Vitamin B12 is not shared by all marine prototrophic bacteria with their environment
Vitamin B12 biosynthesis potential of different bacteriaB vitamins play a key role in complex marine microbial interactions as they are obligatory cofactors in various essential metabolic reactions in all living organism [13, 14, 39,40,41]. An exciting fact about B12 is that genes for synthesis of this complex cofactor have never made the transition to the eukaryotic kingdom, although it is required by both prokaryotes and eukaryotes. De novo synthesis is restricted to a minor fraction of bacteria and archaea, thus, suggesting that the ability to synthesise B12 is disproportionate to its demand in nature [1, 4]. This phenomenon can be observed in various habitats, for example in the soil microbiome, where the proportion of B12 producers is less than one tenth [8]. Similar findings have been shown for the microbiome on human skin, where only 1% of the core species are predicted to produce B12 de novo, while 39 % of the species are predicted to use B12 for metabolism [42]. In order to adequately answer this fundamental question regarding the balance between B12 availability and consumption, we should aim to better understand the synthesis potential of individual prototrophic prokaryotes.Here we present intra- and extracellular B12 concentrations of various B12 prototrophic, alphaproteobacterial strains. The concentration of intracellular B12 differs widely between the various heterotrophic bacteria examined. Converted, B12 molecules detected per cell ranged between 664 to 26,619 in the analysed bacterial cultures, including B12-provider and B12-retainer. Such strong variation in intracellular B12 concentrations have already been shown for a number of other prokaryotes, including Archaea, heterotrophic bacteria, and cyanobacteria [11, 34]. Also, in these studies, the detected intracellular B12 values differed up to three orders of magnitude and showed values similar to the ones we detected. Whether factors such as cell size, which we did not consider in our analysis, or the exact growth phase in which we took the samples had an influence on the strong variation cannot be clarified here. It is quite conceivable that different B12 requirements of the individual cells or different regulatory mechanisms of B12 synthesis played a decisive role for the intracellular B12 concentrations. Nevertheless, we can conclude that not only the genetic B12 biosynthetic potential within a microbial community is decisive, but rather which prokaryote is actually present is crucial for the availability of B12.The extracellular concentrations of B12 detected in M. algicola and P. inhibens were about 8 and 256 times lower than respective intracellular levels. For example, M. algicola secreted about 936 B12 molecules per cell, which was roughly 85 times more as detected for P. inhibens. On the basis of the detected B12 demand of T. pseudonana determined by the bioassay, we can calculate that the eukaryote requires roughly 135,000 B12 molecules per cell, if we base the limitation of cell number solely on B12 availability. Thus, it would take about 144 living M. algicola cells that release B12 to cover the requirements for the growth of one T. pseudonana cell. In fact, the bacterial cell numbers in the stationary phase of the B12-provider-diatom co-cultures were at least 110 times higher than the cell numbers of T. pseudonana. These calculations are all based on ideal laboratory conditions, with sufficient supply of inorganic nutrients and organic substrates and may differ in natural environments where viral infections or sloppy feeding can lead to cell disruption and subsequent release of intracellular B12 [43, 44]. Also, B12 requirement of T. pseudonana cells can vary under different growth conditions. For example, it has been shown that growth of T. pseudonana even with 1 pM of B12 can result in a significant change in the metabolite pool of the diatom, which in turn may have implications for the interaction with bacteria [45]. Nevertheless, our data give a first approximate insight into the interplay between B12-producers and -consumers in the world of microorganisms.Bacterial effects on the growth of T. pseudonana
Growth characteristics of T. pseudonana in co-culture show not only the obligatory provision of B12 by bacteria but also other bacterial factors that influence growth. For example, we observed that Sulfitobacter litoralis, a representative of the Roseobacter group, showed inhibitory behaviour towards the diatom. Other studies have shown that Roseobacter group isolates can produce inhibitory substances, roseobacticides, which can suppress the growth of eukaryotic phototrophs [46]. The provision of B12 leads to a promotion in growth and, at the same time, growth of the diatom is inhibited. One reason for the different growth characteristics of the diatoms observed in co-culture with different bacteria could be the adaptation to different habitats where the bacterial isolates naturally occur.In contrast to these observations, Celeribacter baekdonensis DSM 27375 significantly stimulated the growth of T. pseudonana. Even though C. baekdonensis did not provide B12 despite being synthesized, its presence in co-culture with B12 addition significantly increased the growth rate and growth yield of T. pseudonana compared to the positive control of the corresponding experimental run. In previous bacterial-diatom co-culture experiments, it has been shown that the excretion of cyclic peptides, diketopiperazines, by a bacterium, significantly increased diatom cell numbers [47]. Another plausible scenario is the synthesis and excretion of indoleacetic acid (IAA) by C. baekdonensis, which is a growth-promoting hormone for diatoms [48]. A similar effect is also conceivable for C. baekdonensis and would be exciting to explore in greater depth.A finding that appears to be overlooked in the context of our actual question is the fact that the expected bacterial cell death does not necessarily lead to the release of B12, which would promote the growth of T. pseudonana, and thus promote the interaction. Even after up to six weeks in co-culture, we cannot observe significant growth of T. pseudonana despite the presence of a bacterial B12 prototroph. This fact highlights the importance of cell lysis mechanisms in nature, for example caused by viral infections or sloppy feeding. Already today, these two natural processes are considered to play a significant role in the turnover of dissolved organic matter [44, 49,50,51] and are likely to also have a decisive influence on the release of B-vitamins in marine ecosystems [23]. Additionally, T. pseudonana is known to secret a B12 binding protein under B12 deficient conditions that has an affinity constant of 2 × 1011 M−1. This protein might help them to acquire B12 from the surroundings, when it is released through bacterial cell lysis mechanism [52]. Other phytoplankton might also have a similar strategy to scavenge B12 from the environment. When intracellular B12 is considered as a reservoir for other B12 auxotrophic microorganisms, then, for example, already 19 M. algicola cells would be sufficient to enable the growth of one T. pseudonana cell.The vital cofactor B12 is not shared by all prototrophic bacteriaAbout half of the marine phytoplankton species are B12 auxotrophs and rely on prototrophic prokaryotes to obtain this essential vitamin [1, 53]. Several co-culture experiments have confirmed that individual marine bacterial isolates, mainly Alphaproteobacteria, enable phytoplankton species to overcome their auxotrophy by providing the essential cofactor [13,14,15,16, 27, 28]. In our study we hypothesised that not all B12 prototrophs share B12 with other microorganisms and to prove that we performed individual co-culture experiments between T. pseudonana and 33 B12 prototrophic bacteria. B12 prototrophy of the bacterial isolates was confirmed by their genetic ability to synthesize B12 (Supplementary table S2) and their ability to grow in B12-free medium. The results of our study support this hypothesis, as we were able to identify one group of bacteria that enables growth of T. pseudonana by the supply of the essential cofactor, B12-providers. On the other hand, we also identified a second group of B12 prototrophic bacteria that did not support the growth of the diatom, the B12-retainers. Moreover, while categorizing them into B12-providers and B12-retainers, we observed that there are species within one genus, such as P. inhibens and P. galleciensis, in which one is a B12-provider and the other is a B12-retainer, respectively, although both of them possess the necessary genes for B12 biosynthesis. Yet, the question remains why some bacteria share the cofactor, and others, despite an obligatory interaction enforced in co-culture, do not. In the following, we describe and discuss three scenarios that we consider plausible, whereby not only one scenario has to be correct, but rather all three can take place in the B12-retainer strains that we have identified.First, biosynthesis of metabolites, such as the energetically costly B12 cofactor, are subject to intracellular regulation. Transcriptional regulation of the B12 biosynthesis pathway determines whether, and in what quantity B12 is synthesised in the cell. For example, sigma factors can alter the specificity of an RNA polymerase for a particular promoter, so that gene expression is enhanced or reduced [54]. In the case of the bacterial isolate Propionibacterium strain UF1, the riboswitch cbiMCbl was identified to regulate the gene expression of the cobA operon and thus controls B12 biosynthesis [55]. It is also known that sufficient availability of B12 can repress B12 biosynthesis gene expression in bacteria [56, 57]. In gram-negative proteobacteria as well as in cyanobacteria, for example, cobalamin (pseudocobalamin, in case of some bacteria) biosynthesis and B12 transport genes are regulated by inhibition of translation initiation, whereas in some gram-positive bacteria gene regulation proceeds by transcriptional antitermination [58]. The mechanisms described above are likely to also occur in the bacterial isolates that we tested. The large difference between the detected intracellular B12 concentrations could therefore be due to differences in gene regulation of the different bacteria and may also have had an influence on the release of B12 in the co-culture with T. pseudonana.Second, cobalamin, which we referred to here as B12 for simplicity, belongs to a group of B12-like metabolites, called cobamides. Each cobamide differs in the lower ligand attached. For example, the common cobamide, cobalamin, which is bioavailable to most microorganisms, carries 5,6-dimethylbenzimidazol (DMB) as its lower ligand, whereas pseudocobalamin synthesised by cyanobacteria in high concentrations in the ocean and being less or not bioavailable to most microorganisms, has adenine attached as its lower ligand [11, 41, 59, 60]. In general, the lower ligands of cobamides can be divided into benzimidazoles, purines, and phenols, and more than a dozen cobamides and cobamide-analogs have already been discovered [61]. However, research into the synthesis and actual diversity of cobamides, especially in marine bacteria and archaea, is still in its infancy. In our study, we were unable to detect intracellular B12 in four out of eight bacterial B12-retainer strains, although the cell counts at the time of sampling should have been sufficient for its detection. However, as is generally the case, our LC-MS analysis only targets cobalamin (B12) with its different upper ligands (adenosyl-, cyano-, methyl-, and hydroxy-cobalamin). Therefore, we cannot exclude the possibility that the here studied bacteria synthesise different cobamides, which are possibly not or less bioavailable to T. pseudonana, and not covered by our analytical measurement method. This speculation was supported by the fact that one of these four B12- retainer strains, Sulfitobacter sp. DFL-23, does not possess the DMB synthesis gene bluB and there was no intracellular B12 detected in this strain (Supplementary table S2 and Table 2). Again, it is difficult to explain this phenomenon solely depending on the presence of annotated DMB synthesis gene, as for Loktanella salsilacus DSM 16199 no bluB gene was annotated, still we detected intracellular B12 in this strain using our detection method (Supplementary table S2 and Table 2).Third, the bacteria we have identified as B12-retainer simply may not have actively released the synthesised B12 into their environment. Regardless of the importance of B12 for the vast majority of living organisms on our planet, its excretion mechanisms are to our knowledge still largely unknown. Its size of more than 1,350 Dalton does not allow sufficient diffusion through the cell membrane, which would enable microbial interactions [32]. Thus, it is likely that an unknown mechanism is required for its release. This assumption is further supported by the fact that we were able to detect intracellular B12 in four of the eight B12-retainer strains and at concentrations comparable to those detected in the B12-provider strains. In addition, we could detect intracellular B12 in P. xiamenensis, but none in its exometabolome. On the other hand, presence of extracellular B12 was detected in the exometabolome of both the provider strains examined, M. algicola and P. inhibens. Our findings show that not all bacteria share the pivotal cofactor with their environment, which has an impact on our current understanding of the marine B12 cycle and presumably in other ecosystems as well. The active exchange of B12 and thus microbial interaction plays a much smaller role than previously assumed for a relatively large number of bacteria. Consequently, for some of the B12 prototrophic bacteria within a community, it is likely that the cofactor is only released upon cell lysis.B12 production in the marine ecosystem and ecological implicationsLooking at the original source of B12 in nature, namely prototrophic bacteria and archaea, the bacteria studied here show pronounced differences between the biosynthetic potentials of the cofactors and the ability to share them with their environment. Thus, the natural source of vitamin B12 within a given ecosystem does not primarily depend on the ratio of prototrophic bacteria, but even more crucially on how much of the cofactor is synthesised by the prototrophic prokaryotes within an ecosystem and is actively released. The fact that some bacteria do not voluntarily share B12 with ambient microorganisms, significantly increases the importance of processes, such as sloppy feeding by zooplankton or virus infections [44, 49,50,51], for the release of vitamins in the marine and likely also other ecosystems.Our results also contribute to the controversially discussed question of whether B12 prototrophic bacteria live in symbiosis with phototrophic microorganisms [13, 30]. Despite numerous co-cultivation experiments demonstrating the obligatory provision of B12 by individual bacteria to phototrophic microorganisms, the decisive question of the mechanism of provision has so far been overlooked [13,14,15,16, 27, 28]. In our view, however, this question is crucial when assessing whether a symbiotic interaction is taking place. Our results support the hypothesis that a bacterial mechanism for the active release is likely to exist, as our experiments distinguish between B12-provider and B12-retainer within prototrophic bacteria. Looking at the ecological niches and the isolation sites of the two respective groups, differences can be identified. Most B12-provider strains were isolated from or discovered in association with eukaryotic microorganisms, whereas most B12-retainer strains were isolated as free-living in the ocean (Supplementary table S4). Moreover, six of the tested bacterial strains were isolated from dinoflagellates and five of them were B12-provider. Since we used a diatom as a B12 auxotrophic organism in our study, it would also be interesting to know if these B12-provider strains also provide B12 to other phytoplankton, such as dinoflagellates. Also, in this study we only studied bacteria from the alphaproteobacteria class, since a large share of them are known to be B12 prototrophs and abundant in the marine ecosystem. For future studies, it would be interesting to see if a similar pattern of B12 provisioning can be observed in bacteria from other classes. Our results indicate that the B12 prototrophy of a bacterium does not necessarily indicate a mutualistic interaction with other auxotrophic microorganisms. However, the bacterial group of B12-provider in particular seems to favour living in close proximity to other microorganisms, which is why the exchange of B12 for e.g. organic compounds can establish itself as a distinct symbiotic interaction between individual microorganisms. More
