Spotlight Selection
Environmental Microbiology
Research Article
6 July 2023

Phytoplankton Producer Species and Transformation of Released Compounds over Time Define Bacterial Communities following Phytoplankton Dissolved Organic Matter Pulses

ABSTRACT

Phytoplankton-bacterium interactions are mediated, in part, by phytoplankton-released dissolved organic matter (DOMp). Two factors that shape the bacterial community accompanying phytoplankton are (i) the phytoplankton producer species, defining the initial composition of released DOMp, and (ii) the DOMp transformation over time. We added phytoplankton DOMp from the diatom Skeletonema marinoi and the cyanobacterium Prochlorococcus marinus MIT9312 to natural bacterial communities from the eastern Mediterranean and determined the bacterial responses over a time course of 72 h in terms of cell numbers, bacterial production, alkaline phosphatase activity, and changes in active bacterial community composition based on rRNA amplicon sequencing. Both DOMp types were demonstrated to serve the bacterial community as carbon and, potentially, phosphorus sources. Bacterial communities in diatom-derived DOM treatments maintained higher Shannon diversities throughout the experiment and yielded higher bacterial production and lower alkaline phosphatase activity compared to cyanobacterium-derived DOM after 24 h of incubation (but not after 48 and 72 h), indicating greater bacterial usability of diatom-derived DOM. Bacterial communities significantly differed between DOMp types as well as between different incubation times, pointing to a certain bacterial specificity for the DOMp producer as well as a successive utilization of phytoplankton DOM by different bacterial taxa over time. The highest differences in bacterial community composition with DOMp types occurred shortly after DOMp additions, suggesting a high specificity toward highly bioavailable DOMp compounds. We conclude that phytoplankton-associated bacterial communities are strongly shaped by the phytoplankton producer as well as the transformation of its released DOMp over time.
IMPORTANCE Phytoplankton-bacterium interactions influence biogeochemical cycles of global importance. Phytoplankton photosynthetically fix carbon dioxide and subsequently release the synthesized compounds as dissolved organic matter (DOMp), which becomes processed and recycled by heterotrophic bacteria. Yet the importance of phytoplankton producers in combination with the time-dependent transformation of DOMp compounds on the accompanying bacterial community has not been explored in detail. The diatom Skeletonema marinoi and the cyanobacterium Prochlorococcus marinus MIT9312 belong to globally important phytoplankton genera, and our study revealed that DOMp of both species was selectively incorporated by the bacterial community. The producer species had the highest impact shortly after DOMp appropriation, and its effect diminished over time. Our results improve the understanding of the dynamics of organic matter produced by phytoplankton in the oceans as it is utilized and modified by cooccurring bacteria.

INTRODUCTION

Phytoplankton are responsible for ~50% of the photosynthesis on Earth (1), but a high fraction of their photosynthetically fixed carbon is released from the cells shortly after fixation (2) in the form of dissolved organic matter (DOMp). The released DOMp serves marine heterotrophic bacteria as an energy and carbon (C) source (3) but also contains nutrients such as nitrogen (N, e.g., amino acids, peptides, and proteins [4, 5]) and phosphorus (P, e.g., DNA and RNA [4]), which are likewise consumed and recycled by bacteria (4, 6, 7). Hence, DOMp-mediated interactions between phytoplankton and bacteria are crucial for understanding the fate of carbon and other elements in the ocean, the functioning of aquatic systems (813), and biogeochemical cycles at a global scale (2, 11, 14).
Different phytoplankton groups and species have different biochemical compositions (1517) and liberate different types of DOMp (8, 1821). While the released DOM is phytoplankton strain specific, it may also reflect the broader phylogenetic origin; e.g., the molecular compositions of DOM derived from different cyanobacteria share higher similarities among each other than among diatom-derived DOM, and vice versa (22). Diatoms, for instance, produce especially polysaccharide-rich DOM (23), whereas cyanobacteria produce smaller and less polar compounds (in comparison to diatoms) (22). Diatoms and cyanobacteria also have different ratios of specific amino acids in their released DOMp (2426). However, in addition to the phylogenetic origin, other factors such as temperature, light, nutrients, and phytoplankton growth phase (reviewed by Mühlenbruch et al. [8]), as well as the type of release (leakage, exudation, viral lysis, or grazing) (27, 28) likely determine the composition of the released DOM.
Differing molecular compositions of DOMp may result in different availabilities of carbon and nutrients, which in turn may impact the accompanying bacterial community. This is because bacteria substantially differ in their ability to utilize different compounds (29). Indeed, the phytoplankton species may drive bacterial community dynamics (21, 25, 30, 31) and shape the accompanying bacterial community (15, 26, 3234). To what magnitude the phytoplankton species determines the accompanying bacterial community in terms of nutrient uptake and dissolved organic carbon (DOC) utilization and which are the responsive phyla, however, are still subjects of debate. Earlier experiments with DOMp addition derived from cyanobacteria and diatoms revealed highly similar transcriptomic profiles of bacteria (35), but also specific pathways for P utilization of either cyanobacterium or diatom DOM (36), and the magnitude of P allocation to the bacterial community by DOMp derived from different phytoplankton is still unknown. Likewise, cyanobacteria and diatoms sometimes trigger the same bacterial communities in bloom events, whereas in other bloom events responsive bacteria differed (32). Some bacteria were found to be specialized in either cyanobacteria- or diatom-derived DOM (35), and bacterial communities responded differently after addition of cyanobacterium and diatom lysates (26).
A second aspect that needs to be considered when analyzing bacterial utilization of DOMp is the temporal transformation of compounds and the concomitant succession of specific bacterial phyla after DOM pulses (32, 3739). The DOMp composition changes with time from highly bioavailable, mostly low-molecular-weight (LMW) to less bioavailable, mostly high-molecular-weight (HMW) compounds, caused by the preferential utilization of labile LMW compounds by bacteria (40). This ultimately leads to the sequestration of recalcitrant carbon, referred to as the microbial carbon pump (41). Analogous to the phytoplankton source specificity of specific bacteria, different bacterial phyla are specialized in the utilization of different DOMp compounds. Typically, members of the Flavobacteria are the principal consumers of HMW compounds, whereas members of the alphaproteobacterial Roseobacter clade prefer LMW compounds (32, 35, 37, 38). However, the same bacterial phyla may adapt to the transformation of DOMp compounds by changing expression patterns of transporters (42), and various HMW compounds of the same molecular weight may reveal tremendous differences in their accessibility due to compositional and structural differences (43).
In a previous set of experiments, we showed that DOMp produced by Prochlorococcus strain MIT9312 provides carbon, nitrogen, and phosphorus to accompanying heterotrophic bacteria. We also showed that the uptake of nitrogen and phosphorus bound to DOMp is preferred over inorganic forms (7). Here, we further explore how the identity of the specific DOMp producers and the time after DOMp pulses impact the accompanying bacterial community dynamics. We addressed the following questions: (i) Are there differences between DOMp from a cyanobacterium and a diatom in terms of the ability of the bacterial community to utilize them as C and P sources? (ii) To what magnitude do the phytoplankton producers impact the accompanying bacterial community? (iii) How do bacterial communities change over time following DOMp pulses? (iv) Which are the primary active bacteria in DOMp utilization from both phytoplankton producers? (v) Does the specificity of heterotrophic bacteria to the phytoplankton producers change with DOMp transformation over time?
In order to begin answering these questions, we added DOMp of two distinct phytoplankton species belonging to globally important genera—the diatom Skeletonema marinoi and the cyanobacterium Prochlorococcus marinus strain MIT9312—to natural bacterial communities and determined the bacterial responses over a time course of 72 h.

RESULTS

Experimental design and nutrient concentrations.

An elaborated description of the experimental set-up is given in Materials and Methods. For better traceability throughout the manuscript, a graphical overview of the experimental design is given as Fig. 1. Briefly, incubation bottles were filled with prefiltered coastal water from the eastern Mediterranean, amended with cyanobacterium and diatom DOM, and placed in flowthrough tanks in the dark (Fig. 1). To monitor inorganic nutrient concentrations during the experiment, we measured PO43–, NOx, and NH4+. PO43– concentrations were ca. 0.1 to 0.2 μM in the control and the diatom DOM treatment, whereas ca. 1 μM PO43– were accidently added with the cyanobacterium DOM. However, neither cyanobacterium DOM treatments with high PO43– concentration, nor control or diatom DOM treatments showed pronounced changes in PO43– concentrations during the experiment (see Fig. S1 in the supplemental material). NOx concentrations at the start of the experiment were ~2.5 μM and remained constant throughout the experiment in the control and cyanobacterium DOM treatments, but decreased after 48 h in the diatom DOM treatment (Fig. S1). NH4+ concentrations ranged between 1 and 15 μM but did not reveal obvious trends over time. Nevertheless, control samples showed slightly higher NH4+ concentrations than the DOMp (diatom and cyanobacterium) treatments, which were, however, significantly higher only after 24 h of exposure (Fig. S1).
FIG 1
FIG 1 Schematic overview of the experimental setup. Prochlorococcus MIT9312 (cyanobacterium) and Skeletonema marinoi (diatom) were grown to the early decline phase. Cell-free DOMp was harvested and mixed with prefiltered natural water. Incubation bottles were placed in darkened flowthrough tanks.

Bacterial cell numbers, bacterial production, and alkaline phosphate activity.

Initial bacterial cell numbers were ca. 2.3 × 106 cells mL−1 and increased after 24 h of exposure in all treatments, but levelled off to the original value after 48 and 72 h in both DOMp treatments, and decreased to ca. 1.8 × 106 cells mL−1 in the controls (Fig. 2A). On the other hand, bulk measurements of bacterial production (BP) increased in all treatments but not in the controls during the experiment. BP differed significantly between the different treatments at all time points (Fig. 2B). After 24 h, BP in the diatom DOM treatment was higher than BP in the controls and the cyanobacterium DOM treatments, whereas after 48 and 72 h both phytoplankton DOM treatments yielded similar BP, which were significantly higher than that of the control (Fig. 2B). To test for phosphorus starvation in the individual treatments, we measured the production of alkaline phosphatase activity (APA). Bulk measurements of APA yielded lower activity in the DOMp treatments than in the controls after 48 h, without any statistical differences between both DOMp producers (Fig. 2C). However, after 24 h, diatom DOM treatments showed ca. 4 times lower APA than the cyanobacterium DOM treatments (Fig. 2C).
FIG 2
FIG 2 (A to C) Bacterial cell numbers (A), bacterial production (B), and alkaline phosphatase activity (APA) (C) for different time points and treatments. The letters in the panels represent the outcomes of Tukey post hoc tests between the treatments for each time point (see Table S1 for further information).

Bacterial community responses to DOMp additions.

In order to evaluate the consequences of different DOMp producers and the succession of DOMp on the richness of the active bacterial communities, we calculated Shannon diversities for the individual treatments and time points. In all samples, the diversity remained in a narrow margin, with values ranging from ca. 5.4 to 6.2 (Fig. 3). Nevertheless, after 24 and 48 h, samples in the diatom DOM treatments revealed significantly higher diversities than the cyanobacterium DOM and control samples at this time point (Fig. 3). The increased diversity in diatom samples was consistent also after 72 h of incubation, but not in a significant way (P = 0.055, Table S1). Next, we examined shared and unique amplicon sequence variants (ASVs) between treatments at different time points using Venn analyses. When shared ASVs between all treatments (control, diatom DOM, cyanobacterium DOM) were compared with each other, no trend over time was obvious (13%, 22%, 20%, and 20% after 6, 24, 48, and 72 h of incubation, respectively), whereas ASVs that were only shared between both DOMp treatments (i.e., DOMp generalists) showed a slight increase over time, with shares of 2.3%, 3.5%, 4.3%, and 4.4% for 6 h, 24 h, 48 h, and 72 h of incubation, respectively (Fig. S2). On the other hand, at the different time points (6, 24, 48, 72 h) exclusive (i.e., specialist) ASVs in cyanobacterium- (22%, 20%, 20%, and 19%) and diatom-derived DOM (18%, 23%, 26%, and 25%, respectively) revealed slightly decreasing (cyanobacterium) and variable (diatom) trends over time (Fig. S2).
FIG 3
FIG 3 Shannon diversities for the different time points and treatments. The letters in the panels represent the outcomes of Tukey post hoc tests between the treatments for each time point (see Table S1 for further information).
We then explored the impact of the different treatments and proceeding times on the active bacterial communities with nonmetric multidimensional scaling (NMDS) and Bray-Curtis dissimilarities. Different treatments (t0, control, diatom DOM, cyanobacterium DOM; Fig. 4A) showed significantly different bacterial communities (analysis of similarity [ANOSIM], R = 0.3, P = 0.001), and this difference remained if the comparison was performed only between the DOMp types (ANOSIM, R = 0.22, P = 0.001; Fig. 4A). If treatments were pooled together and samples grouped into different incubation times (0 h, 6 h, 24 h, 48 h, and 72 h), significant differences in communities between the different time points could be detected (ANOSIM, R = 0.4, P = 0.001; Fig. 4A).
FIG 4
FIG 4 Effects of different treatments and proceeding times on bacterial communities. (A) Nonmetric multidimensional scaling (NMDS) for the different treatments and time points combined with EnvFit analyses. The red squares indicate the calculated centers of the different treatments; the arrow indicates the direction of proceeding time. (B) Bray-Curtis dissimilarities of bacterial communities between different time points in the same treatment. (C) Bray-Curtis dissimilarities of bacterial communities between different treatments at the same time point. (B and C) Letters above the box-plots refer to the outcomes of Tukey post hoc tests for differences over time in the same treatment (B) and for time-dependent differences between different treatments (C). Further information and additional tests (different treatments, same time point comparisons [B] and different treatment comparisons at the same time point [C]) are provided in Table S1.
In all treatments, time-dependent changes in the active bacterial community were significantly higher in the first 24 h of exposure (Bray-Curtis dissimilarities between t0 and 6 h as well as between 6 h and 24 h) than dissimilarities between later time points (Fig. 4B). Analogous to comparisons between different time points in the same treatments, community differences between different treatments at the same time point were most pronounced after 6 h, and the overall highest difference was obtained between diatom and cyanobacterium DOM treatments (Fig. 4C). The differences between bacterial communities in diatom and cyanobacterium DOM treatments, however, diminished after 24 h but successively increased again after 48 and 72 h (Fig. 4C).
General patterns revealing bacterial community differences between different treatments and incubation times were confirmed by abundance distributions of the overall 100 most abundant ASVs (defined as the sums of the subsampled reads for all samples; Fig. 5). A dendrogram of the ASVs yielded three major clusters, which were largely driven by taxonomy, with Saccharospirillaceae, Spongiispira, SAR11, and KI89a ASVs in cluster I, Ascidiaceihabitans, Glaciecola, and Aureimarina ASVs dominating cluster II, and a diverse consortium of ASVs (including Synechococcus, Rhodobacteraceae, and others) in cluster III (Fig. 5). However, if the 100 most abundant ASVs were analyzed for the dominant ASVs in the different treatments (t0, control, cyanobacterium, and diatom DOM, where all time points of the three latter were pooled), t0 samples were dominated by ASVs from SAR11 and Synechococcus, whereas 11 ASVs of Saccharospirillaceae, 1 Spongiispira ASV and 1 Glaciecola ASV dominated the controls (Fig. 5; Fig. S3). On the other hand, the cyanobacterium DOM treatments were exclusively dominated by ASVs of Glaciecola, whereas two Flavobacteriaceae ASVs, one Ascidiaceihabitans ASV, one SAR11 ASV, and four Glaciecola ASVs dominated the diatom DOM samples (Fig. 5; Fig. S3A). Additional tests for DOMp generalists, i.e., ASVs that dominate DOMp samples independent of the producer species (assigned treatments: all samples containing DOMp, t0, control), yielded 21 Glaciecola and 2 Flavobacteriaceae ASVs (Fig. S3B). If different incubation times were analyzed for their dominant ASVs (and the different treatments pooled), t0 samples were defined by a broad array of bacterial genera and ASVs, with the phototrophic Synechococcus and heterotrophic SAR11, the OM43 clade, Ascidiaceihabitans, AEGEAN-169, and the NS2b, NS4, NS5, and NS9 marine group contributing several ASVs. The 6 h incubation samples were especially dominated by ASVs of the Rhodobacteraceae, 24 h incubations solely by Glaciecola ASVs, 48 h incubations by 7 ASVs of Glaciecola and 1 ASV from Pseudoalteromonas, and the 72 h incubations by 11 ASVs of Saccharospirillaceae and 1 Aureimarina ASV (Fig. 5; Fig. S3C).
FIG 5
FIG 5 Heatmap of the overall 100 most abundant ASVs for the different treatments and time points. The dendrogram on the left side of the heatmap is based on hierarchical clustering of ASVs. The table on the right side of the heatmap provides ASV taxonomy (first column with color code) and dominant ASVs for the different treatments (t0, diatom DOM, cyanobacterium DOM, control, backed with light-gray), as well as different time points (0 h, 6 h, 24 h, 48 h, 72 h, different treatments pooled, backed with brown-gray; Fig. S3A and C). Dominant ASVs of the treatments/time points are indicated by filled boxes in the color of the respective taxonomy. ASVs of the same genus are defined by numbers in the taxonomic color code.

DISCUSSION

In this study, we described the bacterial utilization of DOMp derived from two primary sources of marine organic matter, i.e., diatoms and cyanobacteria (36), in coastal waters of the eastern Mediterranean in winter. We found similar bacterial responses to diatom and cyanobacterial DOM additions in terms of bacterial cell numbers, bacterial production (BP), and alkaline phosphatase activity (APA). Nevertheless, faster bacterial responses in BP and APA following the addition of diatom DOM suggest greater accessibility of carbon and phosphorus from these DOM compared to cyanobacterium DOM. Furthermore, via 16S rRNA sequencing we demonstrated that DOMp from different producers causes increases in the number and/or activity of different bacterial amplicon sequence variants (ASVs). However, the most pronounced differences between active bacterial communities of both DOMp types were obtained shortly after DOMp additions (6 h), indicating that DOMp from different producers likely differs in the highly bioavailable DOMp fraction. We also observed bacterial community successions over time, which we hypothesize are due to a shift from taxa effective in the utilization of highly bioavailable, producer-specific DOMp compounds shortly after the DOMp pulse toward ASVs specialized in less bioavailable compounds at the later time points. In summary, we conclude that bacterial communities experiencing phytoplankton DOMp pulses are relevantly defined by the original DOMp composition as well as temporal succession in substrate quality following the DOMp pulse.

Effects of DOMp additions on bacterial cell numbers, bacterial production, and alkaline phosphatase activity.

We confirmed bacterial utilization of phytoplankton DOMp as a carbon source with measurements of bacterial cell numbers and production (BP) (Fig. 2) and found indications of phosphorus utilization due to lowered alkaline phosphatase activity (APA) in DOMp treatments (Fig. 2). Absolute cell numbers appear to be surprisingly high for coastal Mediterranean waters, with ~2.2 × 106 cells mL−1 at t0 (Fig. 2A). Yet these values correspond well to our previous study in winter (also ~2.2 × 106 cells mL−1 at t0), Eigemann et al. [7], as well as to a time series study in the eastern Mediterranean (~0.7 to 1.4 × 106 cells mL−1 in January/February, Raveh et al. [44]), both sampled from the same station. Likewise, increased BP in both DOMp treatments corroborated bacterial utilization of phytoplankton-derived DOC, with values close to those measured by Raveh et al. (44) (~0.8 to 1.6 μg C h−1 l−1 in January/February versus ~0.5 μg C h−1 l−1 at t0 in the present study) as well as to our previous experiment (t0 ~0.13 fg C cell−1 h−1 in Eigemann et al. [7] versus ~0.2 fg C cell−1 h−1 in this study).
Despite identical inorganic P concentrations, diatom DOM treatments revealed significantly lower APA compared to controls (Fig. 2C; Fig. S1), suggesting that diatom DOM was used as an organic P source. This finding is strengthened by findings that bacterial communities in the Mediterranean Sea compensate for low inorganic nutrient levels with organic forms (45, 46), and phytoplankton DOM represents a substantial source of it (27). We added approximately 1 μM PO43– with the cyanobacterium DOM (Fig. S1), which makes a definitive statement on P utilization derived from cyanobacterium DOM difficult. However, in our previous study we showed a preferential utilization of organic nutrient forms provided from cyanobacterium DOM compared to inorganic forms (7). We may have experienced a similar preferential organic P utilization in the present study, as suggested by constant inorganic nutrient levels with contemporaneous decreasing APA over time in the cyanobacterial DOM treatments (Fig. 2; Fig. S1).

Faster accessibility of diatom DOM and producer-specific effects on bacterial diversity.

DOMp differ in its composition between different phytoplankton groups (16, 21, 22), as well as between species in the same group (13, 22), which has consequences for their accessibility to the bacterial community (24, 25, 36). Our results suggest a faster and/or higher accessibility of diatom DOM for bacterial biomass synthesis and phosphorus utilization, due to approximately 2 times higher BP and 2.5 times lower APA than bacteria exposed to the cyanobacterium DOM after 24 h of exposure (Fig. 2; bulk measurements). These differences were not attributed to differing cell abundances between treatments (Fig. S4; per-cell measurements). These results seem surprising at a glance, considering that we added less carbon with the diatom than cyanobacterium DOM (14 versus 25 μM; see “Experimental Setup,” below) and also added inorganic PO43– with the cyanobacterium DOM (see above). However, these results are supported by Kieft et al. (36), who found a higher 13C enrichment in bacteria exposed to13C-labeled diatoms than those exposed to cyanobacterium DOM after 15 h, and a mass spectrometric analysis of both DOMp types, where diatom DOM contains a higher fraction of P (although no P uptake of bacteria was measured) (24). However, we cannot rule out an impact of the bacterial community composition on BP outcomes (47), since we found slightly higher cell numbers in cyanobacterium DOM treatments (in contrast to BP and APA) after 24 h (Fig. 2). Thus, cyanobacterial DOM could potentially induce the growth of different and less active organisms.
Higher Shannon diversities throughout the experiment, furthermore, are in accordance with outcomes recorded by Landa et al. (24), who also found higher bacterial richness in diatom-derived DOM treatments, and suggest that diatom DOM can be used by a larger fraction of the bacterial community than can the cyanobacterium DOM, or that taxa are competitively excluded from growing on cyanobacterial DOM (Fig. 3). The first finding is supported by a more diverse DOM composition of diatoms (compared to cyanobacteria) (22), the production of easily accessible DOM such as the polysaccharide laminarin (48), and more evenly distributed individual sugars and amino acids in diatom DOM (which infer a higher diversity of polymers and thus can be used by a broader share of bacteria) (24). Cyanobacterial DOM, for their part, contain lipo-polysaccharides (49) and fucose containing polysaccharides (24), which require sophisticated degradation pathways. Nevertheless, cyanobacterium and diatom DOM treatments experienced different DOC and P concentrations, which may have impacted bacterial responses, and we did not perform any measurements on DOM composition. Nevertheless, because the responses to cyanobacterial DOM were not consistent with the nutrient or DOC concentrations (higher APA when PO43– was higher and lower BP despite higher DOC concentrations), we postulate that the differences between the two DOMp sources are due to DOM quality and not quantity. Further, higher DOCp concentrations were utilized by a larger fraction of heterotrophic bacteria (50), suggesting that the higher bacterial diversity in our diatom DOM treatments (which experienced lower DOC concentrations than the cyanobacterium DOC treatments) indeed derives from DOM quality and not quantity.

DOMp producer-specific ASVs.

In samples of both phytoplankton producers, we found multiple different ASVs of Alphaproteobacteria, Gammaproteobacteria, Bacteroidetes, Verrucomicrobia, Firmicutes, Marinimicrobia, Latescibacteriota, and Actinobacteria as abundant members (Table S2). The vast majority of abundant ASVs in DOMp treatments, however, were assigned to Gammaproteobacteria, the Bacteroidetes family Flavobacteriaceae, and the Rhodeobacterales clade of Alphaproteobacteria (Table S2). These three bacterial groups are primary utilizers of DOMp and are independent from the producer species and time after DOMp pulses. They are well known as dominant members in association with phytoplankton and/or bacterial communities after phytoplankton DOMp addition (31, 32, 3638, 5156).
At the ASV level, however, we identified 17 different Glaciecola ASVs in cyanobacteria and 4 different Glaciecola ASVs in diatom DOM samples dominating the respective treatments (Fig. S3). Glaciecola (Gammaproteobacteria, Alteromonadaceae) were the only ASVs to specifically respond to the cyanobacterium DOMp samples, which is similar to the results of our previous study (7). Similar results were also obtained by Sarmento and Gasol (54), who found preferential uptake of Prochlorococcus DOM by Alteromonadaceae, and by Kearney et al. (57), who identified specific associations of Alteromonadaceae to picocyanobacteria. However, we identified Glaciecola ASVs in the communities responding to both DOMp types (Fig. S3A and C), and Glaciecola reached high relative abundances under conditions dominated by different phytoplankton species in natural, coastal blooms (58). Thus, we suggest that Glaciecola might be considered a generalist phytoplankton-associated bacterium (or DOMp generalist; Fig. S3B). Apart from the four Glaciecola ASVs discussed above, diatom DOM samples were dominated by two Flavobacteriaceae ASVs, as well as one SAR11 and one Ascidiaceihabitans ASV (Fig. S3A). These assignments are supported by numerous studies that found general associations of Proteobacteria and Bacteroidetes with diatoms (reviewed by Amin et al. [59]). However, despite several specific phytoplankton-bacterium associations that were identified in our study as well as others, one should be careful to draw general conclusions. First, we indeed defined several ASVs as phytoplankton producer specific, but one should keep in mind that other ASVs were also abundant in either diatom or cyanobacterium samples, whose nonconsideration as producer specific might be a result of statistical cutoffs, which may not reflect smooth transitions in natural systems. Second, some studies show different results. For example, we identified some Glaciecola ASVs responding to cyanobacterium DOM, whereas Landa et al. (24) found it specific to diatom DOM (and did not find it in cyanobacterium DOM treatments). Further, Rhodobacterales were specifically associated with picocyanobacteria (36, 57), and Prochlorococcus supported the growth of SAR11 in coculture experiments (51), whereas we assigned Ascidiaceihabitans (Rhodobacterales) and SAR11 only to diatom treatments. Thus, general assignments of specific phytoplankton-bacterium associations are challenging, which might be attributed to differences in abiotic parameters (which are important determinants of bacterial associations with phytoplankton [60]) as well as to different bacterial source communities (which greatly impact associations with phytoplankton [61]).

Diminishing impact of the DOMp producer on bacterial communities over time.

We found indications of a strong impact of the phytoplankton producer on bacterial communities shortly after DOMp addition and a diminishing impact of the producer with increasing time, suggested by the most pronounced producer-specific responses as well as bacterial community differences between both DOMp types shortly after DOMp additions (Fig. 2 and 4). We hypothesize that this is caused by a high fraction of producer-specific compounds in the highly bioavailable DOMp fractions, whereas less bioavailable compounds may have a more general composition. This assumption is supported by increases in shared ASVs between both DOMp types over time (when, presumably, highly bioavailable compounds decrease; Fig. S2), the decrease of Bray-Curtis dissimilarities from 6 to 24 h of exposure (Fig. 4C), and the different outcomes for BP, APA, and cell numbers after 24 h, but not after 48 and 72 h (Fig. 2). These results are different from those of Landa et al. (24), who suggested that labile compounds have a minor effect on the bacterial community and that pronounced effects of the producer species occur if enzymatic machineries are required. In their study, however, a continuous supply of cyanobacterium as well as diatom DOM was applied, which is different from our study, in which a single pulse of DOM was added. This may have masked a succession from highly to less bioavailable compounds, and indeed, the fraction of shared bacterial operational taxonomic units (OTUs) between cyanobacteria and diatom DOM treatments increased with time, analogous to our study. Our assumption is further supported by lower recurrences of phytoplankton species than that of heterotrophic bacterial OTUs days to weeks after phytoplankton blooms between years in a 4-year time series of the German bight (38), which may be a result of high similarities in the less bioavailable fraction of phytoplankton DOM derived from different species. Finally, a mechanistic inference study of the same 4-year data set revealed ~3 times higher recurrences of DOM species than of phytoplankton species between years in spring blooms (62), which, taking into account that highly bioavailable compounds may be rapidly consumed, further supports our hypothesis.
Despite observing the most pronounced bacterial community changes between 0 and 6 h as well as 6 and 24 h of exposure (Fig. 4B), successions of the bacterial community occurred throughout the incubation period (see Bray-Curtis dissimilarities between 48 and 72 h of exposure, Fig. 4B), enabling the identification of time point-dominating bacterial ASVs (Fig. S3C). Bacterial community successions after DOMp pulses presumably follow the exploitation of different DOMp compounds over time, from highly bioavailable compounds shortly after the pulse to less bioavailable compounds at later time points (37). The broad consortium of different bacteria dominating t0 changed to four Rhodobacter (Alphaproteobacteria) ASVs and one each of Glaciecola (Gammaproteobacteria) and the NS2b marine group (Flavobacteriaceae) ASVs dominating the 6 h exposure (Fig. S3C). The high share of Alphaproteobacteria at this time point underlines their task as first responders after DOMp pulses, possibly as a result of their high expression of ABC and TRAP transporters that enable the uptake of various LMW compounds (36, 37). After 24 and 48 h, we only identified Gammaproteobacteria (and only one non-Glaciecola ASV) as highly dominant at those time points, and after 72 h, eleven Gammaproteobacteria (all Saccharospirillaceae), and only one Flavobacterium were identified as highly dominant (Fig. S3C). Interestingly, these outcomes reflect the patterns from Teeling et al. (37), who also found a succession of peaking gammaproteobacterial clades, after the primary response of Alphaproteobacteria to a diatom bloom. The succession of Alphaproteobacteria to Gammaproteobacteria and, to a lesser extent, Flavobacteria is a recurrent phenomenon after DOMp pulses (38) and can be explained by the genetic “toolkit” that enables the utilization of numerous HMW compounds (37, 63).

Limitations and synthesis.

While our mesocosm experiments were designed to mimic natural conditions (i.e., phytoplankton blooms producing DOMp), there are several inherent issues with such bottle incubations. These can be seen by the increased bacterial production in the controls (Fig. 2B), similar temporal developments of bacterial communities in all treatments (including the controls; Fig. 4A), and an increase in shared ASVs between all treatments with increasing time (Fig. S2). Indeed, it is almost impossible to completely omit bottle effects in these types of experiments (e.g., Elovaara et al. [64] and Luria et al. [65]). Nevertheless, despite of obvious community changes in the controls from t0 to 6 h of incubation, the highest community differences for these time steps were achieved between both DOMp treatments, suggesting the system was responding primarily to the experimental manipulation (Fig. 4C). Furthermore, we found pronounced differences between the DOMp from the two phytoplankton producers in APA, BP, and cell numbers after 24 h of exposure (Fig. 2) and defined producer-specific ASVs, as well as DOMp generalist ASVs (Fig. S3B). Thus, the observed bottle effects did not prohibit the gain of valuable information.
We also added different amounts of cyanobacteria and diatom DOC (see “Experimental Setup”, below) and treated both DOMp types as “black boxes”; i.e., we did not characterize their composition. As a consequence, we can only assume that the observed differences in responsive bacterial phyla, BP, and APA derive from differences in the DOM quality. The assumption that different phytoplankton groups yield different DOM qualities is supported by numerous studies that revealed significant differences in the DOM of different phytoplankton (18, 19, 25) and, specifically, between that of cyanobacteria and diatoms (22, 24, 26, 54). Another aspect that is not fully addressed in our study is the phenotypic impact on phytoplankton DOM, i.e., the impact of abiotic and biotic factors on the quantity and quality of the released DOMp (8). We tried to account for the most important factors, such as the phytoplankton growth phase (both cultures were harvested at the early decline phase) and light and temperature settings (see “Experimental Setup”, below) but might have experienced a colimitation of N and P in the diatom culture and a N limitation in the cyanobacterium culture at the point of DOMp harvest (Fig. S1). This is another factor that could have affected the DOMp and subsequent community response.
Contextualizing our results in terms of those studies with a similar focus suggests that bacterial diversity, APA, BP, and cell numbers of phytoplankton accompanying bacterial communities are (at least partly) determined by the phytoplankton producer species (Fig. 6). These differential responses may be caused by different DOMp compositions released by different phytoplankton groups and species (8, 18, 19), specifically those of cyanobacteria and diatoms (20, 22) (Fig. 6). Additionally, the bacterial community participates in the succession of DOMp from highly bioavailable to less bioavailable (37, 66), and we were able to assign time point-defining ASVs (Fig. S3C). However, decreasing differences with increasing time of bacterial communities in the same treatment (Fig. 4B), as well as large community differences between both DOMp treatments after 6 h of incubation (Fig. 4C), suggest a particularly high impact of the host shortly after DOMp addition (Fig. 6). After exploitation of species-specific highly bioavailable compounds, the remaining DOMp of different phytoplankton may become more similar, and bacterial communities might be especially impacted by the DOMp accessibility (Fig. 6), i.e., the state of DOMp succession from highly bioavailable to less bioavailable compounds.
FIG 6
FIG 6 Summary figure of the obtained results (indicated by *) and outcomes of similar studies on bacterial utilization of cyanobacterium- and diatom-derived DOM. a, Kieft et al. (36); b, Sarmento and Gasol (54); c, Landa et al. (24); d, Eigemann et al. (7); e, Tada and Suzuki (26). If different outcomes at different time points were obtained, a time specification is given after the reference. (Right side) Differently colored bacteria indicate taxonomically different bacteria, with higher similarities between colors indicating higher taxonomical similarities. The approximating colors of diatom and cyanobacterial DOM with progressing time indicate the succession of DOM compounds from highly to less bioavailable, which might become more similar.

MATERIALS AND METHODS

Experimental setup.

The experiment was performed in winter from 14 January 2019 until 17 January 2019 at the Israel Oceanographic and Limnological Research center in Haifa, Israel. Nalgene bottles (20 L) were filled with 18 L of 50-μm prefiltered seawater (intake pipe at 5-m depth, ca. 50 m from shore), amended with phytoplankton DOM, and placed in 1 m³ natural seawater flowthrough tanks darkened with black plastic sheets to minimize the effect of the remaining phytoplankton. The setup did not account for grazers, because neglectable effects of grazers on bacterial community compositions were shown in previous mesocosm experiments in the Mediterranean (67). We added 25 μM cyanobacterium (Prochlorococcus marinus MIT9312) DOC, in order to produce outcomes comparable with those of similar studies (see Eigemann et al. [7]), and 14 μM diatom (Skeletonema marinoi) DOC (due to technical issues, several DOMp-containing bottles broke at an air transfer) to the background concentration of the seawater. Diatoms and cyanobacteria are the two primary sources of marine organic carbon (36), where Prochlorococcus contributes up to 8.5% of oceanic photosynthesis and 50% of surface open-ocean chlorophyll (68), whereas diatoms execute up to 40% of primary production in coastal areas (69). Each of the three treatments (control, diatom DOM, cyanobacterium DOM) contained four biological replicates, and samples were taken at t0 (50 μm prefiltered seawater; nutrients, RNA, APA, BP, cell numbers), after 1 (only nutrients), 6 (nutrients and RNA), 24, 48, and 72 h (APA, BP, RNA, cell numbers, nutrients). The graphical overview of the experimental setup is given as Fig. 1.

Culture conditions and harvesting of DOMp.

Prochlorococcus marinus MIT9312 was grown under constant light (20 μ E m−2 sec−1) at 22°C in Pro99 medium, where the NH4+ concentration was reduced from 800 μM to 100 μM, resulting in the cells entering stationary stage due to N starvation (70). Skeletonema marinoi was grown in F/2 plus Si medium (71) with adjusted nutrient concentrations (final concentration of 200 μM N and 11.5 μM P) at 45 μ E m−2 sec−1 at 17°C in a conditioning cabinet. To obtain cell-free DOMp, batch cultures were harvested at the early decline phase by centrifugation, followed by filtration through a 0.22-μm polycarbonate filter. DOC concentrations of the phytoplankton cultures were ~1,100 μM for Skeletonema marinoi (measured via high-temperature combustion on a Shimadzu TOC-V machine) and ~800 μM for Prochlorococcus marinus (measured on a Shimadzu ASI-L autoanalyzer). Phytoplankton-derived dissolved organic matter (DOMp) from the early decline phase may reflect natural DOM compositions better than that from exponentially growing cultures (12) and includes DOM from phytoplankton exudation as well as cell lysis. DOMp from both types of phytoplankton was kept at −20°C until utilization (on dry ice during air transport).

Nutrient measurements.

Nutrients were measured colorimetrically following Hansen and Koroleff (72) by means of a Seal analytical QuAAtro constant flow analyzer after prefiltering through 0.2-μm pore-width filters. Concentrations of PO43–, NOx (NO2 + NO3), and NH4+ in the different treatments over the course of the experiment are given in Fig. S1.

Cell numbers, bacterial production, alkaline phosphatase activity, RNA extraction, DNA digestion, cDNA synthesis, and sequencing.

All analyses were performed as described in Eigemann et al. (7). Brief descriptions of the techniques and protocols used are given below.
(i) Cell numbers. Duplicate samples were fixed with cytometry-grade glutaraldehyde (0.125% final concentration), flash-frozen with liquid nitrogen and stored at −80°C until analysis. For measurements, samples were thawed in the dark at room temperature, stained with SYBR green I (Molecular Probes/Thermo Fisher) for 10 min at room temperature, vortexed, and run on a BD FACSCanto II flow cytometry analyzer system (BD 146 Biosciences).
(ii) Bacterial production (BP) and alkaline phosphatase activity (APA). BP was measured using the [4,5-3H]-leucine incorporation method (73). Samples were amended with 100 nmol of leucine L−1 (Perkin Elmer; specific activity, 156 Ci mmol−1) and incubated for 4 h in the dark at ambient surface seawater temperature (~19°C). Incubations were stopped by the addition of 100 μL of ice-cold 100% trichloroacetic acid (TCA). APA was determined by the 4-methylumbeliferyl phosphate (MUF-P; Sigma M8168) method according to reference 74. Substrate was added to a final concentration of 50 μM and incubated in the dark at ambient temperature for 4 h (same as BP).
(iii) RNA extraction, DNA digestion, and cDNA synthesis. In order to analyze the active members of the bacterial community, rRNA was extracted and transcribed into cDNA. Approximately 1.5 L of each incubation bottle was filtered directly onto 0.2-μm polycarbonate filters and extracted with TRI reagent. The remaining DNA was digested using the Turbo DNA-free kit (Invitrogen) using the manufacturer’s instructions, and successful digestions were tested by PCRs. RNA was transcribed into cDNA using MultiScribe reverse transcriptase following the manufacturer’s instructions (Invitrogen).
(iv) Sequencing and sequence analyses. Sequencing was performed using primers 515F and 926R (75) and an Illumina MiniSeq mid-output flow cell. All sequences were analyzed using the DADA2 (76) pipeline and the software packages R (77) and RStudio (78). Briefly, forward and backward primers were trimmed, forward and backward reads were truncated to 280 and 210 bases after quality inspections, respectively, and after merging of forward and backward sequences, only a consensus length between 404 and 417 bases was accepted. For taxonomic assignment, Silva database version 138 (79) was used. All chloroplasts, mitochondria, archaea, eukaryotes, and amplicon sequence variants (ASVs) without any taxonomic affiliation were discarded from downstream analyses. The complete ASV table with absolute, subsampled read counts, and meta data for all samples is accessible as Table S3. After inspections of rarefaction curves, samples with less than 10,154 reads were discarded from further analyses (t0, replicate four [41 reads], and cyanobacterium DOM, replicate four, 6 h of incubation [1,955 reads]), and all remaining samples were subsampled to 10,154 reads. One additional sample was discarded from downstream analyses because of a sandy RNA filter appearance and NMDS clustering far away from all other samples (control, replicate four, 6 h of incubation). Thus, from the original 52 samples, 49 were used for elaborated analyses.

Statistics.

We tested for differences between the treatments at individual time points with Shannon indices, cell numbers, alkaline phosphatase activity (APA), and bacterial production (BP). Variance homogeneities between the different treatments for each time point were tested with Levene tests with the R package car (80). If homogeneities were given, ANOVAs with subsequent Tukey post hoc tests were calculated. If no homogeneities of variances were given, Kruskal-Wallis tests with subsequent Tukey-Nemenyi post hoc tests were executed with the R package PMCMR (81). To test for unique and shared ASVs between treatments, we merged biological replicates as follows: All subsampled, absolute read counts of >0 were set to 1, and subsequently, the mean abundance of the biological replicates was calculated for each treatment. An ASV was set as present if the calculated mean for the treatment was ≥0.5. With these matrices, Venn counts were calculated and Venn diagrams were generated with the limma package (82). Changes in bacterial communities were examined with Shannon diversities, nonmetric multidimensional scaling (NMDS), Bray-Curtis dissimilarities between treatments and time points, and heatmaps of the overall 100 most abundant ASVs. Differences of Bray-Curtis dissimilarities between samples were tested with ANOVAs (homogeneities of variances given) or Kruskal-Wallis tests (homogeneities of variances not given), with subsequent Tukey post hoc or Tukey-Nemenyi post hoc tests, respectively (see above). NMDS was executed with Bray-Curtis distances in the vegan package (83) with subsampled absolute ASV reads. To test for differences between bacterial communities, analyses of similarities (ANOSIM) as well as envfit (environmental fits on ordinations) analyses were conducted using the vegan package. Heatmaps for the overall 100 most abundant subsampled ASVs were executed using the Heatmap3 package (84). To test if specific ASVs of these 100 most abundant ASVs are associated with specific time points or treatments (control, diatom DOM, cyanobacterium DOM), linear discriminant analysis (LDA) effect size (LEfSe) analyses (85) were executed with the online Galaxy tool at https://huttenhower.sph.harvard.edu/galaxy. Time points and treatments were assigned as class, and samples were assigned as subjects. Then, a “one-against-all” strategy was applied with normalization and the LDA effect size threshold was set to 2.5. All analyses, except denoted otherwise, were executed with R (77) and RStudio (78). Statistical outcomes are summarized in Table S1. All graphics, except denoted otherwise, were executed with the ggplot2 package (86) and refined with the freeware Inkscape (https://inkscape.org).

Data availability.

Forward and backward sequence reads were deposited at the European Nucleotide Archive under project PRJEB44710, accession number ERA4147793, study ERP128782. Sample identification and meta-data information for sequences are given in Table S4. Measurements, calibrations and calculations of cell numbers, bacterial production (BP), and alkaline phosphatase activity (APA) are given in Table S5.

ACKNOWLEDGMENTS

We thank Stefan Green (DNA Services Facility at the University of Illinois at Chicago) for the amplicon sequencing and Christian Burmeister for nutrient analyses. We also thank Tom Reich, Dalit Roth-Rosenberg, Tal Luzzatto-Knaan, Noam Nago, and Natalia Belkin for excellent help with the experiment.
This work was supported by the Human Frontier Science Program (HFSP) through grant number RGB 0020/2016 (D.S., M.V., and H.-P.G.), by the National Science Foundation, United States-Israel Binational Science Foundation Program in Oceanography (grant number 1635070/2016532 to D.S.) and by the Israel Ministry of Science and Technology (grant number 3-17404 to D.S.).

Supplemental Material

File (aem.00539-23-s0001.pdf)
File (aem.00539-23-s0002.csv)
File (aem.00539-23-s0003.csv)
File (aem.00539-23-s0004.csv)
File (aem.00539-23-s0005.xlsx)
ASM does not own the copyrights to Supplemental Material that may be linked to, or accessed through, an article. The authors have granted ASM a non-exclusive, world-wide license to publish the Supplemental Material files. Please contact the corresponding author directly for reuse.

REFERENCES

1.
Field CB, Behrenfeld MJ, Randerson JT, Falkowski P. 1998. Primary production of the biosphere: Integrating terrestrial and oceanic components. Science 281:237–240.
2.
Azam F, Malfatti F. 2007. Microbial structuring of marine ecosystems. Nat Rev Microbiol 5:782–791.
3.
Azam F, Fenchel T, Field J, Gray JS, Meyer-Reil LA, Thingstad F. 1983. The ecological role of water-column microbes in the sea. Mar Ecol Prog Ser 10:257–263.
4.
Roth-Rosenberg D, Aharonovich D, Omta AW, Follows MJ, Sher D. 2021. Dynamic macromolecular composition and high exudation rates in Prochlorococcus. Limnol Oceanogr 66:1759–1773.
5.
Beliaev AS, Romine MF, Serres M, Bernstein HC, Linggi BE, Markillie LM, Isern NG, Chrisler WB, Kucek LA, Hill EA, Pinchuk GE, Bryant DA, Wiley HS, Fredrickson JK, Konopka A. 2014. Inference of interactions in cyanobacterial-heterotrophic co-cultures via transcriptome sequencing. ISME J 8:2243–2255.
6.
Riemann L, Holmfeldt K, Titelman J. 2009. Importance of viral lysis and dissolved DNA for bacterioplankton activity in a P-limited estuary, Northern Baltic Sea. Microb Ecol 57:286–294.
7.
Eigemann F, Rahav E, Grossart H-P, Aharonovich D, Sher D, Vogts A, Voss M. 2022. Phytoplankton exudates provide full nutrition to a subset of accompanying heterotrophic bacteria via carbon, nitrogen and phosphorus allocation. Environ Microbiol 24:2467–2483.
8.
Mühlenbruch M, Grossart HP, Eigemann F, Voss M. 2018. Mini-review: phytoplankton-derived polysaccharides in the marine environment and their interactions with heterotrophic bacteria. Environ Microbiol 20:2671–2685.
9.
Seymour JR, Amin SA, Raina JB, Stocker R. 2017. Zooming in on the phycosphere: the ecological interface for phytoplankton-bacteria relationships. Nat Microbiol 2:17065.
10.
Cole JJ. 1982. Interactions between bacteria and algae in aquatic ecosystems. Annu Rev Ecol Syst 13:291–314.
11.
Amin SA, Hmelo LR, van Tol HM, Durham BP, Carlson LT, Heal KR, Morales RL, Berthiaume CT, Parker MS, Djunaedi B, Ingalls AE, Parsek MR, Moran MA, Armbrust EV. 2015. Interaction and signalling between a cosmopolitan phytoplankton and associated bacteria. Nature 522:98–101.
12.
Christie-Oleza JA, Sousoni D, Lloyd M, Armengaud J, Scanlan DJ. 2017. Nutrient recycling facilitates long-term stability of marine microbial phototroph-heterotroph interactions. Nat Microbiol 2:17100.
13.
Grossart HP, Czub G, Simon M. 2006. Algae-bacteria interactions and their effects on aggregation and organic matter flux in the sea. Environ Microbiol 8:1074–1084.
14.
York A. 2018. Marine biogeochemical cycles in a changing world. Nat Rev Microbiol 16:259–259.
15.
Sison-Mangus MP, Jiang S, Kudela RM, Mehic S. 2016. Phytoplankton-associated bacterial community composition and succession during toxic diatom bloom and non-bloom events. Front Microbiol 7:1433.
16.
Biddanda B, Benner R. 1997. Carbon, nitrogen, and carbohydrate fluxes during the production of particulate and dissolved organic matter by marine phytoplankton. Limnol Oceanogr 42:506–518.
17.
Ríos A, Fraga F, Pérez FF, Figueiras FG. 1998. Chemical composition of phytoplankton and particulate organic matter in the Ría de Vigo (NW Spain). Scientia Marina 62:257–271.
18.
Aluwihare LI, Repeta DJ. 1999. A comparison of the chemical characteristics of oceanic DOM and extracellular DOM produced by marine algae. Mar Ecol Prog Ser 186:105–117.
19.
Castillo CR, Sarmento H, Álvarez-Salgado XA, Gasol JM, Marraséa C. 2010. Production of chromophoric dissolved organic matter by marine phytoplankton. Limnol Oceanogr 55:446–454.
20.
Romera-Castillo C, Sarmento H, Alvarez-Salgado XA, Gasol JM, Marrasé C. 2011. Net production and consumption of fluorescent colored dissolved organic matter by natural bacterial assemblages growing on marine phytoplankton exudates. Appl Environ Microbiol 77:7490–7498.
21.
Meon B, Kirchman DL. 2001. Dynamics and molecular composition of dissolved organic material during experimental phytoplankton blooms. Marine Chemistry 75:185–199.
22.
Becker J, Berube P, Follett C, Waterbury J, Chisholm S, DeLong E, Repeta D. 2014. Closely related phytoplankton species produce similar suites of dissolved organic matter. Front Microbiol 5:111.
23.
Myklestad SM. 2000. Dissolved organic carbon from phytoplankton, p 111–148. In Wangersky PJ (ed), Marine chemistry. Springer, Berlin, Germany.
24.
Landa M, Cottrell MT, Kirchman DL, Kaiser K, Medeiros PM, Tremblay L, Batailler N, Caparros J, Catala P, Escoubeyrou K, Oriol L, Blain S, Obernosterer I. 2014. Phylogenetic and structural response of heterotrophic bacteria to dissolved organic matter of different chemical composition in a continuous culture study. Environ Microbiol 16:1668–1681.
25.
Sarmento H, Romera-Castillo C, Lindh M, Pinhassi J, Sala MM, Gasol JM, Marrase C, Taylor GT. 2013. Phytoplankton species-specific release of dissolved free amino acids and their selective consumption by bacteria. Limnol Oceanogr 58:1123–1135.
26.
Tada Y, Suzuki K. 2016. Changes in the community structure of free-living heterotrophic bacteria in the open tropical Pacific Ocean in response to microalgal lysate-derived dissolved organic matter. FEMS Microbiol Ecol 92:fiw099.
27.
Livanou E, Lagaria A, Psarra S, Lika K. 2017. Dissolved organic matter release by phytoplankton in the context of the dynamic energy budget theory. Biogeosciences 42:1–33.
28.
Xiao X, Guo W, Li X, Wang C, Chen X, Lin X, Weinbauer MG, Zeng Q, Jiao N, Zhang R. 2021. Viral lysis alters the optical properties and biological availability of dissolved organic matter derived from prochlorococcus picocyanobacteria. Appl Environ Microbiol 87:e02271-20.
29.
Gómez-Consarnau L, Lindh MV, Gasol JM, Pinhassi J. 2012. Structuring of bacterioplankton communities by specific dissolved organic carbon compounds. Environ Microbiol 14:2361–2378.
30.
Rooney-Varga JN, Giewat MW, Savin MC, Sood S, LeGresley M, Martin JL. 2005. Links between phytoplankton and bacterial community dynamics in a coastal marine environment. Microb Ecol 49:163–175.
31.
Grossart HP, Levold F, Allgaier M, Simon M, Brinkhoff T. 2005. Marine diatom species harbour distinct bacterial communities. Environ Microbiol 7:860–873.
32.
Buchan A, LeCleir GR, Gulvik CA, Gonzalez JM. 2014. Master recyclers: features and functions of bacteria associated with phytoplankton blooms. Nat Rev Microbiol 12:686–698.
33.
Stock W, Blommaert L, De Troch M, Mangelinckx S, Willems A, Vyverman W, Sabbe K. 2019. Host specificity in diatom–bacteria interactions alleviates antagonistic effects. FEMS Microbiol Ecol 95:fiz171.
34.
Jackrel SL, Yang JW, Schmidt KC, Denef VJ. 2021. Host specificity of microbiome assembly and its fitness effects in phytoplankton. ISME J 15:774–788.
35.
Beier S, Rivers AR, Moran MA, Obernosterer I. 2015. The transcriptional response of prokaryotes to phytoplankton-derived dissolved organic matter in seawater. Environ Microbiol 17:3466–3480.
36.
Kieft B, Li Z, Bryson S, Hettich RL, Pan C, Mayali X, Mueller RS. 2021. Phytoplankton exudates and lysates support distinct microbial consortia with specialized metabolic and ecophysiological traits. Proc Natl Acad Sci USA 118:e2101178118.
37.
Teeling H, Fuchs BM, Becher D, Klockow C, Gardebrecht A, Bennke CM, Kassabgy M, Huang S, Mann AJ, Waldmann J, Weber M, Klindworth A, Otto A, Lange J, Bernhardt J, Reinsch C, Hecker M, Peplies J, Bockelmann FD, Callies U, Gerdts G, Wichels A, Wiltshire KH, Glockner FO, Schweder T, Amann R. 2012. Substrate-controlled succession of marine bacterioplankton populations induced by a phytoplankton bloom. Science 336:608–611.
38.
Teeling H, Fuchs BM, Bennke CM, Kruger K, Chafee M, Kappelmann L, Reintjes G, Waldmann J, Quast C, Glockner FO, Lucas J, Wichels A, Gerdts G, Wiltshire KH, Amann RI. 2016. Recurring patterns in bacterioplankton dynamics during coastal spring algae blooms. Elife 5:e11888.
39.
Hach PF, Marchant HK, Krupke A, Riedel T, Meier DV, Lavik G, Holtappels M, Dittmar T, Kuypers MMM. 2020. Rapid microbial diversification of dissolved organic matter in oceanic surface waters leads to carbon sequestration. Sci Rep 10:13025.
40.
Grossart H-P, Ploug H. 2001. Microbial degradation of organic carbon and nitrogen on diatom aggregates. Limnol Oceanogr 46:267–277.
41.
Jiao N, Herndl GJ, Hansell DA, Benner R, Kattner G, Wilhelm SW, Kirchman DL, Weinbauer MG, Luo T, Chen F, Azam F. 2011. The microbial carbon pump and the oceanic recalcitrant dissolved organic matter pool. Nat Rev Microbiol 9:555–555.
42.
Francis TB, Bartosik D, Sura T, Sichert A, Hehemann J-H, Markert S, Schweder T, Fuchs BM, Teeling H, Amann RI, Becher D. 2021. Changing expression patterns of TonB-dependent transporters suggest shifts in polysaccharide consumption over the course of a spring phytoplankton bloom. ISME J 15:2336–2350.
43.
Arnosti C, Wietz M, Brinkhoff T, Hehemann JH, Probandt D, Zeugner L, Amann R. 2021. The biogeochemistry of marine polysaccharides: sources, inventories, and bacterial drivers of the carbohydrate cycle. Annu Rev Mar Sci 13:81–108.
44.
Raveh O, David N, Rilov G, Rahav E. 2015. The temporal dynamics of coastal phytoplankton and bacterioplankton in the eastern Mediterranean Sea. PLoS One 10:e0140690.
45.
Krom MD, Emeis KC, Van Cappellen P. 2010. Why is the eastern Mediterranean phosphorus limited? Progr Oceanogr 85:236–244.
46.
Krom M, Kress N, Berman-Frank I, Rahav E. 2014. Past, present and future patterns in the nutrient chemistry of the Eastern Mediterranean, p 49–68. In Goffredo S, Dubinsky Z (ed), The Mediterranean Sea: its history and present challenges. Springer Netherlands, Dordrecht, The Netherlands.
47.
White AE, Giovannoni SJ, Zhao Y, Vergin K, Carlson CA. 2019. Elemental content and stoichiometry of SAR11 chemoheterotrophic marine bacteria. Limnol Oceanogr Lett 4:44–51.
48.
Becker S, Scheffel A, Polz MF, Hehemann JH. 2017. Accurate quantification of laminarin in marine organic matter with enzymes from marine microbes. Appl Environ Microbiol 83:e03389-16.
49.
Bhatnagar M, Bhatnagar A. 2019. Diversity of polysaccharides in cyanobacteria, p 447–496. In Satyanarayana T, Johri BN, Das SK (ed), Microbial diversity in ecosystem sustainability and biotechnological applications: volume 1 microbial diversity in normal & extreme environments. Springer Singapore, Singapore.
50.
Sarmento H, Morana C, Gasol JM. 2016. Bacterioplankton niche partitioning in the use of phytoplankton-derived dissolved organic carbon: quantity is more important than quality. ISME J 10:2582–2592.
51.
Becker JW, Hogle SL, Rosendo K, Chisholm SW. 2019. Co-culture and biogeography of Prochlorococcus and SAR11. ISME J 13:1506–1519.
52.
Riemann L, Steward GF, Azam F. 2000. Dynamics of bacterial community composition and activity during a mesocosm diatom bloom. Appl Environ Microbiol 66:578–587.
53.
Pinhassi J, Sala MM, Havskum H, Peters F, Guadayol O, Malits A, Marrasé C. 2004. Changes in bacterioplankton composition under different phytoplankton regimens. Appl Environ Microbiol 70:6753–6766.
54.
Sarmento H, Gasol JM. 2012. Use of phytoplankton-derived dissolved organic carbon by different types of bacterioplankton. Environ Microbiol 14:2348–2360.
55.
Goto S, Tada Y, Suzuki K, Yamashita Y. 2017. Production and reutilization of fluorescent dissolved organic matter by a marine bacterial strain, Alteromonas macleodii. Front Microbiol 8:507.
56.
Dadaglio L, Dinasquet J, Obernosterer I, Joux F. 2018. Differential responses of bacteria to diatom-derived dissolved organic matter in the Arctic Ocean. Aquat Microb Ecol 82:59–72.
57.
Kearney SM, Thomas E, Coe A, Chisholm SW. 2021. Microbial diversity of co-occurring heterotrophs in cultures of marine picocyanobacteria. Environ Microbiome 16:1.
58.
Pontiller B, Martínez-García S, Joglar V, Amnebrink D, Pérez-Martínez C, González JM, Lundin D, Fernández E, Teira E, Pinhassi J. 2022. Rapid bacterioplankton transcription cascades regulate organic matter utilization during phytoplankton bloom progression in a coastal upwelling system. ISME J 16:2360–2372.
59.
Amin SA, Parker MS, Armbrust EV. 2012. Interactions between diatoms and bacteria. Microbiol Mol Biol Rev 76:667–684.
60.
Eigemann F, Schulz-Vogt HN. 2019. Stable and labile associations of microorganisms with the cyanobacterium Nodularia spumigena. Aquat Microb Ecol 83:281–293.
61.
Eigemann F, Hilt S, Salka I, Grossart HP. 2013. Bacterial community composition associated with freshwater algae: species specificity vs. dependency on environmental conditions and source community. FEMS Microbiol Ecol 83:650–663.
62.
Mayerhofer MM, Eigemann F, Lackner C, Hoffmann J, Hellweger FL. 2021. Dynamic carbon flux network of a diverse marine microbial community. ISME Commun 1:50.
63.
Lombard V, Golaconda Ramulu H, Drula E, Coutinho PM, Henrissat B. 2014. The carbohydrate-active enzymes database (CAZy) in 2013. Nucleic Acids Res 42:D490–D495.
64.
Elovaara S, Eronen-Rasimus E, Asmala E, Tamelander T, Kaartokallio H. 2021. Contrasting patterns of carbon cycling and dissolved organic matter processing in two phytoplankton–bacteria communities. Biogeosciences 18:6589–6616.
65.
Luria CM, Amaral-Zettler LA, Ducklow HW, Repeta DJ, Rhyne AL, Rich JJ. 2017. Seasonal shifts in bacterial community responses to phytoplankton-derived dissolved organic matter in the western Antarctic Peninsula. Front Microbiol 8:2117.
66.
Attermeyer K, Hornick T, Kayler ZE, Bahr A, Zwirnmann E, Grossart HP, Premke K. 2014. Enhanced bacterial decomposition with increasing addition of autochthonous to allochthonous carbon without any effect on bacterial community composition. Biogeosciences 11:1479–1489.
67.
Baltar F, Palovaara J, Unrein F, Catala P, Horňák K, Šimek K, Vaqué D, Massana R, Gasol JM, Pinhassi J. 2016. Marine bacterial community structure resilience to changes in protist predation under phytoplankton bloom conditions. ISME J 10:568–581.
68.
Biller SJ, Berube PM, Lindell D, Chisholm SW. 2015. Prochlorococcus: the structure and function of collective diversity. Nat Rev Microbiol 13:13–27.
69.
Yool A, Tyrrell T. 2003. Role of diatoms in regulating the ocean's silicon cycle. Global Biogeochem Cycles 17.
70.
Grossowicz M, Roth-Rosenberg D, Aharonovich D, Silverman J, Follows MJ, Sher D. 2017. Prochlorococcus in the lab and in silico: the importance of representing exudation. Limnol Oceanogr 62:818–835.
71.
Guillard RRL. 1975. Culture of phytoplankton for feeding marine invertebrates, p 29–60. In Smith WL, Chanley MH (eds), Culture of marine invertebrate animals. Springer, Boston, MA.
72.
Hansen HP, Koroleff F. 1999. Determination of nutrients, p 159–228, Methods of seawater analysis. John Wiley & Sons, New York, NY.
73.
Simon M, Alldredge AL, Azam F. 1990. Bacterial carbon dynamics on marine snow. Mar Ecol Prog Ser 65:205–211.
74.
Thingstad TF, Mantoura RFC. 2005. Titrating excess nitrogen content of phosphorous-deficient eastern Mediterranean surface water using alkaline phosphatase activity as a bio-indicator. Limnol Oceanogr Methods 3:94–100.
75.
Walters W, Hyde ER, Berg-Lyons D, Ackermann G, Humphrey G, Parada A, Gilbert JA, Jansson JK, Caporaso JG, Fuhrman JA, Apprill A, Knight R. 2016. Improved bacterial 16S rRNA gene (V4 and V4-5) and fungal internal transcribed spacer marker gene primers for microbial community surveys. mSystems 1:e00009-15.
76.
Callahan BJ, McMurdie PJ, Rosen MJ, Han AW, Johnson AJA, Holmes SP. 2016. DADA2: High-resolution sample inference from Illumina amplicon data. Nat Methods 13:581–583.
77.
R Development Team. 2020. R: a language and environment for statistical computing. R Foundation for Statistical Computing, Vienna, Austria.
78.
R Studio Team. 2020. RStudio: integrated development for R. RStudio, PBC, Boston, MA.
79.
Quast C, Pruesse E, Yilmaz P, Gerken J, Schweer T, Yarza P, Peplies J, Glöckner FO. 2013. The SILVA ribosomal RNA gene database project: improved data processing and web-based tools. Nucleic Acids Res 41:D590–D596.
80.
Fox J, Weisberg S. 2019. An {R} companion to applied regression, 3rd ed. Sage, Thousand Oaks, CA.
81.
Pohlert T. 2014. The Pairwise Multiple Comparison of Mean Ranks Package (PMCMR). R package. https://CRAN.R-project.org/package=PMCMR.
82.
Ritchie ME, Phipson B, Wu D, Hu Y, Law CW, Shi W, Smyth GK. 2015. limma powers differential expression analyses for RNA-sequencing and microarray studies. Nucleic Acids Res 43:e47.
83.
Oksanen J, Kindt R, Legendre P, O’Hara B, Simpson GL, Solymos P, Stevens MHH, Wagner H. 2008. vegan: Community Ecology Package. http://vegan.r-forge.r-project.org/.
84.
Zhao S, Guo Y, Sheng Q, Shyr Y. 2014. Heatmap3: an improved heatmap package with more powerful and convenient features. BMC Bioinformatics 15:P16.
85.
Segata N, Izard J, Waldron L, Gevers D, Miropolsky L, Garrett WS, Huttenhower C. 2011. Metagenomic biomarker discovery and explanation. Genome Biol 12:R60.
86.
Ginestet C. 2011. ggplot2: elegant graphics for data analysis. J R Stat Soc Series A Stat Soc 174:245–246.

Information & Contributors

Information

Published In

cover image Applied and Environmental Microbiology
Applied and Environmental Microbiology
Volume 89Number 726 July 2023
eLocator: e00539-23
Editor: Jennifer B. Glass, Georgia Institute of Technology
PubMed: 37409944

History

Received: 31 March 2023
Accepted: 19 June 2023
Published online: 6 July 2023

Permissions

Request permissions for this article.

Keywords

  1. diatom
  2. Prochlorococcus
  3. phytoplankton DOM
  4. phytoplankton-bacterium interactions
  5. dissolved organic matter

Contributors

Authors

Water Quality Engineering, Technical University of Berlin, Berlin, Germany
Leibniz-Institute for Baltic Sea Research, Warnemuende, Germany
Eyal Rahav
Israel Oceanographic and Limnological Research, Haifa, Israel
Hans-Peter Grossart
Leibniz-Institute of Freshwater Ecology and Inland Fisheries, Berlin, Germany
Potsdam University, Potsdam, Germany
Dikla Aharonovich
Leon H. Charney School of Marine Sciences, University Haifa, Israel
Maren Voss
Leibniz-Institute for Baltic Sea Research, Warnemuende, Germany
Leon H. Charney School of Marine Sciences, University Haifa, Israel

Editor

Jennifer B. Glass
Editor
Georgia Institute of Technology

Notes

The authors declare no conflict of interest.

Metrics & Citations

Metrics

Note:

  • For recently published articles, the TOTAL download count will appear as zero until a new month starts.
  • There is a 3- to 4-day delay in article usage, so article usage will not appear immediately after publication.
  • Citation counts come from the Crossref Cited by service.

Citations

If you have the appropriate software installed, you can download article citation data to the citation manager of your choice. For an editable text file, please select Medlars format which will download as a .txt file. Simply select your manager software from the list below and click Download.

View Options

Figures

Tables

Media

Share

Share

Share the article link

Share with email

Email a colleague

Share on social media

American Society for Microbiology ("ASM") is committed to maintaining your confidence and trust with respect to the information we collect from you on websites owned and operated by ASM ("ASM Web Sites") and other sources. This Privacy Policy sets forth the information we collect about you, how we use this information and the choices you have about how we use such information.
FIND OUT MORE about the privacy policy