|
|
||||||||
1 Institute of Marine Studies, University of Plymouth, Drake Circus, Plymouth PL4 8AA, UK
2 School of Ocean Sciences, University of Wales - Bangor, Marine Science Laboratories, Menai Bridge, Anglesey LL59 5EY, UK
* Author for correspondence and current address: School of Biological Sciences, Victoria University of Wellington, PO Box 600, Wellington, New Zealand
| Abstract |
|---|
|
|
|---|
| Introduction |
|---|
|
|
|---|
Hosts may acquire zooxanthellae either by maternal inheritance or from the surrounding seawater. Maternal inheritance is probably the rarer mechanism in the tropics. For example, while some reef corals inherit their symbionts (Lewis, 1974; Kojis and Quinn, 1981; Richmond, 1981; Babcock et al., 1986; Glynn et al., 1991), the vast majority of coral species spawn gametes that lack zooxanthellae (Babcock et al., 1986).
In contrast to tropical symbioses, for temperate symbioses transmission modes have been identified in only a few cases. These include the soft coral Capnella gaboensis, which inherits zooxanthellae from the parent colony (Farrant, 1986); the scleractinian coral Astrangia danae, which spawns zooxanthella-free gametes (Szmant-Froelich et al., 1980); and a small number of sea anemones, the majority of whose ova contain algal symbionts (reviewed by Shick, 1991; Muller-Parker and Davy, 2001). Moreover, the cellular events leading to the acquisition of zooxanthellae and their eventual restriction to the hosts endodermal cells have been reported for tropical scleractinian corals (Hirose et al., 2000, 2001), soft corals (Benayahu et al., 1988, 1992; Benayahu and Schleyer, 1998), and jellyfish (Montgomery and Kremer, 1995), but not for temperate corals or sea anemones.
The sea anemone Anthopleura ballii (Cocks) is locally abundant along the southwestern coasts of Europe, where it is found from intertidal regions to depths of about 25 m (Manuel, 1988; Turner, 1988; Davy et al., 1996, 1997). The zooxanthellae harbored by A. ballii belong to the genus Symbiodinium (Davy et al., 1997), though they have yet to be subjected to molecular characterization. In this study, we documented cellular events from gametogenesis through to planula development in A. ballii, paying particular attention to the transmission and distribution of zooxanthellae within the hosts tissues.
| Materials and Methods |
|---|
|
|
|---|
Microscopical examination
Spawning of A. ballii, which is dioecious, was induced during summer. This was done by exposing anemones to air for between 3 and 5 h. The expelled gametes were collected by pipette and maintained in 100-ml sterile flasks containing artificial seawater at 15 °C. Fertilization occurred within hours and, every second day, the embryos were pipetted into new flasks, which also contained artificial seawater. This procedure ensured that the only possible source of zooxanthellae was the adult anemone.
Gametes, fertilization, and subsequent early development were examined by taking samples, first at hourly intervals and later once daily, for microscopical observation. A careful search for zooxanthellae was made, using interference contrast microscopy, by optical sectioning at each developmental stage. A Leitz Dialux 20 microscope with Vario-orthomat photographic system was employed, and a photographic record of early development was produced. In addition, cellular events occurring during gametogenesis were documented, again using interference contrast microscopy. This was made possible by anesthetizing anemones in equal parts artificial seawater and 7.5% MgCl2 · 6H2O for 12 to 24 h, and then teasing gametes out of the gonads.
| Results |
|---|
|
|
|---|
|
|
Fertilization
The gametes were shed into open water, where fertilization occurred. Each released ovum was surrounded by numerous sperm, which were aggregated between the cytospines. Fertilization usually occurred within 3.5 h of spawning, and unfertilized ova disintegrated after about 7 h, liberating their zooxanthellae. Spermatozoa were still active at this stage and became inactive after 2032 h.
Cleavage
About 3.5 h after spawning, 2-, 4-, 8-, and 16-cell blastulas were observed (4- to 16-cell stages shown in Fig. 1C-1E; 2-cell stage not shown). Cleavage was equal, radial, complete (i.e., holoblastic), and rapid, dividing the embryo into a ball of cells (blastomeres). Due to the initial localization of the zooxanthellae, symbionts were distributed unevenly in the blastomeres, being concentrated in those cells at only one end of the embryo. The zooxanthellae remained just inside the cell membrane of each blastomere. Blastulas of 32 cells were observed after about 5.5 h, and blastulas of 64 cells or more were apparent after 6 h. The blastomeres became ever smaller due to repeated cleavage, and after 8 h, a coeloblastula consisting of many cells and one cell layer was formed (Figs. 1F, 2B). The blastomeres were now 2030 µm in diameter, even in size, and rarely contained more than one zooxanthella each (Fig. 1G). The cytospines were resorbed and replaced by cilia, which soon began to exhibit the characteristic metachronal rhythm that rendered the coeloblastula motile.
Gastrulation
Few gastrulae (Figs. 1H-I, 2C-E) were seen, suggesting that this developmental stage is very short. Twenty hours after spawning, the motile coeloblastulas began to show a slight depression at the pole about which the algae were concentrated. Gastrulation by invagination (and perhaps epiboly) followed (Figs. 1H, 2C), with the blastomeres aggregated, at first, around the blastopore. Gastrulation led to the formation of an embryo with two cell layers encompassing a central cavitythe archenteron (Figs. 1I, 2D). During gastrulation, almost all of the blastomeres containing zooxanthellae moved into the endoderm from around the blastopore region. Only very occasionally were zooxanthellae seen in the ectoderm, and many of these cells appeared to disintegrate (Fig. 1J).
Planulation
After 27 h, most embryos had become late gastrulae or early planulae (Figs. 1K, 2E). By this stage, the developing larvae had shown no growth, remaining about 300 µm in diameter. However, after 2 days, most planulae began to elongate along their vertical axis, tapering slightly towards the posterior end. The zooxanthellae were clearly visible, aggregated in striations running the length of the endoderm. The surface of each larva was completely ciliated, and an apical tuft of longer cilia was visible. After 3 days, the larvae began to exhibit signs of differentiation (Figs. 1L, 2F), with the development of nematocysts, and a ciliated actinopharynx, which replaced the blastopore. Between 3 and 5 days, the planulae began to grow to about 400600 µm in length and 300 µm in diameter, even though they were not fed. The number of zooxanthellae also increased (not quantified), and dividing zooxanthellae were seen frequently. As the mesenteries developed, it became clear that most zooxanthellae were located along these structures. Interestingly, only one aposymbiotic planula was observed throughout the course of this work (Fig. 1M), which is consistent with the absence of aposymbiotic A. ballii at the field site (Lough Hyne). Tentacle rudiments were seen very occasionally in some planulae (Fig. 2G). Although care was taken to isolate the surviving planulae, they could not be kept alive for more than 7 days and so settlement was not observed.
| Discussion |
|---|
|
|
|---|
Gametogenesis and zooxanthella acquisition
In A. ballii, the endodermal tissue surrounding the developing oocytes was heavily laden with zooxanthellae, though infected oocytes were never observed. In contrast, spawned ova almost always harbored zooxanthellae, indicating that infection must occur at, or just prior to, release. We could not ascertain whether infection occurs in the gonadal tissue or after the ova have been released into the coelenteron. But, as the anemones were kept in artificial (and so zooxanthella-free) seawater in sterile flasks, and as spawning occurred in air, we can be certain that the zooxanthellae were of maternal origin, and that infection occurs prior to release into the surrounding seawater and hence prior to fertilization.
While the mode of zooxanthella acquisition has been determined in relatively few species of cnidarians, early indications are that infection prior to fertilization is quite uncommon. For example, the vast majority of scleractinian corals investigated do not harbor zooxanthellae in their eggs (Szmant-Froelich et al., 1980; Babcock et al., 1986; Harrison and Wallace, 1990), though some species of Pocillopora and Montipora do release zooxanthellate ova (Babcock et al., 1986; Harrison and Wallace, 1990; Glynn et al., 1991; Hirose et al., 2001). Of note, the eggs of the hard coral Montipora digitata become infected just 24 h prior to spawning (Harrison and Wallace, 1990), suggesting that the delayed infection seen in A. ballii eggs also occurs in some other hosts. Furthermore, in brooding species like the soft corals Xenia umbellata and Anthelia glauca, where zooxanthellae are transmitted maternally, infection does not occur until the later stages of embryogenesis or larval development (Benayahu et al., 1988; Benayahu and Schleyer, 1998).
The mechanism of entry into the ovum is unknown, but may well be similar to that described for the soft coral Litophyton arboreum (Benayahu et al., 1992). In L. arboreum, zooxanthellae pass through gaps in the mesogloeal covering of the oocytes, accumulate in the perioocytic zone, and ultimately bulge through the oolema and enter the mature oocyte. A similar "phagocytosis" of algal symbionts has been reported for the oocytes of several scleractinian corals (Hirose et al., 2001), as well as for the freshwater Hydra viridissima (Campbell, 1990).
Spawning, early development, and the localization of zooxanthellae
The sperm of A. ballii are similar to those of other Anthopleura spp. (Siebert, 1973). Moreover, as in other symbiotic Anthozoa, the heads are too small (3.5 x 2.3 µm) to act as vectors for paternal transmission of zooxanthellae; zooxanthellae in A. ballii are about 10 µm in diameter (Turner, 1988; Davy et al., 1996).
During the early stages of development, and throughout cleavage, the zooxanthellae remain localized at one end of the embryo. By the time a coeloblastula forms, most zooxanthellae are located in individual blastomeres, at one end of the coeloblastula. That this positioning is of paramount importance for the ultimate localization of the zooxanthellae becomes evident during gastrulation, when zooxanthellae are situated within invaginating blastomeres and so become localized within the endoderm. Indeed, the mechanism is so successful that "stray" zooxanthellae, which end up in the ectoderm, are rare (Fig. 1J).
The initial localization of zooxanthellae seen here is similar to that seen in the corals Pocillopora verrucosa and P. eydouxi (Hirose et al., 2000). However, as in some other coral species (Szmant-Froelich et al., 1980, 1985), gastrulation in P. verrucosa and P. eydouxi occurs via delamination rather than invagination. This means that, in marked contrast to events observed in A. ballii, blastomeres containing zooxanthellae move into the blastocoel of developing embryos, eventually filling the space and forming a stereogastrula (Hirose et al., 2000). The precise mechanism by which the zooxanthellae move into the blastocoel is unknown.
Planulation
As stated above, the position of zooxanthellae in the embryo, and the subsequent localization of zooxanthellae in the endodermis by invagination, means that "stray" zooxanthellae in the epidermal cells of planulae are very rare. A similar paucity of stray zooxanthellae was also reported for the reef corals Pocillopora verrucosa and P. eydouxi (Hirose et al., 2000). However, the planulae of some scleractinian corals (Szmant-Froelich et al., 1985; Schwarz et al., 1999), soft corals (Farrant, 1986; Benayahu et al., 1988, 1992; Benayahu and Schleyer, 1998), and jellyfish (Montgomery and Kremer, 1995) may contain zooxanthellae in their epidermal cells more frequently. In these cases, the zooxanthellae infect either the planulae or, as in the jellyfish Linuche unguiculata, both the embryos and planulae (Montgomery and Kremer, 1995), as opposed to the gametes. The zooxanthellae may then be transferred to the endodermal tissue via cell migration (Montgomery and Kremer, 1995) or trans-mesogloeal passages (Benayahu, 1997; Benayahu and Schleyer, 1998). Alternatively, stray zooxanthellae may degrade in the host or be expelled as a result of being harbored by an inappropriate cell type. Degrading zooxanthellae have been observed in planulae of the scleractinian corals Stylophora pistillata, Seriatopora caliendrum, and Pocillopora verrucosa, though always in the endodermis, rather than the epidermis (Titlyanov et al., 1998).
Mode of transmission as a function of latitude
Symbiotic invertebrates are abundant in tropical seas and regularly release zooxanthellae into the surrounding seawater (Hoegh-Guldberg et al., 1987); viable zooxanthellae are also released in the feces of numerous corallivorous predators (Muller-Parker, 1984). This may result in low selective pressure for the evolution of maternal inheritance in tropical regions, as zooxanthellae are readily available from exogenous sources to infect potential hosts (Buddemeier and Fautin, 1993; Kinzie et al., 2001).
In contrast, while transmission mechanisms have been investigated in relatively few species of zooxanthellate invertebrate, initial observations (including those presented here) suggest that maternal transmission of zooxanthellae is more likely to occur in temperate regions than in the tropics (reviewed by Muller-Parker and Davy, 2001). A predominance of maternal transmission mechanisms at high latitudes would not be surprising, as it could be related to a scarcity of exogenous sources of zooxanthellae and, therefore, selection against hosts that acquire their symbionts from exogenous supplies (Muller-Parker and Davy, 2001). Indeed, a scarcity of sources of zooxanthellae could explain why the temperate coral Astrangia danae, which does not acquire its zooxanthellae maternally, is sometimes found devoid of these symbionts (Szmant-Froelich et al., 1980). In addition, maternal transmission, combined with the ability of temperate algal-invertebrate symbioses to tolerate a wide range of environmental variables (Kevin and Hudson, 1979; Squire, 2000; Howe and Marshall, 2001), could explain the persistence of zooxanthellate organisms at high latitudes (Davy et al., 1997; Muller-Parker and Davy, 2001). More analyses of zooxanthellar transmission mechanisms at different latitudes, and of the ecological advantages conveyed by symbioses in nutrient-rich temperate waters, will help resolve this matter.
| Acknowledgments |
|---|
| Footnotes |
|---|
| Literature Cited |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
A. Scott and P. L. Harrison Embryonic and Larval Development of the Host Sea Anemones Entacmaea quadricolor and Heteractis crispa Biol. Bull., October 1, 2007; 213(2): 110 - 121. [Abstract] [Full Text] [PDF] |
||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |