Metagenomic insights into diazotrophic communities across Arctic glacier forefields

Abstract

Microbial nitrogen fixation is crucial for building labile nitrogen stocks and facilitating higher plant colonisation in oligotrophic glacier forefield soils. Here, the diazotrophic bacterial community structure across four Arctic glacier forefields was investigated using metagenomic analysis. In total, 70 soil metagenomes were used for taxonomic interpretation based on 185 nitrogenase (nif) sequences, extracted from assembled contigs. The low number of recovered genes highlights the need for deeper sequencing in some diverse samples, to uncover the complete microbial populations. A key group of forefield diazotrophs, found throughout the forefields, was identified using a nifH phylogeny, associated with nifH Cluster I and III. Sequences related most closely to groups including Alphaproteobacteria, Betaproteobacteria, Cyanobacteria and Firmicutes. Using multiple nif genes in a Last Common Ancestor analysis revealed a diverse range of diazotrophs across the forefields. Key organisms identified across the forefields included Nostoc, Geobacter, Polaromonas and Frankia. Nitrogen fixers that are symbiotic with plants were also identified, through the presence of root associated diazotrophs, which fix nitrogen in return for reduced carbon. Additional nitrogen fixers identified in forefield soils were metabolically diverse, including fermentative and sulphur cycling bacteria, halophiles and anaerobes.

Figure 1.

Relationship between normalised nif gene abundance (nifHDKNE) and concentration of TOC and TN per gram of soil, across all sampling sites. Nif gene abundance values are normalised against the bacterial single copy housekeeping gene, rpoB, for each metagenome. Values across the different forefields are noted, including: Midtre Lovenbreen (Ml), Russel (Rl), Rabots (Rb) and Storglaciaren (St).

Figure 2.

nifH maximum likelihood phylogeny of sample sequences (bold) and sequenced samples derived from NCBI GenBank and UniProtKB. Most sample sequences were obtained from the nifH phylogeny of Deslippe and Egger (2006). For study samples, the Sample ID is given, corresponding to Table S1 (Supporting Information). For sequenced samples, the database, organism name and gene are given. Bootstrap support values are given, based on 1000 tree iterations. The nifH clusters (derived from Zehr et al.2003) are denoted by leaf colours (Cluster I-IV). The tree is rooted on Cluster IV, as this group contains divergent ‘nif-like’ sequences (Zehr et al.2003). Key groups containing sample sequences are noted, including Firmicutes, Cyanobacteria, Alphaproteobacteria, Betaproteobacteria and Deltaproteobacteria.

Figure 3.

Taxonomic distribution of nif (HDKNE) genes for each forefield at the genus level: Midtre Lovenbreen Ml (A), Russell Rl (B), Rabots Rb (C), Storglaciaren St (D). The total nif gene sequence count for each site was 42, 15, 13 and 91, respectively.