DNA virus


A DNA virus is a virus that has DNA as its genetic material and replicates using a DNA-dependent DNA polymerase. The nucleic acid is usually double-stranded DNA but may also be single-stranded DNA. DNA viruses belong to either Group I or Group II of the Baltimore classification system for viruses. Single-stranded DNA is usually expanded to double-stranded in infected cells. Although Group VII viruses such as hepatitis B contain a DNA genome, they are not considered DNA viruses according to the Baltimore classification, but rather reverse transcribing viruses because they replicate through an RNA intermediate. Notable diseases like smallpox, herpes, and the chickenpox are caused by such DNA viruses.

Group I: dsDNA viruses

organization within this group varies considerably. Some have circular genomes while others have linear genomes. Some families have circularly permuted linear genomes. Others have linear genomes with covalently closed ends.
A virus infecting archaea was first described in 1974. Several others have been described since: most have head-tail morphologies and linear double-stranded DNA genomes. Other morphologies have also been described: spindle shaped, rod shaped, filamentous, icosahedral and spherical. Additional morphological types may exist.
Orders within this group are defined on the basis of morphology rather than DNA sequence similarity. It is thought that morphology is more conserved in this group than sequence similarity or gene order which is extremely variable. Three orders and 31 families are currently recognised. A fourth order—Megavirales—for the nucleocytoplasmic large DNA viruses has been proposed. This proposal has yet to be ratified by the ICTV. Four genera are recognised that have not yet been assigned a family.
Fifteen families are enveloped. These include all three families in the order Herpesvirales and the following families: Ascoviridae, Ampullaviridae, Asfarviridae, Baculoviridae, Fuselloviridae, Globuloviridae, Guttaviridae, Hytrosaviridae, Iridoviridae, Lipothrixviridae, Nimaviridae and Poxviridae.
Bacteriophages belonging to the families Tectiviridae and Corticoviridae have a lipid bilayer membrane inside the icosahedral protein capsid and the membrane surrounds the genome. The crenarchaeal virus Sulfolobus turreted icosahedral virus has a similar structure.
The genomes in this group vary considerably from ~10 kilobases to over 2.5 megabases in length. The largest bacteriophage known is Klebsiella Phage vB_KleM-RaK2 which has a genome of 346 kilobases.
The virophages are a group of viruses that infect other viruses.
A virus with a novel method of genome packing infecting species of the genus Sulfolobus has been described. As this virus does not resemble any known virus it has been classified into a new family, the Portogloboviridae.
Another Sulfolobus infecting virus—Sulfolobus ellipsoid virus 1—has been described. This enveloped virus has a unique capsid and may be classified into a new taxon.

Host range

Species of the order Caudovirales and of the families Corticoviridae and Tectiviridae infect bacteria.
Species of the order Ligamenvirales and the families Ampullaviridae, Bicaudaviridae, Clavaviridae, Fuselloviridae, Globuloviridae, Guttaviridae, Tristromaviridae and Turriviridae infect hyperthermophilic archaea species of the Crenarchaeota.
Species of the order Herpesvirales and of the families Adenoviridae, Asfarviridae, Iridoviridae, Papillomaviridae, Polyomaviridae and Poxviridae infect vertebrates.
Species of the families Ascovirus, Baculovirus, Hytrosaviridae, Iridoviridae and Polydnaviruses and of the genus Nudivirus infect insects.
Species of the family Mimiviridae and the species Marseillevirus, Megavirus, Mavirus virophage and Sputnik virophage infect protozoa.
Species of the family Nimaviridae infect crustaceans.
Species of the family Phycodnaviridae and the species Organic Lake virophage infect algae. These are the only known dsDNA viruses that infect plants.
Species of the family Plasmaviridae infect species of the class Mollicutes.
Species of the family Pandoraviridae infect amoebae.
Species of the genus Dinodnavirus infect dinoflagellates. These are the only known viruses that infect dinoflagellates.
Species of the genus Rhizidiovirus infect stramenopiles. These are the only known dsDNA viruses that infect stramenopiles.
Species of the genus Salterprovirus and Sphaerolipoviridae infect species of the Euryarchaeota.

Taxonomy

The classification of DNA viruses has undergone a revision by the International Committee on the Taxonomy of Viruses in 2020.
Three realms are recognised: Duplodnaviria, Monodnaviria and Varidnaviria. The Duplodnaviria contain several types of double stranded DNA viruses. The Monodnaviria all the single stranded viruses. The Varidnaviria are a taxon of viruses united by the presence of use of vertically-folded jelly roll capsid.
Within the Duplodnaviria one kingdom and two phlya are recognised. The phylum Peploviricota has one order - Herpesvirales. The phylum Uroviricota has one order - Caudovirales.
Within the Monodnaviria four kingdoms are recognised.
Within the Varidnaviria two kingdoms are recognised.
A group of double stranded DNA viruses have been found in fish that appear to be related to the herpesviruses.
Another group of viruses that infect fish has been described.

NCLDVs

The asfarviruses, iridoviruses, mimiviruses, phycodnaviruses and poxviruses have been shown to belong to a single group,—the large nuclear and cytoplasmic DNA viruses. These are also abbreviated "NCLDV". This clade can be divided into two groups:
It is probable that these viruses evolved before the separation of eukaryoyes into the extant crown groups. The ancestral genome was complex with at least 41 genes including the replication machinery up to four RNA polymerase subunits at least three transcription factors capping and polyadenylation enzymes the DNA packaging apparatus and structural components of an icosahedral capsid and the viral membrane.
The evolution of this group of viruses appears to be complex with genes having been gained from multiple sources. It has been proposed that the ancestor of NCLDVs has evolved from large, virus-like DNA transposons of the Polinton/Maverick family. From Polinton/Maverick transposons NCLDVs might have inherited the key components required for virion morphogenesis, including the major and minor capsid proteins, maturation protease and genome packaging ATPase.
Another group of large viruses—the Pandoraviridae—has been described. Two species—Pandoravirus salinus and Pandoravirus dulcis—have been recognized. These were isolated from Chile and Australia respectively. These viruses are about one micrometer in diameter making them one of the largest viruses discovered so far. Their gene complement is larger than any other known virus to date. At present they appear to be unrelated to any other species of virus.
An even larger genus, Pithovirus, has since been discovered, measuring about 1.5 µm in length. Another virus—Cedratvirus—may be related this group.

Pleolipoviruses

A group known as the pleolipoviruses, although having a similar genome organisation, differ in having either single or double stranded DNA genomes. Within the double stranded forms have runs of single stranded DNA. These viruses have been placed in the family Pleolipoviridae. This family has been divided in three genera: Alphapleolipovirus, Betapleolipovirus and Gammapleolipovirus.
These viruses are nonlytic and form virions characterized by a lipid vesicle enclosing the genome. They do not have nucleoproteins. The lipids in the viral membrane are unselectively acquired from host cell membranes. The virions contain two to three major structural proteins, which either are embedded in the membrane or form spikes distributed randomly on the external membrane surface.
This group includes the following viruses:
Although bacteriophages were first described in 1927, it was only in 1959 that Sinshemer working with phage Phi X 174 showed that they could possess single-stranded DNA genomes. Despite this discovery, until relatively recently it was believed that most DNA viruses contained double-stranded DNA. Recent work, however, has shown that single-stranded DNA viruses can be highly abundant in seawater, freshwater, sediments, terrestrial and extreme environments, as well as metazoan-associated and marine microbial mats. Many of these "environmental" viruses belong to the family Microviridae. However, the vast majority has yet to be classified and assigned to genera and higher taxa. Because most of these viruses do not appear to be related, or are only distantly related to known viruses, additional taxa will have to be created to accommodate them.
;Archaea
Although ~50 archaeal viruses are known, all but two have double stranded genomes. These two viruses have been placed in the families Pleolipoviridae and Spiraviridae

Taxonomy

The taxonomy of DNA viruses was significantly revised in 2020. All the single stranded DNA viruses have been placed in a single realm - Monodnaviria. Within this realm four kingdoms are recognised.
A division of the circular single stranded viruses into four types has been proposed. This division seems likely to reflect their phylogenetic relationships.
Type I genomes are characterized by a small circular DNA genome, with the Rep protein and the major open reading frame in opposite orientations. This type is characteristic of the circoviruses, geminiviruses and nanoviruses.
Type II genomes have the unique feature of two separate Rep ORFs.
Type III genomes contain two major ORFs in the same orientation. This arrangement is typical of the anelloviruses.
Type IV genomes have the largest genomes of nearly 4-kb, with up to eight ORFs. This type of genome is found in the Inoviridae and the Microviridae.
Given the variety of single stranded viruses that have been described this scheme—if it is accepted by the ICTV—will need to be extended.

CRESS DNA viruses

All known eukaryotic ssDNA viruses have icosahedral capsids. With the exception of the family Bidnaviridae and Anelloviridae, all eukaryotic ssDNA viruses encode homologous rolling-circle replication initiation proteins with characteristic N-terminal endonuclease domains and C-terminal superfamily three helicase domains. A name for this group of viruses has been proposed—circular Rep-encoding single-strand DNA viruses. It has been proposed that CRESS-DNA viruses have evolved from bacterial plasmids, from which they inherited the Rep genes.

Cruciviridae

A group of ssDNA viruses whose Rep proteins show homology to ssDNA viruses from the families Geminiviridae, Circoviridae, and Nanoviridae, while their coat protein is related to those of ssRNA viruses from the family Tombusviridae and unclassified oomycete-infecting viruses. The name Cruciviridae has been proposed for this group.

Host range

The families Bidnaviridae and Parvoviridae have linear genomes while the other families have circular genomes. The Bidnaviridae have a two part genome and infect invertebrates. The Inoviridae and Microviridae infect bacteria; the Anelloviridae and Circoviridae infect animals ; and the Geminiviridae and Nanoviridae infect plants. In both the Geminiviridae and Nanoviridae the genome is composed of more than a single chromosome. The Bacillariodnaviridae infect diatoms and have a unique genome: the major chromosome is circular : the minor chromosome is linear and complementary to part of the major chromosome. Members of the Spiraviridae infect archaea. Members of the Genomoviridae infect fungi.

Molecular biology

All viruses in this group require formation of a replicative form—a double stranded DNA intermediate—for genome replication. This is normally created from the viral DNA with the assistance of the host's own DNA polymerase.

Recently classified viruses

In the 9th edition of the viral taxonomy of the ICTV the Bombyx mori densovirus type 2 was placed in a new family—the Bidnaviridae on the basis of its genome structure and replication mechanism. This is currently the only member of this family but it seems likely that other species will be allocated to this family in the near future.
A new genus—Bufavirus—was proposed on the basis of the isolation of two new viruses from human stool. Another member of this genus—megabat bufavius 1—has been reported from bats. The human viruses have since been renamed Primate protoparvovirus and been placed in the genus Protoparvovirus.
Another proposed genus is Pecovirus. These are similar in organisation to the Smacovirus but share little sequence similarity.
;Genomoviridae
The most recently introduced family of ssDNA viruses is the Genomoviridae.
The family includes 9 genera, namely Gemycircularvirus, Gemyduguivirus, Gemygorvirus, Gemykibivirus, Gemykolovirus, Gemykrogvirus, Gemykroznavirus, Gemytondvirus and Gemyvongvirus.
The genus name Gemycircularvirus stands for Gemini-like myco-infecting circular virus. the type species of the genus Gemycircularvirus—Sclerotinia sclerotiorum hypovirulence associated DNA virus 1—is currently the only cultivated member of the family. The rest of genomoviruses are uncultivated and have been discovered using metagenomics techniques.
Another genus has been proposed—Gemybolavirus.
;Human isolates
Isolates from this group have also been isolated from the cerebrospinal fluid and brains of patients with multiple sclerosis.
An isolate from this group has also been identified in a child with encephalitis.
Viruses from this group have also been isolated from the blood of HIV+ve patients.
;Animal isolates
Ostrich faecal associated ssDNA virus has been placed in the genus Gemytondvirus. Rabbit faecal associated ssDNA virus has been placed in the genus Gemykroznavirus.
Another virus from this group has been isolated from mosquitoes.
Ten new circular viruses have been isolated from dragonfly larvae. The genomes range from 1628 to 2668 nucleotides in length. These dragonfly viruses have since been placed in the Gemycircularviridae.
Additional viruses from this group have been reported from dragonflies and damselflies.
;Plants and fungi
Three viruses in this group have been isolated from plants.
A virus—Cassava associated circular DNA virus—that has some similarity to Sclerotinia sclerotiorum hypovirulence associated DNA virus 1 has been isolated. This virus has been placed in the Gemycircularviridae.
Some of this group of viruses may infect fungi.

''[Smacoviridae]''

A new family, the Smacoviridae, has been created for a number of single-stranded DNA viruses isolated from the faeces of various mammals. Smacoviruses have circular genomes of ~2.5 kilobases and have a Rep protein and capsid protein encoded in opposite orientations. 43 species have been included in this family which includes six genera—Bovismacovirus, Cosmacovirus, Dragsmacovirus, Drosmacovirus, Huchismacovirus and Porprismacovirus.

Unassigned species

A number of additional single stranded DNA viruses have been described but are as yet unclassified.

Human isolates

Viruses in this group have been isolated from other cases of encephalitis, diarrhoea and sewage.
Two viruses have been isolated from human faeces—circo-like virus Brazil hs1 and hs2—with genome lengths of 2526 and 2533 nucleotides respectively. These viruses have four open reading frames. These viruses appear to be related to three viruses previously isolated from waste water, a bat and from a rodent.
This appears to belong to a novel group.
A novel species of virus—human respiratory-associated PSCV-5-like virus—has been isolated from the respiratory tract. The virus is approximately 3 kilobases in length and has two open reading frames—one encoding the coat protein and the other the DNA replicase. The significance—if any—of this virus for human disease is unknown presently.

Animal viruses – vertebrates

An unrelated group of ssDNA viruses, also discovered using viral metagenomics, includes the species bovine stool associated circular virus and chimpanzee stool associated circular virus. The closest relations to this genus appear to be the Nanoviridae but further work will be needed to confirm this. Another isolate that appears to be related to these viruses has been isolated from pig faeces in New Zealand. This isolate also appears to be related to the pig stool-associated single-stranded DNA virus. This virus has two large open reading frames one encoding the capsid gene and the other the Rep gene. These are bidirectionally transcribed and separated by intergenic regions. Another virus of this group has been reported again from pigs. A virus from this group has been isolated from turkey faeces. Another ten viruses from this group have been isolated from pig faeces. Viruses that appear to belong to this group have been isolated from other mammals including cows, rodents, bats, badgers and foxes.
Another virus in this group has been isolated from birds.
Fur seal feces-associated circular DNA virus was isolated from the faeces of a fur seal in New Zealand. The genome has 2 main open reading frames and is 2925 nucleotides in length. Another virus—porcine stool associated virus 4—has been isolated. It appears to be related to the fur seal virus.
Two viruses have been described from the nesting material yellow crowned parakeet —Cyanoramphus nest-associated circular X virus and Cyanoramphus nest-associated circular K virus Both viruses have two bidirectional open reading frames. Within these are the rolling-circle replication motifs I, II, III and the helicase motifs Walker A and Walker B. There is also a conserved nonanucleotide motif required for rolling-circle replication. CynNCKV has some similarity to the picobiliphyte nano-like virus and CynNCXV has some similarity to the rodent stool associated virus.
A virus with a circular genome—sea turtle tornovirus 1—has been isolated from a sea turtle with fibropapillomatosis. It is sufficiently unrelated to any other known virus that it may belong to a new family. The closest relations seem to be the Gyrovirinae. The proposed genus name for this virus is Tornovirus.
Another faecal virus—feline stool-associated circular DNA virus—has been described.

Animal viruses – invertebrates

Among these are the parvovirus-like viruses. These have linear single-stranded DNA genomes but unlike the parvoviruses the genome is bipartate. This group includes Hepatopancreatic parvo-like virus and Lymphoidal parvo-like virus. A new family Bidensoviridae has been proposed for this group but this proposal has not been ratified by the ICTV to date. Their closest relations appear to be the Brevidensoviruses.
A virus—Acheta domesticus volvovirus—has been isolated from the house cricket. The genome is circular, has four open reading frames and is 2,517 nucleotides in length. It appears to be unrelated to previously described species. The genus name Volvovirus has been proposed for these species. The genomes in this genus are ~2.5 nucleotides in length and encode 4 open reading frames.
Two new viruses have been isolated from the copepods Acartia tonsa and Labidocera aestiva—Acartia tonsa copepod circo-like virus and Labidocera aestiva copepod circo-like virus respectively.
A virus has been isolated from the mud flat snail. This virus has a single stranded circular genome of 2351 nucleotides that encodes 2 open reading frames that are oriented in opposite directions. The smaller open reading frame encodes a protein with similarities to the Rep proteins of circoviruses and plasmids. The larger open reading frame has no homology to any currently known protein.
An unusual—and as yet unnamed—virus has been isolated from the flatworm Girardia tigrina. Because of its genome organization, this virus appears to belong to an entirely new family. It is the first virus to be isolated from a flatworm.
From the hepatopancreas of the shrimp a circular single stranded DNA virus has been isolated. This virus does not appear to cause disease in the shrimp.
A circo-like virus has been isolated from the shrimp. The 1,777-nucleotide genome is circular and single stranded. It has some similarity to the circoviruses and cycloviruses.
Ten viruses have been isolated from echinoderms. All appear to belong to as yet undescribed genera.
A filamentous virus—Apis mellifera filamentous virus—has been described. It appears to be unrelated to other DNA viruses.

Plants

A circular single stranded DNA virus has been isolated from a grapevine. This species may be related to the family Geminiviridae but differs from this family in a number of important respects including genome size.
Several viruses—baminivirus, nepavirus and niminivirus—related to geminvirus have also been reported.
A virus—Ancient caribou feces associated virus—has been cloned from 700-y-old caribou faeces.
A new virus with a three part single stranded genome has been reported. This seems likely to be a member of a new family of viruses.

Marine and other

More than 600 single-stranded DNA viral genomes were identified in ssDNA purified from seawater. These fell into 129 genetically distinct groups that had no recognizable similarity to each other or to other virus sequences, and thus many likely represent new families of viruses. Of the 129 groups, eleven were much more abundant than the others, and although their hosts have yet to be identified, they are likely to be eukaryotic phytoplankton, zooplankton and bacteria.
A virus—Boiling Springs Lake virus—appears to have evolved by a recombination event between a DNA virus and an RNA virus. The genome is circular and encodes two proteins—a Rep protein and a capsid protein.
Further reports of viruses that appear to have evolved from recombination events between ssRNA and ssDNA viruses have been made.
A new virus has been isolated from the diatom Chaetoceros setoensis. It has a single stranded DNA genome and does not appear to be a member of any previously described group.
A virus—FLIP —has been isolated from a lake. This virus has a circular ssDNA genome and an internal lipid membrane enclosed in an icosahedral capsid. The capsid organisation is he capsid organization pseudo T = 21 dextro. The major capsid protein has two β-barrels. The capsid organisation is similar to bacteriophage PM2—a double stranded bacterial virus.

Satellite viruses

es are small viruses with either RNA or DNA as their genomic material that require another virus to replicate. There are two types of DNA satellite viruses—the alphasatellites and the betasatellites—both of which are dependent on begomoviruses. At present satellite viruses are not classified into genera or higher taxa.
Alphasatellites are small circular single strand DNA viruses that require a begomovirus for transmission. Betasatellites are small linear single stranded DNA viruses that require a begomovirus to replicate.

Phylogenetic relationships

Introduction

Phylogenetic relationships between these families are difficult to determine. The genomes differ significantly in size and organisation. Most studies that have attempted to determine these relationships are based either on some of the more conserved proteins—DNA polymerase and others—or on common structural features. In general most of the proposed relationships are tentative and have not yet been used by the ICTV in their classification.

ds DNA viruses

Herpesviruses and caudoviruses

While determining the phylogenetic relations between the various known clades of viruses is difficult, on a number of grounds the herpesviruses and caudoviruses appear to be related.
While the three families in the order Herpesvirales are clearly related on morphological grounds, it has proven difficult to determine the dates of divergence between them because of the lack of gene conservation. On morphological grounds they appear to be related to the bacteriophages—specifically the Caudoviruses.
The branching order among the herpesviruses suggests that Alloherpesviridae is the basal clade and that Herpesviridae and Malacoherpesviridae are sister clades. Given the phylogenetic distances between vertebrates and molluscs this suggests that herpesviruses were initially fish viruses and that they have evolved with their hosts to infect other vertebrates.
The vertebrate herpesviruses initially evolved ~ and underwent subsequent evolution on the supercontinent Pangaea. The alphaherpesvirinae separated from the branch leading to the betaherpesvirinae and gammaherpesvirinae about to. The avian herpes viruses diverged from the branch leading to the mammalian species. The mammalian species divided into two branches—the Simplexvirus and Varicellovirus genera. This latter divergence appears to have occurred around the time of the mammalian radiation.
Several dsDNA bacteriophages and the herpesviruses encode a powerful ATP driven DNA translocating machine that encapsidates a viral genome into a preformed capsid shell or prohead. The critical components of the packaging machine are the packaging enzyme which acts as the motor and the portal protein that forms the unique DNA entrance vertex of prohead. The terminase complex consists of a recognition subunit and an endonuclease/translocase subunit and cuts viral genome concatemers. It forms a motor complex containing five large terminase subunits. The terminase-viral DNA complex docks on the portal vertex. The pentameric motor processively translocates DNA until the head shell is full with one viral genome. The motor cuts the DNA again and dissociates from the full head, allowing head-finishing proteins to assemble on the portal, sealing the portal, and constructing a platform for tail attachment. Only a single gene encoding the putative ATPase subunit of the terminase is conserved among all herpesviruses. To a lesser extent this gene is also found in T4-like bacteriophages suggesting a common ancestor for these two groups of viruses. Another paper has also suggested that herpesviruses originated among the bacteriophages.
A common origin for the herpesviruses and the caudoviruses has been suggested on the basis of parallels in their capsid assembly pathways and similarities between their portal complexes, through which DNA enters the capsid. These two groups of viruses share a distinctive 12-fold arrangement of subunits in the portal complex. A second paper has suggested an evolutionary relationship between these two groups of viruses.
It seems likely that the tailed viruses infecting the archaea are also related to the tailed viruses infecting bacteria.
A study involving 600 herpes genomes and 2000 caudoviral genomes suggested that an evolutionary relationship exists between these order.

Large DNA viruses

The NCLDV group along with three other families and the phage Sulfolobus turreted icosahedral virus and the satellite virus Sputnik all possess double β-barrel major capsid proteins suggesting a common origin.
Several studies have suggested that the family Ascoviridae evolved from the Iridoviridae. A study of the Iridoviruses suggests that the Iridoviridae, Ascoviridae and Marseilleviridae are related with Ascoviruses most closely related to Iridoviruses.
The family Polydnaviridae may have evolved from the Ascoviridae. Molecular evidence suggests that the Phycodnaviridae may have evolved from the family Iridoviridae. These four families may form a clade but more work is needed to confirm this.
Some of the relations among the large viruses have been established. Mimiviruses are distantly related to Phycodnaviridae. Pandoraviruses share a common ancestor with Coccolithoviruses within the family Phycodnaviridae. Pithoviruses are related to Iridoviridae and Marseilleviridae.
Based on the genome organisation and DNA replication mechanism it seems that phylogenetic relationships may exist between the rudiviruses and the large eukaryal DNA viruses: the African swine fever virus, Chlorella viruses and poxviruses.
Based on the analysis of the DNA polymerase the genus Dinodnavirus may be a member of the family Asfarviridae. Further work on this virus will required before a final assignment can be made.
It has been suggested that at least some of the giant viruses may originate from mitochondria.

Other viruses

Based on the analysis of the coat protein, Sulfolobus turreted icosahedral virus may share a common ancestry with the Tectiviridae.
The families Adenoviridae and Tectiviridae appear to be related structurally.
Baculoviruses evolved from the nudiviruses.
The Hytrosaviridae are related to the baculoviruses and to a lesser extent the nudiviruses suggesting they may have evolved from the baculoviruses.
The Nimaviridae may be related to nudiviruses and baculoviruses.
The Nudiviruses seem to be related to the polydnaviruses.
A protein common to the families Bicaudaviridae, Lipotrixviridae and Rudiviridae and the unclassified virus Sulfolobus turreted icosahedral virus is known suggesting a common origin.
Examination of the pol genes that encode the DNA dependent DNA polymerase in various groups of viruses suggests a number of possible evolutionary relationships. All know viral DNA polymerases belong to the DNA pol families A and B. All possess a 3'–5'-exonuclease domain with three sequence motifs Exo I, Exo II and Exo III. The families A and B are distinguishable with family A Pol sharing 9 distinct consensus sequences and only two of them are convincingly homologous to sequence motif B of family B. The putative sequence motifs A, B, and C of the polymerase domain are located near the C-terminus in family A Pol and more central in family B Pol.
Phylogenetic analysis of these genes places the adenoviruses, bacteriophages and the plant and fungal linear plasmids into a single clade. A second clade includes the alpha- and delta-like viral Pol from insect ascovirus, mammalian herpesviruses, fish lymphocystis disease virus and chlorella virus. The pol genes of the African swine fever virus, baculoviruses, fish herpesvirus, T-even bacteriophages and poxviruses were not clearly resolved. A second study showed that poxvirus, baculovirus and the animal herpesviruses form separate and distinct clades. Their relationship to the Asfarviridae and the Myoviridae was not examined and remains unclear.
The polymerases from the archaea are similar to family B DNA Pols. The T4-like viruses infect both bacteria and archaea and their pol gene resembles that of eukaryotes. The DNA polymerase of mitochondria resembles that of the T odd phages.
The virophage—Mavirus—may have evolved from a recombination between a transposon of the Polinton family and an unknown virus.
The polyoma and papillomaviruses appear to have evolved from single-stranded DNA viruses and ultimately from plasmids.

ss DNA viruses

The evolutionary history of this group is currently poorly understood. An ancient origin for the single stranded circular DNA viruses has been proposed.
Capsid proteins of most icosahedral ssRNA and ssDNA viruses display the same structural fold, the eight-stranded beta-barrel, also known as the jelly-roll fold. On the other hand, the replication proteins of icosahedral ssDNA viruses belong to the superfamily of rolling-circle replication initiation proteins that are commonly found in prokaryotic plasmids. Based on these observations, it has been proposed that small DNA viruses have originated via recombination between RNA viruses and plasmids.
Circoviruses may have evolved from a nanovirus.
Given the similarities between the rep proteins of the alphasatellites and the nanoviruses, it is likely that the alphasatellites evolved from the nanoviruses. Further work in this area is needed to clarify this.
The geminiviruses may have evolved from phytoplasmal plasmids. The Genomoviridae and the Geminividiae appear to be related.
Based on the three-dimensional structure of the Rep proteins the geminiviruses and parvoviruses may be related.
The ancestor of the geminiviruses probably infected dicots.
The parvoviruses have frequently invaded the germ lines of diverse animal species including mammals, fishes, birds, tunicates, arthropods and flatworms. In particular they have been associated with the human genome for ~98 million years.
Members of the family Bidnaviridae have evolved from insect parvoviruses by replacing the typical replication-initiation endonuclease with a protein-primed family B DNA polymerase acquired from large DNA transposons of the Polinton/Maverick family. Some bidnavirus genes were also horizontally acquired from reoviruses and baculoviruses.
Polyomaviruses, papillomaviruses and parvoviruses may have descended from unrelated circular Rep-encoding single-stranded DNA viral ancestors.

Bacteriophage evolution

Since 1959 ~6300 prokaryote viruses have been described morphologically, including ~6200 bacterial and ~100 archaeal viruses. Archaeal viruses belong to 15 families and infect members of 16 archaeal genera. These are nearly exclusively hyperthermophiles or extreme halophiles. Tailed archaeal viruses are found only in the Euryarchaeota, whereas most filamentous and pleomorphic archaeal viruses occur in the Crenarchaeota. Bacterial viruses belong to 10 families and infect members of 179 bacterial genera: most of these are members of the Firmicutes and γ-proteobacteria.
The vast majority are tailed with and only 230 are polyhedral, filamentous or pleomorphic. The family Siphoviridae is the largest family. The tailed phages appear to be monophyletic and are the oldest known virus group. They arose repeatedly in different hosts and there are at least 11 separate lines of descent.
All of the known temperate phages employ one of only three different systems for their lysogenic cycle: lambda-like integration/excision, Mu-like transposition or the plasmid-like partitioning of phage N15.
A putative course of evolution of these phages has been proposed by Ackermann.
Tailed phages originated in the early Precambrian, long before eukaryotes and their viruses. The ancestral tailed phage had an icosahedral head of about 60 nanometers in diameter and a long non contractile tail with sixfold symmetry. The capsid contained a single molecule of double stranded DNA of about 50 kilobases. The tail was probably provided with a fixation apparatus. The head and tail were held together by a connector. The viral particle contained no lipids, was heavier than its descendant viruses and had a high DNA content proportional to its capsid size. Most of the genome coded for structural proteins. Morphopoietic genes clustered at one end of the genome, with head genes preceding tail genes. Lytic enzymes were probably coded for. Part of the phage genome was nonessential and possibly bacterial.
The virus infected its host from the outside and injected its DNA. Replication involved transcription in several waves and formation of DNA concatemers.
New phages were released by burst of the infected cell after lysis of host membranes by a peptidoglycan hydrolase. Capsids were assembled from a starting point, the connector and around a scaffold. They underwent an elaborate maturation process involving protein cleavage and capsid expansion. Heads and tails were assembled separately and joined later. The DNA was cut to size and entered preformed capsids by a headful mechanism.
Subsequently, the phages evolved contractile or short tails and elongated heads. Some viruses become temperate by acquiring an integrase–excisionase complex, plasmid parts or transposons.
A possible evolutionary pathway using vesicles rather than a protein coat has been described in the archaeal plasmid pR1SE.