A Novel Freshwater to Marine Evolutionary Transition Revealed within Methylophilaceae Bacteria from the Arctic Ocean

Abstract

Bacteria inhabiting polar oceans, particularly the Arctic Ocean, are less studied than those at lower latitudes. Discovering bacterial adaptations to Arctic Ocean conditions is essential for understanding responses to the accelerated environmental changes occurring in the North. The Methylophilaceae are emerging as a model for investigating the genomic basis of habitat adaptation, because related lineages are widely distributed across both freshwater and marine ecosystems. Here, we investigated Methylophilaceae diversity in the salinity-stratified surface waters of the Canada Basin, Arctic Ocean. In addition to a diversity of marine OM43 lineages, we report on the genomic characteristics and evolution of a previously undescribed Methylophilaceae clade (BS01) common to polar surface waters yet related to freshwater sediment Methylotenera species. BS01 is restricted to the lower-salinity surface waters, while OM43 is found throughout the halocline. An acidic proteome supports a marine lifestyle for BS01, but gene content shows increased metabolic versatility compared to OM43 and evidence for ongoing genome-streamlining. Phylogenetic reconstruction shows that BS01 colonized the pelagic ocean independently of OM43 via convergent evolution. Salinity adaptation and differences in one-carbon and nitrogen metabolism may play a role in niche differentiation between BS01 and OM43. In particular, urea utilization by BS01 is predicted to provide an ecological advantage over OM43 given the limited amount of inorganic nitrogen in the Canada Basin. These observations provide further evidence that the Arctic Ocean is inhabited by distinct bacterial groups and that at least one group (BS01) evolved via a freshwater to marine environmental transition. IMPORTANCE Global warming is profoundly influencing the Arctic Ocean. Rapid ice melt and increased freshwater input is increasing ocean stratification, driving shifts in nutrient availability and the primary production that supports marine food webs. Determining bacterial responses to Arctic Ocean change is challenging because of limited knowledge on the specific adaptations of Arctic Ocean bacteria. In this study, we investigated the diversity and genomic adaptations of a globally distributed group of marine bacteria, the Methylophilaceae, in the surface waters of the Arctic Ocean. We discovered a novel lineage of marine Methylophilaceae inhabiting the Arctic Ocean whose evolutionary origin involved a freshwater to marine environmental transition. Crossing the salinity barrier is thought to rarely occur in bacterial evolution. However, given the ongoing freshening of the Arctic Ocean, our results suggest that these relative newcomers to the ocean microbiome increase in abundance and, therefore, ecological significance in a near-future Arctic Ocean.

Keywords: climate change; genome evolution; marine microbiology; metagenomics; methanol.

FIG 1

Phylogenetic analysis of 16S rRNA genes from Methylophilaceae from Canada Basin metagenomes and a diversity of aquatic ecosystems. The tree was inferred using maximum likelihood (500 bootstraps) and GTR + gamma distribution (four categories) with invariant site model of evolution and the nearest-neighbor interchange heuristic search method. The tree was rooted using Methylibium as an outgroup to the Methylophilaceae. Sequences from the current study are highlighted in red. Only bootstrap values of >60 are included in the tree.

FIG 2

Diversity and biogeography of Methylophilaceae based on ITS variants recovered from Canada Basin metagenomes. (A) Phylogenetic analysis of the Methylophilaceae group across various aquatic regions and depths using the ITS region. The tree was inferred using maximum likelihood (500 bootstraps) and a GTR + gamma distribution (four categories) with invariants sites model of evolution and the nearest-neighbor interchange heuristic search method. Sequences from the current study are highlighted in red. Only bootstrap values of >60 are included in the tree. (B) Abundance of six ITS variants based on summed coverage in metagenome assemblies. (C) Principal coordinate analysis ordination of Bray-Curtis dissimilarities of Arctic samples based on summed coverage of six ITS variants.

FIG 3

Phylogenomic comparison of BS01 with representative Methylophilaceae genomes. (A) Maximum likelihood phylogenetic analysis of a concatenated alignment of 48 orthologs shared between all Methylophilaceae genomes. Values at the nodes are bootstrap values (100 pseudoreplicates). (B) Scatterplot comparing G+C content and genome size. (C) Whole-proteome pI values versus relative frequency in select Methylophilaceae genomes from freshwater and marine habitats.

FIG 4

Reconstruction of methylotrophic metabolism in BS01 and comparison to other Methylophilaceae. (A) Distribution of central one-carbon metabolism genes. (B) Gene expression pattern for central carbon metabolism pathways in Canada Basin surface waters revealed through fragment recruitment of metatranscriptomics against Met-BS01-1 and HTCC2181 genomes. (C) Quantification of BS01 and OM43 abundances in the Canada Basin using qPCR analysis of xoxF4 gene abundance. Error bars indicate standard deviation. DCM, deep chlorophyll maximum; PWW, Pacific winter water.

FIG 5

Reconstruction of nitrogen metabolism in BS01 and comparison to other Methylophilaceae. (A) Distribution of central nitrogen metabolism genes. (B) Gene expression pattern for central nitrogen metabolism pathways in Canada Basin surface waters revealed through fragment recruitment of metatranscriptomics against Met-BS01-1 and HTCC2181 genomes.

FIG 6

Biogeography of BS01 across the global ocean and estuaries. (A) Phylogenetic analysis of the Methylophilaceae family across multiple aquatic regions and depths using the XoxF4 methanol dehydrogenase protein recovered from metagenomes. The tree was inferred using maximum likelihood (500 bootstraps) and JTT + gamma distributed with invariants (four categories) sites model of evolution, with nearest-neighbor interchange heuristic search method. Colored sequences are those from the Arctic Ocean (red), the Antarctic (green), or estuaries (blue). Only bootstrap values of >60 are included in the tree. (B) Distribution of BS01 and OM43-A revealed through fragment recruitment of aquatic metagenomes against Met-BS01-1 and HTCC2181 genomes reported as reads per kilobase of the MAG per gigabase of metagenome (RPKG). The diagonal lines are to signify no metagenome data for that water column feature.