The Battle between Retroviruses and APOBEC3 Genes: Its Past and Present

Abstract

The APOBEC3 family of proteins in mammals consists of cellular cytosine deaminases and well-known restriction factors against retroviruses, including lentiviruses. APOBEC3 genes are highly amplified and diversified in mammals, suggesting that their evolution and diversification have been driven by conflicts with ancient viruses. At present, lentiviruses, including HIV, the causative agent of AIDS, are known to encode a viral protein called Vif to overcome the antiviral effects of the APOBEC3 proteins of their hosts. Recent studies have revealed that the acquisition of an anti-APOBEC3 ability by lentiviruses is a key step in achieving successful cross-species transmission. Here, we summarize the current knowledge of the interplay between mammalian APOBEC3 proteins and viral infections and introduce a scenario of the coevolution of mammalian APOBEC3 genes and viruses.

Keywords: APOBEC3; Vif; arms race; coevolution; gene diversification; lentivirus.

Figure 1

Association of host factors with viral cross-species transmission. (a) Usage of viral receptors for infection. In the case of SARS-CoV-2, its putative ancestral virus, a bat CoV, uses the receptor molecule(s) expressed on bat cells (not yet identified) (left). It is expected that a bat CoV was transferred to humans by a spillover event (although there may have been certain intermediate hosts between bats and humans). At first, the ancestral bat CoV may not have efficiently used human ACE2 as its receptor for infection (middle). However, through adaptive evolution, the bat CoV began to be able to efficiently utilize human ACE2 as the receptor and evolved into SARS-CoV-2 (right). (b) Hijacking of a proviral host protein. In the case of avian influenza A virus, viral proteases, particularly PB2, utilize avian ANP32A for efficient replication in avian cells (left). However, the avian flu polymerase cannot utilize human ANP32A (middle). Through adaptive evolution, the E627K mutation in the PB2 protein enabled the hijacking of human ANP32A for efficient replication in human cells (right). (c) Antagonism of an antiviral host protein. SIVcpz Vif can degrade and counteract the antiviral APOBEC3G protein of its host species (the chimpanzee) (left). In a spillover event, SIVcpz was transferred to gorillas. However, SIVcpz Vif cannot counteract the antiviral APOBEC3G protein of the new host (the gorilla) (middle). Through adaptive evolution, the virus acquired an M16E mutation in Vif, which enabled the virus to counteract gorilla APOBEC3G protein and efficiently replicate (right).

Figure 2

Birth of the APOBEC3G gene. Because the genomes of members of Monotremata (e.g., platypuses) and Marsupialia (e.g., kangaroos) do not encode APOBEC3 genes in the canonical locus (i.e., the locus sandwiched between the CBX6 and CBX7 genes), it was previously assumed that APOBEC3 genes were acquired in the common ancestor of Placentalia. Ito, Gifford, and Sato [62] revealed that the APOBEC3G gene (i.e., the Z2-Z1-type gene) is encoded in the genomes of Hominidae (e.g., humans and chimpanzees), Old World monkeys (e.g., rhesus macaques), and New World monkeys (e.g., night monkeys) but not in those of prosimians (e.g., bushbabies). These results suggest that the APOBEC3G gene (i.e., the Z2-Z1-type gene) was generated in the common ancestor of Simiiformes. Because the timing of the birth of the APOBEC3G gene overlapped with a period of high ERV invasion in the primate genome (shown as a heatmap) [62], it is speculated that retroviral invasions were a driving force in the generation of the APOBEC3G gene. MYA, million years ago.