The A128T resistance mutation reveals aberrant protein multimerization as the primary mechanism of action of allosteric HIV-1 integrase inhibitors

Abstract

Allosteric HIV-1 integrase (IN) inhibitors (ALLINIs) are a very promising new class of anti-HIV-1 agents that exhibit a multimodal mechanism of action by allosterically modulating IN multimerization and interfering with IN-lens epithelium-derived growth factor (LEDGF)/p75 binding. Selection of viral strains under ALLINI pressure has revealed an A128T substitution in HIV-1 IN as a primary mechanism of resistance. Here, we elucidated the structural and mechanistic basis for this resistance. The A128T substitution did not affect the hydrogen bonding between ALLINI and IN that mimics the IN-LEDGF/p75 interaction but instead altered the positioning of the inhibitor at the IN dimer interface. Consequently, the A128T substitution had only a minor effect on the ALLINI IC50 values for IN-LEDGF/p75 binding. Instead, ALLINIs markedly altered the multimerization of IN by promoting aberrant higher order WT (but not A128T) IN oligomers. Accordingly, WT IN catalytic activities and HIV-1 replication were potently inhibited by ALLINIs, whereas the A128T substitution in IN resulted in significant resistance to the inhibitors both in vitro and in cell culture assays. The differential multimerization of WT and A128T INs induced by ALLINIs correlated with the differences in infectivity of HIV-1 progeny virions. We conclude that ALLINIs primarily target IN multimerization rather than IN-LEDGF/p75 binding. Our findings provide the structural foundations for developing improved ALLINIs with increased potency and decreased potential to select for drug resistance.

Keywords: Allosteric Inhibitors; Drug Resistance; HIV-1; Infectious Diseases; Integrase; Pharmacology; Protein-DNA Interactions; Protein-Drug Integrations; Protein-Protein Interactions; Structural Biology.

FIGURE 1.

Catalytic activities of WT and A128T INs. A, strand transfer reaction products (STP; upper panels) and 3′-processing products (19-P; lower panels). Lane 1, 21-mer DNA substrate (21-S) without IN; lanes 2–4, increasing concentrations (0.5, 1, and 2 μm) of WT IN added to the reactions; lanes 5–7, increasing concentrations (0.5, 1, and 2 μm) of A128T IN added to the reactions. B, concerted integration results. The positions of 32-mer donor and supercoiled (SC) target DNA substrates, as well as full-site (FS) and half-site (HS) integration products, are indicated. Lane 1, DNA markers (Bioline Quanti-Marker, 1 kb); lanes 2 and 3, target DNA; lane 4, WT IN (2.4 μm) added to donor and target DNA substrates without LEDGF/p75; lanes 5–7, LEDGF/p75 added with decreasing concentrations (2.4, 1.2, and 0.6 μm) of WT IN; lane 8, A128T IN (2.4 μm) added to donor and target DNA substrates without LEDGF/p75; lanes 9–11, LEDGF/p75 with decreasing concentrations (2.4, 1.2, and 0.6 μm) of A128T IN.

FIGURE 2.

Chemical structures of ALLINI-1 and ALLINI-2.

FIGURE 3.

Effects of ALLINI-1 on 3′-processing activities and IN-LEDGF/p75 binding of WT and A128T INs. A, dose-response effects of ALLINI-1 on 3′-processing activities of WT (●) and A128T (□) INs. B, dose-response effects of ALLINI-1 on the IN-LEDGF/p75 binding for WT (●) and A128T (□) INs. The IC₅₀ values and S.E. obtained from curve fittings are given in Table 1.

FIGURE 4.

Effects of ALLINI-1 on multimerization of WT and A128T INs. Shown are the dose-response effects of ALLINI-1-induced multimerization of WT (●) and A128T (□) INs. The HTRF signal observed due to the dynamic exchange of IN subunits in the absence of the inhibitor is considered 100% base line. The mean values of three independent experiments are shown.

FIGURE 5.

Size exclusion chromatography demonstrating differential multimerization of WT and A128T INs in the presence of ALLINI-1. Shown are the elution profiles of 20 μm WT IN in the absence (A) and presence (B) of 80 μm ALLINI-1. The elution times and respective estimated oligomeric states for the indicated peak are summarized in supplemental Table 3. Also shown are the elution profiles of 20 μm A128T IN in the absence (C) and presence (D) of 80 μm ALLINI-1. The elution times and respective estimated oligomeric states of A128T peaks are summarized in supplemental Table 4.

FIGURE 6.

Solubility of the complex of WT IN and ALLINI-1. WT IN was incubated with the indicated concentrations of ALLINI-1 and then subjected to centrifugation. The supernatant (S) and pellet (P) fractions were analyzed by SDS-PAGE, and IN was detected by anti-His antibody.

FIGURE 7.

Overlay of crystal structures of ALLINI-1 bound to A128T and WT IN CCDs. Ala-128 and its corresponding ligand ALLINI-1 are colored yellow, whereas Thr-128 and the respective ALLINI-1 molecule are colored magenta. Hydrogen bonds between ALLINI-1 molecules and the backbones of Glu-170 and His-171 are shown by yellow (for WT IN) and magenta (for A128T IN) dashed lines. Subunits 1 and 2 are colored cyan and gray, respectively.

FIGURE 8.

Effects of ALLINI-1 on WT and A128T HIV-1 p24 production and infectivity. Upper panel, HEK293T cells were transfected with either WT or A128T IN provirus. HIV-1 particles were produced in the presence or absence of ALLINI-1 for 24 h, and cell-free Gag was measured by HIV-1 p24 ELISA. Lower panel, the indicated WT or A128T IN cell-free virus equivalent to 4 ng of HIV-1 p24 was used to infect TZM-bl cells, and luciferase assay was performed at 48 h post-infection. The luciferase signal, obtained in the absence of ALLINI-1 (Me₂SO alone) for WT or A128T IN, was set to 100%. The average values from at least triplicate infections are shown, and error bars represent S.D.