Akt inhibitors as an HIV-1 infected macrophage-specific anti-viral therapy
© Chugh et al. 2008
Received: 12 December 2007
Accepted: 31 January 2008
Published: 31 January 2008
Skip to main content
© Chugh et al. 2008
Received: 12 December 2007
Accepted: 31 January 2008
Published: 31 January 2008
Unlike CD4+ T cells, HIV-1 infected macrophages exhibit extended life span even upon stress, consistent with their in vivo role as long-lived HIV-1 reservoirs.
Here, we demonstrate that PI3K/Akt inhibitors, including clinically available Miltefosine, dramatically reduced HIV-1 production from long-living virus-infected macrophages. These PI3K/Akt inhibitors hyper-sensitize infected macrophages to extracellular stresses that they are normally exposed to, and eventually lead to cell death of infected macrophages without harming uninfected cells. Based on the data from these Akt inhibitors, we were able to further investigate how HIV-1 infection utilizes the PI3K/Akt pathway to establish the cytoprotective effect of HIV-1 infection, which extends the lifespan of infected macrophages, a key viral reservoir. First, we found that HIV-1 infection activates the well characterized pro-survival PI3K/Akt pathway in primary human macrophages, as reflected by decreased PTEN protein expression and increased Akt kinase activity. Interestingly, the expression of HIV-1 or SIV Tat is sufficient to mediate this cytoprotective effect, which is dependent on the basic domain of Tat – a region that has previously been shown to bind p53. Next, we observed that this interaction appears to contribute to the downregulation of PTEN expression, since HIV-1 Tat was found to compete with PTEN for p53 binding; this is known to result in p53 destabilization, with a consequent reduction in PTEN protein production.
Since HIV-1 infected macrophages display highly elevated Akt activity, our results collectively show that PI3K/Akt inhibitors may be a novel therapy for interfering with the establishment of long-living HIV-1 infected reservoirs.
A hallmark of HIV pathogenesis is the loss of CD4+ T cells in HIV-1 infected patients. Infected CD4+ T cells initially undergo cell cycle arrest at G2 caused by a viral accessory protein, Vpr, and eventually cytolysis [1, 2]. However, the cell fate and molecular consequences of non-dividing target cells of HIV-1 such as macrophages and microglia are poorly understood. We recently reported that in contrast to HIV-1 infected CD4+ T cells, infection in primary human macrophages and a microglial cell line (CHME5) leads to an extended life span and elevated survival against apoptotic stresses . We also showed that in the HIV-1 transduced CHME-5 microglial cell line, this cytoprotective phenotype is induced by intracellular expression of HIV-1 Tat, which plays a primary role in the transcriptional activation of the HIV-1 LTR [4, 5].
HIV-1 infected microglia, brain macrophages, are known to secrete various toxic products such as the Tat and Envelope (Env) proteins, which lead to the death of neighboring neurons and eventually HIV-1 associated dementia (HAD) in the infected host [6–9]. In addition to the secretion of viral proteins, it is known that in the central nervous system (CNS) HIV-1 infected microglia produce nitric oxide (NO), which contributes to the establishment of a highly apoptotic environment in close proximity to infected microglia [10–12]. Even though non-dividing HIV-1 target cells are exposed to these toxic conditions nearby, it has been reported that both microglia and tissue macrophages continue to produce virus for prolonged periods of time. Indeed, a number of studies have suggested that these non-dividing HIV-1 target cells serve as long-living viral reservoirs [13–15].
The PI3K/Akt cell survival pathway has been extensively studied, and has been recognized as a promising target for anti-cancer therapies because its activation is a key cellular event during tumorigenesis . Once PI3K and Akt kinase are activated upon apoptotic stress, they further transduce signals to a series of downstream regulators of cell survival. In its normal state, the PI3K/Akt pathway is negatively regulated by PTEN (phosphatase tensin homolog), which converts PIP3 to PIP2 . We recently observed in our microglial cell line model, that the PI3K inhibitors wortmannin and LY294002 were able to render HIV-1 infected CHME5s susceptible to cell death following an apoptotic stimulus .
In this report, we employed primary human macrophages, an important HIV-1 target cell type and viral reservoir, and investigated the specific molecular mechanisms involved in the modulation of the PI3K/Akt pathway. Importantly, we provide virological evidence that supports the application of anti-PI3K/Akt reagents as a potential anti-HIV-1 strategy to eradicate long-living HIV-1 infected human macrophages and to prevent HIV-1 production from these viral reservoirs.
Since the Akt pathway is a well-characterized pathway for cell survival and HIV-1 infected macrophages exhibit an enhanced survival phenotype, we next tested whether the delayed viral production following exposure to Akt inhibitor and stress (SNP) was related to the induction of cell death in HIV-1 infected macrophages. For this test, we quantified cell death under four experimental conditions using the Live/Dead assay (Figure 1E). This assay uses fluorescent dyes which distinguish live cells (green) from dead cells (red) on the basis of intracellular esterase activity (viable) and incorporation of the ethidium homodimer (nonviable). As expected, treatment with either SNP or either of the PI3K/Akt inhibitors alone did not induce significant amounts of cell death in infected macrophages. However, HIV-1 infected macrophages exposed to both SNP and the PI3K/Akt inhibitors clearly displayed a high percentage of cell death (as shown by the extensive red staining in Figure 1E; results for wortmannin, Akt inhibitor VIII and Miltefosine were similar; data not shown). Macrophages treated with heat-inactivated YU-2 underwent high levels of cell death following SNP treatment and combined treatment with inhibitor and SNP, further supporting our observation of an extended survival phenotype in HIV-1 infected macrophages (data shown in previous manuscript, ). The percentage of cell death induced under each condition is shown below the panel (Figure 1E). These data suggest that the decrease in viral production is secondary to the induction of cell death, following exposure to PI3K/Akt inhibitors.
We also examined PTEN mRNA levels following transduction of our pseudotyped HIV-GFP vector in macrophage by reverse transcriptase PCR (RT-PCR). As shown in Figure 2D, pseudotyped HIV vector-transduced macrophages displayed drastically decreased levels of mRNA compared to the heat-inactivated vector control. The observed decrease in mRNA was more pronounced than the decrease in PTEN protein levels. This is probably because the half life of endogenous PTEN protein is relatively long at about 30 hours . Collectively, these data demonstrate that the cytoprotective effect in macrophages upon infection with HIV-1 YU-2 or transduction with HIV-GFP is likely due to the downregulation of PTEN mRNA and protein levels, which can facilitate the activation of the PI3K/Akt survival pathway.
Since plasma membrane recruitment of Akt kinase typically results in increased phosphorylation and activation of Akt, we hypothesized that HIV-1 infection might lead to an increase in Akt kinase activity. Once phosphorylated and activated, the Akt kinase phosphorylates a series of downstream signals including GSK3β [28–31]. To test our hypothesis, we prepared cell lysates from HIV-GFP transduced macrophages and employed an Akt kinase activity assay which uses active Akt kinase from cell lysates to phosphorylate GSK3β substrate. As shown in Figure 3B, macrophages transduced with HIV-GFP displayed an approximately 40-fold increase in Akt kinase activity over the cells treated with heat-inactivated vector.
We also tested Akt activity in the CHME5 cell line (Figure 3B), and similar results were obtained although the increase in kinase activity was substantially less, due to a high level of basal Akt activity. Interestingly, pre-treatment of vector-transduced CHME5 cells with a potent Akt kinase inhibitor, Akt inhibitor IV , reduced Akt kinase activity to a basal level similar to that observed in cells treated with heat-inactivated vector (Figure 3B). This confirms that the HIV-1 induced increase in survival of both primary macrophages as well as CHME5 cells is likely a result of increased Akt kinase activity. Importantly, this data also supports that the decrease of HIV-1 production by inhibitor treatment, which was observed in Figure 1, is likely due to induction of cell death via inhibition of the Akt survival pathway
To test this, we performed an in vitro binding assay based on the hypothesized competition between Tat and PTEN for binding to p53 (Figure 4B). p53-containing cell lysates were incubated for a defined length of time with either an irrelevant control protein (BSA) or full-length Tat (101 amino acids) to allow binding. Each lysate was then incubated with normalized amounts of PTEN expressing cell lysates (10 μg). p53 complexes were collected by FLAG-tag immunoprecipitation and examined by Western blot analysis using antibodies directed against the C-terminal V5-tag of PTEN. As seen in Figure 4B, the binding of PTEN to p53 was drastically reduced following incubation with Tat, compared to the BSA control. Instead of using purified tat protein, we could have co-expressed Tat in CHME5 cells. However, since intracellular Tat can decrease p53 levels, this may make it technically difficult to pull down detectable levels of p53. These data support our model circuit in which intracellular Tat prevents PTEN from binding to p53 by interacting with the p53 C-terminal domain. This molecular competition event may facilitate the activation of the PI3K/Akt survival pathway in HIV-1 infected macrophages.
Next, we performed similar experiments using CHME5 cell sublines expressing either wildtype or one of two mutant Tat constructs, the basic domain mutant R49Q/K50E or the transactivation mutant C21G (cysteine-rich domain: see Figure 6A). As expected, the C21G Tat mutant exhibited a defect in transactivator function while the construct harboring the basic domain mutation failed to decrease p53 activity (data not shown). Importantly, the basic domain mutant retained transactivation activity similar to wildtype Tat (data not shown). The alterations in p53 activity following expression of Tat 49/50 could be due to the abrogation of the interaction between p53 and Tat, since this binding is known to occur through the basic domain . Next, we tested the survival ability of the CHME5 sublines (wild type and mutant) by exposing cells to E. coli lipopolysaccharide (LPS) and cycloheximide (CHX), and analyzing for the induction of cell death. As shown in Figure 5B, the C21G Tat mutant was still able to exert the cytoprotective effect of wildtype Tat in CHME5 cells while the R49Q/K50E basic domain mutant Tat failed to protect CHME5 cells from the apoptotic stress of LPS/CHX treatment.
We further tested the effect of these Tat mutants on cellular PTEN levels. For this, CHME5 sublines stably transfected with plasmids encoding pcDNA3.1, wild type, C21G or R49Q/K50E Tat protein were analyzed for PTEN expression by Western blotting. Levels of PTEN protein were normalized by α-tubulin protein levels. As shown in Figure 5C, CHME5 cells expressing the R49Q/K50E Tat mutant failed to decrease PTEN protein levels, while CHME5 cells expressing the C21G Tat mutant displayed reduced levels of PTEN similar to wildtype Tat. These data are consistent with the cytoprotective phenotypes of the cells expressing these mutants (Figure 5B). Together, these data suggest that mutations in the Tat basic domain may alter binding of HIV-1 Tat to p53, resulting in increased PTEN levels and consequently an increased incidence of cell death.
SIV and HIV Tat contain a stretch of conserved cysteine residues in the transactivation domain as well as a region rich in basic residues, as shown in the sequence comparison of the cysteine rich and basic domains of HIV-1 (YU-2), SIVmac239 and SIVPBJ Tat proteins (Figure 6A). Therefore, we tested whether the expression of SIV Tat could also induce extended survival of CHME5 cells. A plasmid expressing either the first exon of HIV-1 Tat (psvTat72: ) or SIVPBJ Tat was transfected into CHME5 cells. We also co-transfected a GFP expression plasmid to identify transfected cells expressing Tat. The transfected cells were exposed to LPS/CHX and their survival capability was monitored with the Live/Dead assay. As shown in Figure 6B, CHME5 cells expressing either psvTat72 or SIVPBJ Tat (GFP+) displayed enhanced survival as compared to control cells transfected with pcDNA3.1 and GFP (green). The percentage of only the transfected, GFP+ cells undergoing cell death (red) is shown below each panel. These data suggest that the C-terminal region of Tat encoded in exon 2 of the Tat gene is not required for the cytoprotective activity of Tat in CHME5 cells. These results also show that SIVPBJ Tat is capable of exerting a cytoprotective effect in CHME5 cells, supporting the possibility that Tat's effects on macrophage/microglial cell survival are conserved among lentiviruses.
Mutational studies revealed a novel cytoprotective role for the basic domain of Tat protein. We also observed a decrease in PTEN binding to p53 in the presence of intracellular Tat. Mutations in the basic domain of Tat likely interfere with the ability of HIV-1 Tat to bind p53, allowing stabilization of p53 by PTEN and increased PTEN levels, resulting in abrogation of the cytoprotective phenotype in primary macrophages. In addition, we found that SIV Tat was also capable of protecting CHME5 cells from death. This supports the possibility that Tat's cytoprotective function may be conserved among HIV-1 and SIV Tat proteins, and that these two lentiviruses may share a mechanism for promoting the extended survival of infected macrophages and microglia. Indeed, it is also known that SIV infected macrophages serve as a long-living viral reservoir . Therefore, SIV-infected macaque models may be promising in further developing Akt inhibitors as a novel antiviral therapeutic.
Most importantly, we also examined the ability of PI3K/Akt inhibitors to induce cell death specifically in HIV-1 infected macrophages exposed to SNP stress, which simulates the in vivo local toxic environment. A significant decrease in HIV-1 production from infected macrophages was observed upon combined treatment with SNP and PI3K/Akt inhibitors. This finding suggests that PI3K/Akt inhibitors may have utility as a potential new anti-HIV therapy that is able to specifically target non-dividing HIV-1 target cells such as macrophages, which play important roles in pathogenesis as long-lived HIV-1 reservoirs. Interestingly, infected macrophages treated with SNP or inhibitor alone did not display any signs of cell death or decreased viral production, whereas infected macrophages treated with both SNP and the PI3K/Akt inhibitors underwent cell death with little viral production. This observation indicates that the inhibitory effect of the PI3K/Akt inhibitors on viral production from infected macrophages requires a stressed environment-as may occur in vivo, in association with immune activation and cytokine production .
More interestingly, the Akt inhibitor Miltefosine, which has undergone multiple clinical trials and has been approved for treatment of breast cancer in Europe and parasite infections in other countries, was also able to inhibit viral production and cell survival in HIV-1 infected macrophages. In addition, we also found that another Akt inhibitor, Perifosine, which is also currently in clinical trials, was able to decrease viral production and induce cell death in HIV-1 infected macrophages (data not shown).
One interesting question is why HIV-1 infected CD4+ T cells undergo cell death. It is plausible that HIV-1 infection (and Tat expression) may promote cell cycle progression in dividing/activated CD4+ T cells. However, in these infected, dividing CD4+ T cells, due to the strong G2 cell cycle arresting activity of HIV-1 Vpr, further progression through the cell cycle and cell survival may be prevented, resulting in cytolysis [44, 45]. Another HIV-1 reservoir cell type is the HIV-1 infected resting memory CD4+ T cell [46, 47]. It would be interesting to investigate whether HIV-1 infection also activates the PI3K/Akt pathway in these cells, and if so, whether treatment with PI3K/Akt inhibitors results in elimination of these cells.
In addition to the large number of macrophage/microglia in the toxic environment of the CNS during infection, it has been reported that many of the cells producing HIV-1 in the lymph nodes, spleen and intestine of infected hosts are macrophages [48, 49]. These tissue macrophages are also known to persistently produce virus for a long period of time, serving as viral reservoirs. Therefore, it is possible that treatment with Akt inhibitors that are unable to cross the blood brain barrier (BBB) would result in eradication of these infected tissue macrophages. Interestingly, however, it was reported that alkyllysophospholipids such as Miltefosine are able to penetrate the BBB [50–53], which supports the potential use of Miltefosine to eradicate viral reservoirs of the CNS.
This study elucidates the molecular and cellular mechanisms involved in the cytoprotective effect of HIV-1 infection in primary human macrophages and indicates the PI3K/Akt pathway as a key contributor to this effect. It is increasingly apparent that many PI3K/Akt inhibitors under development as anti-cancer therapy are safe and well-tolerated in both experimental animals and humans [54–57]. Indeed, several inhibitors including Miltefosine have been approved for treatment of human cancers. This further supports the possible use of PI3K/Akt inhibitors for anti-HIV therapy and targeting of long-lived viral reservoirs.
Primary human monocyte-derived macrophages were isolated from human buffy coats and differentiated as previously described . The CHME5 microglial cell line was maintained as described previously . M-tropic HIV-1 YU-2 was prepared using human PBMCs , and VSV-G pseudotyped HIV-1 vectors expressing EGFP and all HIV proteins except Nef and Env were prepared as described  and used to transduce primary human macrophages. Vector titers were determined using CHME5 cells, and the p24 EIA was performed for each vector or virus preparation following manufacturer's protocol (PerkinElmer). The plasmid encoding the first exon of Tat, psvTat72, was obtained from the NIH AIDS reagent program. The p53-FLAG plasmid constructed by Dr. Thomas Roberts  was purchased from Addgene (plasmid 10838). A plasmid encoding the PTEN gene was a generous gift from Dr. Jim Miller (University of Rochester). Using this plasmid, a linker sequence followed by the V5 tag sequence was introduced by PCR. After construction of PTEN-V5 tag, the tagged gene was inserted into pcDNA3.1+Hygro (Invitrogen) using the Kpn I and Xho I restriction sites.
The EGFP-PHAkt fusion gene from pEGFP-PHAkt  was cloned into pShuttle-CMV prior to recombination into pAdEasy (Stratagene). Recombinant adenoviral stocks (Ad.CMV-EGFP-PHAkt) were then generated following transfection in HEK293A cells using methods provided by the supplier (Stratagene). The virus was purified by CsCl density gradient centrifugation and the viral titer was determined by real-time PCR on a BioRad icycler (Hercules, CA) using a Taqman probe and primers that amplified a small portion of the Adenovirus hexon gene .
Primary human macrophages (5 × 104 cells) were infected with HIV-1 YU-2 (MOI = 40) for 48 hours. Heat-inactivated YU-2 was used as a control. Cells were washed with DPBS and transduced with Ad.CMV-EGFP-PHAkt (MOI = 3000) for 24 hours. Positive control cells were treated with epidermal growth factor (EGF, Sigma) for 15 minutes before fixation with 3% formaldehyde. For inhibition of membrane localization, infected macrophages were treated with Miltefosine (10 μM) 24 hours post-infection for 48 hours prior to fixation. Macrophages were visualized for the localization of the PH domain of Akt by examining GFP fluorescence on a Leica microscope (200×).
Primary human macrophages (1 × 106) and CHME5 (1 × 106) cells were transduced with pseudotyped HIV-GFP vector (MOI of 40 for macrophages and MOI of 1 for CHME5 cells), giving ~95% transduction. CHME5 cells were treated with Akt inhibitor IV (.2 μM) for 24 hours following transduction with the HIV-GFP vector where specified. 48 hours post-transduction, cells were lysed using ELB lysis buffer before performing the Akt kinase activity assay (Cell Signaling) as per the manufacturer's protocol. Following incubation with the GSK3β fusion protein, 6 × SDS stop buffer was added and samples were loaded onto an SDS 8% (w/v) polyacrylamide gel and then transferred to nitrocellulose membrane (Hybond, Amersham Biosciences). Using the antibodies supplied, GSK3β-P levels were detected by Western blot analysis. Protein levels were normalized using β-tubulin as a loading control. Each assay was performed in triplicate.
Macrophages were transduced with either the HIV-GFP vector or an adenoviral vector expressing GFP +/- Tat. 24 hours post adenoviral transduction or 5 days post HIV vector transduction, cells were lysed for RNA isolation. cDNA was then synthesized from the RNA samples using the Qiagen cDNA synthesis kit (Qiagen, CA) as per the manufacturer's protocol. RT-PCR was then performed using the following primers for PTEN: F primer – 5' TTTGAAGACCATAACCCACCA 3'; R primer – 5' CCATAGAAATCTAGGGCCTCT 3'. The β-actin RT-PCR was performed with the primers as previously described .
Cell lysates were prepared in ELB buffer supplemented with protease inhibitors (Sigma) and phosphatase inhibitor cocktail (Sigma) and samples (10–20 μg) were applied to an SDS 8% (w/v) polyacrylamide gel. The expression of the proteins of interest was detected by probing with the PTEN (138G6) rabbit monoclonal antibody (Cell Signaling) or M2 FLAG mouse antibody (Sigma, 1:1000). Donkey anti-rabbit Ig or sheep anti-mouse Ig (Amersham Biosciences, 1:5000) for secondary antibody followed by ECL detection using the SuperSignal West Femto kit (Pierce). For a loading control, blots were probed for α-tubulin (Cell Signaling) followed by sheep anti-mouse IgG (Amersham Biosciences). Expression of the protein of interest in each sample was normalized to either β-tubulin or p53-FLAG levels for analysis using ImageJ software (NIH), and ratios were determined from experiments in triplicate.
CHME5 cells were transfected with either a plasmid encoding p53-FLAG or PTEN-V5 tag. After 24 hours, cells were lysed in ELB buffer. Following normalization of protein concentration of each lysate, p53-FLAG-containing lysate (10 μg) was incubated with Tat101 protein (Xeptagen, 2 μg/ml) and allowed to bind for 30 minutes at 4°C on a rocking platform. Following this incubation, an equal protein amount (10 μg) of PTEN-V5 lysate was added and allowed to bind for an additional 30 minutes at 4°C. The combined lysates were then applied to EZview Red anti-FLAG M2 affinity gel (Sigma) for immunoprecipitation as per the manufacturer's protocol. After elution of the p53 complexes with 6× SDS buffer, lysates were applied to an SDS 8% (w/v) polyacrylamide gel and Western analysis was performed as described above. PTEN expression was detected using a mouse anti-V5 tag antibody (Serotec, 1:1000) followed by sheep anti-mouse IgG (Amersham Biosciences, 1:5000). ECL detection was then performed as described above and the membrane was reprobed for p53-FLAG as a loading control. Each binding ratio was performed in triplicate.
Sodium nitroprusside (Sigma) was used at 1 mM for treatment of macrophages. The live/dead assay was then performed as previously described  and percent cell death is shown. For analysis of induction of cell death in the CHME5 Tat and control sublines, cells were exposed to CHX (10 μg/ml) and E. coli serotype O26:B6 LPS (Sigma) at a concentration of 50 μg/ml or 100 μg/ml for 24 hours. For the live/dead assay, cells were cotransfected with a plasmid encoding GFP and a construct expressing either SIVPBJ Tat or psvTat72 (NIH AIDS Reagent Program). Following treatment with LPS (50 μg/ml) and CHX (10 μg/ml), cells were analyzed for cell death using the Live/Dead Cytotoxicity/Viability assay (Molecular Probes) as per the manufacturer's protocol. Images were taken at a magnification of 200 × 24 h post-treatment using a fluorescence microscope (Leica). Each assay was performed in triplicate.
The control (pcDNA3.1) and wildtype Tat (pTat101) expressing CHME5 sublines were constructed as previously described . For creation of the transactivation mutant C21G, the sequence at amino acids 20/21 was changed to GCCGGC by site-directed mutagenesis while amino acids 49 and 50 (Arg and Lys) were changed to CAGGAG (Glu and Gln) to create the basic domain mutant. The C21G and 49/50 Tat mutants were then cloned into the pcDNA3.1+Hygro plasmid (Invitrogen). To create stably expressing CHME5 sublines, cells were transfected with the above constructs and selected using hygromycin for two weeks. The resulting CHME5 sublines were referred to as C21G (transactivation mutant) and Tat 49/50 (basic domain mutant). Three independent subline clones were isolated and used for characterization. The cell lines were tested for transactivation activity using a luciferase assay measuring activation of the HIV-1 LTR. A plasmid encoding the HIV-1 LTR promoter region fused to a luciferase cassette was transfected into the CHME5 cell lines and 24 hours later, cells were lysed and the luciferase assay was performed as previously described .
In order to construct the pseudotyped vector system expressing the Tat 49/50 mutant, the Quickchange II XL kit (Stratagene) was used. The following primers were used to target amino acids 49 and 50 of the Tat gene, changing the AGGAAG sequence to CAGGAG: F primer – 5' GGCATCTCCTATGGCCAGG AGAAGCGGAGACAGCG 3'; R primer – 5' CGCTGTCTCCGCTTCTCCTGG CCATAGGAGATGCC 3'. The sequence of the vector was confirmed by sequencing.
Primary human macrophages were infected with HIV-1 YU-2 (1 × 105 pg). Three days post-infection, cells were washed with DPBS to eliminate the presence of virus. After washing, cells were cultured either in media alone, media containing SNP (1 mM) only, media supplemented with the PI3K inhibitor wortmannin (100 nM, Sigma) or the specific Akt inhibitors IV, VIII (200 nM, 105 nM, Calbiochem)  or Miltefosine (5 uM, Cayman Chemicals) only or a combination of inhibitors and SNP. Infected macrophages were cultured for 12 days, during which time viral supernatants were collected and fresh media with inhibitors or SNP was also added every 3 days. The p24 levels contained in each viral supernatant sample at the various time points was monitored using the p24 enzyme-linked immunosorbent assay (Beckman-Coulter) according to the manufacturer's protocol. Viral production was plotted over time and asterisks denote undetectable p24 levels. On day 12, cells were analyzed for the induction of cell death using the live/dead assay as described above. Bright fields and merged (red + green) fields are shown. Green cells represent viable cells while red-stained cells correspond to dead or dying cells. The average percentage of cell death induced was determined from three independent experiments and is shown with the SD.
This study was supported by NIH grants (BK, R01 AI049881; PC, AI064136; SD, PO1 MH064570; VP, R01 AI49057; SM, RO1 NS 054578; CM, CNPQ fellowship 133328/2006-6 (Brazil)).
This article is published under license to BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.