Printer Friendly

Acquisition of dwarf male "harems" by recently settled females of Osedax roseus n. sp. (Siboglinidae; Annelida).


Osedax, a genus of bone-eating siboglinid annelids, was first described from a whale carcass at 2893-m depth in Monterey Bay, California (Rouse et al., 2004). This initial description involved two new species, Osedax rubiplumus Rouse et al., 2004, and O. frankpressi Rouse et al., 2004. Soon after, two additional species were discovered--O. mucofloris Glover et al., 2005, from a 125-m whale-fall in Swedish waters, and O. japonicus Fujikura et al., 2006, from a 200-m whale--fall off Japan-suggesting that Osedax occurs worldwide at a range of depths. These unusual worms have no mouths or digestive tracts, but unlike other siboglinids (vestimentiferans and frenulates), Osedax does not host chemoautotrophic symbionts. Instead, it employs a complex ramifying root system to penetrate bones and obtain nutrition via heterotrophic bacteria that degrade organic compounds in the bone (Goffredi et al., 2005, 2007). Only female Osedax have these roots and the body form that characterizes this group. In contrast, the males are paedo-morphic dwarfs that appear to have arrested development at a metatrochophore larval stage (Rouse et al., 2004). Males of Osedax frankpressi and O. rubiplumus range from 0.3 to 1.0 mm in length, respectively, and they occur in "harems" that live within the lumen of a female's tube.

As part of our ongoing studies on whale-fall communities, a series of whale carcasses were sunk at various depths in Monterey Bay (Braby et al., 2007). The first whale in this series (5 October 2004) was a juvenile blue whale deposited at 1018 m (= Whale-1018). Two months after its deployment (10 Dec. 2004) the lateral processes of exposed vertebrae had been colonized by an unknown species of Osedax that was referred to as Osedax "MB3" or "rosy" (Braby et al., 2007; Goffredi et al., 2007). Here we formally describe this species as Osedax roseus n. sp. We also present data on female recruitment, density and growth on whalebones, and the acquisition by female Osedax of dwarf male harems during 6 months at Whale-1018. We discuss the results in relation to our earlier hypothesis (Rouse et al., 2004) that sex in Osedax may be environmentally determined. We also compare the occurrence of dwarf males in Osedax with other animals (Ghiselin, 1974; Vollrath, 1998), particularly with the annelid echiuran Bonellia, which is known to have environmentally determined sex (Agius, 1979).

Materials and Methods


Osedax specimens examined in this study were sampled from whale carcasses deposited at three depths in Monterey Bay. Whale-1018 was sunk on 5 October 2004 at 1018 m in Monterey Canyon (36.772[degrees]N, 122.083[degrees]W), and Whale-1820 was sunk on 20 March 2006 (36.708[degrees]N, 122.105[degrees]W). Details involving the deployment of these carcasses, the frequency of submersible visits, and the measurement of environmental variables were previously reported (Braby et al., 2007). Whale-633 was sunk on 11 April 2007 in Soquel Canyon (36.802[degrees]N and 121.994[degrees]W) and visited only once, on 8 August 2007. Most of the specimens of Osedax roseus n. sp. examined in this study were sampled from Whale-1018, which was visited several times during 8 months following its deployment (Table 1). The site occurs at the lower limit of the oxygen minimum zone in Monterey Bay--[O.sub.2] concentration = 0.448 [+ or -] 0.066 ml/l, practical salinity = 34.44 [+ or -] 0.07, and temperature = 3.99 [+ or -] 0.29[degrees]C--all of which are seasonally stable at this depth. The Whale-1820 site is deeper and colder (2.24 [degrees]C [+ or -] 0.019), and has a higher oxygen concentration (1.377 ml/l [+ or -] 0.014) and a salinity of 34.55 [+ or -] 0.012. Whale-633 site is shallower and warmer (5.0 [+ or -] 0.16[degrees]C), and has an intermediate oxygen concentration (0.405 [+ or -] 0.011 ml/l) and a salinity of 34.10 [+ or -] 0.67.

Transverse processes with an upward-facing surface were broken off from vertebrae of Whale-1820 on four occasions (Table 1) for the assessment of female density, growth, and acquisition of dwarf males. These bone surfaces were chosen as being the most appropriate and easy to collect and analyze over a time series. The destruction of all the transverse processes on Whale-1820 by tanner crabs (Chionoecetes tanneri) led to the premature cessation of sampling of bones for this study. Bones that included Osedax were preserved in 95% ethanol for molecular analyses or preserved for gross morphology in 4% formalin and seawater or, on one occasion (Sample B, Table 1), phosphate-buffered 3% glutaraldehyde. Specimens for light microscopy (LM), transmission and scanning electron microscopy (TEM and SEM) were preserved in 3% glutaraldehyde in either 0.1 mol [1.sup.-1] cacodylate or 0.1 mol [1.sup.-1] phosphate buffer with 0.3 mol [1.sup.-1] sucrose (pH 7.8). Histological specimens for LM were dehydrated and then embedded in paraffin. Sections 6 [micro]m thick were cut with a microtome and treated with Mallory's trichrome stain before examination through a Leica DMR compound microscope. Tissue samples for TEM were post-fixed for 80 min in 1% osmium tetroxide after several buffer rinses, dehydrated in a graded ethanol series, and embedded in Spurr's epoxy resin. Semi-thin (1 [micro]m) and ultrathin (80 nm) sections were cut with a Leica UltracutS ultramicrotome. Semithin sections were stained with toluidine blue and examined with a Leica DMR compound microscope. Ultrathin sections were stained with uranyl acetate and lead citrate and viewed through a Philips CM100 transmission electron microscope. Whole females were removed from their tubes for SEM, after post-fixation for 80 min in 1% osmium tetroxide. They were rinsed in filtered water and cleared of tube remnants by light sweeping with a paintbrush. The females were dehydrated in a graded ethanol series, sputter-coated with platinum, and examined with a Philips XL20 scanning electron microscope.

Specimens for confocal laser scanning microscopy (CLSM) were fixed for 2 h in 4% paraformaldehyde in 0.15 mol [1.sup.-1] phosphate-buffered saline (PBS) with 10% sucrose, pH 7.4, rinsed, and stored in PBS (with 0.05% Na[N.sub.3]). Transverse and sagittal slices of the trunks from three specimens were prepared with a scalpel. The sections were pre-incubated for 1 h in PSA (PBS with 0.1% Triton-X 100, 0.05% Na[N.sub.3], 0.25% BSA, 0.6% horse serum, and 10% sucrose) and subsequently incubated for 12 h at room temperature in the two primary antibodies mixed 1:1 (monoclonal mouse anti-acetylated [alpha]-tubulin and polyclonal rabbit anti-serotonin, 5-HT). Subsequently, the specimens were rinsed and incubated for 12 h with two secondary antibodies mixed 1:1 (anti-mouse and anti-rabbit, conjugated with fluorophores CY5 and TRITC, respectively). Thereafter the animals were rinsed and incubated for 40 min in FITC-labeled phalloidin solution (0.17 [micro]mol [1.sup.-1] phalloidin in PBS). Finally, the animals were rinsed in PBS (without Na[N.sub.3]) and mounted between two cover slips in Vectashield (Vector Laboratories, Burlingame, CA) containing DAPI to counterstain DNA by producing blue fluorescence. Antibody binding specificity was tested by omitting primary antibodies, but otherwise treating specimens as described above. Preparations were investigated with a Leica TCS SP5 confocal laser scanning microscope at Adelaide Microscopy, University of Adelaide.

Type specimens are deposited at the Los Angeles County Museum (LACM) and the Scripps Institution of Oceanography, Benthic Invertebrates Collection (SIO-BIC). All novel DNA sequences obtained in this study were deposited in GenBank (ace. nos. DQ996625-DQ996628, EU032469-EU032484, EU164760-EU164773).

Assessment of female density and growth and counts of males

After soaking a transverse process in seawater to remove the fixative, the uppermost surface was determined by the presence of more sediment and a greater density of Osedax females. Females were viewed through a Leica MZ8 stereo-microscope and counted, and most of them were dissected from the surrounding bone with their tubes and ovisacs intact, though the filamentous extensions of the roots could not be extracted. Individual females were removed from their gelatinous tubes, and crown length, trunk length, and width of the trunk posterior to its junction with the crown were measured. The reproductive condition of females was assessed by looking for eggs in the oviduct, which is clearly visible running along the trunk. The transparent gelatinous tubes were examined for the presence of males that might be attached by their posterior hooks to the inner surface of the tube in proximity to the female. For ease of locating and counting the translucent males, females were generally removed from their tubes by pulling them out from the posterior end. The males are generally retained within the tube, though the trunk and crown of each female was also checked. The surface area of the bone fragments was then determined from photographs, using ImageJ (Rasband, 2006).

Statistical analyses of female size distributions were conducted with JMP statistical software (ver. 7.0.1, SAS Institute, 2000). Homogeneity of variances was assessed with Levene's test (Levene, 1960), and normality was assessed with the Shapiro-Wilks test (Shapiro and Wilk, 1965). Because variances were heterogeneous across samples and normality was violated in one sample, we conducted a nonparametric analysis of sample means with the Kruskal-Wallis test (Kruskal and Wallis, 1952). Analyses of the male frequency distributions were conducted using the POISSON function in Excel (ver. 11.3.5, Microsoft Corp.), according to the equation: f(k;[lambda]) = [e.sup.-[lambda]][[lambda].sup.k]/k!, where f is the expected frequency of k males, and A is the empirical mean number of males. Degrees of freedom (df) for [chi square] tests were adjusted to account for pooling of categories with expected values < 5.

Molecular methods and estimates of diversity

Molecular methods for obtaining mitochondrial COI (mt-COI) sequences were detailed in previous publications (Rouse et al., 2004; Goffredi et al., 2004). Vestimentiferan-specific primers COIF and COIR (Nelson and Fisher, 2000) were used to amplify a 1200-base pair (bp) fragment of mtCOI. However, to be consistent with previous studies of mtCOI diversity in Osedax and other deep-sea polychaetes (Rouse et al., 2004), all estimates obtained for this study were based on a 540-bp region from the 5'-end of this gene. We used DnaSP ver. 4.10.9 (Rozas et al., 2003) to estimate the following diversity parameters: S the number of segregating (polymorphic) sites; H the number of haplotypes; h haplotype diversity (equation 8.4 in Nei, 1987), [pi] nucleotide diversity (equation 10.5 in Nei, 1987); and [theta] per site from S (equation 10.3 in Nei, 1987).


Female density, growth, and fecundity

The densities of female specimens of Osedax roseus n. sp. did not change across the first three samples, A-C (1.59 to 1.63/[cm.sup.2]) and they were lowest in sample D at 1.06/[cm.sup.2] (Table 1). Sizes of the females (length of crown + trunk) varied significantly among the four samples (Fig. 1A-D, left panels), but showed no consistent increase in mean size over time, with sample C having the largest average size of females. Variances of the four size-frequency distributions were not homogeneous (Levene's F = 17.23, df = 3, P < 0.001), and sizes were not normally distributed in samples B and D (Shapiro-Wilks W). All these samples were platykurtic (negative kurtosis values), but the direction of skewness varied among samples. A nonparametric one-way ANOVA identified significant differences among the four sample means (Kruskal-Wallis [chi square] = 50.12, df = 3, P < 0.001). The females in sample B were smaller and had narrower variance than the other samples, but this difference may be a consequence of fixation methods, as sample B was preserved in glutaraldehyde rather than the formalin used in samples A, C, and D (Table 1). Also, the four bone samples might have been exposed to colonization by Osedax at different times.


Evidence for more than one round of recruitment is evident in sample D, which exhibited significant platykurtosis and the greatest size variance. The increased variance probably resulted from the subsequent growth of smaller (presumably younger) females, many of which did not have eggs (Fig. 1D). Nonetheless, the percentage of females with eggs (black bars) increased progressively with time (28.3%, 30.1%, 46.6%, 71.8%) from sample A to D (Fig. 1).

Male harems

All the females previously examined for size and fecundity were also examined for the presence of dwarf males in their tubes (Fig. 1A-D, right panels). Of 319 females, 46.7% lacked males, 24.1% had only one male, and 29.2% had two or more males. The maximum number of males in a harem was 14, found in the tubes of two females from the last sample, D. Only sample D had harem sizes with six or more males (n of harems [greater than or equal to] 6 = 23). The mean sizes of male harems differed significantly among the four samples (Kruskal-Wallis [chi square] = 50.12, df = 3, P < 0.001), increasing progressively with time from less than 1 in samples A-C (January, February, and April), to more than 3 by sample D in July (A = 0.333, B = 0.614, C = 0.863, D = 3.504). Pairwise comparisons (Tukey-Kramer HSD, for unequal sample sizes) showed that the females in sample D were significantly larger than in the preceding three samples, which were not significantly different from each other. Frequency distributions of male harem sizes were Poisson distributed for samples A, B, and C; the means and variances of each sample were approximately equal. In contrast, the male harem size distribution was distinctly non-Poisson for sample D; variance [approximately equal to] 3 X mean. Sample D was bimodal, with one mode on females with no males and a second mode on females with 4 males. Thus, the zero class in sample D results from younger females with no eggs. This hypothesis is supported by the significant positive relationship ([R.sup.2] = 0.426, F = 74.9, df = 1, P < 0.001) between harem size and female size (Fig. 2). Smaller and presumably younger females host fewer males.

Systematic description

Phylum Annelida Lamarck, 1809

Order Canalipalpata Rouse & Fauchald, 1997

Family Siboglinidae Caullery, 1914

Osedax roseus new species

Figures 3-7

Previously referred to as Osedax Monterey Bay sp. 3 "rosy," Osedax sp. MB3 ("rosy") and "rosy" (Goffredi et al., 2007); Osedax "rosy" (Braby et al, 2007).

Type material. Monterey Bay, California (36.772[degrees]N, 122.083[degrees]W; 1018-m depth), ROV Ventana dive number V2689 (07/18/2005): holotype, mature adult female (SIO-BIC A979); allotypes, three males from tube of holotype (SIO-BIC A980); paratypes, 10 females and ~100 males (SIO-BIC A981); 10 females and 17 males (LACM POLY 2189).

Other material examined. Several hundred specimens. Monterey Bay, California: 36.772[degrees], 122.083[degrees], 1018-m depth; ROV Ventana dives V2621 (SIO-BIC A986-987), V2643 (SIO-BIC A988-A992), V2689 (SIO-BIC A983); Monterey Bay, California: 36.802[degrees]N, 121.994[degrees]W, 633-m depth; Ventana dive V3061 (SIO-BIC A984-A985); Monterey Bay, California: 36.708[degrees]N 122.105[degrees]W, 1820-m depth; Tiburon dive T1048.

Etymology. From Latin roseus for rosy or pink color.

Description. Holotype fixed in situ in vertebral transverse process (Fig. 3A) and subsequently dissected from bone matrix (Fig. 4B). Trunk and crown, emergent from bone, encased in cylindrical transparent tube with 0.5-mm-thick walls. Contracted crown, comprising four palps and an oviduct, 10.5 mm long. Palps bright red and up to 60 mm in length in life (Fig. 3B), inner margin has pair of white longitudinal stripes (Fig. 3C). Outer palp margin with pinnules starting 1 mm from base (Fig. 5C, E). Each palp with pair of blood vessels (Fig. 6C), with pinnules also vascularized (Fig. 6A, B) Oviduct extends distally 3 mm from trunk (Fig. 5C) between ventral pair of palps (Figs. 3D; 4C; 5A, E; 6A, C), and then runs visibly along trunk (Figs. 3D; 4A, C, D; 5A, E, H; 6E, F) before disappearing at trunk base. Oviduct filled with ellipsoid eggs (mean diameters: 132 X 89 [micro]m; n = 10) with obvious nucleus (Fig. 4C, D). Dorsal collar a semicircular flap (Figs. 3D, E; 4B, C; 5B, D, F), greenish in life (Fig. 3E). Holotype trunk 7.1 mm long with thickened base 1.8 mm long (Fig. 4B); 0.7 mm wide anteriorly, 0.9 mm wide at base, and 1.1 mm wide at junction with ovary (Fig. 4B). No mouth, anus, or obvious intestine present. Trunk encloses two major longitudinal blood vessels (Figs. 3D; 4A; 6E, F), otherwise heavily muscularized with thin outer circular muscle layer beneath epidermis (Fig. 6E, F), thick layer of longitudinal muscle surrounding central blood vessels (Fig. 6D-F). Secretory glands present among inner longitudinal muscle tissue, presumably responsible for gelatinous tube secretion (Fig. 6D, E). Six discrete smaller longitudinal bundles, 30-50 [micro]m wide, apparently muscle, lie just below circular muscle band, distributed around trunk (Fig. 6E, F); running length of trunk. Trunk surface smooth (Figs. 3D; 4A, B), generally lacking cilia, shows ridges on heavily contracted specimens (Fig. 5C, D, F), as does oviduct (Fig. 5E). Two broad ciliated patches, laterally, on either side of collar (Fig. 5D), narrow and extend along trunk as lateral ciliary bands (Figs. 3E; 4D; 5E-I). Nerve system visualized through CLSM of anti-serotonin immunoreactivity shows ventral nerve cord, apparently intra-epidermal, separated into two parts along at least part of trunk, lying between oviduct and circular muscle layer (Fig. 6F). Two other areas of intra-epidermal nerve activity in trunk opposite to oviduct (Fig. 6F). Lower immunoreactivity levels suggest these are not the ventral nerve cord. Fine lateral and denser dorsal peripheral mesh of epidermal nerves run along trunk outside circular muscles. Fine mesh of nerves extending towards center of the trunk from inside circular muscle layer, most distinct at connections to lateral ciliary bands and six discrete longitudinal bundles (Fig. 6F). Mesenteries or septa not apparent in trunk.


Remainder of holotype, comprising ovisac and "roots" (Fig. 4B), dissected from whalebone. Ovisac irregular in shape, except for pair of lateral projections that extend parallel to trunk (Figs. 3D; 4A, B; 5B, C, I; 6D). Ovisac 3 mm by 2 mm by 1 mm. Ovary full of oocytes at various stages of development (Fig. 4E, F). Central oviduct leads to oviduct running along trunk (Figs. 3D, E; 4A, E). Major blood vessels within ovisac (Figs. 3D; 4A, E). Lateral anterior projections, incomplete in dissected holotype, also contain oocytes (Figs. 4E; 6D). One lateral anterior projection branches out to form bulbous branching roots as a sheet 17.5 mm long (Fig. 4B). Holotype with three other major roots (Fig. 4B) continuous with ovisac extending with numerous knob-like projections through whalebone; one 8-mm sheet directly posterior to ovisac, broken; second, a 3-mm posterior lobe; and third, a 10.5-mm mass. Roots of specimens dissected alive from bone and then fixed show more lobate organization (Figs. 3D; 4A; 5B, C). Roots contain large blood vessels, bacteriocyte layer, and epidermis. Epidermis of roots superficially smooth (Fig. 5C, G) but with numerous microvilli (Fig. 4H). Bacteriocyte layer lies beneath epidermis, and bacteriocytes house large numbers (Fig. 4G) of bacteria 4-5 [micro]m long (Fig. 41).

Allotypes, three dwarf males, present in tube of holotype female, 130 [micro]m to 210 [micro]m long (Fig. 7A, D). Males found only in lumen of tube, close to female body, but not attached to it (Figs. 6G; 7D); instead they anchor to inner margin of the tube. Anteriorly, males with a ciliary band, putatively a retained larval prototroch (Fig. 7B, C, E). Body of males with thin longitudinal muscles (Fig. 7C) and otherwise filled with sperm, spermatids (Figs. 6G; 7B, C, E). Early spermatids anterior, just behind prototroch; more mature spermatids, forming large cluster, posteriorly (Fig. 7B, E). Anterior sperm duct carries mature sperm (filiform with spiral nucleus) anterior to putative prototroch (Fig. 7C). Some males with yolk apparent internally (Figs. 6G; 7E). Posteriorly, 16 hooks with capitium teeth emergent; handles 15-20 [micro]m (Fig. 7B). Capitium with six to eight teeth (Fig. 7B, C, E).




Two paratype females with associated males were also deposited at LACM. One of these has seven males (average length 171 [micro]m) in the tube and has a trunk length of 6.3 mm, a crown length of 5.5 mm, with the oviduct extending 3.3 mm into the crown. This paratype has four distinct roots ranging from 4 to 7 mm long. The second paratype has two males (190 [micro]m and 195 [micro]m length, respectively) and has a trunk length of 4.75 mm, a crown length 10.75 mm, with the oviduct extending 3.1 mm into the crown. This paratype has six distinct roots ranging from 3 to 8.3 mm long. In addition to these paratypes, we examined more than 300 females and tubes attributable to Osedax roseus n. sp. for this study. Female size distributions and number of associated males are treated in greater detail above. Female sizes (contracted trunk + crown length) range from 1.6 mm to 24.5 mm long (Figs. 1, 2).

Females of Osedax roseus n. sp. are distinguished from the four other described Osedax species by the presence of the semicircular lobe-shaped collar dorsally at the anterior end of the trunk (Figs. 3E; 5F). Osedax roseus is most similar to Osedax rubiplumus in the organization of the palps and pinnules and root structure. The palps of both species have the pinnules oriented outward, while those of O. frankpressi and O. mucofloris are oriented inward (Glover et al., 2005; Rouse et al., 2004). Osedax japonicus is the only other currently known Osedax species to have the palp pinnules all oriented outward (Fujikura et al., 2006), but it is smaller than O. roseus, has a very short oviduct extending from the trunk, and has pale-colored palps. The roots of O. rubiplumus and O. roseus n. sp. preserved in situ appear quite similar (G. Rouse, pers. obs.), being branching with distinct nodules, and differ from those described for the other three Osedax species. It seems, however, that roots preserved in situ in the bone appear different from those that were dissected out while the worm was still alive and then preserved (contrast Fig. 4A with 4B). Osedax roseus n. sp. differs markedly from O. rubiplumus in having a much smaller adult size for females, males, and spawned eggs. Also, the two interior white stripes on each palp are unique to O. roseus n. sp. Osedax frankpressi has striped palps, but these are visible on the outer surface of the palp (Rouse et al., 2004), though this striping may in fact be homologous, and only the pinnule arrangement differs between the two species. The newly spawned eggs of O. roseus n. sp. (132 X 89 [micro]m) are markedly smaller than those of deeper-dwelling Osedax species (O. rubiplumus and O. frankpressi; 151 X 121 [micro]m and 146 X 117 [micro]m, respectively; Rouse et al., 2004) and are similar in size to, though still larger than, those of O. mucofloris (85-90 [micro]m) and O. japonicus (100 [micro]m) (Glover et al., 2005; Fujikura et al., 2006).

A CLSM analysis of the trunk (Fig. 6F) of O. roseus n. sp. was undertaken to locate the nerve cord and thus establish the dorsal-ventral axis for Osedax. It is known that in vestimentiferans the nerve cord is split in two along the vestimentum and unites posteriorly in the trunk (van der Land and Norrevang, 1977), wherease in frenulates the nerve cord is a diffuse plexus in some areas and a single cord in others (Ivanov, 1963; Southward, 1993). In all siboglinids studied to date, the nerve cord is intra-epidermal, and this appears to be the case in Osedax. The stronger anti-serotonin activity leads us to suggest that the ventral nerve cord lies in the epidermis on either side of the oviduct (Fig. 6F). The only other intra-epidermal activity was two separate areas (ien) opposite the oviduct (Fig. 6F), but these were less distinct and asymmetric, and we discount this as the ventral nerve cord pending further investigation. In either case, the fact that the nerve cord of Osedax is split means that the female trunk may be homologous to the vestimental region of Vestimentifera and the anterior end of the trunk of Frenulata. The six unusual longitudinal muscle bundles also show a higher degree of innervation than other muscle areas in the trunk. This suggests that these muscles may be responsible for a rapid retraction response by the female. The location of the oviduct along the ventral part of the trunk is somewhat surprising since the equivalent (though paired) ducts in frenulates and vestimentiferans open in a dorsal position (Ivanov, 1961; Webb, 1977). However, it must be noted that frenulate and vestimentiferan oviducts travel ventrally inside the trunk before opening dorsally.

The dwarf males of O. roseus n. sp. are on average about 180 [micro]m long--much smaller than males of O. rubiplumus (400 [micro]m-1.1 mm long)--and are similar in size to those of O. frankpressi (150-250 [micro]m). The overall anatomy of the males is essentially similar (unpubl. data). Males attach by their hooks to the lumen surface of the female's tube, and not directly to the female body. Thus, when the female retracts/contracts her trunk, the males stay in the same place.

Molecular systematics and demographic genetics

We examined a 540-bp sequence from the 5'-end of mitochondrial COI (mtCOI) in a sample of 51 Osedax roseus n. sp. females sampled from three whale-falls (Whale-1820, n = 2; Whale-1018, n = 34; and Whale-634, n = 15). Thirty unique haplotypes were observed (GenBank acc. nos. DQ996625-DQ996628, EU032469-EU032484, EU164760-EU164773). We also examined a short segment (226 bp) of this gene from a single O. roseus male found in the tube of one female (GenBank #EU164774). Pairwise sequence divergence between O. roseus n. sp. and all other named as well as presently undescribed Osedax species for which COI sequences are reported was minimally 16.6% (Braby et al., 2007). This value greatly exceeds the mean sequence divergence within these genetically and morphologically distinct evolutionary lineages. We do not present a phylogenetic tree because the majority of internal nodes are not well supported when based on COI sequences alone. A multi-locus phylogenetic analysis of Osedax species, involving mitochondrial COI and 16S rRNA and nuclear 18S and 28S rRNA, is presently underway, but a preliminary analysis of these data and morphological evidence suggest Osedax roseus n. sp. is most closely related to O. rubiplumus.


Altogether, 35 mutations were identified at 33 segregating sites (S) along a 540-bp region from the 5'-end of mtCOI. Thirty haplotypes were observed, and haplotypic diversity (h) was 0.922. On average, 4.202 (k) nucleotide substitutions existed between pairs of haplotypes, and the average number of differences per site ([pi]) was 0.00778. The normalized nucleotide diversity per site ([theta]) based on the number of segregating sites (S) was 0.01358. This estimate of [theta] was used to infer the effective number of breeding females ([N.sub.e(f)]) contributing to this Osedax roseus n. sp. population, according to the relationship [theta] = 2[[N.sub.e(f)][mu]] for a mitochondrial gene. Assuming that [mu] [approximately equal to] 4 X [10.sup.-9] nucleotide mutations per site per generation (for justification, see Rouse et al., 2004), [N.sub.e(f)] was estimated to be about 1.7 X [10.sup.6]--somewhat higher than that reported for Osedax frankpressi and 0. rubiplumus in Rouse et al. (2004).


Osedax roseus n. sp. constitutes the fifth named species of Osedax since initial description of this genus in 2004. Our studies of Osedax from whale carcasses sunk at various depths in Monterey Bay have identified additional species (Braby et al., 2007) that remain to be formally described (unpubl. data). The time-series analysis of female size distributions and male harem sizes in Osedax roseus n. sp. add considerably to our burgeoning knowledge about this unusual group of polychaete annelids. The anatomical studies included in this description of O. roseus n. sp. also clearly identified the ventral nerve cord in females, thereby resolving the question of dorsal-ventral orientation; and they characterized "fast" muscles that are presumably responsible for contraction of the trunk and plumes into the gelatinous tube that houses the worm. However, two important annelid features--the presence of segmentation and nephridia--have not yet been identified in Osedax females.

This is the first analysis of growth in Osedax. Growth rates are poorly understood in most siboglinid polychaetes, as they have not been characterized in frenulates and are highly variable in vestimentiferans. Riftia pachyptila, a vestimentiferan that lives in association with ephemeral hydrothermal vents, has among the fastest growth rates known for any marine invertebrate (Lutz et al., 1994); whereas the cold-seep vestimentiferans Lamellibrachia luymesi and See-piophilia jonesi are long-lived and extremely slow growing (Fisher et al., 1997; Cordes et al., 2007). Osedax roseus n. sp. also exhibits the rapid growth and early maturation expected for a species that occupies a relatively ephemeral environment. When we first examined the carcass of Whale-1018 only 2 months after its deployment, its vertebrae had already been cleared of most soft tissues by mobile scavengers. Although bones were not sampled during this initial visit, careful examination of the high-definition video revealed patches of small Osedax (Braby et al., 2007). Osedax roseus n. sp. was first collected and positively identified with mtDNA one month later. Although less than 3 months old, 28% of the females were already producing eggs, and 27% of the females hosted one or more dwarf males (Fig. 1A). Mean sizes of the females did not increase during subsequent samples for a variety of reasons, including possible artifacts due to different fixation methods or the time taken to retrieve and process bones and the effect this might have had on the contraction of the females when preserved. It is also possible that somatic growth of females is extremely rapid and that they reached adult, or near adult, size within only a few months of settling. They would then be free to devote more resources to egg production. In support of this hypothesis, it is notable that the proportion of spawning females increased continuously to 71.8% during the sampling period. Female densities were lower and their sizes were bimodal in sample D, suggesting a second round of female recruitment to exposed bone. The overall density of female worms in this sample was also lower (Table 1), but it is unclear whether this bimodality is indicative of mortality in older females and hence space becoming available. Unfortunately, the regular sampling of transverse processes for analysis could not continue beyond sample D owing to the activity of the crab Chionoecetes tanneri. Further time-series analyses of recruitment and growth of Osedax may shed light on these issues. Nonetheless, the present analysis of O. roseus n. sp. clearly reveals that females rapidly colonize whalebones at 1000-m depth, quickly grow, reach sexual maturity, and then acquire dwarf males. Some mature females must wait to acquire dwarf males before they can successfully spawn; conversely, others gain males before they are sexually mature (Fig. 2). This difference is presumably a function of the haphazard nature of larval settlement across the bone surface, or the location of females in optimal or non-optimal places for gathering larvae from the water column.


The sequencing of a 226-bp fragment of mtCOI (GenBank #: EU164774) from a small sperm-containing, larval-like organism found in the tube of a female Osedax provides unequivocal molecular support for the morphological interpretation presented here, and in Rouse et al. (2004), that these represent Osedax dwarf males. Except for differences in size, Osedax roseus n. sp. males are essentially similar (Fig. 7) to those of 0. rubiplumus and 0. frankpressi (Rouse et al., 2004). Dwarf males were not originally reported for O. mucofloris and O. japonicus (Fujikura et al., 2006; Glover et al., 2005), but have now been discovered (Glover, Dahlgren, and Rouse, pers. obs.; Fujikura, pers. obs.). Dwarf males occur in all six species of Osedax that are currently known from Monterey Bay (Braby et al., 2007), and these species represent phylogenetic lineages that encompass the Swedish and Japanese Osedax (unpubl. data.), so this result is not surprising. Osedax males seem to be arrested larval forms that appear to have only yolk reserves to develop gametes. Posterior chaetae and the anterior ciliary band resembling a prototroch and posterior chaetae are characteristics of metatrochophore larvae of other siboglinids such as Riftia and Siboglinum (Gardiner and Jones, 1993; Southward, 1999). Such extreme sexual dimorphism involving paedomorphosis is unique among siboglinids, as all other known species of frenulates and vestimentiferans have equal-sized sexes (Ivanov, 1963; Bakke, 1990; Gardiner and Jones, 1993). As far as is known, fertilization is internal in siboglinids, via spermatophores in frenulates (Southward, 1999) or spermatozeugmata in vestimentiferans (Hilario et al., 2005). Fertilization is presumably internal in Osedax, as the dwarf males lie in the lumen of the female's tube, generally near the oviduct. Mature sperm gather anteriorly in the body of the male and a sperm duct runs into the head (Fig. 7C), where sperm presumably exit the body (Rouse et al., 2004), but the mechanism of sperm delivery to eggs remains unknown, as does the site of fertilization. Eggs naturally spawned from a female's oviduct (Fig. 4C) are generally fertilized, because they usually rapidly begin cleavage (unpubl. data). Although the originally described Osedax females had large numbers of males in their tubes (up to 114 in O. rubiplumus and 80 in O. frankpressi), Osedax roseus had on average only 3.5 in the last sample taken and nearly half of the 319 females studied from all samples had no males. It may be that the number of dwarf males in O. mucofloris and O. japonicus is also low and so they were easily overlooked.


Ghiselin (1974) proposed that dwarf males are most likely to evolve when population density is low, females are sedentary or hard to find, and a fitness premium exists on females as opposed to males. They occur in some echiuran and dinophilid annelids (Prevedelli and Vandini, 1999), and in marine and terrestrial animal taxa as diverse as bivalves, cephalapods, barnacles, copepods, spiders, rotifers, and fish (reviewed in Vollrath, 1998). Notable cases involving dwarf parasitic or commensal males occur in deep-sea anglerfish (Shedlock et al., 2003; Pietsch, 2005), and in some sessile filter-feeding barnacles (Klepal, 1987) that occur at very low densities. In these situations, males risk not finding a female, so rapid maturation and reduced size of males is favored as a means of securing a mate (Andersson, 1994). Local population densities of Osedax may be very high on a given whale carcass (Goffredi et al., 2004). Additionally, the present molecular data along with previous estimates (Rouse et al., 2004) suggest that the number of Osedax females contributing to larval pools is immense in Monterey Bay. Nonetheless, the overall density of adult females will be a function of the spatial and temporal frequency of suitable bones. Rough estimates of whale-fall densities in the northeast Pacific suggest that they might occur as frequently as one every 5 to 16 km (Smith and Baco, 2003), but we know little of the distribution of other marine mammal bones that may be suitable for Osedax. Furthermore, the longevity of these food patches and their availability for Osedax are poorly known. Large whale carcasses that fall in anoxic basins, such as those off southern California in the Santa Catalina Basin, persist for many decades (Smith and Baco, 2003), but Osedax is not abundant on these carcasses. In contrast, representatives of the genus are diverse and abundant at all depths in Monterey Bay, and whale bones decompose very rapidly, even within the oxygen minimum zone, due to the actions of these worms (Braby et al., 2007).

Rouse et al. (2004) hypothesized that sex is environmentally determined in Osedax, with the larvae that settle on bones maturing as females and the larvae that subsequently land on females becoming males. A model for environmental sex determination (ESD) with dwarf males is found in bonellid echiuran annelids such as Bonellia viridis (Baltzer, 1934; Schembri and Jaccarini, 1978; Jaccarini et al., 1983; Berec et al., 2005). When sexually undifferentiated Bonellia larvae land on the extended feeding proboscis of a female, they are transformed into dwarf males by the action of a masculinizing hormone, bonellin (Agius, 1979). Male harem size was correlated with female size in O. rubiplumus (Rouse et al., 2004), and this is also supported by the present study (Fig. 2). Many of the small (presumably young) females of O. rubiplumus did not host males, and larger females had an average of 30 males per tube, resulting in an overall sex ratio of 17:1 (Rouse et al., 2004). This shift in sex ratio from female-biased to strongly male-biased suggested that a Bonellia-like ESD system might be operating in Osedax. The present analysis of males and females of Osedax roseus n. sp. and the earlier data on O. rubiplumus are clearly consistent with the ESD hypothesis for Osedax. The density of Osedax roseus n. sp. females on bones did not increase across samples A through D, spanning more than 6 of the 9 nine months post-deployment, but the number of males present in females' tubes increased continuously from 0.333 to 3.5 males per female. Perhaps the extended palps of adult females limit the access of subsequently settling Osedax larvae to bone surfaces (Fig. 3B). Consequently, these larvae may be transformed into males if they settle on the palps of females and are drawn into their tubes. The growth of male harem sizes supports this scenario (Fig. 1, right panel). Male harem sizes in samples A, B, and C had a Poisson distribution, which is consistent with a random accumulation of rare events, but male harem sizes were bimodal in sample D (Fig. 1D, right). Harem sizes increased to a modal number of four males in larger females, with a second mode centered on zero in small females that were not yet fecund. This interpretation is supported by a significant positive relationship between female size (and hence presumably age) and male harem size in sample D (Fig. 2). An alternative hypothesis is that Osedax larvae may be male as a "default" condition and that contact with bone and symbiotic bacteria triggers development into the female Osedax form. Controlled laboratory experiments on Osedax larvae will be required to explore these hypotheses further.


We thank the crews and pilots of the R/Vs Western Flyer and Point Lobos and ROVs Tiburon and Ventana for their technical support and patience. We are grateful to Fredrik Pleijel for allowing the use of his photographs of O. roseus n. sp. The staff of Adelaide Microscopy (University of Adelaide) was essential for the TEM, SEM, and CLSM results. Thanks to Kats Fujikura (JAMSTEC, Japan), Adrian Glover (Natural History Museum, London), and Thomas Dahlgren (Tjarno Marine Biological Laboratory, Sweden) for sharing their new observation on dwarf males of Osedax. This manuscript benefited from useful comments from reviews by Adrian Glover, Steve Gardiner, an anonymous reviewer, and the Editor. Funding for this project was provided by the Monterey Bay Aquarium Research Institute (The David and Lucile Packard Foundation), NSF Award #0334932 (under the Assembling the Tree of Life program) to investigate the Tree of Life for Protostomes, and SIO startup funds to GWR.

Literature Cited

Agius, L. 1979. Larval settlement in the echiuran worm Bonellia viridis: settlement on both the adult proboscis and body trunk. Mar. Biol. 53: 125-129.

Andersson, M. 1994. Sexual Selection. Princeton University Press, Princeton, NJ.

Bakke, T. 1990. Pogonophora. Pp. 37-48 in Reproductive Biology of Invertebrates, Vol. 4, Part B, Fertilization, Development, and Parental Care, K. G. Adiyodi and R. G. Adiyodi, eds. John Wiley, Chichester.

Baltzer, F. 1934. Experiments on sex development in Bonellia viridis. Biol. Bull. 10: 101-108.

Berec, L., J. Schembri, and S. Boukal. 2005. Sex determination in Bonellia viridis (Echiura: Bonelliidae): population dynamics and evolution. OIKOS 108: 473-484.

Braby, C. E., G. W. Rouse, S. B. Johnson, W. J. Jones, and R. C. Vrijenhoek. 2007. Bathymetric and temporal variation among Osedax boneworms and associated megafauna on whale-falls in Monterey Bay, California. Deep-Sea Res. I 54: 1773-1791.

Cordes, E. E., D. C. Bergquist, M. L. Redding, and C. R. P. Fisher. 2007. Patterns of growth in cold-seep vestimenferans including See-piophila jonesi: a second species of long-lived tubeworm. Mar. Ecol. 28: 160-168.

Fisher, C. R., I. A. Urcuyo, M. A. Simpkins, and E. Nix. 1997. Life in the slow lane: growth and longevity of cold-seep vestimentiferans. Mar. Ecol. 18: 83-94.

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, Vol. 1, Onychophora, Chilopoda and Lesser Protostomata, F. W. Harrison and M. E. Rice. eds. Wiley-Liss, New York.

Ghiselin, M. 1974. The Economy of Nature and the Evolution of Sex. University of California Press, Berkeley, CA.

Glover, A. G., B. Kallstrom, 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. 272: 2587-2592.

Goffredi, S. K., C. K. Paull, K. Fulton-Bennett, L. A. Hurtado, and R. C. Vrijenhoek. 2004. Unusual benthic fauna associated with a whale fall in Monterey Canyon. California. Deep-Sea Res. I 51: 1295-1306.

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 and functional diversity of microbial symbionts associated with new species of Osedax polychaete worms. Appl. Environ. Microbiol. 73: 2314-2323.

Hilario, A., C. M. Young, and P. A. Tyler. 2005. Sperm storage, internal fertilization, and embryonic dispersal in vent and seep tube-worms (Polychaeta: Siboglinidae: Vestimentifera). Biol. Bull. 208: 20-28.

Ivanov, A. V. 1961. The structure of the genital system of Pogonophora. Syst. Zool. 10: 24-34.

Ivanov, A. V. 1963. Pogonophora. Academic Press, London.

Jaccarini, V., P. J. Schembri, and M. Rizzo. 1983. Sex determination and larval sexual interaction in Bonellia viridis Rolando (Echiura: Bonelliidae). J. Exp. Mar. Biol. Ecol. 66: 25-40.

Klepal, W. 1987. A review of the comparative anatomy of the males in cirripedes. Oceanogr. Mar. Biol. Annu. Rev. 25: 285-351.

Kruskal, W. H., and W. A. Wallis. 1952. Use of ranks in one-criterion variance analysis. J. Am. Stat. Assoc. 47: 583-621.

Levene, H. 1960. Robust tests for the equality of variances. Pp. 278-292 in Contributions to Probability and Statistics, I. Olkin, ed. Stanford University Press, Stanford, CA.

Lutz, R. A., T. M. Shank, D. J. Fornari, R. M. Haymon, M. D. Lilley, K. L. Von Damm, and D. Desbruyeres. 1994. Rapid growth at deep-sea vents. Nature 371: 663-664.

Nei, M. 1987. Molecular Evolutionary Genetics. Columbia University Press, New York.

Nelson, K., and C. R. Fisher. 2000. Absence of cospeciation in deep-sea vestimentiferan tube worms and their bacterial endosymbionts. Symbiosis 28: 1-15.

Pietsch, T. W. 2005. Dimorphism, parasitism, and sex revisited: modes of reproduction among deep-sea ceratioid anglerfishes (Teleostei: Lophiiformes). Ichthyol. Res. 52: 207-236.

Prevedelli, D., and R. Z. Vandini. 1999. Survival, fecundity and sex ratio of Dinophilus gyrociliatus (Polychaeta: Dinophilidae) under different dietary conditions. Mar. Biol. 133: 231-236.

Rasband, W. S. 2006. ImageJ. Ver. 1.34. U.S. National Institutes of Health, Bethesda, Maryland. [Online] Available: [30 November 2007].

Rozas, J., J. C. Sanchez-DelBarrio, X. Messeguer, and R. Rozas. 2003. DnaSP, DNA polymorphism analyses by the coalescent and other methods. Bioinformatics 19: 2496-2497.

Rouse, G. W., S. K. Goffredi, and R. C. Vrijenhoek. 2004. Osedax: bone-eating marine worms with dwarf males. Science 305: 668-671.

SAS Institute. 2000. JMP[R] User's Guide. SAS Institute, Cary, NC.

Schembri, P. J., and V. Jaccarini. 1978. Some aspects of the ecology of the echiuran worm Bonellia viridis and associated infauna. Mar. Biol. 47: 55-61.

Shapiro, S. S., and M. B. Wilk. 1965. An analysis of variance test for normality (complete samples). Biometrika 52: 591-611.

Shedlock, A. M., T. W. Pietsch, M. G. Haygood, P. Bentzen, and M. Hasegawa. 2003. Molecular systematics and life history evolution of anglerfishes (Teleostei: Lophiiformes): evidence from mitochondrial DNA. Steenstrupia 28: 129-144.

Smith, C. R., and A. R. Baco. 2003. Ecology of whale falls at the deep-sea floor. Oceanogr. Mar. Biol. Annu. Rev. 41: 311-354.

Southward, E. C. 1993. Pogonophora. Pp. 327-369 in Microscopic Anatomy of Invertebrates, Vol. 12, Onychophora, Chilopoda and Lesser Protostomata., F. W. Harrison and M. E. Rice, eds. Wiley-Liss, New York.

Southward, E. C. 1999. Development of Perviata and Vestimentifera (Pogonophora). Hydrobiologia 402: 185-202.

van der Land, J., and A. Norrevang. 1977. Structure and relationships of Lamellibrachia (Annelida, Vestimentifera). K. Dan. Vidensk. Selsk. Biol. Skr. 21: 1-102.

Vollrath, F. 1998. Dwarf males. Trends Ecol. Evol. 13: 159-163.

Webb, M. 1977. Studies on Lamellibrachia barhami (Pogonophora) II. The reproductive organs. Zool. Jahrb. Abt. Anat. Ont. Tiere 97: 455-481.

G. W. ROUSE (1,*), K. WORSAAE (2), S. B. JOHNSON (3), W. J. JONES (3), AND R. C. VRIJENHOEK (3)

(1) Scripps Institution of Oceanography, 9500 Gilman Drive, La Jolla, California 92093-0202; (2) Marine Biological Laboratory, University of Copenhagen, Denmark; and (3) Monterey Bay Aquarium Research Institute, Moss Landing, California 95039

Received 18 July 2007; accepted 10 October 2007.

* To whom correspondence should be addressed. E-mail:
Table 1 Summary of sample data, including time since deployment, for

                      Time   No. of       Area
Sample  Date          (mo.)  females      ([cm.sup.2])

--      12 Oct. 2004  2      n.a.         n.a.
A       5 Jan. 5005   3       60           36.8
B       1 Feb. 2005   4       83           52.1
C       14 Apr. 2005  6       73           44.8
D       18 Jul. 2005  8      159[dagger]  149.5

        Worm density
Sample  (per [cm.sup.2])  Preservative*  Purpose

--                        n.a.           Video and mapping
A                         E, F           DNA, morphology
B       1.63              G              SEM, TEM, morphology
C       1.59              F              Morphology
D       1.62              F, P           Morphology, CLSM

* Preservatives: E = 95% ethanol; F = 10% seawater formalin; G =
glutaraldehyde; P = paraformaldehyde; n.a. = not applicable.
[dagger] 56 females not removed from bone for measurement and counts of
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:Rouse, G.W.; Worsaae, K.; Johnson, S.B.; Jones, W.J.; Vrijenhoek, R.C.
Publication:The Biological Bulletin
Article Type:Report
Geographic Code:1USA
Date:Feb 1, 2008
Previous Article:Comparison of gamete compatibility between two blue mussel species in sympatry and in allopatry.
Next Article:Ionic composition of endolymph and perilymph in the inner ear of the oyster toadfish, Opsanus tau.

Terms of use | Privacy policy | Copyright © 2018 Farlex, Inc. | Feedback | For webmasters