Uncovering novel bacterial and archaeal diversity: genomic insights from metagenome-assembled genomes in Cuatro Cienegas, Coahuila

Abstract

A comprehensive study was conducted in the Cuatro Ciénegas Basin (CCB) in Coahuila, Mexico, which is known for its diversity of microorganisms and unique physicochemical properties. The study focused on the "Archaean Domes" (AD) site in the CCB, which is characterized by an abundance of hypersaline, non-lithifying microbial mats. In AD, we analyzed the small domes and circular structures using metagenome assembly genomes (MAGs) with the aim of expanding our understanding of the prokaryotic tree of life by uncovering previously unreported lineages, as well as analyzing the diversity of bacteria and archaea in the CCB. A total of 325 MAGs were identified, including 48 Archaea and 277 Bacteria. Remarkably, 22 archaea and 104 bacteria could not be classified even at the genus level, highlighting the remarkable novel diversity of the CCB. Besides, AD site exhibited significant diversity at the phylum level, with Proteobacteria being the most abundant, followed by Desulfobacteria, Spirochaetes, Bacteroidetes, Nanoarchaeota, Halobacteriota, Cyanobacteria, Planctomycetota, Verrucomicrobiota, Actinomycetes and Chloroflexi. In Archaea, the monophyletic groups of MAGs belonged to the Archaeoglobi, Aenigmarchaeota, Candidate Nanoarchaeota, and Halobacteriota. Among Bacteria, monophyletic groups were also identified, including Spirochaetes, Proteobacteria, Planctomycetes, Actinobacteria, Verrucomicrobia, Bacteroidetes, Candidate Bipolaricaulota, Desulfobacteria, and Cyanobacteria. These monophyletic groups were possibly influenced by geographic isolation, as well as the extreme and fluctuating environmental conditions in the pond AD, such as stoichiometric imbalance of C:N:P of 122:42:1, fluctuating pH (5-9.8) and high salinity (5.28% to saturation).

Keywords: Archean Eon; extreme environments; geographic isolation; microbial diversification; microbial ecology; microbial evolution; phylogenomics.

Figure 1

AD site in Cuatro Ciénegas Basin (CCB). (A) Aerial view of the site. (B) Photo of CCB in 2016 when the site was first explored. The orange circles mark the prominent areas of the site that were investigated. (C) Dome-shaped structures called Archaean Domes. Photo credit: David Jaramillo.

Figure 2

Taxonomic assignment of the sequencing reads from the AD site in CCB. (A) Taxonomic classification with Kaiju at the phyla level using metagenomic reads. Labels above correspond to sampling area orange circles C) or domes D), sampling depth [0–10 cm (1) to 40–50 cm (5)], and sampling year 2016 to 2022. (B) The relative abundance of the most representative phyla. Pearson correlation coefficient values (r) between relative abundance and sampling depth.

Figure 3

Summary of MAGs assembled from sequencing reads from the AD site in CCB. (A) Number of MAGs obtained per sample. (B) MAG classified at the genus level. (C) Number of MAGs obtained per sampling depth (D) Relative abundance MAG per sample/total count.

Figure 4

Phylogenomic placement of MAGs from the AD on the Archaea (A) and Bacteria (B) domain tree. The phylogenomic tree was reconstructed using GTDB-tk. The branches in red show the retrieved AD lineages.

Figure 5

Phylogenomic tree including both Archaeal and Bacterial domains of the unique MAGs from the AD site in CCB. Amino acid substitutions were modeled using the WAG + F + I + R10 model, chosen according to BIC. The concentric circles with different dots show the sample origin, from surface to 40–50 cm. Functional annotation was performed with MEBS software, which uses the relative entropy measure H′(i) to detect enriched protein domains. The obtained H′ values (in bits) capture the extent to which a given Pfam domain informs the metabolism of interest. In this case, domains with H′ values close to or greater than 1 correspond to the most informative Pfam domains, while low H′ values (close to 0) indicate non-informative ones.

Figure 6

Heat map of the proportion of predicted KEGG orthologs (KO). Panel (A) shows the functional annotation for Archeal MAG, while panel (B) shows the functional annotation for the most abundant MAGs of the Bacteria domain. The color bar represents the relative abundance of KEGG orthologs (KOs) within each genome, measured as a percentage of the total number of coding sequences. The scale ranges from 0 to 10%, indicating the proportion of KOs relative to the genome’s coding sequences.

Figure 7

Metabolic map of MAG C1M08_23 of Methanohalobium (A) and MAG D4M11_7 (family Desulfatiglandaceae) (B). Some metabolic pathways of carbon and nitrogen fixation, and sulfur metabolism are depicted. The red outline ovals denote genes that are absent in the metabolic pathway. The plots were done using biorender.

Figure 8

Phylogenetic trees illustrating the evolutionary relationships of the Candidate Bipolaricaulota (A), PVC (B), and Cyanobacteria (C) groups, conducted to verify monophyletic groupings. Amino acid substitutions were modeled using the LG + F + R9 model for Candidate Bipolaricaulota, LG + F + R7 for PVC, and LG + F + R5 for Cyanobacteria. The leaves highlighted in red represent individuals from MAGs obtained from the AD site, showcasing a monophyletic grouping.