Conserved and convergent organization in the optic lobes of insects and isopods, with reference to other crustacean taxa.

The shared organization of three optic lobe neuropils-the lamina, medulla, and lobula-linked by chiasmata has been used to support arguments that insects and malacostracans are sister groups. However, in certain insects, the lobula is accompanied by a tectum-like fourth neuropil, the lobula plate, characterized by wide-field tangential neurons and linked to the medulla by uncrossed axons. The identification of a lobula plate in an isopod crustacean raises the question of whether the lobula plate of insects and isopods evolved convergently or are derived from a common ancestor. This question is here investigated by comparisons of insect and crustacean optic lobes. The basal branchiopod crustacean Triops has only two visual neuropils and no optic chiasma. This finding contrasts with the phyllocarid Nebalia pugettensis, a basal malacostracan whose lamina is linked by a chiasma to a medulla that is linked by a second chiasma to a retinotopic outswelling of the lateral protocerebrum, called the protolobula. In Nebalia, uncrossed axons from the medulla supply a minute fourth optic neuropil. Eumalacostracan crustaceans also possess two deep neuropils, one receiving crossed axons, the other uncrossed axons. However, in primitive insects, there is no separate fourth optic neuropil. Malacostracans and insects also differ in that the insect medulla comprises two nested neuropils separated by a layer of axons, called the Cuccati bundle. Comparisons suggest that neuroarchitectures of the lamina and medulla distal to the Cuccati bundle are equivalent to the eumalacostracan lamina and entire medulla. The occurrence of a second optic chiasma and protolobula are suggested to be synapomorphic for a malacostracan/insect clade.

Neurons and glia in the midline of the higher crustacean Orchestia cavimana are generated via an invariant cell lineage that comprises a median neuroblast and glial progenitors.

Midline cells are a common feature of both insects and crustaceans. Midline cells in the insects Schistocerca americana and Drosophila melanogaster have been shown to give rise to pairs of either neurons or glial cells (midline precursor) as well as to repeatedly generate neurons (median neuroblast) or both neurons and glia (median neuroglioblast). This study addresses midline cell lineages in a higher crustacean, the amphipod Orchestia cavimana. In vivo labeling of single midline cells shows that the resulting cell lineage is invariant and that these cells act as progenitors for sets of three glial precursors and one median neuroblast. The progeny are restricted to parasegmental units. The glial precursors give rise to three pairs of glial cells; two of them enwrap the commissures. The median neuroblast gives rise to about 10 cells that differentiate into 3 classes of neurons. The presence of median neuroblasts is also shown for another higher crustacean, the isopod Porcellio scaber using BrdU labeling. This is the first study to analyze the cell lineage of crustacean neurons generated by early ectodermal precursors. A comparison with those of insects demonstrates both conservation and change during the evolution of arthropods.

Distal-less expression in embryos of Limulus polyphemus (Chelicerata, Xiphosura) and Lepisma saccharina (Insecta, Zygentoma) suggests a role in the development of mechanoreceptors, chemoreceptors, and the CNS.

The homeobox gene Distal-less (Dll) is well known for its participation in the development of arthropod limbs and their derivatives. Dll activity has been described for all groups of arthropods, but also for molluscs, echinoderms and vertebrates. Generally, Dll participates in the establishment of the proximo-distal-axis and differentiation along this axis. During our investigation of the expression pattern in the silverfish Lepisma saccharina and the horseshoe crab Limulus polyphemus, we found several expressions in late stages which cannot be explained with the "normal" limb-specific function. The antenna, cerci and terminal filament of the silverfish show a striped expression; single cells on the labrum, mandibles, maxillary palps and anal valves are also strongly stained by the Dll antibody. In addition to cell groups in the developing ganglia of the CNS, in the coxal endites and several nerve cells in femur and the trochanter of the prosomal limbs, the whole prosomal shield of Limulus polyphemus is surrounded by Dll-positive cell clusters. Furthermore, the lateral processes of the opisthosoma and the edges of the opisthosomal appendages are Dll positive. To get an indication of the cell fate of these regions, we examined hatched larvae and juvenile stages of both species with the SEM. We found a striking correlation of these Dll-positive areas and different sense organs, especially mechanoreceptors. Since many sense organs in arthropods are situated on the limbs, interpretation of the Dll expression in limbs is problematical. This has critical implications for comparative analysis of Dll expression patterns between arthropods and for the claim of homology between limb-like structures. Furthermore, we discuss the possibility of convergent appendage evolution in various bilaterian groups based on the improvement of spatial sensory resolution.

The phylogenetic relationships of "predatory water-fleas" (Cladocera: Onychopoda, Haplopoda) inferred from 12S rDNA.

Within the Cladocera, the water-fleas, four major taxa can be distinguished: Anomopoda, Ctenopoda, Haplopoda, and Onychopoda. Haplopoda and Onychopoda are called "predatory water-fleas." The Haplopoda is monotypic; its only representative, Leptodora kindtii, is common in palearctic and nearctic freshwater bodies. The Onychopoda show a remarkable geographic distribution. Most of the described species are restricted to the Caspian Sea, the Aral Sea, and peripheral areas of the Black Sea, including the Sea of Azov--all remnants of the Eastern Paratethys. The remaining onychopods are either freshwater inhabitants or marine animals, widespread in the world oceans. We present molecular evidence for a sister group relationship between Haplopoda and Onychopoda within the Cladocera. The Onychopoda and its three families are monophyletic. We suggest an independent invasion into the Ponto-Caspian basin at least three times, twice originating in the palearctic freshwater bodies and once starting from the world oceans.

The evolutionary transformation of phyllopodous to stenopodous limbs in the Branchiopoda (Crustacea)--is there a common mechanism for early limb development in arthropods?

Arthropods and in particular crustaceans show a great diversity concerning their limb morphology. This makes the homologization of limbs and their parts and our understanding of evolutionary transformations of these limb types problematical. To address these problems we undertook a comparative study of the limb development of two representatives of branchiopod crustaceans, one with phyllopodous the other with stenopodous trunk limbs. The trunk limb ontogeny of a 'larger branchiopod', Cyclestheria hislopi ('Conchostraca') and the raptorial cladoceran Leptodora kindtii (Haplopoda) has been examined by various methods such as SEM, Hoechst fluorescent stain and expression of the Distal-less gene. The early ontogeny of the trunk limbs in C. hislopi and L. kindtii is similar. In both species the limbs are formed as ventrally placed, elongate, subdivided limb buds. However, in C. hislopi, the portions of the early limb bud end up constituting the endites and the endopod of the phyllopodous filtratory limb in the adult, whereas in L. kindtii, similar limb bud portions end up constituting the actual segments in the segmented, stenopodous, and raptorial trunk limbs of the adults. Hence, the portions of the limbs corresponding to the endites of the phyllopodous trunk limbs in C. hislopi (and other 'larger branchiopods') are homologous to the segments of the stenopodous trunk limbs in L. kindtii. It is most parsimonious to assume that the segmented trunk limbs in L. kindtii have developed from phyllopodous limbs with endites and not vice versa. This study has demonstrated at least one way in which segmented limbs have been derived from phyllopodous, multi-lobate limbs during evolution. Similar pathways can be assumed for the evolution of stenopodous, segmented and uniramous limbs in other crustaceans. Irrespective of the differences in the adult limb morphology, the early patterning of arthropod limbs seems to follow a similar principle.

Cell lineage of the midline cells in the amphipod crustacean Orchestia cavimana (Crustacea, Malacostraca) during formation and separation of the germ band.

Cell lineages of identified midline cells were traced in the amphipod Orchestia cavimana (Crustacea, Malacostraca) by in vivo labelling. Midline cells are a common phenomenon in the germ band of crustaceans and insects. Studies in midline cells of Drosophila showed an origin from separate, paired anlagen and a differentiation into three types of cells. The in vivo labelling of midline cells of Orchestia demonstrates that they originate from the same material as the neural and epidermal ectoderm, divide in a stereotyped cell division pattern and give rise to at least two different types of cells. During the following evolutionarily derived mode of germ band elongation in Orchestia, a morphogenetic process is intercalated that separates germ band halves. On the level of single cells, it can be shown that midline cells are the only ectodermal cells that bridge the large distance between the separated parts. The cells are stretched extensively but do not proliferate. Comparing the midline cells of Orchestia with non-malacostracan crustaceans and insects, the results favour the hypothesis that midline cells are a distinct population of cells homologous in crustaceans and insects.

The pattern of Distal-less expression in the mouthparts of crustaceans, myriapods and insects: new evidence for a gnathobasic mandible and the common origin of Mandibulata.

We examined embryos of representatives of crustaceans, myriapods and insects with respect to DII expression in the mouthparts. In order to examine the relationships between mandibular DII expression and the occurrence of a mandibular palp we compared amphipod, isopod and decapod crustacean species. In species with mandibular palps, DII expression is maintained throughout development and is restricted to the palps. The species lacking a palp as an adult show only transient DII expression in early embryonic stages. Furthermore, we studied mandibular DII expression in the myriapod Glomeris marginata that lacks like all myriapods mandibular palps as an adult. The expression pattern is similar to that in crustaceans lacking a palp as an adult. We examined entognathous and ectognathous insects. No sign of mandibular expression could be detected. It is shown that the distal parts of the mandibular appendage were reduced in several steps and lineages independently up to a total loss. Furthermore, we studied DII expression in the first and second maxillae. Except for Glomeris and the collembolans, the first maxillae of all species show a similar pattern of three lobes expressing DII: the outer expression marks the maxillary palp and the inner two mark the outgrowing endites (galea and lacinia of insects). In the first maxillae of collembolans only two expression areas could be detected. In palpless adult first maxillae of isopod crustaceans a transitory embryonic palp occurs which is also DII positive. In the second maxillae of insects, isopod and amphipod crustaceans only two DII-positive lobes occur. Our data suggest a gnathobasic character of the mandibles of crustaceans, myriapods and insects supporting the monophyly of Mandibulata sensu Snodgrass. The interpretation of DII expression patterns and its limits are critically evaluated.

How far does cell lineage influence cell fate specification in crustacean embryos?

In crustaceans, invariant cell lineages have been shown to occur (i) in early cleavages of several taxa and (ii) in the course of formation and differentiation of the post-naupliar germ bands in malacostracans. Work on early cleavages is still in its infancy. In contrast, the generation and proliferation of mesoteloblasts and ectoteloblasts and the subsequent proliferation and differentiation of bandlet cells have been studied in members of several subgroups of Malacostraca. Similarities and differences have been determined in order to interpret the interdependencies of the steps in the differentiation process. Some of these steps are highly conserved, as in the case of the generation of four pairs of mesoteloblasts, others are prone to phylogenetic change, as in the case of the primary ring of 19 ectoteloblasts which has been altered at least twice in evolution. A stereotyped cleavage pattern in the germ band has been shown to be independent of the origin of the precursor cells. The question whether neuroblasts in crustaceans and insects are homologous or are the result of convergent evolution is still open. However, the homology of early differentiating neurons in crustaceans and insects seems to be well established. In addition, similarities in the expression patterns of the engrailed gene are likely to be homologous and point to a close relationship between these two groups.

Cell lineage and cell fate in crustacean embryos--a comparative approach.

Malacostracan crustaceans undergo a complex and stereotyped cleavage pattern during formation of and segmentation in their post-naupliar germ band. This pattern has been studied in several malacostracan representatives with respect to morphogenesis and expression of the engrailed gene. Although this cell lineage pattern is specific and invariant in each species, comparative analyses reveal subtle differences between different parts of individual germ bands and between germ bands of different species. We conclude that despite the elaborate cleavage pattern, cell fate specification is not closely linked to cell lineage. Furthermore, some aspects of the evolutionary alterations of germ band formation and segmentation in annelids and arthropods are discussed.

Expression of the Engrailed Gene Reveals Nine Putative Segment-Anlagen in the Embryonic Pleon of the Freshwater Crayfish Cherax destructor (Crustacea, Malacostraca, Decapoda).

Segment formation in the embryonic pleon of the freshwater crayfish Cherax destructor was analyzed by using the monoclonal antibody mAb 4D9 against the product of the segment-polarity gene engrailed. As in other body regions, engrailed is expressed in transverse stripes in the posterior portion of segments in the pleon. Nine engrailed stripes are formed in the pleon. The anterior six stripes correspond to the six pleon segments of adult eumalacostracan crustaceans. The uropods are clearly the appendages of the sixth pleon segment. The seventh engrailed stripe marks the anlage of a seventh ganglion. Stripes eight and nine are transient and disappear before morphogenesis begins. The engrailed stripes seven to nine are interpreted as vestiges of ancestral segments. The seventh segment anlage is thus a recapitulation of the seventh pleonic segment, which is retained in recent adult leptostracans and is considered to be part of the malacostracan ground plan. The stripes eight and nine might point still further back into the phylogeny of crustaceans or even mandibulates. The use of rhodamine-labeled phalloidin reveals that the terminal ganglion of adult crayfish is the fusion product of the anlagen of the sixth and seventh pleonic ganglia and an eighth hemiganglion that is devoid of engrailed expression.

Serially homologous engrailed stripes are generated via different cell lineages in the germ band of amphipod crustaceans (Malacostraca, Peracarida).

A monoclonal antibody (mAb 4D9) was used to analyze engrailed expression in amphipod embryos. As in other arthropods, engrailed is expressed in iterated transverse stripes in the germ band. In the anterior region these stripes are generated without a recognizable division pattern, and their appearance and formation show some irregularities. In the posterior region of the germ band, engrailed expression is correlated with a stereotyped cell division pattern resulting in a highly ordered formation and array of stripes. The engrailed positive cells mark the anterior border of genealogical units, which therefore can be compared with parasegments in Drosophila. Expression starts in the mandibular segment and proceeds first anteriorly and subsequently in a posterior direction. Initial stripes are one cell wide. The widening of stripes is caused by both division of engrailed positive cells and recruitment of new cells that did not previously express engrailed. The widening process is related to segment formation as the intersegmental furrows are established behind the engrailed expressing cells, which are restricted to the posterior portion of the forming segments. A comparison of the modes of engrailed expression in different segments suggests that initial engrailed expression is independent of a certain cell lineage or division pattern. The comparison of the development of the early engrailed stripes in different insects and crustaceans reveals some similarities which show that early engrailed expression is not necessarily clonally inherited.

Expression of engrailed can be lost and regained in cells of one clone in crustacean embryos.

In three species of higher crustaceans (Malacostraca) the expression of engrailed has been analysed in relation to the development of the cell division pattern in the germ band. The species differ in the timing of initial en expression. Compared to Cherax destructor and Neomysis integer the onset of en expression in Orchestia cavimana is delayed and appears one cell cycle later. In Cherax and Neomysis cells of the posterior margin of early en stripes lose en expression. This phenomenon does not occur in Orchestia. In a second step the en stripes widen both by division of en positive cells and de novo expression at the posterior margin of the en stripes. The widening phase is similar among all investigated species. In Cherax and Neomysis the cells with de novo en expression are derivatives of cells, which have ceased to express en one cell cycle before. The results in higher crustaceans suggest that neither initiation nor maintenance of en expression is controlled by lineage restrictions and that early en expression is not clonally transmitted. Furthermore, some aspects of boundaries and fields in embryos are discussed.