Roles of core nosZ denitrifiers in enhancing denitrification activity under long-term rice straw retention

Abstract

The denitrification process is known to contribute to soil nitrogen (N) loss, which is strongly affected by fertilization strategies; however, the effects of distinct straw retention modes on soil denitrification activity have rarely been discriminated and the underlying mechanisms remain unclear. This study coupled field and incubation experiments to explore the characteristics of soil denitrification activity, soil and standing water physicochemical properties, and the abundance, community diversity, and co-occurrence network of nosZ denitrifiers, based on a paddy field implementing 10-year straw retention under a rice-wheat rotation system. Four straw retention treatments with equivalent chemical fertilizers were applied, namely no straw (NS), wheat straw only (WS), rice straw only (RS), and wheat and rice straw (WRS). Results indicated a significant increase (by 41.93-45.80% when compared to that with NS) in the soil denitrification activity with RS and WRS. Correspondingly, treatments with rice straw retention resulted in the development of a similar community composition (P < 0.05), structure (P = 0.001), and more positively interconnected network, as well as similar specific keystone taxa of nosZ denitrifiers, relative to those in non-rice straw mode. Under long-term rice straw retention conditions, the core nosZ-denitrifying phylogroups shifted (r = 0.83, P < 0.001), with the recruitment of keystone taxa from the phyla Bacteroidetes and Euryarchaeota playing a key role in enhancing denitrification activity and stimulating N loss. Accordingly, in a rice-wheat rotation field, the practice of wheat straw retention in a single season is recommended because it will not markedly sacrifice soil N availability impaired by the denitrification process.

Keywords: denitrification activity; denitrifiers; nosZ gene; paddy field; straw retention.

Figure 1

Denitrification enzyme activities (A) and the abundance of the nosZ gene (B) under the four fertilization regimes. Error bars denote the standard errors (n = 3) and are accompanied by different lowercase letters indicating significant differences (P < 0.05) among the four treatment groups according to Tukey’s HSD post-hoc test. NS, no straw; RS, rice straw only; WS, wheat straw only; WRS, rice straw and wheat straw.

Figure 2

Circular maximum likelihood phylogenetic tree (A) and the community compositions (B), presented based on genus and phylum levels, respectively, of the denitrifying bacteria among treatment groups. The tree is based on the nosZ gene sequences of the most abundant operational taxonomic units. Genera in (A) are color-coded by phylum in (B). NS, no straw; RS, rice straw only; WS, wheat straw only; WRS, rice straw and wheat straw.

Figure 3

Differences in the community compositions of denitrifying bacteria at family (A) and genus (B) levels and in the bacterial community structures (C) among treatment groups. The statistical differences were calculated using Tukey’s HSD post-hoc test (marked with asterisks, P < 0.05, n = 3), Kruskal-Wallis H-test (P < 0.05, n = 3), and PERMANOVA test (P = 0.001, n = 6) in (A–C), respectively. NS, no straw; RS, rice straw only; WS, wheat straw only; WRS, rice straw and wheat straw.

Figure 4

Co-occurrence networks (A) and the keystone taxa (B) of denitrifying bacterial operational taxonomic units (OTUs) based on their topological roles, and the relationships of keystone taxa with soil environmental predictors calculated based on the random forest model (C). Non-R and R modes integrate treatments without (NS+WS) and with (RS+WRS) rice straw retention, respectively. Nodes in (A) are colored based on the phylum level, and their sizes are proportional to the number of degrees; the red and blue edges represent positive and negative correlations between two nodes, respectively, the thicknesses of which are proportional to the value of Spearman’s correlation coefficients. Keystone taxa of co-occurrence networks represented by hubs and connectors in (B) are categorized by threshold lines of Zi = 2.5 and Pi = 0.625, respectively. P-values < 0.05, < 0.01 and < 0.001 are indicated using *, **, and *** in (C), respectively. NS, no straw; RS, rice straw only; WS, wheat straw only; WRS, rice straw and wheat straw.

Figure 5

Structural equation models describing the direct effects of physicochemical and bacterial community characteristics of paddy soil on denitrification enzyme activities under non-R (A) and R (B) modes, respectively.