Doolittle, W. Pattern pluralism and the Tree of Life hypothesis. Natl Acad. USA , — The ancient virus world and evolution of cells. Direct 1 , 29 Podolsky, S. The role of the virus in origin-of-life theorizing. Greuet, C. Taylor, F. Google Scholar. The 1. Science , — Giant viruses, giant chimeras: the multiple evolutionary histories of Mimivirus genes. BMC Evol. Article Google Scholar. Lecointre, G. Species sampling has a major impact on phylogenetic inference.
Lartillot, N. A Bayesian mixture model for across-site heterogeneities in the amino-acid replacement process. Download references. You can also search for this author in PubMed Google Scholar. Reprints and Permissions.
Yet viruses cannot be included in the tree of life. Nat Rev Microbiol 7, — Download citation. Issue Date : August Anyone you share the following link with will be able to read this content:. Sorry, a shareable link is not currently available for this article.
Provided by the Springer Nature SharedIt content-sharing initiative. Advanced search. Sign up for the Nature Briefing newsletter — what matters in science, free to your inbox daily. Skip to main content Thank you for visiting nature. Download PDF. You have full access to this article via your institution. Figure 1: Taxon-rich phylogenetic tree of clamp loader proteins. Full size image. References 1 Moreira, D. Google Scholar 15 Raoult, D. Article Google Scholar 17 Lecointre, G.
View author publications. Rights and permissions Reprints and Permissions. Copy to clipboard. Search Search articles by subject, keyword or author. Show results from All journals This journal. Close banner Close. Advances in sequencing technology and comparative genomics have expanded our understanding of the evolutionary relationships between viruses and cellular organisms.
Genomic and metagenomic data have revealed that co-evolution between viral and cellular genomes involves frequent horizontal gene transfer and the occasional co-option of novel functions over evolutionary time.
From the giant, ameba-infecting marine viruses to the tiny Porcine circovirus harboring only two genes, viruses and their cellular hosts are ecologically and evolutionarily intertwined. When deciding how, if, and where viruses should be placed on the ToL, we should remember that the Tree functions best as a model of biological evolution on Earth, and it is important that models themselves evolve with our increasing understanding of biological systems.
While this adaptive strategy can be linked to theoretical models on the emergence of parasitism Koonin et al. In fact, there is evidence to suggest that many novel genes originate in viruses and that gene flow between viruses and their hosts is dominated by host acquisition of viral genes Forterre and Prangishvili, These three hypotheses on viral origins do not cover the full breadth of possibilities.
They do, however, provide a good framework from which to interpret the outpouring of results from comparative genomic analyses that focus on ancient evolutionary events. Some viruses have only a handful of genes, while others have hundreds.
DNA viruses generally have more genes than RNA viruses and, within each of these categories, ds viruses tend to have more genes than ss. Giant viruses containing thousands of genes were first discovered in They are the largest members of the phylum Nucleocytoviricota that multiply within molecular virus factories in the host cytoplasm and they primarily infect species of ameba. Their gene repertoire includes informational genes formerly thought to be exclusive to cells, a finding that led to a rethinking of the very notion of viruses Brandes and Linial, Given what is known about HGT, it is sensible to ask if these informational genes were acquired from cellular hosts or have a more ancient origin, perhaps predating the common ancestor of all modern cells.
The latter scenario promotes the virus-first hypothesis or, alternatively, the reduction hypothesis where giant viruses evolved to lose the full genetic toolkit required for independent existence, gradually adapting to a parasitic lifestyle. Virus-first pushes Nucleocytoviricota back to a pre-cellular origin, while reduction sees them evolving from a primitive cell that existed before LUCA Moelling and Broecker, Nucleocytoviricota were proposed as a fourth domain of life in A phylogenetic tree was built from a subset of informational genes, showing this group of viruses to be clearly distinct from Bacteria, Archaea, and Eukarya.
Some translational genes were predicted to have been horizontally transferred from eukaryotes, suggesting a complex genetic history of ancient vertical transmission accompanied by HGT from other domains of life Boyer et al. Forterre et al. The fourth domain hypothesis was later criticized for a failure to account for non-phylogenetic signals in the sequence data. Williams et al. This finding was later backed up by phylogenomic analyses showing giant viruses evolving multiple times from smaller Nucleocytoviricota ancestors.
While Nucleocytoviricota do not form a separate domain, recent evidence suggests they played an important role in the evolution of modern eukaryotes. This idea is supported by a recent analysis of eight conserved proteins in Nucleocytoviricota that splits the phylum into two superclades and suggests that two transfers of DNA-dependent RNAP happened, one from each clade, from ancestral giant viruses to proto-eukaryotes Guglielmini et al.
These studies further highlight the influence of viruses on the evolution of cellular lineages. The story of giant viruses is reminiscent of the difficulties of studying the ancient past by means of information rooted in the present.
Inferring ancient evolutionary events from modern molecular data is like walking a tightrope, finding a balance between being too careful, and missing the opportunity to advance novel concepts.
The dismissal of a viral supergroup tells a similar story. Protein FSFs are shared between viruses and cells, suggesting distant common ancestry. The abundance of these FSFs distributed across cellular lifeforms and the seven Baltimore viral classes was used in a phylogenomic exploration of viral origins and evolution. The results mistakenly suggested that all viruses originated as a supergroup from a primitive cell before the existence of the common ancestor of all modern cells. This is an interesting concept that has been questioned by more recent analyses that highlight systemic errors biasing the outcome and interpretation of a viral supergroup.
Harish et al. There is therefore no viral supergroup that originated as a monophyletic clade from primitive cells. Likely, viruses did not evolve just once. Viral supergroups and extra domains present simplistic scenarios where viruses remain largely separated from the evolution of their cellular hosts. But viruses can be viewed more as a strategy and less as a single lineage that originated in a single time and place.
They are more likely to have a multifaceted history, fully embraced by the concepts of the biological revolution brought about by genomics and HGT. There is no reason to assume that a strategy as successful as virion production arose just once. There is no universal gene that ties all viruses together in a phylogenetic framework. This was once possible only for closely related viruses, but virus gene-sharing networks have shown that the virosphere is more connected than previously thought.
Iranzo et al. The network consisted of 19 modules, forming five major and three minor supermodules. Eleven of these modules included tailed bacteriophages Caudovirales , highlighting the diversity of these viruses. They also discovered 14 viral hallmark genes VHGs , which accounted for most of the inter-module connections. These hallmark genes included essential structural proteins and those involved in virus replication.
Two major capsid proteins double jelly roll and the HKlike acted as network hubs for the two largest supermodules: 1 HKlike: tailed bacteriophages and herpesviruses Figure 5 and 2 double jelly roll: the putative order Megavirales and smaller viruses, as well as polintons, which are large DNA transposons Iranzo et al.
In a separate study, Bin Jang et al. This finding is not surprising given the exponential increase in available virus genomes and the immense genetic diversity of the virosphere. Figure 5. Taken from Iranzo et al. Internal structure of the Caudovirales supermodule. A bipartite graph is displayed, linking bacterial modules to the genes they share. Proteins that represent hub nodes in the gene-sharing network are labeled. License: CC Attribution 4.
Viral hallmark genes tell us that viruses have a global organization, even if every species of virus cannot be brought together in a single phylogenetic model. More intriguing still is the presence of a so-called palm domain that is also found in the reverse transcriptase enzymes of RNA and DNA retroviruses. There is evidence that these enzymes form a monophyletic group, covering five of the seven Baltimore classes of viruses as well as group II introns, a large family of retroelements that multiply by splicing in and out of bacterial DNA Koonin et al.
The RNA viruses, including the two classes of retroviruses, have been elevated to a new taxonomic rank, the realm Riboviria, in a recently proposed megataxonomy of the virus world Koonin et al. Alternatively, they might be the descendants of ancient cells that have no modern counterparts. More specifically, the replication modules of RNA viruses predated LUCA, but this says nothing about their capsids and related structural proteins—the origin and evolution of viral capsids tells a different story.
Numerous capsid-like structures are present in cells. A good example of these is bacterial microcompartments BMCs. BMCs form shells that compartmentalize certain biochemical reactions in the cytoplasm. They are composed of two shell proteins, BMC-H and BMC-P, that form an icosahedral assembly bearing a striking morphological resemblance to viral capsids. The similarity ends there, however, as neither protein shares structural similarity with viral capsid proteins.
Current evidence suggests a cellular origin of BMCs and, indeed, the recruitment by viruses of many cellular structural proteins Krupovic and Koonin, It is hypothesized, based on conserved protein structures, that the SJR-CP was derived from ancestral cellular carbohydrate- or nucleotide-binding proteins. The protein was co-opted by a parasitic RNA replicator that likely behaved much like plasmids or transposases do today. The combination of a replication module with a structural module gave rise to the first modern viruses.
It is becoming clear that the evolutionary histories of viruses and other MGEs are inseparable. It is also clear that cellular life has not evolved separately from the genetic parasites that have evolved to exploit it. The origin of RNA viruses is currently explained by a hybrid of two hypotheses, virus-first and escape, where the replication module has a virus-first origin and the structural module has an escape origin Figure 6.
What about the origin of DNA viruses? Do they tell a similar story of conflicting evolutionary histories? Nowhere is this more revealing than for the multiple, chimeric origins of ssDNA viruses. Three lineages of ssDNA viruses—inoviruses, pleolipoviruses, and microviruses—evolved independently from RCRE plasmids by co-opting a filamentous, polymorphic, and SJR capsid protein, respectively.
A virus with plasmid origins, coupled with the co-option of existing viral capsids from a different type of replicator is a fascinating scenario. A plasmid ancestor begs an obvious question: if there is a place for viruses on the ToL, why not plasmids too? We can logically ask the same question about other selfish replicators such as DNA transposases, which, after all, show distant homology to viral sequences Iranzo et al.
Figure 6. The evolution of modern plasmids and transposons from ancient MGEs is also depicted. Single-stranded DNA viruses have recently been given a realm of their own: Monodnaviria. Despite a highly variable number of genes, and evidence of HGT between the two supergroups, they have been split into two realms, Varidnaviria and Duplodnaviria, suggesting an ancient and independent origin for both realms Koonin et al.
RNA replication and reverse transcription are unique to viruses and MGEs, except for cases of co-option and subsequent adaptation into cellular processes Koonin et al. It is therefore likely that these types of replication existed before LUCA, back in the primordial world. It is interesting to note that two dsDNA virus groups, papillomaviruses and polyomaviruses, originated from a ssDNA ancestor Kazlauskas et al.
It appears that the viral strategy of genome propagation is an ongoing experiment among biological entities. It is obvious that there is no single branch into which viruses can be placed. It is likely that many viruses are a hybrid of genes from divergent lineages, existing both before and after the emergence of LUCA Koonin et al.
Viruses also played major roles in the origin and evolution of numerous cellular lineages, perhaps even driving the emergence of the three cellular domains of life Forterre, Where does this leave us with viruses and their place on the ToL?
Accepting the ToL as a dynamic model of the evolution of biological entities on Earth, viruses should rightly be included in these models. The question then becomes not if viruses have a place on the ToL, but how and where should they be placed?
It is a difficult task, however, to put these principles into practice. A comprehensive, digital representation of all these trees removes the network component because the evolutionary history of each gene is treated separately. This is somewhat simplistic since genes can gain and lose domains over time Nasir et al. Separate gene trees hardly paint a complete picture of evolution though since genes interact and often replicate together within cells.
A network structure could still be added to the FoL by connecting individual gene trees or specific gene tips to each other to represent higher level organization at the genome, organismal, or species level. In principle, such a digital forest with network components could record both the vertical transmission of genes within lineages and the horizontal transmission of genes across lineages. For example, the lytic and lysogenic lifecycles of viruses and their association with host species and genomes could be represented Figure 7.
It is important to note that models are only useful if they can be used to answer questions. For many purposes, the standard species ToL might be perfectly sufficient. For understanding the co-evolution of viruses and other MGEs with cellular life, such a tree is inadequate.
Figure 7. A simple diagram of the phylogenetic forest of gene trees with network components linking different genes together. The diagram shows four gene lineages, two viral and two cellular.
Arrows connect genes present on the same genome to their respective viral or bacterial species, while the lysogenic virus is connected to species A, showing that it is a prophage whose genes reside on the same genome as species A.
Genetic parasites are an inevitable outcome of replicator systems Koonin et al. Conserved, ancient structural proteins have revealed an entangled evolutionary history of all MGEs. Krupovic et al. Whatever the details of the multiple origins and evolutions of viruses, there is no reason to exclude them from our models of biological evolution on Earth. The difficult part will be to build these models to competently represent the origin and co-evolution of viruses with cellular life.
Life is the outcome of billions of years of experimentation on a planetary scale, with processes that we are only beginning to fathom, and the outcomes appear to be an almost unlimited number of dynamic strategies for replicators to exist and multiply in the world. We have discovered the importance of HGT in genome evolution only relatively recently, yet its molecular basis likely predated the evolution of cooperation. Genetic cooperation and genome organization were therefore preceded by selfishly splicing replicators Dawkins, It is also possible, although speculative, that all modern MGEs are descendants of ancient replicators that existed before cooperative behavior.
The scientific community will never fully agree on the living nature of viruses and other MGEs. We favor an open-minded view in this article, but we think the living nature of viruses does not ultimately matter as much as the fact that they are evolving biological entities that have co-evolved with cellular life and engaged in regular HGT with their hosts, likely playing pivotal roles in cellular evolution.
Our understanding of life is limited but growing. We need dynamic and evolving models that can answer our questions about the nature of biology. In this review, we argue that viruses should be included in future models of biological evolution—models that have historically been represented by the ToL. These models will need to be digital and multi-dimensional in nature.
They will also be very difficult to create. But every model of reality is necessarily a construct. Our brains themselves are comprised of cooperating and competing neuronal modules and sub-modules Rutishauser et al. What we know from the scientific method is that some views are less false than others. Both authors conceived of the review topic and layout, wrote the final draft of the manuscript, and approved the final version. HH wrote a complete draft of the manuscript. The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.
Arinkin, V. Jump ahead with a twist: DNA acrobatics drive transposition forward. Bapteste, E. Do orthologous gene phylogenies really support tree-thinking? BMC Evol. Benner, S. Defining life. Astrobiology 10, — Google Scholar. Berche, P. Louis Pasteur, from crystals of life to vaccination. Berkemer, S. A new analysis of archaea-bacteria domain separation: variable phylogenetic distance and the tempo of early evolution.
Bernhardt, H. The RNA world hypothesis: the worst theory of the early evolution of life except for all the others a. Direct Bin Jang, H. Taxonomic assignment of uncultivated prokaryotic virus genomes is enabled by gene-sharing networks.
Boussau, B. Parallel adaptations to high temperatures in the Archaean eon. Nature , — Boyer, M. Phylogenetic and phyletic studies of informational genes in genomes highlight existence of a 4 domain of life including giant viruses. PLoS One 5:e Braga, L. Bacterial diversification in the light of the interactions with phages: the genetic symbionts and their role in ecological speciation.
Brandes, N. Giant viruses-big surprises. Viruses The not so universal tree of life or the place of viruses in the living world. B Biol. Burkhardt, R. Lamarck, evolution, and the inheritance of acquired characters. Genetics , — Callanan, J. Expansion of known ssRNA phage genomes: from tens to over a thousand.
Castelle, C. Catchpole, R. The evolution of reverse gyrase suggests a nonhyperthermophilic last universal common ancestor. Chakravorty, S. A detailed analysis of 16S ribosomal RNA gene segments for the diagnosis of pathogenic bacteria. Methods 69, — Chiang, Y. Genetic transduction by phages and chromosomal islands: the new and noncanonical. PLoS Pathog. Cleland, C. Life without definitions.
Synthese , — Life Evol. Darwin, C. London: John Murray. Dawkins, R. The Selfish Gene. Viruses as a survival strategy in the armory of life. Life Sci. Dion, M.
Phage diversity, genomics and phylogeny. Doolittle, W.
0コメント