Impact of HIV-1 capsid polymorphisms on viral infectivity and susceptibility to lenacapavir

Abstract

Lenacapavir (LEN) is a first-in-class capsid (CA) inhibitor for the treatment and prevention of HIV-1 infection. While LEN has shown potent antiviral activity across all major HIV-1 subtypes, the impact of existing HIV-1 CA sequence diversity on the activity of LEN remains to be determined. Here, we identified natural polymorphisms within the LEN-binding site and assessed each for their impact on viral infectivity and susceptibility to LEN. Using a co-crystal structure of LEN in complex with a CA hexamer, we identified 29 binding site residues within five angstroms of LEN and analyzed each for naturally occurring polymorphisms across a multiclade collection of >10,000 unique HIV-1 gag sequences. Eleven of these CA residues, including five (M66, Q67, K70, N74, and A105) previously associated with LEN resistance when mutated, were invariant across these sequences. The remaining 18 residues showed one or more substitutions with a ≥0.5% prevalence for a total of 54 CA polymorphisms. When introduced as site-directed mutants (SDMs) in an NL4.3-based reporter virus and evaluated for infectivity and drug susceptibility in MT-4 cells, 74% (40/54) showed impaired infectivity (0.01%-77% of wild type), with 96% (46/48) exhibiting minimal change (less than threefold) in susceptibility to LEN. While CA substitutions L56V and N57H conferred high-level resistance to LEN (72- and 4,890-fold, respectively), both variants showed diminished replication capacity in primary T-cells relative to the wild-type virus. Collectively, these results indicate that existing CA natural HIV-1 sequence diversity within the LEN-binding site is rare and should minimally impact LEN efficacy in treatment-naïve individuals.IMPORTANCEHIV-1 capsid protein mediates multiple essential functions throughout the viral replication cycle, making it an attractive target for therapeutic intervention. Lenacapavir (LEN), a first-in-class HIV-1 capsid inhibitor, is being evaluated as a long-acting option in multiple ongoing clinical studies for HIV treatment and prevention. Twice-yearly lenacapavir is approved in multiple countries for the treatment of adults with multi-drug-resistant HIV-1 in combination with other antiretrovirals, and its investigational use for pre-exposure prophylaxis has shown 99.9%-100% efficacy in preventing HIV infection among a broad and geographically diverse range of study participants. In this report, we investigated how HIV-1 sequence diversity within the LEN binding site may impact virus replication capacity and sensitivity to LEN. Our data demonstrate high capsid sequence conservation across a large and diverse collection of HIV-1 variants, with the majority of naturally occurring capsid polymorphisms having a detrimental effect on viral infectivity and minimal impact on susceptibility to LEN.

Keywords: HIV; antiretroviral resistance; lenacapavir; polymorphism; viral fitness.

Fig 1

HIV-1 CA residues within five angstroms of LEN. (A) LEN bound to cross-linked CA hexamer protein. Left, top surface view of CA hexamer (gray) with LEN (yellow) bound at the interface formed between one CA monomer (cyan ribbon) and its adjacent CA monomer (green ribbon). Right, close-up view of LEN in its binding site. (B and C) Residues defining the LEN-binding site in adjacent CA monomers. CA amino acid side chains within five angstroms of LEN are highlighted in yellow and orange for CA monomer 1 and 2, respectively.

Fig 2

HIV-1 CA sequence conservation in clinical isolates. (A) Percent CA amino acid identity among 5,689 unique HIV-1 subtype B sequences relative to HIV-1 HXB2 reference sequence. Top, schematic illustrating the organization of a CA monomer (amino acids 1–231) composed of an N-terminal domain (NTD, blue), C-terminal domain (CTD, yellow), beta-hairpin (bH), alpha-helices (H1-11, individual boundaries denoted by lower black bars), cyclophilin A binding loop (gray), and major homology region (MHR, green). The 29 CA amino acids identified within five angstroms of LEN at its binding site are denoted above, with those previously implicated in the loss of LEN susceptibility highlighted in red. Bottom, percent CA sequence conservation plot with orange circles highlighting locations of LEN binding site residues. (B) Heat map depicting percent conservation of LEN CA-binding site residues by HIV-1 subtype. The number of unique sequences analyzed for each subtype is shown in Table 1. CRF, circulating recombinant form.

Fig 3

Infectivity of HIV-1 CA SDMs in MT-4 cells. WT and site-directed CA mutant single-cycle reporter HIV-1_NL4.3 were produced in HEK293T cells by transient transfection, and the HIV content for each was determined by p24 antigen enzyme-linked immunosorbent assay (ELISA)using a single serial dilution of each sample and quantified across three samples within the linear range of the assay. MT-4 cells were infected in duplicate with serially diluted, p24-normalized WT and CA mutant viruses and developed 3 days later by One-Glo addition. Center line and error bars represent mean ± SD infectivity values, expressed as a percentage of the WT virus, obtained from at least three independent experiments. P values for each mutant (n = 3 replicate cell cultures for S41A and I73F mutants and 4–8 replicate cell cultures for all others) relative to the WT (n = 12 replicate cell cultures) determined by Brown-Forsythe and Welch ANOVA test. ****P < 0.0001; ***P < 0.001; **P < 0.01; ns, not significant (P > 0.05).

Fig 4

Fold resistance of single-cycle reporter HIV-1 strain NL4.3 encoding naturally occurring CA variants to LEN and a control antiretroviral agent (BIC) in MT-4 cells. Rectangles and error bars represent mean ± SD. EC₅₀ FC values for each CA mutant relative to the WT virus from three independent experiments (n = 3 biological replicates each). The red dotted line defines the cut-off for drug resistance.

Fig 5

Structural impact of CA substitutions conferring loss of susceptibility to LEN. (A) Branching at the Cβ of the L56V mutation introduces steric clashing (red dashed line) with the central di-fluoro-benzyl group of LEN. (B) N57 forms three hydrogen bonds (purple dotted lines) with LEN, whereas the N57H variant forms only one.

Fig 6

Replication competency of LEN-resistant naturally occurring CA variants. Outgrowth kinetics for replication-competent reporter HIV-1 encoding WT or each of the two LEN-resistant variants L56V and N57H in primary human CD4+ T cells obtained from each of two independent donors. Symbols and error bars represent mean ± SD luminescence values expressed relative to the WT virus from six replicate cell cultures in each of two independent experiments.