In situ identification and localization of bacteria associated with ...

2 downloads 0 Views 389KB Size Report
betaproteobacteria. Large clusters of Cytophaga-Flavobacterium-Bacteroides (CFB) were labelled and observed in the cytoplasm of the dinoflagellate cells, but ...
Eur. J. Phycol. (2002), 37 : 523–530. # 2002 British Phycological Society DOI : 10.1017\S0967026202003955 Printed in the United Kingdom

523

In situ identification and localization of bacteria associated with Gyrodinium instriatum (Gymnodiniales, Dinophyceae) by electron and confocal microscopy

E L S A A L V E R C A1 *, I S A B E L L E C . B I E G A L A2 *, G A B R I E L L E M . K E N N A W A Y3 , J A N E L E W I S3 A N D S U S A N A F R A N C A1 " LME, Instituto Nacional de SauT de Dr Ricardo Jorge, Av. Padre Cruz, 1649-016 Lisbon codex, Portugal # Station Biologique de Roscoff, CNRS UPR 9042, UniversiteT Pierre et Marie Curie, Place Georges Teissier, BP 74, 29682 Roscoff Cedex, France $ Phytosciences Research Group, School of Biosciences, University of Westminster, 115 New Cavendish Street, London W1M 8JS, UK (Received 10 March 2002 ; accepted 2 August 2002) The presence of intracellular bacteria in the dinoflagellate Gyrodinium instriatum Freudenthal & Lee has previously been described but the bacterial flora associated with this species has not been characterized. In this study, new results of transmission electron microscopy (TEM) and in situ hybridization using several bacterial group-specific oligonucleotide probes are presented. The long-term association of endocytoplasmic and endonuclear bacteria with G. instriatum has been confirmed. All endonuclear and most of the endocytoplasmic bacteria labelled were identified as belonging to the betaproteobacteria. Large clusters of Cytophaga-Flavobacterium-Bacteroides (CFB) were labelled and observed in the cytoplasm of the dinoflagellate cells, but were absent from the nucleus. Gammaproteobacteria were only observed outside the dinoflagellates. No alphaproteobacteria were detected either free-living or intracellular. Empirical observation of intracellular CFB reflected a degradation process of moribund dinoflagellate cells, whereas the systematic colonization of dinoflagellate nucleoplasm by betaproteobacteria suggested a true symbiotic relationship. Natural colonization may have occurred, perpetuated by vertical transmission of intracellular bacteria to the dinoflagellate daughter cells, via a pool of bacteria sequestered within the nucleus. Dividing bacteria were observed in the nucleus and equilibrium may be maintained by release of endonuclear bacteria to the cytoplasm through nuclear envelope constrictions. Key words : confocal microscopy, dinoflagellates, Gyrodinium instriatum, in situ hybridization, intracellular bacteria, oligonucleotide probes, TEM, ultrastructure

Introduction Bacteria and dinoflagellates are important members of the coastal plankton and their metabolism has a major input to pelagic energy flow and nutrient cycling through the water column (Cole, 1982 ; Doucette, 1995). Like other protozoa, both nonphotosynthetic and photosynthetic dinoflagellates can actively prey on bacteria and other protozoa (Schnepf & Elbra$ chter, 1992 ; Jacobson & Anderson, 1996 ; Uchida et al., 1997 ; Skovgaard, 2000). Besides trophic interactions between the two organisms, bacteria and dinoflagellates can also develop close associations. The occurrence of bacteria-like particles within cells of cultured dinoflagellates was first described by Silva (1962). Subsequent transmission electron microscopy (TEM) studies proCorrespondence to : E. Alverca. Fax j351 21 7526400. e-mail elsa.alverca!insa.min-saude.pt * These two authors have contributed equally to this work.

vided further evidence of the presence of bacterialike structures within several heterotrophic and photosynthetic dinoflagellate species (Gold & Pollingher, 1971 ; Silva, 1978 ; Lucas, 1982 ; Rausch de Traubenberg et al., 1995 ; Doucette et al., 1998 ; Lewis et al., 2001). Generally these relationships have been postulated to be ‘ symbiotic ’ as the two organisms coexist without apparent symptoms and in some cases the association is maintained for a long time. Additionally, it has been suggested that bacteria associated with algal cells may play a direct or indirect role in phycotoxin production (Silva, 1982 b ; Kodama et al., 1990 ; Franca et al., 1995 ; Gallacher et al., 1997 ; To$ be et al., 2001). Gyrodinium instriatum Freudenthal & Lee is an athecate photosynthetic dinoflagellate, which is often abundant in seawater and lagoons or inlets along the Portuguese coast. TEM observations of this species showed the occurrence of numerous intracellular bacteria both in the cytoplasm and in

Manz et al. (1996)

Manz et al. (1992)

Manz et al. (1992)

Materials and methods Culture conditions

HRP, horseradish peroxidase ; ISH, in situ hybridization. aEscherichia coli numbering of the ribosomal RNA operon.

50 16S (319–336) TGGTCCGTGTCTCAGTAC Cytophaga-Flavobacterium cluster of CFB phylum CF319a

50 23S (1027–1043) GCCTTCCCACATCGTTT Gammaproteobacteria

GCCTTCCCACTTCGTTT

GAM42a

50 23S (1027–1043)

HRP HRP – 50 36 36 GCTGCCTCCCGTAGGAGT CGTTCG(C\T)TCTGAGCCA CGTTC(A\C)TTCTGAGCCA

Eubacteria Alphaproteobacteria Mitochondria that have one mismatch with ALF1b Betaproteobacteria EUB338R ALF1b ALF1b competitor BET42a

16S (338–355) 16S (20–35) 16S (20–35)

% formamide in ISH buffer Targeta site (rRNA positions) Sequence (5h-3h) of probe Specificity Probe

Table 1. Oligonucleotide probes used

the nucleoplasm (Silva, 1982 a). However, this was not the case for all clonal cultures isolated from several natural populations, as some clones were found without endonuclear bacteria (Silva & Franca, 1985). To further our understanding of the interactions between dinoflagellates and their associated bacteria the identification of intracellular and free bacteria is essential. Identification of bacterial groups in marine environments has been made possible in the past decade by the application of molecular techniques (Giovannoni et al., 1990). Bacterial associations with dinoflagellates have often been probed with destructive techniques, involving DNA extraction and sequence analysis (Doucette & Tricks, 1995 ; Seibold et al., 2001 ; Hold et al., 2001 a, To$ be et al., 2001). Recently, Biegala et al. (2002) used non-destructive techniques, i.e. whole cell hybridization of 16S rRNA-targeted oligonucleotide probes in conjunction with confocal microscopy, to identify attached or intracellular bacteria within dinoflagellates. One of the major advantages of the latter technique was to allow rapid identification and three-dimensional localization of bacteria within photosynthetic host cells. The aim of the present study was to localize and identify at class level the free and intracellular bacteria associated with a strain of Gyrodinium instriatum maintained in laboratory culture for 18 years.

HRP or not when used as competitor HRP or not when used as competitor HRP

Amann et al. (1990) Modified from Manz et al. (1992) Simon et al. (unpublished)

524

HRP-labelled

Reference

E. Alverca et al.

The strain of Gyrodinium instriatum (LME 184, Lisbon, Portugal) used in this study was isolated in 1982 from Sto Andre! ’s Lagoon, Portugal (Silva & Franca, 1985). The culture was maintained in a mixture of Provasoli’s media ASP1jASP2 (1 : 1) (Provasoli, 1963) until 1998, when Guillard’s medium f\2 was used (Sigma-Aldrich ; Guillard & Ryther, 1962). The culture was maintained at 19–21 mC on a 14 : 10 light : dark cycle. Cell fixation Cells were typically harvested for fixation in mid- and late-exponential growth phase. For ultrastructural studies, a sample of the dinoflagellate culture was pre-fixed in 0n1 % glutaraldehyde (Merck) for 1 h at room temperature. Cells were then concentrated by centrifugation and fixed for 1 h at room temperature with 4 % paraformaldehyde (Merck), 3 % glutaraldehyde (Karnowsky, 1965), in 0n1 M PIPES buffer (piperazine N,Nh-bis 2ethanosulphonic acid, Sigma-Aldrich) with 14 % sucrose and rinsed twice in 0n1 M PIPES buffer with 14 % sucrose. The secondary fixative was made up of two solutions : (1) 1 % OsO (EMS) made up in water to which 3 % % potassium ferricyanide w : v (Sigma-Aldrich) was added and (2) 6 % sodium iodate (Sigma-Aldrich) made up in 0n1 M PIPES buffer. Equal volumes of the two solutions

Bacteria associated with G. instriatum

525

Figs 1–6. Transmission electron micrographs of Gyrodinium instriatum. Fig. 1. Clusters of bacteria (arrows) in the cytoplasm (C) and dividing bacteria (arrowhead). Fig. 2. Intranuclear bacterial clusters in the nucleus (N) periphery. Note isolated bacteria (arrow) among the chromosomes (Chr). Fig. 3. DNA fibrils (arrow) extending from the dinoflagellate chromosome (Chr) to the bacteria (Ba). Fig. 4. Bacteria being released from the nucleus (N) to the cytoplasm (C) (arrow). Fig. 5. Dividing bacteria (arrowhead) inside the nucleus (N). Fig. 6. Degrading bacteria (arrow) in cytoplasmic vacuole. Scale bars represent 1 µm.

were combined (A. Page, personal communication). Samples were left overnight at room temperature or at 4 mC. Cells were rinsed in water, dehydrated in an ethanol series and embedded in Spurr resin (Agar Scientific ; Spurr, 1969). For fluorescence in situ hybridization (FISH), cells were fixed with 1 % paraformaldehyde in PBS buffer (phosphate-buffered saline) at 4 mC for 24 h, collected on inorganic filter membranes and dehydrated in an ethanol series (Amann, 1995). The filters were then stored in the dark at room temperature to await FISH experiments.

Probes Oligonucleotide probes (listed in Table 1) were obtained from Interactiva (St. Malo, France) and labelled with HRP (Horseradish Peroxidase) (Roche Diagnostics) as described by Urdea et al. (1988) and Amann et al. (1992). Fluorescence in situ hybridization The protocol used for TSA-FISH (tyramide signal amplification–fluorescence in situ hybridization) was

E. Alverca et al. according to Biegala et al. (2002). The hybridized cells immobilized on filters were kept at 4 mC in the dark. Confocal observations were made within 2 weeks of preparation without significant loss of fluorescence. Microscopy Ultrathin sections were cut on an LKB 8800 Ultratome III microtome (Bromma, Sweden) with a Diatome diamond knife (Bienne, Switzerland) and counterstained with uranyl acetate and lead citrate (Reynolds, 1963). Stained grids were examined on a Philips EM 301 TEM (Eindhoven, The Netherlands). Fluorescence images were acquired with a confocal laser scanning microscope (CLSM) Fluoview (Olympus Optical, Tokyo, Japan) equipped with an argon–krypton laser (643-OLYM-A03 Omnichrome, Melles Griot, Carlsbad, CA, USA) and a pulsed laser (Mira 900, Coherent, Santa Clara, CA, USA).

Results TEM TEM observations revealed the presence of numerous intracellular bacteria in the majority of the cell sections, both in the cytoplasm (Fig. 1) and in the nucleus (Figs 2, 3). Endonuclear bacterial structures were observed dispersed throughout the nucleoplasm, aggregated in dense clusters, less frequently isolated between the chromosomes and usually surrounded by a transparent halo (Fig. 2). Endonuclear bacteria were mainly observed at the periphery of the nucleus adjacent to the nuclear envelope, while the chromosomes were aggregated in a more central region of the nucleus (Fig. 2). Few bacteria established close connections with the chromosomes. When such cases occurred DNA fibrils could be seen extended towards the bacteria through an electron-transparent zone (Fig. 3). Occasionally, invaginations of the nuclear envelope were observed enclosing single bacteria or clusters of bacteria that appeared to be about to be released into the cytoplasm by constriction of the nuclear membrane (Fig. 4). Actively dividing bacteria were observed in the nucleus (Fig. 5) and more rarely in the cytoplasm (Fig. 1). Bacteria in the cytoplasm were organized in clusters surrounded by a host membrane or, alternatively, isolated among the dinoflagellate organelles and separated from them by a transparent region (Fig. 1). Furthermore, some sections showed the presence of cytoplasmic vacuoles containing electron-dense masses and concentric multilamellar structures (Fig. 6). Fluorescence in situ hybridization Endocytoplasmic and endonucleoplasmic bacteria were successfully labelled with the EUB338R probe (specific for the Bacteria domain) in all Gyrodinium instriatum cells examined (Fig. 7). Hybridized bac-

526 teria were abundant in the nucleus and cytoplasm and were present either singly or in clusters. Bacteria present in dividing dinoflagellates were also strongly labelled with this probe (Fig. 8 a). Hybridization with probes (Table 1) specific for gram-negative alpha-, beta- and gammaproteobacteria and Cytophaga-Flavobacterium-Bacteroides (CFB) showed that endonucleoplasmic and endocytoplasmic bacteria were mainly labelled by the BET42a probe (Figs 9 a, 10 a, 11 a, 12 a). Endonuclear betaproteobacteria were mostly observed at the nuclear periphery, although a few were located between the dinoflagellate chromosomes (Fig. 10 a). No intracellular alpha- and gammaproteobacteria were detected in any of the cells examined (Figs 9 a, 11 a). Some dinoflagellates showed intense endocytoplasmic fluorescence when hybridized with the probe targeting CFB, although no endonuclear labelling was observed. Bacteria labelled with the CF319a probe were associated in large clusters (15i5 µm ; Fig. 12 a). All extracellular bacteria were labelled with the EUB338R probe (data not shown), and most belonged to the beta- and gammaproteobacteria. Very few extracellular bacteria were labelled with the CF319a probe and none were labelled with the alphaproteobacteria. Discussion The presence of intracellular bacteria in G. instriatum has been described at a morphological level in numerous studies (Silva, 1982 a, b ; Silva & Franca, 1985). However, the present work identifies for the first time the bacterial flora associated with this dinoflagellate using whole-cell hybridization of 16S rRNA oligonucleotide probes. Among the different classes of eubacteria probed, representatives of betaproteobacteria were found to be the dominant microorganisms intracellularly and extracellularly. Furthermore, they were the only group of bacteria identified within the nucleoplasm of G. instriatum. This result is surprising as endosymbiotic bacteria of eukaryotic organisms usually belong to the alpha and gammaproteobacteria (Garrity, 2001). Nevertheless, symbiotic relationships of betaproteobacteria with some groups of insects have been described by Du et al. (1994), Fukatsu & Nikoh (2000) and Von Dohlen et al. (2001). To our knowledge, the present study reports for the first time the presence of intracellular betaproteobacteria in dinoflagellates or other free-living protists (Fokin et al., 1996 ; Fritsche et al., 1999 ; Seibold et al., 2001 ; Schweikert & Meyer, 2001). Betaproteobacteria have been considered rare in marine ecosystems but dominate freshwater environments (Glo$ ckner et al., 1999). Nevertheless, betaproteobacteria have recently been observed at high concentrations in marine sediments (Nold et al., 2000).

Bacteria associated with G. instriatum

527

Figs 7–12. Confocal laser scanning microscope 0n7 µm optical sections through Gyrodinium instriatum cells immobilized on a filter. Figs 7 a, 8 a, 9 a, 10 a, 11 a, 12 a excited at 488 nm and hybridized with specific probes (Table 1) labelled with fluorescein using TSA-FISH technique. Figs 7 b, 8 b, 9 b, 10 b, 11 b, 12 b excited at 380 nm and stained with DAPI. Fig. 7 a. Intracellular bacteria in the cytoplasm (C) and in the nucleus (N) labelled with the EUB338-HRP probe, present both isolated and in clusters. Fig. 8 a. Dividing cell showing the same distribution of labelled intracellular eubacteria as observed in Fig. 7 a. Fig. 9 a. Optical section of a cell with no intracellular bacteria labelled with the ALF1b-HRP probe. Fig. 10 a. Three dinoflagellate cells with endocytoplasmic and endonuclear bacteria (arrows) labelled with the BET42a-HRP probe. Fig. 11 a. Optical section of two dinoflagellates with no intracellular bacteria labelled with GAM42a probe. Fig. 12 a. Large clusters of endocytoplasmic bacteria labelled with the CF319a-HRP probe. Scale bars represent 20 µm.

Marine bacterioplankton is dominated by alphaand gammaproteobacteria as well as CFB (Glo$ ckner et al., 1999 ; Cottrell & Kirchman, 2000 ; Hagstro$ m et al., 2000 ; Riemann et al., 2000), which occur either as free-living organisms or attached to phytoplankton. These three classes of bacteria have already been found associated with different species of dinoflagellates (Lafay et al., 1995 ; Hold et al., 2001 a). A recent study (Hold et al., 2001 b) indicated that the majority of the bacteria associated with four different dinoflagellate clonal cultures were alphaproteobacteria, and fluctuations in abundance were observed at different phases of the growth cycle. The absence of alphaproteobacteria either in the culture medium or associated with this clone of G. instriatum is therefore surprising. There are several possible explanations for this discrepancy. Firstly the results produced in the present study may reflect the bacterial community present at the time of sampling for hybridization experiments.

Secondly, the dominance of alphaproteobacteria in dinoflagellate cultures, as previously mentioned by other authors, might not reflect their real abundance in nature. All these studies used polymerase chain reaction (PCR)-based techniques, which are known to preferentially amplify a particular genus or group of target organisms. And finally, in natural environments alphaproteobacteria have been described attached to live or detrital particles less frequently than CFB and gammaproteobacteria (Riemann et al., 2000 ; Cottrell & Kirchman, 2000). It is thus probable that when G. instriatum cells were isolated for the establishment of this clonal culture, no alphaproteobacteria were attached to the dinoflagellates. In contrast, CFB, gamma and betaproteobacteria may have been recovered from nature in the clonal culture. In a few G. instriatum cells, strong cytoplasmic colonization by CFB was observed. Mixotrophy is a nutritional strategy quite common among different

E. Alverca et al. dinoflagellate species (Hansen, 1991 ; Schnepf & Elbra$ chter, 1992 ; Jacobson & Anderson, 1996 ; Skovgaard, 2000), including G. instriatum (Uchida et al., 1997). Therefore, it is possible that the observed cytoplasmic CFB were taken up as prey and maintained as isolated clusters in the cytoplasm. However, these bacteria were rare as free-living organisms in the culture medium, making this hypothesis unlikely. Additionally, the FISH signal observed was very strong, which would not be the case if the target cells were being digested. It seems more reasonable to assume that the large CFB clusters observed corresponded to colonization of senescent or dead dinoflagellate cells by this group of bacteria. This bacterial strategy is common within the CFB group, which have been shown to develop after the collapse of a phytoplankton bloom (Riemann et al., 2000). Additionally, from pureculture studies, it is known that members of this class are able to aerobically degrade a large spectrum of substrates ranging from various proteins, carbohydrates, pesticides and insecticides to complex macromolecules (Bernardet et al., 1996). Although the latter hypothesis is the most probable, an intimate association between these two organisms, such as a symbiosis or parasitism, cannot be ruled out. It has recently been reported that members of the CFB group establish symbiotic relationships with other protozoa such as amoebae (Horn et al., 2001). The nature of the relationship between G. instriatum cells and bacteria from the CFB is unknown, but it seems quite different from the one established with betaproteobacteria, both in the number of dinoflagellate cells colonized (it represents only 2 % of the observed cells versus 100 % of cells with betaproteobacteria) and in the intracellular location of the two bacterial types. The presence of endonuclear bacteria in dinoflagellates is rare, and to our knowledge has been reported in a limited number of species (Silva, 1978 ; Silva & Franca, 1985). The origin of the endonuclear betaproteobacteria in G. instriatum is not fully understood. It can be hypothesized that under certain environmental conditions, such as low nutrient levels, free-living bacteria may be taken up as prey, with those colonizing the nucleus avoiding digestion in the cytoplasm. This route of nuclear infection has been previously suggested by Go$ rtz (1986) to explain the origin of endonucleobiosis in ciliates. Nuclear colonization by prokaryotic microbes is commonly observed in protozoa such as euglenoids (Leedale, 1969) and ciliates (Go$ rtz et al., 1989 ; Kawai & Fujishima, 2000). Intranuclear bacteria may have advantages over those that colonize the cytoplasm because : (1) they ensure stability of the association by homogeneous distribution to daughter cells during cell division – in G. instriatum the nuclear envelope remains intact

528 during mitosis and (2) nuclear bacteria could benefit from the nuclear pools of metabolites, renewed at each host division (Go$ rtz, 1986). TEM observations showed dividing bacteria within the nucleus of dinoflagellate cells (Fig. 5), without apparent prejudice to the host’s normal growth, which is similar to the growth rate of clonal cultures without endocellular bacteria (data not shown). Similar TEM observations have been made on this species by Silva & Franca (1985), who suggested the maintenance of a stable equilibrium between the two organisms. As described by these authors, a decrease in the host’s division rates may lead to an increase in intranuclear bacteria, some of which are subsequently expelled into the cytoplasm via vesicles pinched off from the nuclear membrane, as a reaction of the host. Supporting this hypothesis are TEM observations showing nuclear envelope constrictions enclosing bacteria (Silva & Franca 1985, figures therein and Fig. 4 in this study), indicating that the nucleus in this G. instriatum clone may act as a reservoir of bacteria which will be either ‘ expelled ’ or digested in the cytoplasm according to the host’s needs (Silva, 1982 b). Some endonuclear bacteria were observed dispersed among or closely apposed to dinoflagellate chromosomes through host DNA fibrils. Similar observations have been made in ciliates, and it has been suggested that certain bacterial endonucleobionts digest host chromatin (Go$ rtz, 1986). Gortz further suggested that in a densely colonized nucleus, the destruction of the chromatin would be significant and probably fatal to the host. The G. instriatum clone under study has been successfully maintained in culture without loss of its endonuclear bacteria for a long period. Therefore, it seems unlikely that G. instriatum possesses endonuclear bacteria that feed directly on chromatin, unless it has an adequate chromatin repair mechanism ensuring cell viability. The transfer of prokaryotic DNA to a eukaryote has been previously reported (Taylor, 1979). The presence of abundant intranuclear bacteria and the close connection established by some with dinoflagellate chromosomes may lead to speculation about an eventual uptake of prokaryotic DNA, or its export and translation by the host. Besides its evolutionary implications, it could act as a mechanism through which endonuclear bacteria induce the production of special metabolites by the host cell. This aspect may be of particular relevance in harmful algal species. The microscopic study of Gyrodinium instriatum cells confirmed the maintenance of the association of this dinoflagellate with intracellular bacteria over a period of 18 years. As the cells were not cultured axenically, the possibility exists that during this period bacteria were taken up and released from\

Bacteria associated with G. instriatum into the medium. However, the long-term presence of intranuclear bacteria suggests a symbiotic relation between the dinoflagellates and these bacteria. Further studies on interactions between dinoflagellates and bacteria are necessary to improve understanding of the way both organisms survive and benefit from the association. Since one of the sources of genetic novelty has been identified as being the repeated fusion of bacterial endosymbionts with host cells (Taylor, 1979), these studies may help to give a better insight into evolutionary processes in the Dinophyceae. Sequencing of bacterial ribosomal genes and the design of probes should be pursued and applied to both cultured and natural samples, thereby enlarging the limited information available about the natural bacterial flora associated with dinoflagellates. Although toxicity tests were not part of this study, this species has previously been described as being toxic (Silva, 1982 b). Therefore, an approach combining the techniques used here with toxin detection testing could add much relevant information regarding the involvement of bacteria in the phenomenon of harmful algal blooms (HAB). Acknowledgements We gratefully acknowledge Dr J. F. Lennon for assistance with confocal microscopy and Dr D. Vaulot for helpful discussion and critical reading of the manuscript. The EC Project FAIR CT96-1558 supported this work. References A, R.I. (1995). In situ identification of microorganisms by whole cell hybridization with rRNA-targeted nucleic acid probes. Mol. Microbiol. Ecol. Manual, 3.3.6 : 1–15. A, R.I., K, L. & S, D.A. (1990). Fluorescentoligonucleotide probing of whole cells for determinative, phylogenetic, and environmental studies in microbiology. J. Bacteriol., 172 : 762–770. A, I., Z, B., S, D.A. & S, K.-H. (1992). Identification of individual prokaryotic cells by using enzymelabeled, rRNA-targeted oligonucleotide probes. Appl. Env. Microbiol., 58 : 3007–3011. B, J.F., S, P., V, M., B, F., K, K. & V, P. (1996). Cutting a gordian knot : emended classification and description of the genus Flavobacterium, emended description of the family Flavobacteriaceae, and proposal of Flavobacterium hydatis nom. nov. (basionym, Cytophaga quatilis Stohl and Tait 1978). Int. J. Syst. Bacteriol., 46 : 128–148. B, I., K, G., A, E., L, J., V, D. & S, N. (2002). Identification of bacteria associated with dinoflagellates (Dinophyceae) Alexandrium spp. using Tyramide Signal Amplification–Fluorescent in situ Hybridization and confocal microscopy. J. Phycol., 38 : 404–411. C, J.J. (1982). Interactions between bacteria and algae in aquatic ecosystems. Annu. Rev. Ecol. Syst., 13 : 291–314. C, M. & K, D. (2000). Natural assemblages of marine proteobacteria and members of the Cytophaga-Flavo-

529 bacter cluster consuming low- and high-molecular-weight dissolved organic matter. Appl. Environ. Microbiol., 66 : 1692–1697. D, G.J. (1995). Interactions between bacteria and harmful algae : a review. Nat. Toxins, 3 : 65–74. D, G.J. & T, C.G. (1995). Characterization of bacteria associated with different isolates of Alexandrium tamarense. In Harmful Marine Algal Blooms (Lassus, P., Arzul, G., ErcaredLeDenn, E., Gentien, P. & Marcaillou-Le Baut, C., editors), 33–38. Lavoisier, Paris. D, G.J., K, M., F, S. & G, S. (1998). Bacterial interactions with harmful algal bloom species : bloom ecology, toxigenesis, and cytology. In Physiological Ecology of Harmful Algal Blooms. NATO ASI series vol. G41 (Anderson, D.M., Cembella, A.D. & Hallegraef, G.M., editors), 619–647. Springer, Berlin. D, Y., ML, G. & C, K.-P. (1994). 16S ribosomal sequence identities of β-proteobacterial endosymbionts in three Crithidia species. J. Bacteriol., 176 : 3081–3084. F, S., B, T., B, J. & G$ , H. (1996). Holospora species infecting the nuclei of Paramecium appear to belong into two groups of bacteria. Eur. J. Protistol., 32 : 19–24. F, S., V, S., M, V., P, L. & D, G.J. (1995). Prokaryotes in association with a toxic Alexandrium lusitanicum in culture. In Harmful Marine Algal Blooms (Lassus, P., Arzul, G., Ercared-LeDenn, E., Gentien, P. & Marcaillou-Le Baut, C., editors), 46–51. Lavoisier, Paris. F, T., H, M., S, S., G, R., S, K. & W, M. (1999). In situ detection of novel bacterial endosymbionts of Acanthamoeba spp. phylogenetically related to members of the order Ricckettsiales. Appl. Environ. Microbiol., 65 : 206–212. F, T. & N, N. (2000). Endosymbiotic microbiota of the bamboo pseudococcid Antonina crawii (Insecta, Homoptera). Appl. Environ. Microbiol., 66 : 643–650. G, S., F, K.J., F, J.M. & B H, H.B. (1997). Evidence for production of paralytic shellfish toxins by bacteria associated with Alexandrium spp. (Dinophyta) in culture. Appl. Env. Microbiol., 63 : 239–245. G, G.M. (2001). Bergey’s Manual of Systematic Bacteriology, 2nd edition. Springer, New York. G, S.J., B, T.B., M, C.L. & F, K.G. (1990). Genetic diversity in Sargasso Sea bacterioplankton. Nature, 345 : 60–63. G$ , F.O., F, B.M. & A, R. (1999). Bacterioplankton compositions of lakes and oceans : a first comparison based on fluorescence in situ hybridization. Appl. Environ. Microbiol., 65 : 3721–3726. G, K. & P, U. (1971). Occurrence of endosymbiotic bacteria in marine dinoflagellates. J. Phycol., 7 : 264–265. G$ , H. (1986). Endonucleobiosis in Ciliates. Int. Rev. Cytol., 102 : 169–213. G$ , H., A, N. & R, H. (1989). Ultrastructure of the infectious and reproductive forms of Holospora obtusa, a bacterium infecting the macronucleus of Paramecium caudatum. J. Gen. Microbiol., 135 : 3079–3085. G, R.R.L. & R, J.H. (1962). Studies on marine planktonic diatoms. I. Cyclotella nana Hustedt and Detonula confervacea (Cleve) Gran. Can. J. Microbiol., 8 : 229–239. H$ , A., P, J. & Z, U.L. (2000). Biogeographical diversity among marine bacterioplankton. Aquat. Microb. Ecol., 21 : 231–244. H, P. (1991). Dinophysis – a planktonic dinoflagellate genus which can act both as a prey and a predator of a ciliate. Mar. Ecol. Prog. Ser., 69 : 201–204. H, G.L., S, E.A., B, T.H. & G, S. (2001 a). Comparison of paralytic shellfish toxin (PST) production by the dinoflagellates Alexandrium lusitanicum NEPCC 253 and Alexandrium tamarense NEPCC 407 in the presence and absence of bacteria. FEMS Microbiol. Ecol., 36 : 223–234. H, G.L., S, E.A., R! , M.S., M, E.W., M, R.B., S, C., S, J.R., P, J.I., B, T.H. & G, S. (2001 b). Characterisation of bacterial communi-

E. Alverca et al. ties associated with toxic and non-toxic dinoflagellates : Alexandrium spp. and Scrippsiella trochoidea. FEMS Microbiol. Ecol., 37 : 161–173. H, M., H, M.D., L, T., S, E.N., M, K.D., M, R. & W, M. (2001). Members of the Cytophaga-Flavobacterium-Bacteroides phylum as intracellular bacteria of acanthamoebae : proposal of Candidatus amoebophilus asiaticus. Environ. Microbiol., 3 : 440–449. J, D. & A, D. (1996). Widespread phagocytosis of ciliates and other protists by marine mixotrophic and heterotrophic thecate dinoflagellates. J. Phycol., 32 : 279–285. K, M.J. (1965). A formaldehyde–glutaraldehyde fixative of high osmolarity for use in electron microscopy. J. Cell Biol., 27 : 137A. K, M. & F, M. (2000). Invasion of the macronucleus of Paramecium caudatum by the bacterium Holospora obtusa : fates of the bacteria and timings of invasion steps. Eur. J. Protistol., 36 : 46–52. K, M., O, T., S, S., H, T. & M, T. (1990). Production of paralytic shellfish toxins by a bacterium Moraxella sp. isolated from Protogonyaulax tamarensis. Toxicon, 28 : 707–714. L, B., R, R., R  T, C., B, V., G, M. & C, R. (1995). Roseobacter algicola sp. nov., a new marine bacterium isolated from the phycosphere of the toxin-producing dinoflagellate Prorocentrum lima. Int. J. Syst. Bacteriol., 45 : 290–296. L, G.F. (1969). Observations on endonuclear bacteria in euglenoid flagellates. Oq sterr. Bot. Z., 116 : 279–294. L, J., K, G., F, S. & A, E. (2001). Bacterium–dinoflagellate interactions : investigative microscopy of Alexandrium spp. (Gonyaulacales, Dinophyceae). Phycologia, 40 : 280–285. L, I.A.N. (1982). Observations on Noctiluca scintillans Macartney (Ehrenb.) (Dinophyceae) with notes on an intracellular bacterium. J. Plankton Res., 4 : 401–409. M, W., A, R., L, W., W, M. & S, K.-H. (1992). Phylogenetic oligodeoxynucleotide probes for the major subclasses of proteobacteria : problems and solutions. Syst. Appl. Microbiol., 15 : 593–600. M, W., A, R., L, W., V, M. & S, K.-H. (1996). Application of a suite of 16S rRNA-specific oligonucleotide probes designed to investigate bacteria of the phylum Cytophaga-Flavobacter-Bacteroides in the natural environment. Microbiology, 142 : 1097–1106. N, S.C., Z, J., D, A.H. & T, J.M. (2000). Pacific Northwest marine sediments contain ammonia-oxidizing bacteria in the β subdivision of the Proteobacteria. Appl. Environ. Microbiol., 66 : 4532–4535. P, L. (1963). Growing marine seaweeds. Proc. 4th Int. Seaweed Symp., 9–17. Biarritz. R  T, C., G, M.-L., S-G, M.-O. & E, D. (1995). The toxic dinoflagellate Prorocentrum lima and its associated bacteria. Eur. J. Protistol., 31 : 318–326.

530 R, E.S. (1963). The use of lead citrate at high pH as an electron-opaque stain in electron microscopy. J. Cell Biol., 17 : 208–212. R, L., S, G. & A, F. (2000). Dynamics of bacterial community composition and activity during a mesocosm diatom bloom. Appl. Environ. Microbiol., 66 : 578–587. S, E. & E$ , M. (1992). Nutrional strategies in dinoflagellates. A review with emphasis on cell biological aspects. Eur. J. Protistol., 28 : 3–24. S, M. & M, B. (2001). Characterization of intracellular bacteria in the freshwater dinoflagellate Peridinium cinctum. Protoplasma, 217 : 177–184. S, A., W, A. & S$ , C. (2001). Diversity of endocytic bacteria in the dinoflagellate Noctiluca scintillans. Aquat. Microb. Ecol., 25 : 229–235. S, E.S. (1962). Some observations on marine dinoflagellate cultures. III. Gonyaulax spinifera (Clap. and Lach.) Dies., Gonyalax tamarensis Leb., and Peridinium trochoideum (Stein) Lemm. Notas Estudos do Inst. Biol. Mar., 26 : 1–21. S, E.S. (1978). Endonuclear bacteria in two species of dinoflagellates. Protistologica, 14 : 113–119. S, E.S. (1982 a). Relationship between dinoflagellates and intracellular bacteria. In Marine Algae in Pharmaceutical Science, vol. 2 (Hoppe, H.A. & Levring, T., editors), 269–288. Walter de Gruyter, Berlin. S, E.S. (1982 b). Toxic clones of Gyrodinium instriatum with endonuclear bacteria. Proc. Vth Int. IUPAC Symposium on Mycotoxins and Phycotoxins, Vienna, Austria, 216–219. S, E.S. & F, S. (1985). The association dinoflagellatebacteria : their ultrastructural relationship in two species of dinoflagellates. Protistologica, 21 : 429–446. S, A. (2000). A phagotrophically derivable growth factor in the plastidic dinoflagellate Gyrodinium resplendens (Dinophyceae). J. Phycol., 36 : 1069–1078. S, A.R. (1969). A low-viscosity epoxy resin embedding medium for electron microscopy. J. Ultrastruct. Res., 26 : 31–43. T, F.J.R. (1979). Symbionticism revisited : a discussion of the evolutionary impact of intracellular symbioses. Proc. R. Soc. Lond., Ser. B, 204 : 267–286. T$ , K., F, C., K, M., G, S. & M, L.K. (2001). Seasonal occurrence at a Scottish PSP monitoring site of purportedly toxic bacteria originally isolated from the toxic dinoflagellate genus Alexandrium. Eur. J. Phycol., 36 : 243–256. U, T., K, T. & M, Y. (1997). Predation by a photosynthetic dinoflagellate Gyrodinium instriatum on loricated ciliates. J. Plankton Res., 19 : 603–608. U, M.S., W, B.D., R, J.A., S, M., C, J. & H, T. (1988). A comparison of non-radioisotopic hybridization assay methods using fluorescent, chemiluminescent and enzyme labeled synthetic oligodeoxyribonucleotide probes. Nucleic Acids Res., 16 : 4937–4956. V D, C.D., K, S., A, S.T. & MM, R. (2001). Mealybug β-proteobacterial endosymbionts contain γ-proteobacterial symbionts. Nature, 412 : 433–436.