Intergenomic comparisons highlight modularity of the denitrification pathway and underpin the importance of community structure for N2O emissions

Abstract

Nitrous oxide (N2O) is a potent greenhouse gas and the predominant ozone depleting substance. The only enzyme known to reduce N2O is the nitrous oxide reductase, encoded by the nosZ gene, which is present among bacteria and archaea capable of either complete denitrification or only N2O reduction to di-nitrogen gas. To determine whether the occurrence of nosZ, being a proxy for the trait N2O reduction, differed among taxonomic groups, preferred habitats or organisms having either NirK or NirS nitrite reductases encoded by the nirK and nirS genes, respectively, 652 microbial genomes across 18 phyla were compared. Furthermore, the association of different co-occurrence patterns with enzymes reducing nitric oxide to N2O encoded by nor genes was examined. We observed that co-occurrence patterns of denitrification genes were not randomly distributed across taxa, as specific patterns were found to be more dominant or absent than expected within different taxonomic groups. The nosZ gene had a significantly higher frequency of co-occurrence with nirS than with nirK and the presence or absence of a nor gene largely explained this pattern, as nirS almost always co-occurred with nor. This suggests that nirS type denitrifiers are more likely to be capable of complete denitrification and thus contribute less to N2O emissions than nirK type denitrifiers under favorable environmental conditions. Comparative phylogenetic analysis indicated a greater degree of shared evolutionary history between nosZ and nirS. However 30% of the organisms with nosZ did not possess either nir gene, with several of these also lacking nor, suggesting a potentially important role in N2O reduction. Co-occurrence patterns were also non-randomly distributed amongst preferred habitat categories, with several habitats showing significant differences in the frequencies of nirS and nirK type denitrifiers. These results demonstrate that the denitrification pathway is highly modular, thus underpinning the importance of community structure for N2O emissions.

Figure 1. Maximum likelihood phylogeny of full-length 16S/18S rRNA sequences from 652 organisms with denitrification genes.

The inner colored ring represents taxonomic affiliation as indicated by the legend. The four outer bar-chart rings show the presence of nirK (orange), nirS (purple), nor (turquoise) and nosZ (magenta). Bar height represents the number of copies (≤4). Bootstrap values >70% are indicated by grey circles, and the scale bar denotes nucleotide substitution rate (GTR+Γ). Classification is based on the SILVA database with denomination according to NCBI taxonomy. For NCBI taxon ID number and project name, see Table S1.

Figure 2. Co-occurrence of nosZ with nirK or nirS in genomes with and without nor genes.

Percentage of genomes with only nosZ (black), nosZ and nirK (dark grey) and nosZ and nirS (light grey) among organisms a) harboring nosZ and nor, b) within nosZ Clade I with nor, c) within nosZ Clade II with nor, d) harboring nosZ without nor, e) within nosZ Clade I without nor, and f) within nosZ Clade II without nor. Six genomes that have both nirS and nirK in addition to nosZ are excluded as well as eight halophilic Archaea that group outside Clade I and II in b) and c).

Figure 3. Gene co-occurrence types and habitat preference.

a) Mosaic plot of nirK (K), nirS (S) and nosZ (Z) co-occurrence types across different habitat categories where tile size reflects the number of occurrences and patterns indicate significant overrepresentation (>2) or underrepresentation (<−2) of co-occurrence patterns, as determined by standardized Pearson residuals from χ² test results (P<0.001). Circles indicate non-occurring combinations. b) The percentage of organisms harboring a nor gene for each combination across the habitat categories. No value indicates non-occurring combinations.