| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Marine Biological Laboratory, Woods Hole, Massachusetts 02543
1 Humboldt-Universität Berlin, Germany.
The relationships among the four arthropod groups (insects, crustaceans, myriapods, chelicerates) have been a point of controversy for many years. According to the traditional view, the myriapods are most closely related to the insects (common taxon "Tracheata" or Antennata), the crustaceans are the closest relatives of the Antennata (common taxon Mandibulata), and the chelicerates are the sister group of the Mandibulata. However, when the characters that have been used to connect the insects and the myriapods were reviewed, the results suggested that they are either suspected of being convergent (i.e., Malpighian tubules, tracheae) or are not significant in a phylogenetic discussion (loss or absence of characters) (1, 2). Moreover, many characters were found to support a close crustacean-insect relationship, a group that is named "Tetraconata" due to the tetrapartite crystalline cones of the ommatidia (3). In addition, several phylogenetic trees based on molecular data suggested a Crustacea-Insecta unit (4, 5, 6). One of the morphological characters that supports a monophyletic Crustacea-Insecta unit is the mode of development of the central nervous system via neuroblasts. Neuroblasts are specific neuroectodermal cells that repeatedly divide unequally, giving rise to a smaller ganglion mother cell that is pushed dorsally into the interior of the embryo and divides once to give rise to neurons or glial cells ("stem cell mode"). The neuroblasts of crustaceans and insects are homologous. In the myriapods so far investigated, ganglia form in a different way and can vary somewhat: clusters of tightly packed cells invaginate into the interior of the embryo. To determine whether the presence of neuroblasts is an apomorphic character for the Tetraconata or a plesiomorphic character introduced earlier in arthropod evolution (and differentiated in myriapods), an outgroup comparison including chelicerates is necessary. If chelicerate ganglia form via neuroblasts, they are probably an old arthropod character and therefore invalid for establishing a crustacean-insect sister group relationship. If chelicerates have a different mode of ganglion formation, neuroblasts can be interpreted as an apomorphic character for the Tetraconata.
Examinations of chelicerate neurogenesis are relatively rare. Weygoldt (7) found that several arachnids form ganglia via invaginations of multiple cells. A substantial immunocytochemical study of neurogenesis in the spider Cupiennius salei recently provided similar results (8). No neuroblasts occur in either case: ganglia are formed through invagination. However, to decide whether an "invagination mode" is an ancestral pattern for chelicerate neurogenesis, it is necessary to study more ancestral species. The horseshoe crab is one of the oldest living chelicerates, showing many ancestral characters during embryogenesis (9). Therefore I examined early neurogenesis in the horseshoe crab Limulus polyphemus using immunocytochemical techniques.
Horseshoe crab embryos of different stages (following staging of Brown and Clapper, 10) were fixed in paraformaldehyde (4%). All antibody stainings (anti-horseradish-peroxidase, Sigma, 48 µg/ml; anti-phopho-histone H3, Upstate Biotech) were performed with slightly changed standard protocols (11). Primarily, I used TRITC-labeled phalloidin (Sigma, 2 µg/ml, incubation 1.25 h), which labels F-actin filaments of the cytoskeleton and can be used to visualize nervous cells and muscle cells. Staining patterns were investigated with a Pascal confocal laser scanning microscope (Zeiss).
The results in the horseshoe crab show a strong similarity to the data from Cupiennius (8). As of stage 14, I found multiple sites of high phalloidin staining (Fig. 1a); the number of these sites increases as development progresses. Both the development and the increase in number of these sites occur in a kind of wave from anterior to posterior, as was described for Cupiennius (8). Deep to the fluorescently stained dots were clusters of invaginating bottle-shaped cells slightly larger than those in the surrounding tissue (Fig. 1a). The strongly stained areas were formed by bundled cell processes that extend to the ventral surface (Fig. 1b). At about stage 17, these invaginating cells could be seen to be involved in the differentiation of neurons. This was also shown by labeling with horseradish-peroxidase antibody, which binds to neurons in all arthropods. Furthermore, experiments on cell proliferation, using phospho-histone-H3 antibody to label mitotic cells, suggested that most if not all of the divisions take place in the apical layer of the neuroectoderm. No proliferation of already invaginated cells could be detected. The proliferation pattern and invaginating activity differed from the neurogenetic "stem cell mode" of insects or crustaceans and was more similar to ganglion formation in several myriapods.
|
Supported by the Grass Foundation and the Deutsche Forschungsgemeinschaft (DFG).
Literature Cited
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
