Experimentally induced endosymbiont loss and re ... induced... · Experimentally induced endosymbiont loss and re-acquirement in the hydrothermal vent bivalve Bathymodiolus azoricus
Post on 04-Jun-2018
217 Views
Preview:
Transcript
www.elsevier.com/locate/jembe
Journal of Experimental Marine Biolo
Experimentally induced endosymbiont loss and re-acquirement in
the hydrothermal vent bivalve Bathymodiolus azoricus
Enikf Kadara,T, Raul Bettencourta,b, Valentina Costaa, Ricardo Serrao Santosa,
Alexandre Lobo-da-Cunhac, Paul Dandod
aIMAR Centre of the University of Azores, Department of Oceanography and Fisheries, Rua Cais de Santa Cruz, 9900 Horta, PortugalbMedical School, University of Massachussets, Worcester, MA 01605, USA
cLaboratory of Cell Biology, Institute of Biomedical Sciences Abel Salazar, University of Porto, and CIIMAR, PortugaldSchool of Ocean Sciences, University of Wales-Bangor Anglesey, LL59 5AB, United Kingdom
Received 29 November 2004; received in revised form 2 December 2004; accepted 12 December 2004
Abstract
Invertebrates harbouring endosymbiotic chemoautotroph bacteria are widely distributed in a variety of reducing
marine habitats, including deep-sea hydrothermal vents. In these species mechanisms of symbiont transmission are likely
to be key elements of dispersal strategies that remained partially unresolved because the early life stages are not
available for developmental studies. To study cessation and re-establishment of symbiosis in the host gill a laboratory
experiment was conducted over 45 days in a controlled set-up (LabHorta) that endeavour re-creation of the
hydrothermal vent chemical environment. Our animal model was the vent bivalve Bathymodiolus azoricus from the
Menez Gwen vent site of the Mid Atlantic Ridge (MAR). Animals were exposed to conditions lacking inorganic S
supply for 30 days, which is vital for their symbionts, and then re-acclimatized in sulphide-supplied seawater for an
additional 15 days.
Gradual disappearance of bacteria from the symbiont-bearing gill cells was observed in animals kept in seawater free of
dissolved sulphide for up to 30 days, and was evidenced by histological, ultrastructural observations and Polymerase Chain
Reaction tests. Following re-acclimatisation in S-supplied seawater, proliferation of sulphur-bacteria in the gill bacteriocytes
confirms the functionality of our sulfide-feeding system in supporting chemoautotrophic symbionts. It may also indicate a
horizontal endosymbiont acquisition, i.e. from the environment to the host by means of phagocytosis-like mechanism involving
special bpit-likeQ structures on the apical cell membrane.
The present work reports the first laboratory set-up successfully used to maintain the hydrothermal vent bivalve B.
azoricus for prolonged periods of time by supplying inorganic sulphur as an energy source for its bacterial
endosymbionts. Survival of symbiont bacteria is a critical factor influencing the host physiology and thus the methods
0022-0981/$ - s
doi:10.1016/j.jem
* Correspondi
E-mail addr
gy and Ecology 318 (2005) 99–110
ee front matter D 2004 Elsevier B.V. All rights reserved.
be.2004.12.025
ng author. Tel.: +351 292 200 417; fax: +351 292 200 400.
ess: Enikokadar@notes.horta.uac.pt (E. Kadar).
E. Kadar et al. / J. Exp. Mar. Biol. Ecol. 318 (2005) 99–110100
reported here represent great potential for future studies of host-symbiont dynamics and for post-capture experimental
investigations.
D 2004 Elsevier B.V. All rights reserved.
Keywords: Hydrothermal vent; Bivalves; Sulfur-oxidizer bacteria; Endosymbiont; Bathymodiolus azoricus
1. Introduction
Deep-sea hydrothermal vent ecosystems are func-
tionally dependent on the association between macro-
organisms and their symbiotic chemolithoautotrophic
bacteria, the presence of reduced chemicals from the
vent effluent and the oxygen in seawater (Fisher et al.,
1989). These conditions are fulfilled in the areas
where the mixing of hydrothermal fluid and seawater
allows the availability of both reduced (sulphide,
methane) and oxidized (oxygen, nitrate) compounds
(Sarradin et al., 1999). Bacterial chemosynthesis
utilises the energy obtained from the oxidation of
the reduced chemicals from hydrothermal fluid for the
fixation of CO2 required for primary production
(Yamanaka et al., 2003).
Bathymodiolus azoricus is the dominant species
of many of the hydrothermal vents of the mid-
Atlantic Ridge (MAR), but the genus has worldwide
distribution. Thirteen species of the genus Bathymo-
diolus have been described to date in both the
Pacific and Atlantic Oceans (Von Cosel and Olu,
1998; Von Cosel et al., 1999; Gustafson et al., 1998).
Bathymodiolus spp. exhibits a mixotrophic existence
being capable to obtain carbon and other nutrients
from several sources. Bathymodiolus thermophilus
from the Pacific ingests particulate organic matter
(Page et al., 1991) and also has endosymbiotic
chemoautotrophic bacteria housed within the gills
(Fiala-Medioni et al., 1986; Fiala-Medioni et al.,
1994; Fiala-Medioni et al., 2002; Fisher et al., 1988).
B. azoricus hosts two metabolically distinct (meth-
anotrophic and thiotrophic) procaryotic endosym-
bionts (Pond et al., 1998). According to Fiala-
Medioni et al. (1986), Bathymodiolus sp. can also
absorb and incorporate free amino acids. This
mixotrophy allows the bivalve to have a broad
distribution relative to the effluent compared to other
vent species, being able to supplement its metabolic
needs by heterotrophy (Fisher et al., 1989). More-
over, larvae of these species appear to be capable of
dispersing hundreds of kilometres along a continuous
ridge system and across spreading centres presenting
a larval stage with feeding that might facilitate
dispersal between ephemeral vent habitats, i.e. non-
symbiotic (Craddock et al., 1995).
B. azoricus is found at different depths in hydro-
thermal vents along theMAR:Menez Gwen (37835VN–388N, 840–870 m depth) and Lucky Strike (37800VN–37835VN, 1700 m depth), Rainbow (36814VN,�2300 m
depth) and at Broken Spur (29810VN,�3000 m depth)
(Southward et al., 2001). Menez Gwen (MG) was the
source of our experimental specimens, because it is the
shallowest site on the Azores Triple Junction (ATJ) of
the MAR, and thus mussels survive without the
specialised pressure gear. It was discovered on the
DIVA 1 cruise in May 1994 (Von Cosel et al., 1999).
Patches of mussels at MG range between densities of
400–700 individuals m�2 clustering in a dilute hydro-
thermal medium (88–100% seawater, according to
Sarradin et al., 1999) with an average of 8.2–8.6 8C and
pH of b6.2–8.0 (in situ measurements during mussel
collection). Total hydrogen sulphide concentrations
have been reported as ranging between 0 and 60 AM l�1
(Sarradin et al., 1999).
Earlier ultrastructural investigations on the gill of
B. thermophilus (Fiala-Medioni et al., 1986; Le
Pennec and Bejaoui, 2001; Dubilier et al., 1998)
revealed that the filaments are mainly composed of
bacteriocytes with a large number of intracellular
bacteria, except for two small, frontal and distal areas
that remained ciliated. However, the presence of a
functional feeding groove and the labial palps, as well
as studies of gut contents (Le Pennec and Bejaoui,
2001) confirm the ability of this bivalve to capture
particulate organic matter and thus having a non-
exclusive dependence on its endosymbiotic bacteria.
Pond et al. (1998) concluded that methanotrophs and
thiotrophs contribute equally to the nutrition of B.
azoricus from Menez Gwen. The preponderance of
the autotrophic vs. heterotrophic nutritional processes
to total nutrition is suggested to be dependent on the
E. Kadar et al. / J. Exp. Mar. Biol. Ecol. 318 (2005) 99–110 101
ecological conditions to which the organisms are
subjected. Consequently, we predict that given the
functional plasticity of the bathimodiolid ctenidium,
changes in nutrient supply would induce changes in
the gill, which in turn would give clues on how
endosymbiont bacteria are incorporated into the host.
Therefore, we set out to elucidate putative ultra-
structural changes in the gill of B. azoricus main-
tained in seawater unsupplied with inorganic S and
methane during a period of 30 days after which
animal were re-acclimatized to conditions similar to
the vent environment. In order to address adaptational
strategies involved in B. azoricus remarkable feeding
flexibility, and also to discriminate specific responses
to nutrient supply variations, we have developed a
simplified system in which sulphide was not supplied.
Hydrogen sulphide (H2S) is generally believed to be
the energy source for the establishment of sulphur
oxidizing symbiotic communities (Arndt et al., 2001)
and thus it was regarded as the limiting factor in our
experiment. In this context, cytomorphological
changes caused by nutrient alterations are described.
The functional laboratory set-up described here
was valuable in providing new insights into main-
tenance strategies enabling the study of B. azoricus
for prolonged periods essential for in depth post-
capture experimental investigations.
2. Materials and methods
2.1. Animal collection, maintenance and experimental
conditions
Mussels were collected at an average depth of
�850 m from Menez Gwen vent site on the MAR,
using the Remote Operated Vehicle (ROV), Victor
6000, aboard the R/V, L’Atalante, during the
SEAHMA I cruise (29th July–14th August, 2002).
The mussels were immediately transferred to fresh,
cooled seawater held in plastic cool boxes during
shipment and then were relocated into 40-l volume
tanks housed within a refrigerated unit (ambient
temperature 8–11 8C, water temperature 8.5 8C) at
atmospheric pressure. Each tank contained 30 animals
and was maintained as a closed system. Seawater for
the aquaria was supplied from a reservoir containing
sand-filtered seawater from an unpolluted bay in Horta,
Azores (38.58N 28.78W). Continuous semi-dark con-
ditions were maintained throughout the experiment.
Animals were kept in sulphide-free seawater for 30
days, followed by an additional 15-day exposure to
sulphide-supplied seawater. Meanwhile, this sulphide-
supplied tank housed 30 mussels already introduced
within 24 h upon collection, to serve as experimental
controls and potential symbiont infection source.
Aeration using ordinary aquarium air diffusers
enabled a water oxygen level below 50% of satu-
ration. Sulphide was supplied from a 20 mM stock
solution prepared as 5-l volume batches using
commercially available Na2S crystals dissolved in
filtered seawater (0.45 Am pore size cellulose acetate
membrane, MilliporeR) which was then adjusted to
pH 8.8–9.2 with 0.5 N HCl. The stock solution was
dispensed using a Masterflex pump (model 77120-62)
at a rate of 2 ml min�1, for 15 min, every 2 h to the
seawater tanks, through a diffusion tube at the base of
the tank. The average tank-concentration of sulphide
was 15 AM (cycle range).
The water in the aquaria was changed every 7–10
days when the animals were transferred to a new tank
(previously supplied with sulphide in order to
minimize stress-induced reactions) for removal of
pseudo-faeces and other organic matter that caused
oxygen depletion, and thus, excessive H2S build up.
Total dissolved sulphide present in a mixture of
H2S+HS�+S2�was measured twice a day by color-
imetry according to Cline (1989). Water temperature,
pH and dissolved oxygen concentrations were also
measured daily. The following groups of five
animals were analysed: group 0, animals freshly
dissected immediately after collection; group 1,
mussels quickly transferred to sulphide-supplied
seawater upon collection and dissected after 15-day
exposure; group 2, mussels maintained in sulphide-
free seawater since initial collection and for 15 days;
group 3, mussels maintained in sulphide-free sea-
water since initial collection and for 30 days; group
4, mussels from group 3 re-acclimatized to sulphide
supplied seawater.
2.2. Sulphide monitoring
Two-milliliter syringes were flushed with pressur-
ised nitrogen several times before sampling. A
sample of 1.6-ml water was taken from above the
0
10
20
30
40
50
60
0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17Time (hours)
Fig. 1. Concentration of dissolved sulphide in the water column
monitored. Measurements were made before and after the 15-min
pumping periods (indicated by arrows) every 2 h.
0
10
20
30
40
50
60
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15
Time (days)
Fig. 2. Concentration of dissolved sulphide in the water column
showing an excessive build up on day 12 followed by water change
Values represent averageFSEM of three samples taken at differen
depths above the mussel clump from the experimental tanks.
E. Kadar et al. / J. Exp. Mar. Biol. Ecol. 318 (2005) 99–110102
densest clump of mussels, 25 cm below the surface.
Immediately afterward, 0.4 ml of Diamine reagent
(N,N-dimethyl-1, 4-phenylendiammoniumdichlorid,
MERCK) was drawn up into the syringe and mixed.
The samples were left to react for 15 min and then
read in a spectrophotometer at 670 nm. Standards
between 0 and 100 AM were made by diluting
immediately from a 10 mM stock solution. Degassed
seawater was used to dilute the standards and make
up the stock solution (Cline, 1989).
2.3. Tissue preparation for light and electron
microscopy
Small (1 mm3) tissue pieces were fixed in modified
Trump’s fixative (3% glutaraldehyde and 3% paraf-
ormaldehyde made up with a fixation buffer contain-
ing: 0.15 M Na-cachodylate, 0.3 M sucrose, 0.2 M
NaCl and 0.008 M CaCl according to Distel and
Felbeck, 1987). Following primary fixation, samples
were washed in 0.1 M cacodylate buffer (pH 7.8),
post-fixed in 1% osmium tetroxide in cacodylate
buffer for 1 h, dehydrated in ethanol and embedded in
Spurr resin (Sigma).
Semi-thin (2 Am) sections were obtained using
diamond knife on a LKB-BROMMA ultramicrotome
and stained with methylene blue. Ultra-thin sections
were mounted on copper grids and were double
stained with uranyl acetate and lead citrate.
Five individuals from each experimental group
were investigated (3 blocks per individual) and
observations were made on filaments detached from
the mid portion of the external demibranchs.
2.4. DNA extraction and PCR amplification
Nucleic acids from whole gill tissues were ex-
tracted according to Sambrook et al. (1989) with
slight modifications. In brief, 100–150 mg of tissue
was homogenized with 400 Al STE buffer and
resulting homogenate treated with 20 Al of proteinaseK and 40 Al of 10% SDS solution for 2 h at 55 8C.Two consecutive phenol/chloroform/isoamyl alcohol
(25:24:1; SIGMA, Molecular Biology Grade) extrac-
tions were performed followed by DNA precipitation
with 45 Al of 5 M NH4OAC mixed to the aqueous
phase. Subsequently, purified DNA was used as
template for PCR amplification. 1 Al of Vent
polymerase (New England Biolabs) was used in 50
Al total volume reactions including 20 pM of each
sense and antisense primers, dNTPs at 0.25 mM and
1�reaction buffer, following manufacturer’s instruc-
tions. Thermo-cycling conditions were performed
according to standard conditions. Briefly, an initial
denaturation step at 94 8C for 2 min (bhot startQ) wasfollowed by 35 cycles of denaturation at 94 8C for 1
min, annealing at 50 8C for 1 min and extension 68
8C, followed by a 7-min final extension at 72 8C.Primers design was based on nucleotide sequences
available from the NCBI public database for the
following genes with GenBank accession nos.
AB073122 and AB178052 respectively: Bathymodio-
lus endosymbiont gene for 16S rRNA (sense
5VAGAGTTTGTTCATGGCTCAGA3Vand antisense
.
t
E. Kadar et al. / J. Exp. Mar. Biol. Ecol. 318 (2005) 99–110 103
5VGAAGGCACCAATCCATCTCTG3V); Bathymo-
diolus endosymbiont ATP sulpfurylase gene (sense
5VGCTTTTCAGACCCGCAACCCC3V and antisense
5VCTTGGTGCCGGAGAGCA GTAC3V). PCR prod-
ucts were examined on 0.8% agarose gel electro-
phoresis according to standard protocol.
Plate 1. Macroscopic and histological changes in B. azoricus exposed to
seawater containing about 10 AM dissolved sulphide. (a) Gross appearance
symbiont starved mussel (group 3). Note colour change from brown to wh
group 0 mussels. Gill filaments are divided into 3 functionally different zon
(BZ). Scale bar 50 Am. (d) Bacteriocytes (B) in the gill of group 0 mussels.
seawater for 15 days (group 2), containing several amoebocytes (arrow
substantially reduced. Scale bar: 50 Am. (f) Very thin bacteriocytes and a
observed in gills of group 3 mussels. Scale bar: 10 Am. (g) Gill filaments fr
15 days following the 30-day sulphide free treatment). Scale bar: 50 Am. (h
slightly thicker than in animals kept in sulphur-free seawater, but gill fila
3. Results
3.1. Sulphide levels in the water column
Sulphide concentration reached a dynamic equili-
brium at about 40 AM in approximately 12 h (Fig. 1)
sulphur-free seawater for 30 days followed by re-acclimatization to
of ctenidium in a freshly collected animal (from group 0) vs. (b) a
ite. Scale bars: 1 cm. (c) Light micrograph of the gill filaments from
es: ciliated frontal zone (CZ), transitory (TZ) and bacteriocyte zones
Scale bar 10 Am. (e) Gill filaments from an animal kept in H2S-free
s) within the lumen. The thickness of the bacteriocyte layer is
n increase in the number and size of amoebocytes (arrows) can be
om group 4 mussels (i.e. reacclimatize to H2S-supplied seawater for
) Bacteriocyte zone of a re-acclimatize animal. The bacteriocytes are
ments have not regained their normal thickness. Scale bar: 10 Am.
Plate 2. Electron micrographs of gill filaments of B. azoricus dissected freshly upon collection (group 0). (a) Bacteriocytes contain both types of
symbionts at the apical region; the smaller rod-shaped ones are sulfur-oxidizer bacteria (Sb) and the larger are methanotrophic bacteria (Mb).
The central nucleus (N) and several large lysosomes (L) with membranous content are also visible. Scale bar: 2 Am. (b) Symbiont bearing
vacuoles with non-mixing bacterial content. A double cell membrane (Gram-) and a central clear zone with DNA strands are visible in the
sulfur-oxidizer bacteria (Sb), and rich membranous content characterize the larger oval shaped methanotrophic bacteria (Mb). Scale bar: 0.5 Am.
E. Kadar et al. / J. Exp. Mar. Biol. Ecol. 318 (2005) 99–110104
when pumped at 2 ml min�1, for 15 min every 2 h in
tanks with seawater. In the presence of animals, levels
were around 10 AM, whereas anaerobic conditions,
possibly due to proliferation of free living bacteria,
produced elevated levels of dissolved sulphide in the
water column on day 12, as shown in Fig. 2.
Noteworthy is the full water change to avoid build-
up of excessive sulphide levels incurring significant
mussel loss.
3.2. Morphological and histological changes resulted
from exposure of mussels to sulphide-free seawater
followed by re-acclimatization
Freshly collected B. azoricus exhibited brown gills
with thick demibranchs due to the proliferation of
Plate 3. Electron micrographs of bacteriocytes from B. azoricus gills under
supplied seawater within 24 h of collection, methanotrophs are no longe
bacteria (Sb) are observed in bacteriocytes. Scale bar: 2 Am. (b) Sulfur-oxi
1Am. (c) and (d) Bacteriocytes from animals kept in H2S-free seawater for
free vacuoles (*) and large size lysosomes (L). Scale bars: 0.5 Am and 0.3
apical cell surface of bacteriocytes from animals re-acclimatized in seawat
Copious amounts of sulfur-oxidizer bacteria (Sb) fill the apical zone of the
be observed in close association with the pit-like structures on the surface
symbionts in the host tissue (Plate 1a). Thirty days
after being transferred to sulphur- and CH4-free
seawater, mussels appeared with significant changes
in colour and thickness of their ctenidium. Gills have
become white and feeble as compared to their natural
aspect (Plate 1b). Sections of filaments from gill
lamellae, revealed a single-layered wall around the
central lumen, and exhibited three typical zones that
were previously defined for other symbiont-bearing
bivalves (Distel and Felbeck, 1987; Fiala-Medioni et
al., 1986). The three typical zones, shown in Plate 1c
are the frontal ciliated zone, the transitional zone of
non-ciliated, non-pigmented and bacteria-free cells
and the bacteriocyte zone composed of cells hosting
two types of symbiotic bacteria. These cells con-
stituted the major part of the filament (Plate 1d).
different treatments. (a) and (b) Group 1 mussels, placed in sulphide-
r present but several empty vacuoles (asterisks) and sulfur-oxidizer
dizer bacteria (Sb) do not show morphological alterations. Scale bar:
30 days (group 3) have a spongy appearance, containing symbiont-
Am, respectively. (e) Pit-like structures (arrows) are evident on the
er containing about 10 AM dissolved sulphide for 15 days (group 4).
bacteriocytes. Scale bar: 1 Am. (f) Sulfur-oxidizer bacteria (Sb) could
of bacteriocytes in re-acclimatized mussels. Scale bar: 0.2 Am.
E. Kadar et al. / J. Exp. Mar. Biol. Ecol. 318 (2005) 99–110 105
E. Kadar et al. / J. Exp. Mar. Biol. Ecol. 318 (2005) 99–110106
Mussels kept in sulphide-free seawater revealed
gill filaments that, although exhibiting the single cell
layer intact, were much slimmer in the bacteriocyte
zone and presented a more vacuolated cytoplasm
(Plate 1e–f). An increase in both the number and the
size of amoebocytes from the central lumen of
filaments was also observed (Plate 1e–f).
Even though the gill filaments and the bacterio-
cytes did not recover to their original aspect by day 15
of re-acclimatization to sulphide-supplied seawater
(Plate 1g, h), many bacteriocyte cells contain copious
number of bacteria. However, light micrographs
revealed that the filaments remained shrunken and
the incidence of binfectedQ cells seemed lower as
compared to freshly collected animals (Plate 1h).
3.3. Cytomorphological changes in the bacteriocytes
of symbiont-starved and re-acclimatized bivalves
Ultrastructural observations on the bacteriocytes
from freshly collected animals revealed that the
major volume of cells was composed of symbionts
that occupied the apical region (Plate 2a). These
vacuoles contained two distinctive types of bac-
teria: the smaller, rod-shaped and more abundant,
sulphur oxidizers, and the larger, oval shaped,
methanotrophic bacteria with rich membranous
content, probably, type I methanotrophs (Plate
2b). The two types of bacteria did not show signs
Fig. 3. PCR detection of B. azoricus endosymbiont bacteria in total geno
rRNA (primer set 1) and ATP-sulfurilase (primer set 2) genes was succe
(group 0, lanes 1 and 2 respectively). Mussels kept in seawater for 30 da
presence of endosymbiont (group 3, lanes 3 and 4). Upon re-acclimatizatio
(lane 5) and ATP sulfurilase (lane 6) is again detected suggesting inter
products are shown (arrows) as well as 1 Kb DNA ladder.
of aggregation within the same vesicle. Despite of
the size and shape modulations (due to sectioning
angle), all sulphur oxidizers shared similar ultra-
structural features, i.e. double membranes (gram
negative) with DNA strands found in the centre of
an electron-translucent area (Plate 2b). The larger
methanotrophs also had double membrane and
often presented multiple electron-translucent areas
with DNA strands and cytoplasmatic membranous
material (Plate 2b). Mussels from group 1 (exper-
imental control) lost their methane-oxidizing sym-
bionts, but not the sulphur ones (Plate 3 a,b).
Ultrastructural changes that occurred as a result of
keeping animals in sulphide-free seawater (groups 3
and 4) were most obvious in bacteriocytes, where the
apical zone appeared spongy due to the loss of
bacteria from vesicles (Plate 3 c,d). The general
morphological changes observed on bacteriocytes
were an increased incidence of lysosomes, and also,
an increase in their size as compared to those in
animals from group 0. The simultaneous shrinkage of
the bacteriocytes confers an appearance of the cell as
if almost its entire volume was occupied by lyso-
somes (Plate 3 c). In addition, the lysosomal content
appeared as in a more advanced degradation stage
with unfolded and more heterogeneous membranous
material.
In mussels from group 4 bacteriocytes resembled
those of experimental controls (group 1) in that the
mic DNA preparations from gill tissues. The presence of both 16S
ssfully detected in samples from mussels dissected upon collection
ys failed to reveal the presence of both genes and consequently the
n to sulphide-supplied seawater (group 4) the presence of 16S rRNA
-animal endosymbiont transmission. Expected PCR amplifications
E. Kadar et al. / J. Exp. Mar. Biol. Ecol. 318 (2005) 99–110 107
sulfur-oxidizers were abundant and the larger methane
oxidizers were absent (compare Plate 3 a,b with e,f).
The apical membrane of bacteriocytes displayed
several bpit-likeQ structures (Plate 3, arrows) that
seemingly allowed the entry of sulphur symbiont
bacteria (Plate 3f).
3.4. PCR amplifications
PCR reactions performed with two specific set of
primers (set 1 and set 2) generated DNA fragments of
the expected size as indicated by agarose gel electro-
phoresis, demonstrating the presence of two bacterial
endosymbiont genes from Bathymodiolus sp.
Primer set 1, specifically designed to detect the
ribosomal gene 16S rRNA from endosymbiont of
Bathymodiolus sp. was successful in PCR tests aimed
at the detection of this gene in DNA extracts from
gills homogenates from mussel group 0 (Fig. 3, lane
1). The specificity of our PCR-based endosymbiont
detection method was further demonstrated with a
second primer set specifically designed to detect the
Bathymodiolus bacterial endosymbiont ATP sulfury-
lase gene of deep-sea hydrothermal vent mussels from
the Japanese Suiyo Seamount (Fig. 3, lane 2). DNA
extracts originated from mussel group 3, failed to
reveal the presence of both 16S rRNA and ATP
sulfurylase genes (Fig. 3 lanes 3 and 4 respectively).
In contrast, gill DNA of mussels from group 4
revealed the presence of endosymbiont (Fig. 3, lanes
5 and 6).
While primer set 2 oligonucleotide sequence is
based on a gene sequence from a different Bathy-
modiolus species, our results indicated that not only
this gene is detected in DNA preparations of B.
azoricus gills (Fig. 3 lane 2), but also confirms the
reappearance of sulphur-oxidizing endosymbiont bac-
teria in mussels from group 4 (Fig. 3 lane 6). Our
results therefore, indicate that both primer sets can
successfully amplify the bacterial symbiont DNA
target of genomic DNA extracted from symbiont-
containing gill tissues of B. azoricus.
4. Discussion and conclusions
The flexible feeding regime of B. azoricus based
on both mixotrophy (filter feeding and symbiosis) and
a dual symbiosis (methanotrophic and thiotrophic)
(Pond et al., 1998) enables the mussel to tolerate wide
temporal and spatial variability of the environmental
factors typical to hydrothermal vents (Johnson et al.,
1994). Thus we have selected the bivalve as a model
organism to investigate the mechanisms underlying
nutritional responses in relation to environmental
variations under controlled laboratory conditions.
The laboratory set-up that was developed in Lab-
Horta, and based on a sulphide feeding system was
successful in maintaining endosymbiosis in the host
vent bivalve. This opens the possibility to use
endosymbiont bacteria, otherwise unculturable under
laboratory conditions, as a new experimental tool in
vent research.
The results obtained on the cyclic sulphide supply
into the experimental aquaria are in agreement with
the previously reported 1-h half-life of dissolved
sulphide at neutral pH in aerobic seawater (Almgren
and Hagstrom, 1974). This regime of pumping a stock
Na2S solution on a 2-hourly basis permitted the
maintenance of a stable average level of 10 AMsulphide inside the aquaria that is within the range of
0.5–18 for the AS reported for mussel beds in Menez
Gwen (Sarradin et al., 1999). Prevention of H2S
depletion in experimental tanks was insured by an
intermittent supply system essential for maintenance
of sulphur oxidizer symbiont bacteria reported sensi-
tive to low levels (Childress et al., 1991). Provided
that mussels are placed in seawater supplied with
dissolved sulphide within 24 h of collection, they
survive over several months (data not shown). Oxy-
gen saturation is another major factor influencing
animal maintenance. Under anoxic conditions anae-
robic sulphide production takes place, as reported by
Arndt et al. (2001) for several sulphur-storing
symbioses, and also confirmed in the present study
(see day 12 in Fig. 2). Unless fresh water change is
ensured, massive mortality occurs in the experimental
tanks. However, oxygen levels should not exceed
30% of saturation, as oxidation reactions would
deplete H2S from the system. Thus, for a successful
long-term experimental set-up, rigorously controlled
conditions are obtained by an enduring re-adjustment
of the three major interacting factors, pH, oxygen and
sulphide concentration. Continuous monitoring of
these factors is essential for long-term maintenance
of B. azoricus in captivity permitting specific inves-
E. Kadar et al. / J. Exp. Mar. Biol. Ecol. 318 (2005) 99–110108
tigations to be conducted with a great advantage over
the costly and both time- and human-resource-
consuming in situ observations. Behavioural observa-
tions as well as toxicological investigations will be
possible to be performed under controlled laboratory
conditions with the prospect of understanding the
mechanisms that enable this species to survive under
the extreme chemical and physical conditions typical
to hydrothermal vents.
Ultrastructural and molecular evidence is presented
for the substantial loss of both bacterial symbionts in
bacteriocytes of mussels kept in sulphide-free sea-
water for up to 30 days. Residual endosymbiont
bacteria might have survived the sulfide-deprivation
treatment, however, our PCR detection system sug-
gests that the endosymbiont bacterial population can
vary from one experimental condition to another and
this population particularly increases when sulfide-
deprived mussels are exposed to an environment
containing sulfide and other untreated mussel indi-
viduals. This, in it self, suggests the existence of a
lateral acquisition of symbionts that does not clearly
excludes a vertical transmission. Additionally, PCR
has been a widely used method in deep-sea environ-
mental microbiology as a molecular tool in the
detection of prokaryotes (Imhoff et al., 2003; Gros
et al., 1996, 1998a, 2003; Won et al., 2003). In spite of
this lack of endosymbiont detection, it is possible
however, that a few dormant cells in the form of
endospore may have remained and started dividing as
soon as sulphide was supplied. If so, these endospores
must be resistant to such an extent that it would
withstand standard genomic DNA extraction proto-
cols. A more likely alternative is the re-infection of
these mussels by means of a passive contamination
through bacteria released in the media by control
animals kept in the same sulphide-supplied tank.
Additionally, when in a separate experiment, sym-
biont-starved mussels were transferred to sulphide
supplied aquarium without other mussels, did no show
signs of bacterial re-infection (data not shown).
Control animals (group 1) did not loose their sulphur
endosymbionts, as demonstrated in our studies, as
opposed to a deficiency in methanotrophs as was
expected from the experimental conditions (no meth-
ane supply). Recovering endosymbiont bacteria from
the autoclaved seawater used to rinse gills prior to
bacterial extraction brings about indirect evidence for
their spontaneous leaking from bacteriocytes. Addi-
tionally, there is evidence originated from studies by
Gros and co-workers on environmental transmission
of sulfur-oxidizer bacteria in various symbiont-har-
bouring bivalves (Gros et al., 2003, 1996, 1998a,
1999). These authors have determined the trans-
mission mechanisms by experimental colonisation of
aposymbiotic Codakia larvae and proposed the bpit-likeQ structures of the membrane as a route for
bacterial access into the bacteriocyte (Gros et al.,
1998b). Such structures were consistently observed in
our ultra structure studies of membrane invaginations
on bacteriocytes found in re-acclimatized mussels’
gills (Plate 3e–f). A third alternative source of
infection, although unlikely, may be the invasion of
the gills by free-living sulphur oxidizing bacteria that
may have proliferated in the sulphide-supplied sea-
water. There is evidence for the existence of free-
living sulphur oxidizer bacteria becoming symbiotic
when entering the water current within the bivalve gill
(Gros et al., 1999). The facultative symbiotic nature of
these sulfur-oxidizers is also inferred in Nelson and
co-workers studies (Nelson et al., 1995) taking in
consideration the overlapping symbiont’s molar
growth yield on thiosulfate, with that of free-living
chemoautotrophs. However, 16S rDNA-based phylo-
genetic studies on sulphur-oxidizing bacterial endo-
symbionts indicated that they are clearly distinct from
free-living sulphur-oxidizing bacteria of the genera
Beggiatoa, Halothiobacillus and Thiomicrospira
(Imhoff et al., 2003). Our PCR-based approach to
detect the presence of endosymbionts of B. azoricus
was designed in a way such, only specific endo-
symbiont target genes would be amplified, and thus
ruling out the possibility of contaminations from free-
living sulfur-oxidizer bacteria (Fig. 3). In light of our
results, we conclude that (a) B. azoricus is able to
survive in the absence of sulphide and resulting
reduction of its endosymbiont population for extended
period of time; (b) endosymbiosis can be experimen-
tally manipulated in that once ceased (nutrient
deprivation) may be regained provided that adequate
chemical conditions in the environment are met; (c)
horizontal endosymbiont transmission is possible via
inter-animal contamination. Recent field experiments
conducted by Raulfs and co-workers (Raulfs et al.,
2004) in which B. termophylus were transplanted
away from venting exits at hydrothermal site on the
E. Kadar et al. / J. Exp. Mar. Biol. Ecol. 318 (2005) 99–110 109
southern East Pacific Rise reached similar conclusions
regarding symbiont loss and consequent ultrastruc-
tural modifications, and thus corroborate our exper-
imental results.
By developing a successful laboratory set-up for
the maintenance of B. azoricus that enables preser-
vation and manipulation of endosymbiosis, we
provide a basis for a more elaborated ecophysio-
logical research, in order to understand the general
principles that govern adaptations to the hydro-
thermal environment.
Acknowledgements
The research was undertaken under the scope of
the research project SEAHMA (Seafloor and sub-
seafloor hydrothermal modelling in the Azores Sea)
funded by FCT (PDCTM/P/MAR/15281/1999). Post-
doctoral fellowship support was jointly offered by the
Portuguese Science Foundation (FCT) and by IMAR,
Portugal to Eniko Kadar (IMAR/FCT-PDOC-012/
2001-EcoToxi).
The authors acknowledge the ROV team and the
Atalante crew for their contribution in sampling and
the technical team that helped running LabHorta, the
laboratory set-up for the animal maintenance. We are
also indebted to Sergio Stefanni for experimental
advice and to Tony Ip for his help in providing
primers and allowing the use of his lab facilities at the
University of Massachusetts.
The experiments carried out for this study comply
with the current pertinent laws in Portugal. [SS]
References
Almgren, T., Hagstrom, A., 1974. The oxidation rate of sulphide in
sea water. Water Research 8, 395–400.
Arndt, C., Gaill, F., Felbeck, H., 2001. Anaerobic sulphur
metabolism in thiotrophic symbioses. Journal of Experimental
Biology 204, 741–750.
Childress, J.J., Fisher, C.R., Favuzzi, J.A., Sanders, N.K., 1991.
Sulphide and carbon-dioxide uptake by the hydrothermal vent
clam, Calyptogena magnifica, and its chemoautotrophic sym-
bionts. Physiological Zoology 64, 1444–1470.
Cline, J.D., 1989. Spectrophotometric determinations of hydrogen
sulphide in natural waters. Limnology and Oceanography 14,
454–459.
Craddock, C., Hoeh, W.R., Lutz, R.A., Vrijenhoek, R.C., 1995.
Extensive gene flow among Mytilid (Bathymodiolus thermo-
philus) populations from hydrothermal vents of the Eastern
Pacific. Marine Biology 124, 137–146.
Distel, D.L., Felbeck, H., 1987. Endosymbiosis in the lucinid
clams Lucinoma aequizonata, Lucinoma annulata and Lucina
floridana—a reexamination of the functional-morphology of
the gills as bacteria-bearing organs. Marine Biology 96 (1),
79–86.
Dubilier, N., Windoffer, R., Giere, O., 1998. Ultrastructure and
stable carbon isotope composition of the hydrothermal vent
mussels Bathymodiolus brevior and B. sp. Affinis brevior from
the North Fiji Basin, western Pacific. Marine Ecology-Progress
Series 165, 187–193.
Fiala-Medioni, A., Metivier, C., Herry, A., Lepennec, M., 1986.
Ultrastructure of the gill of the hydrothermal-vent Mytilid
Bathymodiolus Sp. Marine Biology 92, 65–72.
Fiala-Medioni, A., Michalski, J.C., Jolles, J., Alonso, C., Montreuil,
J., 1994. Lysosomic and lysozyme activities in the gill of
bivalves from deep hydrothermal vents. Comptes Rendus De L
Academie Des Sciences Serie Iii-Sciences De La Vie-Life
Sciences 317, 239–244.
Fiala-Medioni, A., McKiness, Z.P., Dando, P., Boulegue, J.,
Mariotti, A., Alayse-Danet, A.M., Robinson, J.J., Cavanaugh,
C.M., 2002. Ultrastructural, biochemical, and immunological
characterization of two populations of the Mytilid mussel
Bathymodiolus azoricus from the Mid-Atlantic Ridge: evidence
for a dual symbiosis. Marine Biology 141, 1035–1043.
Fisher, C.R., Childress, J.J., Arp, A.J., Brooks, J.M., Distel, D.,
Favuzzi, J.A., Felbeck, H., Hessler, R., Johnson, K.S., Kenni-
cutt, M.C., Macko, S.A., Newton, A., Powell, M.A., Somero,
G.N., Soto, T., 1988. Microhabitat variation in the hydrothermal
vent mussel, Bathymodiolus-Thermophilus, at the rose garden
vent on the Galapagos rift. Deep-Sea Research. Part A,
Oceanographic Research Papers 35, 1769–1791.
Fisher, C.R., Brooks, J.M., Childress, J.J., Felbeck, H., Hessler,
R.R., Johnson, K.S., Macko, S.A., Moe, A., Nelson, D.,
Somero, G.N., 1989. Microhabitat requirements of the hydro-
thermal vent mussel, Bathymodiolus thermophilus. American
Zoologist 29, A81.
Gros, O., et al., 1996. Environmental transmission of a sulfur-
oxidizing bacterial gill endosymbiont in the tTropical lucinid
bivalve Codakia orbicularis . Applied and Environmental
Microbiology 62, 2324–2330.
Gros, O., et al., 1998a. Putative environmental transmission of
sulfur-oxidizing bacterial symbionts in tropical lucinid bivalves
inhabiting various environments. FEMS Microbiology Letters
160, 257–262.
Gros, O., Frenkiel, L., Moueza, M., 1998b. Gill filament differ-
entiation and experimental colonization by symbiotic bacteria in
aposymbiotic juveniles of Codakia Orbicularis (Bivalvia:
Lucinidae). Invertebrate Reproduction and Development 34,
219–231.
Gros, O., Duplessis, M.R., Felbeck, H., 1999. Embryonic develop-
ment and endosymbiont transmission mode in the symbiotic
clam Lucinoma aequizonata (Bivalvia : Lucinidae). Invertebrate
Reproduction and Development 36, 93–103.
E. Kadar et al. / J. Exp. Mar. Biol. Ecol. 318 (2005) 99–110110
Gros, O., et al., 2003. Detection of the free-living forms of sulfide-
oxidizing gill endosymbionts in the lucinid habitat (Thalassia
testudinum environment). Applied and Environmental Micro-
biology 69, 6264–6267.
Gustafson, R.G., Turner, R.D., Lutz, R.A., Vrijenhoek, R.C., 1998.
A new genus and five new species of mussels (Bivalvia,
Mytilidae) from deep-sea sulphide/hydrocarbon seeps in the
Gulf of Mexico. Malacologia 40, 63–112.
Imhoff, J.F., Sahling, H., Suling, J., Kath, T., 2003. 16S rDNA-
based phylogeny of sulphur-oxidising bacterial endosymbionts
in marine bivalves from cold-seep habitats. Marine Ecology-
Progress Series 249, 39–51.
Johnson, K.S., Childress, J.J., Beehler, C.L., Sakamoto, C.M., 1994.
Biogeochemistry of hydrothermal vent mussel communities—
the deep-sea analog to the intertidal zone. Deep-Sea Research
Part I-Oceanographic Research Papers 41, 993–1011.
Le Pennec, M., Bejaoui, N.A., 2001. The conquest of reduced deep
marine ecosystems by Mytilid mussels. Bulletin de la Societe
Zoologique de France 126, 121–127.
Nelson, D.C., Hagen, K.D., Edwards, D.B., 1995. The gill symbiont
of the hydrothermal vent mussel Bathymodiolus-thermophilus is
a psychrophilic, chemoautotrophic, sulphur bacterium. Marine
Biology 121, 487–495.
Page, H.M., Fialamedioni, A., Fisher, C.R., Childress, J.J., 1991.
Experimental-evidence for filter-feeding by the hydrothermal
vent mussel, Bathymodiolus-thermophilus. Deep-Sea Research
Part A-Oceanographic Research Papers 38, 1455–1461.
Pond, D.W., Bell, M.V., Dixon, D.R., Fallick, A.E., Segonzac, M.,
Sargent, J.R., 1998. Stable-carbon-isotope composition of fatty
acids in hydrothermal vent mussels containing methanotrophic
and thiotrophic bacterial endosymbionts. Applied and Environ-
mental Microbiology 64, 370–375.
Raulfs, E.C., Macko, S.A., Van Dover, C.L., 2004. Tissue and
symbiont condition of mussels (Bathymodiolus thermophilus)
exposed to varying levels of hydrothermal activity. Journal of
the Marine Biological Association of the United Kingdom 84
(1), 229–234.
Sambrook, J., Fritsch, E.J., Maniatis, T., 1989. Molecular cloning: A
laboratory manual. 2nd ed. Cold Spring Harbour Laboratory
Press, Cold Spring Harbour, New York.
Sarradin, P.M., Caprais, J.C., Riso, R., Kerouel, R., Aminot, A.,
1999. Chemical environment of the hydrothermal mussel
communities in the lucky strike and Menez Gwen vent fields,
Mid-Atlantic Ridge. Cahiers De Biologie Marine 40, 93–104.
Southward, E.C., Gebruk, A., Kennedy, H., Southward, A.J.,
Chevaldonne, P., 2001. Different energy sources for three
symbiont-dependent bivalve molluscs at the Logatchev hydro-
thermal site (Mid-Atlantic Ridge). Journal of the Marine
Biological Association of the United Kingdom 81, 655–661.
Von Cosel, R., Olu, K., 1998. Gigantism in Mytilidae. A new
Bathymodiolus from cold seep areas on the Barbados accre-
tionary Prism. Comptes Rendus De L Academie Des Sciences.
Serie Iii, Sciences De La Vie-Life Sciences 321, 655–663.
Von Cosel, R., Comtet, T., Krylova, E.M., 1999. Bathymodiolus
(Bivalvia : Mytilidae) from hydrothermal vents on the Azores
triple junction and the Logatchev hydrothermal field, Mid-
Atlantic Ridge. Veliger 42, 218–248.
Won, Y.J., Hallam, S.J., O’Mullan, G.D., Pan, I.L., Buck, K.R.,
Vrijenhoek, R.C., 2003. Environmental acquisition of thiotro-
phic endosymbionts by deep-sea mussels of the genus
Bathymodiolus. Applied and Environmental Microbiology 69,
6785–6792.
Yamanaka, T., Mizota, C., Fujiwara, Y., Chiba, H., Hashimoto, J.,
Gamo, T., Okudaira, T., 2003. Sulphur-isotopic composition of
the deep-sea mussel Bathymodiolus marisindicus from cur-
rently active hydrothermal vents in the Indian Ocean. Journal
of the Marine Biological Association of the United Kingdom 83,
841–848.
top related