Unexpected novel clade III type nitrous oxide-reducing bacteria from incubated lake sediments

Abstract

Nitrous oxide-reducing bacteria (N₂ORB) play a pivotal role in regulating N₂O emissions in aquatic ecosystems, with clade I and clade II nosZ-harboring microorganisms representing well-recognized contributors to microbial N₂O consumption. Beyond conventional N₂ORB, the recently identified clade III nosZ from soil may represent a previously overlooked potential N₂O sink, yet the distribution and characterization remain largely unexplored in aquatic ecosystems. Here we established microcosm systems using sediments from five lakes and subjected them to warming temperature gradients to investigate the diversity and genomic characteristics of N₂ORB. Hidden Markov model (HMM)-based analyses identified a total of 45 nonredundant nosZ sequences, including 12 affiliated with clade III nosZ. Clade III nosZ accounted for 10.2%–40.6% of total nosZ genes, indicating that clade III nosZ-harboring N₂ORB is widespread and non-negligible. Reconstruction of metagenome-assembled genomes (MAGs) identified four phylogenetically novel clade III nosZ-harboring N₂ORB, with these MAGs showing low average amino acid identity to their closest known reference genomes. These MAGs showed different denitrification gene inventories, with MAG33 lacking identifiable genes for upstream N₂O-producing steps, suggesting a potential non-denitrifying N₂O reducer. They also encoded oxygen-related stress-response genes, suggesting a potential ability to perform N₂O respiration in the presence of oxygen. Unlike canonical clade I/II nosZ clusters, clade III nosZ-harboring MAGs lacked typical accessory genes and instead exhibited distinct neighboring transporter- and cytochrome-related genes. Together, our results provide evidence for the occurrence of clade III nosZ-harboring N₂ORB in non-soil ecosystems and expand current understanding of their genomic traits.

Introduction

Nitrous oxide (N₂O) is a potent greenhouse gas that depletes stratospheric ozone and contributes to global warming [1, 2]. The primary biological sink for N₂O is reduction to dinitrogen (N₂), which is mediated by N₂O-reducing bacteria (N₂ORB) via N₂O-reductase (NosZ) encoded by the nosZ gene [3, 4]. N₂ORB were initially classified into two clades (clade I and clade II) [3, 5] based on amino acid sequence divergence. These clades often differ in host ecophysiology, including the co-occurrence of upstream denitrification genes and their potential roles as complete denitrifiers or non-denitrifying N₂O reducers. A recent study identified a previously unrecognized protein family in soil, namely, the lactonase-type nosZ clade (clade III nosZ) [6], greatly expanding the known diversity of N₂ORB. Clade III nosZ sequences differ markedly from the typical nosZ clades, yet retain structural features associated with N₂O reduction. Therefore, exploring clade III nosZ-harboring N₂ORB is critical for uncovering previously unrecognized N₂O sink pathways and refining our understanding of N₂O cycling in ecosystems.

Lake ecosystems are important components of the global nitrogen cycle, and N₂O emissions have received considerable attention [7, 8]. The diversity and activity of clade I and clade II nosZ-harboring N₂ORB in aquatic environments have been relatively well documented [9, 10]. In contrast, the recently identified clade III nosZ, also referred to as lactonase-type nosZ, has been reported mainly from soil, groundwater, and marine environments [6], while its diversity, abundance, host taxonomy, and genomic traits in lake ecosystems remain unclear. Owing to the absence of upstream denitrification genes [11], some N₂ORB are incapable of nitrate/nitrite reduction and therefore act as net sinks for N₂O [12]. However, the limited understanding of these N₂ORB likely leads to an underestimation of the N₂O sink capacity in aquatic systems [13]. Therefore, a comprehensive characterization of clade III nosZ-harboring N₂ORB is essential to better constrain their ecological role and potential contribution to N₂O mitigation. Here, we employed metagenomic approaches to investigate the diversity, phylogeny, and genomic potential of N₂ORB in laboratory-incubated lake sediments. By combining hidden Markov model (HMM)-based identification with metagenome-assembled genome (MAG) reconstruction, we aimed to (i) characterize the distribution of different types of N₂ORB, with a particular focus on clade III nosZ, and (ii) evaluate the genotypes of clade III nosZ-harboring N₂ORB. Our findings provide genomic evidence for previously overlooked clade III nosZ diversity in lake sediments and establish a basis for future studies assessing their activity and ecological relevance in freshwater N₂O cycling.

Results and discussion

To assess the diversity of N₂ORB in laboratory-incubated lake sediments (Texts S1 and S2), we constructed an HMM based on 264 reference NosZ sequences (including clade I, II, and III nosZ) to identify nosZ-encoding sequences in the assembled contigs [6] (Text S3). In total, 45 unique nosZ sequences were recovered, including 29 affiliated with the clade II nosZ and 12 with the clade III nosZ (Fig. 1A). Taxonomically, clade III nosZ genes were affiliated with the phyla Nitrospirota, Nitrospinota, Pseudomonadota, and Myxococota. In comparison, clade II nosZ genes showed broader taxonomic diversity across the seven phyla, whereas clade I nosZ genes were affiliated with Pseudomonadota. The mean abundances of clades I, II, and III nosZ were 3.33, 28.95, and 11.26 TPM, respectively (Fig. S1). Although clade II nosZ genes predominated across lakes, clade III nosZ genes accounted for 10.2%–40.6% of the combined abundance of all nosZ genes (Fig. 1B–C), indicating that clade III nosZ-harboring N₂ORB are widespread and non-negligible members that contribute to N₂O consumption. Despite the distinct warming scenarios among the lakes, warming did not significantly restructure the nosZ-containing community (Fig. S2), suggesting that the N₂ORB community composition is relatively stable under current warming conditions.

Figure 1.

Phylogeny and temperature responses of nosZ clade in lake sediments. (A) Maximum-likelihood phylogenetic tree of nosZ genes, constructed using a combination of HMM searches and motif-based filtering, together with reference nosZ sequences. (B) Ternary phase diagram showing the relative abundance distribution of Clade I, Clade II, and Clade III nosZ across three temperature incubation treatments (T, T + 1.5, and T + 4). Each axis represents the relative abundance (%) of the respective clade. For any point in the ternary phase diagram, the three values sum to 100%. (C) The clade III/total nosZ ratio across five lakes under different treatments.

To dissect the genotypes of specific N₂ORB, MAGs were reconstructed from the sediment samples using MetaBAT2 [14] (Text S4; Fig. 2). HMM-based classification of nosZ sequences identified 14 MAGs as putative N₂ORB [6]. All MAGs exhibited estimated contamination below 10% and completeness above 70% (Table S1). Among them, four MAGs (MAG15, MAG29, MAG33, and MAG48) harbored the clade III nosZ gene and were assigned to the phyla Nitrospinota, Desulfobacterota, Nitrospirota, and Nitrospirota, respectively (Fig. S3). These phyla have previously been reported to contain clade III nosZ mainly in soil, groundwater, and marine environments [6]. To the best of our knowledge, this is the first evidence of clade III nosZ-harboring MAGs in lake ecosystems. Compared with their closest known relatives, all four MAGs harboring the clade III nosZ gene showed amino acid identity (AAI) values less than 50%, suggesting that they represent previously uncharacterized clade III nosZ-harboring N₂ORB lineages (Table S2). MAG48 encoded a near-complete denitrification gene set, whereas MAG15, MAG29, and MAG33 lacked detectable nar/nxr-related genes, suggesting limited potential for nitrate reduction. Notably, the nitric oxide reductase genes norB/norC were not detected in MAG33, suggesting that it may primarily function in N₂O consumption rather than contribute to N₂O production. This finding expands the diversity of non-denitrifying N₂ORB, which have previously been predominantly reported within clade II nosZ [4]. However, due to the incomplete recovery of MAG33 (72.4% completeness), the absence of norB/norC may reflect genome incompleteness or true absence, warranting further validation. Additionally, all clade III nosZ-harboring MAGs carried ccoN, cydA, SOD2, and ahpC, suggesting that they are capable of microaerobic respiration and possess mechanisms to tolerate oxidative stress [15, 16]. We also observed that eight MAGs (MAG5, MAG14, MAG21, MAG22, MAG32, MAG39, MAG43, and MAG44) contained the clade II nosZ gene but lacked at least one denitrification gene. In addition, two MAGs (MAG23 and MAG24) harboring the clade I nosZ gene, of which MAG23 possessed a complete denitrification pathway. These findings highlight clade III nosZ-harboring N₂ORB as a previously unrecognized and potentially important component of N₂ORB.

Figure 2.

Characterization of reconstructed nosZ-harboring MAGs. Phylogenetic tree of the MAGs and their closest neighbors based on the analysis of 50 marker genes. Quality metrics, including contamination and completeness, are shown together with AAI (%) values calculated using TypeMat in MiGA Online by comparing each MAG against reference genomes. Abundance and gene inventories of the nosZ-harboring MAGs are displayed. The different colors of the MAGs represent the corresponding phylum. The heatmap illustrates MAG abundance across samples and temperature treatments. Filled circles indicate detected genes, whereas unfilled circles indicate undetected genes. Detailed gene information is provided in Supplementary Information Table S3.

A gene-cluster analysis of 14 nosZ-containing MAGs (Fig. S4) showed that clade I nosZ-harboring MAGs and a subset of clade II nosZ-harboring MAGs possess conserved accessory-gene modules associated with canonical nosZ, including nosR, nosD, nosF, nosY, and nosL. In addition, many clade II nosZ-harboring MAGs encode a putative nosB [4, 6, 12]. In contrast, clade III nosZ-harboring MAGs lacked identifiable canonical nos accessory genes. Instead, genes encoding cytochromes and transporter-related proteins were found in the genomic vicinity of the clade III nosZ. In the clade III nosZ-harboring MAG15 and MAG29, two adjacent genes encode an ABC transporter ATP-binding protein and a permease that show sequence similarity to canonical NosF and NosY, respectively. However, these genes were located at a substantial distance from nosZ and were not accompanied by NosD or NosL. In light of the recently described clade III/L-nosZ architecture, such dispersed transporter genes should be described as NosF/NosY-like homologues rather than annotated as canonical NosF or NosY [6]. Whether these genes contribute to N₂O reduction in these MAGs remains to be experimentally validated.

In summary, this study provides evidence for the occurrence of clade III nosZ-harboring N₂ORB in a non-soil environment [6], expanding their known distribution beyond previously reported habitats. We recovered previously unrecognized clade III nosZ-harboring N₂ORB from sediment-derived microcosms inoculated with lakes, suggesting that this overlooked lineage may also be present in other yet underexplored ecosystems (e.g. aquatic and engineering systems). These findings further broaden our current understanding of the diversity and ecological distribution of clade III nosZ-harboring N₂ORB. Notably, the detection of non-denitrifying clade III nosZ-harboring N₂ORB suggests that this lineage may represent a previously underestimated component of potential N₂O sink. However, their in situ distribution, functional activity, and ecological relevance remain to be confirmed. In particular, the contribution of clade III nosZ-harboring N₂ORB to N₂O mitigation in natural environments requires further investigation, as their role may have been long overlooked in N₂O cycling. Therefore, further investigation of their genotypic and phenotypic traits is needed to better constrain their contribution to the N₂O sink.

Author contributions

Y.Z. conceived and designed the research. F.S., S.W., X.W., Y.S., and Y.Z. performed the experiments. F.S. and S.W. analyzed the experimental and sequencing data, with contributions from K.S., J.L., L.Z., F.W., and Y.Z. wrote the manuscript. All the authors reviewed and approved the submitted version.

Supplementary material

Supplementary material is available at ISME Communications online.

Conflicts of interest

The authors declare that they have no competing interests.

Funding

This research was funded by the NSFC (grant nos. 42577258 and 42207295), the Guangdong Basic and Applied Basic Research Foundation (grant nos. 2025A1515011262 and 2026A1515011016), the Science and Technology Foundation of Guangzhou (grant no. 2025A04J3659), the Guangdong Foundation for the Program of Science and Technology Research (grant no. 2023B1212060044), the Guangdong Major Project of Basic and Applied Basic Research (grant no. 2023B0303000006), and the GDAS’ Project of Science and Technology Development (grant nos. 2020GDASYL-20200101002 and 2022GDASZH-2022010104-2).

Data availability

The raw metagenomic sequencing data have been deposited in the NCBI database accession number PRJNA1459056.