Dominance of Candidatus Scalindua species in anammox community revealed in soils with different duration of rice paddy cultivation in Northeast China

Abstract

The anaerobic ammonium-oxidizing (anammox) bacteria play an important role in the oxygen-limited zone for nitrogen cycling, but their roles in agricultural ecosystems are still poorly understood. In this study, soil samples were taken from the rhizosphere and non-rhizosphere and from surface (0-5 cm) and subsurface (20-25 cm) layers with 1, 4, and 9 years of rice cultivation history on the typical albic soil of Northeast China to examine the diversity and distribution of anammox bacteria based on 16S rRNA gene and hydrazine oxidoreductase encoding gene (hzo). By comparing these soil samples, no obvious difference was observed in community composition between the rhizosphere and non-rhizosphere or the surface and subsurface layers. Surprisingly, anammox bacterial communities of these rice paddy soils were consisted of mainly Candidatus Scalindua species, which are best known to be dominant in marine and pristine environments. The highest diversity was revealed in the 4-year paddy soil based on clone library analysis. Phylogenetic analysis of 16S rRNA gene and deduced HZO from the corresponding encoding gene showed that most of the obtained clones are grouped together with Candidatus Scalindua sorokinii, Candidatus Scalindua brodae, and Candidatus Scalindua spp. of seawater. The obtained clone sequences from all samples are distributed in two subclusters that contain sequences from environmental samples only. Tentative new species were also discovered in this paddy soil. This study provides the first evidence on the existence of anammox bacteria with limited diversity in agricultural ecosystems in Northern China.

Fig. 1

Neighbor-joining tree of phylogenetic analysis of 16S rRNA gene based on primer set Brod541F–Amx820R. Numbers in parenthesis indicate how many clones of that species were among all the anammox bacterial clones in the particular sample (*clones that included in the constrained groups, number of clones were included in the brackets followed after the sample name). (From top to bottom: triangle 1: S12 (3), S14 (1), S15 (1), S16 (2), S17 (1), S18 (1), S19 (2); triangle 2: S12 (7), S13 (7), S14 (10),S16 (13), S18 (1), S20 (11); triangle 3: S14 (1), S15 (1), S19 (3); triangle 4: S10 (3), S11 (5), S13 (2), S14 (3), S15 (4), S16 (3), S17 (11), S18 (1), S19 (9), S20 (4))

Fig. 2

Neighbor-joining tree of phylogenetic analysis of 16S rRNA gene sequences based on primer set Amx368–820. The reference sequences are shown in italics, while the clones identified from this study are shown in bold. Numbers in brackets showed the quantity of clones belonging to that particular species. Bootstrap value was not shown when lower than 50

Fig. 3

Neighbor-Joining tree showing the phylogenetic analysis of hzo gene and known anammox bacterial hzo gene based on deduced protein sequences. Clones from GenBank were shown in italic, and clones identified from this study were shown in bold. The numbers in brackets show the number of times a sequence was detected among all the tested clones of a sample. Bootstrap values (1,000 replicates) higher than 50 % were shown here and the scale bar represents 10 % of sequence divergence

Fig. 4

Dendrogram of the hierarchical clustering analysis of different characterized paddy samples of 16S rRNA gene sequences constructed by the UniFrac unweighted Jacknife Environment Clusters statistical method. Distances are shown in the scale bar below

Fig. 5

Ordination diagrams of paddy soil calculated with the unweighted UniFrac PCoA analysis using 16S rRNA gene sequences. Plot of the first two principal coordinate axes (P1 and P2) is shown here and the distributions of each assemblage (designated with the sample details) in response to the axes are shown on the plot

Fig. 6

Dendrogram of the hierarchical clustering analysis of different characterized paddy samples hzo gene sequences constructed by the UniFrac unweighted Jacknife Environment Clusters statistical method. Distances were shown in the scale bar below

Fig. 7

Ordination diagrams of paddy soil calculated with the unweighted UniFrac PCoA analysis using the hzo gene sequences. Plot of the first two principal coordinate axes (P1 and P2) is shown here and the distributions of each assemblage (designated with the sample details) in response to the axes are shown on the plot