Host antiviral proteins engage in evolutionary arms races with viruses, in which both sides rapidly evolve at interaction interfaces to gain or evade immune defense. For example, primate TRIM5α uses its rapidly evolving ‘v1’ loop to bind retroviral capsids, and single mutations in this loop can dramatically improve retroviral restriction. However, it is unknown whether such gains of viral restriction are rare, or if they incur loss of pre-existing function against other viruses. Using deep mutational scanning, we comprehensively measured how single mutations in the TRIM5α v1 loop affect restriction of divergent retroviruses. Unexpectedly, we found that the majority of mutations increase weak antiviral function. Moreover, most random mutations do not disrupt potent viral restriction, even when it is newly acquired via a single adaptive substitution. Our results indicate that TRIM5α’s adaptive landscape is remarkably broad and mutationally resilient, maximizing its chances of success in evolutionary arms races with retroviruses.

1Host antiviral proteins engage in evolutionary arms races with viruses, in which both sides 2 rapidly evolve at interaction interfaces to gain or evade immune defense. For example, primate 3 TRIM5α uses its rapidly evolving "v1" loop to bind retroviral capsids, and single mutations in this 4 loop can dramatically improve retroviral restriction. However, it is unknown whether such gains 5 of viral restriction are rare, or if they incur loss of pre-existing function against other viruses. 6Using deep mutational scanning, we comprehensively measured how single mutations in the 7 TRIM5α v1 loop affect restriction of divergent retroviruses. Unexpectedly, we found that the 8 majority of mutations increase antiviral function. Moreover, most random mutations do not 9 disrupt potent viral restriction, even when it is newly acquired via single adaptive substitutions. 10Our results indicate that TRIM5α's adaptive landscape is remarkably broad and mutationally 11 resilient, maximizing its chances of success in evolutionary arms races with retroviruses. 13Here, we investigated the adaptive landscape of antiviral specificity conferred by the 1 rapidly evolving, capsid-binding v1 loop of TRIM5α. To our surprise, we found that, rather than 2 the evolutionary landscape of TRIM5α being narrowly constrained among all possible amino 3 acid substitutions, the majority of random mutations in the v1 loop resulted in gains of antiviral 4 restriction. We found that the primary v1 loop determinant for TRIM5α's restriction of HIV-1 and 5 other lentiviruses is its net electrostatic charge. Furthermore, both rhesus and human TRIM5α 6 proteins are highly resilient to mutation, in that they withstand more than half of all possible 7 single amino acid mutations in the v1 loop without compromising their antiviral restriction 8 abilities. This unexpectedly permissive landscape allows TRIM5α to sample a wide variety of 9 mutations to maximize its chances of success in arms races with retroviruses. 10 11 RESULTS 12 A deep mutational scan of the TRIM5α v1 loop 13 Despite their rapid evolution, primate TRIM5α orthologs have sampled relatively limited amino 14 acid diversity at rapidly evolving positions within the capsid-binding v1 loop (Figure 1A). For 15 example, although single amino acid changes at residue 332 are responsible for dramatic 16 differences in antiviral restriction (Y. Li et al., 2006), this residue repeatedly toggles between just 17 three amino acids. The limited diversity is not due to evolutionary inaccessibility, since most 18 amino acids that can be sampled with single nucleotide changes are not observed among19primate TRIM5α orthologs ( Figure 1B). There are two alternative explanations for this restricted 20 diversity. First, it might suggest that adaptive gain-of-function mutations in TRIM5α are rare, 21 with TRIM5α's evolutionary landscape mainly consisting of fitness valleys with only a few 22 mutational avenues to reach fitness peaks ( Figure 1C). Conversely, the limited diversity might 23 be a consequence of epistatic in...

No abstract