Comparative Genomic Insights into the Evolution of Halobacteria-Associated " Candidatus Nanohaloarchaeota"

Abstract

Members of the phylum "Candidatus Nanohaloarchaeota," a representative lineage within the DPANN superphylum, are characterized by their nanosized cells and symbiotic lifestyle with Halobacteria. However, the development of the symbiosis remains unclear. Here, we propose two novel families, "Candidatus Nanoanaerosalinaceae" and "Candidatus Nanohalalkaliarchaeaceae" in "Ca. Nanohaloarchaeota," represented by five dereplicated metagenome-assembled genomes obtained from hypersaline sediments or related enrichment cultures of soda-saline lakes. Phylogenetic analyses reveal that the two novel families are placed at the root of the family "Candidatus Nanosalinaceae," including the cultivated taxa. The two novel families prefer hypersaline sediments, and the acid shift of predicted proteomes indicates a "salt-in" strategy for hypersaline adaptation. They contain a lower proportion of putative horizontal gene transfers from Halobacteria than "Ca. Nanosalinaceae," suggesting a weaker association with Halobacteria. Functional prediction and historical events reconstruction disclose that they exhibit divergent potentials in carbohydrate and organic acid metabolism and environmental responses. Globally, comparative genomic analyses based on the new families enrich the taxonomic and functional diversity of "Ca. Nanohaloarchaeota" and provide insights into the evolutionary process of "Ca. Nanohaloarchaeota" and their symbiotic relationship with Halobacteria. IMPORTANCE The DPANN superphylum is a group of archaea widely distributed in various habitats. They generally have small cells and have a symbiotic lifestyle with other archaea. The archaeal symbiotic interaction is vital to understanding microbial communities. However, the formation and evolution of the symbiosis between the DPANN lineages and other diverse archaea remain unclear. Based on phylogeny, habitat distribution, hypersaline adaptation, host prediction, functional potentials, and historical events of "Ca. Nanohaloarchaeota," a representative phylum within the DPANN superphylum, we report two novel families representing intermediate stages, and we infer the evolutionary process of "Ca. Nanohaloarchaeota" and their Halobacteria-associated symbiosis. Altogether, this research helps in understanding the evolution of symbiosis in "Ca. Nanohaloarchaeota" and provides a model for the evolution of other DPANN lineages.

Keywords: DPANN superphylum; comparative genomics; evolution; horizontal gene transfer; symbiosis; “Candidatus Nanohaloarchaeota”.

FIG 1

Phylogeny of the phylum “Ca. Nanohaloarchaeota.” (a) Phylogenomic tree based on the 122 single-copy ubiquitous proteins in GTDB. It was obtained by pruning the tree in Fig. S1. Briefly, the best-fit model of LG+F+G4 was chosen, and a consensus tree based on ultrafast bootstrap approximation of 1,000 times is presented. (b) Average AAI matrix for the genomes of “Ca. Nanohaloarchaeota.” The data are rounded by omitting decimal fractions smaller than 0.5 and counting all others (including 0.5) as 1. The background (from yellow to orange) is colored according to the threshold AAIs of species, genus, and family (95.0, 65.0, and 45.0%, respectively). The genomes sharing an AAI of more than 95.0% with other genomes of higher completeness or low contamination are marked by red, and they were abandoned in subsequent research.

FIG 2

Relative abundance of the three families of the phylum “Ca. Nanohaloarchaeota” in soda-saline lake samples and the related enrichment cultures. The relative abundance was estimated based on the percentage of the reads that mapped onto the MAG obtained from the soda-saline lake samples, because most MAGs assembled from other samples were not detected (Table S2). NHA21, “Ca. Nanohalalkaliarchaeaceae”; NHA24, NHA20, NHA23, and NHA-2, “Ca. Nanoanaerosalinaceae”; NHA25, NHA26, NHA29, and NHA-1, “Ca. Nanosalinaceae.” The metagenomes of brine, surface sediment, and deep sediment were published in previous reports (24, 29). In their sample names, HC and DK represent Hutong Qagan Lake and Habor Lake, respectively, and the numbers represent the salinities of brines; in the brine and surface sediment, W and S represent water (or brine) and sediment, respectively; in the deep sediment, the numbers following “Sed” show the salinities of the pore water. The enrichment culture is described in Materials and Methods.

FIG 3

Horizontal gene transfer inference in the genomes of the phylum “Ca. Nanohaloarchaeota.” The taxonomic percentage of horizontal gene transfer donors for each strain was inferred using the HGTector pipeline with a GTDB taxonomy-based database based on the archaeal genomes in GTDB release 202. The genomes that are affiliated with the families “Ca. Nanosalinaceae,” “Ca. Nanoanaerosalinaceae,” and “Ca. Nanohalalkaliarchaeaceae” (NHAA) are marked.

FIG 4

Comparison of isoelectric point profiles and amino acid compositions. (a) Isoelectric point profiles of the predicted proteomes of “Ca. Nanoanaerosalinaceae,” “Ca. Nanohalalkaliarchaeaceae,” and reference species. The y axis shows the frequencies of proteins in the proteomes at each isoelectric point. The isoelectric point of each protein was predicted based on the amino acid sequence. The isoelectric point profiles with a bin width of 0.1 are shown. Haloferax mediterranei ATCC 33500 (GCA_000306765.2) and Escherichia coli O157:H7 strain Sakai (GCA_000008865.2) represent acid-shifted salt-in halophiles and nonhalophiles, respectively. Numbers in parentheses are average isoelectric points (more details are presented in Table S3). (b) Percentage of acidic amino acid glutamate, aspartate, and both. The composition was calculated from the predicted proteome based on the genome sequence. D, aspartate; E, glutamate; D+E, the sum of glutamate and aspartate.

FIG 5

Functional potentials of the phylum “Ca. Nanohaloarchaeota” lineages. (a) Dot plot showing the presence or absence of genes involved in metabolism and environmental response in the members of the two novel families and the percentage in the family “Ca. Nanosalinaceae.” Solid and hollow dots indicated presence and absence in the genomes, respectively. Transparent blues show the percentage of genes in 14 “Ca. Nanosalinaceae” genomes. (b) Reconstruction of functional potentials in three families of “Ca. Nanohaloarchaeota.” The process was estimated based on the genes involved in genetic information processing, metabolism, and environmental stress response. Solid dots with different colors indicate the presence of the process or gene(s) in the three families, i.e., “Ca. Nanohalalkaliarchaeaceae” (NHAA) represented by NHA21, “Ca. Nanoanaerosalinaceae,” and “Ca. Nanosalinaceae,” while hollow dots indicate the absence of the process or gene(s). Glc, glucose; Glc-1P, glucose 1-phosphate; Glc-6P, glucose 6-phosphate; F-6P, fructose 6-phosphate; F-1,6P₂, fructose 1,6-bisphosphate; DHAP, dihydroxyacetone phosphate; G-3P, glyceraldehyde 3-phosphate; PEP, phosphoenolpyruvate; Glu, glutamate; OAA, oxaloacetate; Cit, citrate; Ict, isocitrate; 2-OG, 2-oxoglutarate; Scn-CoA, succinyl-CoA; Scn, succinate; Fmr, fumarate; Che, chemotaxis; GlcN-6P, glucosamine 6-phosphate; GlcN-1P, glucosamine 1-phosphate; GlcNAc-1P, N-acetylglucosamine 1-phosphate; UDP-GlcNAc, UDP-N-acetylglucosamine; PHB, poly-hydroxybutyrate. Enzyme abbreviations are listed in Table S4.

FIG 6

History event approximation in the phylum “Ca. Nanohaloarchaeota.” (a) Ancestral reconstruction tree of “Ca. Nanohaloarchaeota.” The consensus tree of ultrafast bootstrap approximation based on 124 single-copy ubiquitous orthogroups in representative genomes is exhibited. The historical events are approximated based on the species tree and 1,629 gene UFBOOT trees. The radii of the black circles at the nodes represent their (inferred) genome sizes. Some internal nodes of interest are marked by Arabic numerals. The bar charts above the horizontal branches represent the numbers of duplications, transfers, originations, and losses (bar heights in the legend correspond to 300 events each). Some branches were extended with dashed lines to fit the width of the bar charts. The branches leading to the phylum EX4484-52 were collapsed. (b) Dot plot showing the functional annotation of the historical event at the interest nodes. The orthogroups that achieved a threshold of 0.3 in the raw reconciliation frequencies are reported. The putative functions of orthogroups were estimated by medoid sequences, which have the highest sum of similarity scores with all other sequences based on the BLOSUM62 substitution matrix.