Resistance to integrase inhibitors

Abstract

Integrase (IN) is a clinically validated target for the treatment of human immunodeficiency virus infections and raltegravir exhibits remarkable clinical activity. The next most advanced IN inhibitor is elvitegravir. However, mutant viruses lead to treatment failure and mutations within the IN coding sequence appear to confer cross-resistance. The characterization of those mutations is critical for the development of second generation IN inhibitors to overcome resistance. This review focuses on IN resistance based on structural and biochemical data, and on the role of the IN flexible loop i.e., between residues G140-G149 in drug action and resistance.

Keywords: AIDS; Elvitegravir; GSK-1265744; GSK-1349572; HIV-1 integrase; Raltegravir; interfacial inhibitors; resistance.

Figure 1

HIV-1 life cycle. HIV-1 first binds cells via interactions between the viral glycoproteins gp120/gp41 and the cellular receptor CD4 and CCR5 or CXCR4 co-receptors. After a conformational change of the complex, gp41 allows the fusion of both viral and cellular membranes, leading to the release within the cytoplasm of the viral core. The viral RNA (two copies of single-stranded RNA) is reverse-transcribed by the viral reverse transcriptase (RT) into double-stranded DNA. This DNA is then processed by the viral integrase (IN) and organized within a large nucleoprotein complex, the pre-integration complex (PIC). Following nuclear translocation, IN catalyzes the integration of this same DNA within the host genome (strand transfer, ST). After transcription, the viral RNA can be exported as genomic RNA (non-spliced) or as mRNAs (spliced RNA). The spliced RNA is translated into viral polyproteins that are cleaved by the viral protease (PR) and associate with genomic RNA at the cell membrane. Maturation by PR leads to new infectious particles released from the infected cell by budding. Drug targets include four different steps of viral replication: entry/fusion, reverse transcription, integration, and maturation (red). (Adapted from [3]).

Figure 2

The integration process. IN catalyzes the integration of the HIV DNA within the cellular DNA in two distinct steps. First, IN cleaves the dinucleotide GT at both 3′ extremities of the viral DNA (red) by 3′-processing (3′-P). After nuclear import, the strand transfer (ST) reaction leads to the integration of the viral DNA (green) into the cellular DNA (blue). Cellular proteins then repair the newly created junctions, cleave the overhangs and fill the gaps, duplicating five bases of the cellular DNA on each side.

Figure 3

HIV-1 and PFV IN structures. A. Comparison of the primary structures of HIV-1 and PFV IN. N-terminal (NTD) and C-terminal domains (CTD) are represented in light gray and catalytic core domains (CCD) in dark gray. The DDE motif is colored in red and mutations conferring resistance to RAL (positions 143, 148 and 155 for HIV-1; 212, 217 and 224 for PFV IN) are highlighted in blue. The flexible loop, comprising amino acids 140–149 for HIV-1 IN or 209–218 for PFV IN, is colored in green. B. Three-dimensional structure of HIV-1 and PFV IN core domains. Colors correspond to scheme A. In addition, amino acids 92 and 140 for HIV-1 IN and 161 and 209 for PFV IN are highlighted in light blue. Cartoon representations were obtained using MacPyMol version 0.99rc6 and the pdb file 2B4F (HIV-1 IN core domain, residue 57–207, with mutations F185K) and 3L2R (PFV IN complete structure with viral DNA and Mg/Zn cations, residue represented 123–269).

Figure 4

Amino acid alignment of the flexible loop region from different retroviruses. Alignment of the amino acid sequences corresponding to the region around the flexible loop of HIV-1, simian immunodeficiency virus (SIV), FIV, avian/Rous sarcoma virus ASV/RSV, xenotropic murine leukemia-related retrovirus (XMRV) and PFV. The flexible loop region is shaded in green. Residues identical in all the retroviruses are colored in red, and conserved residues with only one amino acid difference existing between the retroviruses are colored in blue. Numbers refer to the first residue for each sequence. Numbering at the top of the alignment corresponds to HIV-1 IN residues involved in resistance to RAL and to the catalytic glutamic acid (E152). Numbering for PFV IN is reported underneath the alignment.

Figure 5

Crystal structures of PFV IN in complex with INSTI. A. Three-dimensional structure of the PFV IN core domain in complex with its viral DNA substrate and RAL (left) or EVG (right) in the presence of magnesium. The core domain (amino acids 123–269) is colored in grey. Positions 212, 217 and 224, corresponding positions 143, 148 and 155 in HIV-1 IN are highlighted in blue. Positions 161 and 209, corresponding to secondary mutation positions 92 and 140 for HIV-1 IN, are highlighted in cyan. Cartoon representations were obtained using MacPyMol version 0.99rc6 and the pdb files 3L2T (RAL) and 3L2U (EVG) containing the PFV IN complete structure with viral DNA and Mg²⁺/Zn²⁺. B. Close view of the active site containing RAL (left) or EVG (right). Only the residues of PFV IN and the bases of the viral DNA within 5 Å of the drug are represented as sticks. Colors are similar to panel A. C. RAL (left) and EVG (right) chemical structures oriented as in the PFV IN structure in panel B.