The evolution of endogenous viral elements

Abstract

Endogenous retroviruses are a common component of the eukaryotic genome, and their evolution and potential function have attracted considerable interest. More surprising was the recent discovery that eukaryotic genomes contain sequences from RNA viruses that have no DNA stage in their life cycle. Similarly, several single-stranded DNA viruses have left integrated copies in their host genomes. This review explores some major evolutionary aspects arising from the discovery of these endogenous viral elements (EVEs). In particular, the reasons for the bias toward EVEs derived from negative-sense RNA viruses are considered, as well as what they tell us about the long-term "arms races" between hosts and viruses, characterized by episodes of selection and counter-selection. Most dramatically, the presence of orthologous EVEs in divergent hosts demonstrates that some viral families have ancestries dating back almost 100 million years, and hence are far older than expected from the phylogenetic analysis of their exogenous relatives.

Figure 1

Processes Involved in the Generation of EVEs, and How EVEs Can Be Used to Estimate the Age of Viruses, Using Both RNA and DNA Viruses as Examples The presence of EVEs in related species A and B and integrated into the same genomic position such that they are orthologous indicates that this integration event occurred prior to the divergence of these two species (species C is an outgroup). If it is known when species A and B diverged, then the minimum age of the insertion event can also be estimated. See Katzourakis and Gifford (2010) for more details. Figure kindly provided by Rob Gifford.

Figure 2

Genomic Structures and Phylogenetic Distribution of Exogenous and Endogenous Bornaviruses, Members of the Order Mononegavirales BDV, exogenous Borna disease virus (shown in red); EBLN, endogenous viruses. For each species, the most intact endogenous elements are shown relative to a representative complete exogenous virus genome. Squares of a specific color on the phylogeny indicate EVEs and the viral gene they represent. Note the bias toward integrated NP sequences. EVEs with poly-A tails are shown, while intact ORFs are marked by an “O” symbol and expressed genes by an “X” symbol. Vertical lines identify noncontiguous genes. Adapted from Katzourakis and Gifford (2010), which should be consulted for more details. Original figure kindly provided by Rob Gifford.

Figure 3

Genomic Structures and Phylogenetic Distribution of Exogenous and Endogenous Filoviruses, Members of the Viral Order Mononegavirales ZEBOV, exogenous Ebola virus Zaire (shown in red); TSD, target site duplication. Other labeling is the same as that in Figure 2. Adapted from Katzourakis and Gifford (2010), which should be consulted for more details. Original figure kindly provided by Rob Gifford.

Figure 4

Evolution of Exogenous and Endogenous Avian Hepadnaviruses The phylogeny of the bird host species is shown in blue on the left of the figure, while that of the hepadnaviruses (both avian and mammalian) is shown in green on the right. Estimates of divergence time are also shown in both cases; note the huge difference between the time scale of the host (millions of years) and virus (thousands of years) trees. The presence or absence of orthologous eZHBV insertions in various bird species (i.e., eZHBVI, eZHBVj, and eZHBVa) are denoted by the “+” and “−” symbols, respectively, while a question mark means uncertain status. The likely phylogenetic placement of the germline integrations producing eZHBVl and eZHBVa are shown by the green lines connecting the host and virus trees. Using estimated bird host divergence times of 25 MYA and 35 MYA, it is conservatively estimated that the integration of eZHBVa and eZHBVl must have occurred at least 19 MYA, as shown by the dashed line. Adapted from Gilbert and Feschotte (2010), which should be consulted for more details. Original figure kindly provided by Clément Gilbert.