• We are sorry, but NCBI web applications do not support your browser and may not function properly. More information
Logo of aemPermissionsJournals.ASM.orgJournalAEM ArticleJournal InfoAuthorsReviewers
Appl Environ Microbiol. Jul 2005; 71(7): 3483–3494.
PMCID: PMC1169014

Phylogenetic Diversity and Specificity of Bacteria Closely Associated with Alexandrium spp. and Other Phytoplankton


While several studies have suggested that bacterium-phytoplankton interactions have the potential to dramatically influence harmful algal bloom dynamics, little is known about how bacteria and phytoplankton communities interact at the species composition level. The objective of the current study was to determine whether there are specific associations between diverse phytoplankton and the bacteria that co-occur with them. We determined the phylogenetic diversity of bacterial assemblages associated with 10 Alexandrium strains and representatives of the major taxonomic groups of phytoplankton in the Gulf of Maine. For this analysis we chose xenic phytoplankton cultures that (i) represented a broad taxonomic range, (ii) represented a broad geographic range for Alexandrium spp. isolates, (iii) grew under similar cultivation conditions, (iv) had a minimal length of time since the original isolation, and (v) had been isolated from a vegetative phytoplankton cell. 16S rRNA gene fragments of most Bacteria were amplified from DNA extracted from cultures and were analyzed by denaturing gradient gel electrophoresis and sequencing. A greater number of bacterial species were shared by different Alexandrium cultures, regardless of the geographic origin, than by Alexandrium species and nontoxic phytoplankton from the Gulf of Maine. In particular, members of the Roseobacter clade showed a higher degree of association with Alexandrium than with other bacterial groups, and many sequences matched sequences reported to be associated with other toxic dinoflagellates. These results provide evidence for specificity in bacterium-phytoplankton associations.

Bacteria and phytoplankton dynamics are thought to be closely linked, and the “phycosphere,” or region immediately surrounding and influenced by phytoplankton cells, is an important bacterial habitat that is distinct from the surrounding water. Phytoplankton excrete organic compounds that represent an important fraction of primary production, form the base of the marine microbial food web (36), and stimulate bacterial growth (51). Bacteria can be free living in the phycosphere (7), can be attached to the surface of the algal cells (34, 57), or can occur as intracellular algal symbionts (38). Bacterium-phytoplankton interactions play a key role in processes ranging from biogeochemical cycling within the microbial loop (6) to biochemically mediated interactions that influence phytoplankton growth, reproduction, cyst formation, and mortality (13). The influence of bacteria on toxic algae has been of particular interest, and bacteria have been implicated in the production and/or modification of algal toxins (29, 30, 33).

Several studies have reported evidence for species-specific interactions between bacteria and phytoplankton, which has led to the conclusion that bacteria can play a major role in controlling phytoplankton dynamics. For example, Fukami et al. (18) found that natural bacterial communities collected during a Gymnodinium nagasakiense bloom inhibited Skeletonema costatum but stimulated G. nagasakiense. Later, Fukami et al. (19) isolated Flavobacterium sp. strain 5N-3, which was found to have algicidal activity against G. nagasakiense but to have no effect on Chattonella antiqua, Heterosigma akashiwo, or S. costatum. These findings have important implications, especially if they hold true for interactions between bacteria and members of the general phytoplankton community that form the basis for carbon cycling in coastal marine environments (32).

A number of studies have shown that bacteria attached to particles are phylogenetically distinct from free-living bacteria (9, 11, 49). This finding indicates that there are selective forces that drive the community succession on particles toward a phylogenetic composition that differs from the composition in the surrounding water. Rooney-Varga et al. (49) recently reported a link between the community dynamics of phytoplankton and particle-associated bacteria in the Bay of Fundy, suggesting that species-specific associations occur in the phycosphere. However, given the complexity of coastal temperate phytoplankton communities, it has not yet been possible to determine whether links between dynamics at the community level are indeed the result of specific interactions between particular bacteria and phytoplankton and, if so, which bacterial species are associated with which phytoplankton. For example, Savin et al. (50) observed as many as 42 phytoplankton species in a surface seawater sample from the Bay of Fundy, which made it impossible to use direct molecular analysis of the attached bacterial community to determine which bacteria were associated with a particular phytoplankton species. Clearly, the process by which an associated bacterial community might be “selected” by an algal species is complex and not well understood (12).

Given the complexity of natural samples, another approach is to characterize bacteria associated with phytoplankton in culture. Many algal isolates are obtained by micropipetting a single algal cell from an environmental sample, thereby producing a clonal culture. In the absence of treatment to render the culture axenic, bacteria that were initially present in the phycosphere and are capable of growing in association with the algae are selected for with successive transfers. In the current study, we investigated the possibility of specific bacterium-phytoplankton associations by analyzing the phylogenetic diversity of bacteria associated with cultures of diverse nontoxic phytoplankton, as well as members of the toxic dinoflagellate genus Alexandrium, isolated from several regions of the world. Our analysis included a survey of bacterial associates of the major phylogenetic groups of phytoplankton common in the Gulf of Maine and similar temperate coastal environments, as well as Alexandrium species from the Gulf of Maine, Japan, Portugal, France, Russia, and the Bering Sea. The results reveal a pattern of specific bacterial associations with toxic Alexandrium species, regardless of the location of isolation.


Cultivation of phytoplankton.

The phytoplankton cultures used in this study are listed in Table Table11 and included 10 Alexandrium cultures, as well as 13 cultures that represent dominant taxa in the Gulf of Maine phytoplankton community, including members of the Dinophyceae, Coscinodiscophyceae, Chlorarchniophyceae, Prymnesiophyceae, Prasinophyceae, Eustigmatophyceae, Dictyochophyceae, and Cryptophyceae. Alexandrium spp. cultures were maintained in modified f/2 medium. The f/2 medium was prepared as described by Guillard and Ryther (27), except that it contained 10−8 M Na2SeO3 and 10−8 M CuSO4 · 5H2O. All other cultures were grown in f/2 medium (25, 27), K medium (31), L1 medium (26), Prov medium (48), or f/2-Si medium (f/2 medium with Na2SiO3 · 9H2O omitted), as shown in Table Table1.1. Phytoplankton cultures were grown at 15°C with cycles consisting of 14 of light and 10 h of darkness (for Alexandrium spp.) or of 13 h of light and 11 h of darkness (for all other cultures); the light intensity was ca. 200 μmol photons m−2 s−1. Alexandrium cultures were kindly provided by D. M. Anderson (Woods Hole Oceanographic Institute, Woods Hole, MA), and all other cultures were obtained from the Provasoli-Guillard National Center for Culture of Marine Phytoplankton at Bigelow Laboratory, Boothbay Harbor, ME. All Provasoli-Guillard National Center for Culture of Marine Phytoplankton and Alexandrium fundyense cultures were isolated from the Gulf of Maine, while Alexandrium tamarense and Alexandrium minutum cultures were isolated from France, the Bering Sea, Portugal, and Japan (Table (Table1).1). All phytoplankton cultures were maintained as unialgal, xenic cultures from the time of original isolation and, with the exception of A. fundyense CB301, were isolated from vegetative cells (not cysts or resting stages).

Phytoplankton cultures analyzed to determine the phylogenetic identities of their bacterial associates

Analysis of attached and free-living bacteria.

In order to compare the phylogenetic composition of bacteria attached to A. fundyense cells with the phylogenetic composition of free-living bacteria, a 20-μm-pore-size sieve was used to separate A. fundyense strain CB301 cells and attached bacteria from free-living bacteria in the culture. A sample of a mid-exponential-phase A. fundyense strain CB301 culture was aseptically washed over a 20-μm-pore-size sieve. A. fundyense cells collected on the 20-μm-pore-size sieve were rinsed carefully with sterile seawater in order to remove free-living bacteria. Cells collected on the sieve were then backwashed with sterile f/2 medium and collected by centrifugation at 5,000 × g for 5 min. Bacterial cells that passed through the sieve (considered free living) were collected by centrifugation at 10,000 × g for 10 min.

PCR and DGGE analysis of 16S rRNA gene fragments.

A PCR-denaturing gradient gel electrophoresis (DGGE) approach was used to analyze the 16S rRNA gene phylogeny of bacteria associated with phytoplankton. Phytoplankton cells and their associated bacteria were subjected to centrifugation at 10,000 × g for 10 min, washed with sterile seawater, and collected by centrifugation at 10,000 × g for 10 min. Genomic DNA extraction, amplification of 16S rRNA gene fragments, DGGE, and sequence analysis were conducted as previously described (49). The DGGE conditions included a 55 to 70% denaturant gradient gel (where 100% was equivalent to 7 M urea and 40% formamide) (46) that was electrophoresed for 16 h at 70 V. For each sample, approximately 1.0 μg PCR product was analyzed by DGGE. After DGGE, isolated bands were excised and pulverized with a sterile mortar and pestle, and DNA was eluted overnight in 50 μl 0.1 M Tris, pH 8.0, at 4°C. Partial 16S rRNA genes were then reamplified from excised bands and analyzed by a second DGGE in order ensure that heteroduplexes were resolved. In addition, for these second DGGEs, bands from different samples that appeared to migrate to the same or similar positions in the original analysis were placed in adjacent lanes to confirm their relative positions. DGGE profiles were analyzed using the Quantity One gel documentation software (Bio-Rad, Hercules, CA) in order to determine the positions of individual bands.

Cloning and sequencing.

Direct sequencing of some of the DGGE bands was unsuccessful, and these bands were therefore reamplified and cloned prior to sequencing. Clones were constructed using a TOPO TA cloning kit (Invitrogen Corp., Carlsbad, CA) with the pCR 2.1-TOPO vector and TOP10 One Shot chemically competent cells as described by the manufacturer. Sequencing reactions were performed using primer 338F according to the instructions of the manufacturer of the sequencing kit (Beckman-Coulter, Fullerton, CA), and sequences were analyzed using a Beckman-Coulter CEQ 2000 automated DNA sequencer in the University of Massachusetts Lowell Biological Sciences Department.

16S rRNA gene sequences were checked for potential chimeras with the Ribosomal Database Project II (RDP II) Chimera Check program (39). Sequences were then aligned with closely related 16S rRNA sequences from GenBank and the RDP II, and alignments were edited manually using the program SeqPup. Unambiguously aligned base positions were used to construct phylogenetic trees with maximum-likelihood and maximum-parsimony methods using PAUP* (Sinauer Associates, Sunderland, MA) and the Weighbor-Joining method using RDP II (39).

Analysis of DGGE results.

Once the DGGE analysis and sequencing were complete, a schematic diagram representing each unique DGGE band was constructed. Any bands found to be related to chloroplast rRNA sequences were assumed to originate from the phytoplankton themselves and were not considered in further analyses. In many cases, multiple bands were found to represent the same sequence and were therefore represented as a single DGGE band position. Using the compiled DGGE information, pairwise similarity values for DGGE profiles were calculated as follows: SAB = MAB/N, where MAB is the number of matches (i.e., the number of bands present in both lane A and lane B for each possible band position) and N is the number of band positions (i.e., the total number of bands in the composite lane) (42). DGGE band intensities were not taken into account.


DGGE profiles and phylogenetic diversity.

DGGE profiles revealed a high degree of variability among bacterial associates of different phytoplankton cultures, and there were relatively few common bands for multiple cultures (Fig. (Fig.11 and and2).2). Because of this complexity in the DGGE profiles, further analysis by multiple DGGEs and sequencing was necessary in order to obtain robust intersample comparisons. A compilation of the DGGE and sequencing data is shown in Fig. Fig.3,3, as described above. There was also substantial variability in the number of bands associated with each culture; as few as one nonchloroplast DGGE band was associated with the Chlorarchnion sp. culture, and as many as seven nonchloroplast DGGE bands were associated with the Tetraselmis sp. culture (Fig. (Fig.33).

FIG. 1.
DGGE profiles of 16S rRNA gene fragments of bacteria in cultures of diverse phytoplankton isolated from the Gulf of Maine. All labeled bands were analyzed by multiple DGGEs and/or sequence analysis. Chimeric sequences were not included in the analysis. ...
FIG. 2.
DGGE profiles of 16S rRNA gene fragments of bacteria in Alexandrium spp. cultures. All labeled bands were analyzed by multiple DGGEs and/or sequence analysis. Chimeric sequences were not included in the analysis. Bands for attached (ATTL01) and free-living ...
FIG. 3.
Schematic representation of DGGE bands of bacteria associated with diverse phytoplankton from the Gulf of Maine and Alexandrium species from around the world. DGGE bands that were sequenced are indicated by “seq.” Sequences that matched ...

Comparison of the DGGE profiles of free-living, attached, and bulk bacterial samples of the A. fundyense strain CB301 culture revealed similar profiles for all three sample types, and all dominant bands were present in all sample types (Fig. (Fig.2).2). Similarly, DGGE analysis of bacteria associated with strain CB301 on multiple occasions and during different growth stages showed the same major bands (data not shown), although the band intensities sometimes varied.

Phylogenetic analysis of DGGE band sequences revealed the dominance of two major groups of bacteria, the Roseobacter group in the alpha-Proteobacteria and the Cytophaga-Flavobacterium-Bacteroides (CFB) phylum. Together, these two groups accounted for 78% of the DGGE bands analyzed. Other groups that were represented in multiple samples included members of the family Alteromonadaceae, the Sphingomonas group, and the Rhizobium-Agrobacterium group and Marinobacter sp. strain PCOB-2 (Table (Table22 and Fig. Fig.3).3). Interestingly, most (62%) of the sequences analyzed were more than 99% similar to their closest relatives in the databases, while only 15% exhibited similarity values of less than 97%.

Similarity values for the closest relatives of 16S rRNA gene sequences retrieved from phytoplankton culturesa

Sequences were considered to show evidence of association with toxic dinoflagellates if they exhibited more than 99% similarity with another sequence retrieved from a dinoflagellate culture or bloom (Table (Table2).2). This similarity value was chosen as the operational limit for defining members of the same phylotype, given PCR and sequencing errors and interoperon sequence variability (1). Sequences found in multiple Alexandrium spp. cultures in the current study were also marked as showing evidence of association with toxic dinoflagellates (Fig. (Fig.3).3). Interestingly, the majority of the bands (68%) in Alexandrium spp. cultures matched other bacteria associated with toxic dinoflagellates, including other Alexandrium species and Gymnodinium species. In contrast, only 19% of the bands from nondinoflagellate cultures matched sequences associated with toxic dinoflagellates. Nondinoflagellate cultures also had a much lower incidence of sequences that matched (i.e., were >99% similar to) sequences in the databases that were found in other algal cultures. For example, for nondinoflagellate cultures, 66% of the sequences with database matches were most closely related to sequences obtained directly from environmental samples and were not specifically associated with algae. This result was in contrast with the fact that 77% of the sequences with database matches from Alexandrium cultures had closest relatives that were directly associated with dinoflagellates. These results provide further evidence that there is selection of specific bacterial species associated with dinoflagellates.

The Roseobacter clade, in particular, was prevalent among sequences that we considered to be associated with toxic dinoflagellates. For example, for the Alexandrium spp. cultures analyzed here, 11 of 19 bands that were considered bands from associates of toxic dinoflagellates were Roseobacter sequences. Other phylogenetic groups that were represented among the associates of toxic dinoflagellates, but were less prevalent, included Marinobacter sp. strain PCOB-2, the Rhizobium-Agrobacterium group, and the CFB phylum. Sequences that fell within the Altermonadaceae family and the Sphingomonas group were associated only with nontoxic phytoplankton cultures.

A phylogenetic tree of the Roseobacter group 16S rRNA gene fragments (Fig. (Fig.4)4) contained a particular cluster of closely related sequences that are associated with the A. fundyense cultures analyzed here and an A. tamarense culture analyzed by Hold et al. (30), and similar topologies were obtained by all three methods used for tree construction (data not shown). While sequences of our DGGE bands did not provide sufficient phylogenetic information for a robust bootstrap analysis of this group, members of this cluster are clearly very closely related. Also evident from Fig. Fig.44 is the divergence between sequences affiliated with Alexandrium spp. and several sequences from the Tetraselmis sp. culture (sequences Trs11C, Trs11E, and Trs11F), the Micromonas pusilla culture (Mcp10B), and the Nannochloropsis granulata culture (Ncl9E and Ncl9D).

FIG. 4.
Phylogenetic tree of 16S rRNA gene sequences belonging to the Roseobacter group. Sequences obtained in the current study are indicated by boldface type. Sequences considered to show evidence of association with toxic dinoflagellates are indicated by asterisks. ...

In contrast to the Roseobacter group sequences, which were at least 98% similar to their closest GenBank relatives (Table (Table2),2), sequences that fell within the CFB group exhibited greater phylogenetic diversity and were frequently not closely related to any reported sequences, and the levels of similarity with their closest GenBank relatives were as low as 90.7% (Table (Table2).2). CFB group sequences that were considered to be from associates of toxic dinoflagellates (Emh3C and Atc5′A) were found to be identical to sequences retrieved from toxic Gymnodinium catenatum cultures (24).

Similarity values for DGGE profiles of bacteria associated with phytoplankton.

While the complexity of the DGGE profiles made it difficult to visually discern patterns across samples, analysis of pairwise similarity values for DGGE profiles revealed several interesting results. Pairwise coefficients (Fig. (Fig.5)5) were calculated using DGGE data compiled in Fig. Fig.3,3, which took multiple DGGEs and sequence analysis into consideration. Interestingly, two areas where there were higher similarity values are evident in Fig. Fig.5;5; these areas are the comparisons among nontoxic phytoplankton from the Gulf of Maine and the comparisons among Alexandrium spp. from the Gulf of Maine and other regions of the world. In fact, if one sequence (which was identical to the Marinobacter sp. strain PCOB-2 sequence and was present in Pseudopedinella elastica, Scrippsiella, A. fundyense strain CB301, and A. tamarense strain ATRU10/1) was excluded, all similarity values for Alexandrium spp. and non-Alexandrium spp. bacterial assemblages were zero. With all of the data considered, a two-tailed t test showed that the similarity values among non-Alexandrium samples were significantly higher than the similarity values among non-Alexandrium samples and Alexandrium samples at a P value of <0.001. Similarly, the similarity values among Alexandrium samples were significantly higher than the similarity values for comparisons between Alexandrium samples and non-Alexandrium samples at a P value of 0.014. There was no significant difference (P = 0.655), however, between Alexandrium cultures from the Gulf of Maine and Alexandrium cultures from around the world (Japan, Portugal, Russia, France, and the Bering Sea).

FIG. 5.
Plot of pairwise similarity values for DGGE profiles of bacterial assemblages associated with phytoplankton cultures.


While bacteria are known to be closely associated with phytoplankton and are thought to influence phytoplankton population dynamics and toxicity, the phylogenetic identity and specificity of bacterium-phytoplankton interactions are only beginning to be explored. In the current study, we conducted an extensive survey of bacteria associated with toxic and nontoxic phytoplankton. Our approach was to use 16S rRNA gene DGGE to analyze bacterial associates of at least one representative of each major phylogenetic group of phytoplankton from the Gulf of Maine, as well as multiple toxic Alexandrium spp. cultures originating from the Gulf of Maine, Japan, coastal Portugal, France, Russia, and the Bering Sea, in order to determine the phylogenetic diversity of the bacterial associates and whether there were specific associations between bacteria and phytoplankton. While the DGGE profiles were complex and highly variable, we found higher levels of similarity for bacterial assemblages from Alexandrium spp. cultures from around the world than for Alexandrium spp. cultures and nontoxic cultures from the same region, indicating that there was selection for certain bacterial phylotypes in Alexandrium cultures. In particular, a cluster in the Roseobacter group was associated with toxic dinoflagellate cultures analyzed here and in other studies (3, 4, 24, 30), as were several CFB strains.

Limitations of the experimental approach.

Given the complexity of natural phytoplankton communities (50) and the fact that Alexandrium spp. can cause paralytic shellfish poisoning at levels as low as 200 cells liter−1, we relied on unialgal cultures to study bacterium-phytoplankton interactions. While a similar approach has been used in other studies (24, 30, 51), this experimental system introduced several potential biases into our analyses. For example, the environment of a unialgal culture growing in nutrient-enriched medium under laboratory conditions is quite different from the environment experienced by a member of a complex community subjected to variable natural conditions. Only bacteria capable of growing with that alga under laboratory conditions persist after successive transfers. Perhaps equally important, only the bacteria associated with the single algal cell used to establish the culture have the opportunity to be present as the culture is grown and transferred. It is unlikely that the bacterial species associated with a single cell accurately represent all bacteria associated with a particular algal population, and it is even more unlikely that they are representative of the bacteria associated with a given alga over a range of ecological and environmental conditions. Nonetheless, analyzing the bacteria in unialgal cultures provides a means to separate individual algal species from their complex natural matrix and is a first step toward understanding bacterium-phytoplankton interactions.

We attempted to minimize potential biases of this experimental system by using several criteria to select cultures for analysis. For our survey of diverse phytoplankton taxa, we focused on cultures from one geographic area (the Gulf of Maine) and included at least one representative of each major taxonomic group of phytoplankton. We felt that the Gulf of Maine provided an ideal geographic area as it is a highly productive, temperate, coastal environment in which diverse eukaryotic phytoplankton are the dominant primary producers and in which Alexandrium blooms occur annually (5, 50, 53). For Alexandrium species cultures, we included five cultures from the Gulf of Maine and then chose other cultures that represented a broad geographic range of this cosmopolitan genus. In order to minimize biases associated with cultivation, we chose cultures with a minimal time since the original isolation and similar growth conditions (Table (Table1).1). Because we were interested in bacteria associated with actively growing phytoplankton, we focused on cultures that originated from vegetative cells from the water column, not dormant stages, such as Alexandrium spp. cysts, found in the sediment. In contrast with other studies of bacterium-phytoplankton associations in cultures (24, 30), we analyzed a relatively large number of diverse phytoplankton cultures in order to obtain a robust comparison of bacterial associates and more accurately determine the specificity of the associations.

Our comparison of DGGE profiles of free-living and attached bacteria associated with A. fundyense strain CB301 (Fig. (Fig.2),2), as well as our analysis of the DGGE profiles of samples taken during different phases of A. fundyense batch culture (results not shown), indicated that there was no qualitative difference in the dominant DGGE bands. These results are similar to those reported by Hold et al. (30) and suggest that it was not necessary to analyze multiple phytoplankton growth stages or free-living and attached bacterial fractions in order to determine the phylogenetic identities of dominant bacteria associated with phytoplankton cultures.

Evidence for specific interactions.

Our results revealed patterns of specific associations between Alexandrium and bacteria. Pairwise similarity coefficients for bacterial species composition in phytoplankton cultures (Fig. (Fig.5)5) revealed significantly higher levels of similarity among Alexandrium cultures than between nontoxic phytoplankton and Alexandrium cultures. This relationship was true regardless of the geographic location from which Alexandrium species cultures were obtained. In addition, compared to diverse nontoxic phytoplankton, the Alexandrium spp. cultures analyzed here contained a higher percentage of bacterial sequences that matched database sequences from other dinoflagellate cultures (Table (Table22 and Fig. Fig.3),3), including cultures of G. catenatum isolated from New Zealand and Japan (24), Alexandrium species from British Columbia, Canada, and Plymouth, United Kingdom, and Scrippsiella trochoidea from British Columbia, Canada (30). While cross-contamination between cultures is always a concern, the fact that these dinoflagellate cultures were isolated from and grown in different regions of the world makes this an unlikely explanation of our results. Similarly, while we cannot rule out selection by artificial laboratory growth conditions, the fact that diverse phytoplankton were cultivated under similar conditions (Table (Table1)1) suggests that factors other than cultivation conditions contributed to the observed similarities. For example, S. costatum, Thalassiosira gravida, N. granulata, Prorocentrum minimum, Scrippsiella sp., and all Alexandrium cultures were cultivated in f/2 medium, yet S. costatum, T. gravida, N. granulata, and P. minimum shared no bacterial associates with Alexandrium cultures. Lastly, phylogenetic analysis of members of the Roseobacter clade, which was the most predominant group among the Alexandrium-associated bacteria, indicated that many sequences from Alexandrium spp. cultures were closely related to each other and divergent from several Roseobacter sequences obtained from cultures of Micromonas pusilla (e.g., Mcp10B), Tetraselmis sp. (Trs11C and Trs11E), and N. granulata (Ncl9D) (Fig. (Fig.4).4). Not surprisingly, these results did not reveal a single bacterium that was consistently found in all toxic or nontoxic dinoflagellate cultures. Instead, they revealed a pattern in which certain bacterial phylotypes were more commonly observed in dinoflagellate cultures than in other cultures. However, given the potential limitations of our approach (i.e., the bacterial associates of each culture were a subset of the bacteria associated with a single phytoplankton cell at the time of isolation), the emergence of such a pattern is especially interesting.

The predominant phylogenetic groups of bacteria found in this study were very similar to those previously found to be associated with algal cultures (3, 4, 24, 30, 41, 51). In fact, more than 60% of all the sequences found were more than 99% similar to their closest database relative, and only three sequences (Afb16B, Atb4′F, and Mcp10C) had no close relatives. Several studies have found a prevalence of Roseobacter spp. associated with algal cultures, including cultures of dinoflagellates (3, 4, 24, 30) and diatoms (51). Similarly, members of the family Alteromonadaceae, the CFB group, and the sphingomonads have been found to be associated with algae (16, 17, 24, 30). While these bacterial families and broader phylogenetic groups appear to be important associates of algae in general, a pattern of specific bacterium-alga associations is evident from the prevalence of closely related phylogenetic clusters within these groups that are associated with closely related phytoplankton.

In the current study, only one nontoxic dinoflagellate (Scrippsiella sp.) was analyzed. Other studies of bacteria associated with dinoflagellates have also focused on toxic species (4, 24, 30). Therefore, it is difficult to determine whether the observed selection of specific bacterial phylotypes in the presence of dinoflagellates was due to the toxicity of Alexandrium species or to some other feature of Alexandrium physiology. Our finding that there were two groups with higher levels of similarity for the bacterial profiles, bacteria from diverse nontoxic phytoplankton and bacteria from Alexandrium (Fig. (Fig.5),5), suggests that toxicity may play a role in selecting for specific bacteria. Saxitoxin has recently been shown to influence sodium and potassium channel activity in prokaryotes and, in saxitoxin-producing cyanobacteria, may maintain homeostasis under sodium stress conditions (47). It has also been shown to mitigate the effects of a sodium channel activator, veratridine, in Vibrio fischeri (47). It is therefore conceivable that the production of saxitoxin by dinoflagellates selects for specific bacteria that are better able to maintain sodium and/or potassium homeostasis in its presence. Alternatively, cell wall components, exudates, production of osmolytes such as dimethyl sulfoniopropionate (DMSP), pigment composition, and life cycle features characteristic of dinoflagellates may select for specific bacterial phylotypes. For example, in some thecate dinoflagellates ecdysis occurs as part of the life cycle (45). During this process the outermost membranes, thecal plates, and flagella are shed during cell division. In natural actively growing populations, these cell coverings could provide a substrate for specific bacterial consortia. Also, hypnocysts formed during the life cycle of Alexandrium have recently been shown to harbor intracellular bacteria that could maintain a long-term connection between the vegetative and dormant stages of Alexandrium. These cyst-associated bacteria could seed germinating populations of vegetative Alexandrium cells, maintaining a bacterial association even without the need for endocytobiosis after each dormancy period (52).

Dominant bacterial associates of other phytoplankton.

The phytoplankton cultures analyzed here represent the full taxonomic breadth of the Gulf of Maine eukaryotic phytoplankton community (Table (Table1).1). While all of the phytoplankton are photosynthetic, they vary greatly in terms of size, life history, motility, cell wall constituents, pigment composition, and cellular storage products. For example, the cell walls of the dinoflagellates analyzed (Alexandrium spp., Scrippsiella sp., and Prorocentrum lima) are comprised of cellulose theca, while those of the diatoms (Chaetoceros tortissimus, S. costatum, and T. gravida) are comprised of silica frustules and those of the cryptophyte Rhodomonas sp. are comprised of an organic periplast (37). Storage products range from starch (dinoflagellates, cryptophytes) to chrysolaminarin and lipids (diatoms, prymnesiophytes) and mannitol (prasinophytes) (37). These differences in the biochemical makeup of divergent phytoplankton species are likely to select for different associated bacterial species that may rely on organic compounds available in the phycosphere. In addition, some phytoplankton are known to produce antibiotics that can exert strong selective pressures (54, 55).

Indeed, several differences were evident in the bacterial assemblages found in diverse nontoxic phytoplankton cultures compared with those found in Alexandrium species cultures. Compared to Alexandrium species associates, these bacteria included a greater prevalence of members of the CFB phylum, the family Alteromonodaceae, and the Sphingomonas group; the latter two were absent from Alexandrium cultures (Fig. (Fig.3).3). While members of the Roseobacter clade were present in diverse nontoxic phytoplankton cultures analyzed here, they were not observed as frequently as they were in dinoflagellate cultures (Fig. (Fig.3).3). In addition, bacteria from nontoxic cultures were more likely to match database sequences retrieved directly from the environment or to have no close relatives in the databases (Table (Table2).2). These results are not surprising given the dearth of studies of (and therefore 16S rRNA gene sequence information for) bacteria associated with phytoplankton other than dinoflagellates.

Potential function of bacterial associates of phytoplankton.

Bacteria have been shown to have many effects on phytoplankton, including algicidal activity (44, 58), stimulation of phytoplankton growth (16), production or modulation of toxicity (20, 21, 35), and inhibition or promotion of cyst formation (2, 18). Bacteria may stimulate phytoplankton growth via the production of vitamins (28), iron chelators (siderophores) (56), and cytokinins (40). Our findings and those of other workers underscore the importance of the Roseobacter clade in associations with phytoplankton and, especially, dinoflagellates. For example, members of the Roseobacter clade were found to be prevalent in cultures of A. tamarense (2, 30), Alexandrium ostenfeldii (4), S. trochoidea (30), G. catenatum (24), and Pfiesteria-like species (3). This clade is known to be a dominant member of coastal bacterial communities (10, 14, 15, 23), and previously, we found that it is ubiquitous across seasonal (February, May, July, and September) and spatial gradients in the northern Gulf of Maine (49). Members of this group possess diverse metabolic capabilities. Roseobacter species have been shown to be capable of degrading lignins and other aromatic ring compounds, including vanillate, coumarate, cinnamate, ferulate, benzoate, and p-hydroxybenzoate (8), which may be present in dinoflagellate phycospheres. Many Roseobacter isolates have been shown to utilize DMSP as both a carbon source and a sulfur source, and it is likely DMSP metabolism is important in Roseobacter-phytoplankton interactions (22, 23, 43). However, because the abilities to produce and consume DMSP are so widespread among phytoplankton and roseobacters, respectively, it seems unlikely that DMSP metabolism resulted in the prevalence of a narrow phylogenetic cluster of roseobacters in the Alexandrium spp. cultures observed here.


Interactions between bacteria and phytoplankton are thought to be important in controlling the dynamics of both communities and yet are only beginning to be understood at the species composition level. Our results provide evidence for specific bacterium-phytoplankton associations, especially between toxic dinoflagellates and members of the Roseobacter clade. This result was supported even though toxic dinoflagellates were isolated from different regions of the world and were grown under conditions that were very similar to those used for diverse phytoplankton taxa. Phylogenetic analysis of bacteria associated with a wide diversity of phytoplankton revealed the prevalence of members of the Roseobacter clade, the CFB phylum, and the Alteromonadaceae family, indicating that members of these groups are well adapted to living in close association with phytoplankton and that specific clusters within these groups are selected for in association with different phytoplankton.


We thank D. M. Anderson and D. Kulis for providing Alexandrium spp. cultures and M. Graves for his assistance with automated DNA sequence analysis.

This work was supported by award OCE-0117820 from the National Science Foundation.


1. Acinas, S. G., L. A. Marcelino, V. Klepac-Ceraj, and M. F. Polz. 2004. Divergence and redundancy of 16S rRNA sequences in genomes with multiple rrn operons. J. Bacteriol. 186:2629-2635. [PMC free article] [PubMed]
2. Adachi, M., T. Kanno, R. Okamoto, S. Itakura, M. Yamaguchi, and T. Nishijima. 2003. Population structure of Alexandrium (Dinophyceae) cyst formation-promoting bacteria in Hiroshima Bay, Japan. Appl. Environ. Microbiol. 69:6560-6568. [PMC free article] [PubMed]
3. Alavi, M., T. Miller, K. Erlandson, R. Schneider, and R. Belas. 2001. Bacterial community associated with Pfiesteria-like dinoflagellate cultures. Environ. Microbiol. 3:380-396. [PubMed]
4. Allgaier, M., H. Uphoff, A. Felske, and I. Wagner-Dobler. 2003. Aerobic anoxygenic photosynthesis in Roseobacter clade bacteria from diverse marine habitats. Appl. Environ. Microbiol. 69:5051-5059. [PMC free article] [PubMed]
5. Anderson, D. M. 1997. Bloom dynamics of toxic Alexandrium species in the northeastern U.S. Limnol. Oceanogr. 4:1009-1022.
6. Azam, F. 1998. Microbial control of oceanic carbon flux: the plot thickens. Science 280:694-696.
7. Blackburn, N., T. Fenchel, and J. Mitchell. 1998. Microscale nutrient patches in planktonic habitats shown by chemotactic bacteria. Science 282:2254-2256. [PubMed]
8. Buchan, A., L. S. Collier, E. L. Neidle, and M. A. Moran. 2000. Key aromatic-ring-cleaving enzyme, protocatechuate 3,4-dioxygenase, in the ecologically important marine Roseobacter lineage. Appl. Environ. Microbiol. 66:4662-4672. [PMC free article] [PubMed]
9. Crump, B. C., E. V. Armbrust, and J. A. Baross. 1999. Phylogenetic analysis of particle-attached and free-living bacterial communities in the Columbia River, its estuary, and the adjacent coastal ocean. Appl. Environ. Microbiol. 65:3192-3204. [PMC free article] [PubMed]
10. Dang, H., and C. R. Lovell. 2000. Bacterial primary colonization and early succession on surfaces in marine waters as determined by amplified rRNA gene restriction analysis and sequence analysis of 16S rRNA genes. Appl. Environ. Microbiol. 66:467-475. [PMC free article] [PubMed]
11. Delong, E. F., D. G. Franks, and A. L. Alldredge. 1993. Phylogenetic diversity of aggregate-attached vs free-living marine bacterial assemblages. Limnol. Oceanogr. 38:924-934.
12. Doucette, G. J. 1995. Interactions between bacteria and harmful algae: a review. Nat. Toxins 3:65-74. [PubMed]
13. Doucette, G. J., M. Kodama, S. Franca, and S. Gallacher. 1998. Bacterial interactions with harmful algal bloom species: bloom ecology, toxigenesis, and cytology, p. 619-647. In D. M. Anderson, A. D. Cembella, and G. M. Hallegraeff (ed.), Physiological ecology of harmful algal blooms, vol. G 41. Springer-Verlag, Berlin, Germany.
14. Eilers, H., J. Pernthaler, F. O. Glockner, and R. Amann. 2000. Culturability and in situ abundance of pelagic bacteria from the North Sea. Appl. Environ. Microbiol. 66:3044-3051. [PMC free article] [PubMed]
15. Eilers, H., J. Pernthaler, J. Peplies, F. O. Glockner, G. Gerdts, and R. Amann. 2001. Isolation of novel pelagic bacteria from the German Bight and their seasonal contributions to surface picoplankton. Appl. Environ. Microbiol. 67:5134-5142. [PMC free article] [PubMed]
16. Ferrier, M., J. L. Martin, and J. N. Rooney-Varga. 2002. Stimulation of Alexandrium fundyense growth by bacterial assemblages from the Bay of Fundy. J. Appl. Microbiol. 92:1-12. [PubMed]
17. Fisher, M. M., L. W. Wilcox, and L. E. Graham. 1998. Molecular characterization of epiphytic bacterial communities on charophycean green algae. Appl. Environ. Microbiol. 64:4384-4389. [PMC free article] [PubMed]
18. Fukami, K., T. Nishijima, H. Murata, S. Doi, and Y. Hata. 1991. Distribution of bacteria influential on the development and the decay of Gymnodinium nagasakiense red tide and their effects on algal growth. Nippon Suisan Gakkaishi 57:2321-2326.
19. Fukami, K., A. Yuzawa, T. Nishijima, and Y. Hata. 1992. Isolation and properties of a bacterium inhibiting the growth of Gymnodinium nagasakiense. Nippon Suisan Gakkaishi 58:1073-1077.
20. Gallacher, S., K. J. Flynn, J. M. Franco, E. E. Brueggemann, and H. B. Hines. 1997. Evidence for production of paralytic shellfish toxins by bacteria associated with Alexandrium spp. (Dinophyta) in culture. Appl. Environ. Microbiol. 63:239-245. [PMC free article] [PubMed]
21. Gallacher, S., K. J. Flynn, J. Leftley, J. Lewis, P. D. Munro, and T. H. Birbeck. 1996. Bacterial production of sodium channel blocking toxins, p. 355-358. In T. Yasumoto, Y. Oshima, and Y. Fukuyo (ed.), Harmful and toxic algal blooms. Intergovernmental Oceanographic Commission of UNESCO, Sendai, Japan.
22. Gonzalez, J. M., R. P. Kiene, and M. A. Moran. 1999. Transformation of sulfur compounds by an abundant lineage of marine bacteria in the alpha-subclass of the class Proteobacteria. Appl. Environ. Microbiol. 65:3810-3819. [PMC free article] [PubMed]
23. Gonzalez, J. M., R. Simo, R. Massana, J. S. Covert, E. O. Casamayor, C. Pedros-Alio, and M. A. Moran. 2000. Bacterial community structure associated with a dimethylsulfoniopropionate-producing North Atlantic algal bloom. Appl. Environ. Microbiol. 66:4237-4246. [PMC free article] [PubMed]
24. Green, D. H., L. E. Llewellyn, A. P. Negri, S. I. Blackburn, and C. J. Bolch. 2004. Phylogenetic and functional diversity of the cultivable bacterial community associated with the paralytic shellfish poisoning dinoflagellate Gymnodinium catenatum. FEMS Microbiol. Ecol. 47:345-357. [PubMed]
25. Guillard, R. R. L. 1975. Culture of phytoplankton for feeding marine invertebrates, p. 29-59. In W. L. Smith and M. H. Chanley (ed.), Culture of marine invertebrate animals. Plenum Publishing Corporation, New York, N.Y.
26. Guillard, R. R. L., and P. E. Hargraves. 1993. Stichochrysis immobilis is a diatom, not a chrysophyte. Phycologia 32:234-236.
27. Guillard, R. R. L., and J. H. Ryther. 1962. Studies of marine planktonic diatoms. I. Cyclotella nana Hustedt and Detonula confervaceae (Cleve) Gran. Can. J. Microbiol. 8:229-239. [PubMed]
28. Haines, K. C., and R. R. L. Guillard. 1974. Growth of vitamin B12-requiring marine diatoms with vitamin B12-producing marine bacteria. J. Phycol. 10:245-252.
29. Hold, G. L., E. A. Smith, T. H. Birkbeck, and S. Gallacher. 2001. 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. [PubMed]
30. Hold, G. L., E. A. Smith, M. S. Rappé, E. W. Maas, E. R. B. Moore, C. Stroempl, J. R. Stephen, J. I. Prosser, T. H. Birkbeck, and S. Gallacher. 2001. Characterisation of bacterial communities associated with toxic and non-toxic dinoflagellates: Alexandrium spp. and Sprippsiella trochoidea. FEMS Microbiol. Ecol. 37:161-173.
31. Keller, M. D., R. C. Selvin, W. Claus, and R. R. L. Guillard. 1987. Media for the culture of oceanic ultraphytoplankton. J. Phycol. 23:633-638.
32. Kerkhof, L. J., M. A. Voytek, R. M. Sherrell, D. Millie, and O. Schofield. 1999. Variability in bacterial community structure during upwelling in the coastal ocean. Hydrobiology 401:139-148.
33. Kodama, M. 1990. Possible links between bacteria and toxin production in algal blooms, p. 52-61. In E. A. Graneli (ed.), Toxic marine phytoplankton. Elsevier Science Publishing Co., Inc., Amsterdam, The Netherlands.
34. Kogure, K., U. Simidu, and N. Taga. 1982. Bacterial attachment to phytoplankton in seawater. J. Exp. Mar. Biol. Ecol. 56:197-204.
35. Kopp, M., G. J. Doucette, M. Kodama, G. Gerdts, C. Schuett, and L. K. Medlin. 1997. Phylogenetic analysis of selected toxic and non-toxic bacterial strains isolated from the toxic dinoflagellate Alexandrium tamarense. FEMS Microbiol. Ecol. 24:251-257.
36. Lancelot, C. 1983. Factors affecting phytoplankton extracellular release in the Southern Bight of the North Sea. Mar. Ecol. Prog. Ser. 12:115-121.
37. Lee, R. E. 1999. Phycology, 3rd ed. Cambridge University Press, Cambridge, United Kingdom.
38. Lewis, J., G. Kennaway, S. Franca, and E. Alverca. 2001. Bacterium-dinoflagellate interactions: investigative microscopy of Alexandrium spp. (Gonyaulacales, Dinophyceae). Phycologia 40:280-285.
39. Maidak, B. L., J. R. Cole, T. G. Lilburn, C. T. Parker, Jr., P. R. Saxman, J. M. Stredwick, G. M. Garrity, B. Li, G. J. Olsen, S. Pramanik, T. M. Schmidt, and J. M. Tiedje. 2000. The RDP (Ribosomal Database Project) continues. Nucleic Acids Res. 28:173-174. [PMC free article] [PubMed]
40. Maruyama, A., M. Maeda, and U. Simidu. 1986. Occurrence of plant hormone (cytokinin)-producing bacteria in the sea. J. Appl. Bacteriol. 61:569-574.
41. Mayali, X., and G. J. Doucette. 2002. Microbial community interactions and population dynamics of an algicidal bacterium active against Karenia brevis (Dinophyceae). Harmful Algae 1:277-293.
42. McCaig, A. E., L. A. Glover, and J. I. Prosser. 2001. Numerical analysis of grassland bacterial community structure under different land management regimens by using 16S ribosomal DNA sequence data and denaturing gradient gel electrophoresis banding patterns. Appl. Environ. Microbiol. 67:4554-4559. [PMC free article] [PubMed]
43. Miller, T. R., and R. Belas. 2004. Dimethylsulfoniopropionate metabolism by Pfiesteria-associated Roseobacter spp. Appl. Environ. Microbiol. 70:3383-3391. [PMC free article] [PubMed]
44. Mitsutani, A., K. Takesue, M. Kirita, and Y. Ishida. 1992. Lysis of Skeletonema costatum by Cytophaga sp. isolated from the coastal water of the Ariake Sea. Nippon Suisan Gakkaishi 58:2159-2167.
45. Morrill, L. C. 1984. Ecdysis and the location of the plasma membrane in the dinoflagellate Heterocapsa niei. Protoplasma 119:8-20.
46. Muyzer, G., E. C. De Waal, and A. G. Uitterlinden. 1993. Profiling of complex microbial populations by denaturing gradient gel electrophoresis analysis of polymerase chain reaction-amplified genes coding for 16S rRNA. Appl. Environ. Microbiol. 59:695-700. [PMC free article] [PubMed]
47. Pomati, F., C. Rossetti, D. Calamari, and B. A. Neilan. 2003. Effects of saxitoxin (STX) and veratridine on bacterial Na+-K+ fluxes: a prokaryote-based STX bioassay. Appl. Environ. Microbiol. 69:7371-7376. [PMC free article] [PubMed]
48. Provasoli, L., J. J. A. McLaughlin, and M. R. Droop. 1957. The development of artificial media for marine algae. Arch. Mikrobiol. 25:392-428. [PubMed]
49. Rooney-Varga, J. N., M. W. Giewat, M. C. Savin, M. LeGresley, and J. L. Martin. 2005. Links between phytoplankton and bacterial community dynamics in a coastal marine environment. Microb. Ecol. 49:163-175. [PubMed]
50. Savin, M. C., J. L. Martin, M. LeGresley, M. W. Giewat, and J. N. Rooney-Varga. 2004. Plankton diversity in the Bay of Fundy as measured by morphological and molecular methods. Microb. Ecol. 48:51-65. [PubMed]
51. Schäfer, H., B. Abbas, H. Witte, and G. Muyzer. 2002. Genetic diversity of ‘satellite’ bacteria present in cultures of marine diatoms. FEMS Microbiol. Ecol. 42:25-35. [PubMed]
52. Schweikert, M. 2003. Cell wall ultrastructure and intracytoplasmic bacteria in hypnocysts of toxic Alexandrium tamarense (Dinophyceae). Protistology 3:138-144.
53. Shumway, S. E., S. Sherman-Caswell, and J. W. Hurst. 1988. Paralytic shellfish poisoning in Maine: monitoring a monster. J. Shellfish Res. 7:643-652.
54. Sieburth, J. M. 1961. Acrlyic acid, an antibiotic principle in Phaeocystis blooms in Antarctic waters. Science 132:676-677. [PubMed]
55. Sieburth, J. M. 1968. The influence of algal antibiosis on the ecology of marine microorganisms, p. 63-94. In M. R. Droop and E. L. F. Woods (ed.), Advances in microbiology of the sea. Academic Press, London, United Kingdom.
56. Soria-Dengg, S., R. Reissbrodt, and U. Hortsmann. 2001. Siderophores in marine coastal waters and their relevance for iron uptake by phytoplankton: experiments with the diatom Phaeodactylum tricornutum. Mar. Ecol. Prog. Ser. 220:73-82.
57. Vaque, D., C. M. Duarte, and C. Marrasé. 1990. Influence of algal population dynamics on phytoplankton colonization by bacteria: evidence from two diatom species. Mar. Ecol. Prog. Ser. 65:201-203.
58. Yoshinaga, I., T. Kawai, and Y. Ishida. 1997. Analysis of algicidal ranges of the bacteria killing the marine dinoflagellate Gymnodinium mikimotoi isolated from Tanabe Bay, Wakayama Pref., Japan. Fish. Sci. (Tokyo) 63:94-98.

Articles from Applied and Environmental Microbiology are provided here courtesy of American Society for Microbiology (ASM)
PubReader format: click here to try


Related citations in PubMed

See reviews...See all...

Cited by other articles in PMC

See all...


Recent Activity

Your browsing activity is empty.

Activity recording is turned off.

Turn recording back on

See more...