Competition for Different Substrates
The shortness of the following section does not reflect the importance we assign to the topic. Our current understanding of the relationship between the growth of individual bacterial populations and the availability of particular substrates or nutrients is still rudimentary. For example, it is established that pelagic bacteria and archaea may incorporate amino acids (
201), but with a few exceptions (see below) it is unclear if different genotypes show preferences for individual components of such mixes. Studies on the interactions between bacteria and specific algae indirectly suggest that substrate quality may play a role in the competition between microbial populations. Pure cultures of various algal species may harbor distinct microbial assemblages (
252). Different microbial communities could be established from identical inocula in seawater micrososms if blooms of specific algal groups were induced (
225). Cottrell and Kirchman reported (
47) that various cooccurring microbial populations in pelagic habitats may be specialized on the degradation and uptake of particular substrate classes. Using microautoradiography, these authors showed that different bacterial genotypes also differed in their uptake of
3H from radiolabeled chitin, proteins, amino acids, and
N-acetylglucosamine. A preference for simple monomers over proteins in bacteria from the SAR11 clade has recently been reported by the same laboratory (
168).
A set of recent investigations on the transfer and processing of phytoplankton-derived organic sulfur compounds has provided a first model for future studies of the relationship between particular bacterial populations and a specific biogeochemical process. Dissolved dimethylsulfonopropionate (dDMSP) is an algal osmolyte that is released during viral lysis or sloppy zooplankton feeding (50, 167). dDMSP is a growth substrate for heterotrophic bacteria (153), and it represents an important precursor of protein-sulfur in marine bacterioplankton (139). An increasing body of evidence is pointing to a few groups of marine bacteria as key elements in dDMSP biogeochemistry. Laboratory studies on isolates of the diverse Roseobacter clade have indicated a widespread ability of members of this lineage to degrade dDMSP and to incorporate the sulfur derived from this organic compound into cellular protein (93). In addition, large populations of bacteria affiliated with the Roseobacter clade were observed during dDMSP-producing algal blooms in the North Atlantic. In two independent studies the horizontal or vertical distribution of microbes from this group was positively correlated to dDMSP concentrations or fluxes (96, 323). Direct evidence of DMSP uptake was recently obtained by microautoradiography (170, 306). This approach also revealed that DMSP uptake may be a feature of different bacterial populations, and bacteria from the SAR11 clade rather than Roseobacter mediated the bulk of DMSP turnover in an offshore planktonic assemblage (169).
Patchiness and Gradients
Selective preference for particular substrates may not be the only factor affecting the competition between aquatic microbes. Specialization on a single resource might in fact be disadvantageous in an energy-deficient environment (
60), and the majority of bacteria and archaea in pelagic marine environments are capable of incorporating mixes of radiolabeled amino acids (
137).
A considerable proportion of the substrates and of bacterial productivity in freshwater and coastal habitats are distributed in organic particles and microscale patches (12). In contrast, the bathypelagic zone of the open oceans represents a rather desert-like environment poor in particulate organic matter. Individual microbial species or phylogenetic lineages within the bacterioplankton thus likely differ in their ability to succeed in habitats with steeper or flatter substrate gradients. An "opportunistic" growth strategy might be widespread among those bacterial groups that successfully colonize organic aggregates or other nutrient-rich microniches. A marine Pseudoalteromonas sp. strain exhibited a significantly shorter growth delay than an Oceanospirillum sp. under "feast-and-famine" batch culture conditions, but the former strain experienced a growth disadvantage if substrate changes were gradual (212). The importance of such bacteria may vary between habitats, and different species of "opportunistically" growing microbes are found in marine and freshwater environments (224).
Representatives of "opportunistic" genera of Gammaproteobacteria that are readily culturable on solid media were generally rare in the bacterioplankton of shallow North Sea waters (62). In contrast, colony-forming bacteria represented a prominent fraction of the microbial assemblages in brackish waters of the coastal Baltic Sea (227, 228). Readily enrichable bacteria were also found in high concentrations in the plankton of a lake with high irregular input of organic carbon from the catchment (35). The identification and quantification of such "r-strategists" within heterotrophic picoplankton assemblages might thus eventually allow deductions about short-term fluctuations in the availability of organic matter in aquatic habitats, e.g., for a biological monitoring of pollution effects. In contrast, many if not most of the typical free-living marine and freshwater bacteria appear to lack the ability to form colonies on solid media (282), and their growth is negatively affected by enhanced substrate levels (289).
It is likely that many bacterial species in the water column combine elements of the oligotrophic and the "opportunistic" growth strategies in their life cycles, e.g., by alternating between free-living and attached growth forms (141). Isolates from freshwater plankton that were initially oligocarbophilic could gradually be adapted to richer conditions (114). Bacteria from the marine NOR5 lineage, which are apparently free-living in the water column, also formed colonies on low-nutrient agar plates, albeit significantly later than other "opportunistic" strains (63). A dual-niche existence may also explain why facultatively anaerobic bacteria appear to be such a common component of coastal bacterioplankton assemblages (4, 243). In shallow waters of the North Sea >80% of free-living Roseobacter spp. cells were able to incorporate glucose at both oxic and anoxic conditions (4). Since the average depth of the German Bight is only 20 m, bacteria from the Roseobacter spp. lineage that colonize aggregated senescent algae (97) might experience temporary anoxia while settled on the sediment surface. These bacteria could be reintroduced into the water column by the periodic resuspension of particulate organic matter (179).
Mortality
So far there is little evidence that a lack of resources is a major cause of bacterioplankton mortality. Cultured strains that are thought to be representative for the pelagic environment are often oligocarbophilic (
238,
259), or they survive extended periods of starvation (
68). Predation and viral lysis are believed to be the key factors that counterbalance microbial growth in the water column (
13,
77,
271). Growth inhibition or cell damage induced by UV radiation may be another important ecological factor in some aquatic habitats (
10,
119,
285).
Nevskia ramosa, a species inhabiting the neuston layer of fresh waters, exhibited elevated resistance to UV (
290). Some alpine lakes exposed to intense levels of UV-B radiation (
148) feature conspicuously high abundances of actinobacteria, a phylogenetic lineage of mainly gram-positive bacteria with a high genomic G+C content (
91,
263,
310). Gram-positive bacterial isolates are often less affected by UV radiation than gram-negative bacteria (
8), and a high genomic G+C content has been suggested to mediate higher resistance to radiation damage (
166).
Viral lysis. The current concepts of specific viral-bacterial interaction in aquatic systems are largely shaped by theoretical models (298), and presently there are almost no investigations about the influence of viral lysis on the coexistence of individual microbial populations in the plankton (260, 317). Viral influence on the growth of different bacteria might act both directly by "killing the winner" (298), but also indirectly via the release of dissolved organic matter and nutrients from lysed pro- and eukaryotic cells (92, 182). Rapid development of resistance to viral infection has been observed in some bacterial strains during continuous-culture experiments (181), but it is unclear if such a process will also occur in natural aquatic assemblages. Recently, changes of the species richness of marine archaea have been reported as a likely consequence of experimentally manipulating viral densities (317). For a detailed discussion of the influence of viruses on aquatic microbial assemblages, readers should refer to specific reviews on the subject (311, 319).
Selective predation. Hetero- and mixotrophic protists, in particular nanoflagellates and ciliates, are the main consumers of picoplankton in the marine and freshwater pelagic zone. Their role in controlling the abundance, biomass, and productivity of microbial assemblages has been amply documented (269, 298). Many of these predators are omnivorous, i.e., they can feed on a large range of bacterial species. However, they are not unselective feeders. In the following we will focus on predation selectivity that is related to prey cell morphology, in order to illustrate the influence of a phenotypic feature on the success of different genotypic microbial populations. Other aspects of microbial predator-prey interactions are discussed elsewhere (25, 82, 131, 215, 277).
Due to their specific particle uptake mechanisms and handling procedures, protistan predators cannot feed on all bacterial shapes or cell size classes with equal efficiency (26, 178). As a consequence, microbial cells within a length range of 1 to 3 µm are preferably ingested by heterotrophic flagellates and ciliates, whereas smaller or larger cells profit from reduced loss rates (130, 186, 276). Even small differences in cell sizes between strains of freshwater ultramicrobacteria may have large effects on their survival rates (27). Microbial assemblages that are exposed to high grazing pressure thus typically harbor high abundances of extremely small cells (222). In contrast, large filamentous bacterial morphotypes and bacteria that form microcolonies are substantially more protected from protistan grazing (115, 273). Filamentous morphotypes accumulate during periods of high predation in fresh waters (134), but they are rarely observed in marine systems. Some strains of freshwater Betaproteobacteria produce microcolonies within a sponge-like matrix of extrapolymerous substances that renders them resistant to flagellate predation (113).
Size-selective predation induces shifts in the genotypic composition of mixed assemblages by imposing different mortality rates on bacterial species with different mean cell sizes (Fig. 8) (221). First evidence for such community changes originated from continuous-cultivation experiments on microbial assemblages that formed stable associations with freshwater algae (221, 234). Selective grazing mortality is moreover believed to set an upper limit to the standing stocks of some microbial populations in the plankton. For example, bacteria from a cosmopolitan lineage of freshwater Betaproteobacteria (beta I) constituted approximately 10% of the summer assemblage in a eutrophic drinking-water reservoir (278). After removal of protists and in situ incubations in dialysis bags, members of the beta I clade increased to almost 30% of total cells within 24 h. It is thus likely that these bacteria contributed disproportionally to the flux of organic carbon from the picoplankton to the higher trophic levels. Some genera of "opportunistically" growing Gammaproteobacteria (Alteromonas, Vibrio, and Pseudoalteromonas) are probably rare in coastal surface picoplankton because they are almost completely suppressed by size-selective grazing (16). This mechanism might also play a role in eliminating pathogenic Vibrio spp. from the water column (165).
On the other hand, grazing-resistant genotypes rapidly accumulate in assemblages exposed to high protistan predation. In nonaxenic continuous cultures of the phytoflagellate Cryptomonas sp., filamentous Betaproteobacteria were only observed if protistan predators were added (279). A filament-forming Comamonas acidovorans strain outcompeted a Vibrio sp. strain after the addition of bacterivorous flagellates to chemostat cocultures (112). Grazing-resistant bacterial morphotypes from different phylogenetic lineages increased rapidly after food web manipulation in water from fishless ponds (132, 147). A substantial enrichment of filamentous Flectobacillus sp. was observed during artificially induced blooms of nanoflagellates in samples from a eutrophic freshwater reservoir (278). Threadlike bacteria of >99% 16S rRNA sequence similarity formed >40% of total bacterial biomass in a mesotrophic lake during high protistan grazing (223). Such a natural enrichment of grazing-resistant genotypes in fresh waters appears to be an ephemeral phenomenon, as these bacteria are typically sensitive to predation by larger, filter-feeding metazooplankton (129).
The interplay between resource availability and mortality in determining the population sizes of different bacterial taxa is poorly understood. Changes in bacterial community composition might be related to changes in the ratio of bacterial mortality rate to growth rate (277). This would imply that communities with apparently stable taxonomic composition may be encountered at very different levels of total microbial productivity or mortality and that profound community changes are induced by shifts from top down to bottom up or vice versa (235). One type of natural model system to study such interactions might be freshwater reservoirs that feature pronounced longitudinal substrate and nutrient gradients between the river influx and the dam areas. Experiments that combine the transplant of water from different positions of the gradient with food web manipulations (80) might eventually shed more light on the complex control of different microbial populations by top-down and bottom-up forces.
Answers to all your Biology Questions
