The Osedax trophosome: organization and ultrastructure.
In most symbiotic associations the symbionts reside only on or in certain body areas, specialized structures, or organs, enabling both partners to manage functioning and control more efficiently. In the marine polychaete family Siboglinidae, a unique symbiont-housing organ termed the trophosome has evolved. Siboglinids completely lack mouth, gut, and anus, and for a long time it was unclear how the worms fulfill their nutritional needs. However, with the discovery of the bacterial symbionts of Riftia pachyptila Jones (1981), the importance of the trophosome became apparent (Cavanaugh et al, 1981). Only after this discovery were endosymbionts found in all other known siboglinids.
To date, Siboglinidae comprise the taxa Frenulata, Osedax, Monilifera (comprising Sclerolinum sensu Ivanov (1994)), and Vestimentifera. Recent molecular studies supported Sclerolinum as sister to Vestimentifera, Osedax as sister to Sclerolinum plus Vestimentifera, and Frenulata as sister to all three (Rouse et al, 2004; Glover et al, 2005). All lack mouth, gut, and anus, but live in obligatory association with bacteria. Although siboglinids live in habitats rich in reduced chemicals (Southward, 1993; Rouse el al, 2004; Vrijenhoek et al, 2009; Bright and Lallier, 2010), only the symbionts of frenulates, Sclerolinum, and vestimentiferans harbor chemolithoautotrophic Gammaproteo-bacteria fixing carbon dioxide. Transfer of fixed organic carbon to the host was demonstrated in the vestimentiferan Riftia pachyptila (Felbeck and Jarchow, 1998; Bright et al., 2000). In contrast Osedax's symbionts belong to the heterotrophic order Oceanospirillales (Gammaproteobacteria), and both host and endosymbionts rely on whale bones for nutrition (Goffredi et al., 2005, 2007).
In general, siboglinids are tube-dwelling annelids with a slender body organized into different functional parts. In frenulates and Sclerolinum those regions from anterior to posterior are the forepart, the trunk, and the segmented opisthosome. The vestimentiferan body can be divided into the obturacular region, the vestimental region, the trunk, and a segmented opisthosome. In all of them the trunk region is the longest section and houses the gonads and the trophosome. However, homology of the body regions across groups is debatable (Southward, 1980; Rouse, 2001; Rouse and Pleijel, 2001).
Although there are detailed descriptions of the morphology and origin of the vestimentiferan trophosome (Bosch and Grasse, 1984a, b; deBurgh, 1986; Southward, 1988; deBurgh et al., 1989; Bright and Sorgo, 2003; Nussbaumer et al., 2006), less is known about the trophosomes of frenulates and Sclerolinum (Ivanov, 1963; Southward et al., 1981; Southward, 1982; Callsen-Cencic and Flugel, 1995). In Osedax a trophosome was described as lacking (Rouse et al., 2004). In frenulates, Sclerolinum, and vestimentiferans the trophosome is situated in the trunk region. In vestimentiferans the trophosome takes up the entire length of the body cavity. In frenulates and Sclerolinum it occurs mainly in the posterior trunk region. Bacteriocytes make up the main part of the vestimentiferan trophosome. In Riftia pachyptila they are derived from visceral mesoderm and organized in numerous interconnecting lobules (Bosch and Grasse, 1984a, b; Bright and Sorgo, 2003; Nussbaumer et al., 2006). In frenulates and Sclerolinum the trophosome has been described as a simple two-layered organ with an inner epithelium of bacteriocytes derived from endoderm and an outer myoepithelium (Southward, 1993).
Osedax was first discovered on whale falls in Monterey Canyon, off the coast of California (Rouse et al., 2004) and differs from previously known siboglinids in several aspects. (1) In contrast to other siboglinids it shows a striking sexual dimorphism, with large females and paedomorphic males resembling larvae (Rouse et al., 2004, 2008; Worsaae and Rouse, 2009). (2) Exteriorly the adult female body can be subdivided into an anterior red crown, a contractile trunk, a bulbous ovisac, and branching roots at the posterior (Rouse et al., 2004). (3) Only females live in bones, with their trunks extending outward and their posterior roots excavating the bone. Males live in the transparent gelatinous tube that encloses the female trunk. (4) Initially, unlike other siboglinids, Osedax was described as lacking a distinct trophosome (Rouse et al., 2004). However, adult females of Osedax house symbionts in bacteriocytes surrounding the ovisac and extending into the extensively branching root structure (Rouse et al., 2004; Glover et al, 2005; Fujikura et al., 2006).
Regardless of whether or not the bacteriocytes are organized in a specialized organ, they are an important adaptation in all siboglinids because they are the interacting unit between symbionts and host. However, the existing data on different trophosomes is incoherent, fragmentary, and conflicting. It seems doubtful that the trophosome is a unique homologous structure within the family Siboglinidae, as both endodermal and mesodermal origins for this organ have been described. To date the origin of the bacteriocytes in Osedax has been unclear, making comparison with other siboglinids impossible. Here we provide the first detailed description of the organization and ultrastructure of the symbiont-housing area in Osedax. This allows us to answer the question of whether Osedax has a trophosome, and if so, how it is organized. Comparison of the symbiont-housing area of Osedax with trophosomes and body regions of other siboglinids is necessary to investigate whether the trophosome is a homologous structure within the Siboglinidae and how it could have evolved in the potential last common ancestor of the family, leading to the evolution of the diverse symbioses seen today.
Materials and Methods
All specimens used in this study were collected in Monterey Canyon, California. One colonization device, similar to the so-called bone tree described by Jones et al. (2008), was modified and deployed at the site of whale-1820 during ROV Tiburon dive T1163 on 16 August 2007 and recovered on 20 December 2007. Modifications included the sectioning of the cow bone femurs. Instead of cutting the bones in half longitudinally, bones were cut in transverse sections into several individual discs. Those discs were then screwed together again, and the pre-sectioned cow bones were fixed on the bone tree. This setup facilitates the collection of intact whole specimens when taking apart the individual disks. Upon recovery, the cow bones harbored dense colonizations of a yet-undescribed species of Osedax sp., referred to as "O. green palp" (Vrijenhoek et al, 2009), that were collected and fixed for the present study.
For transmission electron microscopy (TEM), one bone disc covered with Osedax specimens was fixed overnight with a mixture of 1.5% acrolein, 3% glutaraldehyde, and 1.5% paraformaldehyde in 0.1 mol [1.sup.-1] cacodylate buffer containing 10% (w/v) sucrose, rinsed with 0.1 mol [1.sup.-1] cacodylate buffer with 10% (w/v) sucrose three times for 10 min, postfixed in 2% [OsO.sub.4] in 0.1 mol [1.sup.-1] cacodylate buffer with 10% (w/v) sucrose for 2 h, rinsed again with the same buffer three times for 10 min, and dehydrated with a graded ethanol series up to 70%. All steps were carried out on ice. Specimens were stored in 70% ethanol at 4 [degrees]C until embedding. Several specimens were dissected out of the bone disc before fixation and then fixed for TEM as described above. For embedding, Osedax "green palp" specimens were dehydrated in an ethanol series up to 100% and after three exchanges were transferred to 100% propylene oxide and rinsed three times for 10 min. After infiltration with a 1:1 AGAR low-viscosity resin and propyleneoxide mixture for 1 h, followed by a 2:1 resin and propyleneoxide mixture overnight, specimens were transferred to pure AGAR low-viscosity resin for 6 h and then embedded in fresh resin and hardened at 60 [degrees]C for 16 h.
For immunohistochemistry and fluorescence in situ hybridization (FISH), additional specimens were fixed in 4% paraformaldehyde in 0.1 mol [1.sup.-1] Sorensen's buffer with 10% (w/v) sucrose at 4 [degrees]C overnight, then rinsed in 0.1 mol [1.sup.-1] Sorensen's buffer with 10% (w/v) sucrose three times for 10 min each, dehydrated in a graded ethanol series up to 70%, and stored until further treatment. Embedding and terminal deoxynucleotidyl transferase dUTP nick end labeling (TUNEL) assays, including positive and negative controls, were carried out as described by Pflugfelder et al. (2009) to check for apoptotic cell death. Nuclei showing a bright signal after the TUNEL assay are referred to as TUNEL-positive, and thus interpreted as undergoing apoptosis. FISH using oligonucleotid probes specific for bacteria (EUB338, Amann et al., 1990, Thermo Scientific) and Gammaproteobacteria (GAM42a, Manz et al., 1992, Thermo Scientific) was carried out as described by Nussbaumer et al. (2006).
For TEM, embedded specimens were sectioned serially, starting at the posterior end, using a Diatome diamond knife on a Reichert Ultracut S microtome. Several semithin (1-[micro]m) sections were cut, followed by at least 10-20 ultrathin (70-[micro]m)sections until the entire roots and ovisac regions were sectioned. Semithin sections were stained with aqueous 1.5% toluidine solution. Ultrathin sections were stained with uranyl acetate and lead citrate by hand and examined with a Zeiss EM 902 transmission electron microscope. One specimen was serially sectioned for light microscopy, stained with basic fuchsin and methylene blue, and examined with a Zeiss Axio Imager Al light microscope.
Three female specimens (T1163_28, T1163_43, and T1163_70) were measured to determine the proportions of different tissue types: ovary, bacteriocytes, body wall (equivalent to epidermis plus muscle layer plus nonsymbiotic cells of trophosome), and coelomic cavity. Their relative fractions were compared for two different regions discriminated earlier on the basis of ultrastruetural characterization into ovisac and roots. For each specimen, five semithin cross sections of each of those two regions were examined using a Zeiss Axio Imager Al microscope and photographed using an Olympus Color View III CCD-camera. Areas were measured using the AnalySIS program, ver. 3.2, and percentages of tissue types were calculated. The number of bacteriocytes per section per region was counted, and the mean area per bacteriocyte was calculated for each section. The width and the length of the trunk (from behind the crown to the start of the bulbous ovisac) were measured. Areas of the different tissue types, number of bacteriocytes, and mean area per bacteriocyte of the different root regions were compared statistically. Using SPSS, ver. 16, a Mann-Whitney U test (n = 30, P [less than or equal to] 0.01) was used to test for differences between the regions, and a univariate analysis of variance (n = 30, P [less than or equal to] 0.01) was calculated to test whether there were differences between the three specimens.
Segment: a morphological unit, repeated antero-posteriorly with the same organization and development (Purschke, 2002; Scholtz. 2002).
Septum or dissepiment: formed by apposing the posterior wall of one segment and the anterior wall of the next (Fransen, 1988; Westheide, 2007).
Diaphragm: any membrane--muscular, epithelial or myoepithelial--that does not separate bordering metameric segments (sensu Southward, 1980, 1993).
Bacteriocyte: host cell containing symbionts (Southward, 1982; Hand, 1987; deBurgh et al., 1989; Gardiner and Jones, 1993).
Nonsymbiotic cell: host cell devoid of symbionts (Bright and Sorgo. 2003).
Peritoneum: noncontractile epithelial lining of the coelom, overlying the splanchnic muscle of the gut and the somatic muscle of the body wall (Rieger and Lombardi, 1987; Fransen, 1988).
Trophosome: symbiont-housing organ in siboglinids (Cavanaugh et al, 1981), now commonly used to refer to any symbiont-housing organs in siboglinids or catenulids (U. Dirks (University of Vienna) et al., unpubl.).
All cow bones deployed for 4 months and recovered from whale-1820 in December 2007 showed dense colonization of Osedax sp. as previously described (Jones el al., 2008). The outer layer of the calcified bone was still forming one contiguous surface, but it possessed many holes through which trunks of Osedax specimens emerged. Root structures of individual specimens were separated from neighboring ones by thin bony walls, surrounding individual roots like rosettes. Thus, the shape of the excavations of the bones followed the pattern of the roots.
The characteristic body regions of Osedax--crown, trunk, bulbous ovisac, and branching roots--as described by Rouse et al. (2004) could also be distinguished exteriorly in the adult females used in this study. Overall, the roots region did not show a clear branching pattern, since the shape of every individual root structure looked different, except for two short processes from the ovisac region extending anteriorly on both sides of the trunk (Fig. 1). The roots region was light- to dark-green in live animals. This green tissue also covered the major peripheral part of the ovisac region. Common to all individuals was the lack of chaetae and the lack of segments within the roots region or following it.
[FIGURE 1 OMITTED]
Internally there was no septum or diaphragm separating the trunk from the ovisac region or the ovisac region from the roots region. In addition, the anteriormost region of the trunk containing the brain was not separated from the posterior trunk region.
The trunk lengths of the three specimens used for morphometric analyses measured 1.1 mm, 1.9 mm, and 3.3 mm, espectively. Even though the trunk length varied, the trunk width (average 379.21 [micro]m [+ or -] 5.62 [micro]m) was similar.
Using light and electron microscopy, we confirmed that symbionts were housed in bacteriocytes (Figs. 2, 3). Those were restricted to a specific tissue within the ovisac and roots regions. Nonsymbiotic cells were mixed in with the bacteriocytes, and together they formed an apolar tissue that lacked apical belt-shaped junctional complexes and was submerged in an inconspicuous extracellular matrix (Rieger and Lombardi, 1987). We will refer to this tissue of bacteriocytes and nonsymbiotic cells as the trophosome. It lies basally to the body wall musculature and apically to the peritoneum lining the body coelom. The trophosome corresponds to the light- to dark-green tissue in the ovisac and roots regions in live animals. However, ultrastructure of the trophosome and the epidermis differs along this longitudinal axis and allows the roots region to be distinguished from the ovisac region (see below). The epidermis in the different body regions is described in Katz et al (2010).
Cross sections showed that the roots region was composed of several tissues and cell types from apical to basal: (1) epidermis with epithelial cells, (2) the musculature with myoepithelial myocytes, (3) the trophosome with apolar bacteriocytes and nonsymbiotic cells, and (4) the non-muscular peritoneum with peritoneocytes lining the body coelom (Figs. 2A, 4). In the ovisac region this organization was kept, but the gonads located in the body cavity added another tissue type. The major dorsal and ventral blood vessels could no longer be identified within the root tissue, but several interspersed blood lacunae were found.
The outermost layer was provided by the single-layered epithelium, followed by a thin and inconspicuous muscle layer. A clear orientation of muscle fibers into longitudinal or circular muscle layers as in the trunk region could no longer be distinguished. The fibrous part was usually wedged between basal portions of epidermal cells. The rest of the muscle cells containing mitochondria usually meandered around the epidermal cells beneath. A prominent mix of large bacteriocytes and nonsymbiotic cells formed an apolar tissue, the trophosome, and was adjacent basally to the muscle layer. Finally, the trophosome was confined by a single-layered peritoneum of thin, flat electron-light cells with characteristic cell junctions toward the center of the worms, thus lining a coelom (Fig. 2A, F, G). Even though the basal matrices of the peritoneum confining the coelom and the concentric muscle layers are indiscernible, the apical cell junctions of the peritoneum are opposite in direction to those of the epidermis.
The comparison of the different tissue types between the two regions showed significant differences in relative contribution for ovary, body wall (epidermis plus muscle layer plus nonsymbiotic cells of trophosome), and coelomic cavity. The ovary was absent or less than 2% in the root region, but it ranged from 14.4% to 18.1% in the ovisac region. Both body wall and coelom took up significantly larger areas in the roots region than in the ovisac region. The fraction of bacteriocyte tissue, the number of bacteriocytes, and the average area per bacteriocyte were the same in both regions (Table 1).
Bacteriocytes were conspicuous large ovoid cells. Most looked transparent and lacked cytoplasm; however, random cells containing some electron-light cytoplasm were found (Fig. 3C, D). FISH using group-specific probes for Bacteria and Gamniaproteobacteria confirmed the occurrence of symbionts within bacteriocytes (Fig. 4). No cell organelles were observed except for a centrally located nucleus, which was round and slightly darker compared to the nuclei of other cells. Besides symbionts, bacteriocytes contained star- or rosette-like structures (Fig. 2B, C, E) of unknown nature. TUNEL-positive nuclei indicative of apoptosis were rarely detected in the roots region, whereas most of the nuclei of the bacteriocytes in the ovisac region stained positive for TUNEL in all tested specimens.
[FIGURE 4 OMITTED]
Nonsymbiotic cells (Figs. 2B, C, F, G; 3C) were characterized by a centrally located oval nucleus and a dense amount of rough endoplasmic reticulum, The latter almost completely filled the cells and gave them an electron-dense appearance. These cells also contained membrane-bound vesicles with electron-dark smooth to granular and fibrillar inclusions.
Table 1 provides the areas and numbers of bacteriocytes found in the cross sections of roots region and ovisac region for the three worms examined. However, those numbers do not reflect significant differences between the roots and the ovisac region regarding the fraction of bacteriocyte tissue (Mann-Whitney U test, P = 0.419) and number of bacteriocytes per section (Mann-Whitney U test, P = 0.740). Also, the mean area per bacteriocyte was similar in both the root and ovisac regions (Mann-Whitney U test, P = 0.078).
Table 1 Mean tissue fractions per region per individual Individual Region Body wall (%) Coelom (%) Ovary Bacteriocyte (%) area (%) T1163_28 roots 80.2 9.5 0.86 8.03 T1163_28 ovisac 48.9 0.6 15.35 8.9 T1163_43 roots 83.9 4.8 0.00 10.7 T1163_43 ovisac 51.1 2.0 18.10 10.3 T1163_70 roots 77.7 11.1 1.61 11.0 T1163_70 ovisac 42.1 1.1 14.42 14.7 Individual Region # of Bacteriocytes Area/bacteriocyte ([mu][m.sup.2]) T1163_28 roots 120.0 864.0 T1163_28 ovisac 102.6 907.5 T1163_43 roots 67.6 740.2 T1163_43 ovisac 219.8 624.2 T1163_70 roots 173.0 849.3 T1163_70 ovisac 118.2 672.8
Symbionts housed in bacteriocytes in the roots region looked different from those in the ovisac region (Fig. 2 vs. Fig. 3). The bacteria in the roots region were intact rods (Fig. 2B, C, F). Some of them seemed to be dividing (Fig. 2C). Some bacteria lay free in the bacteriocytes and others were housed in vacuoles, with sometimes five or more within one vacuole (Fig. 2B, C, D). In the ovisac region the symbionts were in the process of being degraded or were already completely degraded. Intact symbionts were only occasionally found within vacuoles; degrading symbionts were always enclosed in vacuoles (Fig. 3D). There was a gradual transition from the roots region with intact symbionts toward those in the process of degradation in the ovisac region (Figs. 2G, D; 3B; 5). At the onset of degradation, symbionts were surrounded by myelin bodies, but the outer and cytoplasmic membranes could still be recognized. In further degraded stages the bacterial cytoplasm became condensed and electron-dark. The symbionts possessed a contracted shape until only myelin bodies were left in the final stage (Fig. 3C, D).
[FIGURE 3 OMITTED]
[FIGURE 5 OMITTED]
We suggest that the term trophosome should be applied to the symbiont-housing tissue in Osedax because of the organizational and functional similarities of tissue that contains endosymbionts, regardless of the origin of the tissue. However, in distinct contrast to the trophosomes of all other siboglinids are the low abundance of endosymbionts and their heterotrophic nature in Osedax. Morphological comparison and molecular data on phylogenetic relationships within siboglinids point to two possible scenarios in the evolution of the trophosome: either the trophosome evolved once in the last common stem species of siboglinids together with reduction of the gut, or it evolved several times independently.
Osedax trophosome organization
The position of the trophosome, being sandwiched between the muscle layer of the body wall and a peritoneum lining the body coelom, clearly points to its origin from somatic mesoderm. Bacteriocytes and nonsymbiotic cells form one entity with the muscular mesodermal part of the body wall surrounding the coelom. In vestimentiferans, the bacteriocytes make up the main part of each trophosome lobule and are derived from visceral mesoderm (Bright and Sorgo, 2003; Nussbaumer et al., 2006). Each lobule is formed by an axial blood vessel, which is surrounded by myoepithelium containing bacteriocytes organized as an epithelium, followed by an apolar tissue of bacteriocytes and confined by a single-layered peritoneum also lining the body coelom (Bosch and Grasse, 1984a, b; Bright and Sorgo, 2003). In some frenulate species the bacteriocyte epithelium surrounds a fluid-filled lumen; in other species this lumen is restricted or even occluded. However, investigations of a frenulate of the genus Siboglinum from the Skagerrak, North Sea, at that time not described but most likely S. poseidoni, suggested that the bacteriocytes might proliferate from the muscle layer (Southward et al., 1981). Later developmental studies provided evidence for an endodermal origin of the trophosome in frenulates (Callsen-Cencic and Flugel, 1995). Without further investigation, an endodermal origin like that in Frenulata was assumed for Sclerolinum, but new evidence points to a mesodermal origin (I. Eichinger, University of Vienna, pers. comm.). Although high diversity exists in the muscle system of annelids and a high degree of modifications and reductions can be found just within siboglinids (Southward, 1975, 1993; Jones, 1981; Lanzavecchia et al., 1988; Gardiner and Jones, 1993; Southward et al., 2005; Tzetlin and Filippova, 2005), transformation of the somatic mesoderm with the integration of the trophosome in the body wall as in Osedax is a unique character.
The dimensions of the trophosomes in the different siboglinid taxa also reflect a high plasticity of this organ. The arrangement of the Osedax trophosome with its relative dimensions and the number and size of its bacteriocytes is similar in the ovisac and roots regions. Bacteriocytes account for only a minor fraction of all tissue types in Osedax, in contrast to the situation in vestimentiferans, where the trophosome fills most of the trunk (Gardiner and Jones, 1993; Bright and Sorgo, 2003). In Sclerolinum and frenulates the trophosome takes up the posterior trunk region, but only about 10% of that volume is bacteriocytes in the frenulate Siboglinum fiordicum (Ivanov, 1963; Southward, 1982, 1993).
On the basis of our morphological analysis and immunohistochemisty, we propose that the bacteriocytes and symbionts in the Osedax trophosome have a specific life cycle that is directed from the roots region toward the ovisac region. Growth of bacteriocytes with intact and dividing symbionts occurs at the tips of the roots. Bacteriocytes at the top of the ovisac region are the oldest, as evidenced by massive apoptosis and symbionts in the process of degradation. The vestimentiferan trophosome has a similar, complex cell cycle with terminal differentiation of bacteriocytes and endosymbionts. However, in Vestimentifera, bacteriocytes and symbionts change from the center of a lobule to its periphery (Bosch and Grasse, 1984a, b; deBurgh et al., 1989; Bright, 2002; Bright and Sorgo, 2003). Host cells in the central zone act as a population of "tissue-specific unipotent bacteriocyte stem cells" (Pflugfelder et al., 2009) housing intact and dividing rod-shaped symbionts. As host cells in the central zone divide, they push older bacteriocytes, which have symbionts transforming into cocci, toward the periphery. In the outermost zone, large coccoid symbionts are digested and host cells undergo programmed cell death via apoptosis. Thus, both the bacteriocyte tissue and the symbiont population are kept in balance (Pflugfelder et al., 2009). No zones or regions have been distinguished in the trophosome of frenulates. Digestion of symbionts appears to be random (Southward, 1982, 1993; Callsen-Cencic and Flugel, 1995).
Osedax symbionts are rod-shaped and reside freely in the cytoplasm. Once they are enclosed within vacuoles they are on the way to degradation. In contrast, vestimentiferan symbionts are encapsulated within a host-derived membrane-bound vacuole in the established trophosome (Cavanaugh. 1983, 1994; Bosch and Grasse, 1984a, b; Bright and Sorgo, 2003). However, during infection, rod-shaped symbionts lie freely in the cytoplasm and between the cells of skin tissue (Nussbaumer et al., 2006). In frenulates, rod-shaped endosymbionts are enclosed in vacuoles (Southward, 1993). Southward (1982) observed the endosymbionts of Sclerolinum lying freely in the cytoplasm, but new evidence shows that they are enclosed within vacuoles (I. Eichinger, University of Vienna, pers. comm.).
Inclusions of unknown nature and function are present in Osedax bacteriocyts as well as in adjacent nonsymbiotic cells. Star-like inclusions found in bacteriocytes resemble the concentric-layered granules within frenulate bacteriocytes and the small electron-dense membrane-bound granules in bacteriocytes of Sclerolinum. In the frenulate Siboglinum atlanticum, granules are composed of calcium and phosphorus (Southward, 1982). The fibrillar inclusions in nonsymbiotic cells of Osedax have not been described in any other siboglinid tissue.
Osedax symbiosis nutrition
We suggest that in the Osedax symbiosis, nutrition must be provided to the symbionts by the host, and not the other way round. Fujikura et al. (2006) already questioned whether the amount of symbionts could provide sufficient nutrients for the symbiosis. In frenulates, the symbionts range from a few to abundant (Southward, 1982, 1993; Callsen-Cencic and Flugel, 1995). In Vestimentifera, bacteriocytes are tightly packed with symbionts (Bosch and Grasse, 1984a, b; Bright and Sorgo, 2003), and nutrients are transferred to the host in two ways: leaking of organic carbon from symbionts and digestion of symbionts (Felbeck, 1981; Felbeck and Jarchow, 1998; Bright et al., 2000).
Osedax symbionts are heterotrophic and do not fix inorganic carbon (Goffredi et al., 2005, 2007), in contrast to chemoautotrophs in other siboglinids (Cavanaugh et al., 1981; Southward et al., 1981, 1986; Schmaljohann and Flugel, 1987; Losekann et al., 2008). Unless Osedax symbionts follow a different mode of nutrition than all other known Oceanospirillales, they need organic compounds for their own subsistence. However, they are housed in an internal tissue, isolated from the environment and presumed nutrient source by several host cell layers and cannot directly access organic bone material. Bone needs to be broken down and taken up by the host before symbionts can use it. Thus, the host has to feed its symbiont, but how this is accomplished and what are the specific nutritional interactions between the two partners still need to be investigated.
Trophosome evolution in Siboglinidae
The different origins of the siboglinid trophosomes reflect a high diversity of this key organ. When addressing the question of homology of the trophosome in Siboglinidae, an evaluation and comparison of body regions among all members of this taxon is inevitable. As in previous studies (Rouse et al., 2004; Glover et al., 2005), no segmented opisthosome or chaetae have been found in adult females. However, opisthosomal
chaetae were observed in Osedax dwarf males (Rouse et al., 2004; Worsaae and Rouse, 2009). The fate of the opisthosome in Osedax females cannot be determined at this point. However, from a functional point of view, an opisthosome with denticulate chaetae used as a digging organ, as in frenulates, or like an anchor, as in vestimentiferans and Sclerolinum (Gardiner and Jones, 1993; Southward, 1993, Southward et al., 2005; Stewart and Cavanaugh, 2006), is not needed in the whale-bone habitat and lifestyle of Osedax females. The ramification of the roots in the bone keeps the worms from being lifted off. As the roots penetrate the bone, they grow and spread further, and their ability to anchor the worms becomes even stronger.
It is unclear whether the trunk region in Osedax is homologous with the regions termed trunk in any other siboglinid. Only the trunk region in Osedax females contains the gonads, the trophosome, and the brain (main concentrations of axons and neurons), whereas the brain is located in the vestimentum in vestimentiferans (developing from the prostomium, peristomium, and anterior part of the first segment) (Nussbaumer et al., 2006), in the cephalic lobe in frenulates and Sclerolinum (Southward, 1993). Consistent with previous studies (Rouse et al., 2004), the roots region, the ovisac region, and the trunk are continuous--without any diaphragm, septum, or any other partition--and can be considered as functional subdivisions of one compartment. The trophosome is situated at the posterior end of this one compartment.
The trunk in Vestimentifera, Frenulata, and Sclerolinum is likely homologous. It develops from the posterior part of the first segment by elongation in vestimentiferans (Nussbaumer et al., 2006) and frenulates (Southward, 1980). A diaphragm (sensu Southward, 1980), which develops later than the first septum cutting off the opisthosome (Norrevang, 1970; Bakke, 1974, 1977; Ivanov, 1975), separates the forepart from the trunk in Frenulata. Neither a diaphragm nor a septum exists between the vestimentum and the trunk in Vestimentifera (Nussbaumer et al., 2006; Bright, pers. obs.). Information on the presence of a diaphragm between forepart and trunk in Sclerolinum is conflicting (Southward, 1961; Ivanov and Selivanova, 1992). In frenulates, chaetae, characteristic for segments, develop on the trunk, but not anterior of the diaphragm. In vestimentiferan larvae, chaetae are located on opisthosomal segments and the posterior of the trunk (Nussbaumer et al., 2006). On the trunk of adults, chaetae are absent according to most authors (Gardiner and Jones, 1993; Schulze, 2001; Southward et al., 2005) except Ivanov (1991).
Rouse (2008) suggested the possible homology of the adult female Osedax trunk with the vestimental region of Vestimentifera and the anterior end of the trunk of Frenulata because of the double-stranded ventral nerve cord running along the Osedax trunk. However, with the data available, it is not clear whether the Osedax trunk derives from one or several segments and how the prostomium and peristomium are involved in its formation. Similarly, development and origin of the trunk in Frenulata and Sclerolinum are still far from understood (Rouse, 2001; Schulze and Halanych, 2003).
Evolution of an alternative feeding source would be required prior to reduction of the digestive system. However, different scenarios may explain how this happened (Fig. 6). The trophosome is either a homologous structure or it developed several times independently. The latter would mean that the reduction of the gut also happened autonomously more than once. In that case, the trophosome would not serve as a morphological character unifying the family Siboglinidae, nor would the nonfunctional, occluded gut (Rouse, 2001) be a synapomorphic character of this taxon.
[FIGURE 6 OMITTED]
If the trophosome is a homologous structure, the last common ancestor would have established a symbiotic association within both germ layers--the endoderm and the mesoderm--and the gut reduction would have happened only once. This stage could have been similar to the infection process of early vestimentiferan larvae, when several germ layers initially become infected (Nussbaumer et al., 2006). Subsequently, in each taxon symbiosis in the respective part would have been reduced. Thus different evolutionary pathways led to the establishment of the trophosomes as diverse as we see them in recent siboglinid taxa today.
Specific developmental directions in accordance with the future symbiosis, especially the gut reduction and the elongation of the posterior part of the first segment into the trunk, might have already developed early in their last common ancestor. In Riftia pachyptila, the infection with putative endosymbionts occurs in settled larval stages when bacteria enter through the skin and migrate into mesodermal tissue, where they initiate the development of the trophosome (Nussbaumer et al., 2006). This process occurs during metamorphosis from the metatrochophore larva into the early juvenile, and it must therefore be tightly coordinated with the formation of other structures of the adult body plan. Certain pathways integrated and fixed during larval and early juvenile development might have restricted potential host areas for the establishment of endosymbiotic associations later. Consequently, developmental constraints might have determined the development of the Osedax symbiosis in the somatic mesoderm.
The life strategy of Osedax to exploit the bones of whales and other mammals (Goffredi et al., 2005; Jones et al., 2008) fits with the assumption that the siboglinids evolved from low-oxygen, sedimented habitats to decaying organic material to hydrocarbon seeps to hydrothermal vents (Schulze and Halanych, 2003). Halanych et al. (2001) suggested that this radiation occurred within the Scleralinum-vestimentiferan clade, but the discovery of Osedax shifts its occurrence earlier. We think this evolutionary pattern should be reflected within the evolution of siboglinid trophosomes. Unfortunately, only limited and sometimes conflicting morphological data for frenulate and Sclerolinum trophosomes exist and clearly need to be updated and clarified. The description of the Osedax trophosome presented here calls for a change in our view of trophosome functioning, development, and siboglinid evolution.
This research was supported by the Austrian Science Fund project # P20282-B17 (to M. B.), the David and Lucille Packard Foundation, and the Forsehungsstipendium of the University of Vienna (to S. K.), which we gratefully acknowledge. We thank Robert C. Vrijenhoek for inviting us on several expeditions; the captain and the crew of the RV Western Flyer; the past and current pilots of the ROVs Tiburon and Doc Ricketts for their skillful sample collections; and Joe Jones, Shannon Williams, and Julio Harvey for their assistance and fun during several whale fall expeditions. We also thank D. Gruber, U. Hormann, G. Spitzer (Department of Cell Imaging and Ultrastructure. University of Vienna), and Ch. Baranyi (Department of Microbial Ecology, University of Vienna).
Amann, R. I., B. J. Binder, R. J. Olson, S. W. Chisholm, R. Devereux, and D. A. Stahl. 1990. Combination of 16S 4RNA-targeted oligonucleotide probes with flow cytometry for analyzing mixed microbial populations. Appl. Environ. Microbiol. 56: 1919-1925.
Bakke, T. 1974. Settling of the larvae of Siboglinum fiordicum Webb (Pogonophora) in the laboratory. Sarsia 56: 57-70.
Bakke, T. 1977. Development of Siboglinum fiordicum Webb (Pogonophora) after metamorphosis. Sarsia 63: 65-73.
Bosch, C., and P.-P. Grasse. 1984a. Cycle partiel des bacteries chimioautotrophes symbiontiques et leurs rapports avec les bacteriocytes chez Riftia pachyptila Jones (Pogonophore Vestimentifere). I. Le trophosome et les bacteriocytes. C. R. Acad. Sci. 299: 371-376.
Bosch, C., and P.-P. Grasse. 1984b. Cycle partiel des bacteries chimioautotrophes symbiontiques et leurs rapports avec les bacteriocytes chez Riftia pachyptila Jones (Pogonophore Vestimentifere). II. L'evolution des bacteries symbiotiques et des bacteriocytes. C. R. Acad. Sci. 299: 413-419.
Bright, M. 2002. Life strategies of thiotrophic ectosymbioses. Pp. 19-32 in The Vienna School of Marine Biology: A Tribute to Jorg Ott, M. Bright, P. C. Dworschak, and M. Stachowitsch, eds. Facultas Universitatsverlag, Vienna.
Bright, M., and F. H. Lallier. 2010. The biology of vestimentiferan tubeworms. Oceanogr. Mar. Biol. Annu. Rev. 48: 211-264.
Bright, M., and A. Sorgo. 2003. Ultrastructural reinvestigation of the trophosome in adults of Riftia pachyptila (Annelida, Siboglinidae). Invertebr. Biol. 122: 347-368.
Bright, M., H. Keckeis, and C. R. Fisher. 2000. An autoradiographic examination of carbon fixation, transfer and utilization in the Riftia pachyptila symbiosis. Mar. Biol. 136: 621-632.
Callsen-Cencic, P., and H. J. Flugel. 1995. Larval development and the formation of the gut of Siboglinum poseidoni Flugel & Langhof (Pogonophora, Perviata). Evidence of protostomian affinity. Sarsia 80: 73-89.
Cavanaugh, C. M. 1983. Symbiotic chemoautotrophic bacteria in marine invertebrates from sulfide-rich habitats. Nature 302: 58-61.
Cavanaugh, C. M. 1994. Microbial symbiosis: patterns of diversity in the marine environment. Am. Zool. 34: 79-89.
Cavanaugh, C. M., S. L. Gardiner, M. L. Jones, H. W. Jannasch, and J. B. Waterbury. 1981. Prokaryotic cells in the hydrothermal vent tube worm Riftia pachyptila Jones: possible chemoautotrophic symbionts. Science 213: 340-342.
deBurgh, M. E. 1986. Evidence for a physiological gradient in the vestimentiferan trophosome: size-frequency distribution analysis of bacterial populations and trophosome chemistry. Can. J. Zool. 64: 1095-1103.
deBurgh, M. E., S. K. Juniper, and C. L. Singla. 1989. Bacterial symbiosis in Northeast Pacific Vestimentifera: a TEM study. Mar. Biol. 101: 97-105.
Felbeck, H. 1981. Chemoautotrophic potential of the hydrothermal vent tube worm, Rifiia pachyptila Jones (Vestimentifera). Science 213: 336-338.
Felbeck, H., and J. Jarchow. 1998. Carbon release from purified chemoautotrophic bacterial symbionts of the hydrothermal vent tube-worm Riftia pachyptila. Physiol. Biochem. Zool. 71: 294-302.
Fransen, M. E. 1988. Coelomic and vascular systems. Pp. 199-213 in The Ultrastructure of Polychaeta, W. Westheide and C. O. Hermans, eds. Gustav Fischer Verlag, Stuttgart.
Fujikura, K., Y. Fujiwara, and M. Kawato. 2006. A new species of Osedax (Annelida: Siboglinidae) associated with whale carcasses off Kyushu, Japan. Zool. Sci. 23: 733-740.
Gardiner, S. L., and M. L. Jones. 1993. Vestimentifera. Pp. 371-460 in Microscopic Anatomy of Invertebrates, Onychophora, Chilopoda and Lesser Protostomata, F. W. Harrison and M. E. Rice, eds. Wiley-Liss, New York.
Glover, A. G., K. Bjorn, C. R. Smith, and T. G. Dahlgren. 2005. World-wide whale worms? A new species of Osedax from the shallow north Atlantic. Proc. R. Soc. B Biol. Sci. 272: 2587-2592.
Goffredi, S. K., V. J. Orphan, G. W. Rouse, L. Jahnke, T. Embaye, K. Turk, R. Lee, and R. C. Vrijenhoek. 2005. Evolutionary innovation: a bone-eating marine symbiosis. Environ. Microbiol. 7: 1369-1378.
Goffredi, S. K., S. B. Johnson, and R. C. Vrijenhoek. 2007. Genetic diversity and potential function of microbial symbionts associated with newly discovered species of Osedax polychaete worms. Appel. Environ. Microbiol. 73: 2314-2323.
Halanych, K. M., R. A. Feldman, and R. C. Vrijenhoek. 2001. Molecular evidence that Sclerolinum brattstromi is closely related to vestimentiferans, not to frenulate pogonophorans (Siboglinidae, Annelida). Biol. Bull. 201: 65-75.
Hand, S. C. 1987. Trophosome ultrastructure and the characterization of isolated bacteriocytes from invertebrate-sulfur bacteria symbioses. Biol. Bull. 173: 260-276.
Ivanov, A. V. 1963. Pogonophora. Academic Press, London.
Ivanov, A. V. 1975. Embryonalentwicklung der Pogonophora und ihre systematische Stellung. Pp. 10-44 in 'The Phylogeny and Systematik Position of Pogonophora, A. Norrevang, ed. Paul Parey, Zoological Central Institute, University of Copenhagen.
Ivanov, A. V. 1991. Monilifera, a new subclass of Pogonophora. Dokl. Akad. Nauk. SSSR 312: 505-507.
Ivanov, A. V. 1994. On the systematic position of Vestimentifera. Zool. Jahrb. Abt. Syst. Okol. Geogr. Tiere 121: 409-456.
Ivanov, A. V., and R. V. Selivanova. 1992. Sclerolinum javanicum sp. n., a new pogonophoran living on rotten wood. A contribution to the classification of Pogonophora. Biol. Morya (Vladivost.) 1-2: 27-33.
Jones, M. L. 1981. Riftia pachyptila Jones: observations on the vestimentiferan worm from the Galapagos Rift. Science. 213: 333-336.
Jones, W. J., S. B. Johnson, G. W. Rouse, and R. C. Vrijenhoek. 2008. Marine worms (genus Osedax) colonize cow bones. Proc. R. Soc. B Biol. Sci. 275: 387-391.
Katz, S., W. Klepal, and M. Bright. 2010. The skin of Osedax (Siboglinidae, Annelida): an ultrastructural investigation of its epidermis. J. Morphol. 271: 1272-1280.
Lanzavecchia, G., M. de Eguileor, and R. Valvassori. 1988. Muscles. Pp. 71-88 in The Ultrastructure of Polychaeta, W. Westheide and C. O. Hermans, eds. Gustav Fischer Verlag, Stuttgart.
Losekann, T., A. Robador, H. Niemann, K. Knittel, A. Boetius, and N. Dubilier. 2008. Endosymbioses between bacteria and deep-sea siboglinid tubeworms from an Arctic Cold Seep (Haakon Mosby Mud Volcano, Barents Sea). Environ. Microbiol. 10: 3237-3254.
Manz, W., R. Amann, W. Ludwig, M. Wagner, and K.-H. Schleifer. 1992. Phylogenetic oligodeoxynucleotide probes for the major subclasses of Proteobacteria: problems and solutions. Syst. Appl. Microbiol. 15: 593-600.
Norrevang, A. 1970. On the embryology of Siboglinum and its implications for the systematic position of the Pogonophora. Sarsia 42: 7-16.
Nussbaumer, A. D., C. R. Fisher, and M. Bright. 2006. Horizontal endosymbiont transmission in hydrothermal vent tubeworms. Nature 441: 345-348.
Pflugfelder, B., S. Cary, and M. Bright. 2009. Dynamics of cell proliferation and apoptosis reflect different life strategies in hydrothermal vent and cold seep vestimentiferan tubeworms. Cell Tissue Res. 337: 149-165.
Purschke, G. 2002. On the ground pattern of Annelida. Org. Divers. Evol. 2: 181-196.
Rieger, R. M., and J. Lombardi. 1987. Ultrastructure of coelomic lining in echinoderm podia: significance for concepts in the evolution of muscle and peritoneal cells, Zoomorphology 107: 191-208.
Rouse, G. W. 2001. A cladistic analysis of Siboglinidae Caullery, 1914 (Polychaeta, Annelida): formerly the phyla Pogonophora and Vestimentifera. Zool. J. Linn. Soc. 132: 55-80.
Rouse, G. W., and F. Pleijel. 2001. Siboglinidae Caullery, 1914. Pp. 202-205 in Polychaetes. Oxford University Press, Oxford.
Rouse, G. W., S. K. Goffredi, and R. C. Vrijenhoek. 2004. Osedax: bone-eating marine worms with dwarf males. Science 305: 668-671.
Rouse, G. W., K. Worsaae, S. B. Johnson, W. J. Jones, and R. C. Vrijenhoek. 2008. Acquisition of dwarf male "harems" by recently settled females of Osedax roseus n. sp. (Siboglinidae; Annelida). Biol. Bull. 214: 67-82.
Schmaljohann, R., and H. J. Flugel. 1987. Methane-oxidizing bacteria in Pogonophora. Sarsia 72: 91-98.
Scholtz, G. 2002. The Articulata hypothesis--or what is a segment? Org. Divers. Evol. 2: 197-215.
Schulze, A. 2001. Ultrastructure of opisthosomal chaetae in Vestimentifera (Pogonophora, Obturata) and implications for phylogeny. Acta Zool. (Stockh.) 82: 127-135.
Schulze, A., and K. M. Halanych. 2003. Siboglinid evolution shaped by habitat preference and sulfide tolerance. Hydrobiologia 496: 199-205.
Southward, A. J., E. C. Southward, P. R. Dando, G. H. Rau, H. Felbeck, and H. Flugel. 1981. Bacterial symbionts and low (13)C/(12)C ratios in tissues of Pogonophora indicate unusual nutrition and metabolism. Nature 293: 616-619.
Southward, A. J., E. C. Southward, P. R. Dando, R. L. Barrett, and R. Ling. 1986. Chemoautotrophic function of bacterial symbionts in small pogonophora. J. Mar. Biol. Assoc. UK 66: 415-437.
Southward, E. C. 1961 Siboga-Expeditie Pogonophora. Siboga-Expeditie series, vol. 25: 1-22. E. J. Brill, Leiden.
Southward, E. C. 1975. Fine structure and phylogeny of the Pogonophora. Symp. Zool. Soc. Land. 36: 235-251.
Southward, E. C. 1980. Regionation and metamerisation in Pogonophora. Zool. Jahrb. Abt. Anat. Ontogenie Tiere 10: 264-275.
Southward, E. C. 1982. Bacterial symbionts in Pogonophora. J. Mar. Biol. Assoc. UK 62: 889-906.
Southward, E. C. 1988. Development of the gut and segmentation of newly settled stages of Ridgeia (Vestimentifera): implications for relationship between Vestimentifera and Pogonophora. J. Mar. Biol. Assoc. UK 68: 465-487.
Southward, E. C. 1993. Pogonophora. Pp. 327-369 in Microscopic Anatomy of Invertebrates, Onychophora, Chilopoda and Lesser Protostomata, F. W. Harrison and M. E. Rice, eds. Wiley-Liss, New York.
Southward, E. C, A. Schulze, and S. L. Gardiner. 2005. Pogonophora (Annelida): form and function. Hydrobiologia 535/536: 227-251.
Stewart, F. J., and C. M. Cavanaugh. 2006. Symbiosis of thiotrophic bacteria with Riftia pachyptila. Pp. 197-225 in Molecular Basis of Symbiosis, J. Overmann, ed. Springer, Berlin.
Tzetlin, A. B., and A. V. Filippova. 2005. Muscular system in polychaetes (Annelida). Hydrobiologia 535/536: 113-126.
Vrijenhoek, R. C., S. B. Johnson, and G. W. Rouse. 2009. A remarkable diversity of bone-eating worms (Osedax: Siboglinidae; Annelida). BMC Biol. 7: 74.
Westheide, W. 2007. Annelida, Ringelwurmer. Pp. 374-439 in Spezielle Zoologie Teil 1: Einzeller und wirbellose Tiere, W. Westheide and R. Rieger, eds. Spektrum Akademischer Verlag, Heidelberg.
Worsaae, K., and G. W. Rouse. 2009. The simplicity of males: dwarf males of four species of Osedax (Siboglinidae; Annelida) investigated by confocal laser scanning microscopy. J. Morphol. 271: 127-142.
SIGRID KATZ (1), *, WALTRAUD KLEPAL (2), AND MONIKA BRIGHT (1)
(1) Department of Marine Biology, Faculty of Life Sciences, University of Vienna, Austria; and (2) Institution of Cell Imaging and Ultrastructure Research, Faculty of Life Sciences, University of Vienna, Austria
* To whom correspondence should be addressed. E-mail: firstname.lastname@example.org
Received 28 December 2009; accepted 10 Febrauary 2011.
|Printer friendly Cite/link Email Feedback|
|Author:||Katz, Sigrid; Klepal, Waltraud; Bright, Monika|
|Publication:||The Biological Bulletin|
|Date:||Apr 1, 2011|
|Previous Article:||Phylogenetic characterization of episymbiotic bacteria hosted by a hydrothermal vent limpet (Lepetodrilidae, vetigastropoda).|
|Next Article:||Organization and microanatomy of the Sclerolinum contortum trophosome (Polychaeta, Siboglinidae).|