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.

INTRODUCTION

HIV-1 and other retroviruses integrate a DNA copy of their viral RNA genome into the genome of the host cell, a process that is mediated by the virally encoded enzyme integrase (IN) (1, 2). The integrated proviruses serve as templates for transcription of new viral RNA genomes and the mRNAs required for production of progeny virions. The catalytic integration reaction that inserts a DNA copy of the viral genome in the host genome presents a target for treating people living with HIV-1 (PLWH) and for preventing new infections.

Two chemical steps are required for integration: (i) 3′-end processing, in which IN cleaves two nucleotides from each 3′-end of the viral DNA, and (ii) DNA strand transfer, in which IN inserts the ends of the viral DNA into target DNA in a one-step transesterification reaction (3). Completion of integration requires removal of two nucleotides from the 5′-ends of the viral DNA, filling in of the single strand gaps between viral and target DNA, and ligation. These latter steps are mediated by cellular enzymes. IN is both necessary and sufficient to carry out both 3′-processing and DNA strand transfer in vitro. Each step of DNA integration, up to formation of the integration intermediate, occurs within a series of stable nucleoprotein complexes (intasomes), involving a multimer of IN and a pair of viral DNA ends. The atomic structures have now been determined for a variety of retroviral intasomes (4–13), including HIV-1 (14, 15), revealing differences in the oligomeric assemblies and how they engage target DNA. The formation of intasome species and the intasome-mediated insertion of the viral DNA (vDNA) into the host genome are essential steps in the viral replication cycle [see (1, 2) for recent reviews].

Because integration is an essential step in viral replication, intasomes are targeted by an important class of antiretroviral drugs, the IN strand transfer inhibitors (INSTIs). INSTIs selectively interact with both the bound viral DNA and IN and, as a consequence, do not bind tightly to free IN (4, 16, 17). INSTIs work by chelating the two Mg²⁺ ions, blocking the active site and inhibiting the strand transfer activity of IN. Five INSTIs have been clinically approved by the U.S. Food and Drug Administration and antiretroviral regimens that include INSTIs are standard of care. These include first-generation INSTIs raltegravir (RAL, approved 2007) and elvitegravir (EVG, approved 2011), as well as second-generation INSTIs dolutegravir (DTG, approved 2013), bictegravir (BIC, approved 2018), and cabotegravir (CAB, approved 2021). INSTIs are widely considered to be among the best therapeutic options for use in combination therapies to treat HIV-infected patients (18).

The major difference between first- and second-generation INSTIs lies in their ability to inhibit mutant forms of IN. Whereas both RAL and EVG can potently inhibit the wild-type (WT) enzyme, drug-resistant mutations (DRMs) quickly develop (19). The second-generation INSTIs can potently inhibit many of the mutant forms of IN that arise during the course of treatment with the first-generation inhibitors (20–22). However, clinical data indicate that resistance to second-generation INSTIs is now a growing problem (23–25). Unfortunately, because HIV-1 replicates rapidly and the viral load is high, DRMs can arise, and viral sequences within an infected individual can differ by 10% or more (26). These data highlight the challenges to developing effective treatments. They also point to the fact that, while resistance is a problem, it can be addressed when the mechanisms are understood and by using this information to develop compounds that inhibit both the WT and the DRM enzymes.

The most frequently encountered DRMs to INSTI therapy include mutations at position Q148H/K/R. The Q148 mutations are commonly found in combination with the G140A/S mutations (23–25, 27) [and, rarely, G140C (27)]. In addition to the pair of DRMs at positions 140/148, numerous other DRMs arise in and around the active site, frequently including E138K (27). Recent structural work using the simian immunodeficiency virus (SIV_rcm) IN as a model for HIV-1 IN led to a proposed mechanism for the resistance of the G140S/Q148H double mutant (28). The mutation Q148H arises first, expelling a water molecule from the secondary coordinating shell of the Mg²⁺ ions and introducing a bulky electropositive residue underneath the bound ligand. The subsequent mutation of the neighboring residue G140S acidifies the Nε2 of H148. Coupling of the residues S140-H148-E152 increases the partial positive charge near the Mg²⁺ ligand. Collectively, these changes weaken chelation of the bound INSTI, leading to a reduction in drug binding. Because there are many complex DRMs that have a change at Q148 (H/K/R), Mg²⁺ chelation of Mg²⁺ was suggested to be the Achilles’ heel of this drug class. However, SIV_rcm serves as a proxy for HIV-1; their active sites, including residues that are encountered in drug-resistant viral clones, are not identical. Moreover, only the double mutant G140S/Q148H was structurally defined, but there are many combinations of mutations at position 148H/K/R which can be paired with the G140A/S mutations and with other nearby mutations, including E138K. Thus, the full consequences of these and related mutations, and the structural basis of resistance in HIV-1 IN, which should help us understand how to develop INSTIs that can overcome resistance caused by mutations at these positions, remain unclear.

INSTIs containing a naphthyridine core have been developed into potent IN inhibitors that rival or even outperform clinical drugs in terms of their ability to inhibit a broad range of DRMs (20, 21, 29, 30). For example, the compound 4d was reported to inhibit certain DRMs, including variants with 140/148 mutations, with over an order of magnitude higher potency than the clinically used drug DTG (21). Structural biology data have shed light on how naphthyridine-containing compounds bind to intasomes from HIV-1 and prototype foamy virus (PFV) (15, 29), revealing insights into the small but important differences in the ways INSTIs bind to PFV and HIV-1 IN (15). However, how and why these compounds retain potency against DRMs remains unclear. Thus, there is a pressing need not only to explain the underlying mechanisms of viral resistance but also to understand why compounds like 4d can potently inhibit resistant forms of IN.

Defining the underlying mechanisms of resistance lags behind clinical tabulation of emerging variants. This is primarily due to a lack of high-resolution structures of resistant variants and complementary tools to probe the resistance mechanisms. There have also been discrepancies in resistance profiles recorded for the 140/148 positions and their behavior in combination with other DRMs. Since mutations in HIV-1 IN that confer drug resistance occur in several distinct combinations, structures of HIV-1 intasomes bound to inhibitors with and without such mutations, complemented by insights on inhibitory potencies and enzyme/viral fitness, are required to understand the details of how these mutations confer resistance and to guide the design of drugs with improved resistance profiles.

Here, we sought to understand the mechanism(s) of resistance caused by mutations at positions 138, 140, and 148. Our results reveal weaknesses in one of the leading clinically used drugs, DTG, and explain the mechanism(s) that underlie its loss of potency against DRMs. We extend these findings to explain how and why mutations at the three residue substitutions lead to drug resistance. Last, we provide insight into how a leading investigational compound, which is currently undergoing preclinical testing, retains considerable potency against the set of DTG-resistant variants we analyzed. Collectively, these data provide mechanistic insights into both drug resistance and its subversion and will enable more direct pathways to designing INSTIs with improved resistance profiles.

RESULTS

Inhibitory potencies of DTG and a leading developmental INSTI 4d against resistant mutants

Novel developmental INSTIs based on the naphthyridine scaffold maintain potency against drug-resistant variants with mutations at position Q148 of HIV-1 IN (15, 20, 21, 29–31). However, whether their potency is maintained across the Q148H/K/R family of DRMs, particularly in association with the E138K and G140A/S mutations, remains unclear. We determined the ability of a leading clinically used second-generation INSTI, DTG, to inhibit mutant viruses containing all possible combinations of E138K + G140A/S + Q148H/K/R DRMs using single-round viral infection assays and compared the data to the most potent preclinical developmental INSTI 4d (Fig. 1A)…