This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, reproduction and adaptation in any medium and for any purpose provided that it is properly attributed. For attribution, the original author(s), title, publication source (PeerJ Preprints) and either DOI or URL of the article must be cited.
Solar salterns are excellent model ecosystems for studying virus-microbial interactions because of their low microbial diversity, environmental stability, and high viral density. By using the power of CRISPR spacers to link viruses to their prokaryotic hosts, we explored virus-host interactions in geographically diverse salterns and related them to carbon cycling. Using taxonomic profiling, we identified hosts such as archaeal Haloquadratum, Halorubrum, and Haloarcula and bacterial Salinibacter, and we found that community composition related to not only salinity but also local environmental dynamics. Characterizing glycerol metabolism genes in these metagenomes suggested Halorubrum and Haloquadratum possess most dihydroxyacetone kinase genes while Salinibacter possesses most glycerol-3-phosphate dehydrogenase genes. We identified CRISPR spacers in the metagenomes with two different methods and found more spacers in the IC21 and C34 salterns compared with the SS19, SS33, and SS37 salterns, suggesting fewer types of CRISPR spacers in the Haloquadratum-majority salterns. After CRISPR detection, spacers were aligned against haloviral genomes to map virus to host. While most alignments linked viruses to Haloquadratum walsbyi, there were groups of interactions with the low abundance taxa Haloarcula and Haloferax. Further examination of the dinucleotide and trinucleotide usage differences between paired viruses and their hosts confirmed viruses and hosts had similar nucleotide usage signatures. Detection of cas genes in the salterns supported the possibility of CRISPR activity. Taken together, our studies suggest similar virus-host interactions exist in different solar salterns and that the glycerol metabolism gene dihydroxyacetone kinase is associated with Haloquadratum and Halorubrum.
This is a revision of this article that was recently submitted for a second round of review at PeerJ.
Flow charts outlining analyses performed in each part of the methods
The sections covered include profiling taxonomic and functional compositions of selected metagenomes, de novo and reference-guided CRISPR detection, mapping virus-host interactions with CRISPR spacers, comparative analysis of virus and host sequence characteristics, and assembling metagenomes and detecting halobacterial cas genes in metagenomic contigs.
Number of CRISPR spacers detected by de novo and reference guided methods, normalized by millions of reads in each metagenome
Spacers were detected either with the de novo detection method (Crass) or the reference-guided method (MetaCRAST) with corresponding maximum edit distances (from 0 to 3) described in Materials and Methods that used a query of 29 halobacterial CRISPR direct repeat sequences. All spacer counts are reported after clustering initially detected spacers with CD-HIT (with a clustering similarity threshold of 0.9).
Network of virus-host interactions generated from spacers detected in the genome of Haloquadratum walsbyi strain C23
The library of haloviral genomes screened is listed in Table S1. Nodes represent either viruses or spacers, while edges represent BLAST alignments linking spacers to viruses. Viruses are marked in orange and spacers in blue. Visualization was performed with Cytoscape.
Taxonomic affiliation of all BLAST hits of cas genes against contigs (not only best hits)
Detected contigs assembled with Newbler and Velvet were aligned against a library of cas genes from three halobacterial orders (Halobacteriales, Haloferacales, and Natrialbales). The taxonomic affiliations of all hits for each contig were tabulated into profiles for each order and assembly method.
The library included 326 contigs more than 500 bp in length assembled from Chula Vista metaviromes (http://data.imicrobe.us/project/view/58) and the following haloviral genomes obtained from NCBI GenBank.