SummaryBiological nitrogen fixation is an important source of fixed nitrogen for the biosphere. Microorganisms catalyse biological nitrogen fixation with the enzyme nitrogenase, which has been highly conserved through evolution. Cloning and sequencing of one of the nitrogenase structural genes, nifH , has provided a large, rapidly expanding database of sequences from diverse terrestrial and aquatic environments. Comparison of nifH phylogenies to ribosomal RNA phylogenies from cultivated microorganisms shows little conclusive evidence of lateral gene transfer. Sequence diversity far outstrips representation by cultivated representatives. The phylogeny of nitrogenase includes branches that represent phylotypic groupings based on ribosomal RNA phylogeny, but also includes paralogous clades including the alternative, non-molybdenum, non-vanadium containing nitrogenases. Only a few alternative or archaeal nitrogenase sequences have as yet been obtained from the environment. Extensive analysis of the distribution of nifH phylotypes among habitats indicates that there are characteristic patterns of nitrogen fixing microorganisms in termite guts, sediment and soil environments, estuaries and salt marshes, and oligotrophic oceans. The distribution of nitrogen-fixing microorganisms, although not entirely dictated by the nitrogen availability in the environment, is nonrandom and can be predicted on the basis of habitat characteristics. The ability to assay for gene expression and investigate genome arrangements provides the promise of new tools for interrogating natural populations of diazotrophs. The broad analysis of nitrogenase genes provides a basis for developing molecular assays and bioinformatics approaches for the study of nitrogen fixation in the environment.

Because viruses of eukaryotic algae are incredibly diverse, sweeping generalizations about their ecology are rare. These obligate parasites infect a range of algae and their diversity can be illustrated by considering that isolates range from small particles with ssRNA genomes to much larger particles with 560 kb dsDNA genomes. Molecular research has also provided clues about the extent of their diversity especially considering that genetic signatures of algal viruses in the environment rarely match cultivated viruses. One general concept in algal virus ecology that has emerged is that algal viruses are very host specific and most infect only certain strains of their hosts; with the exception of viruses of brown algae, evidence for interspecies infectivity is lacking. Although some host-virus systems behave with boom-bust oscillations, complex patterns of intraspecies infectivity can lead to host-virus coexistence obfuscating the role of viruses in host population dynamics. Within the framework of population dynamics, host density dependence is an important phenomenon that influences virus abundances in nature. Variable burst sizes of different viruses also influence their abundances and permit speculations about different life strategies, but as exceptions are common in algal virus ecology, life strategy generalizations may not be broadly applicable. Gaps in knowledge of virus seasonality and persistence are beginning to close and investigations of environmental reservoirs and virus resilience may answer questions about virus inter-annual recurrences. Studies of algal mortality have shown that viruses are often important agents of mortality reinforcing notions about their ecological relevance, while observations of the surprising ways viruses interact with their hosts highlight the immaturity of our understanding. Considering that just two decades ago algal viruses were hardly acknowledged, recent progress affords the optimistic perspective that future studies will provide keys to unlocking our understanding of algal virus ecology specifically, and aquatic ecosystems generally.

Algal-virus-specific PCR primers were used to amplify DNA polymerase (pol) gene fragments from geographically isolated natural virus communities. Natural algal virus communities were obtained from coastal sites in the Pacific Ocean in British Columbia, Canada, and the Southern Ocean near the Antarctic peninsula. Genetic fingerprints of algal virus communities were generated using denaturing gradient gel electrophoresis (DGGE). Sequencing efforts recovered 33 sequences from the gradient gel. Of the 33 sequences examined, 25 encoded a conserved amino acid motif indicating that the sequences were pol gene fragments. Furthermore, the 25 pol sequences were related to pol gene fragments from known algal viruses. In addition, similar virus sequences (>98% sequence identity) were recovered from British Columbia and Antarctica. Results from this study demonstrate that DGGE with degenerate primers can be used to qualitatively fingerprint and assess genetic diversity in specific subsets of natural virus communities and that closely related viruses occur in distant geographic locations. DGGE is a powerful tool for genetically fingerprinting natural virus communities and may be used to examine how specific components of virus communities respond to experimental manipulations.

To examine algal virus (Phycodnaviridae) genetic diversity in freshwater environments, gene fragments were cloned and sequenced from a river and a reservoir in Colorado, USA, and 2 different lakes in Ontario, Canada using PCR methods that target a diverse subset of known Phycodnaviridae DNA polymerase genes. Numerous phycodnavirus gene sequences were obtained from every sample, and rarefaction analysis of the sequence libraries demonstrated that virus richness was variable among different sample locations, and among samples collected from the same location at different times. Phylogenetic analysis of the unique sequences from each sample indicated that most sequences from the same geographic region (i.e. Colorado or Ontario) clustered together, but several exceptions were also observed. Phylogenetic analysis also demonstrated that the sequences obtained were more closely related to sequences from cultivated marine phycodnaviruses belonging to the genus Prasinovirus than to those from cultivated freshwater phycodnaviruses from the genus Chlorovirus. Overall, phycodnavirus sequences originating from cultivated marine viruses and marine clone libraries were not genetically distinct from the freshwater phycodnavirus sequences reported in this study.

Viruses infecting the harmful bloom-causing alga Phaeocystis globosa (Prymnesiophyceae) were readily isolated from Dutch coastal waters (southern North Sea) in 2000 and 2001. Our data show a large increase in the abundance of putative P. globosa viruses during blooms of P. globosa, suggesting that viruses are an important source of mortality for this alga. In order to examine genetic relatedness among viruses infecting P. globosa and other phytoplankton, DNA polymerase gene (pol) fragments were amplified and the inferred amino acid sequences were phylogenetically analyzed. The results demonstrated that viruses infecting P. globosa formed a closely related monophyletic group within the family Phycodnaviridae, with at least 96.9% similarity to each other. The sequences grouped most closely with others from viruses that infect the prymnesiophyte algae Chrysochromulina brevifilum and Chrysochromulina strobilus. Whether the P. globosa viruses belong to the genus Prymnesiovirus or form a separate group needs further study. Our data suggest that, like their phytoplankton hosts, the Chrysochromulina and Phaeocystis viruses share a common ancestor and that these prymnesioviruses and their algal host have coevolved.Phaeocystis globosa is a phytoplankton species with unicellular and colonial stages that can form dense blooms in temperate coastal waters of the North Sea (up to 10 8 liter Ϫ1 [7,10,11]) The ability to generate high biomass, dominate the phytoplankton community for extended periods, and produce dimethylsulfoniopropionate and dimethyl sulfide makes this alga a model species for biogeochemical studies (26,30). Because its diverse roles in element and trophic dynamics are amplified during blooms, the wax and wane of P. globosa affects the function and structure of pelagic food webs (7). Its influence on pelagic food webs may depend on the factors responsible for bloom termination. Earlier studies have shown that the collapse of P. globosa blooms can be sudden and that cell lysis was a major source of mortality (7, 9). Through cell lysis, high concentrations of organic matter are released into the water column and remineralized by heterotrophs, thereby stimulating the microbial food web and affecting nutrient fluxes (4,18,19,21,55).Viruses are known lytic agents of phytoplankton (6, 46). With increased awareness of the potential regulation of phytoplankton dynamics by viruses, efforts have been made to establish virus-phytoplankton model systems in culture. Although virus-like particles have been found in most classes of eukaryotic unicellular photoautotrophs (38, 52), at present, there are only approximately a dozen phytoplankton viruses in culture. Most of these infect bloom-forming species and about half infect harmful algal bloom species (20,25,35,42,47,48). Although viruses have been implicated in the demise of phytoplankton blooms (3, 4, 12, 34), there is no evidence that viruses are responsible for the termination of P. globosa blooms.The present work focuses on the isolation of viruses infecting P. globosa...