A cell-based strategy to assess intrinsic inhibition efficiencies of HIV-1 reverse transcriptase inhibitors

Abstract

During HIV-1 reverse transcription, there are increasing opportunities for nucleos(t)ide (NRTI) or nonnucleoside (NNRTI) reverse transcriptase (RT) inhibitors to stop elongation of the nascent viral DNA (vDNA). In addition, RT inhibitors appear to influence the kinetics of vDNA synthesis differently. While cell-free kinetic inhibition constants have provided detailed mechanistic insight, these assays are dependent on experimental conditions that may not mimic the cellular milieu. Here we describe a novel cell-based strategy to provide a measure of the intrinsic inhibition efficiencies of clinically relevant RT inhibitors on a per-stop-site basis. To better compare inhibition efficiencies among HIV-1 RT inhibitors that can stop reverse transcription at any number of different stop sites, their basic probability, p, of getting stopped at any potential stop site was determined. A relationship between qPCR-derived 50% effective inhibitory concentrations (EC50s) and this basic probability enabled determination of p by successive approximation. On a per-stop-site basis, tenofovir (TFV) exhibited 1.4-fold-greater inhibition efficiency than emtricitabine (FTC), and as a class, both NRTIs exhibited an 8- to 11-fold greater efficiency than efavirenz (EFV). However, as more potential stops sites were considered, the probability of reverse transcription failing to reach the end of the template approached equivalence between both classes of RT inhibitors. Overall, this novel strategy provides a quantitative measure of the intrinsic inhibition efficiencies of RT inhibitors in the natural cellular milieu and thus may further understanding of drug efficacy. This approach also has applicability for understanding the impact of viral polymerase-based inhibitors (alone or in combination) in other virus systems.

FIG 1

Cell-based approach for monitoring reverse transcription vDNA products of various length by quantitative real-time PCR. (A) Schematic representation of HIV-1 reverse transcription stages. Minus-strand DNA synthesis (indicated in blue) is initiated from a ${\text{tRNA}}_3^{\text{Lys}}$ primer, and U5 and R regions of the vRNA (indicated in black) are first copied, forming a minus-strand strong-stop vDNA product. This minus-strand product is then transferred to the 3′ end of the vRNA and the U3 region is immediately copied. Continued minus-strand synthesis copies viral enzymatic (pol) and structural (gag) genes, while plus-strand synthesis (indicated in red) is initiated discontinuously from polypurine tracts (cPPT and PPT) in the opposite direction. This plus-strand strong-stop vDNA product is then transferred to the nascent end of the minus-strand vDNA and the primer-binding site (PBS) region is fully copied. (B) Linear representation of completed minus-strand vDNA products of different lengths monitored by qPCR. Locations of all TaqMan primer-probes are represented by the arrowheads. Conventional primer sets monitor completion of early/short (RU5 and U3) and late/long (GAG and PBS) vDNA products. Unconventional primer sets monitor completion of intermediate (1K, 2K, 4K, 6K, and 8K) vDNA products, where “K” represents approximately 1,000 nucleotides in length. The actual length of each vDNA product is indicated adjacent to each primer-probe set.

FIG 2

Kinetics of reverse transcription during a single cycle of infection. MT-2 cells were pretreated (−24 h) with 10-fold-adjusted (10×) EC₅₀ levels of TFV (▽), FTC (□), or EFV (○) prior to infection, along with a no-drug control (⧫). At indicated time points postinfection, cells were lysed, and vDNA product copy numbers quantified in replicate by qPCR were expressed as a percentage of the response in the no-drug control-treated cells measured at 10 h postinfection and set to 100%. The impact of RT inhibitors was assessed for early reverse transcription of short vDNA products by monitoring for minus-strand strong stop (RU5) (A) and late reverse transcription of long vDNA products by monitoring for second-strand transfer (PBS) (B). Data represent the means ± standard errors of the means of two independent experiments.

FIG 3

Relative accumulation of vDNA products of different lengths. MT-2 cells treated with 1-fold- or 10-fold-adjusted EC₅₀s of TFV, FTC, or EFV were harvested at 6 h postinfection. Each vDNA product was quantified in replicate by qPCR and expressed as a percentage of the response of the same product quantified in the absence of drug, set to 100%. Data represent the means ± standard errors of the means of two independent experiments, fit by nonlinear regression using a single-exponential decay function. Best-fit parameters and values are shown in Table 1.

FIG 4

Stochastic process of RT inhibition at potential stop sites. A set of events {s₁, s₂, s₃} for RT to be stopped at sites 1, 2, and 3 and another set of events {ns₁, ns₂, ns₃} for RT not to be stopped at sites 1, 2, and 3 are defined.