SEARCH

SEARCH BY CITATION

Summary

  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgements
  8. References
  9. Supporting Information

The shrimp Rimicaris exoculata from hydrothermal vents on the Mid-Atlantic Ridge (MAR) harbours bacterial epibionts on specialized appendages and the inner surfaces of its gill chamber. Using comparative 16S rRNA sequence analysis and fluorescence in situ hybridization (FISH), we examined the R. exoculata epibiosis from four vents sites along the known distribution range of the shrimp on the MAR. Our results show that R. exoculata lives in symbiosis with two types of filamentous epibionts. One belongs to the Epsilonproteobacteria, and was previously identified as the dominant symbiont of R. exoculata. The second is a novel gammaproteobacterial symbiont that belongs to a clade consisting exclusively of sequences from epibiotic bacteria of hydrothermal vent animals, with the filamentous sulfur oxidizer Leucothrix mucor as the closest free-living relative. Both the epsilon- and the gammaproteobacterial symbionts dominated the R. exoculata epibiosis at all four MAR vent sites despite striking differences between vent fluid chemistry and distances between sites of up to 8500 km, indicating that the symbiosis is highly stable and specific. Phylogenetic analyses of two mitochondrial host genes showed little to no differences between hosts from the four vent sites. In contrast, there was significant spatial structuring of both the gamma- and the epsilonproteobacterial symbiont populations based on their 16S rRNA gene sequences that was correlated with geographic distance along the MAR. We hypothesize that biogeography and host–symbiont selectivity play a role in structuring the epibiosis of R. exoculata.


Introduction

  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgements
  8. References
  9. Supporting Information

The alvinocaridid shrimp Rimicaris exoculata (Williams and Rona, 1986) is endemic to hydothermal vents on the Mid-Atlantic Ridge (MAR) (Schmidt et al., 2008a). Large swarms containing as many as 3000 shrimp per m2 (Gebruk et al., 2000) aggregate on hydrothermal vent chimneys in the mixing zone between electron donor-rich vent fluids and the surrounding oxidized seawater. The source of nutrition for R. exoculata is unclear, but a large chemoautotrophic bacterial biomass would be needed to support such dense swarms at hydrothermal vents in the deep sea where the input of organic matter from photosynthesis is extremely low (Van Dover, 2000).

The stable isotopic composition of adult shrimp indicates a chemosynthetic food source (Van Dover et al., 1988; Rieley et al., 1999). Unlike the bathymodiolin mussels they co-occur with on the MAR, which rely on endosymbiotic methane- and sulfur-oxidizing bacteria for their nutrition (Robinson et al., 1998; Pimenov et al., 2002; Duperron et al., 2006), R. exoculata appears to have no endosymbiotic bacteria (Van Dover et al., 1988). Instead, the shrimp host a dense covering of epibiotic bacteria on specialized appendages within the gill chamber (Fig. 1). The morphology of the shrimp shows adaptation to the symbiosis, as the mouthparts within the gill chamber are atypically large and densely covered with setae, to which the ectosymbionts are attached (Van Dover et al., 1988; Casanova et al., 1993; Gebruk et al., 1993; Komai and Segonzac, 2008). A nutritional role has been suggested for the ectosymbionts, but it is unclear how energy could be transferred from the ectosymbionts to the host. Rimicaris exoculata does not have a reduced gut like many chemosynthetic hosts with endosymbiotic bacteria that rely on their endosymbionts for nutrition (Van Dover et al., 1988). To gain nutrition from the bacterial ectosymbionts, R. exoculata could either take up organic compounds from them through trans-epidermal transfer, or ingest them. Two possible ways in which the shrimp could ingest their symbionts are: (i) the shrimp harvest their ectosymbionts using their modified feeding appendages, and transfer them to the mouth (Gebruk et al., 1993), or (ii) the shrimp ingest their exuvae after moulting (Segonzac et al., 1993). While the benefit of the association to the shrimp host remains unclear, chemosynthetic ectosymbionts would represent a rich source of nutrition, enabling the shrimp to live, indirectly, from the inorganic energy sources abundant at hydrothermal vents. The bacteria most likely benefit from the association as the shrimp position themselves in the mixing zone between electron donor-rich vent fluids and the surrounding seawater, allowing the ectosymbionts stable and simultaneous access to electron donors and acceptors.

image

Figure 1. Morphology of the Rimicaris exoculata symbiosis. A. The R. exoculata gill chamber with carapace removed, showing the mouthparts (Sc and Ex) to which the shrimp epibionts are attached. Sc, scaphognathite; Ex, exopodite; Gi, gill. Scale bar = 5 mm. B. Scaphognathite dissected out of the shrimp. Scale bar = 5 mm. C. Closer view of the scaphognathite setae. The filamentous epibiotic bacteria can be seen as a white fuzzy material. D. Scanning electron microscope image showing the filamentous epibionts. Insert: a ‘thick’ filament is indicated with a white arrowhead, a ‘thin’ filament with a yellow arrowhead. Scale bar = 10 µm.

Download figure to PowerPoint

The metabolism of the ectosymbionts has not been clearly identified. Autotrophy has been demonstrated by RuBisCO activity (Wirsen et al., 1993; Cavanaugh and Robinson, 1996; Polz et al., 1998) and the incorporation of radioactively labelled inorganic carbon (Galchenko et al., 1989; Jannasch et al., 1991; Polz et al., 1998). A number of different electron donors have been suggested to fuel the symbiosis. The observation of internal sulfur globules in shrimp ectosymbionts from the Trans-Atlantic Geotransverse (TAG) vent field led to the conclusion that the ectosymbionts are chemoautotrophic sulfur-oxidizing bacteria (Gebruk et al., 1993). At the Rainbow vent field, the ectosymbionts are associated with iron oxyhydroxide minerals that appear to have been precipitated by a biological rather than chemical process (Gloter et al., 2004), and this led to the hypothesis that the ectosymbionts at this vent field might gain their energy by iron oxidation (Gloter et al., 2004; Zbinden et al., 2004). This would be a novel process, as all currently known chemosynthetic symbioses rely on the oxidation of methane or reduced sulfur compounds (Cavanaugh et al., 2006; Dubilier et al., 2008).

Hydrothermal vents on the MAR can either be ultramafic- or basalt-hosted. In basalt-hosted systems, the end-member vent fluids are enriched in sulfide and depleted in hydrogen and methane. In contrast, fluids in ultramafic-hosted systems are enriched in hydrogen and methane, and depleted in sulfide (Charlou et al., 2002; Schmidt et al., 2007). The geological setting of the vent fields has been hypothesized to influence the diversity of the free-living bacterial community (Perner et al., 2007), but this has not yet been investigated for symbiotic bacteria. Schmidt and co-workers (2008b) modelled energy budgets at two MAR vent sites and suggested that the diversity of the R. exoculata epibiosis might differ between the ultramafic-hosted Rainbow and basalt-hosted TAG sites, based on thermodynamic predictions of the energy available from the oxidation of different electron donors.

The first molecular studies of the R. exoculata epibionts were from shrimp collected at the Snake Pit vent field on the MAR (Polz and Cavanaugh, 1995). These showed that they all belong to a single phylotype within the Epsilonproteobacteria, despite the presence of various morphotypes (Polz and Cavanaugh, 1995). A recent study of R. exoculata from the Rainbow vent field on the MAR suggested that the epibiont diversity might be higher than previously assumed based on 16S rRNA gene sequences and ultrastructural observations, but fluorescence in situ hybridization was not used to distinguish between epibionts and casually associated bacteria or contaminants (Zbinden et al., 2008).

In this study, we analysed the phylogeny of R. exoculata and their epibionts from four vent fields, Rainbow, TAG, Logatchev and South MAR. These four vent fields are separated by up to 8500 km along the MAR and span the known distribution range of R. exoculata. Two of the vents, Rainbow and Logatchev, are ultramafic-hosted, while the two others, TAG and South MAR, are basalt-hosted (Table S1). Our aim was to re-examine the diversity of the shrimp ectosymbionts based on observations that multiple morphotypes occur on the shrimp (Polz and Cavanaugh, 1995; Zbinden et al., 2004; Zbinden et al., 2008), and that epibiont diversity might differ between ultramafic- and basalt-hosted vent fields.

Results

  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgements
  8. References
  9. Supporting Information

Host phylogeny

Alignment of the mitochondrial cytochrome oxidase subunit I (COI) genes from 12 R. exoculata individuals, three from each of the four vent sites, identified one shared and five non-shared substitutions. The single shared substitution at position 525 is shared by the three TAG shrimp and one Logatchev shrimp. Since the COI gene was so highly conserved in shrimp from geographically distant sampling sites, we analysed an additional mitochondrial marker gene encoding cytochrome b (CytB). Although the CytB gene contained more substitutions than the COI gene in R. exoculata populations (four shared, 11 non-shared substitutions), geographic clades also could not be identified based on this gene (Fig. S1).

Diversity of R. exoculata epibionts

16S rRNA clone libraries were constructed for the same 12 individuals used for host phylogenetic analyses. Between 53 and 90 clones were partially sequenced for each individual, and clones were assigned to groups with > 99% sequence similarity (Table 1). Seven groups dominated the clone libraries from all four vent fields, of which five belonged to the Epsilonproteobacteria (Epsilon 1–5), and two belonged to the Gammaproteobacteria (Gamma 1 and 2) (Table 1).

Table 1.  Clone library results.
 Rainbow (RB)TAG (TG)Logatchev (LG)South MAR (SM)
Individual No.123456101112131415
  1. Number of partial sequences found in each clone library that belonged to the 5 Epsilon symbiont and 2 Gamma symbiont groups. Within each of these 7 groups, sequences shared > 99% identity. Number of full sequences analysed is shown in parentheses. CFB (Cytophaga-Flavobacteria-Bacteroidetes) and Other include phylogenetically diverse sequences that were not found in all individuals.

Epsilon 1 31 (3)48 (2)         
Epsilon 2   26 (2)8 (2)2 (2)      
Epsilon 3      56 (3)55 (2)26 (1)   
Epsilon 4 4 (1)8         
Epsilon 5         18 (1)27 (2)15 (2)
Gamma 158 (2)26 (2)7 (2)15 (1)62 (1)49 (1)10 (2)2 (3)12 (3)   
Gamma 2         17 (2)18 (2)31 (1)
CFB5173713369951121
Other13135 111063625199
Total No. of partial sequences767969538365888190657582

FISH analyses showed that the epsilon- and gammaproteobacterial groups that dominated the clone libraries also dominated the bacterial community on R. exoculata from all four vent fields. To show this, we did three-colour hybridizations with probes specific to the epsilon- and gammaproteobacterial groups found in the clone libraries the general bacterial probe EUB I-III (Amann et al., 1990; Daims et al., 1999), and 4′,6-diamidino-2-phenylindole (DAPI) staining (Fig. 2) (see Table S2 for probes used in this study). The specific probes hybridized with filamentous epibionts that could be distinguished from each other based on their morphology (Fig. 2). The probes specific to the gammaproteobacterial sequences hybridized with filaments that had coccoid-shaped cells and a diameter of approximately 1 µm, while the probes specific to the epsilonproteobacterial sequences hybridized with filaments that were 3 µm in diameter (Fig. 2).

image

Figure 2. Fluorescence in situ hybridization. A. Confocal laser scanning micrograph of a cross-section through a R. exoculata scaphognathite from Individual 1 from the Rainbow vent site, showing the two symbionts attached to the scaphognathite seta (Se). The Gamma symbiont (green) was hybridized with the Rexogam1268RT probe targeting the Rainbow and TAG Gamma symbiont sequences, and the Epsilon symbiont (red) was hybridized with the Epsilon 1 probe targeting the Rainbow Epsilon 1 symbiont. B–E. Epifluorescence micrographs of cross sections through R. exoculata scaphognathite setae (Se) showing the specificity of the site-specific probes designed for the Epsilon symbionts. The DAPI stain is shown in blue, probe signals for the Gamma symbionts in light green, and probe signals for the Epsilon symbionts in pink. (B) South MAR R. exoculata Individual 17 hybridized with the South MAR Epsilon 5 probe and the Rexogam1268LS Gamma probe. (C) The same South MAR individual as in (B), hybridized with the Rexogam1268LS Gamma probe and the Epsilon 3 probe. The Logatchev Epsilon 3 probe shows no signals when hybridized with the South MAR individual. (D and E) Images from Logatchev Individual 10, hybridized in (D) with the South MAR Epsilon 5 probe and in (E) with the Logatchev Epsilon 4 probe. No signals are seen on the Logatchev individual with the South MAR Epsilon 5 probe. (D) and (E) are also hybridized with the Rexogam1268LS probe for the Gamma symbionts. Scale bars = 20 µm. The scale bar in (B) applies to images (B–E).

Download figure to PowerPoint

The relative abundance of the 16S rRNA clone sequences from the epsilon- and gammaproteobacterial groups varied considerably both among individuals from the same site and individuals from different sites (Table 1). Both groups were present in 11 of the 12 shrimp examined, and their relative abundance in the clone libraries was consistent with observations of their relative abundance by FISH. Only one individual had no epsilonproteobacterial sequences in its 16S rRNA clone library (Individual 1 in Table 1). However, this symbiont was clearly present on this individual based on FISH, albeit at low abundance.

16S rRNA gene phylogeny of the epsilonproteobacterial symbionts

The epsilonproteobacterial 16S rRNA sequences from the R. exoculata clone libraries fell in a clade that included previously published sequences from R. exoculata epibionts from the Snake Pit vent field (Polz and Cavanaugh, 1995), and clone sequences from the gut of R. exoculata from the Rainbow vent field (Zbinden and Cambon-Bonavita, 2003) (Fig. 3). The closest cultured relative was Sulfurovum lithotrophicum, a rod-shaped sulfur-oxidizing chemolithoautotroph isolated from a hydrothermal vent in the Western Pacific (Inagaki et al., 2004) (91.9–92.9% sequence identity to the R. exoculata Epsilon 1–5 symbionts, Fig. 3). Sequences from bacteria associated with other hydrothermal vent invertebrates, the gastropod Crysomallon squamiferum, the barnacle Vulcanolepas osheai, the crab Kiwa hirsuta and the polychete worms Alvinella pompejana and Paralvinella palmiformis also fell between the R. exoculata Epsilon symbionts (Fig. 3B). A number of sequences from free-living hydrothermal vent bacteria also belonged to this clade (Fig. 3B).

image

Figure 3. 16S rRNA phylogeny of the epsilonproteobacterial symbiont. Maximum likelihood phylogeny of the R. exoculata Epsilon symbiont 16S rRNA genes. The numbers in parentheses refer to the number assigned to each individual. (Table S3). The Gammaproteobacterium Vibrio cholerae (accession number AY494843) was used as an outgroup (arrow). A. Relationship of the R. exoculata Epsilon symbionts to other members of the Epsilonproteobacteria. B. Expanded view of boxed area in (A). Phylogeny of the R. exoculata Epsilon symbionts and their close relatives, both invertebrate-associated and free-living. Colours indicate the geographic location of the sampling sites. Bars indicate 10% estimated sequence divergence. Only bootstrap values (100 re-samplings) over 60 are shown.

Download figure to PowerPoint

16S rRNA gene phylogeny of the gammaproteobacterial symbionts

The gammaproteobacterial 16S rRNA sequences from the R. exoculata clone libraries formed a clade with bacteria associated with the hydrothermal vent animals C. squamiferum and K. hirsuta (Fig. 4). The closest relatives of this clade were free-living bacteria from a carbonate chimney at Lost City, and a clade of bacteria associated with V. osheai and K. hirsuta. The closest cultured free-living relative to the R. exoculata Gamma symbionts was Leucothrix mucor, a filamentous sulfur-oxidizer (Grabovich et al., 1999), with 90.2–90.8% sequence identity. The most closely related symbiont sequences were from the endosymbionts of the siboglinid tubeworms Oligobrachia mashikoi (88.5–90.0% sequence identity) and O. haakonmosbiensis (88.5–89.4% sequence identity). The clade containing the R. exoculata Gamma 1 and 2 sequences always grouped with L. mucor and the O. haakonmosbiensis symbionts in all phylogenetic analyses, but the relationship of this group to the methane- and sulfur-oxidizing endosymbionts of other chemosynthetic invertebrates, and to free-living methane- and sulfur-oxidizing bacteria was not consistent between treeing methods.

image

Figure 4. 16S rRNA phylogeny of the gammaproteobacterial symbiont. Maximum likelihood phylogeny of the R. exoculata Gamma symbiont 16S rRNA genes. The numbers in parentheses refer to the number assigned to each individual (Table S3). The Alphaproteobacterium Rhodocyclus tenuis (accession number D16208) was used as an outgroup (arrow). A. Relationship of the R. exoculata Gamma symbionts to other members of the Gammaproteobacteria. Methane-oxidizing endosymbionts (MOX) are shaded in green and sulfur-oxidizing endosymbionts in yellow (SOX). Bar indicates 10% estimated sequence divergence. B. Phylogeny of the R. exoculata Gamma symbionts and their close relatives, both invertebrate-associated and free-living. Colours indicate the geographic location of the sampling. Bar indicates 5% estimated sequence divergence. Only bootstrap values (100 re-samplings) over 60 are shown.

Download figure to PowerPoint

Biogeography of the R. exoculata symbionts

The R. exoculata Epsilon 1–5 symbionts formed geographic groups based on 16S rRNA analyses that were supported by maximum likelihood, parsimony, and neighbour-joining methods (Fig. 3B). With the exception of Rainbow, shrimp from each vent site harboured only a single group of sequences with > 99% identity. At Rainbow, the shrimp harboured two phylogenetically distinct groups, with > 99% sequence identity within each group: the dominant Epsilon 1 group, and the low-abundance group Epsilon 4. Sequence identities between the Epsilon 1–5 groups varied from 93.5% to 97.5%. Sequences from the ultramafic-hosted sites Rainbow and Logatchev did not group together to the exclusion of sequences from the basalt-hosted sites TAG and South MAR. Statistical analysis of the epibiont sequences showed a significant correlation between genetic distance based on the 16S rRNA gene and geographic distance along the MAR (r = 0.43, P < 0.001).

Unlike the Epsilon symbionts, the Gamma symbiont 16S rRNA sequences did not differ sufficiently to resolve geographic clades. The Gamma 2 symbiont sequences from the South MAR vent did, however, form a separate cluster to sequences from the northern MAR vent fields Rainbow, TAG and Logatchev (Gamma 1) in all treeing methods (Fig. 4B). Although site-specific geographic clades could not be resolved, a significant correlation was found between genetic and geographic distance for the Gamma symbiont (r = 0.85, P < 0.001).

To ensure that the site-specific differences in 16S rRNA found in the clone libraries are genuine and not caused by PCR or cloning bias, we designed probes for in situ detection of the epibiont groups (summarized in Table S2). For the Gamma symbionts, two probes were designed; one for the symbiont sequences from the northernmost vent fields, Rainbow and TAG, and one for those from the southernmost vent fields, Logatchev and South MAR (Table S2). The probe for the Rainbow and TAG Gamma symbionts bound specifically to all of the thinner filaments (Gamma morphotype) in R. exoculata individuals from the two target sites, and no signals were observed with this probe in individuals from either the Logatchev or South MAR vent fields. Conversely, the Logatchev and South MAR probe specifically bound all filaments with the Gamma morphology on R. exoculata from the target sites, and showed no signals on Rainbow or TAG individuals.

The Epsilon symbiont sequences were more divergent than the Gamma symbiont sequences, and this allowed us to design site-specific FISH probes for these groups. The Epsilon probes were used to identify the epibionts on R. exoculata from a single vent field, and tested against non-target organisms from the three other vents. For example, the probe targeting the South MAR Epsilon 5 symbionts bound specifically to the epsilonproteobacterial epibionts of shrimp from South MAR, and showed no signals when hybridized with shrimp from the Rainbow, TAG, Logatchev and vents (Fig. 2).

The probe targeting the Logatchev Epsilon 3 symbiont has no mismatches to the Rainbow Epsilon 4 symbiont, and could therefore be used to examine the relative abundance of the Epsilon 1 and 4 symbionts on Rainbow shrimp (Fig. S2). The Epsilon 1 symbionts were more abundant than the Epsilon 4 symbionts, just as the Epsilon 1 sequences were numerically dominant over the Epsilon 4 sequences in the clone libraries (Table 1). To confirm that the signals on the Rainbow shrimp were due to the presence of the Epsilon 4 and not the Logatchev Epsilon 3 symbiont, we also used another probe, RexoepsLG86, which perfectly matches the Logatchev Epsilon 3 symbiont, but has 1 mismatch to the Rainbow Epsilon 4 symbiont (Fig. S2). This probe did not hybridize with bacteria on the Rainbow shrimp.

For the free-living relatives of the Epsilon symbionts, there was no obvious geographic trend in their 16S rRNA phylogeny, or for invertebrate-associated bacteria that fell within the R. exoculata epibiont clade (Fig. 3B); for example, environmental sequences from hydrothermal fluids at both the Logatchev and South MAR vent fields fell throughout the tree and did not group according to geography. Thus, there appears to be no genetic structuring of the free-living relatives of the Epsilon symbionts. We could not analyse the distribution of free-living relatives of the Gamma symbionts, as no sequences from free-living bacteria were available that fell within the Gamma symbiont clade.

Discussion

  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgements
  8. References
  9. Supporting Information

Rimicaris exoculata has a dual symbiosis with filamentous Epsilon- and Gammaproteobacteria

The R. exoculata symbiosis was previously considered to be composed of a single epsilonproteobacterial phylotype based on 16S rRNA sequencing, FISH, and slot-blot hybridization of individuals from the Snake Pit vent site (Polz and Cavanaugh, 1995). A recent study showed that additional bacterial phylotypes were present in clone libraries from shrimp from the Rainbow vent site (Zbinden et al., 2008), but they were not investigated with FISH. We identified a second, novel symbiont type, belonging to the Gammaproteobacteria, as a dominant member of the R. exoculata epibiosis at all four MAR vent sites studied here. We did not have specimens of R. exoculata from Snake Pit to re-examine the diversity of the epibionts at this site, but it is likely that the Gamma symbionts are also present on Snake Pit shrimp, and were missed in the original study (Polz and Cavanaugh, 1995).

Both the Gamma and Epsilon symbionts were present on all shrimp from all four vent sites. Their relative abundance varied considerably, even among individuals collected at the same time from the same site. Given the high motility of the shrimp it is unlikely that this variability reflects vent fluid chemistry. An alternative explanation is that there is a sequential colonization pattern of the two symbiont types that is linked to the molting cycle of the shrimp (Corbari et al., 2007). We did not have enough individuals from different molting stages to test this hypothesis, but based on our observations, freshly molted shrimp were first colonized by the Epsilon symbionts while later stages were dominated by Gamma symbionts.

The R. exoculata Gamma symbionts fell in a clade consisting exclusively of bacteria associated with hydrothermal vent animals (the scaly snail Crysomallon squamiferum and the Yeti crab K. hirsuta, Fig. 4). The energy sources used by the bacteria associated with these vent animals are not known, but key genes for sulfur oxidation and sulfate reduction could be amplified from K. hirsuta epibiotic material (Goffredi et al., 2008). The closest relatives of the vent symbiont clade are the endosymbionts of Oligobrachia spp. tubeworms, and the free-living, filamentous bacterium L. mucor. The endosymbiont of O. haakonmosbiensis has been identified as a sulfur oxidizer (Lösekann et al., 2008), while the metabolism of the O. mashikoi symbiont is unclear (Kimura et al., 2003). Leucothrix mucor was originally characterized as a chemoorganoheterotroph (Brock, 1966), but was later shown to grow chemolithoheterotrophically, using reduced sulfur compounds as an energy source (Grabovich et al., 1999). The close phylogenetic relationship of the R. exoculata Gamma symbionts to L. mucor and the sulfur-oxidizing symbiont of O. haakonmosbiensis could indicate that the R. exoculata Gamma symbionts use reduced sulfur compounds as an energy source. However, both free-living and symbiotic methane-oxidizing bacteria were the sister group to the clade containing the R. exoculata Gamma symbionts, Oligobrachia spp. symbionts and L. mucor, although this branching order received very little bootstrap support (Fig. 4). The metabolic capabilities of the R. exoculata Gamma symbionts therefore remain unclear.

Epibionts of hydrothermal vent animals

Close relatives of both the R. exoculata Gamma and Epsilon symbionts identified in this study have also been found in clone libraries from three other hydrothermal vent animals: the Yeti crab K. hirsuta (Goffredi et al., 2008), the scaly snail C. squamiferum (Goffredi et al., 2004) mentioned above and the barnacle V. osheai (Suzuki et al., 2009). The presence of close relatives of both symbiont types on such diverse hosts could be due to multiple horizontal transmission events in both bacterial lineages. It was recently shown that a single gene is sufficient to alter the host range of bioluminescent V. fischeri symbionts (Mandel et al., 2009), which could explain how such events occur. Multiple horizontal transmission events have been proposed for Spiroplasma bacterial endosymbionts in phylogenetically unrelated host insects (Haselkorn et al., 2009). Unlike the diverse insect hosts of Spiroplasma spp. that share common habitats, there is no currently known hydrothermal vent site where the distribution ranges of R. exoculata, K. hirsuta, V. osheai and C. squamiferum overlap. It is possible that such a vent site exists, but has not yet been discovered. Alternatively, free-living forms of the symbiotic bacteria that can disperse between vent sites might allow horizontal transmission between unrelated hosts that do not co-occur.

Role of the epibionts

A nutritional role has been proposed for the ectosymbionts of K. hirsuta, but this is still unresolved, as the crab host has never been observed to feed off its epibionts (Goffredi et al., 2008). ‘Lau sp. A’, a relative of V. osheai that has a morphologically similar ectosymbiosis was also hypothesized to feed off its epibionts by using modified mandibles to comb the filamentous bacteria from their cirral setae (Southward and Newman, 1998). All R. exoculata individuals we investigated from four separate MAR vent fields had both Gamma and Epsilon symbionts. While this suggests that the association is obligate for the shrimp host, the role of the symbionts in shrimp nutrition remains unclear. This is a common theme for all currently known ectosymbiotic associations at hydrothermal vents, where a clear contribution to host nutrition has not yet been shown.

Host biogeography

The genes encoding COI and CytB have been used successfully in previous studies of shrimp phylogeography to differentiate geographically separate populations within a species, at geographic scales ranging from tens of kilometers (for example, Page et al., 2008) to hundreds of kilometers (for example, Teske et al., 2007; Hunter et al., 2008). Despite this, neither of these genes could resolve site-specific populations of R. exoculata at hydrothermal vents included in this study, which were separated by up to 8500 km along the MAR. This suggests that there is significant migration and therefore gene flow between populations over vast geographic distances (Vrijenhoek, 1997). High migration rates have been hypothesized for R. exoculata populations along the North MAR (Creasey et al., 1996; Shank et al., 1998). In addition, larval properties such as a planktotrophic lifestyle (Tyler and Young, 1999) and the presence of storage compounds of likely photosynthetic origin (Pond et al., 1997; Pond et al., 2000) indicate a prolonged larval stage capable of long-distance dispersal. Although we could not distinguish geographically separate R. exoculata populations based on analysis of mitochondrial genes, more variable markers such as the nuclear internal transcribed spacer (Chu et al., 2001) or microsatellites (Meng et al., 2009) might identify phylogeographic patterns that are not visible at the level of the COI or CytB gene.

Symbiont biogeography

Although geographically separate host populations could not be identified, the symbiont populations of R. exoculata at the four vent sites investigated showed significant spatial structuring, which we verified with FISH. We were only able to investigate a limited number of individuals from each vent site due to sampling limitations; however, for two of the four vent sites, different individuals were used for 16S rRNA gene sequencing and FISH, meaning that a total of six individuals were analysed from each of these two sites, and the results of both methods were always consistent. Mantel tests showed significant spatial structuring of the symbiont populations and indicated that geography affects the diversity of shrimp epibionts, not vent geochemistry.

The R. exoculata symbionts are attached to the outside surfaces of the shrimp, and must recolonize the shrimp exoskeleton after each molt. In addition, a free-living form of the Epsilon symbiont has been shown to make up a substantial proportion of the free-living community on sulfides at the Snake Pit vent site (Polz and Cavanaugh, 1995). They are therefore most likely horizontally transmitted. Assuming horizontal transmission of the symbionts, two models could explain the spatial structuring and biogeographic pattern we saw in the distribution of symbiont populations. In the first model, free-living symbiont populations are genetically isolated due to the existence of barriers to gene flow between vent fields. In this model, dispersal events are rare, and diversification of the symbiont population at a particular site is greater than transport of the symbionts, either free-living forms or host-attached, that would cause the mixing of populations. If this is the case, then the spatial structuring of symbiont populations would be expected to reflect the diversity of free-living forms of the symbionts. For example, only the Epsilon 2 symbiont would be found in the free-living community at the TAG vent site. Biogeography has been hypothesized to be a significant factor in structuring the diversity of hydrothermal vent fauna (Van Dover et al., 2002), and was recently shown to be the major factor in structuring populations of endosymbiotic bacteria at vents on the northern MAR (DeChaine et al., 2006).

In the second model, we assume that the free-living symbiont populations are not spatially structured along the MAR. For example, free-living forms of the Rainbow symbiont occur at other vent sites, and the TAG, Logatchev and South MAR symbionts occur at the Rainbow site in a free-living form. The structuring of symbiont populations at each vent site would therefore be due to specific colonization of the hosts by their symbionts from the pool of diverse free-living forms. This model requires highly specific recognition mechanisms between the hosts and their symbionts, and indeed these are known to play a critical role in horizontally transmitted animal-bacteria endosymbioses (Ruby, 2008). The association between the squid Euprymna scolopes and its luminescent bacterial symbionts Vibrio fischeri is one of the best understood model systems for symbiont recognition in horizontally transmitted associations (Nyholm and McFall-Ngai, 2004; Ruby, 2008). Remarkably, only a single regulatory gene is needed for conferring V. fischeri strains that can normally not colonize E. scolopes with the ability to infect these hosts (Mandel et al., 2009). A recent study of the nematode Laxus oneistus ectosymbiosis showed the role of host-expressed lectins in symbiont recognition and attachment to the worm cuticle (Bulgheresi et al., 2006), and a host-expressed lectin in the clam Codakia orbicularis was hypothesized to be involved in the recognition of its sulfur-oxidizing symbionts (Gourdine et al., 2007). The introduction of a single lectin gene into a plant that normally hosts nitrogen-fixing Rhizobium leguminosarum bv. viciae confers the ability to be colonized by R. leguminosarum bv. trifolii (Diaz et al., 1989). These last examples explain how specific recognition factors could be present in the R. exoculata populations at each vent field despite the lack of genetic structuring of their mitochondrial marker genes for COI and CytB: it is possible that symbiont specificity is determined by lectins or other genes that we did not analyse in this study.

It is not currently possible to determine which of the two models above best explains the observed geographic structuring of the R. exoculata symbionts. Only a very limited number of sequences are currently available from free-living bacteria related to the R. exoculata symbionts. These all belong to the Epsilonproteobacteria and do not group according to their geography. For example, sequences from free-living bacteria at the South MAR and Logatchev vent sites fell throughout the tree, and did not group exclusively with the symbiont sequences from these sites (Fig. 3B). This limited data set indicates that specific host-symbiont recognition drives the geographic grouping of the R. exoculata symbionts, but the number of sequences from free-living relatives of the shrimp symbionts is currently much too limited to provide sufficient support for either model. Extensive analyses of the free-living gamma- and epsilonproteobacterial populations from MAR vent sites are needed to better understand the processes causing the geographic structuring of the R. exoculata symbiont populations.

Conclusions and outlook

It is now known that free-living forms of horizontally transmitted symbionts can occur in the environment (Lee and Ruby, 1994; Miethling et al., 2000; Gros et al., 2003; Harmer et al., 2008), and that the abundance of the free-living forms can be correlated with the presence of the host (Lee and Ruby, 1994; Harmer et al., 2008). Surprisingly however, no study to date has investigated the spatial or geographic structure of free-living symbiont populations, although understanding their distribution patterns is crucial for determining the factors responsible for this structuring. This study provides the basis for future investigations of free-living forms of the R. exoculata epibionts. Comparing the distribution patterns of symbiotic bacteria and their free-living counterparts will provide insights into evolutionary processes such as migration, geographic isolation and symbiont-host interactions, and how these have shaped the diversity of symbiotic bacteria.

Experimental procedures

  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgements
  8. References
  9. Supporting Information

Sampling and storage

Rimicaris exoculata were collected at four vent fields along the MAR with a slurp gun on remotely operated vehicles (ROVs) (see Table S1 for the cruises, ROV and ship names, and chief scientists involved in the sampling). Samples were processed on board immediately where possible, or a maximum of 12 h after retrieval. They were either frozen for DNA or fixed for FISH analysis. Samples for FISH were fixed at 4°C for 4–10 h in 2% formaldehyde in 0.2 mm filtered seawater. After fixation, samples were washed 3 times at 4°C for 30 min in 0.2 µm filtered seawater, then stored at −20°C in a 50% ethanol 50% filtered seawater solution.

DNA extraction and PCR amplification

Genomic DNA was extracted separately from scaphognathite, exopodite and carapace tissue (Fig. 1) with the FastDNA SPIN kit for soil (Qbiogene, Carlsbad, CA, USA). DNA was stored in aliquots at −20°C. The 16S rRNA gene was amplified from pooled scaphognathite, exopodite, and carapace DNA of 3 individuals from each sampling site. The universal bacterial primers 8F and 1492R (Muyzer et al., 1995) were used for the amplification step. The reaction mixtures for PCR amplification contained 50 pmol of each primer, 2.5 µmol of each deoxynucleotide triphosphate, 1× Eppendorf buffer, 1 U of Eppendorf Taq polymerase and approximately 200 ng of genomic DNA. The final volume was adjusted to 50 µl with sterile water. The PCR program involved an initial denaturation step at 95°C for 5 min, followed by 25 cycles of 95°C for 1 min, 42°C for 1.5 min, and 72°C for 2 min, with a final elongation step at 72°C for 10 min. PCR bias was minimized by using only 25 (Rainbow) or 20 (TAG, Logatchev, South MAR) cycles, and pooling 4 (Rainbow) or 10 (TAG, Logatchev, South MAR) separate PCRs for each individual. PCR products were purified with a QIAquick PCR purification kit (QIAGEN, Hilden).

Cloning and sequencing of 16S rRNA genes

Purified PCR products were ligated at 4°C overnight with the pGEM-T Easy vector (Promega). The ligation product was used for transformation with a TOPO-TA kit (Invitrogen, Carlsbad, CA, USA). Clone libraries of 96 clones per shrimp were constructed. The insert size of white Escherichia coli colonies was controlled by PCR screening with vector primers M13F and M13R, using 0.5 µl of an overnight liquid culture as a template in a 20 µl reaction mixture. Partial sequencing of all positive clones was done using ABI BigDye and an ABI PRISM 3100 genetic analyser (Applied Biosystems, Foster City, CA, USA). Sequences were imported into BioEdit (Hall, 1997–2001) and aligned with ClustalW. Sequence groups were identified by visual inspection of the alignment. The phylogenetic affiliation of each group was determined by using blast (Altschul et al., 1997) for a few representative sequences from each group. Clones from the identified ectosymbiont groups were randomly chosen for full sequencing. Full sequences were assembled using Sequencher (http://www.genecodes.com).

Phylogenetic analyses

Sequences covering most of the 16S rRNA gene (900–1400 nt) were used for phylogenetic analyses. These were aligned against close relatives in ARB (Ludwig et al., 2004) using the Silva small subunit alignment (http://www.arb-silva.de; Pruesse et al., 2007). The automatic alignment was refined by hand. Maximum likelihood, parsimony and neighbour-joining phylogenies were calculated in ARB using 30% (Epsilon symbiont phylogeny, 1408 columns used for calculation) and 50% (Gamma symbiont phylogeny, 1405 columns used for calculation) positional variability filters. Positional variability filters were calculated using ARB with > 200 sequences across the Epsilonproteobacteria (for the Epsilon symbiont phylogeny), or the Gammaproteobacteria (for the Gamma symbiont phylogeny). A termini filter was used for all analyses, removing the primer sequences. Bootstrapping was done with the PhyML package in ARB, with 100 re-samplings.

Amplification and sequencing of shrimp mitochondrial genes

The mitochondrial cytochrome oxidase subunit I (COI) gene was amplified with the primers LCO1490 and HCO2198 (Folmer et al., 1994) using the following PCR cycling conditions: initial denaturation at 95°C for 5 min, followed by 36 cycles at 95°C for 1 min, 43°C for 1.5 min and 72°C for 2 min, then a final elongation step at 72°C for 10 min. The mitochondrial cytochrome b gene was amplified with primers designed for decapod crustaceans, Cybf and Cybr (Harrison, 2001), using the following cycling conditions: initial denaturation at 95°C for 5 min, followed by 36 cycles at 95°C for 1 min, 50°C for 1.5 min and 72°C for 2 min, then a final elongation step at 72°C for 10 min. PCR products were used for direct sequencing (see above).

Probe design and fluorescence in situ hybridization (FISH)

Oligonucleotide probes were designed either with the probe design function in ARB or by visually identifying a suitable target site in the ARB alignment. Vent field-specific probes were designed for the Epsilon symbionts, and were tested by cross-hybridization with samples from all other vent sites included in the study. It was not possible to design vent field-specific probes for the Gamma symbionts, as the 16S rRNA sequences were too closely related. In this case, two probes were designed, one targeting the Logatchev and South MAR sequences, the other targeting the Rainbow and TAG sequences. Probes for the Gamma symbiont sequences had at least two mismatches to all other sequences in the NCBI database.

Whole scaphognathite tissues were embedded in either paraffin or Steedman's wax (Steedman, 1957) and 6 µm thick sections cut with an RM 2165 microtome (Leica, Germany). The sections were collected on Superfrost Plus slides (Roth, Germany). Wax was removed from paraffin sections by washing in Roti-Histol (Roth, Germany) 3 times for 10 min each, and from Steedman's sections by washing in 96% ethanol 3 times for 5 min each. Sections were circled with a wax pen (PAP-pen, Kisker Biotech, Steinfurt, Germany), then rehydrated in an ethanol series consisting of 1 min in 96% ethanol, 1 min in 80% ethanol, then 1 min in 50% ethanol. Sections were hybridized in a buffer (0.9 M NaCl, 0.02 M Tris/HCl pH 8.0, 0.01% SDS, with the appropriate formamide concentration) containing probes at an end concentration of 5 ng µl−1. Sections were hybridized for 3 h at 46°C, then washed for 30 min at 48°C with buffer (0.1 M NaCl, 0.02 M Tris/HCl pH 8.0, 0.01% SDS, 5 mm EDTA), then rinsed in distilled water. To stain all DNA, sections were covered in a 1% DAPI solution, left for 3 min, rinsed with distilled water, then dipped in 96% ethanol and air dried. Sections were mounted in a mixture of Citifluor and Vectashield and examined using both a fluorescence microscope (Zeiss Axioskop, Germany) and a confocal laser-scanning microscope (Zeiss CLSM 510, Germany).

Scanning electron microscopy

For scanning electron microscopy, R. exoculata specimens were fixed as above. Single filaments of the scaphognathite were dehydrated in an ethanol series (70–100%) and 100% acetone, transferred to hexamethyldisilazane and air dried. Dried specimens were placed on carbon adhesive tabs on an aluminium stub, sputtered with gold and viewed in a Philips XL20 at 15 kV.

Statistical analysis

To test the hypothesis that genetic distances between the epibiont 16S rRNA sequences were correlated with their geographical distances, Mantel tests using 1000 permutations were performed using the R (v.2.8.0) package Vegan ( http://www.r-project.org). Geographical distances were estimated using Google Earth v.4.3 (http://earth.google.de). Along-ridge distances were estimated for MAR sites, as the axial valley is hypothesized to create corridors for the dispersal of organisms between vent fields (Tyler and Young, 2003).

Nucleotide sequence accession numbers

The sequences from this study are available through GenBank under the following accession numbers: FM203374–FM203407 and FN393020–FN393030 (symbiont 16S rRNA sequences), FN392996–FN393007 (R. exoculata COI gene) and FN393008–FN393019 (R. exoculata cytochrome b gene).

Acknowledgements

  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgements
  8. References
  9. Supporting Information

We would like to thank the captains and crews of the research vessels and ROVs that contributed to the sampling effort (Table S1), and Valérie Cueff-Gauchard (Exomar 2005), Olav Giere (M64/1) and Frank Zielinski (M64/2) for their help with the on-board work. Many thanks to Dennis Fink (MPI Bremen) for his expertise in Adobe Photoshop, to Claudia Bergin (MPI Bremen) for an introduction to confocal microscopy, and to Sandra Schoettner (MPI Bremen) and Erwan Roussel (Ifremer) for help in the lab. This work was supported by the Max Planck Society, the Priority Program 1144 (From Mantle to Ocean: Energy- Material- and Life-Cycles at Spreading Axes; contribution number 44) of the German Research Foundation (DFG), the DFG Cluster of Excellence at MARUM, Bremen, and a Marie Curie research training network (MOMARnet) scholarship to J.M.P. We are grateful to the Census of Marine Life working group ChEss for their support of research on the biogeography of chemosynthetic ecosystems.

References

  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgements
  8. References
  9. Supporting Information
  • Altschul, S.F., Madden, T.L., Schäffer, A.A., Zhang, J., Zhang, Z., Miller, W., and Lipman, D.J. (1997) Gapped BLAST and PSI-BLAST: a new generation of protein database search programs. Nucleic Acids Res 25: 33893402.
  • Amann, R.I., Binder, B.J., Olson, R.J., Chisholm, S.W., Devereux, R., and Stahl, D.A. (1990) Combination of 16S ribosomal RNA-targeted oligonucleotide probes with flow-cytometry for analyzing mixed microbial populations. Appl Environ Microbiol 56: 19191925.
  • Brock, T.D. (1966) The habitat of Leucothrix mucor, a widespread marine microorganism. Limnol Oceanogr 11: 303307.
  • Bulgheresi, S., Schabussova, I., Chen, T., Mullin, N.P., Maizels, R.M., and Ott, J.A. (2006) A new C-type lectin similar to the human immunoreceptor DC-SIGN mediates symbiont acquisition by a marine nematode. Appl Environ Microbiol 72: 29502956.
  • Casanova, B., Brunet, M., and Segonzac, M. (1993) L'impact d'une épibiose bactérienne sur la morphologie fonctionelle de crevettes associées à l'hydrothermalisme médio-atlantique. Cah Biol Mar 34: 573588.
  • Cavanaugh, C.M., and Robinson, J.J. (1996) CO2 fixation in chemoautotrophic-invertebrate symbioses: expression of form I and form II RubisCO. In Microbial Growth on C1 Compounds. Lidstrom, M.E., and Tabita, F.R. (eds). Dordrecht, the Netherlands: Kluwer, pp. 285292.
  • Cavanaugh, C.M., McKiness, Z.P., Newton, I.L.G., and Stewart, F.J. (2006) Marine chemosynthetic symbioses. In The Prokaryotes: An Evolving Electronic Resource for the Microbial Community. Dworkin, M., Falkow, S.I., Rosenberg, E., Schleifer, K.-H., and Stackebrandt, E. (eds). New York, USA: Springer, pp. 475507.
  • Charlou, J.L., Donval, J.P., Fouquet, Y., Jean-Baptiste, P., and Holm, N. (2002) Geochemistry of high H2 and CH4 vent fluids issuing from ultramafic rocks at the Rainbow hydrothermal field (36°14′N, MAR). Chem Geol 191: 345359.
  • Chu, K.H., Li, C.P., and Ho, H.Y. (2001) The first internal transcribed spacer (ITS-1) of ribosomal DNA as a molecular marker for phylogenetic and population analyses in Crustacea. Mar Biotechnol 3: 355361.
  • Corbari, L., Zbinden, M., Cambon-Bonavita, M.A., Gaill, F., and Compere, P. (2007) Bacterial symbionts and mineral deposits in the branchial chamber of the hydrothermal vent shrimp Rimicaris exoculata: relationship to moult cycle. Aquat Biol 1: 225238.
  • Creasey, S., Rogers, A.D., and Tyler, P.A. (1996) Genetic comparison of two populations of the deep-sea vent shrimp Rimicaris exoculata (Decapoda: Bresiliidae) from the Mid-Atlantic Ridge. Mar Biol 125: 473482.
  • Daims, H., Bruhl, A., Amann, R., Schleifer, K.H., and Wagner, M. (1999) The domain-specific probe EUB338 is insufficient for the detection of all Bacteria: development and evaluation of a more comprehensive probe set. Syst Appl Microbiol 22: 434444.
  • DeChaine, E.G., Bates, A.E., Shank, T.M., and Cavanaugh, C.M. (2006) Off-axis symbiosis found: characterization and biogeography of bacterial symbionts of Bathymodiolus mussels from Lost City hydrothermal vents. Environ Microbiol 8: 19021912.
  • Diaz, C.L., Melchers, L.S., Hooykaas, P.J.J., Lugtenberg, B.J.J., and Kijne, J.W. (1989) Root lectin as a determinant of host-plant specificity in the Rhizobium-legume symbiosis. Nature 338: 579581.
  • Dubilier, N., Bergin, C., and Lott, C. (2008) Symbiotic diversity in marine animals: the art of harnessing chemosynthesis. Nat Rev Microbiol 6: 725739.
  • Duperron, S., Bergin, C., Zielinski, F., Blazejak, A., Pernthaler, A., McKiness, Z.P., et al. (2006) A dual symbiosis shared by two mussel species, Bathymodiolus azoricus and Bathymodiolus puteoserpentis (Bivalvia: Mytilidae), from hydrothermal vents along the northern Mid-Atlantic Ridge. Environ Microbiol 8: 14411447.
  • Folmer, O., Black, M.B., Hoeh, W.R., Lutz, R.A., and Vrijenhoek, R.C. (1994) DNA primers for amplification of mitochondrial cytochrome C oxidase subunit I from metazoan invertebrates. Mol Mar Biol Biotechnol 3: 294299.
  • Galchenko, V.F., Pimenov, N.V., Lein, A.I., Galkin, S.V., Moskalev, L.I., and Ivanov, M.V. (1989) Autotrophic CO2 assimilation in tissues of prawn Rimicaris exoculata from the Mid-Atlantic Ridge hydrothermal area. Doklady Akademii Nauk SSSR 308: 14781481.
  • Gebruk, A.V., Pimenov, N.V., and Savvichev, A.S. (1993) Feeding specialization of bresiliid shrimps in the TAG site hydrothermal community. Mar Ecol Prog Ser 98: 247253.
  • Gebruk, A.V., Southward, E.C., Kennedy, H., and Southward, A.J. (2000) Food sources, behaviour, and distribution of hydrothermal vent shrimps at the Mid-Atlantic Ridge. J Mar Biol Assoc UK 80: 485499.
  • Gloter, A., Zbinden, M., Guyot, F., Gaill, F., and Colliex, C. (2004) TEM-EELS study of natural ferrihydrite from geological-biological interactions in hydrothermal systems. Earth Planet Sci Lett 222: 947957.
  • Goffredi, S.K., Waren, A., Orphan, V.J., Van Dover, C.L., and Vrijenhoek, R.C. (2004) Novel forms of structural integration between microbes and a hydrothermal vent gastropod from the Indian Ocean. Appl Environ Microbiol 70: 30823090.
  • Goffredi, S.K., Jones, W.J., Erhlich, H., Springer, A., and Vrijenhoek, R.C. (2008) Epibiotic bacteria associated with the recently discovered Yeti crab, Kiwa hirsuta. Environ Microbiol 10: 26232634.
  • Gourdine, J.P., Markiv, A., and Smith-Ravin, J. (2007) The three-dimensional structure of codakine and related marine C-type lectins. Fish Shellfish Immunol 23: 831839.
  • Grabovich, M.Y., Muntyan, M.S., Lebedeva, V.Y., Ustiyan, V.S., and Dubinina, G.A. (1999) Lithoheterotrophic growth and electron transfer chain components of the filamentous gliding bacterium Leucothrix mucor DSM 2157 during oxidation of sulfur compounds. FEMS Microbiol Lett 178: 155161.
  • Gros, O., Liberge, M., Heddi, A., Khatchadourian, C., and Felbeck, H. (2003) Detection of the free-living forms of sulfide-oxidizing gill endosymbionts in the lucinid habitat (Thalassia testudinum environment). Appl Environ Microbiol 69: 62646267.
  • Hall, T. (19972001) BioEdit. [WWW document]. URL http://www.mbio.ncsu.edu/BioEdit/bioedit.html
  • Harmer, T.L., Rotjan, R.D., Nussbaumer, A.D., Bright, M., Ng, A.W., DeChaine, E.G., and Cavanaugh, C.M. (2008) Free-living tube worm endosymbionts found at deep-sea vents. Appl Environ Microbiol 74: 38953898.
  • Harrison, J.S. (2001) Phylogeny, biogeography, and speciation in the genus Austinixa and related genera (Crustacea: Brachyura: Pinnotheridae). PhD Thesis. College Station, TX, USA: Texas A&M University.
  • Haselkorn, T.S., Markow, T.A., and Moran, N.A. (2009) Multiple introductions of the Spiroplasma bacterial endosymbiont into Drosophila. Mol Ecol 18: 12941305.
  • Hunter, R.L., Webb, M.S., Iliffe, T.M., and Bremer, J.R.A. (2008) Phylogeny and historical biogeography of the cave-adapted shrimp genus Typhlatya (Atyidae) in the Caribbean Sea and Western Atlantic. J Biogeogr 35: 6575.
  • Inagaki, F., Takai, K., Nealson, K.H., and Horikoshi, K. (2004) Sulfurovum lithotrophicum gen. nov., sp. nov., a novel sulfur-oxidizing chemolithoautotroph within the Epsilonproteobacteria isolated from Okinawa Trough hydrothermal sediments. Int J Syst Evol Microbiol 54: 14771482.
  • Jannasch, H.W., Wirsen, C.O., and Molyneaux, S.J. (1991) Chemosynthetic microbial activity at the 23° and 26°N Mid-Atlantic Ridge vent sites. Ridge Events 19.
  • Kimura, H., Sato, M., Sasayama, Y., and Naganuma, T. (2003) Molecular characterization and in situ localization of endosymbiotic 16S ribosomal RNA and RuBisCO genes in the pogonophoran tissue. Mar Biotechnol 5: 261269.
  • Komai, T., and Segonzac, M. (2008) Taxonomic review of the hydrothermal vent shrimp genera Rimicaris Williams & Rona and Chorocaris Martin & Hessler (Crustacea: Decapoda: Caridea: Alvinocarididae). J Shellfish Res 27: 2141.
  • Lee, K.H., and Ruby, E.G. (1994) Effect of the squid host on the abundance and distribution of symbiotic Vibrio fischeri in nature. Appl Environ Microbiol 60: 15651571.
  • Lösekann, T., Robador, A., Niemann, H., Knittel, K., Boetius, A., and Dubilier, N. (2008) Endosymbioses between bacteria and deep-sea siboglinid tubeworms from an Arctic Cold Seep (Haakon Mosby Mud Volcano, Barents Sea). Environ Microbiol 10: 32373254.
  • Ludwig, W., Strunk, O., Westram, R., Richter, L., Meier, H., Yadhukumar, et al. (2004) ARB: a software environment for sequence data. Nucleic Acids Res 32: 13631371.
  • Mandel, M.J., Wollenberg, M.S., Stabb, E.V., Visick, K.L., and Ruby, E.G. (2009) A single regulatory gene is sufficient to alter bacterial host range. Nature 458: 215217.
  • Meng, X.H., Wang, Q.Y., Jang, I.K., Liu, P., and Kong, J. (2009) Genetic differentiation in seven geographic populations of the fleshy shrimp Penaeus (Fenneropenaeus) chinensis based on microsatellite DNA. Aquaculture 287: 4651.
  • Miethling, R., Wieland, G., Backhaus, H., and Tebbe, C.C. (2000) Variation of microbial rhizosphere communities in response to crop species, soil origin, and inoculation with Sinorhizobium meliloti L33. Microb Ecol 40: 4356.
  • Muyzer, G., Teske, A., Wirsen, C.O., and Jannash, H.W. (1995) Phylogenetic relationships of Thiomicrospira species and their identification in deep-sea hydrothermal vent samples by denaturing gradient gel electrophoresis of 16S rDNA fragments. Arch Microbiol 164: 165172.
  • Nyholm, S.V., and McFall-Ngai, M.J. (2004) The winnowing: establishing the squid-Vibrio symbiosis. Nat Rev Microbiol 2: 632642.
  • Page, T.J., Short, J.W., Humphrey, C.L., Hillyer, M.J., and Hughes, J.M. (2008) Molecular systematics of the Kakaducarididae (Crustacea: Decapoda: Caridea). Mol Phylogenet Evol 46: 10031014.
  • Perner, M., Kuever, J., Seifert, R., Pape, T., Koschinsky, A., Schmidt, K., et al. (2007) The influence of ultramafic rocks on microbial communities at the Logatchev hydrothermal field, located 15°N on the Mid-Atlantic Ridge. FEMS Microbiol Ecol 61: 97109.
  • Pimenov, N.V., Kalyuzhnaya, M.G., Khmelenina, V.N., Mityushina, L.L., and Trotsenko, Y.A. (2002) Utilization of methane and carbon dioxide by symbiotrophic bacteria in gills of Mytilidae (Bathymodiolus) from the Rainbow and Logachev hydrothermal fields on the Mid-Atlantic Ridge. Microbiol 71: 587594.
  • Polz, M.F., and Cavanaugh, C.M. (1995) Dominance of one bacterial phylotype at a Mid-Atlantic Ridge hydrothermal vent site. Proc Natl Acad Sci USA 92: 72327236.
  • Polz, M.F., Robinson, J.J., Cavanaugh, C.M., and Van Dover, C.L. (1998) Trophic ecology of massive shrimp aggregations at a Mid-Atlantic Ridge hydrothermal vent site. Limnol Oceanogr 43: 16311638.
  • Pond, D., Dixon, D., and Sargent, J. (1997) Wax-ester reserves facilitate dispersal of hydrothermal vent shrimps. Mar Ecol Prog Ser 146: 289290.
  • Pond, D.W., Gebruk, A., Southward, E.C., Southward, A.J., Fallick, A.E., Bell, M.V., and Sargent, J.R. (2000) Unusual fatty acid composition of storage lipids in the bresilioid shrimp Rimicaris exoculata couples the photic zone with MAR hydrothermal vent sites. Mar Ecol Prog Ser 198: 171179.
  • Pruesse, E., Quast, C., Knittel, K., Fuchs, B.M., Ludwig, W.G., Peplies, J., and Glockner, F.O. (2007) SILVA: a comprehensive online resource for quality checked and aligned ribosomal RNA sequence data compatible with ARB. Nucleic Acids Res 35: 71887196.
  • Rieley, G., Van Dover, C.L., Hedrick, D.B., and Eglinton, G. (1999) Trophic ecology of Rimicaris exoculata: a combined lipid abundance stable isotope approach. Mar Biol 133: 495499.
  • Robinson, J.J., Polz, M.F., Fiala-Medioni, A., and Cavanaugh, C.M. (1998) Physiological and immunological evidence for two distinct C1-utilizing pathways in Bathymodiolus puteoserpentis (Bivalvia: Mytilidae), a dual endosymbiotic mussel from the Mid-Atlantic Ridge. Mar Biol 132: 625633.
  • Ruby, E.G. (2008) Symbiotic conversations are revealed under genetic interrogation. Nat Rev Microbiol 6: 752762.
  • Schmidt, K., Koschinsky, A., Garbe-Schonberg, D., De Carvalho, L.M., and Seifert, R. (2007) Geochemistry of hydrothermal fluids from the ultramafic-hosted Logatchev hydrothermal field, 15°N on the Mid-Atlantic Ridge: temporal and spatial investigation. Chem Geol 242: 121.
  • Schmidt, C., Le Bris, N., and Gaill, F. (2008a) Interactions of deep-sea vent invertebrates with their environment: the case of Rimicaris exoculata. J Shellfish Res 27: 7990.
  • Schmidt, C., Vuillemin, R., Le Gall, C., Gaill, F., and Le Bris, N. (2008b) Geochemical energy sources for microbial primary production in the environment of hydrothermal vent shrimps. Mar Chem 108: 1831.
  • Segonzac, M., Desaintlaurent, M., and Casanova, B. (1993) Enigma of the trophic adaptation of the shrimp Alvinocarididae in hydrothermal areas along the Mid-Atlantic Ridge. Cah Biol Mar 34: 535571.
  • Shank, T.M., Lutz, R.A., and Vrijenhoek, R.C. (1998) Molecular systematics of shrimp (Decapoda: Bresiliidae) from deep-sea hydrothermal vents, I: enigmatic ‘small orange’ shrimp from the Mid-Atlantic Ridge are juvenile Rimicaris exoculata. Mol Mar Biol Biotechnol 7: 8896.
  • Southward, A.J., and Newman, W.A. (1998) Ectosymbiosis between filamentous sulphur bacteria and a stalked barnacle (Scalpellomorpha, Neolepadinae) from the Lau Back Arc Basin, Tonga. Cah Biol Mar 39: 259262.
  • Steedman, H.F. (1957) Polyester wax – new ribboning embedding medium for histology. Nature 179: 13451345.
  • Suzuki, Y., Suzuki, M., Tsuchida, S., Takai, K., Horikoshi, K., Southward, A.J., et al. (2009) Molecular investigations of the stalked barnacle Vulcanolepas osheai and the epibiotic bacteria from the Brothers Caldera, Kermadec Arc, New Zealand. J Mar Biol Assoc UK 89: 727733.
  • Teske, P.R., Froneman, P.W., Barker, N.P., and McQuaid, C.D. (2007) Phylogeographic structure of the caridean shrimp Palaemon peringueyi in South Africa: further evidence for intraspecific genetic units associated with marine biogeographic provinces. Afr J Mar Sci 29: 253258.
  • Tyler, P.A., and Young, C.M. (1999) Reproduction and dispersal at vents and cold seeps. J Mar Biol Assoc UK 79: 193208.
  • Tyler, P.A., and Young, C.M. (2003) Dispersal at hydrothermal vents: a summary of recent progress. Hydrobiologia 503: 919.
  • Van Dover, C.L. (2000) The Ecology of Deep-Sea Hydrothermal Vents. Princeton, NJ, USA: Princeton University Press.
  • Van Dover, C.L., Fry, B., Grassle, J.F., Humphris, S., and Rona, P.A. (1988) Feeding biology of the shrimp Rimicaris exoculata at hydrothermal vents on the Mid-Atlantic Ridge. Mar Biol 98: 209216.
  • Van Dover, C.L., German, C.R., Speer, K.G., Parson, L.M., and Vrijenhoek, R.C. (2002) Marine biology – evolution and biogeography of deep-sea vent and seep invertebrates. Science 295: 12531257.
  • Vrijenhoek, R.C. (1997) Gene flow and genetic diversity in naturally fragmented metapopulations of deep-sea hydrothermal vent animals. J Hered 88: 285293.
  • Williams, A.B., and Rona, P.A. (1986) Two new caridean shrimps (Bresiliidae) from a hydrothermal field on the Mid-Atlantic Ridge. J Crustacean Biol 6: 446462.
  • Wirsen, C.O., Jannasch, H.W., and Molyneaux, S.J. (1993) Chemosynthetic microbial sctivity at Mid-Atlantic Ridge hydrothermal vent sites. J Geophys Res 98: 96939703.
  • Zbinden, M., and Cambon-Bonavita, M.A. (2003) Occurrence of Deferribacterales and Entomoplasmatales in the deep-sea alvinocarid shrimp Rimicaris exoculata gut. FEMS Microbiol Ecol 46: 2330.
  • Zbinden, M., Le Bris, N., Gaill, F., and Compere, P. (2004) Distribution of bacteria and associated minerals in the gill chamber of the vent shrimp Rimicaris exoculata and related biogeochemical processes. Mar Ecol Prog Ser 284: 237251.
  • Zbinden, M., Shillito, B., Le Bris, N., De Montlaur, C.D., Roussel, E., Guyot, F., et al. (2008) New insigths on the metabolic diversity among the epibiotic microbial communitiy of the hydrothermal shrimp Rimicaris exoculata. J Exp Mar Biol Ecol 359: 131140.

Supporting Information

  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgements
  8. References
  9. Supporting Information

Fig. S1. Parsimony network of host cytochrome b genes. Parsimony network inferred from 484 positions of the cytochrome b gene of Rimicaris exoculata from Rainbow (red), TAG (green), Logatchev (orange) and South MAR (blue). Each circle represents one sequence (individual number listed in circles), except for Individuals 1 and 12, which were identical. Lines connecting genotypes are one nucleotide difference, small white circles represent hypothetical unsampled ancestors. Calculated with the TCS program (Clement et al., 2000).

Fig. S2. A. Fluorescence in situ hybridization of R. exoculata Individual 1 from the Rainbow vent field with the Epsilon 1 probe (pink) showing the abundance of the Epsilon 1 symbiont. The Gamma symbiont is shown in green (hybridized with the probe Rexogam1268RT).

B. FISH of the same individual with the Epsilon 3 probe (pink) showing the abundance of the Epsilon 4 symbiont. The Gamma symbiont is shown as above.

C. FISH of the same individual with the RexoepsLG86 (5′-ctcgtcagccagtgc-3′) with competitor (5′-ctcgtcagccagtac-3′), showing that the signals in (B) are due to the presence of the Epsilon 4 symbiont, and not the Logatchev Epsilon 3. DAPI stain is shown in blue in all 3 images. The scale bar in (C) (20 μm) applies to all images.

Table S1. Summary of sampling sites and cruises.

Table S2. FISH probes designed for this study.

Table S3. Individual numbers used for sampling, and corresponding name used in text and figures.

Please note: Wiley-Blackwell are not responsible for the content or functionality of any supporting materials supplied by the authors. Any queries (other than missing material) should be directed to the corresponding author for the article.

FilenameFormatSizeDescription
EMI_2129_sm_Figure_legend.doc27KSupporting info item
EMI_2129_sm_Fig_S1.tif170KSupporting info item
EMI_2129_sm_Fig_S2.tif6702KSupporting info item
EMI_2129_sm_Table_S1-S3.doc117KSupporting info item

Please note: Wiley Blackwell is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.