Structural and inhibition studies of the RNase H function of xenotropic murine leukemia virus-related virus reverse transcriptase

Abstract

RNase H inhibitors (RNHIs) have gained attention as potential HIV-1 therapeutics. Although several RNHIs have been studied in the context of HIV-1 reverse transcriptase (RT) RNase H, there is no information on inhibitors that might affect the RNase H activity of other RTs. We performed biochemical, virological, crystallographic, and molecular modeling studies to compare the RNase H function and inhibition profiles of the gammaretroviral xenotropic murine leukemia virus-related virus (XMRV) and Moloney murine leukemia virus (MoMLV) RTs to those of HIV-1 RT. The RNase H activity of XMRV RT is significantly lower than that of HIV-1 RT and comparable to that of MoMLV RT. XMRV and MoMLV, but not HIV-1 RT, had optimal RNase H activities in the presence of Mn²⁺ and not Mg²⁺. Using hydroxyl-radical footprinting assays, we demonstrated that the distance between the polymerase and RNase H domains in the MoMLV and XMRV RTs is longer than that in the HIV-1 RT by ∼3.4 Å. We identified one naphthyridinone and one hydroxyisoquinolinedione as potent inhibitors of HIV-1 and XMRV RT RNases H with 50% inhibitory concentrations ranging from ∼0.8 to 0.02 μM. Two acylhydrazones effective against HIV-1 RT RNase H were less potent against the XMRV enzyme. We also solved the crystal structure of an XMRV RNase H fragment at high resolution (1.5 Å) and determined the molecular details of the XMRV RNase H active site, thus providing a framework that would be useful for the design of antivirals that target RNase H.

Fig 1

Dependence of RNase H activity on metal and time for HIV-1, MoMLV, and XMRV RTs. (A) A 50 nM concentration of Cy3-T_r35/P_d22 was incubated with 20 nM HIV-1 RT or MoMLV RT or 100 nM XMRV RT for the indicated times. Lengths of cleavage products are indicated. (B) Quantification of cleavage products shown in panel A. Percent RNA-DNA duplex was plotted as a function of time to determine the rate of cleavage, k (min⁻¹), by each enzyme.

Fig 2

Use of site-specific Fe²⁺ footprinting assay to determine the translocation state of RTs bound to DNA. The translocation state of RTs was determined using hydroxyl radical footprinting. A 50 nM concentration of 5′-Cy3-T_d43/P_d30-ddA chain terminated with ddA was incubated with HIV-1 RT (600 nM) or XMRV RT (1 μM) and a 0 to 500 μM concentration of the next incoming nucleotide (dTTP) to induce the formation of posttranslocation complexes or a 0 to 100 μM concentration of PFA to induce the formation of pretranslocation complexes. The complexes were treated for 5 min with ammonium iron sulfate (1 mM) and resolved on polyacrylamide–7 M urea gel. For HIV-1 RT, cleavages at positions −18 and −17 from the 3′-OH of the primer strand indicate pre- and posttranslocation complexes, respectively. In the case of XMRV RT, the corresponding complexes are at positions −18 and −19, suggesting a slightly longer distance between the polymerase and RNase H active sites.

Fig 3

Inhibition of RNase H activity of HIV-1 and XMRV RTs and isolated XMRV and p15-Ec HIV RNases H by RNHIs. (A) RNHIs used in this study. (B) A 20 nM concentration of RT or p15-Ec HIV RNase H or 200 nM XMRV RNase H was preincubated with increasing concentrations of RNHIs at room temperature for 5 min. Reactions were initiated by the addition of 100 nM RNase H substrate HTS-1 for the RTs (250 nM HTS-1 for XMRV and p15-Ec HIV RNases H) in the presence of 5 mM MgCl₂ (HIV-1 RT) or 0.5 mM MnCl₂ (XMRV enzymes and p15-Ec). After 20 min of incubation at 37°C, reactions were quenched with EDTA. The results from dose-response experiments were plotted using GraphPad Prism 4, and IC₅₀s were obtained at midpoint concentrations. (C) Reactions were performed as described above, with minor variation. A 20 nM concentration of enzyme and 100 nM RNase H substrate HTS-2 were used for the RTs, while 20 nM p15-Ec HIV RNase H or 200 nM XMRV RNase H and 250 nM HTS-2 substrate was used for the RNase H assays. Reaction mixtures were incubated for 20 min at 37°C.

Fig 4

Effect of preincubation conditions on inhibition of RNase H. We used HIV-1 RT (A), XMRV RT in the presence of Mg²⁺ (B), XMRV RT in the presence of Mn²⁺ (C), XMRV RNase H in the presence of Mg²⁺ (D), and XMRV RNase H in the presence of Mn²⁺ (E). Each reaction mixture contained 20 nM RT or 200 nM RNase H, 500 nM NAPHRHI, 5 mM MgCl₂ or 0.5 mM MnCl₂, and 250 nM RNA-DNA (HTS-1) substrate. Assays were performed as described in Materials and Methods. The enzymes were preincubated with inhibitor (NAPHRHI), MgCl₂ or MnCl₂, and/or RNA-DNA (HTS-1) substrate under the following conditions at 37°C for 5 min prior to starting the reaction: ■, enzyme plus NAPHRHI; ▲, enzyme plus NAPHRHI plus MgCl₂ or MnCl₂; ▼, enzyme plus NAPHRHI plus HTS-1; ♦, enzyme plus HTS-1; •, enzyme plus HTS-1 plus MgCl₂ or MnCl₂ (no-inhibitor control). The reactions were initiated by the addition of the missing components. The fluorescence signal was normalized to the highest fluorescence value of the uninhibited reaction to obtain percent activity.

Fig 5

Crystal structure of the isolated XMRV ΔC RNase H domain. (A) 2F_o-F_c electron density map of XMRV ΔC RNase H, displayed at σ = 2.0. (B) Stereo view of the isolated XMRV ΔC RNase H domain structure in cartoon representation. The structure contains an internal mixed β-sheet (shown in blue), composed of four parallel strands and one antiparallel strand, and four α-helices (shown in red). One Mg²⁺ ion (shown as a green sphere) is coordinated in the active site. Images were made using PyMOL (74).

Fig 6

Sequence alignment of XMRV, MoMLV, HIV-1, and B. halodurans (Bh) RNases H. All deleted residues from the XMRV ΔC RNase H structure are underlined in green. Residues contacting the RNA template are highlighted in red, residues contacting the DNA primer are highlighted in blue and residues contacting both the RNA template and DNA primer are highlighted in blue with red letters. Hydrophobic core residues are highlighted in gray. Conserved active-site residues are highlighted in yellow.

Fig 7

Comparison of the RNase H active sites of isolated XMRV RNase H ΔC and HIV-1 RT-dsDNA. (A) The active site of XMRV RNase H ΔC. One Mg²⁺ ion (green sphere) is coordinated by the conserved catalytic residues Asp524, Glu562, and Asp583 (shown as sticks) and two water molecules (red spheres [Wat1 and Wat2]). (B) The RNase H active site of HIV-1 RT-dsDNA (PDB ID, 3KJV). One Mg²⁺ ion (green sphere) is coordinated by the conserved catalytic residues Asp443, Glu478, and Asp498 (shown as sticks). The other conserved active-site residue, Asp549, is also shown. The dsDNA is shown in light gray. Images were made using PyMOL (74).

Fig 8

Interactions of RNases H with RNA-DNA substrates. (A) Molecular model showing potential interactions of XMRV ΔC RNase H (cyan cartoon) with an RNA-DNA substrate. The molecular model was built using the crystallographic coordinates of unliganded XMRV ΔC RNase H and an RNA-DNA oligonucleotide as described in Materials and Methods. The location of helix C (which would contact the primer strand) is shown by the arrow. (B) Cartoon representation of the HIV-1 RNase H complex with a polypurine tract RNA-DNA substrate (1HYS, orange cartoon). (C) Cartoon representation of the B. halodurans RNase H complex with an RNA-DNA substrate (1ZBI, blue cartoon). The RNA templates are shown as light gray sticks and spheres, while the DNA primers are shown as dark gray sticks and spheres. RNase H residues that interact with the nucleic acid are shown as space-filling spheres in their respective colors. Locations of the Mg²⁺ ions are shown as green spheres, and the RNA base containing the scissile phosphate is shown in pink. Images were made using PyMOL (74).