , posterior Hox genes), induction of the claw fold and germinal matrix ( shh ), and/or proliferation of epidermal cells in the claw matrix ( Msx1 , Msx2 ). The prevalence of convergences and parallelisms in nail and claw structure among mammals underscores the existence of multiple morphogenetic pathways for evolutionary change in distal limb appendages.     

Hox genes and the crustacean body plan

The Crustacea present a variety of body plans not encountered in any other class or phylum of the Metazoa. Here we review our current knowledge on the complement and expression of the Hox genes in Crustacea, addressing questions related to the evolution of body architecture. Specifically, we discuss the molecular mechanisms underlying the homeotic transformation of legs into feeding appendages, which occurred in parallel in several branches of the crustacean evolutionary tree. A second issue that can be approached by the comparative study of Hox genes and their expression in the Crustacea bears on the homology of the abdomen. We discuss whether the so-called "abdominal" tagma of the crustaceans is homologous to the abdomen of insects. In addition, the homology of the abdomen between malacostracan and non-malacostracan crustaceans has also been questioned. We also address the question of the molecular developmental basis of the apparent lack of an abdomen in barnacles. We discuss these issues in relation to the problem of constraint versus adaptation in evolution.     

The making of a feather: Homeoproteins,retinoids and adhesion molecules

We have been using feather development as a model for understanding the molecular basis of pattern formation and to explore the roles of homeoproteins, retinoids and adhesion molecules in this process. Two kinds of homeobox (Hox) protein gradients in the skin have been identified: a ‘microgradient’ within a single feather bud and a ‘macrogradient’ across the feather tract. The asynchronous alignment of different Hox macrogradients establishes a unique repertoire of Hox expression patterns in skin appendages within the integument, designated here as the ‘Hox codes of skin appendages’. It is hypothesized that these Hox codes contribute to the phenotypic determination of skin appendages. High doses of retinoic acid cause a morphological transformation between feather and scale, while low doses of retinoic acid cause an alteration of the axial orientation of skin appendages. We have tested the ability of molecules directly involved in the feather formation process to mediate the action of the Hox codes, and surmise that adhesion molecules are potential candidates. Using specific Fabs to suppress the activity of adhesion molecules, we have found that L-CAM is involved in the formation of the hexagonal pattern, N-CAM is involved in mediating dermal condensations, tenascin is involved in feather bud growth and elongation, and integrin β-1 is essential for epithelial-mesenchymal interactions. More work is in progress to fully understand the molecular pathways regulating the feather formation process.     

Selection on bristle length has the ability to drive the evolution of male abdominal appendages in the sepsid fly Themira biloba

Many exaggerated and novel traits are strongly influenced by sexual selection. Although sexual selection is a powerful evolutionary force, underlying genetic interactions can constrain evolutionary outcomes. The relative strength of selection vs. constraint has been a matter of debate for the evolution of male abdominal appendages in sepsid flies. These abdominal appendages are involved in courtship and mating, but their function has not been directly tested. We performed mate choice experiments to determine whether sexual selection acts on abdominal appendages in the sepsid Themira biloba. We tested whether appendage bristle length influenced successful insemination by surgically trimming the bristles. Females paired with males that had shortened bristles laid only unfertilized eggs, indicating that long bristles are necessary for successful insemination. We also tested whether the evolution of bristle length was constrained by phenotypic correlations with other traits. Analyses of phenotypic covariation indicated that bristle length was highly correlated with other abdominal appendage traits, but was not correlated with abdominal sternite size. Thus, abdominal appendages are not exaggerated traits like many sexual ornaments, but vary independently from body size. At the same time, strong correlations between bristle length and appendage length suggest that selection on bristle length is likely to result in a correlated increase in appendage length. Bristle length is under sexual selection in T. biloba and has the potential to evolve independently from abdomen size.