Printer Friendly

Continued growth and cell proliferation into adulthood in the notochord of the appendicularian Oikopleura dioica.


The subphylum Urochordata (or Tunicata) of the phylum Chordata includes three classes: Ascidiacea, Thaliacea, and Appendicularia. Urochordate species exhibit diverse forms, including free-swimming, sessile, and colonial. In most urochordates the larva has a chordate body plan with a trunk or head region and a motile tail containing a notochord and a dorsal nerve cord (Katz, 1983; Meinertzhagen and Okamura, 2001). Ascidians, represented by the most studied of the urochordates, Ciona intestinalis, undergo a metamorphosis from a free-swimming larva into a solitary sessile adult. The metamorphic transition involves the loss of tail structures, including the notochord. The salps and doliolids (Thaliacea), closely related to ascidians, also undergo a metamorphosis through which the notochord (if initially formed at all) is lost, but they maintain a pelagic life, in some cases as colonial forms (Bone, 1998). In contrast, the appendicularians, represented by Oikopleura dioica, retain the larval body form with notochord and dorsal nerve cord into adulthood. For this reason they are also called Larvaceans.

Urochordates have become popular model organisms in the effort to deduce information about the common chordate ancestor. Recent genomic comparison has provided evidence that urochordates, not cephalochordates, represent the closest living relatives of vertebrates (Delsuc et al, 2006), although this has been contended (Bourlat et al., 2006). The genomes of three urochordate species have been sequenced--Ciona intestinalis (Dehal et al., 2002), Ciona savignyi (Vinson et al., 2005), and Oikopleura dioica (annotation in progress). Some analyses suggest that chordates descended from a free-swimming urochordate such as Oikopleura (Wada, 1998; Swalla et al., 2000).

The notochord is one of the representative traits of the phylum Chordata, together with a dorsal tubular nervous system (Ruppert and Barnes, 1994). In most larval urochordates the notochord provides a stiff skeletal element for the tail, but it must also be flexible to facilitate swimming and other movements. During embryonic development in the vertebrates, it also serves as a source of inductive signals that contribute to the patterning of adjacent neuroectoderm, paraxial mesoderm, and endoderm (Yamada et al., 1991; Koseki et al., 1993; Cleaver et al., 2000). The early morphogenetic patterns of notochord development are similar throughout the chordate radiation, but strikingly distinct patterns emerge at the histological level as development proceeds. In vertebrates, the notochord eventually becomes a nucleus for the condensation of mesenchyme in the formation of the vertebrae. In this process it can acquire a variety of forms in the adult, ranging from a continuous vertebral core, as in fish and amphibians, to a segmented form as the nucleus pulposus of the intervertebral discs, as in most mammals (Kardong, 1998). The cephalochordate notochord is one example of an evolutionary novelty because it is made up of muscle cells that are innervated through contact with the ventral aspect of the nerve cord (Flood, 1975). Unlike vertebrates, the tunicates and cephalochordates never convert their notochord into a vertebral column or backbone. In ascidians like Ciona intestinalis, the notochord, along with other tail structures, is completely resorbed on metamorphosis (Satoh, 1994; Godeaux et al., 1998). Oikopleura dioica retains a chordate body plan, and also its notochord, throughout development, and undergoes only a minor metamorphosis that consists of a simple shift in the orientation of the tail with respect to the trunk (tail-shift) from co-linear to almost perpendicular to the trunk axis. (Bone, 1998).

Notochord development has been extensively studied in Ciona intestinalis. When first visible, the notochord consists of a stack of 40 cells that goes through a series of morphological stages resulting in the formation of a hollow tube. The early notochord cell lineage in Ciona was first described by Conklin (1905) and later by Ortolani (1955) and Nishida (1997); live imaging of notochord formation was performed by Miyamoto and Crowther (1985). A comparative description of ascidian notochord development is found in Burighel and Cloney (1997) and includes four characteristic stages (stage I-IV). The notochord rudiment is formed during the blastula stage. Intricate vertical and horizontal cell rearrangements ("convergence and extension") produce an elongated rod of 40 disc-shaped cells ("stack of coins," stage I), well described in Ciona intestinalis (Miyamoto and Crowther, 1985), Acidia callosa (Cloney, 1964), Boltenia villosa, and Corella inflate (Munro and Odell, 2002a, b). This process is apparently common to many chordates, including amphioxus and some vertebrates (Mookerjee et al., 1953; Flood, 1975; Cloney, 1990). The notochord cells then become reduced in diameter to form short cylinders. In most ascidian species (both solitary and colonial) the next developmental step (stage II) is a process of vacuolarization, in which small lenticular pockets form at the boundaries between neighboring notochord cells and then enlarge to vacuoles. The timing and position of the first vacuoles differ among the ascidians (Burighel and Cloney, 1997). During stage III, the notochord cells become irregular in shape as the vacuoles expand and eventually coalesce to form a confluent hollow lumen with an inner matrix. During stage IV, the 40 notochord cells become flattened against the notochord sheath in an epithelial manner, and the fully developed notochord becomes a tubular structure closed at both ends and filled with an acellular matrix of secretory products.

Early histological studies addressed notochord development to the larval stage in the appendicularians Oikopleura dioica and Oikopleura longicauda (Fol, 1872; Seeliger, 1893; Martini 1909). The process is similar to that described in ascidians, the largest difference being that the notochord in the newly hatched larva consists of 20 cells instead of 40 cells (Burghiel and Cloney, 1997). The ultrastructure of the adult Oikopleura dioica notochord was studied by Olsson (1965), who described a lumen covered by a thin epithelium of flattened cells and containing a sulphur-rich proteinaceous secretion. None of these earlier reports provide measurements of tail length or counts of notochord cells.

The Brachyury (T) gene is essential for notochord development (Cunliffe and Ingham, 1999), and its expression pattern has been studied in both appendicularians (Oikopleura dioica--Bassham and Postlethwait. 2000: Oikopleura longicauda--Nishino et al., 2001) and ascidians (Halocynthia roretzi--Yasuo and Satoh, 1994, 1998; Ciona intestinalis--Corbo et al., 2001; Satoh, 2003). Bassham and Postlethwaite (2000) covered embryonic stages up to hatching, and thus focused on the notochord rudiment and on the similarities between Oikopleura and ascidians. Nishino et al. (2001) followed notochord development past the early hatching stage and noted that cell numbers appear to increase in the juvenile, but did not provide cell counts or measurements of tail length.

Here we use fluorescence confocal microscopy and BrdU labeling to provide the first quantitative analysis of appendicularian notochord development and growth, including cell numbers and morphologies at different developmental stages. We report features of cell division and relative movement from the newly hatched larva to the adult that correspond to stages I-IV of notochord development seen in ascidians. We then show that with further maturation and growth of the Oikopleura notochord, cell numbers reach levels far exceeding the 20 cells seen in the hatchling and the 40 cells seen in the ascidian larva. Continued BrdU incorporation in notochord cells parallels the increase in cell number, and expression of a cell-cycle marker demonstrates that this is associated with cell division. The extensive proliferation of notochord cells into adulthood is similar to what is seen in vertebrates and is likely to be necessary for maintaining the mechanical properties of the appendicularian tail as it lengthens and is utilized in the adult behavioral repertoire.

Materials and Methods


Populations of Oikopleura dioica Fol, 1872 were collected from fjords in the area of Bergen, Norway, and cultured at 14-15 [degrees]C in 6-liter beakers with constant stirring. The duration of the life cycle was 5-6 days. Mature males and females were placed together in 4-liter volumes of seawater and allowed to spawn. Metamorphosis took place 12-14 h after spawning, at which time the animals expanded their first house and began filter-feeding. The animals then grew continuously and on day 5 the gonads increased dramatically in size, surpassing the size of the head/trunk at spawning. Cultures were diluted 1:6 on day 1 and 1:1 on day 2 after spawning, and then animals were pipetted into clean seawater on each of the following days. The animals were maintained on cultured algal strains of Isochrysis galbana, Chaetoceros calcitrans, and Synecococcus sp., supplemented with Rhodomonas sp. from day 3 onward. Animals for BrdU labeling were obtained by in vitro fertilization. This was accomplished by collecting oocytes from mature females in glass spawning bowls (about 25-ml volume) and adding a diluted suspension of sperm obtained from the ejaculates of 1-2 mature males in sterile filtered seawater. Development continued at room temperature, 18-20 [degrees]C, and was rapid, with the first cell cleavages occurring after 20-30 min and subsequent cleavages every 5 min. The blastula stage was reached at 1.5 h post-spawning, hatching at about 3.5-4 h, and the tail shift stage at 12-14 h. For BrdU labeling, particular care was taken to use batches of fertilized eggs that were synchronized in their initial cell divisions. Hereafter, developmental timepoints are abbreviated as follows: H, hours postfertilization; D, days postfertilization.

Phalloidin and To-Pro-3 staining

Animals were fixed for phalloidin staining in 4% paraformaldehyde in 0.1 mol [1.sup.-1] 3-(N-morpholino)propanesulfonic acid (MOPS), 0.5 mol [1.sup.-1] NaCl, 2 mmol [1.sup.-1] EGTA pH 8.0, and 0.2% Triton X-100, overnight at 4 [degrees]C. All staining procedures were performed in 0.2-ml PCR tubes. All washes and incubations used 0.2-ml volumes and lasted 5 min or more, without agitation. The animals were washed three times in PBST (PBS with 0.1% Tween-20). In an Eppendorf tube, Phalloidin-Alexa488 in methanol (Molecular Probes) was air-dried in the dark to evaporate the methanol (MeOH inhibits actin detection), and then reconstituted in PBST (3% w/v) containing 1% BSA. The animals were incubated in this solution overnight at 4[degrees]C. They were then washed three times in PBST and stained with To-Pro-3 iodide (Molecular Probes) diluted 1:500 in PBST for 4 h at room temperature. A brief wash was performed in PBST, and the specimens were placed in mounting medium (Vectashield, Vector Laboratories) on depression slides (VWR) and stored at -20[degrees]C until examined on a confocal microscope.


Animals were fixed for immunohistochemistry for 1 h on ice in 4% paraformaldehyde in 0.1 mol [1.sup.-1] MOPS (pH 8.0) and 0.5 mol [1.sup.-1] NaCl, and stored in 0.1 mol [1.sup.-1] MOPS and 0.5 mol [1.sup.-1] NaCl. They were subsequently washed in PBS (pH 7.0) with 0.1% Tween-20 and 2 mmol [1.sup.-1] EDTA (washing solution) four times (5 min or more per wash) prior to immunohistochemistry.

All immunohistochemical procedures were performed in 0.5-ml PCR tubes for ease of visualization. The incubation times used below are crucial to allow sufficient antibody penetration (to preserve tissue integrity, harsher treatments such as proteinase K incubation were not used). All washes and incubations used 0.5-ml volumes and lasted 5 min or more, without agitation. Immunohistochemistry was performed as described in Soviknes et al. (2005). Pre-immune blocking in "blocking solution" (3% acetylated bovine serum albumin (BSA, Sigma) and 0.1% Tween-20 in PBS) was carried out overnight at 4[degrees]C Blocking solution without primary antiserum was used as a negative control in all experiments. The animals were then washed four times in washing solution and incubated for 3 days at 4[degrees]C in primary antiserum (Phospho-histone-H3, Cell Signalling Technology, 9706S,) diluted 1:100 in blocking solution. They were then washed five times in washing solution and post-fixed in 4% paraformaldehyde with 0.1 mol [1.sup.-1] MOPS and 0.5 mol [1.sup.-1] NaCl for 1 day or overnight at 4[degrees]C. They were again washed four times in washing solution and then incubated with the secondary antibody (Goat-anti-mouse HRP-conjugated, DAKO) diluted 1:500 in blocking solution for 3 days at 4[degrees]C. This was followed by another five washes in washing solution. Fluorescent staining of the secondary antibody was performed using the TSA amplification reagent (Perkin Elmer), diluted 1:500 in the amplification diluent (Perkin Elmer) and incubated 2-4 h at room temperature. A brief wash with 2 mmol [1.sup.-1] EDTA in PBS preceded nuclear staining using To-Pro-3 iodide (Molecular Probes) diluted 1:500 in 2 mmol [1.sup.-1] EDTA in PBS and incubation overnight at 4[degrees]C. Specimens were placed in mounting medium (Vectashield, Vector Laboratories) on depression slides (VWR). Coverslips were added and sealed with nail polish before storage at -20[degrees]C until examination with a confocal microscope.

BrdU labeling

A 3 mmol [1.sup.-1] BrdU solution was prepared from BrdU powder (formula weight: 307g/mol, Sigma) dissolved in sterile filtered seawater for 2 h. The solution was then diluted to 1 mmol [1.sup.-1] and 100 [micro]mol [1.sup.-1] in sterile filtered seawater for incubation of embryos.

Two approaches were used for BrdU labeling. One was to incubate different batches of embryos at sequential developmental stages, using a 15-min pulse of 1 mmol [1.sup.-1] BrdU, and then letting the animals develop until 15 h for assessment of BrdU labeling. This provided information in preliminary experiments about the general window of time during which cells became postmitotic. The other approach, which was used to obtain more precise information about spatiotemporal patterns of cell genesis, and which has provided the data documented in detail here, was to incubate different batches of embryos with 100 [mciro]mol [1.sup.-1] BrdU continuously, starting at different developmental stages, until 9 h, for assessment of BrdU labeling. This approach labels all cells that undergo DNA synthesis (for example, at mitosis) after the start of BrdU incubation, leaving any cells that have become postmitotic by the start of incubation unlabeled. In both cases, embryos were washed well with sterile filtered seawater at the end of BrdU incubation.

After washing, the animals were fixed overnight at 4[degrees]C in 4% paraformaldehyde in 0.1 mol [1.sup.-1] MOPS (pH 8.0), 0.5 mol [1.sup.-1] NaCl, 0.5 mol [1.sup.-1] EGTA, and stored in 0.1 mol [1.sup.-1] MOPS and 0.5 mol [1.sup.-1] NaCl for up to 1 week at 4[degrees]C. They were then washed in PBS (pH 7.0) containing 0.2% Triton X-100 (PBST) four times for 5 min or more per wash. A DNase incubation was performed to expose the incorporated BrdU as epitope. Animals were incubated in PBST containing 1% BSA, 1 mmol [1.sup.-1] MgS[O.sub.4], 10 u/ml DNase I overnight at 4[degrees]C and then rinsed in PBST containing 1% BSA and 2 mmol [1.sup.-1] EDTA. Primary antibody incubation was in PBST containing 1% BSA and anti-BrdU (1:100, Accurate Chemical, catalog number YM8006, mouse monoclonal antibody clone YU-4) for 3 days at 4[degrees]C. Preparations were then washed in PBST three times, and a postfixation was performed with 4% paraformaldehyde in PBS (pH 7.4), for 1 h at 4[degrees]C. Preparations were again washed in PBST three times. Blocking was performed in PBST containing 1% BSA for 1 h at room temperature. Secondary antibody incubation was in PBST containing 1% BSA and FITC-conjugated donkey anti-mouse IgG (1:1000, Jackson ImmunoResearch, catalog number 715-095-151) for 2 days at 4[degrees]C. Preparations were again washed in PBST three times prior to nuclear staining using To-Pro-3 iodide (1:100, Molecular Probes) in PBS containing 2 mmol [1.sup.-1] EDTA. After a brief wash in PBST, the preparations were placed on glass slides in mounting medium (Vectashield, Vector Laboratories) under coverslips.


In prior studies (Soviknes et al., 2005, 2007) differential interference (Nomarski) optics using a Leica DMLB microscope equipped with a 40X oil immersion objective (numerical aperture 1.00) were used initially to locate and identify the cerebral and caudal ganglia, the first brain nerves, and the dorsal nerve cord. With a short period of training, the same structures could readily be located and identified in fluorescence images obtained by confocal microscopy. In the present study we used only confocal fluorescence microscopy, with a Leica TCS laser scanning confocal microscope equipped with both 40X and 63X oil immersion objectives (numerical aperture 1.25 and 1.40, respectively). Image stacks of about 1-[micro]m optical sections were acquired using Leica PowerScan 2.5 software.


Using phalloidin staining for actin and nuclear staining with To-Pro-3, we were able to follow with confocal microscopy the cellular events associated with notochord development (Figs. 1, 2). In the newly hatched embryo, the notochord has the typical "stack of coins" appearance of stage I. The perimeter of each notochord cell is strongly stained with phalloidin and there are clear cell boundaries between the different notochord cells. The notochord nuclei are centrally positioned within each notochord cell (Fig. 1A). In Figure 1, the 1st, 11th, and 19th notochord cells (n1, n11, and n19) are labeled to facilitate counting. In the hatchling, n1 appears rounded anteriorly, whereas n2-n19 are more disc-shaped. The cell posterior to n19 (n20) has a more knob-like appearance. At this stage other cell types, such as the developing nerve cord and the two rows of muscle cells, can be identified easily (see also Soviknes et al., 2007).



About half an hour later, at 4-h postfertilization (4H; Fig. 1B), the notochord cells are still disc-shaped and have centralized nuclei, but they have elongated. Cell n20 maintains its knob-like appearance. Just prior to this stage, the tail has undergone a 90-degree twist in the longitudinal axis related to the trunk, such that the original dorsal aspect of the notochord (along which the clearly visible caudal nerve cord courses) now lies on the left side. A row of endodermal cells lies in close relation to the ventral aspect (now right side) of the notochord. In all figures here, however, the notochord is oriented with the original dorsal aspect up.

At 5H (Fig. 1C), the notochord has elongated more, and two additional cells appear, protruding from each side of the notochord between n19 and n20. These are seen more clearly at 6H (inset, Fig. 1C). The origin and fate of these two cells is unclear. They are not obviously present at that position at later stages, nor have we observed similar cells associated with the notochord at other positions at later stages. We have therefore not defined them as notochord cells and do not include them in notochord cell counts (see below). The cell at the position of n20 now has a more compact nucleus and much denser nuclear staining.

At 5H, the first signs of muscle fibers can be detected along the length of the notochord. The phalloidin staining of cell-membrane-associated actin between adjacent notochord cells appears to become thinner in the central area of contact, and small actin-free bubbles start to appear. Both of these features are probably the earliest stages of vacuolization, which is clearly seen in the middle portion of the notochord by 7H (Fig. ID), by which time the vacuoles have increased in size and appear to fuse between neighboring cells. Vacuolization spreads rapidly to the anterior and posterior ends of the notochord (inset, Fig. 1D). The n20 cell, with the compact nucleus and dense nuclear staining, now closely resembles what has been called the terminal cell (t-cell, Nishino et al., 2001). The fusion of vacuoles into a continuous lumen is almost complete by 8H (Fig. 1E). The notochord cells change shape as a consequence, becoming more flattened cells lying dorsal and ventral to the lumen. In the specimen shown in Figure 1E, only n1 and n4 still have a central position, not yet shifted dorsally or ventrally. At this stage the caudal ganglion is well defined and the different clusters of neurons in the caudal nerve cord acquire their axial positions.

Until 8H, the number of notochord cells has remained constant at 20 (Fig. 3), but already at this stage there are initial signs of notochord cell mitosis, as seen with nuclear staining in Figure 1E (white asterisk). By 9H, proliferation of notochord cells is obvious (Figs. 1F, 3). The notochord now appears with a complete lumen circumscribed by 32 notochord cells, with one at each end and the others positioned in two rows, one dorsal and one ventral. Proliferation appears to be more rapid in the dorsal row, which contains 17 cells compared to 13 cells in the ventral row at this stage. The t-cell no longer appears as part of the notochord proper, but rather as a caudal appendage. The cytoplasm of the elongated, epithelial-like notochord cells has become strongly stained by To-Pro-3, presumably reflecting a high concentration of cytoplasmic RNA.

By 12H (Fig. 1G, H), the anterior third of the notochord alone contains more than 20 cells. Now there are clearly four rows of cells--one dorsal, one ventral, and one on each side. The notochord nuclei have become more uniform in morphology, rounded when viewed en face and flattened when viewed tangential to the notochord circumference. They are distributed relatively evenly around the perimeter in the transverse plane to provide complete coverage of the notochord lumen. The first and last notochord cells are located at the respective termini of the notochord and thereby close the lumen.

During the next 2.5 days, the number of notochord cells continues to increase, having quadrupled from 8H to Dl postfertilization and doubling again from Dl to D3 (Fig. 3A). From the 12H stage to adult stages, the only major morphological change is that the increasing numbers of notochord cells become distributed more evenly along the notochord, as seen at D3 (Fig. 1I, J), but the spacing is different for the lateral cells and the dorsal and ventral cells. Thus, in the anterior end (Fig. 1I), n2-n5 are positioned somewhat intermediate to the dorsal, ventral, and lateral rows, whereas at the posterior end (Fig. 1J), lateral cells are absent, the last six cells being found only in the dorsal and ventral rows. In contrast, it is apparent that the middle portion of the notochord has about two lateral cells per dorsal notochord cell (Fig. 1I, J).

To assist the reader in discerning the notochord and notochord cells in the panels of Figure 1, we have traced these structures for Figure 1 (E-J) in Figure 2.

In Figure 3B, the length of the notochord from the most anterior cell to the most posterior cell is shown at different stages. There is a steady increase in notochord length from hatching to D3, and also from 4H to 8H when notochord cell number is constant. Thus, at first the notochord grows through the elongation of the initial complement of 20 notochord cells; thereafter, growth continues through the addition of new cells.

To confirm that the increase in the number of notochord cells involves cell proliferation, we first examined the notochord in a series of embryos and larvae that were labeled with BrdU in another study of neurogenesis (Soviknes and Glover, 2007). In Figure 4, the temporal pattern of BrdU incorporation up to the 9H stage is shown using a cumulative labeling approach. In the period from 4H to 8H, when the number of notochord cells is constant at 20, substantial numbers of notochord cells continue to synthesize DNA, suggesting a high degree of endoreduplication. BrdU incorporation continues in many notochord cells at least until 15H, in parallel with the increase in notochord cell number, as demonstrated in BrdU pulse-labeling experiments (Fig. 5 A, B). Since the BrdU labeling does not distinguish between endoreduplication and mitosis, however, we also used the cell-cycle marker phospho-histone-H3 to assess the presence of mitotic cells. At 9H, during the first doubling of notochord cell number, scattered phospho-histone-H3-immunopositive notochord cell nuclei were present along the length of the notochord (Fig. 5C-I). The fraction of notochord cells expressing this marker in any given 9H preparation was low (1-3 per preparation), consistent with a short half-life of the phospho-histone-H3 protein. Phospho-histone-H3-immunopositive notochord cell nuclei were also seen in D3 preparations but were very scarce (not shown), consistent with a lack of synchronization of notochord cell mitoses during the later phase of proliferation.


Early notochord development

The early development of the notochord and the expression of the notochord-specifying gene Brachyury (T) is similar in Oikopleura dioica, Oikopleura longicauda, and ascidians (Bassham and Postlethwait, 2000; Nishino et al., 2001; Yasuo and Satoh, 1994, 1998). The principal difference is that the Oikopleura species generate 20 notochord cells at hatching (Bassham and Postlethwait, 2000; Nishino et al., 2001, respectively), whereas ascidians generate 40 (Burighel and Cloney, 1997). Bassham and Postlethwait (2000) observed 19 Brachyury-expressing cells tightly stacked in the nascent notochord prior to hatching, plus one additional cell posterior to the last notochord cell that evidently became incorporated into the notochord secondarily. We see 20 notochord cells in a row at hatching, with no sign of a separation between the last two cells. However, the last cell (n20) has a knob-like appearance different from the disc shape of nl-n19, and as in Oikopleura longicauda, eventually appears to separate from the end of the notochord to give rise to the so-called "terminal-cell" (t-cell; Bassham and Postlethwait, 2000; Nishino et al., 2001). The t-cell is first identifiable at 6H by its oblong nucleus and dense nuclear staining (inset Fig. 1C, marked with a bent arrow). At 7H it no longer appears to be integrated into the notochord and only 19 disc-shaped notochord cells can be seen. The function of the t-cell is not known.



At 5H, two cells appear between n19 and n20 and protrude laterally from the notochord axis (Fig. 1C, and inset). In an earlier report we showed that the ninth pair of muscle cell nuclei can be identified at 8H (Soviknes et al., 2007), but the origin of this pair of muscle cells is unclear. We suggest that the two cells that appear between n19 and n20 give rise to the ninth pair of muscle cells. At 6H, these cells are aligned with the eighth pair of muscle cell nuclei, and the ninth muscle cell pair seen later is positioned in the same location just anterior to the notochord tip.

It is apparent that the origins of the most caudal elements in the Oikopleura tail are complex, and there are indications that at least two cell types, muscle and the t-cell, may derive from notochordal mesoderm. Lineage tracing combined with dynamic imaging methods will be required to resolve this issue.

Formation of the tubular notochord

In the ascidian Ciona, the notochord rudiment completes initial cell rearrangements to form a rod of 40 disc-shaped cells organized as a "stack of coins" (Miyamoto and Crowther, 1985; Cloney, 1990). Development from this point through the formation of the definitive tubular notochord with epithelial notochord cells has been classified into stages (I-IV; Burighel and Cloney, 1997). In the appendicularian Oikopleura longicauda, the initial rod-like notochord contains 19 cells in a stack-of-coins arrangement, plus one posterior knob-like cell (Nishino et al., 2001). Development in this species has also been described in a series of stages (stages 1-6; Galt and Fenaux, 1990). The stack-of-coins stage is stage I in ascidians and stage 1 in Oikopleura longicauda. In ascidians, the onset of vacuolization, stage II, and the process of vacuole coalescence, stage III, correspond to early and late stage 3 in Oikopleura longicauda (stage 2, skipped). Finally, the tubular notochord form in ascidians, designated as stage IV, corresponds to stages 4-6 in Oikopleura longicauda. We have now extended the developmental description of notochord development in Oikopleura dioica through the terminal stage defined for ascidians and Oikopleura longicauda, and find that the pattern of morphological events is essentially the same. We can thus present a timeline of notochord development in Oikopleura dioica, as shown in Table 1.

It is interesting to note that in some species of acidians, the notochord does not progress through the full sequence of developmental stages. Some arrest development as early as stage I, and others at stages II or III (Cloney, 1990). Among doliolid tunicates, some species follow the ascidian pattern, whereas others do not develop a notochord structure at all (Godeaux, 1990). These phylogenetic variants demonstrate that the developmental events underlying the formation of the larval notochord are subject to modification, presumably linked to differences in larval and adult lifestyle and tail function. The shared element of a freely swimming larval form in Ciona and the two Oikopleura species may thus be related to the conservation of notochord developmental pattern in these species.


The proliferating notochord

Appendicularians maintain their tail throughout life, and as solitary, freely swimming filter-feeders, both their lifestyle and their adult chordate body plan differ from those of other tunicate groups (Fenaux, 1998; Fenaux et al., 1998). By contrast, most ascidian and doliolid larvae have a chordate body plan but resorb the tail and notochord at metamorphosis (Cloney, 1990; Godeaux, 1990). In the appendicularian Oikopleura longicauda, it has been noted that the number of notochord cells in the adult must exceed the 20 cells seen in the larva (Cloney, 1990; Nishino et al., 2001), but no previous reports have counted the number of adult notochord cells or described the anatomy of the adult notochord, aside from an ultrastructural study performed in Oikopleura dioica (Olsson, 1965).

Here we have shown that within an hour after attaining the terminal notochord developmental stage previously described for ascidians and Oikopleura longicauda, the number of notochord cells in Oikoleura dioica begins to increase, and continues to increase for at least 2.5 days, reaching by this point an 8-fold increase over the initial number of 20. This increase in cell number parallels the increase in overall tail length that occurs as the animal continues to grow through adulthood. Presumably a similar increase occurs in Oikopleura longicauda. Thus, in Oikopleura, the formative stages of notochord development are succeeded by a dramatic proliferative phase that is clearly necessary to support structural growth. Similar growth-related proliferation occurs in the vertebrate notochord. This phylogenetic variant is of particular interest in terms of the regulation of the cell cycle in notochord cells and the control of cell intercalation that must occur to maintain a constant tubular organization as the notochord lengthens. An interesting feature of the cell intercalation is that the lateral notochord cells and the dorsal and ventral notochord cells are regulated differentially. There are about 20% more lateral notochord cells than dorsal and ventral notochord cells, and lateral cells are not found in the most posterior part of the notochord (see Fig. 1J). The difference might be related to the anatomy of the Oikopleura tail and the way the tail is used in swimming movements. The fins of the tail extend from the dorsal and ventral aspects, and when swimming, the tail oscillates on the bilateral axis. It seems likely that this exercises a higher mechanical stress on the lateral aspects of the notochord and at the same time an increased demand for flexibility, necessitating a higher density of lateral cells to resist mechanical strain during the bending movements.

An alternative explanation for the increase in notochord cell number is that cells are recruited into the notochord from outside germinal sources. The fact that BrdU incorporation continues in notochord cells after the initial complement of 20 cells has been attained is highly suggestive of intrinsic proliferation, and this is demonstrated directly by the expression of the cell-cycle marker phospho-histone-H3. However, BrdU incorporation in the period from 4H (4-h postfertilization) to 8H could also represent endoreduplication, because the number of notochord cells is constant during this period. Endoreduplication is a common feature of epithelial cells in Oikopleura (Ganot and Thompson, 2002) and could be present in the notochord as well. Assessment of ploidy in future studies will settle this issue. Another important question for future studies is how the cell cycle is regulated in the Oikopleura notochord to generate the proliferative phases demonstrated here.


We thank Daniel Chourrout for providing the generous research facilities at the Sars International Centre for Marine Molecular Biology; Jean-Marie Bouquet, Helen Eikeseth Ottera, and David Osborne for maintaining the Oikopleura culture; and Di Jiang for critically reviewing the manuscript. This research was supported by a grant from the Norwegian Research Council ("OIKOGEN," 146653/431).

Literature Cited

Bassham, S., and J. Postlethwait. 2000. Brachyury (T) expression in embryos of a larvacean urochordate, Oikopleura dioica, and the ancestral role of T. Dev. Biol. 220: 322-332.

Bone, Q. 1998. The Biology of Pelagic Tunicates. Oxford University Press, New York.

Bourlat, S. J., T. Juliusdottir, C. J. Lowe, R. Freeman, J. Aronowicz, M. Kirschner, E. S. Lander, M. Thorndyke, H. Nakano, A. B. Kohn, et al. 2006. Deuterostome phylogeny reveals monophyletic chordates and the new phylum Xenoturbellida. Nature 444: 85-88.

Burighel, P., and R. A. Cloney. 1997. Urochordata: Ascidiacea. Pp. 221-347 in Microscopical Anatomy of Invertebrates: Hemichordata, Chaetognatha and the Invertebrate Chordates, Vol. 15, R. W. Harrison and E. E. Ruppert, eds. Wiley Liss, New York.

Cleaver, O., D. W. Seufert, and P. A. Krieg. 2000. Endoderm patterning by the notochord: development of the hypochord in Xenopus. Development 127: 869-879.

Cloney, R. A. 1964. Development of the ascidian notochord. Acta Embryol. Morphol. Exp. 7: 111-130.

Cloney, R. A. 1990. Urochordata-Ascidiacea. Pp. 391-451 in Reproductive Biology of Invertebrates, Vol. 4, Part B, K. G. Adiyodi and R. G. Adiyodi, eds. Wiley, Chichester, UK.

Conklin, E. G. 1905. The organization and cell lineage of the ascidian egg. J. Acad. Nat. Sci. 13: 1-119.

Corbo, J. C., A. Di Gregorio, and M. Levine. 2001. The ascidian as a model organism in developmental and evolutionary biology. Cell 106: 535-538.

Cunliffe, V. T., and P. W. Ingham. 1999. Switching on the notochord. Genes Dev. 13: 1643-1646.

Dehal, P., Y. Satou, R. K. Campbell, J. Chapman, B. Degnan, et al. 2002. The draft genome of Ciona intestinalis: insights into chordate and vertebrate origins. Science 298: 2157-2167.

Delsuc, F., H. Brinkmann, D. Chourrout, and H. Philippe. 2006. Tunicates and not cephalochordates are the closest living relatives of vertebrates. Nature 439: 965-968.

Fenaux, R. 1998. Life history of the Appendicularia. Pp. 151-159 in The Biology of Pelagic Tunicates. Oxford University Press, New York.

Fenaux, R., Q. Bone, and D. Deibel. 1998. Appendicularian distribution and zoogeography. Pp. 251-2641 in The Biology of Pelagic Tunicates, Q. Bone, ed. Oxford University Press, New York.

Flood, P. R. 1975. Fine structure of the notochord of amphioxus. Symp. Zool. Soc. Lond. 36: 81-104.

Fol, H. 1872. Etudes sur les Appendiculaires du Detroit de Messine. Mem. Soc. Phys. H. 21: 445-499.

Galt, C. P., and R. Fenaux. 1990. Urochordata-Larvacea. Pp. 471-500 in Reproductive Biology of Invertebrates, Vol. 4, Part B, K. G. Adiyodi and R. G. Adiyodi, eds. Wiley, Chichester, UK.

Ganot, P., and E. M. Thompson. 2002. Patterning through differential endoreduplication in epithelial organogenesis of the chordate, Oikopleura dioica. Dev. Biol. 252: 59-71.

Godeaux, J., Q. Bone, and J.-C. Braconnot. 1998. Anatomy of Thaliacea. Pp. 1-24 in The Biology of Pelagic Tunicates, Q. Bone, ed. Oxford University Press, New York.

Godeaux, J. E. A. 1990. Urochordata -Thaliacea. Pp. 453-469 in Reproductive Biology of Invertebrates, Vol. 4, Part B, K. G. Adiyodi and R. G. Adiyodi, eds. Wiley, Chichester, UK.

Kardong, K. V. 1998. Vertebrates: Comparative Anatomy, Function, Evolution. 2nd ed. McGraw Hill, New York.

Katz, M. J. 1983. Comparative anatomy of the tunicate tadpole, Ciona intestinalis. Biol. Bull. 164: 1-27.

Koseki, H., J. Wallin, J. Wilting, Y. Mizutani, A. Kispert, C. Ebensperger, B. G. Herrmann, B. Christ, and R. Balling. 1993. A role for Pax-1 as a mediator of notochordal signals during dorsoventral specification of vertebrae. Development 119: 649-660.

Martini, E. 1909. Studien uber die Konstanz histologischer Elemente I Oikopleura longicauda. Z. Wiss. Zool. 92: 563-626.

Meinertzhagen, I. A., and Y. Okamura. 2001. The larval ascidian nervous system: the chordate brain from its small beginnings. Trends Neurosci. 7: 401-410.

Miyamoto, D. M., and R. J. Crowther. 1985. Formation of the notochord in living ascidian embryos. J. Embryol. Exp. Morphol. 86: 1-17.

Mookerjee, S., E. M. Deuchar, and C. H. Waddington. 1953. The morphogenesis of the notochord in amphibia. J. Embryol. Exp. Morphol. 1: 399-409.

Munro, E. M., and G. M. Odell. 2002a. Morphogenetic pattern formation during ascidian notochord formation is regulative and highly robust. Development 129: 1-12.

Munro, E. M., and G. M. Odell. 2002b. Polarized basolateral cell motility underlies invagination and convergent extension of the ascidian notochord. Development 129: 13-24.

Nishida, H. 1997. Cell lineage and timing of fate restriction, determination and gene expression in ascidian embryos. Semin. Cell Dev. Biol. 4: 359-365.

Nishino, A., Y. Satou, M. Morisawa, and N. Satoh. 2001. Brachyury (T) gene expression and notochord development in Oikopleura longicauda (Appendicularia, Urochordata). Dev. Genes. Evol. 211: 219-231.

Olsson, R. 1965. Comparative morphology and physiology of the Oikopleura notochord. Isr. J. Zool. 14: 213-220.

Ortolani, G. 1955. The presumptive territory of the mesoderm in the ascidian germ. Experientia 11: 445-446.

Ruppert, E. E., and R. D. Barnes. 1994. Invertebrate Zoology, 6th ed. Saunders College Publ., Fort Worth, TX.

Satoh, N. 1994. Developmental Biology of Ascidians. Cambridge University Press, New York.

Satoh, N. 2003. The ascidian tadpole larva: comparative molecular development and genomics. Nat. Rev. Genet. 4: 285-295.

Seeliger, O. 1893. Einige Beobachtungen uber die Bildung des ausseren Mantels der Tunicaten. Z. Wiss. Zool. 56: 488-505.

Soviknes, A. M., and J. C. Glover. 2007. Spatiotemporal patterns of neurogenesis in the appendicularian Oikopleura dioica. Dev. Biol. 311: 264-275

Soviknes, A. M., D. Chourrout, and J. C. Glover. 2005. Development of putative GABAergic neurons in the appendicularian urochordate Oikopleura dioica. J. Comp. Neurol. 490: 12-28.

Soviknes, A. M., D. Chourrout, and J. C. Glover. 2007. Development of the caudal nerve cord, motoneurons, and muscle innervation in the appendicularian urochordate Oikopleura dioica. J. Comp. Neurol. 503: 224-243.

Swalla, B. J., C. B. Cameron, L. S. Corley, and J. R. Garey. 2000. Urochordates are monophyletic within the deuterostomes. Syst. Biol. 49: 52-64.

Vinson, J. P., D. B. Jaffe, K. O'Neill, E. K. Karlsson, N. Stange-Thomann, et al. 2005. Assembly of polymorphic genomes: algorithms and application to Ciona savignyi. Genome Res. 15: 1127-1135.

Yamada, T., M. Placzek, H. Tanaka, J. Dodd, and T. M. Jessell. 1991. Control of cell pattern in the developing nervous system: polarizing activity of the floor plate and notochord. Cell 64: 635-647.

Yasuo, H., and N. Satoh. 1994. An ascidian homolog of the mouse Brachyury (T) gene is expressed exclusively in notochord cells at the fate restricted stage. Dev. Growth. Differ. 36: 9-18.

Yasuo, H., and N. Satoh. 1998. Conservation of the developmental role of Brachyury in notochord formation in a urochordate, the ascidian Halocynthia roretzi. Dev. Biol. 200: 158-170.

Wada, H. 1998. Evolutionary history of free-swimming and sessile lifestyles in urochordates as deduced from 18S rDNA molecular phylogeny. Mol. Biol. Evol. 15: 1189-1194


Sars International Centre for Marine Molecular Biology, University of Bergen, Bergen High Technology Centre, Thormohlensgt. 55, N-5008 Bergen, Norway

Received 18 April 2007; accepted 21 September 2007.

* To whom correspondence should be addressed. E-mail:
Table 1 Correlation of timing and staging of notochord morphogenesis in
Oikopleura and ascidian species

Oikopleura dioica,  Oikopleura longicauda,
time in hours       stage                   Ascidian stage

4                   1                       I
6                   3 (early)               II
7                     (late)                III
8                   4                       IV
COPYRIGHT 2008 University of Chicago Press
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 2008 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Author:Soviknes, Anne Mette; Glover, Joel C.
Publication:The Biological Bulletin
Article Type:Author abstract
Date:Feb 1, 2008
Previous Article:Effects of cyclic hypoxia on gene expression and reproduction in a grass shrimp, Palaemonetes pugio.
Next Article:Transitions in morphology, nematocyst distribution, fluid motions, and prey capture during development of the scyphomedusa Cyanea capillata.

Terms of use | Privacy policy | Copyright © 2019 Farlex, Inc. | Feedback | For webmasters