Salty secrets of Halobacterium salinarum AD88: a new archaeal ecotype isolated from Cuatro Cienegas Basin

Abstract

The Cuatro Cienegas Basin (CCB) in Mexico, represents a unique ecological habitat, characterized by extreme and fluctuating conditions, providing a window into ancient evolutionary processes. This basin, characterized by hypersalinity and phosphorus scarcity, harbors diverse microbial communities that exhibit remarkable adaptations to oligotrophic conditions. Among these, Halobacterium salinarum, a halophilic archaeon known for its polyploid genome and metabolic versatility, has been extensively studied as a model for extremophile survival. However, only a limited number of H. salinarum strains have been successfully cultured and characterized to date. Here, we report the isolation and genomic analysis of a novel Halobacterium salinarum strain, AD88, from microbial mats at the Archaean Domes site in the CCB. This strain displays unique genomic features, including smaller plasmid sizes and distinctive metabolic pathways for phosphorus and sulfur utilization. Comparative analyses with other Halobacterium strains revealed genetic innovations, such as genes involved in sulfolipid biosynthesis, enabling membrane stability in phosphorus-depleted environments, and adaptations for horizontal gene transfer, which facilitate genomic flexibility in response to environmental pressures. This study reveals that H. salinarum AD88 is the first recorded diploid strain of Halobacterium, a feature previously undocumented in this genus. Phylogenomic reconstruction positioned AD88 tightly within the Halobacterium clade, reflecting its evolutionary history within the genus. Pangenome analysis further highlighted the open nature of the Halobacterium genus, with AD88 contributing novel accessory genes linked to ecological specialization. These findings emphasize the evolutionary significance of the CCB as a natural laboratory for studying microbial adaptation and expand our understanding of archaeal genomic diversity and functional innovation under extreme conditions.

Introduction

Halobacterium salinarum, a gram-negative extremophilic archaeon, is an emerging model organism for studying halophiles due to its ability to persist and grow in high-salinity environments [1, 2]. Its sophisticated mechanisms for ion regulation, osmoregulation, and phototrophic growth exemplify versatility for stress resistance and energy acquisition [3, 4]. Characterized by a distinct red pigmentation attributed to bacteriorhodopsin [5], H. salinarum withstands high salt concentrations and exhibits extraordinary resistance to extreme temperature and radiation [6,7,8]. Its discovery and successful isolation have facilitated studies of their unique genomic and physiological properties, improving our understanding of archaeal adaptation strategies.

Despite its frequent detection in hypersaline environmental samples, easy culturing, and tractability, only a few different H. salinarum isolates have been successfully cultured and maintained under laboratory conditions. Currently, there are only five distinct isolates available in culture collections: H. salinarum NRC 34,001 [9], H. salinarum R1 [10], H. salinarum NRC-1 [11], H. salinarum 91-R6 [12], and H. salinarum KBTZ01 [13]. These isolates exhibit variations in their gene content due to adaptation to specific environments and their unique evolutionary contingencies, contributing to the accessory genome of each one of them. Overall, the genome of Halobacterium salinarum is highly dynamic and polyploid, typically possessing between 10 and 30 copies of its genome per cell [14, 15]. With an average genome size of ~ 2.0 Mbp and high G + C content (60%) [2, 16], Halobacterium sp. NRC-1, generally holds two 191 and 365 kbp-containing mega-plasmids, pNRC100 and pNRC200, respectively, which encode several genes essential for the organism’s survival in saline environments [17,18,19].

Traditional pan-genome analyses employed extensively in well-researched taxa showcase the biogeography of diversity and gene distribution across phylogenetic and environmental correlations [20]. Applying these analyses to species with only a few cultured members, such as H. salinarum, can reveal genetic novelty and potential responses to environmental cues and resource availability. An example of such a response has been observed in the genome of Bacillus coahuilensis from the Cuatro Ciénegas Basin (CCB) [21]. The genomic analysis of B. coahuilensis revealed its ability to produce sulfolipids instead of phospholipids, as it lacks the genes for producing phosphorus-rich teichoic acids and polyanionic teichuronic acids [21].

The Cuatro Cienegas Basin (CCB) within the Chihuahuan Desert, Mexico, is a unique and ecologically significant region featuring extreme and fluctuating conditions, such as low phosphorus, high radiation, alkaline conditions, and high mineral-pool content, resulting in high salinity levels [22,23,24,25]. Recent studies have shown that a particular site at CCB harbors an extensive archaeal diversity. This site, named Archaean Domes (AD) given the microbial mats that form dome-like structures under wet conditions, is particularly rich in members of the Euryarchaeota phylum, which includes halophiles and methanogens, showing a relative abundance of approximately 14% over seven years [25,26,27,28]. Despite reports of a relative abundance of over 30% Euryarchaeota at the site in 2019 [27], obtaining an axenic culture of most organisms within this domain has remained elusive.

We hypothesize that Halobacterium salinarum AD88, isolated from CCB, has acquired genetic adaptations absent in previously studied strains, given the distinct ecological pressures present in the Cuatro Cienegas Basin (CCB) and limited number of cultured H. salinarum strains. Specifically, we expect that the extreme and fluctuating conditions of the CCB, such as low phosphorus availability and high salinity, have driven genomic innovations in AD88, including the evolution of novel metabolic pathways, such as secondary responses to phosphorus depletion, and structural genomic changes.

To test this hypothesis, we conducted a genomic analysis of H. salinarum AD88 and compared its genome to the twelve available Halobacterium strains and twenty-three additional genomes from the Halobacteriales order. Distinctive adaptations to oligotrophy, particularly to phosphorus limitations, were found, much as smaller plasmid sizes and strong evidence of a single genome duplication event, becoming the first diploid Halobacterium salinarum recorded to date.

Materials & methods

Sampling and culture conditions

Microbial mats (~ 5 cm of depth) were collected from the Archaean Domes site at Rancho Pozas Azules, Cuatro Ciénegas, Coahuila, Mexico (26° 49′ 41.7″ N, 102° 01′ 28.7″ O) using sterile Falcon tubes. For primary isolation, a microbial mat sample was mixed with liquid AM2 medium in a 50 mL Falcon tube to create a suspension. AM2 is a custom-designed medium containing (per liter): 21.2 g bacteriological agar, 2.4 g dextrose, 5 g peptone (Oxoid), 5 g yeast extract, 3 g Na₃C₆H₅O₇, 0.2 g CaCl₂•2 H₂O, 50 g MgCl₂•6 H₂O, 20 g MgSO₄•7 H₂O, 2 g KCl, and 250 g NaCl. A 100 µL aliquot of the suspension was plated onto AM2 agar medium using ⌀ 0.5 mm glass beads for 30 s to facilitate sample dispersion. After four weeks of incubation, distinct colonies were observed and selected based on morphological differences. For the secondary isolation phase, each distinct colony was streaked onto ATCC-213 agar medium, which contained (per liter): 20 g bacteriological agar, 2.5 g tryptone, 10 g yeast extract, 0.2 g CaCl₂•6 H₂O, 10 g MgSO₄•7 H₂O, 5 g KCl, and 250 g NaCl. New individual colonies were then re-streaked onto fresh ATCC-213 plates to ensure the isolation of pure cultures. This process was repeated until axenic cultures were obtained, which were subsequently verified through whole-genome sequencing. Both AM2 and ATCC-213 media were adjusted to pH 7.4, and all incubations were carried out at 37 °C. Once the strain was successfully obtained in axenic culture, it was grown in liquid ATCC-213 medium and incubated at 37 °C/ 250 rpm, until reaching exponential phase at day 6 for subsequent analyses, including scanning electron microscopy (SEM) imaging and genomic DNA extraction.

SEM imaging

Cells grew in ATCC-213 liquid media until reaching the exponential phase and prepared for scanning electron microscopy (SEM). A liquid culture was centrifuged in 1.5 ml Eppendorf tubes to obtain a pellet. The pellet was rinsed twice with 1 ml of 0.1 M of cacodylate buffer, for 5 min. The solution was extracted with a pipette in order do not disturb the pellet, which later was resuspended in the fixing solution containing 2.5% glutaraldehyde, 0.1 M cacodylate buffer, 4.7 M NaCl and 80 mmol MgSO₄•7H₂O and incubated for 1 h at room temperature. The homogenized mix was centrifuged to obtain a new pellet and stored at 4 °C overnight. Post-fixing was performed using 1% osmium tetroxide in 0.1 M of cacodylate buffer, dehydration steps were done using 50, 70,80,95,100% EtOH, during 15 min/each, followed by critical point-drying using Tousimis Autosamdri 815, purge timer setting 2, and sputter-coating with 5.19 nm (or 10.35 nm) Pt with Leica EM ACE600 sputter coater. The Hitachi SU-8230 cold field-emission SEM was used. The sample was imaged with 1 kV (or 3 kV or 5 kV) accelerating voltage.

DNA extraction and sequencing

Genomic DNA extraction was performed as follows: Biomass pellets were obtained from agar plates and resuspended in a solution consisting of 400 µL phenol-chloroform and 400 µL QTP lysis buffer. The lysis buffer composition per liter included 40 mL Triton X-100, 100 mL 20% SDS, 20 mL 5 M NaCl, 10 mL 2 M Tris (pH 8), 7 mL EDTA (pH 8), and milliQ water up to 10 mL. The mixture was placed into 2 mL Eppendorf tubes containing 0.2 µL of 0.1 mm Ø glass beads. Tubes were subjected to agitation in a shaker for 1 min, followed by incubation on ice for 1 min. This process was repeated three times. For DNA precipitation, the aqueous phase was transferred to a new 2 mL Eppendorf tube, supplemented with 25 µL sodium acetate, and filled to the top with 95% ethanol. The mixture was incubated at − 20 °C overnight. The tubes were centrifuged, and the DNA pellets were washed twice with 70% ethanol. The gDNA was resuspended in 30 µL of TE buffer containing RNase and stored at − 20 °C. Sequencing was performed at the Microbial Genome Sequencing Center, Pittsburg, PA, using Illumina MiSeq technology, paired-end 2 × 150 bp reads.

Genome assembly and annotation

We filtered the raw sequencing data using FastQC (v0.11.8) [29] and Trimmomatic [30] (v0.39) with the following parameters: ILLUMINACLIP: adapters.fa:2:20:10 LEADING:10 TRAILING:3 SLIDINGWINDOW:4:20 MINLEN:36. Assembly of reads was performed using SPAdes (v3.15.3) [31] with the following parameters spades.py -1 file_R1.fastq − 2 file_R2.fastq. The functional annotation was carried out using the Prokka v1.14.6 [32]. The contig file in FASTA format was prepared as the input for the annotation process with the next options: prokka --kingdom Archaea --outdir output_directory --prefix output_prefix contigs_genome.fasta. Taxonomic identification was performed using GTDB-tk (v2.3.2) [33]. The functional annotation pipeline integrated various bioinformatics tools, including Prodigal [34] for gene prediction, RNAmmer [35] for rRNA identification, and Aragorn [36] for tRNA detection. The resulting annotated genome included information on protein-coding genes, rRNA, tRNA, and other genomic features. To make sure there was not contamination in the culture, we checked genome quality using CheckM v.2 [37] using default parameters (checkm lineage_wf < bin folder > < output folder> ).

For the assembly of plasmids from sequencing data, we used the bioinformatics tool plasmidSPAdes, an extension of the SPAdes assembler optimized specifically for plasmid sequences. The DNA sequencing data obtained were preprocessed to ensure high-quality reads, removing low-quality sequences and potential contaminants. Next, we executed the plasmidspades.py script with the paired end read files (reads_1.fastq and reads_2.fastq) as input and specified an output directory for the assembly results. The command used was: plasmidspades.py -1 reads_1.fq -2 reads_2.fq -o output_directory. The assembly process was conducted in a computational environment with sufficient memory and storage resources to handle the workload. The resulting contigs included representatives of the plasmids present in the sample, which were subsequently analyzed for characterization and functional annotation.

Pan genome analysis

To find possible genomic differences in Halobacterium salinarum AD88 genome with respect to other Halobacteria genomes reported in the NCBI database, we first determined the pangenome of 35 complete sequenced genomes from the Halobacteriales order (Supplementary Material Table 1) using get_homologues.pl software [38] with the OMCL algorithm using the parameters recommended by the developer (C min %coverage in BLAST pairwise alignments (range = 1-100, default = 75), E max E-value (default = 1e-05,max = 0.01), S min % sequence identity in BLAST query/subj pairs (range = 1-100], default = 1 OMCL), N min BLAST neighborhood correlation (range = 0,1,default = 0 OMCL]) [39]. After determining the pangenomic matrix, we filtered out the orthologous gene families only present in our genome with a homemade R script. Functional annotation was carried out with kofamscan [40] of the families of possible orthologous genes of this analysis and were subsequently visualized with the keggdecoder software [41] designed to parse and visualize a KEGG-Koala output to determine the completeness of biogeochemically-relevant metabolic pathways (Supplementary Material Table 2).

BlastP alignments

Homology searches for sulfur pathway and plasmid-encoded genes were performed using BLASTP v2.15.0 [42, 43] with default parameters. A local protein sequence database was constructed from 12 Halobacterium reference genomes and the Halobacterium salinarum AD88 genome using the NCBI BLAST + makeblastdb utility. The query sequences (sulfur pathway and plasmid-encoded proteins) were aligned against the database with the following settings: BLOSUM62 substitution matrix, E-value threshold of 10, gap opening penalty of 1, gap extension penalty of 1, compositional adjustment (comp_based_stats 2), and word size of 3. Results were generated in tabular format (-outfmt 6) and filtered to retain hits with significant sequence identity (Supplementary Material Tables 3 and Table 4).

Phylogenomic inference

For phylogenetic inference we used the species alignment file generated by the OrthoMCL (OMCL) algorithm [44], identifying core homologous genes of 12 Halobacterium genus reference genomes and 23 Halobacteriales reference genomes from NCBI, and implemented the get_homologues software [38]. From the resulting core-genome gene set, only those encoding ribosomal subunits were selected. These sequences were aligned with mafft v7.490 [45] and concatenated into a single matrix. The optimal amino acid substitution model selection, maximum likelihood phylogenetic tree reconstruction, and nodal support analysis using bootstrap (with 1000 replicates) were performed with IQ-TREE v2.1.4 [46]. The resulting topology was graphically visualized and edited using the iTOL program [47].

To evaluate the similarity between the analyzed genomes, we calculated the Average Nucleotide Identity (ANI) using the ANIb method (ANI by BLAST) [48]. This approach allows the determination of nucleotide identity between pairs of complete genomes through local alignments using BLAST+. The analysis was performed with the pyani package [49], v.0.2.9 in Python. The ANI calculation was conducted using the anib module of pyani, which implements bidirectional alignments through BLAST+, v.2.16.0 [50]. All possible pairs between the genomes were compared (A vs. B & B vs. A), and the results included the average nucleotide identity, alignment coverage, and the total aligned length for each pair. Then we compared Halobacterium salinarum AD88 with Halobacterium zhouii (the species with the lowest ANI to AD88) and with Halobacterium salinarum NRC-34,001 (the species with the highest ANI to AD88). This detailed comparison highlights the variation in nucleotide identity across closely related and more distantly related species. For individual synteny comparisons between H. salinarum AD88 and H. salinarum NRC 34,001, and H. salinarum AD88 and H. zhouii XZYTJ26, the ANIb method was implemented using default parameters (k-mer:16, --min_fraction:0.2, --frag_len:3000), plots were visualized in proksee [51].

Ploidy analysis

Previously, it has been reported that some haloarchaea contain multiple copies of their genomes. We applied a bioinformatic approach to determine the ploidy of H. salinarum AD88. For this prediction, GenomeScope 2.0 [52] takes as input the k-mer spectrum, performs a non-linear least-squares optimization to fit a mixture of negative binomial distributions, and outputs estimates for genome size, repetitiveness, and heterozygosity rates. To visualize the results of GenomeScope 2.0, we used Smudgeplots, an approach to visualize genome structure and infer ploidy directly from the k-mers present in sequencing reads. The analyzed genome has an average sequencing depth of 60X, ensuring high accuracy in estimating genomic parameters and supporting robust predictions of ploidy and genome organization.

Results and discussion

A needle in a haystack

The newly discovered ecotype, Halobacterium salinarum AD88, was isolated from a microbial mat (Fig. 1a) found within the Cuatro Cienegas Basin (CCB), in an extensive culturing assay spanning five years. The basin is home to one of the few examples of living stromatolites and microbial mats, holding a broad spectrum of microbial taxa [53]. In the Archaean Domes (AD), a specific vernal pond within the CCB measuring 35 m x 10 m x 20 cm (L x W x D), microbial communities exhibit dynamic structural and functional adaptations in response to water availability [25]. Under hydrated conditions, these communities form a dome-like structure, encapsulating methane gas within the central cavity. Conversely, during dry conditions, the dome-like structures collapse, becoming flattened. Further analysis of their microbial composition highlighted the complexity and diversity of these communities, in which a significant presence of archaea and a diverse array of viruses were identified. Archaea historically accounted for less than 2% of the community, yet recent studies have identified a large archaeal diversity holding not only Euryarchaeota phylum and TACK superphyla, but also Asgard and DPANN superphyla. Halobacterium salinarum, in particular, have been identified within these communities with a 0.03969% relative abundance [25]. The isolation of H. salinarum AD88 exemplifies the power of combining synthetic and experimental approaches, key to study ancient taxa.

Phenotypic characterization

Colonies grown in agar plates were small, smooth and round, displaying the typical red coloration after incubation for 7 days at 37ºC. Most isolates of extremely halophilic archaea form red-colored colonies as a result of C-50 carotenoid pigments secretion, which include bacteriorhodopsins and betaines [54]. Their synthesis is triggered by high-intensity light and osmotic stress, respectively. Under conditions of low oxygen availability, haloarchaea increase the production of bacteriorhodopsins to exploit light energy for generating a proton gradient across the cell membrane, which is then used to synthesize ATP. Scanning Electron Microscopy Imaging revealed characteristic pleomorphic rod-shaped cells (Fig. 1b and c), and standard microscopic analysis showed flagellum-mediated motility, supporting the identification of two flagella coding genes found in one of the assembled plasmids.

Fig. 1

Microbial Structures in CCB. (a) A CCB microbial mat characterized by stratified, colorful layers used for isolation of H. salinarum AD88. The upper most salt crust, composed primarily of halite crystals, often harbors halophilic microorganisms, including extremophilic archaea such as Halobacterium spp. and some halophilic bacteria. Beneath the crust, stratified microbial layers exhibit distinct colors due to the presence of pigments like chlorophylls (green layer), carotenoids (orange-red), and phycobiliproteins, reflecting the diversity of cyanobacteria, purple sulfur bacteria, and other anaerobes in deeper strata, (b) SEM image of a cluster of Halobacterium salinarum AD88 cells. The cells exhibit pleomorphic morphology, including variations in size and shape, a characteristic feature of haloarchaea. This structural adaptability aids survival in extreme saline conditions, facilitating aggregation and biofilm formation, which enhance resource acquisition and stress resistance, (c) Scanning Electron Micrograph (SEM) of a single rod-shaped cell of Halobacterium salinarum AD88, with a measured length of approximately 3.09 μm

Halobacterium salinarum AD88 has smaller plasmid sizes

The assembly of the draft genome obtained 60 scaffolds with an N50 = 374,983, an L50 = 3, a GC content of 66.1% and an estimated size of 2,520,761 bp. The genome has a 99% of completeness and less than 1% of contamination according with CheckM standards (a microbial genome quality assessment tool) [37]. Functional annotation yielded 47 tRNAs, a single copy of the 5 S, 16 S, and 23 S ribosomal genes, and 2,587 coding sequences.

We further identified and assembled two plasmids present in the H. salinarum AD88 genome, displaying smaller plasmid sizes than those that have been reported previously for this species, possibly reflecting adaptation to the CCB environment (Fig. 2) [55]. Plasmid I with a GC content of 63.93%, contains a single contig of 22,589 bp and encodes 25 coding sequences. Plasmid II, consisting of 12 contigs with a total length of 115,924 bp, GC %= 56.82, encodes 118 coding sequences, of which only 50.84% are non-hypothetical proteins. Environmental pressures shape plasmid content, at least as much as they influence plasmid size. However, there is no direct evidence that plasmid size alone is an adaptation mechanism to extreme environments. Plasmid gene composition varies with habitat, carrying different accessory genes reflecting local selective needs [56]. Organisms in long-term stable or nutrient-poor extreme habitats often exhibit genome reduction, loosing non-essential genes to conserve energy. Despite predictions that costly plasmids should be lost, mechanisms like host-plasmid coadaptation (Bouma and Lenski 1988) and gene transfer can maintain them. Thus, smaller plasmid sizes in extreme environments may result from a combination of adaptation and relaxed selection rather than direct selective pressure alone [57].

Studies of plasmids isolated from various archaeal species have shown a great diversity in gene content and innovation in replication strategies [58, 59]. The role of plasmid-borne genes in haloarchaea species goes beyond acquisition of beneficial genes, as these plasmids usually encode genes essential for survival [60]. Host cells are more likely to retain plasmid-encoded genes when these genes provide a higher fitness by positive frequency-dependent selection, despite the natural tendency for plasmids to be lost during cell division (segregation loss) [61, 62].

In this study, the genetic content of H. salinarum AD88 plasmid I reveals potential adaptive responses that are consistent with observed trends in haloarchaeal plasmids, encoding essential elements such as a cysteine desulfurase (sufS) and an accessory transpersulfurase protein (sufE), which form an iron-sulfur carrier complex (EC 2.8.1.7) [63, 64], playing a role in respiration, gene regulation, DNA repair and replication [65]. It also facilitates proper assembly and transfer of Fe-S clusters, which are critical to the function of redox-active enzymes and electron transport chains, particularly in environments where energy acquisition is challenging [66, 67]. A notable aspect of H. salinarum AD88 is that a single copy of sufS gene, the catalytic subunit, is located in Plasmid I, whereas in other H. salinarum strains, these genes are typically found in multiple copies within the chromosome and plasmids (Supplementary Material Table 4). Given that Fe-S cluster biosynthesis is essential for cellular survival in extreme environments, the localization of these genes in AD88 plasmid may represent an adaptive strategy, potentially acquired through horizontal gene transfer. This unique genomic arrangement could provide regulatory advantages or increased genomic plasticity, reinforcing the role of plasmids as reservoirs of adaptive genes in haloarchaea [68]. Additionally, the plasmid contains fla/Che operon genes, key in the response to environmental stimuli through chemotaxis and typically located in the main chromosome [69]. H. salinarum strains are light energy-transducing systems and the encoded proteins enable the microorganism to process and respond to this stimulus through its flagellar motor [70, 71]. CheA regulates the signal transmission, while CheW connects it to sensory receptors that detect light [72]. The methylation system involving CheB and CheR allows H. salinarum to modulate its sensitivity to changing light conditions, ensuring appropriate movement towards light sources [73, 74].

Plasmid II is characterized by a high density of IS (Insertion Sequence) and ISH (Insertion Sequences of Halophiles) elements, the latter are unique to halophilic archaea [75, 76], which vary in sequence and structure but share the common feature of having inverted repeats at their ends and a gene encoding a transposase enzyme [75, 77]. The clustering of IS elements creates hotspots of genetic variability, influencing the expression and regulation of neighboring genes and facilitating horizontal gene transfer, thereby enhancing the organism evolutionary adaptability by inducing mutations and genomic rearrangements [78]. Other Halobacterium strains, such as NRC-1, also display an accumulation of frequent ISH insertions possibly mediating rapid adaptation to environmental stresses [79]. Interestingly, a potential homolog of the COQ5 gene was found within the largest plasmid. This gene catalyzes the only C-methylation involved in the biosynthesis of coenzyme Q in humans and the yeast Saccharomyces cerevisiae [80]. An additional homologue has been reported in E. coli, named UbiE [81]. Both UbiE and COQ5 are members of a family of methyltransferases involved in the biosynthesis of menaquinone and ubiquinone [81]. Members of this clade are widely distributed among bacteria and eukaryotes but are absent in archaea [82]. Future research should address specifically whether this gene performs a similar function as UbiE in E. coli and COQ5 in S. cerevisiae.

Noteworthy, the reduction in plasmid size in AD88, coupled with a high GC content, suggests an evolutionary response to the selective pressures of the CCB environment, where low nutrient availability, particularly the scarcity of phosphorus, could favor smaller, less metabolically demanding plasmids. Larger plasmids, which often carry additional genes, impose metabolic costs that make them energetically costly to maintain under nutrient-limited conditions. These costs can lead to a selective disadvantage, reducing the microorganism’s fitness, as resources that would otherwise be used for growth and survival are diverted to maintain and replicate larger plasmids. A reduction in the reproductive success of individuals harboring larger plasmids, could lead to a decrease in their frequency within the population over time and drive the fixation of smaller plasmid variants. Consequently, smaller plasmids would enable AD88 ecotype to distribute more energy towards essential cellular functions, such as, protein synthesis and replication in response to CCB specific environmental conditions [83, 84].

Fig. 2

Genomic and functional characterization of plasmids in the Halobacterium salinarum AD88 genome. The figure displays circular maps of two plasmids, Plasmid I (22,589 bp) and Plasmid II (115,924 bp), highlighting their genomic structure and annotated genes. The outermost rings represent annotated genes categorized by their functions. Genes associated with antibiotic resistance are labeled in red (e.g., tetracycline resistance gene tet, β-lactam resistance gene OXA, and macrolide resistance genes erm and msr). Mobile genetic elements, such as transposases and insertion sequences, regions involved in horizontal gene transfer, replication and plasmid stability are highlighted in purple, including ori regions and partitioning systems. The concentric circles indicate GC content (black) and GC skew (innermost), visualizing nucleotide composition patterns for each plasmid

Ploidy reduction in Halobacterium salinarum AD88

The genomic analysis showed consistent patterns in the distribution of read coverage and k-mer frequencies, suggesting a ploidy of two in the genome of Halobacterium salinarum AD88 (Fig. 3). The observation that AD88 possesses only two copies of its genome is intriguing given the conventional understanding that H. salinarum strains typically exhibit high levels of genome copy number [85]. The standard expectation within halophiles is the maintenance of multiple copies of their genomes, often ranging from 10 to 25 copies per cell, to cope with the extreme osmotic stress in their natural saline environments [14]. For other strains of H. salinarum, the cellular ploidy undergoes dynamic changes throughout the growth phases. During the exponential phase, fast-growing cells exhibit an average of approximately 25 copies of the chromosome, which decreases to 15 copies during the early stationary phase [14]. This reduction in ploidy is not influenced by variation in strain growth rate, as cultures with a twofold lower growth rate maintain the same chromosome copy number. Similarly, in Haloferax volcanii, the genome copy number remains high during the exponential phase, with an average of 18 copies per cell, but decreases to 10 copies upon entering the stationary phase [14]. The high ploidy level has been attributed to various factors, including the need for increased gene dosage to counteract the deleterious effects of high salt concentrations on DNA stability and replication fidelity [86].

Fig. 3

Smudge plot for the genome of H. salinarum AD88 reveals consistent read coverage and k-mer frequency patterns, suggesting a ploidy of two

Polyploids can have several evolutionary advantages in comparison to monoploid species, such as gene redundancy, allowing mutations to occur in one copy of a gene without losing the original wild-type gene sequence, resistance against conditions that induce double strand breaks (DSB’s), and generally having lower spontaneous mutation rates [87]. A single genome duplication in AD88 strain could be linked to specific adaptations or niche preferences of this particular isolate. For genome duplication to be sustained by natural selection, its phenotypic benefits must outweigh the significant energetic and resource costs [88]. It is possible that the strain has evolved alternative mechanisms to mitigate stress, allowing it to succeed with fewer genome copies compared to other strains or species within the same genus. However, the reduced genome copy number in AD88 raises questions about the trade-offs between genome duplication and cellular fitness in extreme environments. While high genome ploidy may confer advantages regarding genomic robustness and adaptation to saline conditions, it also incurs metabolic costs associated with DNA replication, maintenance, and resource allocation [89, 90]. Therefore, the observed low ploidy level in our strain may reflect a balance between the benefits of genome duplication and the metabolic demands imposed by maintaining multiple copies of the genome.

Metabolic adaptations in response to phosphorus scarcity

Amidst this genetic dynamism, there is a deeper narrative of ecological interconnectedness. Microbial life in CCB is partly a collective eco-evolutionary response to a complex web of abiotic interactions [53]. The scarcity of phosphorus and high mineral content in CCB appear to be the principal factors shaping the genetic landscape, where beneficial mutations are fixed, and the selection of genes involved in nutrient acquisition and metabolic versatility confers new traits that enhance fitness.

While comparing metabolic capabilities of AD88 with other Halobacterium strains, we identified 20 orthologous gene families (Fig. 4). The identification of gene families shared between AD88, and a limited number of reference genomes suggests selective pressure driving the retention of these genes in response to similar environmental conditions. However, evaluating the homology of pathway genes using a 70% sequence identity threshold, did not detect a fully conserved presence of those pathways in Halobacterium sp. GSL 19, H. salinarum NRC 34001, and Halobacterium sp. NRC 1. Nevertheless, the absence of full pathway integrity at this threshold does not necessarily indicate that these strains lack the pathway entirely. It is possible that alternative genes or functionally analogous pathways fulfill the same metabolic role, or that sequence divergence in these strains reduces homology detection despite functional conservation.

Notably, AD88 shared more metabolic similarities to H. salinarum 91-R6 strain than any other, particularly in genes that code for phosphonate and phosphate transporters, and those involved in sulfolipid biosynthesis. Given that the CCB is characterized by persistent phosphorus limitation, it provides an ideal environment where sulfolipid substitution could occur as an adaptive strategy to reduce cellular phosphorus demand while maintaining membrane integrity. Sulfolipids can replace phospholipids in membrane structures, reducing cellular phosphorus demand and offering a selective advantage in environments where phosphorus is scarce [91, 92]. Genetic studies have demonstrated that sulfolipid biosynthesis is upregulated under phosphate limitation and that mutants lacking sulfolipids exhibit growth defects, confirming their essential role in maintaining membrane function [93, 94]. This metabolic adaptation is reminiscent of the findings in Bacillus coahuilensis [21], an endemic bacterial species uniquely adapted to the extreme and oligotrophic conditions encountered in the Cuatro Ciénegas Basin (CCB), where the presence of sulfolipid-related genes was also identified as a critical adaptation for survival. The ability to utilize sulfolipids as a substitute for phospholipids suggests a convergent evolutionary strategy among microorganisms in the basin. Further experimental validation is needed to confirm whether Halobacterium salinarum AD88 substitutes phosphate forms with sulfolipids under phosphorus limitation. Lipidomic analyses comparing membrane composition under different phosphorus conditions, along with transcriptomic and mutant studies, could determine if sulfolipid biosynthesis is actively regulated in response to phosphorus stress.

Fig. 4

Maximum-likelihood phylogenetic tree of 39 conserved ribosomal protein families from the core genome. The tree was inferred using the WAG + F + I + R3 substitution model, selected as the best-fit model via the Bayesian Information Criterion (BIC). Branch support values, derived from the ultrafast bootstrap algorithm with 1,000 replicates

Further functional annotation revealed genes associated with a complete and specialized pathway for assimilatory sulfate reduction, including PAPSS (3’-phosphoadenosine 5’-phosphosulfate synthase), sat (sulfate adenylyltransferase), cysNC (bifunctional enzyme CysN/CysC), cysH (phosphoadenosine phosphosulfate reductase), and cysJ (sulfite reductase - NADPH flavoprotein alpha-component) (Fig. 5). This set of genes encode a complete enzymatic process that spans from the initial activation of sulfate to its final reduction to sulfide, allowing the synthesis of essential sulfur-containing molecules. Unlike dissimilatory reduction, which is primarily for energy production, assimilatory sulfate reduction is an energy-dependent process that requires ATP. This system’s primary function is biosynthetic, incorporating reduced sulfur into cellular components and supporting anabolic processes [95]. To investigate the uniqueness of the assimilatory sulfate reduction genes found in Halobacterium salinarum AD88, we performed BlastP alignments of all sulfur pathway genes against the 12 Halobacterium species used in this study (Supplementary Material Table 3). Our results confirm that AD88 harbors specific distinctive genes, with no full-length homologs identified in other Halobacterium species. While no homologs were found with 100% sequence identity across complete genes, we observed 100% identity in shorter sequence segments (4–5 AA). These findings suggest that while these genes may be unique to AD88 in their entirety, some conserved sequence motifs exist within related species. The absence of full-length homologs does not preclude the possibility of functional or distant homologs within the genus or other haloarchaea.

These unique genetic features represent a distinct sulfur-reduction system, which, along with genes coding for resistance to toxic ions such as arsenic and heavy metals, are thought to be additional strategies to cope with phosphorus depletion. Arsenic reduction systems have been previously identified in other archaeal genomes isolated from CCB, such as Halorubrum sp., found in co-cultured with Marinococcus luteus [96]. Phosphorus and arsenic are chemical analogs commonly found in oligotrophic environments, and they can substitute for each other in biological processes [97]. The ability of proteins to interact with multiple substrates, known as promiscuity, often serves as a foundation for the evolution of new protein functions. Thus, the chemical similarities between arsenate and phosphate allow for cross-reactivity between the two, which could facilitate an evolutionary shift. The cross-reactivity might enable organisms to transition from relying primarily on phosphate for metabolic processes to using arsenate, or the reverse, as environmental conditions demand. This flexibility in substrate use may have driven evolutionary changes, allowing AD88 to adapt to an environment with varying levels of these two compounds [98, 99]. Overall, these adaptations represent significant evidence of the evolutionary pressures that shape the microbial communities in CCB in general and the AD site in particular, facilitating their survival through genetic flexibility, and demonstrating a remarkable example of how environmental stress can drive genomic innovations and diversification.

Fig. 5

Sulfur metabolism pathways in a Halobacterium salinarum AD88. The figure illustrates sulfate uptake via the CysPUWA transporter and its subsequent activation to APS and PAPS. Assimilatory sulfate reduction converts sulfate to sulfide for amino acid biosynthesis (e.g., L-cysteine and L-serine), while dissimilatory pathways reduce sulfate to sulfide for energy generation. Intermediates like thiosulfate and trithionate are processed by specific enzymes, integrating sulfur metabolism with broader pathways, including carbon fixation and amino acid synthesis. The oxidation states of sulfur compounds are highlighted throughout the pathways

Evolutionary relationships within Haloarchaea

We conducted a phylogenomic reconstruction to position the newly identified strain within an evolutionary context and assess its relationship to established Halobacteriales taxa (Fig. 4). Analyzing their evolutionary distances provided crucial insights into the genetic divergence and relationships among halophilic archaea. The phylogenomic reconstruction reveals that Halobacterium salinarum AD88 clusters tightly alongside laboratory strains like NRC-1 and R1 [10], displaying ultrashort branches and reflecting recent divergence from a common ancestor. The close genetic relationships within Halobacterium clade suggest that recent evolutionary pressures have maintained high genetic similarity, likely due to similar environmental conditions and adaptive strategies. We propose that their evolutionary adaptations have been driven more by microenvironmental pressures not involving large-scale genetic changes within a relatively recent evolutionary timeframe, which might explain their close phylogenetic clustering. This highlights the importance of integrating phylogenetic, genomic, and ecological data to better understand microevolution in halophilic organisms.

Meanwhile, the deeper evolutionary divergences observed with other species points to the long-term evolutionary isolation and specialization of lineages that have adapted to distinct ecological niches, like Haloarcula, Halococcoides and Halodesulfurarcheum genera. From a broader haloarchaeal evolution perspective, the clear separation of Halobacterium from other genera (like Haloferax, Haloarcula,) reinforces that haloarchaea have diversified into multiple lineages that, despite sharing extreme halophily, have long independent evolutionary histories​. Such insights can inform how haloarchaeal species may have radiated into different ecological niches, since some clusters consist of strains isolated from similar habitats (salt mines, salterns, etc.), indicating that geographic or environmental factors played a role in their divergence [100].

The tree topology also identifies Halobacterium zhouii as one of the more basal (early-branching) members of the genus​, indicating it diverged earlier relative to others, this pattern is consistently observed in other phylogenies [101, 102]. This aligns with our nucleotide identity analysis, which shows that H. zhouii exhibits the lowest synteny levels compared to AD88, further supporting its early divergence within the genus. A more detailed analysis of nucleotide identity levels, aimed refine the genetic relationships among Halobacterium strains, is discussed later in this study.

Pangenome openness provides evidence of ongoing genetic diversification

The genome comparison between 13 Halobacterium strains, including H. salinarum AD88, suggests that this genus has an open pangenome (Fig. 6) [103]. We observe that new gene families are continuously incorporated as more strains or species are discovered, with each new genome within Halobacteria bringing ~ 137 new genes to the pool of pan-genes [104]. In halophilic archaea, such as Halobacterium salinarum, the open pangenome structure may be an adaptive mechanism, enabling the species to prosper in hypersaline environments through extensive gene acquisition and loss. This characteristic is often seen in organisms exposed to dynamic and heterogeneous environments, as genetic flexibility allows for continuous adaptation and survival [105].

Pan genome analysis identified 3,744 homologous gene groups, of which 1,072 genes form part of the core genome, including those necessary for fundamental biological processes, indicating a shared evolutionary foundation [106, 107]. Within the core (%) genome we found genes related to DNA replication and a photoreactivation repair system, such as photolyases coding genes phr1 and phr2, which directly reverse UV-induced DNA damage, specifically pyrimidine dimers, ensuring the maintenance of genome integrity [108]. Additionally, 1,325 genes were categorized as softcore, defined as a set of genes found in 95% of the genomes [109]. Meanwhile, 1,182 were identified as shell genes, which are remaining moderately conserved genes, and present at intermediate frequencies in 15–95% of the strains [38, 110]. Both softcore and shell genes show the genetic variability within the genus, suggesting that different species have evolved unique traits for adaptation to the specific environments they were isolated from [111]. Surprisingly, ars operon genes arsA, arsD, arsR2, and arsM were identified as part of the shell genome, and not in any of the extrachromosomal replicons, as they usually occur in pNRC100 in Halobacterium sp. NRC-1 [112]. Last, we found 1,236 cloud genes, which appear in less than 15% of the strains and could indicate a reservoir of genetic diversity that may be involved in niche specialization [111, 113]. As part of the cloud genome, we found the pleiotropic regulator bat gene, which controls light-sensing mechanisms and bacteriorhodopsin and retinal chromophore synthesis. We also observed copA and copB genes involved in metal homeostasis and regulate copper transport, sufS involved in the biosynthesis of iron-sulfur clusters and contributes to survival under varying redox conditions, and CRISPR-Cas associated genes, including cas1, cas2, and cas6.

The high accessory genome in Halobacterium species shows the adaptive importance of genetic exchange mechanisms, particularly horizontal gene transfer (HGT), which is well-documented in haloarchaea and facilitated by genomic island mobility and homologous recombination [105, 114]. These mechanisms have been pivotal in haloarchaeal evolution, enabling transitions from autotrophic anaerobes, as seen in methanogenic ancestors [115] to aerobic lifestyles through gene acquisitions from bacterial sources like Deinococcus radiodurans and Bacillus species [11, 114]; Moreover, HGT has played a continuous and critical role in adaptation by providing haloarchaea access to new genetic variation, enabling adaptations that overcome evolutionary limitations. While cumulative mutations may trap species in local adaptive peaks, unable to explore better fitness options, HGT between haloarchaeal species introduces new genes far more rapidly than would occur via mutation without HGT [116]. HGT opens up evolutionary pathways that would be inaccessible, offering new adaptive opportunities and helping reverse deleterious mutations that may accumulate during periods of purifying selection [116].

Fig. 6

Pangenome and core-genome analysis of the studied strains. The pangenome was determined using the OrthoMCL algorithm, and the distribution of orthologous families is visualized in the Upset plot. The core-genome, observed in the first bar of the Upset plot, consists of 1,073 orthologous families. It was defined as the set of genes present in all strains and identified through the consensus of three algorithms: OrthoMCL (OMCL), Bidirectional Best Hits (BDBH), and COGtriangles

Given the close phylogenomic relationship between Halobacterium salinarum AD88 and other Halobacterium strains, we conducted this comparative analysis to gain deeper insights into their genetic divergence and evolutionary proximity. This approach allowed us to further resolve their relationships at a finer scale, identifying potential genomic variations that contribute to their differentiation within the genus. When comparing the genomes of different Halobacterium strains, distinct patterns emerge at the nucleotide identity levels (Fig. 7a). This comparative analysis illustrates how genomic data align with evolutionary and taxonomic distinctions, emphasizing that average nucleotide identity (ANI) values can serve as reliable indicators of genetic relatedness and evolutionary history. ANI values above 95% indicate strains are thought to be part of the same species, and intra-species ANI values tend to cluster closely around this threshold, especially in stable environments. Comparisons between Halobacterium salinarum strains show high ANI values ranging between 98.6 − 99.9%, indicating a high degree of genomic conservation. Within the same species, H. salinarum AD88 shows the highest similarity to H. salinarum NRC 34001, an isolate from a Canadian salted buffalo hide, with an ANI of 99.4%. The high synteny is represented by the nearly continuous alignment of homologous regions, with minimal interruptions or translocations (Fig. 7b). Such conserved synteny suggests that these strains have undergone limited structural genomic changes since diverging from a common ancestor, likely due to similar environmental pressures that shaped their evolutionary trajectories and preservation of gene function and regulatory networks crucial for survival​​.

At the interspecies level, AD88 shows high similarity to Halobacterium sp. NRC-1 (ANI of 99.1%) isolated from a stable, rich salt flat of Utah, where consistent high salinity and nutrient availability differ markedly from CCB’s dynamic ecosystem. The high similarities suggests shared adaptations for thriving in salt-rich habitats, likely contributing to their genomic overlap. However, it remains intriguing that despite their different isolation sources, they maintain such a degree of nucleotide identity. This similarity emphasizes how shared environmental stressors, such as high salinity, can lead to parallel adaptations through convergent evolution, even when the specific ecological contexts differ.

In contrast, AD88 displays significant nucleotide differences when compared to Halobacterium zhouii XZYJT26, isolated from saline soil of Mangkang ancient solar saltern in Tibet, China, showing only 80% identity, and fragmented synteny characterized by multiple breaks, rearrangements, and scattered homologous regions (Fig. 7b). The nucleotide differences can be attributed to distinct geographic and climate conditions; while the Tibetan region has a high-altitude, cold, and arid climate, CCB is characterized by a warm, semi-arid environment, fostering unique evolutionary divergences that distinguish microbial strains. The observed evolutionary divergence, rooted in environmental and taxonomic differences, is closely tied to broader mechanisms of genome evolution, such as, changes in gene organization. As microbes evolve, their genomes undergo rearrangements, such as inversions, transpositions, and duplications [117]. These processes disrupt the original order of genes. As a result, only a small fraction of genes retain the same arrangement (synteny) between distantly related organisms, which tend to increase with greater sequence divergence [118]. However, certain gene clusters maintain synteny even across distantly related organisms, which can signal that these genes are functionally conserved to ensure efficiency in specific niches. When synteny is preserved over large evolutionary distances, it likely reflects strong selection pressure to maintain the arrangement, indicating the importance of these genes, as their order and proximity are critical for their function, due to co-regulation, participation in the same metabolic pathways, and physical interaction in protein complexes [119].

Fig. 7

Average Nucleotide Identity (ANI) Analysis Among Halobacterium Strains. (a) The heatmap shows the ANI values between various strains of Halobacterium genus. Strains of the same species, such as Halobacterium salinarum exhibit high ANI values, reflecting significant genomic conservation, while comparisons between strains from different species, indicates greater evolutionary divergence, (b) The synteny plots of H. salinarum AD88 and related strains illustrate a continuum of evolutionary divergence, where conserved synteny and rearranged regions mirrors the dual demands of stability and innovation. While conserved synteny between H. salinarum AD88 and H. salinarum NRC 34001 are important for maintaining functional core pathways, rearranged regions between H. salinarum AD88 and H. zhouii XZYTJ26 give the evolutionary flexibility required for niche specialization

Conclusion

Using an approach that includes culture-dependent and culture-independent techniques, we describe the genomic characteristics of the Halobacterium salinarum AD88 strain. Striking differences compared to other Halobacteria were uncovered, as its genomic architecture is diploid and with smaller novel plasmids. Additionally, genomic analysis suggests that AD88 may compensate for phosphorus scarcity through an enhanced reliance on sulfur metabolism, including sulfolipid biosynthesis pathways that replace phospholipids in membrane composition. This shift toward sulfur-based adaptations would support the hypothesis that nutrient availability plays a fundamental role in shaping the evolutionary trajectory of extreme halophiles. Phylogenomic analysis confirmed AD88’s placement within the Halobacterium clade, revealing its close evolutionary relationship with other H. salinarum strains. While its positioning in the tree suggests a high degree of similarity, nucleotide sequence identity levels reveal specific genetic differences that distinguish it from closely related strains. These findings highlight AD88 as a genetically distinctive strain and raise questions about the evolutionary forces that shaped its divergence, prompting a reevaluation of our understanding of genome dynamics in halophilic archaea. While phylogenomic methods provide a useful framework for distinguishing species-level differences in archaea, they fail to capture the full complexity of genomic variations. Comparative genomic studies across diverse archaeal taxa have revealed patterns of conservation and divergence in core genes, genes of unknown function, and mobile genetic elements, shaped by both phylogenetic and ecological pressures. Investigating the role of conserved and accessory genes in facilitating ecological specialization and maintaining population structures across evolutionary gradients could illuminate the mechanisms regulating archaeal speciation, offering new insights into how genomic diversity and functional adaptations arise and are maintained in extreme environments. Further exploration of these dynamics through integrative approaches, combining genomics, transcriptomics, and environmental data would advance our ability to refine archaeal taxonomy and deepen our understanding of their evolutionary trajectories.