Skip to main page content
Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
Review
. 2012 Sep;76(3):667-84.
doi: 10.1128/MMBR.00007-12.

Interactions Between Diatoms and Bacteria

Affiliations
Free PMC article
Review

Interactions Between Diatoms and Bacteria

Shady A Amin et al. Microbiol Mol Biol Rev. .
Free PMC article

Abstract

Diatoms and bacteria have cooccurred in common habitats for hundreds of millions of years, thus fostering specific associations and interactions with global biogeochemical consequences. Diatoms are responsible for one-fifth of the photosynthesis on Earth, while bacteria remineralize a large portion of this fixed carbon in the oceans. Through their coexistence, diatoms and bacteria cycle nutrients between oxidized and reduced states, impacting bioavailability and ultimately feeding higher trophic levels. Here we present an overview of how diatoms and bacteria interact and the implications of these interactions. We emphasize that heterotrophic bacteria in the oceans that are consistently associated with diatoms are confined to two phyla. These consistent bacterial associations result from encounter mechanisms that occur within a microscale environment surrounding a diatom cell. We review signaling mechanisms that occur in this microenvironment to pave the way for specific interactions. Finally, we discuss known interactions between diatoms and bacteria and exciting new directions and research opportunities in this field. Throughout the review, we emphasize new technological advances that will help in the discovery of new interactions. Deciphering the languages of diatoms and bacteria and how they interact will inform our understanding of the role these organisms have in shaping the ocean and how these interactions may change in future oceans.

Figures

Fig 1
Fig 1
Micrographs of representative diatom species. (A) Light micrographs of diatoms. Clockwise from the top left corner: Striatella unipunctata, Odontella sp., Stephanopyxis turris, Pseudo-nitzschia sp., Thalassiosira sp., Cylindrotheca sp., Asterionellopsis glacialis, Skeletonema costatum, Grammatophora oceanica, and Chaetoceros sp. Images are courtesy of Colleen Durkin. (B) Scanning electron microscopy (SEM) images of diatoms. Clockwise from the top left corner: Didymosphenia geminate (Lyngbye), valve view; Lauderia annulata with bacteria, girdle view showing attachment of two cells; Thalassionema nitzschioides (Grunow), valve view showing the rounded valve end; Coscinodiscus sp. after sexual reproduction in a culture restored large cells (top) from small gametangial cells (bottom); Chaetoceros didymus with bacteria; Asterionellopsis glacialis with bacteria; Actinoptychus senarius; Lithodesmium undulatum, girdle view; and Stephanopyxis turris (Greville) (center). Images are courtesy of Julie Koester (Coscinodiscus sp.) and Mark Webber (the rest).
Fig 2
Fig 2
Maximum-likelihood phylogenetic tree of the bacterial domain, highlighting heterotrophic taxa most commonly found associated with diatoms. Also shown are the autotrophic nitrogen-fixing bacteria (Cyanobacteria) known to be associated with diatoms. Bacterial phyla are color coded and labeled in the corresponding colored ring. Taxa reported to be associated with diatoms in culture or field samples are labeled in the outer ring. Boldface genera were reported in two or more independent studies. The tree is based on a concatenated alignment of 31 conserved predicted proteins from 350 bacterial species with whole genome sequences (188). Asterisks indicate taxonomic positions that are estimated from nearest 16S neighbor on the tree because they were not included in the original alignment. Rhodovulum nearest 16S neighbor, Rhodobacter (164); Ruegeria nearest 16S neighbor, Silicibacter (190); Stappia nearest 16S neighbor, Xanthobacter (105); Limnobacter nearest 16S neighbors, Burkholderia and Cupriavidus (111); Neptunomonas nearest 16S neighbor, Marinomonas (192); Halomonas nearest 16S neighbor, Chromohalobacter (6); Alteromonas and Glaciecola nearest 16S neighbor, Pseudoalteromonas (85); Sulfitobacter and Staleya nearest 16S neighbor, Roseobacter (106); Croceibacter, Aequorivita, and Lacinutrix nearest 16S neighbor, Flavobacterium (10); Reichenbachia and Gelidibacter nearest 16S neighbor, Cytophaga (112, 175); Winogradskyella nearest 16S neighbor, Gramella (108); Maribacter nearest 16S neighbor, Bacteroides (10); Richelia and Calothrix nearest 16S neighbor, Nostoc (97a). Abbreviations: β-proteo, Betaproteobacteria; δ-proteo, Deltaproteobacteria; Ac, Acidobacteria; ε-proteo, Epsilonproteobacteria; A, Aquificae; Bacteroid, Bacteroidetes; C, Chlorobi; C/V, Chlamydiae and Verrucomicrobia; P, Planctomycetes; S, Spirochaetes; Cyano, Cyanobacteria; Ch, Chloroflexi; F, Fusobacteria; Sy, Synergistetes; T, Thermotogae; and D, Deinococcus-Thermus. (Tree modified from reference by permission of Macmillan Publishers Ltd.)
Fig 3
Fig 3
Diffusive boundary layer around a spherical diatom cell (black circle) under different flow scenarios. (A and B) Stationary cell (pure diffusion) with a 4-μm diameter (A) or with a 20-μm diameter (B); (C) sinking cell in the absence of flow; (D) cell in shear flow. Panels A and B demonstrate the effect of cell size on the concentration gradients of a compound excreted from the cell surface and consequently on the size of the diffusive boundary layer. Panels C and D show the effect of sinking or swimming and shear flow on the shape of the diffusive boundary layer of a spherical cell. Panel D shows compression of the boundary layer in the y and z directions (with z being into the page) and stretch in the x direction in response to a uniaxial flow. Axes represent the distance away from the cell center either in μm (A and B) or in cell radii (C and D), and contours represent concentration ranges in μM (A and B) or in percentage of surface concentration (C and D). Panels A and B were calculated using equations 1 and 2 for cell surface and bulk (seawater) concentrations of 10 μM and 0.01 μM, respectively. Panels C and D are based on data from reference .
Fig 4
Fig 4
Structures of hydrophobic signaling molecules produced by diatoms and bacteria. (A) Examples of quorum-sensing autoinducers produced by bacteria shown to have a role in interkingdom signaling: 3-oxo-C12-HSL and C4-HSL (Pseudomonas aeruginosa), 3-oxo-C12-HSL (Vibrio fischeri), 3-oxo-C8-HSL (Agrobacterium tumefaciens), and 3-oxo-C16:1-HSL (Sinorhizobium meliloti). (B) Pheromones produced by diatoms.
Fig 5
Fig 5
Depictions of microbial interactions. Examples of bacterial interactions with diatoms are shown, categorized as competitive (purple), synergistic (orange), and parasitic (blue). An example of horizontal gene transfer (HGT) as evidence for past associations between diatoms and bacteria is also shown (green). Small rectangles with flagella represent bacteria; the large dark green rectangle (center) represents a diatom cell. Key reactions involved in the interactions are depicted within respective cells. The background gradient and dashed line represent the phycosphere. In the nutrient-poor regions of the oceans, bacteria and diatoms often encounter limiting resources for which competition may be established. Bacteria and diatoms both produce membrane-bound enzymes, such as alkaline phosphatase and 5′-nucleotidases (triangles), to break down dissolved organic phosphorus (DOP). Unless coupled directly to an uptake mechanism, consequent release of orthophosphate into the surrounding seawater may result in competitive uptake by both organisms, with bacteria generally outcompeting diatoms at low concentrations (173). One way to avoid such competition is to establish a synergistic mutualism, in which an exchange of resources increases the fitness of both organisms. For example, some diatoms lack a B12-independent methionine synthase gene (metE) and thus require an exogenous source of vitamin B12 to synthesize methionine. Many bacteria can produce B12, thus providing a source for diatoms (70). In return, the bacteria would benefit from dissolved organic matter (DOM) produced by the diatoms. Nitrogen-fixing cyanobacteria have been shown to supply fixed nitrogen to diatoms, likely in the form of ammonia or dissolved organic nitrogen (57). Some bacteria produce the highly photolabile siderophore vibrioferrin, which has been shown recently to supply soluble iron to algal cells, presumably in exchange for DOM (4). In this case, bacterial growth is significantly enhanced in the presence of the algae, providing further evidence for a mutualistic interaction. A subset of bacteria possesses the ability to utilize a by-product of algal photorespiration, glycolate, with the potential consequence of shifting bacterial community dynamics over diel cycles (101). In the category of parasitic interactions, we include interactions that directly parasitize or defend against such parasitism. Desbois et al. have shown that the fatty acid eicosapentaenoic acid (EPA) produced by the diatom P. tricornutum is an effective antimicrobial agent against known human pathogens (45). The compound appears to be less effective against marine bacteria. Algicidal bacteria can attack diatoms through attachment and/or release of enzymes that degrade the diatom's organic matrix, which protects the silica frustule from dissolution. Furusawa et al. demonstrated the attachment of a Saprospira sp. to a diatom and subsequent release of fibril-like materials that appear to aid in decomposition of the cell wall (60). The algicidal bacterium Kordia algicida appears to produce a diffusible protease responsible for the killing activity (137). In a similar system, Lee et al. characterized the protease from a Pseudoalteromonas sp. as a serine protease (107). Horizontal gene transfer (HGT) represents the footprint of ancient associations between bacteria and diatoms. Horizontal transfer of genetic material, hypothesized to originate from bacteria in close associations with diatoms, has shaped the metabolic potential of modern diatom species. In one example, a key offshoot pathway of the urea cycle in diatoms, catalyzed by carbamate kinase (CBK), appears to be encoded by a gene with no phylogenetic associations with other known eukaryotes (2).
None
None
None

Similar articles

See all similar articles

Cited by 147 articles

See all "Cited by" articles

Publication types

LinkOut - more resources

Feedback