With a pinch of salt: metagenomic insights into Namib Desert salt pan microbial mats and halites reveal functionally adapted and competitive communities

Abstract

The hyperarid Namib Desert is one of the oldest deserts on Earth. It contains multiple clusters of playas which are saline-rich springs surrounded by halite evaporites. Playas are of great ecological importance, and their indigenous (poly)extremophilic microorganisms are potentially involved in the precipitation of minerals such as carbonates and sulfates and have been of great biotechnological importance. While there has been a considerable amount of microbial ecology research performed on various Namib Desert edaphic microbiomes, little is known about the microbial communities inhabiting its multiple playas. In this work, we provide a comprehensive taxonomic and functional potential characterization of the microbial, including viral, communities of sediment mats and halites from two distant salt pans of the Namib Desert, contributing toward a better understanding of the ecology of this biome.

Keywords: CRISPR-Cas; functional diversity; gene transfer agent; halite; horizontal gene transfer; microbial mat; playa; salt pan; taxonomic diversity; virus-host interactions.

Fig 1

Map of the central Namib Desert with the sampling location distribution. It clearly shows the extensive northern gravel plains and the southern dune fields. The red triangles indicate the location of the two sampled playas—Hosabes (Ho) and Eisfeld (Ei)—and the black dot represents the Gobabeb—Namib Research Institute. Inset photographs depict the salt pan streams of Hosabes (A) and Eisfeld (B) and a close-up of the collected samples: microbial mat (C) and the dark (D) and red (E) halite rocks. Image adapted from ESA, CC BY-SA 3.0 IGO.

Fig 2

Namib Desert microbial community mat and halite sample diversity. (A) Principal component plot of the Namib stream mat and halite samples based on gene taxonomy at order level obtained using the DESeq2 package. Taxonomic diversity was compared by location (e.g., Eisfeld or Hosabes) and sublocation (e.g., source, sink, or halite). (B) Relative taxonomic classification of the metagenomic open reading frames. Proteobacteria dominate the salt pan samples, while members of the Euryarchaea are predominant in both halites. H, Hosabes; E, Eisfeld; α, Alphaproteobacteria; β, Betaproteobacteria; γ, Gammaproteobacteria; δ, Deltaproteobacteria; Hb, Halobacteriaceae; Hf, Haloferacaceae; N, Natrialbaceae.

Fig 3

Functional potential of the Namib salt pan and halite microbial communities. Panels depict the carbon (A), nitrogen (B), and sulfur (C) cycles from the mat (left) and halite (right) metagenomic data. Black arrows represent steps of the cycle present in the community, while gray arrows represent pathways not detected in the metagenomes. Thickness of arrows is proportional to the marker gene counts for each pathway. Main taxa with the genetic potential for each metabolic step are shown in italics.

Fig 4

CRISPR-Cas type I and type III systems are widespread in the Namib Desert salt pan microbial communities. (A) Relative abundance of prokaryotic defense system KO terms in the Namib Desert stream mat and halite metagenomes. T-A, toxin-antitoxin; R-M, restriction modification; DNA-PT, DNA phosphorothioation. (B) Relative abundance of prokaryotic defense systems annotated using the PADLOC. The graph indicates the percentage of the total defense loci identified. (C) Relative abundance of CRISPR-Cas systems in each of the metagenomic data sets. (D) Phylogenetic tree of the Cas9 protein sequences present in the Namib salt pan metagenomes. Colored branches indicate a reference protein sequence obtained from the RefSeq protein database, while black branches indicate protein sequences obtained from the metagenomic data. Sequences belonging to the three described Cas9 categories are shaded in purple (II-A), green (II-C), or orange (II-B). The symbols at the end of the branch indicate the location within the salt pan from where the metagenomic sequence was obtained (red, Eisfeld; yellow, Hosabes; blue, halite; circle, source; square, sink).

Fig 5

Genome-based network of shared protein content. Each node represents a viral genome, and edges represent statistically significant relationships between the protein profiles of those viral genomes. Groups composed exclusively of RefSeq viruses were excluded for clarity. Viral clusters of interest are indicated with the prefix “VC” followed by their corresponding number and/or circled in black. Inset: network of Caudoviricetes indicating the largest viral clusters with mVir sequences.

Fig 6

Overview of the genomic organization of GTA-like and virus-like contigs from mVir clusters with GTA-signal. (A) Comparison of Rhodobacter capsulatus GTA (RcGTA) to the largest contig of VC_69 and GTA-like or virus-like contigs from VC_89. (B) Comparison of VC_255 contigs to Roseobacter phage RDJL Phi1 virus. (C) Genomic organization of VC_309 contigs. The shades of red represent pairwise protein similarity, with stronger red as the most similar. Genes are shown as boxes with an arrow indicating their orientation in the genome and colored according to their assigned functional category: blue, virus structure and assembly; orange, DNA replication; green, transcription; purple, virus exit; yellow, defense mechanisms; red, integration; pink, metabolism.