Genetic diversity in TRIMa gives insights into the impact of retroviruses during primate evolution

Retroviruses have been colonizing vertebrate hosts for hundreds of millions of years, leaving a calling card of integrated retroviral sequence transmitted between generations (Gifford and Tristem 2003). An early discovery in research into the current HIV-1 pandemic was the finding that this particular retrovirus had a limited host range, being restricted to humans and apes. The barrier to infection in other primate cells was identified in 2004 as being mediated by a protein named TRIM5a, which acted to block HIV-1 replication soon after viral entry into cells and before reverse transcription (Fig. 14.8) (Stremlau et al. 2004).

The discovery of TRIM5a was made after a screen of cDNA clones prepared from HIV-1 resistant rhesus macaque cells and tested for their ability to protect an HIV-1 susceptible human cell line (Stremlau et al. 2004). Subsequent small interfering RNA knockdown experiments to endogenous TRIM5a in rhesus cells removed the block to HIV-1 replication naturally occurring in these cells. This blocking protein was, however, specific to HIV-1 as almost no protection was found to the related lentivirus, SIVmac, when the cDNA for rhesus TRIM5a was introduced into human cells. Indeed the pattern of restriction of viral replication varies significantly between primate species and between viruses, with human TRIM5a poorly restricting HIV-1 but efficiently restricting N-MLV, a murine y retrovirus (Perron et al. 2004).

TRIM5a was found to act by binding to the retroviral capsid core, leading to accelerated uncoating and degradation of the viral capsid (Chatterji et al. 2006; Stremlau et al. 2006). A particular domain on TRIM5a was found to be involved in recognizing the viral capsids entering the cell and to be a critical determinant of species-specific restriction of primate lentiviruses. This pattern recognition molecule is denoted the B30.2/SPRY domain (Fig. 14.9). Between primate species, there is evidence of marked sequence diversity in TRIM5a including deletions, insertions, and nucleotide substitutions and duplications (Newman and Johnson 2007). The greatest amino acid variation is seen in the B30.2/SPRY domain. The domain comprises a conserved central hydrophobic core with a binding surface comprised of protruding loops of variable length containing non-conserved amino acids (Fig. 14.9). It appears no coincidence that the key region of this domain determining retroviral specificity (dubbed the SPRY 'patch') should also have the most evidence of positive selection, with variation in the protein found in different primate lineages thought to reflect past infections with pathogenic retroviruses (Box 14.5) (Sawyer et al. 2005; Stremlau et al. 2005; Yap et al. 2005).

Different primate species encode different TRIM5a proteins, each with a different antiretroviral restriction activity (Song et al. 2005).There is marked sequence variation between Old and New World monkeys with specific regions and individual amino acids altering TRIM5a activity. For example, substitution of arginine (found in humans and chimpanzees) to proline (found in rhesus monkeys) in position 332 of human TRIM5a within the SPRY patch restricts HIV-1 (Yap et al. 2005; Li et al. 2006c).

Moreover there is evidence that amino acid variation at this site will cause retrovirus-specific restriction. An ancient endogenous retrovirus reconstructed from the chimp and gorilla genomes (PtERVI) was found to be restricted by human TRIM5a but this restriction was lost on substituting the arginine (R) for a glutamine (Q), the hominid ancestral residue (see Fig. 14.9). In contrast, this change in amino acid improved the ability of the TRIM5a to restrict HIV-1 (Kaiser et al. 2007). This remarkable piece of scientific work, resurrecting an extinct retrovirus found at more than 100 copies in chimpanzee and gorilla genomes but absent from the human genome, suggests that TRIM5a became fixed in our early human ancestors based on its ability to confer resistance to a retrovirus such as PtERV1 some 3 to 4 million years ago and that this may, in part, be responsible for our current low resistance to HIV-1 (Kaiser et al. 2007). A large body of work has now clearly demonstrated that evolution of diversity in TRIM5a has arisen due to species-specific ancestral retroviral challenges.

Resequencing TRIM5 from indigenous human subjects from geographically distinct regions (Sawyer et al. 2006) as well as HIV infected and exposed but seronegative individuals (Speelmon et al. 2006) has revealed many SNPs, including five common nonsynonymous polymorphisms. Analysis of the distribution of SNPs has shown a marked

Figure 14.8 Site of action of TRIM5a and APOBEC3G during HIV-1 infection. TRIM5a acts to restrict HIV-1 after viral entry into the cell by binding to the retroviral capsid core. APOBEC3G acts later, but this is critically dependent on whether viral Vif protein is present. (A) In wild-type HIV-1, the integrated HIV-1 provirus is transcribed and translated. The Vif protein acts to bind human APOBEC3G, targeting it for ubiquitination and proteosomal degradation. (B) In contrast with HIV-1/Avif, no Vif protein is synthesized and APOBEC3G is therefore not degraded but rather can become incorporated into the nascent virion and restrict further infection. Redrawn and reprinted from Holmes et al. (2007), copyright 2007, with permission from Elsevier.

Figure 14.8 Site of action of TRIM5a and APOBEC3G during HIV-1 infection. TRIM5a acts to restrict HIV-1 after viral entry into the cell by binding to the retroviral capsid core. APOBEC3G acts later, but this is critically dependent on whether viral Vif protein is present. (A) In wild-type HIV-1, the integrated HIV-1 provirus is transcribed and translated. The Vif protein acts to bind human APOBEC3G, targeting it for ubiquitination and proteosomal degradation. (B) In contrast with HIV-1/Avif, no Vif protein is synthesized and APOBEC3G is therefore not degraded but rather can become incorporated into the nascent virion and restrict further infection. Redrawn and reprinted from Holmes et al. (2007), copyright 2007, with permission from Elsevier.

0 0

Post a comment