IMG/VR v.2.0: an integrated data management and analysis system for cultivated and environmental viral genomes

A Review on Viral Metagenomics in Extreme Environments

Viruses are the most abundant biological entities in the biosphere, and have the ability to infect Bacteria, Archaea, and Eukaryotes. The virome is estimated to be at least ten times more abundant than the microbiome with 107 viruses per milliliter and 109 viral particles per gram in marine waters and sediments or soils, respectively. Viruses represent a largely unexplored genetic diversity, having an important role in the genomic plasticity of their hosts. Moreover, they also play a significant role in the dynamics of microbial populations. In recent years, metagenomic approaches have gained increasing popularity in the study of environmental viromes, offering the possibility of extending our knowledge related to both virus diversity and their functional characterization. Extreme environments represent an interesting source of both microbiota and their virome due to their particular physicochemical conditions, such as very high or very low temperatures and >1 atm hydrostatic pressures, among others. Despite the fact that some progress has been made in our understanding of the ecology of the microbiota in these habitats, few metagenomic studies have described the viromes present in extreme ecosystems. Thus, limited advances have been made in our understanding of the virus community structure in extremophilic ecosystems, as well as in their biotechnological potential. In this review, we critically analyze recent progress in metagenomic based approaches to explore the viromes in extreme environments and we discuss the potential for new discoveries, as well as methodological challenges and perspectives.

A Phage Protein Aids Bacterial Symbionts in Eukaryote Immune Evasion

Phages are increasingly recognized as important members of host-associated microbiomes, with a vast genomic diversity. The new frontier is to understand how phages may affect higher order processes, such as in the context of host-microbe interactions. Here, we use marine sponges as a model to investigate the interplay between phages, bacterial symbionts, and eukaryotic hosts. Using viral metagenomics, we find that sponges, although massively filtering seawater, harbor species-specific and even individually unique viral signatures that are taxonomically distinct from other environments. We further discover a symbiont phage-encoded ankyrin-domain-containing protein, which is widely spread in phages of many host-associated contexts including human. We confirm in macrophage infection assays that the ankyrin protein (ANKp) modulates the eukaryotic host immune response against bacteria. We predict that the role of ANKp in nature is to facilitate coexistence in the tripartite interplay between phages, symbionts, and sponges and possibly many other host-microbe associations.

The Virome of Cerebrospinal Fluid: Viruses Where We Once Thought There Were None

Traditionally, medicine has held that some human body sites are sterile and that the introduction of microbes to these sites results in infections. This paradigm shifted significantly with the discovery of the human microbiome and acceptance of these commensal microbes living across the body. However, the central nervous system (CNS) is still believed by many to be sterile in healthy people. Using culture-independent methods, we examined the virome of cerebrospinal fluid (CSF) from a cohort of mostly healthy human subjects. We identified a community of DNA viruses, most of which were identified as bacteriophages. Compared to other human specimen types, CSF viromes were not ecologically distinct. There was a high alpha diversity cluster that included feces, saliva, and urine, and a low alpha diversity cluster that included CSF, body fluids, plasma, and breast milk. The high diversity cluster included specimens known to have many bacteria, while other specimens traditionally assumed to be sterile formed the low diversity cluster. There was an abundance of viruses shared among CSF, breast milk, plasma, and body fluids, while each generally shared less with urine, feces, and saliva. These shared viruses ranged across different virus families, indicating that similarities between these viromes represent more than just a single shared virus family. By identifying a virome in the CSF of mostly healthy individuals, it is now less likely that any human body site is devoid of microbes, which further highlights the need to decipher the role that viral communities may play in human health.

CRISPR-Cas System of a Prevalent Human Gut Bacterium Reveals Hyper-targeting against Phages in a Human Virome Catalog

Bacteriophages are abundant within the human gastrointestinal tract, yet their interactions with gut bacteria remain poorly understood, particularly with respect to CRISPR-Cas immunity. Here, we show that the type I-C CRISPR-Cas system in the prevalent gut Actinobacterium Eggerthella lenta is transcribed and sufficient for specific targeting of foreign and chromosomal DNA. Comparative analyses of E. lenta CRISPR-Cas systems across (meta)genomes revealed 2 distinct clades according to cas sequence similarity and spacer content. We assembled a human virome database (HuVirDB), encompassing 1,831 samples enriched for viral DNA, to identify protospacers. This revealed matches for a majority of spacers, a marked increase over other databases, and uncovered "hyper-targeted" phage sequences containing multiple protospacers targeted by several E. lenta strains. Finally, we determined the positional mismatch tolerance of observed spacer-protospacer pairs. This work emphasizes the utility of merging computational and experimental approaches for determining the function and targets of CRISPR-Cas systems.

Metagenomic Analysis of the Diversity of DNA Viruses in the Surface and Deep Sea of the South China Sea

A metagenomic analysis of the viral community from five surface and five deep sea water (>2000 m below the surface, mbs) samples collected from the central basin of the South China Sea and adjacent Northwest Pacific Ocean during July-August 2017 was conducted herein. We builded up a South China Sea DNA virome (SCSV) dataset of 29,967 viral Operational Taxonomic Units (vOTUs), which is comparable to the viral populations from the original Tara Ocean and Malaspina expeditions. The most abundant and widespread viral populations were from the uncultivated viruses annotated from the viral metagenomics. Only 74 and 37 vOTUs have similarity with the reported genomes from the cultivated viruses and the single-virus genomics, respectively. The community structures of deep sea viromes in the SCSV were generally different from the surface viromes. The carbon flux and nutrients (PO₄ and NOx) were related to the surface and deep sea viromes in the SCSV, respectively. In the SCSV, the annotated vOTUs could be affiliated to the cultivated viruses mainly including Pelagibacter (SAR11) phage HTVC010P, Prochlorococcus phages (P-GSP1, P-SSM4, and P-TIM68), Cyanophages (MED4-184 and MED4-117) and Mycobacterium phages (Sparky and Squirty). It indicated that phage infection to the SAR11 cluster may occur ubiquitously and has significant impacts on bathypelagic SAR11 communities in the deep sea. Meanwhile, as Prochlorococcus is prominently distributed in the euphotic ocean, the existence of their potential phages in the deep sea suggested the sedimentation mechanism might contribute to the formation of the deep sea viromes. Intriguingly, the presence of Mycobacterium phages only in the deep sea viromes, suggests inhabitance of endemic viral populations in the deep sea viromes in the SCSV. This study provided an insight of the viral community in the South China Sea and for the first time uncovered the deep sea viral diversity in the central basin of the South China Sea.

Diversity patterns of bacteriophages infecting Aggregatibacter and Haemophilus species across clades and niches

Aggregatibacter and Haemophilus species are relevant human commensals and opportunistic pathogens. Consequently, their bacteriophages may have significant impact on human microbial ecology and pathologies. Our aim was to reveal the prevalence and diversity of bacteriophages infecting Aggregatibacter and Haemophilus species that colonize the human body. Genome mining with comparative genomics, screening of clinical isolates, and profiling of metagenomes allowed characterization of 346 phages grouped in 52 clusters and 18 superclusters. Less than 10% of the identified phage clusters were represented by previously characterized phages. Prophage diversity patterns varied significantly for different phage types, host clades, and environmental niches. A more diverse phage community lysogenizes Haemophilus influenzae and Haemophilus parainfluenzae strains than Aggregatibacter actinomycetemcomitans and “Haemophilus ducreyi”. Co-infections occurred more often in “H. ducreyi”. Phages from Aggregatibacter actinomycetemcomitans preferably lysogenized strains of specific serotype. Prophage patterns shared by subspecies clades of different bacterial species suggest similar ecoevolutionary drivers. Changes in frequencies of DNA uptake signal sequences and guanine–cytosine content reflect phage-host long-term coevolution. Aggregatibacter and Haemophilus phages were prevalent at multiple oral sites. Together, these findings should help exploring the ecoevolutionary forces shaping virus-host interactions in the human microbiome. Putative lytic phages, especially phiKZ-like, may provide new therapeutic options.

Characterization of the virome of Paracoccus spp. (Alphaproteobacteria) by combined in silico and in vivo approaches

Bacteria of the genus Paracoccus inhabit various pristine and anthropologically-shaped environments. Many Paracoccus spp. have biotechnological value and several are opportunistic human pathogens. Despite extensive knowledge of their metabolic potential and genome architecture, little is known about viruses of Paracoccus spp. So far, only three active phages infecting these bacteria have been identified. In this study, 16 Paracoccus strains were screened for the presence of active temperate phages, which resulted in the identification of five novel viruses. Mitomycin C-induced prophages were isolated, visualized and their genomes sequenced and thoroughly analyzed, including functional validation of their toxin-antitoxin systems. This led to the identification of the first active Myoviridae phage in Paracoccus spp. and four novel Siphoviridae phages. In addition, another 53 prophages were distinguished in silico within genomic sequences of Paracoccus spp. available in public databases. Thus, the Paracoccus virome was defined as being composed of 66 (pro)phages. Comparative analyses revealed the diversity and mosaicism of the (pro)phage genomes. Moreover, similarity networking analysis highlighted the uniqueness of Paracoccus (pro)phages among known bacterial viruses.

A Viral Ecogenomics Framework To Uncover the Secrets of Nature's "Microbe Whisperers"

Microbes drive critical ecosystem functions and affect global nutrient cycling along with human health and disease. They do so under strong constraints exerted by viruses, which shape microbial communities’ structure and shift host cell metabolism during infection.

Microbes drive critical ecosystem functions and affect global nutrient cycling along with human health and disease. They do so under strong constraints exerted by viruses, which shape microbial communities’ structure and shift host cell metabolism during infection. While the majority of viruses and their associated impacts remain poorly characterized, a number of mechanisms by which viruses alter microbial cells and ecosystems have already been revealed. Here I outline how a comprehensive host-resolved mapping of viral sequence space will enable a thorough characterization of virus-encoded mechanisms for microbial manipulation. With soon-to-be millions of virus genomes obtained from metagenomes, one of the major challenges resides in the development of methods for high-throughput and high-resolution virus-host pairing, before multi-omics approaches can be leveraged to fully decipher virus-host dynamics in nature. Beyond novel fundamental biological knowledge, these studies will likely provide new molecular tools enabling a precise engineering of microbial cells and communities.

Taxonomic assignment of uncultivated prokaryotic virus genomes is enabled by gene-sharing networks

Microbiomes from every environment contain a myriad of uncultivated archaeal and bacterial viruses, but studying these viruses is hampered by the lack of a universal, scalable taxonomic framework. We present vConTACT v.2.0, a network-based application utilizing whole genome gene-sharing profiles for virus taxonomy that integrates distance-based hierarchical clustering and confidence scores for all taxonomic predictions. We report near-identical (96%) replication of existing genus-level viral taxonomy assignments from the International Committee on Taxonomy of Viruses for National Center for Biotechnology Information virus RefSeq. Application of vConTACT v.2.0 to 1,364 previously unclassified viruses deposited in virus RefSeq as reference genomes produced automatic, high-confidence genus assignments for 820 of the 1,364. We applied vConTACT v.2.0 to analyze 15,280 Global Ocean Virome genome fragments and were able to provide taxonomic assignments for 31% of these data, which shows that our algorithm is scalable to very large metagenomic datasets. Our taxonomy tool can be automated and applied to metagenomes from any environment for virus classification.

The Promises and Pitfalls of Machine Learning for Detecting Viruses in Aquatic Metagenomes

Tools allowing for the identification of viral sequences in host-associated and environmental metagenomes allows for a better understanding of the genetics and ecology of viruses and their hosts. Recently, new approaches using machine learning methods to distinguish viral from bacterial signal using k-mer sequence signatures were published for identifying viral contigs in metagenomes. The promise of these content-based approaches is the ability to discover new viruses, with no or few known relatives. In this perspective paper, we examine the use of the content-based machine learning tool VirFinder for the identification of viral sequences in aquatic metagenomes and explore the possibility of using ecosystem-focused models targeted to marine metagenomes. We discuss the impact of the training set composition on the tool performance and the current limitation for the retrieval of low abundance viral sequences in metagenomes. We identify potential biases that could arise from machine learning approaches for viral hunting in real-world datasets and suggest possible avenues to overcome them.

Enterococcus faecalis Countermeasures Defeat a Virulent Picovirinae Bacteriophage

Enterococcus faecalis is an opportunistic pathogen that has emerged as a major cause of nosocomial infections worldwide. Many clinical strains are indeed resistant to last resort antibiotics and there is consequently a reawakening of interest in exploiting virulent phages to combat them. However, little is still known about phage receptors and phage resistance mechanisms in enterococci. We made use of a prophageless derivative of the well-known clinical strain E. faecalis V583 to isolate a virulent phage belonging to the Picovirinae subfamily and to the P68 genus that we named Idefix. Interestingly, most isolates of E. faecalis tested—including V583—were resistant to this phage and we investigated more deeply into phage resistance mechanisms. We found that E. faecalis V583 prophage 6 was particularly efficient in resisting Idefix infection thanks to a new abortive infection (Abi) mechanism, which we designated Abiα. It corresponded to the Pfam domain family with unknown function DUF4393 and conferred a typical Abi phenotype by causing a premature lysis of infected E. faecalis. The abiα gene is widespread among prophages of enterococci and other Gram-positive bacteria. Furthermore, we identified two genes involved in the synthesis of the side chains of the surface rhamnopolysaccharide that are important for Idefix adsorption. Interestingly, mutants in these genes arose at a frequency of ~10⁻⁴ resistant mutants per generation, conferring a supplemental bacterial line of defense against Idefix.

The 26th annual Nucleic Acids Research database issue and Molecular Biology Database Collection

The 2019 Nucleic Acids Research (NAR) Database Issue contains 168 papers spanning molecular biology. Among them, 64 are new and another 92 are updates describing resources that appeared in the Issue previously. The remaining 12 are updates on databases most recently published elsewhere. This Issue contains two Breakthrough articles, on the Virtual Metabolic Human (VMH) database which links human and gut microbiota metabolism with diet and disease, and Vibrism DB, a database of mouse brain anatomy and gene (co-)expression with sophisticated visualization and session sharing. Major returning nucleic acid databases include RNAcentral, miRBase and LncRNA2Target. Protein sequence databases include UniProtKB, InterPro and Pfam, while wwPDB and RCSB cover protein structure. STRING and KEGG update in the section on metabolism and pathways. Microbial genomes are covered by IMG/M and resources for human and model organism genomics include Ensembl, UCSC Genome Browser, GENCODE and Flybase. Genomic variation and disease are well-covered by GWAS Catalog, PopHumanScan, OMIM and COSMIC, CADD being another major newcomer. Major new proteomics resources reporting here include iProX and jPOSTdb. The entire database issue is freely available online on the NAR website (https://academic.oup.com/nar). The NAR online Molecular Biology Database Collection has been updated, reviewing 506 entries, adding 66 new resources and eliminating 147 discontinued URLs, bringing the current total to 1613 databases. It is available at http://www.oxfordjournals.org/nar/database/c.

The Promises and Pitfalls of Machine Learning for Detecting Viruses in Aquatic Metagenomes

Tools allowing for the identification of viral sequences in host-associated and environmental metagenomes allows for a better understanding of the genetics and ecology of viruses and their hosts. Recently, new approaches using machine learning methods to distinguish viral from bacterial signal using k-mer sequence signatures were published for identifying viral contigs in metagenomes. The promise of these content-based approaches is the ability to discover new viruses, with no or few known relatives. In this perspective paper, we examine the use of the content-based machine learning tool VirFinder for the identification of viral sequences in aquatic metagenomes and explore the possibility of using ecosystem-focused models targeted to marine metagenomes. We discuss the impact of the training set composition on the tool performance and the current limitation for the retrieval of low abundance viral sequences in metagenomes. We identify potential biases that could arise from machine learning approaches for viral hunting in real-world datasets and suggest possible avenues to overcome them.