- Laboratory Culture
- Cleavages, gastrulation, and neurulation
- Cell lineages and fate map up to hatching larvae
- Larval development
- Juvenile development and gametogenesis
Recent advances in our understanding of the molecular and cellular mechanisms that govern development owe a lot to model organisms with short life cycles in which genetic approaches are feasible, such as Drosophila and Caenorhabditis elegans. Oikopleura dioica Fol 1872, an appendicularian (or larvacean), is a candidate model organism for several reasons: It has a life cycle of only 5 days at 20 C (Fenaux 1998a). Inland and multi-generation culture methods have been established. A draft genome sequence is now available. The animal is transparent during development. Morphogenesis and cell lineages during embryogenesis are well described. And the juvenile and adult anatomies have been comprehensively described.
Reviews by Galt & Fenaux (1990) and Alldredge (1976) give a general introduction to appendicularians. The Oikopleura tadpole builds and lives inside a so-called house (oikos in Greek), which is made from acelluar materials secreted from the oikoplastic epithelium (trunk epidermis) (Fig. 1A, B; Fenaux 1986; Flood & Deibel 1998; Thompson et al. 2001). The house is a complex filtration apparatus by which the animal can feed on a highly concentrated suspension of particles filtered from seawater. The tail generates a water current through the house when the tadpole is within the house. The tadpole leaves and discards the house 5 to 10 times a day, forming a new one in just 10 min. Appendicularians are second or third most abundant among the zooplankton community and serve as food for other zooplankton and fishes (Alldredge 1976). Their discarded houses are the main component of marine snow. Thus, appendicularians have ecological importance in marine food web.
Appendicularians are planktonic tunicates (urochordates), and retain a swimming tadpole shape throughout their life (Fig. 1A, B, see also Fig. 3), whereas other tunicates resorb the tail during metamorphosis. Only 3 families and 69 species of appendicularians have been described so far (Fenaux 1993). Together with ascidians and thaliaceans, they are the closest relatives of the vertebrates (Fig. 1C) (Garstang 1928; Delsuc et al. 2006; Bourlat et al. 2006). 18S rDNA sequences suggest that appendicularians are the most basal among these tunicates (Wada 1998; Swalla et al. 2000), although their phylogenetic position among the tunicates is still controversial (Stach & Turbeville 2005). Oikopleura dioica is the sole dioecious (or dioicous) species so far reported among the Tunicata. They represent a simplified body plan of chordates, characterized by a tail with a notochord and dorsal neural tube, an endostyle, and a gill aperture. The structure of the adult body is relatively simple and consists of a small number of cells; for example, there are only 20 notochord cells at hatching and 20 non-fused muscle cells even in the tail of a mature adult (Nishino et al. 2000, 2001; Bassham & Postlethwait 2000; reviewed in Fenaux 1998; Nishino & Satoh 2001).
Fig. 1.The appendicularian Oikopleura dioica and its phylogenetic position. Dorsal (A) and lateral (B) views of the animal in its ghouseh. Carbon particles from India ink were added to seawater and were concentrated by the house. Seawater comes into the house from the bilateral inlet filters and goes out from the ventral outlet pore. The particles in seawater are highly concentrated by the nets of the food-trap filter and introduced into the trunk via the mouth. Digestive organs within the trunk are made visible with the black particles. (C) Phylogenetic tree within the phylum Chordata.
Oikopleura dioica is abundant worldwide in coastal regions. It tolerates extremes of temperature and salinity, and is the only species that has been cultured so far in laboratories. Paffenhofer (1973) first reported successful culture in the laboratory. Now it is routinely cultured in laboratories in Villefranche-sur-Mer (France), Bergen (Norway), and Eugene, Oregon (USA), and in our lab.
The current culture methods were developed in Bergen as described in Chioda et al. (2002). In our lab, the animals are cultured in artificial seawater (RohtoMarine, Rei-Sea Co., Tokyo) in 5-L plastic beakers under constant stirring with an acrylic paddle (15 rpm) at 20 C (Fig. 2A) (Fujii et al. 2008). Ten grams of activated charcoal is added to each beaker in order to keep the seawater in good condition. The life cycle is 5 days. Cultures are diluted five fold on day 1, and then approximately 100 animals are transferred to clean artificial seawater on day 3 by pipetting, and allowed to naturally spawn on day 5 (Fig. 2C). The animals are fed with a cocktail of live algae and cyanobacterium of various sizes (Fig. 2B), namely Isochrysis galbana (approximately 4 Êm, 3000 cells/mL), Chaetoceros calcitrans (3?6 Êm, 3000 cells/mL), Rhinomonas reticulata (12 Êm, 1000 cells/mL), and Synecococcus sp. (1.5 Êm, 4.0 mL of culture / 5 L) on days 1 and 2, and then at double those concentrations from days 3 to 5. The animals grow very fast each day, and grow well under these culture conditions (see Fig. 11). Growing animals can trap only foods that are intermediate in size between the mesh size of the inlet filter and the food-trap filter of the house (Fig. 1A, B). This is why they are fed foods of various sizes. We have cultured the animals for 2 years over a hundred generations without introducing new animals from the ocean.
The life cycle is shown in Fig. 3. At 20 C, they hatch 3 h after fertilization. Tail shift (see below) of the larva takes place at 10 h, then the juvenile becomes fully functional, making its first house and starting to take food. They mature to males or females with full-grown gonads in 5 days, then spawn and die soon after. Developmental time roughly doubles at 13 C (Fig. 3B). Each female spawns 40 to 400 eggs depending on the amount of food available.
Fig. 2.Laboratory culture. (A) The animals are cultured in artificial seawater in 5-L plastic beakers with an acrylic paddle at 20 C. 10 g of activated charcoal is added to each beaker. (B) Culture of algal strains in plastic bags. O. dioica is fed with a mixture of Isochrysis galbana, Chaetoceros calcitrans, Synecococcus sp., and Rhinomonas reticulata. (C) The animalfs life cycle is 5 days (D1?D5). Cultures are diluted on day 1, and then approximately 100 animals are transferred to clean artificial seawater on day 3. They naturally spawn on day 5.
Fig. 3.Life cycle of O. dioica at (A) 20 C and (B) 13 C, showing durations of embryogenesis and of larval and juvenile development.
Oikopleura dioica has a tiny genome of only 72 Mb, the smallest ever found in a chordate (ascidian Ciona, 160 Mb; amphioxus Branchiostoma, 550 Mb; fugu, 400 Mb; human 3000 Mb) (Seo et al. 2001). Shotgun sequences obtained at a high coverage (14?) are estimated to represent 99.7% of all coding nucleotides. The first assembly of the draft genome is available at Genoscope (http://www.genoscope.cns.fr/externe/Francais/Projets/Projet_HG/). In spite of the small genome, the number of genes is estimated as 15 000 (comparable to Ciona, and approximately half of human). This indicates a high gene density within the genome (1 gene every 5 kb). Indeed, intergenic sequences are very short, and 62% of introns are smaller than 50 bp. The positions of introns in each gene diverge from those in other animals (Edverdsen et al. 2004). Comparison of lengths of some genes between O. dioica and human revealed that genes measuring 0.8, 1.6, and 2.2 kb in O. dioica correspond to genes measuring 15, 30, and 35 kb in human, respectively. The small size of the genome and high density of genes will facilitate chemical and insertional mutagenesis in this organism. It is still premature to conclude a link between the small genome and the short life cycle or rapid embryogenesis only from the data so far available.
Several groups have used molecular and comparative approaches for studying the development of this animal by cloning several developmentally regulated genes (Bassham & Postlethwait 2000, 2005; Thompson et al. 2001; Canestro et al. 2005; Canestro & Postlethwait 2007; see also Nishino et al. 2001 for O. longicauda). A conspicuous character of O. dioica is the partial loss of Hox genes (Seo et al. 2004; Edvardsen et al. 2005), specifically the central genes Hox3, 5, 6, 7, and 8, and the Hox cluster is totally fragmented and dispersed within the genome. However, the anterior?posterior order of Hox gene expression seems to keep spatial colinearity. A similar situation is seen in ascidians (Ikuta et al. 2004; Ikuta & Saiga 2005), although ascidian lost different Hox cluster genes (Hox7, 8, 9, and 11). The body plan of chordate ancestors (probably similar to that of cephalochordates; Fig. 1C), may have been simplified in the tunicate lineage, resulting in relaxed pressure to conserve a full set of the Hox gene cluster. Alternatively, embryogenesis with determinate and invariant cell lineages may reduce the requirement for several Hox genes (Seo et al. 2004; Edvardsen et al. 2005). Another unique character of O. dioica is the presence of spliced-leader RNA trans-splicing (Ganot et al. 2004), which is known in a limited and sporadic distribution among eukaryotes, and only in C. elegans and ascidians among metazoans so far (Vandenberghe et al. 2001; Satou et al. 2006). At least 25% of O. dioica mRNA has spliced-leader RNA in the 5? ends, although the biological significance of trans-splicing is still unclear. These unique characters that are commonly found in appendicularians and ascidians support their close relationship.¢Back to CONTENTS
Artificial fertilization can be done by mixing fully matured eggs and sperm dissected from gonads, or naturally spawned eggs and sperm. Sperm can be kept frozen in liquid nitrogen (K. Ouchi and HN, unpublished). The fine structure of sperm has been described in Flood & Afzelius (1978). Each sperm is 30 Êm long and has a typical shape with head and tail. The fine structure of eggs and events at fertilization have been described in Holland et al. (1988). Unfertilized eggs are approximately 80 Êm in diameter, and are arrested at meiotic metaphase I. There are no accessory cells inside or outside the vitelline membrane.
By 30 s after fertilization, the surface of the egg becomes rough. This is the first sign of fertilization. The narrow perivitelline space does not expand after fertilization. In ascidians, movements of the egg cytoplasm, called ooplasmic segregation, take place after fertilization (Conklin 1905; Nishida 1994). These movements transform the symmetry of eggs from radial to bilateral. During the movements, several localized maternal mRNAs that play crucial roles in cell fate specification and embryonic axis determination are brought to the future posterior pole and establish the secondary embryonic axis perpendicular to the primary animal?vegetal axis (Nishida & Sawada 2001; Negishi et al. 2007; reviewed in Nishida 2002, 2005; Prodon et al. 2007). In O. dioica eggs, however, ooplasmic movement is not obvious (Holland et al. 1988), and there is no mitochondrion-rich myoplasm comparable to that in ascidian eggs (HN, unpublished). So how unfertilized eggs of O. dioica with radial symmetry turn into bilaterally symmetrical embryos remains elusive. The embryogenesis of O. dioica is a kind of simplified form of that of ascidians: The fate maps of both organisms are similar, as mentioned below, but O. dioica does not share the ooplasmic movements just after fertilization. Identification of Oikopleura homologs of localized mRNAs in ascidians and examination of their localization will shed light on this issue.
Even though Oikopleura has a small number of embryonic cells and can be cultured readily in the laboratory, only few reports describing its embryogenesis have appeared so far (Delsman 1910; Fenaux 1998a; Nishino & Satoh 2001). Among these, that of Delsman (1910) is remarkably detailed and comparable to that of Conklinfs (1905) monumental paper on ascidian embryogenesis. Delsman investigated the cleavage pattern in amazing detail using embryos at various stages collected directly from the ocean with a plankton net. He managed to identify each blastomere during the cleavage and gastrulation stages.
Eggs and embryos are highly transparent and thus suitable for observation with differential interference contrast (DIC) optics (Fig. 4) (Fujii et al. 2008). The emission of two polar bodies at 15 min is followed by the first cleavage at 35 min at 13 C. Early cleavages take place every 15 min. After the cleavage stage, the first extensive morphogenetic movement, gastrulation, starts at 2 h at the 32-cell stage (Fig. 4G). Every vegetal blastomere is internalized during gastrulation. The tailbud stage is attained at 4 h (Fig. 4K), then the tail elongates and the embryo becomes bent ventrally. Notochord cells aligned in a single row become evident at 4.5 h (Fig. 4L?N). Tailed larvae hatch from the vitelline membrane at 6 h (Fig. 4O) (see also Supplemental Movies of the embryogenesis in Fujii et al. 2008). There is no obvious structure in the head and trunk regions, but the larvae occasionally beat their tail. Fine structures of hatching larvae are described in Stach (2007). The tail shows a remarkable 90 counterclockwise rotation relative to the trunk when viewed from the posterior, with neural tube on the left side (Fig. 4N). The rotation of the tail reflects the direction of the tail beat. In ascidians, lancelets (Amphioxus), and lower vertebrates, the tail beats in a left?right direction, whereas in appendicularians it beats in a dorso?ventral direction.
Fig.4. Embryogenesis of O. dioica. Developmental times and stages are indicated below each photo. (A) Unfertilized egg, 80 Êm in diameter. (B) Fertilized egg after emission of the second polar body. (C) 2-cell stage. (D) 4-cell stage. (E?G) 8-, 16-, and 32-cell embryos, respectively. The name of each blastomere is indicated in (E and F). The embryo in (G) is gastrulating from the left side of the photo. (H?L) Late gastrula to tailbud stages. Photos were taken every 30 min. (M, N) Right and ventral views during tail elongation. In the optical section of the tail, muscle (Mu), notochord (Not), nerve cord (NC), and endodermal strand (ES) are visible. The tail has rotated counterclockwise by 90 when viewed from the posterior direction with the nerve cord on the left side. (O) A tadpole larva is hatching from the vitelline membrane (VM). ant, anterior; pos, posterior; ani, animal pole; veg, vegetal pole. (Modified from Fujii et al. 2008 with permission.)
Cleavages, gastrulation, and neurulation
Fujii et al. (2008) recently reassessed Delsmanfs (1910) descriptions of cleavage pattern and morphogenetic movements in detail using updated techniques such as DIC observation, time-lapse video recording, and 3D reconstruction with confocal microscopy. Movies of full embryogenesis, gastrulation, and neurulation are available as supplementary materials with the paper. We confirmed that the descriptions by Delsman were both precise and accurate, except for one new finding of unequal division of B1 blastomeres, as described below. Cleavage showed an invariant pattern among individuals. The cleavage pattern is somewhat complicated, as the pattern is roughly, but not precisely, bilateral. This makes it difficult to recognize the orientation of embryos, as they lack a straight midline, unlike ascidian embryos. The detection of genes that are expressed in specific subsets of blastomeres has significantly advanced our understanding of early cell-fate-specification processes in ascidian embryos (Nishida 2005). Similarly, the precise identification of blastomeres would be essential for clarifying patterns of gene expression in O. dioica. For example, expression of brachyury and muscle actin is initiated in notochord and muscle precursor blastomeres at the 32- and 64-cell stages, respectively (Bassham & Postlethwait 2000; Nishino et al. 2000, 2001).
The first two cleavage planes are meridional (Figs. 4C, D). The first cleavage generates the left and right blastomeres, and the second divides the anterior and posterior blastomeres, as in ascidian embryos, which are equivalent to the dorsal and ventral blastomeres, respectively, in amphibian embryos. (Refer to Nishida 2005 for differences in conventional definitions of the anterior?posterior and dorsal?ventral axes in tunicate and amphibian embryos.) The anterior-left and posterior-right blastomeres of the 4-cell embryo are invariably formed slightly closer to the vegetal pole, where they are in contact with each other. On the other hand, the anterior-right and posterior-left blastomeres are in contact at the animal pole. This is the first sign of invariant left?right asymmetry in the O. dioica embryo. The third cleavage occurs horizontally, generating four smaller vegetal blastomeres and four larger animal blastomeres (Fig. 4E). As described later, all the vegetal blastomere descendants are internalized during gastrulation.
At the 8-cell stage, each blastomere is named as in the ascidian system (Delsman 1910). Animal cells are indicated by a lower-case letter, and vegetal cells by a capital. Anterior cells are named gah and gAh and posterior cells are named gbh and gBh. The cells of the right half when viewed from the vegetal pole (future dorsal side) with the anterior pole upward are labeled by underlining (Fig. 4E). From the 16-cell stage, Delsman (1910) applied the nomenclature system used in annelid and mollusc embryos. All the descendant cells of a given blastomere in the 8-cell embryo are indicated with the character of the blastomere. After each division, a numeral (1 or 2) is added, 1 for the daughter cell closer to the animal pole. For instance, blastomere gAh divides into A1 and A2. Then the A1 blastomere divides into A11 and A12, and A2 divides into A21 and A22. Details of the cleavage pattern up to the 32-cell stage are reported in Fujii et al. (2008) (see also Fig. 8).
Fig.5. Gastrulation. Ingressing cells during the first phase (A-D) and the second phase (E-G) are indicated by pseudo-yellow color. (A?D) First phase. Sequential images of time-lapse video taken from 105 to 122 min after fertilization at 13 C. Name of each blastomere is indicated. Eight yellow-colored blastomeres at the vegetal pole are ingressing. The embryonic surface becomes covered by six green-colored blastomeres by epiboly. These six green cells are arranged like an asterisk at the completion of ingression of the yellow cells (D). B11 cells, sister cells of B12, generated by unequal division are shown in blue. Anterior is up. (E?G) Second phase, from 122 to 145 min. The B12 cells have divided into B122 and B121 and are now colored yellow. Six yellow-colored cells are internalized sequentially in the order B121, B122, and B11 and covered by the b-line cells. (Reprinted from Fujii et al. 2008 with permission.)
Fig.6. Neurulation. Ingressing neural precursor cells are indicated by pseudo-yellow color. Sequential images of time-lapse video taken from 120 to 130 min after fertilization at 13 C. Anterior is at the bottom left, and the vegetal pole is at the upper right. Four yellow cells of A1 descendants in the anterior region divide once along the anterior?posterior axis and are then internalized. The surface is covered by green a-line cells. (Reprinted from Fujii et al. 2008 with permission.)
Two rounds of unequal cell division occur in the posterior-vegetal B-line blastomeres. The B blastomere pair of the 8-cell embryo divides into the larger B2 and the smaller B1 pair. In the second round, B1 blastomeres at the posterior pole of the 16-cell embryo divide unequally into the larger B12 and the smaller B11 again (see Fig. 8, posterior views) (Fujii et al. 2008, see also Supplemental Movies in Fujii et al. 2008). Thus, the cleavage pattern of the O. dioica embryo shows striking similarity to that of ascidians, in which three rounds of unequal division take place at the posterior pole of the vegetal hemisphere from the 8- to 64-cell stages (Conklin, 1905). Eventually, the smallest blastomeres at the posterior pole of the 64-cell ascidian embryo are fated to become primordial germ cells (Tomioka et al. 2002; Shirae-Kurabayashi et al. 2006). It is uncertain whether the resulting tiny B11 cell (see Fig. 8, brown cell pair) of O. dioica is a primordial germ cell, but comparison with ascidian embryos suggests that this is plausible. The posterior blastomeres in ascidians have a special subcellular structure, the centrosome-attracting body (CAB) (Hibino et al. 1998; Nishikata et al. 1999; Iseto & Nishida 1999; Negishi et al. 2007). During unequal cleavages, the microtubule bundle extending from the posterior-most centrosome of the pair is connected to the CAB in the posterior cortex. Then, as the microtubule bundle shortens, the interphase nucleus shifts posteriorly and approaches the CAB. Consequently, an asymmetrically located mitotic apparatus is formed, and unequal division occurs. The CAB is a multifunctional structure in ascidian embryos (reviewed in Nishida et al. 1999; Nishida 2002; Prodon et al. 2007), and it also serves as an mRNA localization scaffold, where some maternal mRNAs are specifically localized. The CAB also contains an electron-dense matrix similar to germplasm. It is likely that appendicularians also posses the CAB structure, as two rounds of unequal division take place at the posterior pole.
In O. dioica, gastrulation is initiated as early as the 32-cell stage, 1 h 50 min after fertilization at 13 C, much earlier than in ascidians (at the 110-cell stage in Halocynthia roretzi, 9 h after fertilization at 13 C) and vertebrates. Internalization of the vegetal cells proceeds by ingression and/or epiboly, keeping the embryonic outline almost spherical. There is no archenteron. All vegetal cells, namely A- and B-line cells, are internalized. The gastrulation process can be subdivided into three phases. First, eight cells at the vegetal pole ingress (Fig. 5A?D), then descendants of B1 blastomeres in the posterior region follow (Fig. 5E?G) (Fujii et al. 2008, see also Supplemental Movies in Fujii et al. 2008). Finally, b22 descendants in the animal hemisphere (muscle precursors) are internalized later (Stach et al. 2008).
After the first phase of gastrulation, neurulation starts at the 64-cell stage, again earlier than in other chordates. It is difficult to precisely define the neurula stage in O. dioica, as the second phase of gastrulation takes place simultaneously with neurulation. In the anterior region, two rows of four cells that originate from the A1 blastomere pair are internalized (Fig. 6), as previously observed in O. longicauda (Nishino and Satoh 2001). There is no neural fold, and folding of the neural plate is not observed (Fujii et al. 2008, see also Supplemental Movies in Fujii et al. 2008). However, at the tailbud stage, four cells constitute a neural tube in optical transverse sections of the nerve cord in the tail (Fig. 4N), as observed in ascidian tadpole larvae. Therefore, tube formation would take place at later stage after neural precursor cells have completely ingressed. In addition to the A1 descendants, a recent cell lineage analysis showed that descendants of the a222 cell (Fig. 6B, C) join the CNS at a later stage (Stach et al. 2008).
Thus, the initial morphogenetic movements ? gastrulation and neurulation ? take place quite early when the embryo still consists of only a small number of cells. The cells are internalized by ingression, not by folding of an epithelial sheet.¢Back to CONTENTS
Cell lineages and fate map up to hatching larvae
Detailed descriptions of cell lineages are especially important for animal embryos that show invariant and determinative cleavage patterns, as is the case for C. elegans and ascidians (Sulston et al. 1983; Nishida, 1987). Recently, Stach and his colleagues have traced complete cell lineages of O. dioica up to the hatching stage by using time-lapse recording with 4D microscopy (Stach et al. 2008). The cell lineages and cell fate were essentially invariant among three specimens examined. A fertilized egg divides 9 or 10 times on average and generates approximately 550 cells at hatching. The most striking feature of O. dioica embryogenesis is that the fates of most blastomeres are restricted to give rise to a single tissue type by the 32-cell stage (Fig. 7). The fate map shows bilateral symmetry (Fig. 8) with the exception that the right A22 blastomere gives rise to notochord (pink) in addition to endoderm (yellow). Every cell in the animal view develops into larval epidermis (green). In the vegetal view, cells at the vegetal pole form endoderm (yellow). The posterior and lateral marginal cells give rise to tail muscle (red). Notochord (pink) and CNS (light blue) are derived from the anterior marginal blastomeres.
The topography in presumptive tissue territories looks quite similar to that in the ascidian fate map (Fig. 8, bottom left; Nishida, 2005), which in turn shows similarity to the amphibian fate map although each territory differs in relative proportion (Fig. 8, bottom right; Kourakis and Smith, 2005). Therefore, the fate map of O. dioica can be regarded as a typical one shared by basal chordates. In ascidians, fate restriction in most blastomeres is completed by the 110-cell stage, in contrast to the 32-cell stage in O. dioica. However, it is notable that in all these animals fate restriction is almost completed just before gastrulation starts. The Oikopleura fate map at the 32-cell stage represents the simplest one in chordates and probably among all animals so far observed.
Fig7. Cell lineage tree of O. dioica from fertilized egg to hatching larva. Each lineage is color-coded after the developmental fate of the cell is tissue-restricted. An arrow at the bottom marks the left?right asymmetry in the formation of notochord in the A22 lineage. (Modified from Stach et al. 2008 with permission.)
Fig8. Fate map and cell fate restriction during cleavage stages. Embryos are viewed from various directions indicated at the top. Orientations of embryos are indicated at the bottom. Note that the vegetal pole (future dorsal side) is down in the anterior, posterior, right, and left views. Arrowed bars connect sister blastomeres. At the 16-cell stages, blastomeres are colored when the fate of the blastomere is restricted to give rise to a single tissue type. At the 32-cell stage, every tissue-forming territory is colored regardless of fate restriction of blastomeres. Color codes are shown at the bottom. Note that most cells at the 32-cell stage are fated to give rise to a single cell type. The fates of heart (purple) and germ cells (brown) are just speculation based on comparison with ascidian embryos, and are not experimentally confirmed. At the bottom left, the fate map of the animal and vegetal halves of the 110-cell ascidian embryo are shown with the same color codes for comparison (Nishida, 2005). White blastomeres are precursors of mesenchyme and trunk lateral cells, which are not present in O. dioica embryos. At the bottom right, the fate map of the left half of the Xenopus blastula is shown with the same color codes (Kourakis and Smith, 2005). Animal pole is up. HM, head mesoderm. Note the similarity of topography in the presumptive tissue territories between appendicularian, ascidian, and amphibian fate maps, although each territory differs in relative proportion. The fate map of O. dioica was drawn using the cleavage pattern data in Fujii et al. (2008) and lineage data in Stach et al. (2008) with permission.
Larval development is subdivided into five stages, finishing with the young juvenile at stage 6 (Fig. 3B and 9) (Fenaux 1976, 1977, 1998a; Galt & Fenaux 1990). Just after hatching (stage 1), there is no obvious structure in the trunk, but larvae develop into fully functional juveniles (stage 6) in 13 h after hatching at 13 C, or in about half the time at 20 C. During these stages the tail continuously elongates and becomes flattened laterally.
Fig9. Larval development is subdivided into five stages (1?5). After the tail shift, stage 6 is the young juvenile stage. Developmental times after fertilization at 13 C and the corresponding stages are indicated above each photo. Larvae develop into fully functional juveniles (stage 6) in 13 h after hatching at 13 C. (B-E) Left-side views. (F) Right-side view after the tail shift. See details in text.
The general organization of organs that are present in the adult form is already achieved by stage 6 (Fig. 10). The only exception is the gonad, which is a small rudiment in the posterior-ventral position (Fig. 10I). Cell divisions in the epidermis and probably in most of the other somatic tissues cease before tail shift (Galt & Fenaux 1990). Therefore, growth of the juvenile results solely from increase of cell volume (Fig. 11). The somatic cells increase their content of DNA to become polyploid (Fenaux 1971; Ganot and Thompson 2002), a result of successive rounds of DNA replication in the absence of karyokinesis and cytokinesis. Organ formation is completed after tail shift (stage 6) while the animal is still small and completely transparent. Every structure in the adult form can be easily observed in anesthetized whole animals at this stage by changing the focal plane (Fig. 10). The tail shift makes the definition of body axes difficult. In this report, body axes that are defined for the trunk region before the tail shift are used for the axes of the whole body after the tail shift, so the mouth is still anterior, although the tadpole swims backwards.
Here the characters of each tissue and organ are briefly summarized. The anatomy of appendicularians is reviewed in detail in Fenaux (1998b). Transverse sections at various levels and an interactive computer 3D reconstruction are available at the EURAPP site (http://www.obs-vlfr.fr/~eurapp/).
Fig.10Anatomy of stage 6 juveniles. (A?O) Serial optical sections. (A?E) Dorsal views of a juvenile from shallow to deep focal plane. (F?I) Right views. (J, K) Left views. (L?O) Ventral views. Blue arrows indicate seawater current. Red arrows represent flow of food. Seawater exits from the spiracles (blue circles). Fecal pellets are discarded from the anus (red circles). (P?V) Serial resin (Technovit) sections from anterior to posterior levels viewed from the posterior side. Dorsal is up. Approximate level of each section is indicated at the bottom of (E). (W) A section of the tail. Notochord is a tube with a lumen. Nerve cord is on the left of the notochord. Each of the dorsal and ventral epidermises is lined by a single flattened muscle cell. See details in text.
The entire epidermis consists of a single layer of epithelium, with no mesodermal lining below it. The only coelomic space is the pericardium. Most regions of the trunk epidermis are highly patterned to secrete the intricate house, and constitute the oikoplastic epithelium; the posterior part of the trunk region, covering the gonad and tail epidermis, does not secrete the house. The oikoplastic epithelium consists of approximately 2000 cells, and is subdivided into bilateral territories according to nuclear sizes and shapes, specific gene expression, and extent of polyploidization (30?1300 fold) (Thompson et al. 2001; Ganot and Thompson 2002). Each territory is related to certain house structures; for example, the Eisen and Fol fields are thought to secret the inlet and food-trap filter, respectively. The tail epidermis is a single layer of squamous epithelium. Left and right protrusions make fins (Fig. 10W). This contrasts to the ascidian larva, whose dorso-ventral fins are made of acellular materials.
The nervous system of O. dioica shows significant complexity, and has been intensively described (Olsson et al. 1990). The brain (or sensory vesicle) is present in the antero-dorsal region (Fig. 10I, J). A cavity on the left side of the brain (Fig. 9F and 10A) contains a statocyte that senses gravity. The statocyte contains a presumed calcareous statolith, and is connected to the brain wall via a stalk and via some sensory cilia emanating from the brain wall (Holmberg 1984). The whole brain consists of approximately 70 cells (Olsson et al. 1990; Soviknes et al. 2005). Axons from the brain are connected to sensory cells present in mouth and pharynx regions, namely, upper lip cells, lower lip cells, pharynx cells, and ventral sense organ cells. The ventral sense organ contains 30 cells and is suggested to be an olfactory organ (Bollner et al. 1986). In addition, a pair of bilateral epidermal sensory bristles, the Langerhans receptors, project out of the trunk (Fig. 10E) and probably sense mechanical stimuli. These cells receive neurite projection from a single cell in the tail ganglion (Holmberg 1986). The descending axons from the brain are connected to the spiracle to control its ciliar movement, and to some epidermis cells in the Eisen and Fol regions, plausibly to control house production.
The tail shows a remarkable 90 counterclockwise rotation relative to the trunk when viewed from the posterior, with neural tube on the left side as mentioned in the section, Embryogenesis. A thick caudal nerve runs posteriorly beyond the stomach (Fig. 10A), turns in the ventral direction, and innervates the tail ganglion (approximately 30 cells including glial cells; Soviknes et al. 2005) on the left side of the proximal region of the tail (Fig. 10J, L). A nerve cord runs along the left side of the tail with sporadic distributions of neuronal cell bodies (Fig. 10E, approximately 30 cells; Soviknes et al. 2007). On the left side of the tail nerve cord, a small hollow is surrounded by four (lateral and medial) ependymal cells in transverse section (Holmberg & Olsson 1984). The nerve cord innervates each dorsal or ventral muscle cell by a single axon of a cholinergic motor neuron (Flood 1973, 1975; Nishino et al. 2000; Soviknes et al. 2007). Recently, spatiotemporal pattern of neurogenesis in O. dioica was described in detail (Soviknes & Glover 2007).
Heart Haemolymph circulates within the body through blood vessels without endothelial cells, constituting an open blood system. There are no blood cells in the haemolymph. The heart is present ventrally between the left and right stomachs (Fig. 10M). It beats very fast, so it is easily recognizable in a living animal. The anatomical description of the Oikopleura heart is poor so far, and further investigations are definitely required.
The notochord is a synapomorphy characteristic of chordates. It is present in the center of the tail throughout its length. The notochord is a tube with a continuous lumen surrounded by flattened notochord cells, and serves as an axial endoskeleton (Fig. 10E, W). It is surrounded by basement membrane and outer fibrous extracellular matrix (Olsson 1965). Notochord consists of 19 disc-shaped cells plus 1 spherical terminal cell (t cell) at hatching, and they further divide to form flattened epithelial sheet surrounding the central lumen during larval development (Fig. 10W) (Nishino et al. 2001); in contrast, an ordinary ascidian larva has 40 notochord cells that do not divide further after hatching. The t cell is present at the posterior tip of the notochord throughout the life (Fig. 10E) and is never integrated into a notochord rod. Its function is as yet unknown.
The dorsal and ventral epidermis of the tail is lined with a single flattened layer of non-fused muscle cells (Fig. 10E, W) (Nishino et al. 2000). At hatching, a single row of eight muscle cells is aligned along the anterior?posterior axis on each side of the tail. During larval development, two additional small muscle cells of unknown origin appear at the tip of the tail (Nishino et al. 2001). Thus, 10 muscle cells line each side. This contrasts to ordinary ascidian larvae, which have 18 to 21 muscle cells arranged in three rows on each side. Muscle fibers are present only on the inner side of each muscle cell, and the epidermal side of the cells is filled with mitochondria. Each muscle cell is innervated by a single axon of a different cholinergic motor neuron (Flood 1973, 1975; Nishino et al. 2000; Soviknes et al. 2007). Cell bodies of ten pairs of the bilateral motor neurons are present along most of the length of the tail nerve cord.Digestive tract
Concentrated food particles with some seawater is introduced into the mouth from the food-trap filter of the house (Fig. 1A, B and 10B, K, P). Within the pharynx, the food is trapped by mucus secreted from the endostyle (Fig. 10I, Q) and brought into the esophagus (Fig. 10B, J, K, red arrows). The seawater goes ventrally to a pair of spiracles (Fig. 10B, D, K, S, T), gill apertures that open ventrally and directly to the outside of the body. A ciliary band just inside the opening drives the water current within the body to draw food particles into the pharynx, and is readily visible in living animals.
The digestive tract shows remarkable left?right asymmetry. Food particles progress into the esophagus, left stomach, and right stomach via a connection by means of the inside cilia (Fig. 10K, H, U, red arrows). Then processed food is delivered into intestine (Fig. 10F, G), rectum, and anus (Fig. 10E, H, M, N, T). Fecal pellets are discarded ventrally from the anus. Each region of the digestive tract has a distinct thickness of epithelium and shows specific fine-structure characters reflecting different physiological roles (Burighel et al. 2001).
In the tail of newly hatched larvae, an endodermal strand is present on the right of the notochord (Fig. 4N). According to Delsman (1912), most cells of the endodermal strand are absorbed into the trunk and give rise to a pair of oral (or buccal) glands (Fig. 10C), although this fact needs verification. Only two cells of the endodermal strand remain in the tail and give rise to two subchordal cells in the adult (Fig. 10E). The functions of oral gland and subchordal cells remain elusive. The presence, number, and distribution of subchordal cells provide good criteria for identifying species among the Oikopleurinae. Interestingly, only members of the genera Oikopleura, Folia, and Stegosoma with oral gland and subchordal cells show bioluminescence of the house (Galt and Fenaux 1990).¢Back to CONTENTS
Juvenile development and gametogenesis
After the tail shift and initiation of feeding on the first day, juveniles grow rapidly (Fig. 11). A gonadal rudiment is present in the posterior-ventral position and grows continuously to extend along the posterior epidermis and eventually occupies the posterior half of the body. Initially on day 5, males and females cannot be distinguished. Sperm and oocyte formation is remarkably rapid, taking a few hours (Nishino & Morisawa 1998). Fine structure during spermatogenesis is described in detail in Martinucci et al. (2005). Spermatogenesis proceeds in a highly synchronized way in the whole testis (syncytium), then a number of sperm are produced.
Oogenesis proceeds in a rather unusual way. The whole ovary is a single syncytium, called a coenocyst (Ganot et al. 2007a, b). Oogenesis is divided into five phases. After the proliferation of germ nuclei in a common cytoplasm connected via cytoplasmic bridges, the fate of each nucleus is determined to be meiotic nucleus or nurse nucleus at phase 1 on day 2, but all nuclei remain within the common cytoplasm. Only meiotic nuclei are destined to be oocytes. Meiotic nuclei enter into the zygotene stage, while nurse nuclei start DNA synthesis to become polyploid, gradually increasing nuclear size in phase 2. In phase 3 on day 4, meiotic nuclei are arrested at prophase I. The coenocyst cytoplasm increases rapidly by active transcription of nurse nuclei. In phase 4, some meiotic nuclei are eventually selected to be oocytes, and grow so that the cytoplasm of the coenocyst is transferred through a ring canal. The ring canal marks the future vegetal pole of the egg. Meiosis is resumed. In final phase 5, full-grown oocytes are arrested at metaphase I (Fig. 11, day 5 female). Other nuclei undergo apoptosis. The size of full-grown oocytes is constant, and the number of oocytes produced by a single female depends on the available amount of food and thus on the final volume of the coenocyst. Efficient numerical adjustment of oocyte production to environmental resources is important to the fitness of this species. How animals select an adequate number of meiotic nuclei to become oocyte nuclei during phase 4 is interesting but unknown.
Under natural conditions, oogenesis proceeds as described above, and the animals spawn on day 5 without a specific spawning cue in lab culture. However, when animals are stressed, for example by high population density, they produce eggs earlier than normal. In this case, the coenocyst is smaller and the number of eggs becomes less. At spawning, the gonad of the female bursts via rupture of the gonadal wall at the specialized spawning anlagen on the dorsal side of the gonad (Ganot et al. 2006). In contrast, sperm come out of a small hole in the sperm duct on the dorsal side, and spawning lasts several minutes. Animals of both sexes die a few hours after spawning, and thus the animal is semelparous.
Fig.11Juvenile development from the end of day 1 (after tail shift) to day 5 (sexual maturation) at 20 C. Photos are shown at the same scale. On day 5, the animals mature sexually, and male and female become distinguishable by the gonad filled with sperm or eggs. They spawn on day 5. See details in text.
Oikopleura dioica is characterized by its simplified life habit and organization, and shows the following features, some of which are eminently advantageous for developmental and genetic studies: (1) Swimming tadpole up to adult stage. (2) Abundance in ocean. (3) Dioecious and semelparous. (4) Quick development and short life cycle. (5) Feasibility of closed and inland culture over generations. (6) Transparency. (7) Small cell number in embryos and adults. (8) Small genome and high density of genes (1 gene / 5 kb).
Taking advantage of the extremely short life cycle of O. dioica, genetic approaches involving transgenesis and mutagenesis would be fruitful. The syncytial coenocyst of the female gonad is a good candidate for a DNA injection site to yield a large number of offspring with introduced DNA. Maintenance of frozen sperm will also facilitate studies using transgenesis. The high density of genes and assembled genome sequences will greatly help mutagenesis studies. Enhancer trap and insertional mutagenesis with the transposon Minos have recently succeeded in ascidians (Sasakura et al. 2005; Awazu et al. 2007; reviewed in Sasakura 2007). Minos should also work in appendicularians. Mutagenesis using transposons could facilitate the identification of mutated genes in the assembled genome sequences. Therefore, it will be a good start with this kind of approach in future studies.
Several particularly interesting issues will make good targets for studies on O. dioica:
Establishment of chordate body plan and cell fate specification during embryogenesis: These have been extensively investigated in ascidians (Satoh, 1994; Nishida, 2005). Comparison of these processes in O. dioica would be interesting because its eggs do not show directed movement of egg cytoplasm just after fertilization, which is required to establish the secondary embryonic axis in ascidians. In addition, cell fate restriction is completed much earlier: at the 32-cell stage rather than at the 110-cell stage in ascidians.
Left?right asymmetries: The cleavage pattern shows left?right asymmetry as early as the 4-cell stage. Juveniles and adults also show remarkable left?right asymmetries in the tail and digestive tract. What factors are involved in the generation of these asymmetries and how are early and late asymmetries causally related?
Pattern formation of oikoplastic epithelium: The epidermis is spatially patterned for the secretion of the intricate house. Such an elaborate pattern would involve various processes of pattern formation. Analysis of these processes will contribute to our understanding of how organisms create precise patterns to generate highly complex but well organized structures.
Polyploidy of somatic cells: How do cells transit from ordinary cell division to DNA synthesis without cell division after the tail shift? What is the cue, and how are the specific numbers of endocycles of different cell types controlled?
Sex determination: O. dioica is the sole dioecious species so far reported in the subphylum Tunicata. Therefore, this character is apparently derived. Nothing is known about sex determination. Is sex genetically controlled? How did the sexes diverge during evolution? Was the evolutional transition from hermaphroditic to dioecious state prosecuted by a few trivial steps? These questions are fundamental to an understanding of the evolution of sex in animals.
Oogenesis: Oogenesis proceeds in a single giant syncytium, the coenocyst. How oocyte production is adjusted to the available resources is unknown. How are fates of nurse cell and meiotic nuclei specified, and then how do animals select an adequate number of meiotic nuclei to become oocytes?
Control of swimming modes: The swimming pattern of larvae is developmentally controlled. In addition, Three kinds of swimming mode are distinguishable in juveniles, with distinct frequency and curvature of tail beat, in juveniles and adults. Juveniles in the house, outside the house, and expanding the house show different swimming modes. How is swimming mode switched by a neural circuit with a small number of neuronal cells?
In addition, detailed further comparison of the genomes of appendicularians, ascidians, and Amphioxus will shed light on the evolutionary history and mechanisms of chordate evolution. In conclusion, O. dioica is an attractive organism for developmental, evolutionary, and physiological studies of chordates, offering considerable promise for future genetic studies.¢Back to CONTENTS