Mechanisms of HIV-1 integrase resistance to dolutegravir and potent inhibition of drug-resistant variants

Abstract

HIV-1 infection depends on the integration of viral DNA into host chromatin. Integration is mediated by the viral enzyme integrase and is blocked by integrase strand transfer inhibitors (INSTIs), first-line antiretroviral therapeutics widely used in the clinic. Resistance to even the best INSTIs is a problem, and the mechanisms of resistance are poorly understood. Here, we analyze combinations of the mutations E138K, G140A/S, and Q148H/K/R, which confer resistance to INSTIs. The investigational drug 4d more effectively inhibited the mutants compared with the approved drug Dolutegravir (DTG). We present 11 new cryo-EM structures of drug-resistant HIV-1 intasomes bound to DTG or 4d, with better than 3-Å resolution. These structures, complemented with free energy simulations, virology, and enzymology, explain the mechanisms of DTG resistance involving E138K + G140A/S + Q148H/K/R and show why 4d maintains potency better than DTG. These data establish a foundation for further development of INSTIs that potently inhibit resistant forms in integrase.

Fig. 1.. Comparison of antiviral potencies of DTG and 4d against a panel of INSTI-resistant mutants.

(A) Chemical structures of DTG and 4d, used for the assays. The EC₅₀ values were determined for either of the two compounds using vectors that carry INSTI-resistant (B) single, (C) double, and (D) triple mutants in a single-round infection assay. The error bars represent SDs of independent experiments, n = 3, performed in triplicate.

Fig. 2.. Enzymatic activities, viral replicative capacities, and coupled residue pairs.

(A) Enzymatic catalytic activities, half-site (HS) and concerted integration (CI), for all combinations of single, double, and triple IN mutants corresponding to changes at positions E138K + G140A/S + Q148H/K/R. The bands corresponding to supercoiled plasmid DNA (SP) and viral DNA are also indicated. (B) Bar graph of the two-end integration data in (A). The integration activity of WT IN was set to 100%, and the activity of mutant IN was presented as a percentage of wt IN. The error bars represent SDs of independent experiments, n = 3, performed in triplicate. (C) Replicative capacities measured in cell culture of the INSTI-resistant mutants using a single-round infection assay. The replicative capacity of wt IN was set to 100%, and the activity of mutant IN was presented as a percentage of wt IN. (D) Double mutant cycle data derived from Potts statistical analyses. As fitness is inversely proportional to the Potts energy, ΔΔE > 0 implies that the mutations are compensatory. The plot shows that many of the residue pairs, most notably G140S + Q148H and G140A + Q148K, are coupled to one another, indicating that compensatory effects between these pairs of positions within the protein drive the observed residue covariation. Residue pairs close to the center of the distribution are not coupled.

Fig. 3.. Drug-resistant mutations E138K/G140A/Q148K induce altered residue configurations in the active site around bound DTG ligand.

Atomic models of the active site from high-resolution intasome (WT and mutant) structures bound to DTG are displayed for (A) the WT intasome, (B to D) the single-mutant variants (B) E138K, (C) G140A, (D) Q148K, (E to G) the double-mutant variants (E) E138K/G140A, (F) E138K/Q148K, (G) G140A/Q148K, and (H) the triple-mutant variant E138K/G140A/Q148K. In the legend, pink arrows indicate the positions of the mutations, while the asterisks refer to the induced changes, red indicating a novel interaction and yellow indicating an alteration of the configuration of the residue. (I) MD simulation of E138K/Q148K (KK) and E138K/G140A/Q148K (KAK) intasomes with DTG bound. The atomic distances, measured from the positively charged amino head of Lys¹⁴⁸ to either of the Mg²⁺ ions, are indicated for the duration of the simulation.

Fig. 4.. Structural insights from comparisons of Lys, Arg, and His mutations at position 148.

Atomic models of the region surrounding the bound DTG from experimental structures of (A) E138K/G140A/Q148K, (B) E138K/G140A/Q148R, and (C) E138K/G140S/Q148H intasomes with DTG bound. (D) Overlay of the three experimentally determined structures. Relevant atomic distances are indicated in the snapshots.

Fig. 5.. Structure-based explanation for 4d potency.

(A and B) Atomic models of (A) DTG or (B) 4d from high-resolution structures of the E138K/G140A/Q148K intasome bound to either of the two drugs, respectively. The intasome is displayed as a surface view, while the drugs are displayed in ball-and-stick. The terminal adenine base is removed, for clarity. (C) Overlay of the two structures in (A) and (B), with all atoms displayed as ball-and-stick. Residues whose positions vary between the two structures are highlighted with a red asterisk. (D) Close-up of the cryo-EM density of the terminal adenosine and its interaction with the bound drug. When the intasome contains the E138K/G140A/Q148K mutation, experimental density is weak for conformer A, which stacks on top of the bound INSTI, when the ligand is DTG, but is pronounced when the ligand is 4d. (E) Quantitative comparison of the rotameric occupancies of the two conformers, A and B, in all structures derived in the current work. Only the structures in which 4d is bound clearly show pronounced conformer A occupancy. (F) Chemical schematic of the π-π stacking between the terminal adenine base and 4d, which helps to maintain the bound drug in the active site of the intasome. A top-down view of the π-π stacking within the structure is shown below. (G) Thermodynamic cycle used for estimating the ligand binding free energies to the target in the two conformers. (H) Estimates of the free energies for the stacked (brown) and the nonstacked (tan) configurations of the terminal adenine base, for WT (left) and E138K/G140A/Q148K (right) DRM intasomes.