Intracellular transactivation of HIV can account for the decelerating decay of virus load during drug therapy

Abstract

Basic virus dynamics models have been essential in understanding quantitative issues of HIV replication. However, several parts of the viral life cycle remain elusive. One of the most critical steps is the start of viral transcription, which is governed by the regulatory protein trans-activator of transcription (Tat) that induces a positive feedback loop. It has been shown that this feedback loop can alternate between two states leading to a transient activation of viral transcription. Using Monte Carlo simulations, we integrate the transactivation circuit into a new virus dynamics model having an age-dependent transactivation rate and reversion into latency. The cycling of infected cells between an activated and latent state results in the typical decelerating decay of virus load following therapy. Further, we hypothesize that the activation of latently infected cells is governed by the basal transcription rate of the integrated provirus rather than the intra- or extracellular environment. Finally, our systems approach to modeling virus dynamics offers a promising framework to infer the extracellular dynamics of cell populations from their intracellular reaction networks.

Figure 1

Dynamics of transactivation and reversion into latency. (A) Representative time plots from stochastic simulation runs of three HIV-infected cells. Each color depicts the concentration of protein P (solid lines) and protein trans-activator of transcription (Tat) (dashed lines) within one infected cell. The dotted line indicates the threshold of transactivation. (B) Probability distribution and age-dependent rate of transactivation. A small amount of pre-integration transcripts (mRNA_n(0)=5) results in early transactivation of HIV-infected cells. (C) Probability distribution and age-dependent rate of reversion. After a time window of a few days, the transactivated cells revert into a latent state at a constant rate (blue line). (D) Probability distribution and age-dependent rate of transactivation of latently infected cells. Reactivation of latently infected cells is determined by the slowest process in the transactivation circuit (basal transcription) and therefore occurs at a constant rate. LTR=1 and all other species are zero except otherwise indicated. k_b=10⁻⁶ s⁻¹ for all simulations.

Figure 2

An age-dependent transactivation model predicts biphasic viral load decline during antiretroviral therapy. Numerical simulations of the partial differential equation (PDE) model from Equation (1) accurately describe the first two phases. Recently infected cells become transactivated (black line) causing the initial shoulder phase of the viral load after treatment (red line). The virus load decays proportional to the number of activated virus-producing cells (blue line). The first phase of decline is followed by a slower second phase, which is caused by reactivation of latently infected cells (green line). Initially, the simulation is run to approach an equilibrium before the number of new infections, I(0,t), is set to zero (dotted line). Parameters are: r=0.019 d⁻¹, α=0.039 d⁻¹, δ=0.7 d⁻¹, μ=0.0 d⁻¹ and p/δ_V=100.

Figure 3

Intracellular trans-activator of transcription (Tat) transactivation kinetics predict a decelerating decay during prolonged antiretroviral therapy. (A) Viral load decay within the first 45 days. The first phase results primarily from the death of activated virus-producing cells, whereas the second phase is determined by the reactivation of latently infected cells. Experimental estimates of the decay rates during the two phases are given by the dashed (Perelson et al, 1997) and dotted (Palmer et al, 2008) lines. (B) Decelerating decay during prolonged periods of therapy approximates experimental observations (dotted lines, Palmer et al, 2008). (C) Similarly, the pool of latently infected cells decays and approximates experimental estimates (dashed lines, Zhang et al, 1999; Finzi et al, 1999; Ramratnam et al, 2000; Siliciano et al, 2003). Parameters used: initial pool of activated, virus-producing cells A₀=3.1 × 10⁷ cells (Chun et al, 1997), initial pool of latently infected cells I₀=1.4 × 10⁶ cells (Chun et al, 1997), death rate of activated, virus-producing cells δ=0.7 d⁻¹ (Perelson et al, 1997) and for simplicity we assume lifelong persistence of the resting CD4⁺ cells, i.e., μ=0.0 d⁻¹ per day. In this figure, f(α) is a normal distribution with μ=0.2 and σ=0.1, truncated at zero and renormalized.