| |
Cell-to-cell spread of HIV permits ongoing replication despite antiretroviral therapy
|
| |
| |
Download the PDF here
Our data indicate that cell-to-cell spread is a likely source of intermittent ongoing replication in the face of ART, and that this is a consequence of some cell-to-cell infections transmitting virus numbers much in excess of what is required to infect a cell in the absence of ART. The large transmitted dose strongly decreases the probability that every transmitted virus will be inhibited by the drugs, and therefore greatly weakens their effect. This replication may adversely affect the immune system, increasing activation11, 12 and cell death13, and could potentially contribute to the maintenance of an HIV reservoir in locations such as lymphoid tissue where cell-to-cell spread occurs. This replication may adversely affect the immune system, increasing activation11, 12 and cell death13, and could potentially contribute to the maintenance of an HIV reservoir in locations such as lymphoid tissue where cell-to-cell spread occurs.
Cell-to-cell Spread of HIV Permits Ongoing Replication
A novel model for ongoing HIV replication in the face of ART is described in Nature by researchers at the California Institute of Technology (1). The persistence of HIV reservoirs could be explained by cell-to-cell spread of the virus that overwhelms drug concentrations in cells.
"How ongoing replication might take place in the presence of ART has remained unclear......The persistence of HIV reservoirs could be explained by cell-to-cell spread of the virus that overwhelms drug concentrations in cells.
Multiple infections occur in vivo and in vitro and are thought to be associated with cell-to-cell spread, a direct transmission mode that minimizes the number of virus particles failing to reach the target cell.
The authors therefore used a co-culture system with infected cells to generate cell-to-cell spread, in the absence or presence of Tenofovir.
Co-culture dramatically decreased sensitivity to drug: Tenofovir decreased cell-free infection 30 fold but caused less than a twofold decrease of coculture infection.
They also used a combination of Tenofovir and Efavirenz and found that cell-free infection was efficiently prevented by these drugs. But infections occuring by cell-to-cell transfer was much less affected by the drugs: at highest drug concentrations, transmission rate was 6 times higher than that of cell-free infection."
http://www.hiv-reservoir.net/index.php/Latest-News-on-HIV-Reservoirs-Eradication/hiv-propagation-on-art.html
-----------------------------------------
Cell-to-cell spread of HIV permits ongoing replication despite antiretroviral therapy - pdf attached
Alex Sigal1, Jocelyn T. Kim1,2, Alejandro B. Balazs1, Erez Dekel3, Avi Mayo3, Ron Milo4 & David Baltimore1
Aug 18 2011 Nature | Letter
Latency and ongoing replication1 have both been proposed to explain the drug-insensitive human immunodeficiency virus (HIV) reservoir maintained during antiretroviral therapy. Here we explore a novel mechanism for ongoing HIV replication in the face of antiretroviral drugs. We propose a model whereby multiple infections2, 3 per cell lead to reduced sensitivity to drugs without requiring drug-resistant mutations, and experimentally validate the model using multiple infections per cell by cell-free HIV in the presence of the drug tenofovir. We then examine the drug sensitivity of cell-to-cell spread of HIV4, 5, 6, 7, a mode of HIV transmission that can lead to multiple infection events per target cell8, 9, 10. Infections originating from cell-free virus decrease strongly in the presence of antiretrovirals tenofovir and efavirenz whereas infections involving cell-to-cell spread are markedly less sensitive to the drugs. The reduction in sensitivity is sufficient to keep multiple rounds of infection from terminating in the presence of drugs. We examine replication from cell-to-cell spread in the presence of clinical drug concentrations using a stochastic infection model and find that replication is intermittent, without substantial accumulation of mutations. If cell-to-cell spread has the same properties in vivo, it may have adverse consequences for the immune system11, 12, 13, lead to therapy failure in individuals with risk factors14, and potentially contribute to viral persistence and hence be a barrier to curing HIV infection.
Current antiretroviral therapy (ART) does not cure HIV infection because low-level viraemia persists from virus reservoirs that are insensitive to ART1. The reservoirs may be long-lived infected cells, cells with latent virus, ongoing cycles of infection termed ongoing replication, or a combination of sources1. How ongoing replication might take place in the face of ART has remained unclear. If ART succeeds in decreasing ongoing HIV replication to very low levels, why does it not eliminate replication completely? Here we explore a novel mechanism for ongoing HIV replication in the presence of ART.
Multiple infections of one cell may propagate at drug concentrations where infection by single particles would die out: if more virions are transmitted per cell, the probability that at least one of the virions escapes the drug should increase (Fig. 1a). To model the effect of multiple infections on drug sensitivity (Supplementary Theory, section 1), we assume infections by individual virions are independent events, each with a probability of escaping the drug and succeeding in infecting the cell. To quantify infection sensitivity to drugs, we introduce the transmission index (TX), which we define as the fraction of cells infected in the presence of drug (Id) divided by the fraction of cells infected in the absence of drug (I). Given: (1) a multiplicity of infection of m infectious units per cell, where m is defined as the product of virus particle number and the probability of infection per virus particle; (2) a concentration of antiretroviral agent d that reduces m by factor f(d), where f(d) ≥ 1. Under these conditions, the transmission index is:
TX has two important limiting regimes: m 1, in which case TX ≈ 1/f(d) and m/f(d) 1, in which case TX ≈ 1. In the first case, where few viruses infect each cell, the infection is sensitive to the effect of the drug, whereas in the second, where many viruses infect each cell, the infection is insensitive.
To test this, we infected the highly infection-permissive MT-4 T-cell line with cell-free HIV encoding yellow fluorescence protein (YFP)15 at low (0.2) and high (100) m in the presence of tenofovir (TFV), a nucleotide reverse transcriptase inhibitor. We determined infected cell number by YFP fluorescence (Supplementary Fig. 1) and observed that infection with cell-free virus at low m was sensitive to TFV across the range of concentrations used. At high m, infection was insensitive to low and intermediate TFV concentrations (Fig. 1b), supporting the model. Thus, multiple cell-free HIV infections per cell recapitulate the insensitivity to drug of an HIV reservoir.
Multiple infections occur in vivo2, 16 and in culture8, 10 and are thought to be associated with cell-to-cell spread2, 8, 9, 10, a directed transmission mode that minimizes the number of virus particles failing to reach the target cell. We therefore used co-culture with infected cells to generate cell-to-cell spread and compared drug sensitivity to infection with cell-free virus. Infection by co-culture occurs both by cell-free virus shed by infected donor cells and by cell-to-cell spread. Administration of cell-free virus lacks a cell-to-cell component-the measured average virus cycle time (1.7 days; Supplementary Fig. 2) would rarely permit cell-free virus infected cells to complete a second round of infection during the experiment (2 days). Therefore, we compared cell-free virus infection and the combination of cell-free virus infection and cell-to-cell spread resulting from co-culture. We used drugs that act far downstream of entry, to ensure any differences between cell-to-cell and cell-free infection are not due to factors that physically inhibit drug action in cell-to-cell spread.
We infected peripheral blood mononuclear cells (PBMCs) in the presence or absence of TFV by co-culture or using cell-free virus. To separate donor from target cells in co-culture, we used HLA-A2-negative donor cells and HLA-A2-positive targets (Supplementary Fig. 3a). Two days post-infection, we determined the fraction of target cells infected using p24 intracellular staining of HLA-A2-positive PBMCs (Fig. 2a, top panel, controls in Supplementary Fig. 3b). Co-culture dramatically decreased sensitivity to drug: TFV decreased cell-free infection ~30-fold but caused less than a twofold decrease of co-culture infection (Fig. 2b). The decline in HLA-A2 expression in the target cells after infection (Supplementary Fig. 3b) is consistent with observations that productive HIV infection downregulates HLA17.
We also used Rev-CEM18 reporter T cells as targets. These cells express green fluorescent protein (GFP) in the presence of HIV early proteins Tat and Rev (Supplementary Fig. 4). To infect Rev-CEM cells, we used either cell-free HIV or co-culture with infected MT-4 cells engineered to be >99% mCherry positive (Supplementary Fig. 5). We excluded GFP/mCherry double-positive cells from the analysis to avoid scoring fused cells as infected (Supplementary Figs 6 and 7). This underestimates co-culture infection because it excludes unfused cell doublets in the process of virus exchange.
To block infection, we applied TFV and the non-nucleoside reverse transcriptase inhibitor efavirenz (EFV) (Fig. 2a, bottom panel, Supplementary Fig. 7). At the highest concentrations used, co-culture TX was over sixfold higher than cell-free infection TX (Fig. 2b). The trend was similar when donors were PBMCs or Rev-CEM cells (Supplementary Fig. 8). Co-culture TX was lower than in PBMC-to-PBMC transmission, suggesting that target cells have an important role in cell-to-cell spread efficiency. The lower drug sensitivity in co-culture was not due to secreted donor cell factors that decrease the susceptibility of target cells to drugs (Supplementary Fig. 9).
We next determined the number of infectious units (m) transmitted. For co-culture, m was previously proposed to have a two-peaked Poisson distribution, one peak corresponding to cell-free virus or some low virus cell-to-cell transmissions, and the second to high virus number transmissions3, 9. We fit a two-peaked Poisson and other distributions to the data (Supplementary Theory, section 2). The two-peaked Poisson fit the data best (Fig. 2b, dotted line, Supplementary Fig. 10). The first peak mean was ~1 infectious unit for both drugs, with 94% and 97% of infections in this peak for TFV and EFV, respectively. The second peak mean was 73 (TFV) and 175 (EFV), with the remaining 6% and 3% of infections in this peak. This predicts that whereas most infections are cell-free or low virus cell-to-cell transmissions, a minority involve very large numbers of viruses. This might seem to imply large numbers of integrations in the absence of drug in the high virus number subset. Arguing against this is our observation of a significantly increased cell death rate with increasing numbers of multiple infections in the absence of drugs (data not shown). Inter-virus interference, such as downregulation of CD4 receptors19, may also limit provirus number.
To investigate whether cell-to-cell spread can lead to HIV replication through multiple virus cycles with ART, we measured the replication ratio (R), defined as fold change in the number of infected cells per virus cycle under conditions where target cells are not limiting: (Ik/I0)1/k. Here k is the number of elapsed virus cycles, Ik is the number of infected cells at virus cycle k, and I0 is the number of infected cells at the start. For expanding infections R > 1, whereas infections with R < 1 ultimately terminate20, 21. Although this assumes synchronized virus cycles, we simulated desynchronization and observed that its effect was negligible at the measured variability in cycle lengths (Supplementary Fig. 11).
To measure R, we tracked infection daily (Methods) in the absence of drug, with 100 μM TFV, or with a combination of EFV, TFV and the nucleoside reverse transcriptase inhibitor emtricitabine (FTC) at their clinical maximum plasma concentrations (Cmax: 10 μM EFV, 2 μM TFV and 10 μM FTC22). The fraction of infected cells was kept low to ensure that target cells were not limiting. R0, RTFV and , the replication ratios with no drug, TFV or at Cmax, were fitted from the data (Fig. 3a, dashed lines). They were 65, 2.5 and 0.95, respectively. RTFV was significantly greater than 1 (P < 0.01), indicating an expanding infection. was slightly lower than 1 in all experiments (Supplementary Fig. 12), indicating an infection slightly below the expansion threshold.
Given the lack of evolution in the plasma in individuals with HIV successfully suppressed by drugs23, 24, ongoing replication can occur if: (1) it is compartmentalized to other locations25, 26, (2) if it is intermittent; (3) the circulating virus is at a fitness maximum24; or some combination of these factors. We obtained . If this is extrapolated in vivo, it follows that ongoing replication cannot persist independently but may have a role if it interacts with another reservoir that primes replication27. To examine this scenario, we performed a stochastic simulation (Methods). As expected for intermittent replication, every infection chain that starts from the introduction of an infected cell from a different reservoir-for example, reactivation from latency-terminates (Supplementary Fig. 14). A constant input of one infected cell per virus cycle results in a steady state where substantial numbers of newly infected cells are generated, but the average number of mutations anywhere on the HIV genome per infected cell is low (~1; Fig. 3b). Because each infection chain is independent, these mutations are expected to be sporadic and not linked by temporal structure.
Evidence for ongoing replication during ART derives from the decrease in virus decline rates28, some HIV sequence divergence29 and long terminal repeat circle formation when the integrase inhibitor raltegravir is included in drug regimens11. At least in some individuals, antiretroviral suppression is close to the ongoing replication threshold: a mutation conferring very low-level resistance to EFV at therapy initiation30 is sufficient to cause ongoing replication, as indicated by increased virological failure risk14. Our data indicate that cell-to-cell spread is a likely source of intermittent ongoing replication in the face of ART, and that this is a consequence of some cell-to-cell infections transmitting virus numbers much in excess of what is required to infect a cell in the absence of ART. The large transmitted dose strongly decreases the probability that every transmitted virus will be inhibited by the drugs, and therefore greatly weakens their effect. This replication may adversely affect the immune system, increasing activation11, 12 and cell death13, and could potentially contribute to the maintenance of an HIV reservoir in locations such as lymphoid tissue where cell-to-cell spread occurs.
|
|
| |
| |
|
|
|
