The type I interferon (IFN-I)-inducible human restriction factor TRIM5α inhibits the infection of human cells by specific nonhuman retroviruses, such as N-MLV and EIAV, but does not generally target HIV-1. However, the introduction of two aminoacid substitutions, R332G and R355G, in the human TRIM5α (huTRIM5α) domain responsible for retroviral capsid recognition leads to efficient HIV-1 restriction upon stable over-expression. CRISPR-Cas-based approaches to precisely edit DNA could be employed to modify TRIM5 in human cells. Toward this aim, we used a DNA transfection-based CRISPR-Cas9 genome editing protocol to successfully mutate TRIM5 to its potentially HIV-1-restrictive version by homology-directed repair (HDR) in HEK293T cells. Nine clones bearing at least one HDR-edited TRIM5 allele containing both mutations were isolated (5.6% overall efficiency), whereas another one contained only the R332G mutation. Of concern, several of these HDR-edited clones contained on-target undesired mutations, and none had all the alleles corrected. Our study demonstrates the feasibility of editing the TRIM5 gene in human cells and identifies the main challenges to be addressed in order to use this approach to confer protection from HIV-1.

Tripartite-motif-containing protein 5 isoform α (TRIM5α) is a cytoplasmic antiretroviral effector upregulated by type I interferons (IFN-I). We previously showed that two points mutations, R332G/R335G, in the retroviral capsid-binding region confer human TRIM5α the capacity to target and strongly restrict HIV-1 upon overexpression of the mutated protein. Here, we used clustered regularly interspaced short palindromic repeats (CRISPR)-Cas9-mediated homology-directed repair (HDR) to introduce these two mutations in the endogenous human TRIM5 gene. We found 6 out of 47 isolated cell clones containing at least one HDR-edited allele. One clone (clone 6) had both alleles containing R332G, but only one of the two alleles containing R335G. Upon challenge with an HIV-1 vector, clone 6 was significantly less permissive compared to unmodified cells, whereas the cell clones with monoallelic modifications were only slightly less permissive. Following interferon (IFN)-β treatment, inhibition of HIV-1 infection in clone 6 was significantly enhanced (~40-fold inhibition). TRIM5α knockdown confirmed that HIV-1 was inhibited by the edited TRIM5 gene products. Quantification of HIV-1 reverse transcription products showed that inhibition occurred through the expected mechanism. In conclusion, we demonstrate the feasibility of potently inhibiting a viral infection through the editing of innate effector genes. Our results also emphasize the importance of biallelic modification in order to reach significant levels of inhibition by TRIM5α.

16The type I interferon (IFN-I)-inducible human restriction factor TRIM5α inhibits the infection of allele containing both mutations were isolated (5.6% overall efficiency), whereas another one 24 contained only the R332G mutation. Of concern, several of these HDR-edited clones contained 25 on-target undesired mutations, and none had all the alleles corrected. We observed a lack of 26 HIV-1 restriction in the cell clones generated, even when cells were stimulated with IFN-I prior 27 to infection. This, however, was partly explained by the unexpectedly low potential for TRIM5α-28 mediated restriction activity in this cell line. Our study demonstrates the feasibility of editing the 29

TRIM5α is a cytoplasmic antiviral effector induced by type I interferons (IFN-I) that has the potential to intercept incoming retroviruses by interacting with their capsid core, leading to uncoating induction and the partial degradation of core components. Most HIV-1 strains escape restriction by human TRIM5α due to a lack of interaction between TRIM5α and its viral molecular target. We previously showed, however, that two point mutations, R332G/R335G, in the capsid-binding region confer human TRIM5α with the capacity to target and strongly restrict HIV-1 upon the overexpression of the mutated protein. Here, we explored the possibility to introduce these two mutations in the endogenous human TRIM5 gene by CRISPR-Cas9-mediated gene editing. For this, we electroporated CRISPR ribonucleoproteins (RNPs) and the donor DNA into Jurkat T lymphocytic cells and isolated clones by limiting dilution. We analyzed 47 clones using specific PCR assays, and found that six clones (13%) contained at least one gene-edited allele. One clone (clone 6) had both alleles edited for R332G, but only one of the two alleles was edited for R335G. Upon challenge with an HIV-1 vector, clone 6 was significantly less permissive compared to unmodified cells, whereas the cell clones with monoallelic modifications were only slightly less permissive. Following IFN-β treatment, the inhibition of HIV-1 infection in clone 6 was significantly enhanced (~50-fold inhibition), whereas IFN-β treatment had no effect on TRIM5α overexpressed by retroviral transduction. Knockdown experiments confirmed that HIV-1 was inhibited by the edited TRIM5 gene products, whereas quantification of HIV-1 reverse transcription products confirmed that inhibition occurred through the expected mechanism. In conclusion, we demonstrate the feasibility of potently inhibiting a viral infection through the editing of innate effector genes, but our results also emphasize the importance of biallelic modification in order to reach significant levels of inhibition by TRIM5α.

TRIM5α is a cytoplasmic antiretroviral effector upregulated by type I interferons (IFN-I). We previously showed that two points mutations, R332G/R335G, in the retroviral capsid-binding region confer human TRIM5α the capacity to target and strongly restrict HIV-1 upon over-expression of the mutated protein. Here, we used CRISPR-Cas9-mediated homology-directed repair (HDR) to introduce these two mutations in the endogenous human TRIM5 gene. We found 6 out of 47 isolated cell clones containing at least one HDR-edited allele. One clone (Clone 6) had both alleles containing R332G but only one of the two alleles containing R335G. Upon challenge with an HIV-1 vector, clone 6 was significantly less permissive compared to unmodified cells, whereas the cell clones with monoallelic modifications were only slightly less permissive. Following IFN-β treatment, inhibition of HIV-1 infection in clone 6 was significantly enhanced (~40-fold inhibition). TRIM5α knockdown confirmed that HIV-1 was inhibited by the edited TRIM5 gene products. Quantification of HIV-1 reverse transcription products showed that inhibition occurred through the expected mechanism. In conclusion, we demonstrate the feasibility of potently inhibiting a viral infection through editing of innate effector genes. Our results also emphasize the importance of biallelic modification in order to reach significant levels of inhibition by TRIM5α.