Editor: Patricia Sobecky
Lack of endosymbiont release by two Lucinidae (Bivalvia) of the genus Codakia: consequences for symbiotic relationships
Article first published online: 16 DEC 2008
Journal compilation © 2008 Federation of European Microbiological Societies. Published by Blackwell Publishing Ltd. No claim to original French government works
FEMS Microbiology Ecology
Volume 67, Issue 2, pages 261–267, February 2009
How to Cite
Brissac, T., Gros, O. and Merçot, H. (2009), Lack of endosymbiont release by two Lucinidae (Bivalvia) of the genus Codakia: consequences for symbiotic relationships. FEMS Microbiology Ecology, 67: 261–267. doi: 10.1111/j.1574-6941.2008.00626.x
- Issue published online: 22 DEC 2008
- Article first published online: 16 DEC 2008
- Received 26 June 2008; revised 26 September 2008; accepted 27 October 2008.First published online 16 December 2008.
- sulfur-oxidizing gill endosymbiont;
- transmission mode
Associations between marine invertebrates and chemoautotrophic bacteria constitute a wide field for the study of symbiotic associations. In these interactions, symbiont transmission must represent the cornerstone allowing the persistence of the association throughout generations. Within Bivalvia, in families such as Solemyidae or Vesicomyidae, symbiont transmission is undoubtedly vertical. However, in Lucinidae, symbiont transmission is described in the literature as ‘environmental’, symbionts being acquired from the environment by the new host generations. Hence, if there is transmission, symbionts should be transmitted from adults to juveniles via the environment. Consequently, we should observe a release of the symbiont by adults. We attempted to detect such a release within two Lucinidae species of the genus Codakia. We sampled 10 Codakia orbicularis and 20 Codakia orbiculata distributed in 10 crystallizing dishes containing filtered seawater. During 1 month of investigation, we analyzed water of the dishes in order to detect any release of a symbiont using catalyzed report deposition-FISH techniques. For 140 observations realized during this period, we did not observe any release of symbionts. This suggests that the idea of host-to-host passage in Lucinidae is inaccurate. We could therefore consider that the transmission mode from generation to generation does not occur within Lucinidae, symbiosis appearing to be advantageous in this case only for the host, and constitutes an evolutionary dead-end for the bacteria.
Since their discovery in hydrothermal vents (Felbeck et al., 1981), associations between marine invertebrates and chemoautotrophic bacteria have been described in more accessible environments such as shallow water sea-grass beds (Berg & Alatalo, 1984), mangrove swamps (Schweimanns & Felbeck, 1985), cold seeps (Juniper & Sibuet, 1987), whale bones (Deming et al., 1997) and sunken woods (Gros & Gaill, 2007; Duperron et al., 2008). These associations are present all over the world in many host taxa as an endo- or an ecto-symbiosis (Gill & Vogel, 1962; Polz et al., 1992; Polz & Cavanaugh, 1995; Haygood & Davidson, 1997; Zbinden et al., 2004). According to 16S rRNA gene sequences, the majority of marine symbionts known to date belong to the Gammaproteobacteria subdivision (Distel et al., 1994).
One of these associations involves Lucinidae (Mollusca: Bivalvia) and sulfur-oxidizing (thioautotrophic) intracellular symbionts (Distel et al., 1988, 1994; Durand & Gros, 1996; Williams et al., 2003). Lucinidae are the most diverse bivalve family, live in a wide variety of marine habitats and are geographically very widespread (Taylor & Glover, 2005). All Lucinidae genera studied to date harbor endosymbiotic thioautotrophic bacteria localized in specialized cells of the gill called bacteriocytes (Fisher, 1990).
As in all endosymbioses, symbiont transmission must constitute the most important factor enabling symbiotic associations to continue. Such transmission can be (1) vertical (from parents to offspring), which implies the integration of the symbiont into the gametes, (2) horizontal (spread of symbionts between contemporary hosts) or (3) environmental. The latter term has been proposed to define the reinfection of the new generation by a stock of microorganisms present in the environment without, however, discussion of the origin of this environmental stock (Le Pennec et al., 1988; Cary & Giovannoni, 1993).
In Lucinidae, no symbionts in ovaries, eggs, embryos and larvae of Codakia orbicularis have been described (Gros et al., 1996), which eliminates the vertical transmission hypothesis. However, aposymbiotic juveniles of C. orbicularis can be infected in the laboratory with crude sand from natural habitats of C. orbicularis (Gros et al., 1996). Later, free-living forms of the C. orbicularis gill endosymbiont were detected in Thalassia testudinum sea-grass bed sediments (Gros et al., 2003). These observations led the authors to conclude that Lucinidae sulfur-oxidizing endosymbionts were environmentally transmitted.
However, environmental transmission requires the passage of the symbiont from an infected host (i.e. adults) to a noninfected host (i.e. aposymbiotic juveniles) via the environment (Fig. 1), a phenomenon that has not yet been described in Lucinidae. Consequently, if transmission occurs, a release of the symbiont by adults in the environment must exist. In this study, we attempted to find it.
Materials and methods
Approximately 40 mm shell length individuals of C. orbicularis (Linné, 1758) and 15 mm shell length Codakia orbiculata (Montagu, 1808) individuals were collected by hand in March 2007 from T. testudinum sea-grass sediments on îlet Cochon (16°12′53.76″N; 61°32′05.74″W) in Guadeloupe (French West Indies).
Clams were divided into 10 crystallizing dishes containing 500 mL of 0.2 μm filtered seawater according to the following distribution: two crystallizing dishes contained a single C. orbicularis; four crystallizing dishes contained two C. orbicularis each; and four crystallizing dishes contained five to six C. orbiculata each.
To avoid symbiont reacquisition due to bivalve filtration, individuals were kept raised halfway up the dishes. No sulfur, food or oxygen was added to the dishes. Every day, ‘old’ seawater was discarded and replaced by ‘fresh’ 0.2 μm filtered seawater. A positive control (0.2 μm filtered seawater containing C. orbicularis gill extract) was also realized in order to check whether symbionts could be detected by our analysis technique.
Various studies have used fluorescent probes to detect environmental bacteria or bacterial symbionts in the environment (Christensen et al., 1999; DeLong et al., 1999; Manz et al., 2000; Gros et al., 2003). Here, as a positive control, we used the universal probe EUB338, which targets 16S rRNA gene of most eubacteria (Amann et al., 1990; Daims et al., 1999), and the specific probe Symco2A, which hybridizes in a specific manner the 16S rRNA gene of the symbiont of C. orbicularis (Gros et al., 1996) and of C. orbiculata, which harbor a symbiont with the same 16S rRNA gene sequence (Durand & Gros, 1996).
Catalyzed report deposition (CARD)-FISH
Histological sections or pieces of filters were used for hybridization using the following probes: EUB338 to target all eubacteria and Symco2A for C. orbicularis or C. orbiculata gill endosymbiont. This protocol follows Pernthaler et al. (2002) up to the washing buffer step. Sections were then washed twice in 1 × phosphate-buffered solution (PBS) [15 min at room temperature (RT)] and incubated in a solution of amplification buffer (1/10 [w/v] dextrane sulfate, 1 × PBS, 2 M NaCl and 1/100 [v/v] 10% blocking reagent), 0.5% diluted H2O2 in 1 × PBS and 2 μL of fluorescent-labeled tyramide for 15 min at 37 °C in the dark. Sections were then washed in 1 × PBS (5 min at RT in the dark), in sterilized water and twice in 50 mL of absolute ethanol. Sections were air-dried, 4′,6-diamino-2-phenylindole (DAPI) counterstained and embedded in DAKO medium before observation under an epi80i fluorescent microscope (NIKON, France).
CARD-FISH detection threshold
We purified a gill extract on percoll, following Caro et al. (2007), and quantified its bacterial concentration by counting on a Malassez cell. We then used this inoculum to constitute samples containing 0, 10, 102, 103, 104, 105, 106 and 107 bacteria mL−1. These samples were treated and analyzed in a manner similar to those used for Symbiont release detection experiments.
Symbiont release detection
The experiment was conducted over 30 days. To investigate the release of symbionts, 50 mL of seawater from the 500-mL experimental batches were sampled and analyzed at T0, every 4 h the first day, once a day until day 5 and then every 5 days (for a total of 140 observations). Every 50-mL sample was fixed in 4% PFA for 3 h at 4 °C and filtered on GTTP filters (Millipore, France). Each section of the filter (1/4 of filter) was analyzed directly or stored at −80 °C before analysis.
Monitoring of the symbiotic population in the gills by CARD-FISH on slides
To follow the evolution of the symbiotic population in the gills, individuals were sacrificed at T0, T0+10 days, T0+20 days and T0+30 days, and gill sections were CARD-FISH analyzed. Slides were prepared following Dubilier et al. (1995), with some modifications: after embedding in paraffin, 4 μm sections were realized and pasted on SuperFrost® slides (Menzel-Glaser®). Slides were used immediately or stored at −20 °C. Sections were then deparaffinized in toluene and rehydrated (100%, 95%, 70% ethanol and distilled water). Slides were washed in 0.2 M HCl (12 min at RT), Tris/HCl 20 mM (10 min at RT), Proteinase K (0.5 μg mL−1 of Tris/HCl 20 mM; 5 min at 37 °C) and finally in Tris/HCl 20 mM for 10 min at RT. Slides were then directly used for CARD-FISH.
Monitoring of the symbiotic population in the gills by transmission electron microscopy (TEM)
After dissection, gill samples were prefixed in 2.5% glutaraldehyde/0.1 M cacodylate buffer (pH 7.2 isosmotic to seawater by addition of sodium chloride and CaCl2) for 2 h, and washed three times in the same isosmotic buffer. Pieces of gill were fixed for a further hour at RT in 1% osmium tetroxide in the same buffer, and then rinsed in distilled water before postfixation, using 2% aqueous uranyl acetate (1 h), dehydration through an ethanol series and propylene oxide before embedding in an Epon–Araldite mixture. Thin sections, 60 nm thick, were cut for TEM observation. They were processed and stained by uranyl acetate and lead citrate using standard TEM methods and observed under a Leo 912 TEM microscope.
The purpose of this study was to check whether sulfur-oxidizing gill endosymbionts of two tropical lucinids belonging to the genus Codakia (C. orbicularis and C. orbiculata) are released by their hosts in the environment. To answer this question, we attempted to detect these non cultivable thioautotrophic bacteria using a specific fluorescent probe targeting 16S rRNA gene sequences from a controlled environment.
CARD-FISH detection level
To ensure that the lack of symbiont detection was not due to a lack of sensitivity of the technique used, we determined the minimal number of bacteria necessary to observe a signal using 0.2 μm filtered seawater containing a purified fraction of C. orbicularis gill endosymbionts extracted from a freshly collected individual of C. orbicularis (Caro et al., 2007) as a positive control. The EUB338 (Fig. 2b) and Symco2A (Fig. 2c) probes were sufficiently effective in detecting at least 10 bacteria mL−1 (minimal density analyzed); the negative control (0.2 μm filtered seawater without a gill extract) remained negative for the two probes (data not shown).
Filters of the different sampling points were hybridized with EUB338 and Symco2A probes. For all the 140 samples analyzed, no positive hybridization with Symco2A was found, whether early [every 4 h without seawater change (T0 to T0+16 h)] or late in the experiment [seawater change every day (T0 to T0+30 days)] (Fig. 2f). In all cases, the internal positive control of hybridization (EUB338) was positive, confirming that bacteria could be detected using our technique (Fig. 2e). A DAPI counterstaining was used to confirm that positive hybridizations represented bacteria, and to eliminate auto fluorescent particles from the sediments (Fig. 2a and d). No host mortality was observed during the experiment.
Evolution of the symbiotic population inside gills
Gill sections of C. orbicularis and C. orbiculata at 0, 10, 20 or 30 days of starvation were hybridized with EUB338 and Symco2A. In both cases, probes hybridized only in the lateral zone of the gill where bacteriocytes are localized. Based on microscopic observations, sections of T0 (Fig. 3a), T0+10 days and T0+20 days (data not shown) revealed no hybridization differences (Fig. 3a). However, on the T0+30 days (Fig. 3b), the sample presented a weaker hybridization signal.
Through the lateral zone of each gill filament, bacteriocytes correspond to large cells (up to 35 μm length) characterized by a rounded apical pole in contact with circulating seawater and a cytoplasm filled with individually enclosed bacteria (Fig. 4a). Bacterial endosymbionts, which could reach 5 μm in length (Fig. 4a), are characterized by numerous sulfur granules (appearing as electron-lucent vesicles after conventional TEM preparation) located in the periplasmic space as described recently using cryo-EFTEM (Lechaire et al., 2008). After 1 month of starvation, modifications could be noted in the gill filaments observed, with a decrease of intracellular bacteria and the appearance of numerous large secondary lysosomes (Fig. 4b), indicating an increase of intracellular digestion of the symbionts.
Symbiont transmission must constitute the most important factor permitting the persistence of symbiotic associations. Within Lucinidae (Mollusca: Bivalvia), associated with the sulfur-oxidizing gill endosymbiont, previous studies (1) revealed an absence of symbionts in the ovaries, eggs, embryos and larvae of various lucinids, (2) showed that juveniles can be infected with crude sand from T. testudinum sea-grass beds and (3) detected the free-living form of the symbiont in these sea-grass beds (Gros et al., 1996, 1998, 1999, 2003). Accordingly, the authors of these studies excluded the possibility of a vertical transmission of symbionts in Lucinidae and concluded that transmission was environmental. Nevertheless, environmental transmission being a passage from an infected host (here adults) to a noninfected host (here juveniles) via the environment, it would require a release of the symbiosis-competent bacteria by the adult clams (Fig. 1). In the present study, using CARD-FISH techniques, we attempted to detect the release of thioautotrophic symbionts by the lucinids C. orbicularis and C. orbiculata and we did not find any evidence of such a release over 1 month of experiment.
No evidence of symbiont release
What hypotheses could explain this result? (1) Our experimental conditions might not be optimal enough to allow symbiont release to occur, although rearing conditions kept the hosts alive in an apparently good condition for at least a month (not a single host died during the whole experiment). (2) The natural release of symbionts could be restricted to a defined period at the end of the larval development after metamorphosis (July–September), to permit a more efficient transmission. The failure to detect a release could then be because our experiment was conducted in April. However, we can infect juveniles with crude sand in the laboratory all the year round and whatever the age of the juvenile (O. Gros, personal observations). These observations suggest that the release of symbionts, if it does take place, could occur efficiently all throughout the year (in sea-grass beds), which makes the ‘period-restricted release’ hypothesis unlikely. (3) The release took place, but was not detected. Considering the sensitivity of the method we used, this is very unlikely because some batches included several individuals and sampling was conducted 24 h after filtered seawater exchange. Moreover, the 30 days of the experiment provided ample opportunities for release to take place, if it really existed. (4) Release could occur at the death of the host. Nevertheless, a transmission strategy based on the death of the host can be successful only in the case of parasitic microorganisms (Ewald, 1987). Now, sulfur-oxidizing endosymbionts associated with marine invertebrates were not considered as deleterious (Stewart et al., 2005). Finally (5), the symbiont release phenomenon does not exist and consequently the environmental transmission hypothesis proposed by Gros et al. (1996) could be invalid.
Evolution of the symbiotic population into the gills
The diminution of the hybridization signal observed on gill sections must be considered because it could reveal a release undetected by CARD-FISH. However, this hypothesis is unlikely because the diminution of the hybridization signal seems large enough to correspond to a release we should have been able to detect. Considering that bivalve gills could harbor c. 3 × 109 bacteria g−1 of gills, and considering that the symbiont population decreases up to 30% per month (A. Caro et al., unpublished data), it may have a release of 3 × 107 bacteria per clam day−1. Considering our detection limit of 10 bacteria mL−1 (i.e. 5000 bacteria per crystallizing dish), this release could not have gone undetected. Alternative explanations for the decrease of the hybridization signal are (1) decreasing metabolic activity due to experimental conditions (related with rRNA copy) or (2) lysosomal degradation of the symbionts into the gills. We favor the latter hypothesis because our TEM observations show a reduced number of bacteria in the bacteriocytes and an increased number of secondary lysosome-like structures (Fig. 4b) similar to those described in the literature (Frenkiel & Mouëza, 1995; Frenkiel et al., 1996; Liberge et al., 2001).
Consequences for symbiotic relationships
Thus, the lack of release-detection in our experimental conditions, although not refuted completely, does not support the environmental transmission mode. However, the absence of vertical transmission, the ability to infect with crude sand, the detection of the free-living form and the absence of symbiont release all agree with the hypothesis that host-to-host transmission does not take place within Lucinidae. If this is so, what consequence is there for the symbiotic relationship?
In chemoautotrophic symbioses, it is commonly supposed that both partners benefit from the symbiosis. On the host side, harboring a symbiont is very important considering the clearly established feeding role of bacteria in this type of symbiosis (Berg & Alatalo, 1984; Fisher, 1990). Indeed, symbiont carbon fixation covers an important part of the host's nutritional needs in chemoautotrophic endosymbioses (Stewart et al., 2005). On the symbiont side, gills would constitute a more favorable and more stable environment than seawater or sediments, and would provide a better access to metabolic substrates (e.g. sulfide, CO2) (Stewart et al., 2005). This pattern leads us to think that the relation is a mutualism. However, in Lucinidae, if the symbiont does not multiply in his host (Caro et al., 2007) and if release does not occur (as suggested by our results), juveniles acquire the bacteria only from a free-living form of the symbiont that multiplies only in the surroundings (Fig. 5). This ‘capture’ enables the host to benefit from the bacteria as described above. On the contrary, for the bacteria enclosed and exploited inside the gills, this relation constitutes an evolutionary dead-end.
TEM images were obtained at the ‘Service de Microscopie Électronique de l'IFR 83-Biologie Intégrative CNRS (Paris, France).’ We thank Dominique Higuet and Stéphane Ronsseray for their fruitful comments, and Malcolm Eden and Denis Poinsot for English revision. T.B. is funded as a PhD student by an MENRT (Ministère de l'Education Nationale, de la Recherche et des Technologies) grant.
- 1990) Combination of 16S rRNA-targeted oligonucleotide probes with flow cytometry for analyzing mixed microbial populations. Appl Environ Microb 56: 1919–1925. , , , , & (
- 1984) Potential of chemosynthesis in molluscan mariculture. Aquaculture 39: 165–179. & (
- 2007) Characterization of the population of the sulfur-oxidizing symbiont of Codakia orbicularis (Bivalvia, Lucinidae) by single-cell analyses. Appl Environ Microb 73: 2101–2109. , , , & (
- 1993) Transovarial inheritance of endosymbiotic bacteria in clams inhabiting deep-sea hydrothermal vents and cold seeps. P Natl Acad Sci USA 90: 5695–5699. & (
- 1999) Counting and size classification of active soil bacteria by fluorescence in situ hybridization with an rRNA oligonucleotide probe. Appl Environ Microb 65: 1753–1761. , & (
- 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: 434–444. , , , & (
- 1999) Visualization and enumeration of marine planktonic archaea and bacteria by using polyribonucleotide probes and fluorescent in situ hybridization. Appl Environ Microb 65: 5554–5563. , , & (
- 1997) Evidence for the microbial basis of a chemoautotrophic invertebrate community at a whale fall on the deep seafloor: bone-colonizing bacteria and invertebrate endosymbionts. Microsc Res Techiq 37: 162–170. , , & (
- 1988) Sulfur-oxidizing bacterial endosymbionts: analysis of phylogeny and specificity by 16S rRNA sequences. J Bacteriol 170: 2506–2510. , , , , , , & (
- 1994) Evidence for phylogenetic congruence among sulfur-oxidizing chemoautotrophic becterial endosymbiont and their Bivalve hosts. J Mol Evol 38: 533–542. , & (
- 1995) Characterization of chemoautotrophic bacterial symbionts in a gutless marine worm (Oligochaeta: Annelida) by phylogenetic 16S rRNA sequence analysis and in situ hybridization. Appl Environ Microb 61: 2346–2350. , , & (
- 2008) Sulphur-oxidizing extracellular bacteria in the gills of Mytilidae associated with wood falls. FEMS Microbiol Ecol 63: 338–349. , , & (
- 1996) Bacterial host specificity of Lucinacea endosymbionts: interspecific variation in 16S rRNA sequences. FEMS Microbiol Lett 140: 193–198. & (
- 1987) Transmission modes and evolution of the parasitism-mutualism continuum. Ann NY Acad Sci 503: 295–306. (
- 1981) Calvin–Benson cycle and sulphide oxidation enzymes in animals from sulphide-rich habitats. Nature 293: 291–293. , & (
- 1990) Chemoautotrophic and methanotrophic symbioses in marine invertebrates. Rev Aquat Sci 2: 399–613. (
- 1995) Gill ultrastructure and symbiotic bacteria in Codakia orbicularis (Bivalvia: Lucinidae). Zoomorphology 115: 51–61. & (
- 1996) Gill structure in Lucina pectinata (Bivalvia: Lucinidae) with reference to hemoglobin in bivalves with symbiotic sulphur-oxidizing bacteria. Mar Biol 125: 511–524. , & (
- 1962) Lysine synthesis and phylogeny: biochemical evidence for a bacterial-type endosymbiont in the protozoon Herpetomonas (Strigomonas) oncopelti. Biochim Biophys Acta 56: 200–201. & (
- 2007) Extracellular bacterial association in gills of “wood mussels”. Cah Biol Mar 48: 103–109. & (
- 1996) Environmental transmission of a sulfur-oxidizing bacterial gill endosymbiont in the tropical lucinid bivalve Codakia orbicularis. Appl Environ Microb 62: 2324–2330. , , , & (
- 1998) Putative environmental transmission of sulfur-oxidizing bacterial symbionts in tropical lucinid bivalves inhabiting various environments. FEMS Microbiol Lett 160: 257–262. , , & (
- 1999) Embryonic development and endosymbiont transmission mode in the symbiotic clam Lucinoma aequizonata (Bivalvia: Lucinidae). Invertebr Reprod Dev 36: 93–103. , & (
- 2003) Detection of the free-living forms of sulfide-oxidizing gill endosymbionts in the lucinid habitat (Thalassia testudinum environment). Appl Environ Microb 69: 6264–6267. , , , & (
- 1997) Small-subunit rRNA genes and in situ hybridization with oligonucleotides specific for the bacterial symbionts in the larvae of the bryozoan Bugula neritina and proposal of “Candidatus endobugula sertula”. Appl Environ Microb 63: 4612–4616. & (
- 1987) Cold seep benthic communities in Japan subduction zones: spatial organization, trophic strategies and evidence for temporal evolution. Mar Ecol Prog Ser 40: 115–126. & (
- 2008) In situ characterization of sulfur in gill-endosymbionts of the shallow water lucinid Codakia orbicularis (Linné, 1758) by cryo-EFTEM microanalysis. Mar Biol 154: 693–700. , , & (
- 1988) Endocytosis and lysis of bacteria in gill epithelium of Bathymodiolus thermophilus, Thyasira flexuosa, and Lucinella divaricata. (Bivalvia: Mollusca). J Shellfish Res 7: 483–489. , & (
- 2001) Lysosomes and sulfide-oxidizing bodies in the bacteriocytes of Lucina pectinata, a cytochemical and microanalysis approach. Mar Biol 139: 401–409. , & (
- 2000) Widefield deconvolution epifluorescence microscopy combined with fluorescence in situ hybridization reveals the spatial arrangement of bacteria in sponge tissue. J Microbiol Meth 40: 125–134. , , , & (
- 2002) Fluorescence in situ hybridization and catalyzed reporter deposition for the identification of marine bacteria. Appl Environ Microb 68: 3094–3101. , & (
- 1995) Dominance of one bacterial phylotype at a Mid-Atlantic Ridge hydrothermal vent site. P Natl Acad Sci USA 92: 7232–7236. & (
- 1992) Chemoautotrophic sulfur-oxidizing symbiotic bacteria on marine nematodes: morphological and biochemical characterization. Microbiol Ecol 24: 312–329. , , , & (
- 1985) Significance of the occurrence of chemoautotrophic bacterial endosymbionts in lucinid clams from Bermuda. Mar Ecol Prog Ser 24: 113–120. & (
- 2005) Chemosynthetic endosymbioses: adaptations to oxic–anoxic interfaces. Trends Microbiol 13: 439–448. , & (
- 2005) Lucinidae (Bivalvia) – the most diverse group of chemosymbiotic molluscs. Zool J Linn Soc 148: 421–438. & (
- 2003) Molecular phylogeny of the Lucinoidea (Bivalvia): non-monophyly and separate acquisition of bacterial chemosymbiosis. J Mollus Stud 70: 187–202. , & (
- 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: 237–251. , , & (