AIDS RESEARCH AND HUMAN RETROVIRUSES Volume 6, Number 12, 1990 Mary Ann Liebert, Inc., Publishers
Activation of Human Immunodeficiency Virus Oxidative Stress SYLVIE
LEGRAND-POELS,1
2
Type 1 by
DOLORES VAIRA,3 JOËL PINCEMAIL,4 ALBERT VAN DE and JACQUES PIETTE1-2
VORST,2
ABSTRACT An important aspect of the infection by the human immunodeficiency virus (HIV-1) type 1 is its clinical latency, suggesting that the virus itself or the provirus may remain latent for extended periods of time after primary infection. Certain heterologous viral proteins or chemical and physical agents are able to reactivate latent virus. Since a common denominator shared by these agents is the ability to cause stress response in cells, we have examined the effects of oxidative stress mediated by hydrogen peroxide (H202) on HIV-1 latently infected promonocytic cell line termed Ul. After exposure to H202 in concentrations ranging from 0.1 to 2 mM, the viability of the Ul cells decreased during 24 h before recovery. At 24 h post stress, the Ul cells began to express virus as assessed by elevated reverse transcriptase activities in culture supernatants. Immunofiuorescence carried out on stressed Ul cells using anti-HIV-1 polyclonal antibodies showed that H202 leads to HIV-1 gene expression activation, but not to a release of viral particles from damaged cells. Additionally, using a HeLa cell line containing integrated the bacterial chloramphenicol acetyl transferase (CAT) gene under the control of the HIV-1 long terminal repeat (LTR), we have shown that oxidative stress mediated by H202 allows transactivation of the viral LTR revealed by intracellular CAT activity. A stimulation factor of around 4 of CAT activity can be reached when these cells are treated with 0.5 mM H202. A lower stimulation factor of CAT activity can be obtained when oxidative stress is mediated by stimulated phagocytes. During inflammatory reactions, blood cells containing latent virus can be exposed to activated oxygen species released by stimulated leukocytes and such a stress could convert a latent into a productive infection, leading to the development of the acquired immune deficiency syndrome.
INTRODUCTION
activation signals,6"9 by certain heterologous viruses like the human herpes simplex virus type 1, ' ° by certain cytokines, " '3 and by physical agents,13~17 suggesting that such external agents can influence the rate of HIV-1 multiplication in the organism. A common denominator shared by several agents is the ability to cause stress responses in cells.1819 A recent study17 has demonstrated that ultraviolet (UV) irradiation, probably damages DNA, which elicits an SOS-like stress response in mammalian cells, thus activating the NFkB transcription factor, and in this way allows its binding to the HIV long terminal repeat (LTR) major enhancer element. The present study has been undertaken to determine if oxidative stress on "
HUMAN
immunodeficiency virus
(HIV) infection is clini-
cally characterized by a long latency period between the time of infection and the onset of the acquired immune deficiency syndrome (AIDS).1 One of the limiting steps in the development of AIDS can be considered as being the activation of the latent provirus. Latent virus is found in cells carrying the CD4 surface antigen, including T-helper cells,2,3 macrophages and their multinucleated derivatives (e.g., the Langerhans cells of the epidermis3-4 and microglia).s In experimental systems, the activation of HIV-1 transcription is achieved by T-cell
'Laboratory of Virology, Department of Microbiology, laboratory of Experimental Physics, 'AIDS Reference Laboratory, and "Laboratory of Biochemistry, University of Liege, Liege. Belgium. 1389
LEGRAND-POELS ET AL.
1390
containing cells can produce HIV-1 reactiva- phorbol myristate acetate (PMA, 8 x 10~8M)orthe¿V-formylmean by oxidative stress is an exposure to met-leu-phe chemotactic peptide (FMLP, 1 p.M) in PBS (pH oxygen-activated species like Superoxide anión (02~), hydro- 7.4) containing 7.5 mM glucose, Ca2+ (2 mM) and Mg2+ (0.5 gen peroxide (H202), hydroxyl radical (OH ), and singlet mM) at 37°C during 30 min. The capacity of respiratory burst oxygen ('02). These chemical entities have a wide potential to was assayed by monitoring the 0/ formation as mentioned injure cells by inducing oxidation products in membranes and in below. nucleic acids.20 The oxidative stress can occur in vivo during, for example, an inflammatory reaction, where the mononuclear radicals detection and polymorphonuclear (PMNs) leukocytes are stimulated and Oxygen then generate Superoxide anión by reducing molecular The determination of Superoxide anión was performed using oxygen.21 This radicalar species can in turn be dismutated into the Superoxide dismutase (SOD) -inhibitable reduction of ferrihydrogen peroxide, an oxidant capable of diffusing inside the cytochrome C.26 The substrate (75 p.M) was added to the cells to promote oxidation reactions at the DNA level.22 In phagocyte stimulation medium and its reduction was followed at addition to these oxygen-activated species, leukocyte activation 550 nm. Superoxide anions and hydroxyl radicals were also also leads to the degranulation phenomenon where other oxidant detected by electron paramagnetic resonance (EPR) using 5,5products, enzymes and Ieukotrienes are released.23 T-helper dimethyl-1-pyrroline-N-oxide (DMPO) as a spin trap. DMPO lymphocytes or other blood cells having integrated HIV-1 virus was added to the solution at 10 mM and EPR spectra were in their chromosomes act as in vivo cellular targets to such recorded using a Varían E9 spectrometer at microwave power of oxidative stress occurring during an inflammatory reaction. 20 mW, a modulation amplitude of 1 Gauss, a modulation In this study, we have investigated, first the effect of hydro- frequency of 9.5 GHz, and a field set at 3425 G. gen peroxide and then the effect of oxidative stress mediated by stimulated monocytes or PMNs. Hydrogen peroxide is the only the Ul cell line to hydrogen peroxide oxygen-species, generated extracellularly, which is able to Exposure of induce DNA strand breaks in many cell types.22 To determine Aliquotsof2.5 x 106U1 cells in 10 ml complete media were the ability of an oxidative stress to affect HIV-1 latency, we have placed in 25 cm2 culture flasks and incubated at 37°C for 24 h. used a nonexpressing promonocytic cell line termed Ul as a Then, various concentrations of H202 (from 0 to 2000 p.M) were target for the oxidation reactions.24 The Ul cell contains two added to the culture medium. Each day, after the oxidative integrated copies of the viral genome, and under normal growth stress, cells were counted using trypan blue exclusion, aliquots conditions shows minimal constitutive expression of the virus. of supernatant fluids and cells were obtained for both the Phorbol ester,24 certains cytokines."12 UV irradiation,16 or determination of a reverse transcriptase (RT) activity and HIV-1 heat shock13 can induce an expression of HIV-1 in these Ul antigen assays, respectively. At 24 and 72 h after exposure to HIV-1 provirus tion. What we
cells. After oxidative stress, an eventual HIV reactivation from Ul cell line could be detected by measuring the supernatant reverse transcriptase (RT) activity or by immunofluorescence (IF) of cytospun cell preparations for viral proteins. To characterize the molecular events leading to virus reactivation, we have also used an HeLa cell line carrying an integrated DNA cartridge containing the bacterial chloramphenicol acetyl'5 transferase (CAT) gene under the control of the HIV-1 LTR. This type of cell line could establish whether or not the viral reactivation occurs through a LTR transactivation.
MATERIALS AND METHODS
H202, the volume of culture medium was doubled in each flask.
HeLa HIV-1 CAT line stimulated phagocytes
exposed to H202
or
to
Various concentrations of H202 (0.1-1.5 mM) were added to approximately 106 HeLa cells in exponential growing conditions. In other experiments, various amounts (1 to 30 x 106) of freshly prepared phagocytes were stimulated with PMA or FMLP as described above and then immediately cocultivated with approximately 106 HeLa cells in exponential growing conditions for 20 h. The surviving cells were harvested 20 h after oxidative stress, counted, and CAT activities were determined on the whole cell extracts as mentioned below.
Established cell lines The HIV-1 latently infected promonocytic cell line, U124 was cultivated in RPMI-1640 medium supplemented with 10% fetal calf serum (FCS). The HeLa HIV-1 CAT line15 was supplied by Kris Valerie (Smith Kline and French Laboratories, King of Prussia, PA) and was grown in Dulbecco's minimal essential medium (DMEM) containing 10% FCS and MEM amino acids.
Monocytes,
PMN preparation, and stimulation
Human mononuclear leukocytes and PMN were isolated from fresh human blood as described by Borgeat and Samuelsson.25 Washed cells were suspended in phosphate-buffered saline (PBS) and counted. The stimulation was carried out using
Reverse
transcriptase (RT) assay
The RT assay was done according to Hoffman.27 Briefly, the virus was purified and concentrated by ultracentrifugation of supernatant fluids (2 h at 131,000 x g at 4°C). The pellets were resuspended in 25 pJ of TNE (10 mM Tris-HCl, 100 mM NaCl, 1 mM EDTA) containing Triton X-100 (0.1%) and left 30 min on ice. Then, 10 p.1 of the virus preparation was incubated at 37°C for 60-90 min with 40 pi of a solution containing 62.5 mM Tris-HCl (pH 7.8), 6.25 mM dithiothreitol, 6.25 mM MgCL, 180 mM KC1, 0.06% Triton X-100, 0.375 mM glutathione, 0.625 mM EGTA, 31 p.g/ml BSA, 2.5% ethylene glycol, 250 mU/ml poly rA-oligodT, and 125 p.Ci/ml[3H]deoxythymidine. The DNA was precipitated with 10% trichloroacetic acid (TCA)
INDUCTION OF HIV-1 FROM LATENCY
1391
presence of 10 p.g of plasmid with a Bio-Rad (Richmond, CA) gene puiser at 300V and 960 p.F. After electroporation, the cells were seeded in 75 cm2 flasks and cultured 72 h before CAT activity determination.
and filtered through 2.4 mm Whatman GF-A filters. Filters were rinsed with 0.01 M sodium pyrophosphate and 95% ethanol and then counted.
Immunofluorescence (IF) assay
CAT activity determination
Immunofluorescence (IF) assays for the detection of HIV-1 proteins in the cells were performed as follows. From each sample, an aliquot of cells was taken and washed twice in PBS. Next, 5 x 20 p.1 (approximately 5 times 2 x 104 cells) of each cell suspension was placed in 5 holes of a slide, dried, and fixed in acetone at -20°C for 10 min. Afterward, three holes were incubated for 30 min at 37°C in a 1:20 dilution of sera from a seropositive individual assessed by Western blotting, the fourth hole ina 1:20 dilution of an HIV-1-negative normal sera control, and the last hole in PBS. The slides were washed in PBS and incubated for another 30 min at 37°C with a 1:10 dilution of sheep antihuman IgG fluorescein labeled antibody and with an equal volume of 1% Evans blue solution. Slides were again washed in PBS and viewed for IF staining using a Leitz Laborlux D fluorescent microscope.
HeLa HIV-1 CAT cells Cells
CAT assays were performed as described by Gorman et al.28 according to Neumann et al.29 The last method is based on the solubility difference in a scintillation cocktail between the [14C]butyryl coenzyme A, which is the substrate of the reaction and the [14C]butyryl chloramphenicol formed by the enzymatic reaction. In this technique, only the [l4C]butyryl chloramphenicol diffuses through the organic phase and can be assayed by scintillation counting. or
RESULTS
Effects of H202 on viability and kinetics of Ul cell growth Initial experiments were designed to characterize the susceptibility of Ul cells to the lethal effect of H202. As shown in Figure 1A, chronically infected Ul cells are highly sensitive to H202, and small increases in H202 concentration lead to a proportional decrease in viability. The viability began to significantly decrease with low H202 concentration (200 p,M) and led
transfection by electroporation
around 50% confluence were trypsinized, collected by centrifugation, and washed with Hepes buffer. The cells were transfected by electroporation in 1 ml of Hepes buffer, in the at
B
H202 concentration ( mM
0)
E
Days post
treatment
FIG. 1. (A) Effects of H202 on the viability of Ul cells. Ul cells were exposed to varying doses of H202 at time 0, replaced in culture for 24 h and then counted by trypan blue exclusion. (B) Effects of H202 on cell proliferation. Ul cells were exposed to varying concentrations of H202 and immediately replaced in culture. They were fed on Day 1 and aliquots were taken daily for cell 1 mM H,0„ and -•- 2 mM HÄ 0.3 mM H70„ a- 0.5 mM H,0, counts, -o without H202, -Q-0 1 mM H202, -
1392
LEGRAND-POELS ET AL.
to a very low survival fraction above 3 mM. Optimal concentrations for an oxidative stress mediated by H202 were between 0.5 mM and 1 mM. The kinetics of cell proliferation also showed a consistent pattern after treatment with H202 and were also highly affected by H202 exposure (Fig. IB). There was an initial cell loss and a slowed growth over the first 24-48 h after H202 exposure, but recovery could be observed after 48 h (for H202 concentrations between 0.1 and 0.3 mM) except for treatments performed with higher concentrations, where a lower recovery could be depicted.
Effects ofH202
on
HIV-1 reactivation in Ul cells
The estimation of the effect of H202 on HIV-1 reactivation in Ul cells was determined by two different techniques: (1) by measurement of a reverse transcriptase (RT) activity in the culture supernatants and (2) by the appearance of intracellular viral antigen in the treated Ul cells. An increasing RT activity could be determined in U1 supernatant fluid after H202 exposure and this phenomenon turned out to be transient with a maximum arising after 72 h (Fig. 2A). Furthermore, a dose-response curve could be demonstrated (Fig. 2B). An increase of H202 concentration up to 1 mM produced a proportional stimulation of RT activity. Above this concentration, the degree of stimulation decreased, probably because of an excessive lethal effect of H202. Over the optimal concentration range, a four- to fivefold stimulation level was reached and these concentrations (0.5-1 mM) led to a viability below 70% (Fig. 1 A). Because the rate of recovery varied as a function of the H202 concentration, experiments were designed to determine RT activities on supernatants containing an identical number of stressed cells. In these conditions, a stimulation factor of around 6 could be observed
for survival rates lower than 20%. These results demonstrated that HIV-1 reactivation from latently infected Ul cells, estimated by the RT activity in the cell supernatants, could be achieved by oxidative stress mediated by H202 and that the triggering factors were probably cellular lesions leading to a decreased viability. To ensure that the appearance of RT activity was due to a stimulation of HIV genes expression and not to a release of viral particles from cells which were damaged by oxidative stress, we attempted to detect an increase of intracellular HIV proteins synthesis by immunofluorescence (IF) assay. Figure 3 shows a representative effect of H202 on the expression of HIV-1 antigens in Ul cells. It was evident that a treatment with H202 at concentrations between 0.5 and 1 mM led to an increase in the frequencies of IF-positive cells and an optimum could be depicted 48 h after the oxidative stress. Positive controls were carried out by stimulation with PMA.
Effects of Hj02 HIV-1
on
CAT~cells
CAT gene
expression
in HeLa
In an attempt to define the molecular steps involved in the HIV-1 reactivation by H202, we used as cellular target, an established cell line (HeLa HIV-1 CAT cells) carrying an integrated DNA cartridge containing the bacterial chloramphenicol acetyl transferase (CAT) gene under the control of the HIV-1 long terminal repeat (LTR). A polyadenylation signal was introduced in the 3' end of the CAT gene. The HeLa HIV-1 CAT cells were then subjected to H202 and after 20 h, the CAT activity was determined on cellular extracts. CAT gene-expression increased in cellular extracts after H202 exposure and, as for the Ul cells, a dose-response curve could be demonstrated
B
o X
E
E
Q-
o
o.
o
>.
u
ra i—
rr
rr
Days post
treatment
H202concentration! mM
(A) Time course appearance of a viral RT activity in Ul cells supernatant fluid in response to H202 exposure. Ul cells exposed to 700 pM H202 (-0-) or were not treated (-X-). Every day after the oxidative stress, approximately 6 x 105 cells were taken, centrifugated and the supernatant fluids were assayed for evidence of RT activity. (B) Effects of different concentrations of H202 on the induction of virus expression by U1 cells. RT assays were performed 72 h after the oxidative stress. For each condition, RT activity was measured on 1 ml of supernatant fluid. FIG. 2. were
INDUCTION OF HIV-1 FROM LATENCY
1393
Immunofluorescence detection of the production of intracellular viral antigens 48 h after the U1 cells stimulation by 0.5 mM H202. After the treatment, the cells were washed, cytospun fixed and incubated with high titered sera from an AIDS patient followed by goat antihuman FITC antibodies. (A) Ul cells treated with 80 nM PMA. (B) Ul cells treated with 0.5 mM H202. (C) Untreated Ul cells. FIG. 3.
1394
LEGRAND-POELS ET AL.
(Fig. 4). Positive controls were carried out by transfecting these cells with a plasmid expressing the HIV-2 tat gene whose product was known to efficiently transactivate the HIV-1 LTR. Maximal stimulation of CAT gene expression ranged from 3- to 4-fold in at least three independent experiments and occurred when H202 concentration approached 500 p,M. This H202 concentration turned out not to be obtainable in vivo, although micromolar concentrations could be reached in the close proximity of stimulated leukocytes and appeared to be sufficient to promote DNA alterations in different target cells.22 Induction of CAT activity could be observed as soon as 5 h after the oxidative stress, and was maximal after 20 h. No CAT activity could be observed 40 h after treatment with H202. These results clearly demonstrated that the first event in the HIV-1 reactivation following oxidative stress mediated by H202 was the LTR transactivation by cellular proteins.
Activation of Human Immunodeficiency Virus Oxidative Stress SYLVIE
LEGRAND-POELS,1
2
Type 1 by
DOLORES VAIRA,3 JOËL PINCEMAIL,4 ALBERT VAN DE and JACQUES PIETTE1-2
VORST,2
ABSTRACT An important aspect of the infection by the human immunodeficiency virus (HIV-1) type 1 is its clinical latency, suggesting that the virus itself or the provirus may remain latent for extended periods of time after primary infection. Certain heterologous viral proteins or chemical and physical agents are able to reactivate latent virus. Since a common denominator shared by these agents is the ability to cause stress response in cells, we have examined the effects of oxidative stress mediated by hydrogen peroxide (H202) on HIV-1 latently infected promonocytic cell line termed Ul. After exposure to H202 in concentrations ranging from 0.1 to 2 mM, the viability of the Ul cells decreased during 24 h before recovery. At 24 h post stress, the Ul cells began to express virus as assessed by elevated reverse transcriptase activities in culture supernatants. Immunofiuorescence carried out on stressed Ul cells using anti-HIV-1 polyclonal antibodies showed that H202 leads to HIV-1 gene expression activation, but not to a release of viral particles from damaged cells. Additionally, using a HeLa cell line containing integrated the bacterial chloramphenicol acetyl transferase (CAT) gene under the control of the HIV-1 long terminal repeat (LTR), we have shown that oxidative stress mediated by H202 allows transactivation of the viral LTR revealed by intracellular CAT activity. A stimulation factor of around 4 of CAT activity can be reached when these cells are treated with 0.5 mM H202. A lower stimulation factor of CAT activity can be obtained when oxidative stress is mediated by stimulated phagocytes. During inflammatory reactions, blood cells containing latent virus can be exposed to activated oxygen species released by stimulated leukocytes and such a stress could convert a latent into a productive infection, leading to the development of the acquired immune deficiency syndrome.
INTRODUCTION
activation signals,6"9 by certain heterologous viruses like the human herpes simplex virus type 1, ' ° by certain cytokines, " '3 and by physical agents,13~17 suggesting that such external agents can influence the rate of HIV-1 multiplication in the organism. A common denominator shared by several agents is the ability to cause stress responses in cells.1819 A recent study17 has demonstrated that ultraviolet (UV) irradiation, probably damages DNA, which elicits an SOS-like stress response in mammalian cells, thus activating the NFkB transcription factor, and in this way allows its binding to the HIV long terminal repeat (LTR) major enhancer element. The present study has been undertaken to determine if oxidative stress on "
HUMAN
immunodeficiency virus
(HIV) infection is clini-
cally characterized by a long latency period between the time of infection and the onset of the acquired immune deficiency syndrome (AIDS).1 One of the limiting steps in the development of AIDS can be considered as being the activation of the latent provirus. Latent virus is found in cells carrying the CD4 surface antigen, including T-helper cells,2,3 macrophages and their multinucleated derivatives (e.g., the Langerhans cells of the epidermis3-4 and microglia).s In experimental systems, the activation of HIV-1 transcription is achieved by T-cell
'Laboratory of Virology, Department of Microbiology, laboratory of Experimental Physics, 'AIDS Reference Laboratory, and "Laboratory of Biochemistry, University of Liege, Liege. Belgium. 1389
LEGRAND-POELS ET AL.
1390
containing cells can produce HIV-1 reactiva- phorbol myristate acetate (PMA, 8 x 10~8M)orthe¿V-formylmean by oxidative stress is an exposure to met-leu-phe chemotactic peptide (FMLP, 1 p.M) in PBS (pH oxygen-activated species like Superoxide anión (02~), hydro- 7.4) containing 7.5 mM glucose, Ca2+ (2 mM) and Mg2+ (0.5 gen peroxide (H202), hydroxyl radical (OH ), and singlet mM) at 37°C during 30 min. The capacity of respiratory burst oxygen ('02). These chemical entities have a wide potential to was assayed by monitoring the 0/ formation as mentioned injure cells by inducing oxidation products in membranes and in below. nucleic acids.20 The oxidative stress can occur in vivo during, for example, an inflammatory reaction, where the mononuclear radicals detection and polymorphonuclear (PMNs) leukocytes are stimulated and Oxygen then generate Superoxide anión by reducing molecular The determination of Superoxide anión was performed using oxygen.21 This radicalar species can in turn be dismutated into the Superoxide dismutase (SOD) -inhibitable reduction of ferrihydrogen peroxide, an oxidant capable of diffusing inside the cytochrome C.26 The substrate (75 p.M) was added to the cells to promote oxidation reactions at the DNA level.22 In phagocyte stimulation medium and its reduction was followed at addition to these oxygen-activated species, leukocyte activation 550 nm. Superoxide anions and hydroxyl radicals were also also leads to the degranulation phenomenon where other oxidant detected by electron paramagnetic resonance (EPR) using 5,5products, enzymes and Ieukotrienes are released.23 T-helper dimethyl-1-pyrroline-N-oxide (DMPO) as a spin trap. DMPO lymphocytes or other blood cells having integrated HIV-1 virus was added to the solution at 10 mM and EPR spectra were in their chromosomes act as in vivo cellular targets to such recorded using a Varían E9 spectrometer at microwave power of oxidative stress occurring during an inflammatory reaction. 20 mW, a modulation amplitude of 1 Gauss, a modulation In this study, we have investigated, first the effect of hydro- frequency of 9.5 GHz, and a field set at 3425 G. gen peroxide and then the effect of oxidative stress mediated by stimulated monocytes or PMNs. Hydrogen peroxide is the only the Ul cell line to hydrogen peroxide oxygen-species, generated extracellularly, which is able to Exposure of induce DNA strand breaks in many cell types.22 To determine Aliquotsof2.5 x 106U1 cells in 10 ml complete media were the ability of an oxidative stress to affect HIV-1 latency, we have placed in 25 cm2 culture flasks and incubated at 37°C for 24 h. used a nonexpressing promonocytic cell line termed Ul as a Then, various concentrations of H202 (from 0 to 2000 p.M) were target for the oxidation reactions.24 The Ul cell contains two added to the culture medium. Each day, after the oxidative integrated copies of the viral genome, and under normal growth stress, cells were counted using trypan blue exclusion, aliquots conditions shows minimal constitutive expression of the virus. of supernatant fluids and cells were obtained for both the Phorbol ester,24 certains cytokines."12 UV irradiation,16 or determination of a reverse transcriptase (RT) activity and HIV-1 heat shock13 can induce an expression of HIV-1 in these Ul antigen assays, respectively. At 24 and 72 h after exposure to HIV-1 provirus tion. What we
cells. After oxidative stress, an eventual HIV reactivation from Ul cell line could be detected by measuring the supernatant reverse transcriptase (RT) activity or by immunofluorescence (IF) of cytospun cell preparations for viral proteins. To characterize the molecular events leading to virus reactivation, we have also used an HeLa cell line carrying an integrated DNA cartridge containing the bacterial chloramphenicol acetyl'5 transferase (CAT) gene under the control of the HIV-1 LTR. This type of cell line could establish whether or not the viral reactivation occurs through a LTR transactivation.
MATERIALS AND METHODS
H202, the volume of culture medium was doubled in each flask.
HeLa HIV-1 CAT line stimulated phagocytes
exposed to H202
or
to
Various concentrations of H202 (0.1-1.5 mM) were added to approximately 106 HeLa cells in exponential growing conditions. In other experiments, various amounts (1 to 30 x 106) of freshly prepared phagocytes were stimulated with PMA or FMLP as described above and then immediately cocultivated with approximately 106 HeLa cells in exponential growing conditions for 20 h. The surviving cells were harvested 20 h after oxidative stress, counted, and CAT activities were determined on the whole cell extracts as mentioned below.
Established cell lines The HIV-1 latently infected promonocytic cell line, U124 was cultivated in RPMI-1640 medium supplemented with 10% fetal calf serum (FCS). The HeLa HIV-1 CAT line15 was supplied by Kris Valerie (Smith Kline and French Laboratories, King of Prussia, PA) and was grown in Dulbecco's minimal essential medium (DMEM) containing 10% FCS and MEM amino acids.
Monocytes,
PMN preparation, and stimulation
Human mononuclear leukocytes and PMN were isolated from fresh human blood as described by Borgeat and Samuelsson.25 Washed cells were suspended in phosphate-buffered saline (PBS) and counted. The stimulation was carried out using
Reverse
transcriptase (RT) assay
The RT assay was done according to Hoffman.27 Briefly, the virus was purified and concentrated by ultracentrifugation of supernatant fluids (2 h at 131,000 x g at 4°C). The pellets were resuspended in 25 pJ of TNE (10 mM Tris-HCl, 100 mM NaCl, 1 mM EDTA) containing Triton X-100 (0.1%) and left 30 min on ice. Then, 10 p.1 of the virus preparation was incubated at 37°C for 60-90 min with 40 pi of a solution containing 62.5 mM Tris-HCl (pH 7.8), 6.25 mM dithiothreitol, 6.25 mM MgCL, 180 mM KC1, 0.06% Triton X-100, 0.375 mM glutathione, 0.625 mM EGTA, 31 p.g/ml BSA, 2.5% ethylene glycol, 250 mU/ml poly rA-oligodT, and 125 p.Ci/ml[3H]deoxythymidine. The DNA was precipitated with 10% trichloroacetic acid (TCA)
INDUCTION OF HIV-1 FROM LATENCY
1391
presence of 10 p.g of plasmid with a Bio-Rad (Richmond, CA) gene puiser at 300V and 960 p.F. After electroporation, the cells were seeded in 75 cm2 flasks and cultured 72 h before CAT activity determination.
and filtered through 2.4 mm Whatman GF-A filters. Filters were rinsed with 0.01 M sodium pyrophosphate and 95% ethanol and then counted.
Immunofluorescence (IF) assay
CAT activity determination
Immunofluorescence (IF) assays for the detection of HIV-1 proteins in the cells were performed as follows. From each sample, an aliquot of cells was taken and washed twice in PBS. Next, 5 x 20 p.1 (approximately 5 times 2 x 104 cells) of each cell suspension was placed in 5 holes of a slide, dried, and fixed in acetone at -20°C for 10 min. Afterward, three holes were incubated for 30 min at 37°C in a 1:20 dilution of sera from a seropositive individual assessed by Western blotting, the fourth hole ina 1:20 dilution of an HIV-1-negative normal sera control, and the last hole in PBS. The slides were washed in PBS and incubated for another 30 min at 37°C with a 1:10 dilution of sheep antihuman IgG fluorescein labeled antibody and with an equal volume of 1% Evans blue solution. Slides were again washed in PBS and viewed for IF staining using a Leitz Laborlux D fluorescent microscope.
HeLa HIV-1 CAT cells Cells
CAT assays were performed as described by Gorman et al.28 according to Neumann et al.29 The last method is based on the solubility difference in a scintillation cocktail between the [14C]butyryl coenzyme A, which is the substrate of the reaction and the [14C]butyryl chloramphenicol formed by the enzymatic reaction. In this technique, only the [l4C]butyryl chloramphenicol diffuses through the organic phase and can be assayed by scintillation counting. or
RESULTS
Effects of H202 on viability and kinetics of Ul cell growth Initial experiments were designed to characterize the susceptibility of Ul cells to the lethal effect of H202. As shown in Figure 1A, chronically infected Ul cells are highly sensitive to H202, and small increases in H202 concentration lead to a proportional decrease in viability. The viability began to significantly decrease with low H202 concentration (200 p,M) and led
transfection by electroporation
around 50% confluence were trypsinized, collected by centrifugation, and washed with Hepes buffer. The cells were transfected by electroporation in 1 ml of Hepes buffer, in the at
B
H202 concentration ( mM
0)
E
Days post
treatment
FIG. 1. (A) Effects of H202 on the viability of Ul cells. Ul cells were exposed to varying doses of H202 at time 0, replaced in culture for 24 h and then counted by trypan blue exclusion. (B) Effects of H202 on cell proliferation. Ul cells were exposed to varying concentrations of H202 and immediately replaced in culture. They were fed on Day 1 and aliquots were taken daily for cell 1 mM H,0„ and -•- 2 mM HÄ 0.3 mM H70„ a- 0.5 mM H,0, counts, -o without H202, -Q-0 1 mM H202, -
1392
LEGRAND-POELS ET AL.
to a very low survival fraction above 3 mM. Optimal concentrations for an oxidative stress mediated by H202 were between 0.5 mM and 1 mM. The kinetics of cell proliferation also showed a consistent pattern after treatment with H202 and were also highly affected by H202 exposure (Fig. IB). There was an initial cell loss and a slowed growth over the first 24-48 h after H202 exposure, but recovery could be observed after 48 h (for H202 concentrations between 0.1 and 0.3 mM) except for treatments performed with higher concentrations, where a lower recovery could be depicted.
Effects ofH202
on
HIV-1 reactivation in Ul cells
The estimation of the effect of H202 on HIV-1 reactivation in Ul cells was determined by two different techniques: (1) by measurement of a reverse transcriptase (RT) activity in the culture supernatants and (2) by the appearance of intracellular viral antigen in the treated Ul cells. An increasing RT activity could be determined in U1 supernatant fluid after H202 exposure and this phenomenon turned out to be transient with a maximum arising after 72 h (Fig. 2A). Furthermore, a dose-response curve could be demonstrated (Fig. 2B). An increase of H202 concentration up to 1 mM produced a proportional stimulation of RT activity. Above this concentration, the degree of stimulation decreased, probably because of an excessive lethal effect of H202. Over the optimal concentration range, a four- to fivefold stimulation level was reached and these concentrations (0.5-1 mM) led to a viability below 70% (Fig. 1 A). Because the rate of recovery varied as a function of the H202 concentration, experiments were designed to determine RT activities on supernatants containing an identical number of stressed cells. In these conditions, a stimulation factor of around 6 could be observed
for survival rates lower than 20%. These results demonstrated that HIV-1 reactivation from latently infected Ul cells, estimated by the RT activity in the cell supernatants, could be achieved by oxidative stress mediated by H202 and that the triggering factors were probably cellular lesions leading to a decreased viability. To ensure that the appearance of RT activity was due to a stimulation of HIV genes expression and not to a release of viral particles from cells which were damaged by oxidative stress, we attempted to detect an increase of intracellular HIV proteins synthesis by immunofluorescence (IF) assay. Figure 3 shows a representative effect of H202 on the expression of HIV-1 antigens in Ul cells. It was evident that a treatment with H202 at concentrations between 0.5 and 1 mM led to an increase in the frequencies of IF-positive cells and an optimum could be depicted 48 h after the oxidative stress. Positive controls were carried out by stimulation with PMA.
Effects of Hj02 HIV-1
on
CAT~cells
CAT gene
expression
in HeLa
In an attempt to define the molecular steps involved in the HIV-1 reactivation by H202, we used as cellular target, an established cell line (HeLa HIV-1 CAT cells) carrying an integrated DNA cartridge containing the bacterial chloramphenicol acetyl transferase (CAT) gene under the control of the HIV-1 long terminal repeat (LTR). A polyadenylation signal was introduced in the 3' end of the CAT gene. The HeLa HIV-1 CAT cells were then subjected to H202 and after 20 h, the CAT activity was determined on cellular extracts. CAT gene-expression increased in cellular extracts after H202 exposure and, as for the Ul cells, a dose-response curve could be demonstrated
B
o X
E
E
Q-
o
o.
o
>.
u
ra i—
rr
rr
Days post
treatment
H202concentration! mM
(A) Time course appearance of a viral RT activity in Ul cells supernatant fluid in response to H202 exposure. Ul cells exposed to 700 pM H202 (-0-) or were not treated (-X-). Every day after the oxidative stress, approximately 6 x 105 cells were taken, centrifugated and the supernatant fluids were assayed for evidence of RT activity. (B) Effects of different concentrations of H202 on the induction of virus expression by U1 cells. RT assays were performed 72 h after the oxidative stress. For each condition, RT activity was measured on 1 ml of supernatant fluid. FIG. 2. were
INDUCTION OF HIV-1 FROM LATENCY
1393
Immunofluorescence detection of the production of intracellular viral antigens 48 h after the U1 cells stimulation by 0.5 mM H202. After the treatment, the cells were washed, cytospun fixed and incubated with high titered sera from an AIDS patient followed by goat antihuman FITC antibodies. (A) Ul cells treated with 80 nM PMA. (B) Ul cells treated with 0.5 mM H202. (C) Untreated Ul cells. FIG. 3.
1394
LEGRAND-POELS ET AL.
(Fig. 4). Positive controls were carried out by transfecting these cells with a plasmid expressing the HIV-2 tat gene whose product was known to efficiently transactivate the HIV-1 LTR. Maximal stimulation of CAT gene expression ranged from 3- to 4-fold in at least three independent experiments and occurred when H202 concentration approached 500 p,M. This H202 concentration turned out not to be obtainable in vivo, although micromolar concentrations could be reached in the close proximity of stimulated leukocytes and appeared to be sufficient to promote DNA alterations in different target cells.22 Induction of CAT activity could be observed as soon as 5 h after the oxidative stress, and was maximal after 20 h. No CAT activity could be observed 40 h after treatment with H202. These results clearly demonstrated that the first event in the HIV-1 reactivation following oxidative stress mediated by H202 was the LTR transactivation by cellular proteins.
