Recovery in methanotrophic activity does not reflect on the methane-driven interaction network after peat mining

Abstract

Aerobic methanotrophs are crucial in ombrotrophic peatlands, driving the methane and nitrogen cycles. Peat mining adversely affects the methanotrophs, but activity and community composition/abundances may recover after restoration. Considering that the methanotrophic activity and growth are significantly stimulated in the presence of other microorganisms, the methane-driven interaction network, encompassing methanotrophs and non-methanotrophs (i.e., methanotrophic interactome), may also be relevant in conferring community resilience. Yet, little is known of the response and recovery of the methanotrophic interactome to disturbances. Here, we determined the recovery of the methanotrophic interactome as inferred by a co-occurrence network analysis, comparing a pristine and restored peatland. We coupled a DNA-based stable isotope probing (SIP) approach using ¹³C-CH₄ to a co-occurrence network analysis derived from the ¹³C-enriched 16S rRNA gene sequences to relate the response in methanotrophic activity to the structuring of the interaction network. Methanotrophic activity and abundances recovered after peat restoration since 2000. 'Methylomonaceae' was the predominantly active methanotrophs in both peatlands, but differed in the relative abundance of Methylacidiphilaceae and Methylocystis However, bacterial community composition was distinct in both peatlands. Likewise, the methanotrophic interactome was profoundly altered in the restored peatland. Structuring of the interaction network after peat mining resulted in the loss of complexity and modularity, indicating a less connected and efficient network, which may have consequences in the event of recurring/future disturbances. Therefore, determining the response of the methane-driven interaction network, in addition to relating methanotrophic activity to community composition/abundances, provided a more comprehensive understanding of the resilience of the methanotrophs.Importance The resilience and recovery of microorganisms from disturbances are often determined with regard to their activity and community composition/abundances. Rarely has the response of the network of interacting microorganisms been considered, despite accumulating evidence showing that microbial interaction modulates community functioning. Comparing the methane-driven interaction network of a pristine and restored peatland, our findings revealed that the metabolically active microorganisms were less connected and formed less modular 'hubs' in the restored peatland, indicative of a less complex network which may have consequences with recurring disturbances and environmental changes. This also suggests that the resilience and full recovery in the methanotrophic activity and abundances do not reflect on the interaction network. Therefore, it is relevant to consider the interaction-induced response, in addition to documenting changes in activity and community composition/abundances, to provide a comprehensive understanding of the resilience of microorganisms to disturbances.

FIG 1

Relative pmoA gene abundance along the density gradient of the [¹³C]CH₄ and [^unlabelledC]CH₄ incubations from the pristine (A) and restored (B) peatlands (mean ± standard deviation [SD]; n = 4). The relative abundance was calculated as the proportion of each fraction over the total sum of the gene abundance for each sample. DNA from the “light” and “heavy” fractions (denoted by arrows) in the [¹³C]CH₄ incubations was used for 16S rRNA gene amplicon sequencing.

FIG 2

Principal-component analysis showing the clustering of the 16S rRNA gene sequences according to the different fractions (light and heavy) and sites (pristine and restored peatlands) of the incubation with [¹³C]CH₄. The circle and triangle indicate pristine and restored peatlands, respectively.

FIG 3

The bacterial (A) and methanotrophic (B) community composition in the starting material (prior to the incubation) and after [¹³C]CH₄ incubation (light and heavy fractions) (mean; n = 4). The 16S rRNA gene sequences affiliated with methanotrophs in panel B were retrieved from the total community in panel A. The 16S rRNA gene sequences affiliated with the methanotrophs were present at <2% of the total community in the starting material, and at ∼20% and ∼74%, respectively, in the pristine and restored peatland after incubation (heavy fraction). P and R denote pristine and restored peatlands, respectively.

FIG 4

Co-occurrence network analysis in the pristine (A) and restored (B) peatlands. The network analysis was derived from the ¹³C-enriched 16S rRNA gene sequences (heavy fraction), representing the metabolically active community of the interaction network. The topological properties of the networks are given in Table 2. Significant connection (P < 0.01) with a SparCC correlation of a magnitude of more than 0.7 (positive correlation, blue edges) or less than −0.7 (negative correlation, red edges) are given. Each node represents a bacterial taxon at the operational taxonomic unit (OTU) level, given to the lowest taxonomic rank (family, genus, or species) when available. The size of the node is proportional to the number of connections, and a darker shade of a color indicates higher betweenness centrality. The top 10 nodes with the highest betweenness centrality, representing the key nodes, are given as triangles, and the number inside the key nodes refers to their affiliation as follows: 1, Rhodospirillales; 2, uncultured bacteria; 3, Burkholderiaceae; 4, Methylomonas (methanotroph); 5, “Candidatus Solibacter”; 6, Gammaproteobacteria; 7, Methylocystis (methanotroph); 8, Babeliales; 9, Pirellulaceae; 10, Methylacidiphilaceae (methanotroph); 11, Bdellovibrio; 12, Magnetospirillaceae; 13, Acidimicrobiia; 14, Sphingobacteriales; 15, Roseiarcus; 16, Beijerinckiaceae; 17, Myxococcales; and 18, Methylomonas paludis (methanotroph). The OTUs representing the microorganisms and betweenness centrality values of each OTU are listed in Table S1 in the supplemental material.