INTRODUCTION
Marteilia is a protozoan parasite belonging to the phylum Paramyxea which affects some marine invertebrate species. In Europe, this parasite has been responsible for important mortalities of flat oysters Ostrea edulis in culture areas (Goulletquer and Héral, Reference Goulletquer and Héral1997). As far as cultivated mussels Mytilus galloprovincialis are concerned, several authors have reported less mortality but a loss of condition and negative effects on reproduction in parasitized individuals (Villalba et al. Reference Villalba, Mourelle, Carballal and López1993; Camacho et al. Reference Camacho, Villalba, Beiras and Labarta1997). Due to the detrimental impact of this parasite on flat oyster culture since the 1970s, M. refringens infection has been listed as a notifiable disease to the OIE (the World Organization for Animal Health). Consequently, the parasite has become a subject for scientific research since its first description in 1968 in the Aber Wrach, Brittany (Comps, Reference Comps1970).
In Europe, 2 species or types of Marteilia, M. refringens (type O) and M. maurini (type M), are responsible for marteiliosis of oysters and mussels, respectively (Le Roux et al. Reference Le Roux, Lorenzo, Peyret, Audemard, Figueras, Vivares, Gouy and Berthe2001). Nevertheless, recent data, based on molecular studies, suggest that both parasites, M. refringens and M. maurini, are conspecific and correspond to 2 types of the same genomic species called M. refringens (López-Flores et al. Reference López-Flores, De La Herran, Garrido-Ramos, Navas, Ruiz-Rejón and Ruiz-Rejón2004), most probably through an alloxenic speciation process, and this nomenclature is used in the current study. Furthermore, a molecular typing study has recently emphasized the possible cross-detection of the parasite types and the potential lack of specificity for their hosts (Novoa et al. Reference Novoa, Posada and Figueras2005). However, these molecular results need confirming by robust epidemiological and ecological datasets, as well as by biological aspects, such as a study of the complete life-cycle of the parasite.
After some unsuccessful attempts to infect mussels and oysters horizontally, the hypothesis of a heterogenic life-cycle for M. refringens was proposed (Berthe et al. Reference Berthe, Pernas, Zerabib, Haffner, Thebault and Figueras1998). Moreover, development of molecular techniques for M. refringens detection has facilitated life-cycle studies (Le Roux et al. Reference Le Roux, Audemard, Barnaud and Berthe1999, Reference Le Roux, Lorenzo, Peyret, Audemard, Figueras, Vivares, Gouy and Berthe2001). The copepod Acartia grani (Paracartia grani) has been suggested as an intermediate host in the life-cycle of M. refringens infecting flat oysters (Audemard et al. Reference Audemard, Le Roux, Barnaud, Collins, Sautour, Sauriau, De Montaudouin, Coustau, Combes and Berthe2002). This copepod appeared infected by M. refringens when the dynamics of the parasite were studied in the ‘claire’ system in Marennes-Oléron. ‘Claires’ are shallow ponds traditionally used for fattening and greening oysters, and the life-cycle of M. refringens was demonstrated to be completed in this environment (Audemard et al. Reference Audemard, Barnaud, Collins, Le Roux, Sauriau, Coustau, Blachier and Berthe2001). However, results obtained in the ‘claire’ model need to be corroborated in a more complex and natural ecosystem (Berthe et al. Reference Berthe, Le Roux, Adlard and Figueras2004).
On the other hand, to date, no intermediate hosts have been proposed for M. refringens infecting mussels. The flux of both M. refringens type O (usually infecting flat oysters) and type M (usually infecting mussel) through the copepod A. grani therefore represents an interesting subject of study in order to improve the understanding of the relationship of the two parasite types with potential intermediate hosts. The experimental infection of A. grani from infected flat oysters and mussels has been successful but the infection patterns of M. refringens in A. grani were different for copepods infected via mussels or via flat oysters. While the development of the parasite was obvious in copepods infected from flat oysters, only early stages of infection were found in A. grani infected from mussels. Neither oysters nor mussels were infected after cohabitation with infected copepods (Carrasco et al. Reference Carrasco, Arzul, Furones, Chollet, Robert, Joly and Berthe2005), as reported in previous studies (Audemard et al. Reference Audemard, Le Roux, Barnaud, Collins, Sautour, Sauriau, De Montaudouin, Coustau, Combes and Berthe2002; Berthe et al. Reference Berthe, Le Roux, Adlard and Figueras2004). Therefore, research on other zooplanktonic species as candidates for Marteilia intermediate hosts was proposed for further work.
Due to their physical and environmental characteristics, the bays of the Delta de l'Ebre in Catalonia (Spain), which are endemic areas for the Marteilia parasite, were proposed as a model study site for Marteilia dynamics and life-cycle studies (Carrasco et al. Reference Carrasco, López-Flores, Alcaraz, Furones, Berthe and Arzul2007). The Ebre bays are larger and more representative of other infected areas than the ‘claire model’, especially for Marteilia infecting mussels because mussel culture is very important in this area. Furthermore, recent studies have reported zooplankton infected by M. refringens in these bays. Preliminary results in such a complex estuarine environment suggested the implication of some zooplankton species in the parasite life-cycle. The Cyclopoida Oithona sp. and an indeterminate Harpaticoida appeared to be infected by the parasite (Carrasco et al. Reference Carrasco, López-Flores, Alcaraz, Furones, Berthe and Arzul2007). A recently developed molecular diagnostic technique based on the intergenic spacer of rDNA (IGS) (López-Flores et al. Reference López-Flores, De La Herran, Garrido-Ramos, Navas, Ruiz-Rejón and Ruiz-Rejón2004) showed high sensitivity and was therefore considered to be an interesting tool for the detection of M. refringens in small organisms such as zooplanktonic species (Carrasco et al. Reference Carrasco, López-Flores, Alcaraz, Furones, Berthe and Arzul2007). In order to improve the understanding of the life-cycle and ecology of the parasite, screening of M. refringens infection in the zooplankton community, using IGS nested PCR, was carried out in Alfacs Bay (Delta de l'Ebre) for a whole year (October 2004-October 2005). The dynamics of the parasite in mussel cultures were also studied and compared with the results obtained from zooplankton.
MATERIALS AND METHODS
Study site
The Delta de l'Ebre is located in the North Western Mediterranean (Fig. 1). Sediments from the River Ebre form and shape this delta, which is mainly comprised of 2 semi-enclosed bays: Fangar Bay in the northern half, and Alfacs Bay in the southern half. Both are exploited for mollusc aquaculture and fisheries. The dynamics of M. refringens in mussels and the zooplankton community were studied in Alfacs Bay for 1 complete year from October 2004 to October 2005. For the purpose of this study, 2 critical environmental parameters, temperature and salinity, were recorded weekly at a depth of 2 metres using a manual conductivity meter (WTW) during the experimental period.
Fig. 1. Study site: Delta de l'Ebre, Catalonia, North East Spain (arrow shows the sampling site).
Mytilus galloprovincialis samples
For this study, mussels Mytilus galloprovincialis were collected from experimental ropes, hung from a commercial raft stocked with Alfacs spat collected naturally during the spring of 2004. Sixty mussels were randomly collected each month during the study period between October 2004 and October 2005. Half of each individual was fixed in 10% formaldehyde in filtered sea water (v/v) for 24–48 h, and afterwards conserved in 70% ethanol (v/v) for histological studies in order to detect the presence of M. refringens. Samples were dehydrated and embedded in paraffin wax. Paraffin blocks were cut at 3 μm and stained with haematoxylin-eosin. Slides were studied under light microscopy at a magnification of 40X, and the prevalence and infection intensity were recorded.
The prevalence was calculated as the number of mussels found infected divided by the number of mussels histologically tested. Infection intensity was calculated according to Villalba et al. (Reference Villalba, Mourelle, Carballal and López1993) in a range from 0 to 5, which represented null infection to high infection intensity. Parasite stages in each positive case were also recorded by following Robledo and Figueras (Reference Robledo and Figueras1995): young stages in the epithelium of the stomach or primary digestive tubules (plasmodia containing 1 or 2 cells), first stages of maturation in primary and secondary tubules (plasmodia containing 2 or more cells), middle stages of maturation (sporangiosori with sporania primordia), mature stages (sporangiosori with mature sporangia) and mature sporangia released in the lumen of hepatopancreatic tubules.
Zooplankton samples
Zooplankton samples were collected every 15 days throughout the study period by means of horizontal hauls made with a Juday-Bogoroy zooplankton net fitted with 100 μm mesh. The net was hauled at 3 knots for a period of 5 min around the mussel ropes. Each zooplankton sample was split into 2 aliquots: one was immediately fixed for 24–48 h in 10% formaldehyde in filtered sea water for taxonomic studies; the second was fixed and conserved in 100% ethanol for molecular analyses.
The predominant zooplankton groups of each sample were taxonomically classified under a stereomicroscope. A minimum of 10 individuals from the most abundant taxa were sorted and transferred into Eppendorf vials containing 96% ethanol for further molecular analysis.
DNA extraction and amplification by PCR
DNA extraction from zooplankton samples was carried out as follows: ethanol-fixed animals were ground and suspended in 10 volumes of extraction buffer (NaCl 100 mm, 10 mm Tris, pH 8, 2·5 mm EDTA, pH 8, SDS 0·5%) containing proteinase K (100 μg/ml). Following an overnight incubation at 55°C, DNA was extracted using a standard protocol involving phenol/chloroform, and precipitation with ethanol. Integrity and quantity of DNA were measured by spectrometry.
Nested-PCR targeting the IGS was performed in 2 consecutive PCR reactions using the 2 primer pairs previously designed by López-Flores et al. (Reference López-Flores, De La Herran, Garrido-Ramos, Navas, Ruiz-Rejón and Ruiz-Rejón2004).
The primers MT-1/MT-2 (first round PCR) and MT-1B/MT-2B (second round PCR), were used to amplify a fragment of the intergenic spacer of the ribosomal genes (IGS) of the parasite. First round PCR assay was carried out according to the standard conditions for Silver-star Taq polymerase (Eurogenetec, Seraing, Belgium) using 10 ng of DNA. After DNA denaturation at 94°C for 5 min, 30 cycles were run with a PTC-100tm thermocycler (MJ. Research, Inc.). The protocol was comprised of denaturation at 94°C for 1 min, annealing at 55°C for 1 min and elongation at 72°C for 1 min. A final elongation for 10 min at 72°C was also performed. In the second round PCR assay, DNA denaturation at 94°C was carried out for 5 min, and 24 cycles were run with a PTC-100tm thermocycler (MJ. Research, Inc.) as follows: denaturation at 94°C for 30 sec, annealing at 55°C for 30 sec and elongation at 72°C for 30 sec. A final elongation for 10 min at 72°C was also performed.
DNA extracted from an infected oyster was used as a positive control for both PCRs. Amplified products were analysed electrophoretically on 1% agarose gels and purified using the QiaClean sequencing kit (QIAGEN). The same protocol was performed for each separate zooplankton taxon in order to evaluate the presence of the parasite.
Nucleotide sequences of purified PCR fragments were determined by Sanger's method (Sanger et al. Reference Sanger, Nicklen and Coulson1977) using the ABI Prism Dye Terminator Cycle Sequencing kit (Applied Biosystems). The identity of the sequences was obtained by comparison with sequences stored in the GenBank data base using BLAST (Altschul et al. Reference Altschul, Madden, Schaffer, Zhang, Zhang, Miller and Lopman1997).
RESULTS
Environmental characteristics
Temperatures decreased from October 2004, when values reached 18°C, to January 2005, which was one of the coldest months (around 8°C) (Fig. 2). During March 2005, water temperatures increased from 8°C to 16°C, and reached 17°C in April. Maximum temperature values were observed during June (29·4°C) and July (28·8°C). Temperature values began to decrease again from 23°C in September to 19°C at the end of October 2005.
Fig. 2. The graph shows the prevalence of Marteilia refringens and the infection intensity level from October 2004 to October 2005 in Alfacs Bay. Temperature and salinity are also represented. The electrophoresis gel shows the zooplankton samples detected as positive by nested PCR during the same period. The positive control corresponds to DNA extracted from a Marteilia-infected flat oyster.
Salinity was usually between 34 and 36 p.s.u., although specific decreases in November 2004 and the end of August 2005 brought salinity values down to around 23 p.s.u.
Marteilia refringens dynamics in Mytilus galloprovincialis
The prevalence of M. refringens in M. galloprovincialis showed a clear seasonal cycle throughout the year (Fig. 2). During autumn and winter (October 2004–March 2005) prevalences (the % of infected mussels) were low, ranging from 6·67% in October and December 2004 to 3·34% in January 2005. In spring, prevalences increased from 5% in March to 13·34% at the end of April. Infection prevalences in May decreased to 6·67% but they increased again during the summer period, reaching 11·67% in June and 26·67% in July, which was the maximum value observed during the year. During September and the autumn season, prevalences of infection began to decrease again (16·67%) and in October 2005 only 10% of the individuals were found infected.
Other parasites, such as some Turbelaria individuals and the Trematoda Protoeces sp. were also recorded in most of the monthly samples, whereas ciliates were only observed in a few cases. Two individuals (one from the June sample and the other one from the October 2005 sample) were affected by neoplasia.
Infection intensity and parasite stages in mussels
Infection intensity was high, level 5 (more than 90% of infected tubules), during April and May, August, September, and also in December (Fig. 2). In these samples, an important part of the infected individuals had most of the digestive tubules infected by M. refringens. Furthermore, all the known developmental stages of Marteilia were present in these samples, from young stages in the epithelium of the stomach (plasmodia containing 1 or 2 cells) to mature stages (sporangiosori) containing sporangia (Fig. 3). Sporangiosori containing mature sporangia and free sporangia were released, in some of the cases, into the lumen of the digestive tubules.
Fig. 3. Different stages of Marteilia refringens development in its host Mytilus galloprovincialis observed with light microscopy (H&E stained). (A) Young stages, plasmodia containing 1 or 2 cells in the epithelium of the stomach. (B) First stages of maturation in primary and secondary tubules, plasmodia containing 2 or more cells. (C) Middle stages of maturation, sporangiosori with sporangia primordial. (D) Mature stages, sporangiosori with mature sporangia. (E) Sporangiosori with mature sporangia in secondary tubules and young stages in a primary tubule. (F) Mature sporangia released in the lumen of hepatopancreatic tubules.
In July, 50% of the individuals infected by Marteilia spp. reached level 4 infection intensity (between 50% and 90% of infected tubules). Most of these mussels showed 85% infected tubules and in some of these individuals only first stages of maturation (plasmodia in the epithelium of the stomach and in secondary digestive tubules) could be observed. About 40% of the mussels collected in July were infected by young stages located in the stomach epithelium (level 1) and 10% of the samples reached level 3 infection (between 10% and 50% of infected tubules). In January and February, 50% and 75% of the infected individuals, respectively, also reached level 4 infection intensity and the presence of mature stages (sporangiosori with mature sporangia) was common.
During June 2005, October 2004 and March 2005, only young stages located in the stomach epithelium were detected.
Marteilia in mixed zooplankton samples
Eight out of 27 zooplankton samples were found positive using the Marteilia IGS nested PCR (Fig. 2). Five of them were collected during spring 2005 (from the end of March to the beginning of June). However, isolated positive samples could also be detected in autumn. The analysed sequences, one sequence from each positive sample, showed a similarity of between 94% and 100% with the Marteilia IGS fragment. Sequences were submitted to GenBank and have the following Accession numbers: AM504129, AM504130, AM504131, AM504132, AM504133, AM504134, AM504135 and AM504136.
The zooplankton of Alfacs Bay and Marteilia in zooplankton species
The zooplankton community consisted of groups of congeneric species (genera Acartia and Oithona), harpacticoids (Euterpina, Clitemnestra), with the occasional presence of Centropages, Clausocalanus and Paracalanus species. Radiolarians (Aulacantha), Cnidarians, ctenophores and appendicularians completed the list of Holoplankton groups. The meroplankton was dominated by larval stages of decapod crustaceans, polychaetes, echinoderm and mollusc larvae, as well as others which were less abundant. Highest zooplankton abundances were recorded in autumn, winter and spring.
The taxa identified in each positive zooplankton sample, as well as their relative abundance, are shown in Table 1. In some samples, up to 4 different taxa were abundant enough to be sorted into Eppendorf vials for further analysis. The Cyclopoida Oithona sp., and different Calanoida species, including Acartia clausi and Acartia discaudata, were present in 50% of the samples found infected by M. refringens (Table 2). Furthermore, other Calanoida such as Acartia italica, Paracalanus spp. and the Harpaticoida Euterpina acutrifons were also present in 25% of the infected samples. The Calanoida Acartia latisetosa and Centrophages sp. were only present in 12·5% of positive samples. Appendicularians, lamellibranch larvae, ctenophores and larvae of crustacean Decapoda (zoea of Brachyura, probably Portumnus sp.) were also present in some samples, but in low abundances.
Table 1. Taxonomic classification and relative abundance of taxa constituting zooplankton samples found to be infected by Marteilia refringens using nested PCR
(+, Weak abundance; ++, medium abundance; +++, high abundance.)
Table 2. Prevalence (%) of taxa in all zooplankton samples detected to be infected by Marteilia refringens using nested PCR
(For example, Acartia discaudata was present in 50% of the Marteilia-positive zooplankton samples.)
Six taxa, Acartia discaudata, A. clausi, A. italica, Oithona sp., Euterpina acutifrons and Brachyura larvae (zoea of Portumnus sp.) appeared infected by M. refringens when tested by IGS nested PCR analysis (Fig. 4). Therefore, ecologically diverse marine crustacean groups, such as calanoida, cyclopoida and harpaticoida copepods, and decapod crustacean larvae were parasitized by M. refringens, although A. discaudata and Oithona sp. were the most frequently detected positive species. The analysed sequences, 1 sequence from each parasitized species sample, showed a similarity of between 94% and 100% with the Marteilia IGS fragment. Sequences were submitted to GenBank and have the following Accession numbers: AM504137, AM504138, AM504139, AM504140, AM504141, AM504142, AM504143, AM504145 and AM504144.
Fig. 4. Electrophoresis gel for the IGS nested Marteilia refringens PCR (358 base pairs) analyses of the different species of zooplankton from the M. refringens-positive zooplankton samples.
DISCUSSION
Recent studies have shown the relevant role that zooplankton species can play in the life-cycle of certain marine protozoan parasites (Audemard et al. Reference Audemard, Le Roux, Barnaud, Collins, Sautour, Sauriau, De Montaudouin, Coustau, Combes and Berthe2002; Skovgaard and Saiz, Reference Skovgaard and Saiz2006). Implication of zooplankton species in the life-cycle of other protozoan bivalve parasites, such as Bonamia ostreae (Haplosporidia), are also currently being studied (Lynch et al. Reference Lynch, Armitge, Wylde, Mulcahy and Culloty2006a, Reference Lynch, Armitge, Coughtan, Mulcahy and Cullotyb). The results of this current study suggest the existence of closely-linked dynamics for the parasite M. refringens in mussel populations and the zooplankton community in the Alfacs Bay ecosystem.
Prevalence values of M. refringens in M. galloprovincialis showed a seasonal cycle. The results suggest that this seasonal cycle is related to the annual water temperature cycle, as already proposed by other authors (Grizel, Reference Grizel1977; Balouet, Reference Balouet1977; Berthe et al. Reference Berthe, Pernas, Zerabib, Haffner, Thebault and Figueras1998, Reference Berthe, Le Roux, Adlard and Figueras2004; Audemard et al. Reference Audemard, Barnaud, Collins, Le Roux, Sauriau, Coustau, Blachier and Berthe2001). The lowest M. refringens prevalence was observed in January, one of the coldest months, whereas in spring, when water temperatures reached values higher than 17°C in April, M. refringens prevalences also increased to double the March values by the end of April. The maximum annual prevalence value was reached in July, which corresponded to the hottest month of the studied period, and prevalences of infection began to decrease again during August. Nevertheless, an unexpected and sudden decrease of prevalence occurred from 13·34% in April to 6·67% in May, while temperatures were increasing to 25°C. This phenomenon could be explained by a release of mature sporangia to the environment. This hypothesis is supported by observation of histological slides from mussels collected during this period that also coincided with the month for which reported intensities were the highest (in April). It could be that, after releasing all mature sporangia, some animals could recover or, alternatively, animals could keep some latent stages of the parasite which are not easy to detect by histopathology.
Compared to the dynamics of M. refringens in flat oysters, in which mature stages are usually not observed during the winter time (Balouet, Reference Balouet1977), mature sporangia could be detected in mussels throughout the year in the present study. This observation is in line with previous observations (Robledo and Figueras, Reference Robledo and Figueras1995). This difference could support the hypothesis of a differentiated ecological behaviour for both types of M. refringens (Carrasco et al. Reference Carrasco, Arzul, Furones, Chollet, Robert, Joly and Berthe2005) or a differentiated susceptibility of the hosts to the infection.
Previous studies undertaken in the Delta de l'Ebre bays have suggested the involvement of zooplankton in the M. refringens life-cycle in this particular ecosystem (Carrasco et al. Reference Carrasco, López-Flores, Alcaraz, Furones, Berthe and Arzul2007). In the present study, using nested PCR, zooplankton appeared to be infected by M. refringens mostly during the spring months, when prevalences began to increase just before the peak of parasitism. This period may correspond to the infection period that some authors suggest for M. refringens in flat oysters, when temperatures reach more than 17°C from June to August (Balouet, Reference Balouet1977; Audemard et al. Reference Audemard, Barnaud, Collins, Le Roux, Sauriau, Coustau, Blachier and Berthe2001). In Alfacs Bay, water temperatures reached 16°C at the end of March, when the first zooplankton sample appeared infected in spring. Previous observations on Marteilia spp. dynamics in mussel populations from the Delta de l'Ebre bays (Carrasco et al. Reference Carrasco, López-Flores, Alcaraz, Furones, Berthe and Arzul2007) also supported the fact that new infections could occur between the end of May and August. Similar observations in cultivated mussels from Galicia were recorded by Robledo and Figueras (Reference Robledo and Figueras1995). During the months preceding the increase of prevalence, the parasite may infect zooplankton populations, thereby leading to further transmission.
The hypothetically important role played by the zooplankton community in the transmission of the parasite would also be supported by the relative proportion of the different developmental stages of M. refringens in mussel samples in spring. In April, the month when most of the zooplankton samples were parasitized, mussels were infected mainly by mature stages of the parasite and, as discussed previously, mature stages were released into the environment with mussel faeces, that in turn could infect zooplankton (Audemard, Reference Audemard2001). In June, parasite stages observed in mussels were only represented by primordial stages, probably because of recent new infection. Similarly for July, only young, and not mature, M. refringens stages were detected in half of the parasitized individuals.
Other infected zooplankton samples were detected in November and December 2004 and September 2005. In September 2005, temperatures were still over 20°C and the infection period was possibly still running. However, the prevalences of M. refringens were beginning to decrease in mussels in August. This decrease shows that further transmission of the parasite was not occurring during this period.
Two other infected zooplankton samples were detected in November and December 2004. These months are supposed to be outside the infection period with temperatures between 10 and 15°C and the occurrence of low M. refringens prevalences in mussels (around 6%). When Audemard and co-workers (Reference Audemard, Sajus, Barnaud, Sautour, Sauriau and Berthe2004) studied the dynamics of M. refringens in Acartia grani populations in the ‘claires’ system between May and October, one positive sample could still be detected in September. This event was attributed to two possible phenomena: a false positive due to the presence of the parasite in the digestive tract following ingestion or its presence on some external part of the body; or the lack of development of the infective stage for the bivalves and production of resistant eggs (Audemard et al. Reference Audemard, Sajus, Barnaud, Sautour, Sauriau and Berthe2004). In any case, the hypothesis of the role of zooplankton populations as alternative hosts cannot be discarded until the transmission of the parasite is experimentally demonstrated.
Nested PCR performed on zooplankton taxa after sorting, revealed the presence of M. refringens in the species which were the most abundant in the Ebre delta. These species also occur in most of the semi-landlocked marine areas in temperate systems, since harbours, estuaries and enclosed bays are usually occupied by groups of congeneric copepods, most of them belonging to the genera Acartia (Alcaraz, Reference Alcaraz1979) and Oithona (Turner, Reference Turner2004), as well as by Appendicularians and larval stages of benthonic organisms, decapod crustaceans, cirripeds, molluscs, amongst others (Riley et al. Reference Riley, Stommel and Bumpus1949). These zooplankton species are characterized by their ability to feed on a wide size-range of suspended food particles (from <1 μm, Appendicularians, to >50 μm, Oithona), such as phytoplankton, components of the microbial loop, detritus, faecal pellets, etc. Some are non-selective filter-feeders, like the mollusc larvae, whereas others create feeding currents that entrap food particles in their capture area, or they are ambush predators preying on relatively large, mobile prey (Alcaraz, Reference Alcaraz1980). The size range of Marteilia (8–40 μm) fits into the optimum feeding size-range of the zooplankton community that occupies the Ebre Delta bays. Previous studies have demonstrated the experimental infection of A. grani by M. refringens via feeding, and differences in the susceptibility of the copepod to the M. refringens type would explain the advance or suppression of the infective process (Carrasco et al. Reference Carrasco, Arzul, Furones, Chollet, Robert, Joly and Berthe2005). Consequently, further studies, including the use of in situ hybridization, will help to discriminate real infections from accidental, cul de sac infections caused by ingestion. Complementary experimental transmission experiments will also be required to demonstrate the nature of zooplankton as potential intermediary hosts as opposed to their ability to act just as an alternative host.
Six different zooplankton taxa appeared infected by M. refringens using nested PCR, including copepods (3 Calanoida, A. discaudata, A. clausi and A. italica; 1 Cyclopoida, Oithona sp.; and 1 Harpacticoida, Euterpina acutifrons), and zoea larval stages of Brachyuran decapods (probably Portumnus sp.). Copepod species more frequently found infected in the present study were A. discaudata and Oithona sp. This is the first detection of M. refringens in A. discaudata, a congeneric species of A. grani, which both coexist in most coastal, landlocked marine systems with similar and superimposed ecological niches (Alcaraz, Reference Alcaraz1983). Oithona sp., a fundamental component of zooplankton in similar environments, was already reported as parasitized in a previous study carried out in this same area (Carrasco et al. Reference Carrasco, López-Flores, Alcaraz, Furones, Berthe and Arzul2007). Moreover, this genus belongs to the Cyclopoida which appeared as infected in the study undertaken in the ‘claire’ model (Audemard et al. Reference Audemard, Barnaud, Collins, Le Roux, Sauriau, Coustau, Blachier and Berthe2001).
This study shows that the dynamics of M. refringens in mussels and zooplankton appeared to be linked in Alfacs Bay. A number of zooplankton species were found to be infected by the parasite using nested PCR. Nevertheless, complementary transmission experiments are required to test the involvement of these species as potential intermediate hosts in the parasite life-cycle. However, if these studies do not give conclusive results, other non-planktonic species may need to be taken into consideration as potential hosts in the life-cycle of M. refringens, such as Cereus pendunculatus (Cnidaria) and Lineus gisserensis (Nematoda) that were previously found infected in the ‘claire’ model (Audemard et al. Reference Audemard, Le Roux, Barnaud, Collins, Sautour, Sauriau, De Montaudouin, Coustau, Combes and Berthe2002).
This work was carried out with funding from an INIA pre-doctoral grant from the Spanish Government and a Marie Curie training site UE grant to N. C., and INIA project RTA04-023 co-ordinated by Dr M. Ramón. Also an IFAPA post-doctoral grant (Junta de Andalucia) to I.L. and MCYT grants CTM2004-02575/MAR and REN2002-04151-C02-01/MAR to M. A. We would like to thank Dr T. Renault and Dr P. Goulletquer for their welcome in the laboratory of IFREMER-La Tremblade, as well as the IRTA staff M. Fernández for environmental data, L. Elandalousi, K. Andre and A. Roque for their advice with molecular tools, C. Rodgers for English revision, and J. Diogene who coordinated the field samplings. Finally, we wish to thank the skippers J. M. Reverté and M. Rosales, as well as the students Joanna Gil and Susana Sanchez for collaborating with samplings, Guillem Riba for informatics advise and especially Olga Bellot and Noèlia Gras for technical support.
