(Received for publication, October 30, 1996, and in revised form, March 10, 1997)
From the a Centre de Recherche en Infectiologie and Département de Microbiologie, f Unité de Recherche en Pédiatrie and Département de Médecine, Centre Hospitalier Universitaire de Québec, Pavillon CHUL, Faculté de Médecine, Université Laval, Ste-Foy (Québec), Canada G1V 4G2 and the h Department of Medicine, McGill University, Montréal, Canada H3A 1A1
Replication of human immunodeficiency virus type
1 (HIV-1) is increased by different cytokines and T cell activators,
also known to modulate tyrosine phosphorylation levels. A novel class of protein tyrosine phosphatase (PTP) inhibitors, peroxovanadium (pV)
compounds, were tested for a putative effect on HIV-1 long terminal
repeat (LTR) activity. We found that these PTP inhibitors markedly
enhanced HIV-1 LTR activity in 1G5 cells, a stably transfected cell
line that harbors an HIV-1 LTR-driven luciferase construct. A direct
correlation between the extent of tyrosine phosphorylation and the
level of HIV-1 LTR inducibility was seen after treatment with three
different pV compounds. Transient transfection experiments were carried
out in several T cell lines, and after addition of pV, a marked
increase in HIV-1 LTR activity was measured. Monocytoid cells were
tested using U937-derived cell lines and were also found to be
sensitive to the pV-mediated potentiating effect on HIV-1 LTR activity.
A significant reduction of the pV-mediated increase in HIV-1 LTR
activity was seen in cells transiently transfected with an HIV-1
LTR-driven luciferase construct bearing a mutation in both NF-B
binding sites although detectable levels of induction remained.
Electrophoretic mobility shift assays allowed the identification of the
nuclear translocation of the NF-
B p50·p65 heterodimer complex
induced by pV compounds. A dominant negative version of the repressor
I
B
mutated on serines 32 and 36 impeded pV-induced NF-
B-dependent luciferase activity. Western blot
analysis showed a clear diminution in the protein level of I
B
starting 30 min after pV treatment of Jurkat E6.1 cells which is
indicative of its degradation. On the other hand, no increase in
tyrosine phosphorylation was observed on I
B
itself. Finally, we
tested the PTP inhibitors on four cell lines latently infected with
HIV-1 and showed a consistent pV-mediated increase in virion
production. Thus, our studies suggest that pV-mediated activation of
HIV-1 LTR activity is controlled by the nuclear translocation of the
NF-
B transcription factor, which is mediated by I
B
serine
phosphorylation and degradation, but also by a still undefined
NF-
B-independent pathway.
One of the most important aspects of the life cycle of the human
immunodeficiency virus type 1 (HIV-1)1 is
the modulation of its replication that is positively regulated by
several cytokines or T cell activators (1). Most of these HIV-1 long
terminal repeat (LTR) inducers act either completely or partially via
the nuclear factor kappa B (NF-B) transcription factor. Although
rarer, certain LTR activators are known to act through
NF-
B-independent pathways and include, for example, sodium butyrate,
an inducer potentially acting on relaxation of chromatin by histone
acetylation (2).
NF-B has been shown to regulate viral transcription via the two
NF-
B sites located in the HIV-1 LTR enhancer region (3). The
predominant form of this transcription factor has been identified as a
heterodimer made of 50-kDa (p50) and 65-kDa (p65 or RelA) subunits (4,
5). The heterodimeric NF-
B complex is sequestered in the cytoplasm
as an inactive precursor complexed with a repressor termed I
B
that masks the nuclear localization signal of the heterodimer complex
(6, 7). Formation of the active nuclear form of NF-
B composed of p50
and p65 subunits are induced after extra- or intracellular stimulation
of cells. These stimulants appear to act primarily by inducing the
release and degradation of I
B
subsequent to its phosphorylation
on both serine 32 and 36 residues (8), thereby allowing the rapid
translocation of NF-
B from the cytoplasm to the nucleus and binding
on regulatory regions of genes bearing the NF-
B binding sites (4, 6, 7).
Ever since it became evident that activation of NF-B involves the
release of the I
B
repressor by direct phosphorylation, it has
been attractive to evaluate the role of protein phosphorylation in
directly or indirectly affecting the NF-
B pathway. Overall, several
protein kinases have been shown to be important in the regulation of
NF-
B translocation including protein kinase A, protein kinase C,
Raf, and p59fyn (9-12), most of which have also been shown to
modulate HIV-1 LTR activity. Further evidence supporting the role
played by phosphorylation events in the control of NF-
B activity
came from the demonstration that okadaic acid, a selective inhibitor of
Ser/Thr phosphatase (13), induces NF-
B translocation in intact cells
(14).
Tyrosine phosphorylation levels have been shown to often increase in
signaling pathways leading to NF-B activation. The level of tyrosine
phosphorylation should thus be of importance in the control of the
pathway(s) leading to the activation and translocation of NF-
B.
Tyrosine phosphorylation is regulated by two opposing enzymatic
activities, namely protein tyrosine kinases (PTK) and protein tyrosine
phosphatases (PTP). These latter include a wide range of different
enzymes, CD45 being the major constituent of T cell surface PTP (15).
Modulation of intracellular phosphotyrosine content by PTP represents a
crucial mechanism by which the cellular response is regulated. This
fact is clearly demonstrated by studies using specific inhibitors
against PTP that have been shown to stimulate T cell activation
(16).
It is of prime importance to gain more knowledge of the molecular
events controlling the transcriptional state of the HIV-1 proviral DNA.
A large array of cellular factors have been shown to interact with
binding sites located within the 5 regulatory region of the viral
genome. Because an increase in both tyrosine phosphorylation and HIV-1
replication is seen following T cell activation, we were interested in
seeing whether the induction of intracellular tyrosyl phosphorylation
events, using new tyrosine phosphatase inhibitors, could modulate HIV-1
LTR activity. In the present study, we have made use of a set of pV
compounds, which represent new potent structurally defined PTP
inhibitors, and have demonstrated their capacity to markedly enhance
HIV-1 LTR activity upon incubation with both T and monocytoid cells. Our results show that the induction is partly
NF-
B-dependent and requires serine phosphorylation of
the I
B
repressor.
Cells were maintained in complete
culture medium made of RPMI 1640 medium supplemented with 10% fetal
bovine serum (Hyclone Laboratories, Logan, UT), glutamine (2 mM), penicillin G (100 units/ml), and streptomycin (100 µg/ml). The lymphoid T cell lines used include Jurkat E6-1 (17),
HPB-ALL (18), and Molt-4 clone 8 (19) cell lines. 1G5 and
Jurkat-tat are clonal cell lines derived from Jurkat E6.1
cells stably transfected with a luciferase gene driven by an HIV-1 long
terminal repeat and a tat expressing vector, respectively
(20, 21). J1.1 and ACH-2 are latently HIV-1-infected cell lines derived
from the parental cell lines Jurkat E6.1 and A3.01, respectively (22,
23). The promonocytic cell line U1 was cloned from U937 cells surviving
acute HIV-1 infection (24), and OM10.1 was similarly isolated from
HL-60 (25). HIV-1 LTR-CAT stably transfected cell lines U38 and M311 each originate from U937 and Molt-4 cells, respectively (26). HPB-ALL
cells were provided by Dr. A. Weiss (Howard Hughes Medical Center, San
Francisco, CA), and Jurkat E6-1 was obtained from the American Type
Culture Collection (Rockville, MD). All other cell lines were supplied
by the AIDS Research and Reference Reagent Program, Division of AIDS,
National Institute of Allergy and Infectious Diseases, NIH (Rockville,
MD). The pLTRX-LUC plasmid was kindly given by Dr. O. Schwartz
(Unité d'Oncologie Virale, Institut Pasteur, Paris, France). The
pLTRX-LUC plasmid contains a 722-base pair XhoI
(644)-HindIII (+78) fragment from HIV-1LAI
placed in front of the luciferase reporter gene (27). We have also used in our studies pLTR-LUC (HIV-1 LTR from strain HXB2) and mutated NF-
B pm
BLTR-LUC plasmids, which were kindly provided by Dr. Calame (Columbia University, NY). These molecular constructs contain the luciferase reporter gene under the control of wild-type
(GGGACTTTCC) or NF-
B mutated (CTCACTTTCC)
HIV-1 LTR (28). The wild-type and dominant negative I
B
expressing
vectors pCMV-I
B
and pCMV-I
B
S32A/36A have been described
previously (29). The p
B-TATA-LUC contains the minimal HIV-1
B
region and a TATA box placed upstream of the luciferase reporter gene
(29). These three latter molecular constructs were generous gifts from
Dr. W. C. Greene (The J. Gladstone Institutes, San Francisco). Carrier
DNA (herring sperm DNA) was purchased from Sigma.
pV molecules were prepared as
described previously (30). Briefly, V2O5 was
dissolved in an aqueous KOH solution and then mixed with 30%
H2O2 and the respective ancillary ligand in
addition to the ethanol for optimal precipitation. Characterization of the pV molecules were carried out by infrared 1H NMR and
vanadium-51 (51V) NMR spectroscopy. Stock solutions of pV
molecules (1 mM in phosphate-buffered saline, pH 7.4) were
kept at 20 °C until used. Sodium orthovanadate (Sigma) was freshly
dissolved in 10 mM HEPES, pH 7.4.
To minimize variations in plasmid transfection
efficiencies, transfected cells were pooled 24 h after
transfection and were next separated into various treatment groups.
Cells (5-10 × 106) were first washed once in a TS
buffer (25 mM Tris-HCl, pH 7.4, 5 mM KCl, 0.6 mM Na2HPO4, 0.5 mM
MgCl2, and 0.7 mM CaCl2) and resuspended in 0.5-1 ml of TS containing 15-30 µg of the indicated plasmid(s) and 500 µg/ml of DEAE-dextran (final concentration). The
cells/TS/plasmid/DEAE-dextran mixture was incubated for 25 min at room
temperature. Thereafter, cells were diluted at a concentration of
1 × 106/ml using complete culture medium supplemented
with 100 µM chloroquine (Sigma). After 45 min of
incubation at 37 °C, cells were centrifuged, resuspended in complete
culture medium, and incubated at 37 °C for 24 h. Transiently
and stably transfected cells were seeded at a density of
105 cells per well (100 µl) in 96-well flat-bottom
plates. Cells were left untreated or were treated with the different
peroxovanadium PTP inhibitors, bpV[HOpic], bpV[bipy], and
bpV[pic] at specified concentrations (10 µM in most
experiments), phytohemagglutinin (PHA-P at 3 µg/ml, Sigma), tumor
necrosis factor- (TNF-
at 2 ng/ml, R & D systems, Minneapolis),
or phorbol myristic acid (PMA at 20 ng/ml, Sigma) in a final volume of
200 µl. Next, these cells were incubated at 37 °C for either 8 or
24 h. All experimental points were done in triplicate. Cell
viability was estimated by the
(3-(4,5-dimethylthiazol-2-yl)-5-(3-carboxymethoxyphenyl)-2-(4-sulfophenyl)-2H-tetrazolium, inner salt assay as described previously (31). Finally, luciferase or
CAT activity was evaluated. Luciferase activity was determined following a modified version of a protocol (32). Briefly, following the
incubation period, 100 µl of cell-free supernatant were withdrawn from each well, and 25 µl of cell culture lysis buffer (25 mM Tris phosphate, pH 7.8, 2 mM dithiothreitol,
1% Triton X-100, and 10% glycerol) were added before incubation at
room temperature for 30 min. An aliquot of cell extract (20 µl) was
mixed with 100 µl of luciferase assay buffer (20 mM
Tricine, 1.07 mM
(MgCO3)4·Mg(OH)2·5 H2O, 2.67 mM MgSO4, 0.1 mM EDTA, 270 µM coenzyme A, 470 µM luciferin, 530 µM ATP, and 33.3 mM dithiothreitol), and the sample was introduced into the
counting chamber of a standard liquid scintillation counter equipped
with a single-photon monitor software (Beckman Instruments, Fullerton,
CA). Total photo events were measured over a 30-s time lapse. CAT
assays were performed with the CAT enzyme-linked immunosorbent assay
following manufacturer's recommendations (Boehringer Mannheim).
Intracellular flow cytometry was performed as follows. Cells (5 × 105) were washed once in PBS, pH 7.4, and were resuspended in 100 µl of fresh complete culture medium before treatment with each peroxovanadium inhibitors (10 µM) for 30 min at 37 °C. Next, cells were washed once with PBS at room temperature and centrifuged 5 min at 300 × g. Cell pellets were then fixed with 25 µl of reagent A (Fix & Perm cell permeabilization kit from CALTAG Laboratories, South San Francisco, CA) and incubated 15 min at room temperature. Cells were washed in PBS, resuspended with 25 µl of reagent B to which was added an anti-phosphotyrosine monoclonal antibody (clone 4G10; Upstate Biotechnology Inc., Lake Placid, NY) or a rabbit anti-Tat antibody (antiserum to HIV-1 Tat from Dr. Brian Cullen, AIDS Research and Reference Reagent Program, Division of AIDS, NIAID, NIH), vortexed gently, and incubated for 15 min at room temperature. Cells were subsequently washed with PBS 1% NaN3 and resuspended with 100 µl of PBS containing a fluorescein isothiocyanate-labeled goat anti-mouse or chicken anti-rabbit IgG antibody (1 µg of total) and further incubated for 15 min at room temperature. Finally, cells were centrifuged and resuspended in 1% paraformaldehyde in PBS before being analyzed by flow cytometry (EPICS XL, Coulter Corp., Miami, FL). Cell surface CD4 expression was monitored as described previously (33).
Immunoblot AssayJurkat E6.1 (1 × 106
cells/ml) were treated with the pV compound bpV[pic] (10 µM) at various times and were resuspended in lysis buffer
consisting of 20 mM Tris-HCl, pH 8.0, 150 mM
NaCl, 2 mM EDTA, 1% Nonidet P-40, 10% glycerol, 0.025 mM p-nitrophenyl guanidinobenzoate, 10 µg/ml
aprotinin, 10 µg/ml leupeptin, 1 mM sodium orthovanadate,
and 10 mM sodium fluoride. The homogenate was vortexed and
incubated for 45 min on ice with intermittent mixing before
centrifugation to remove cellular debris. Protein content was
determined by the commercial BCA Protein Assay Reagent (Pierce).
Cleared cellular lysates or immunoprecipitated samples were
electrophoresed on a 12% SDS-polyacrylamide gel electrophoresis, and
the gel was transferred to a polyvinylidene difluoride membrane (Millipore, Mississauga, Ontario). Immunoblot detection was performed with either monoclonal anti-phosphotyrosine P-Tyr-02 (a kind gift from
Dr. Vaclav Horejsi, Institute of Molecular Genetics, Prague, Czech
Republic), 4G10 (Upstate Biotechnology, Inc., Lake Placid, NY), or PY20
(Transduction Laboratories, Lexington, KY) antibodies diluted 1:2000,
or rabbit polyclonal antibodies specific for the C-terminal portion of
IB
(kindly provided by Dr. W. C. Greene, The J. Gladstone
Institute, San Francisco, CA) diluted 1:6000. Detection was carried out
using either peroxidase-conjugated goat anti-mouse at a 1:20,000
dilution or peroxidase-conjugated sheep anti-rabbit antibodies at
1:10,000 dilution. Immunoreactive bands were visualized using the
enhanced chemiluminescence detection system according to the
manufacturer's instructions (Amersham Corp.).
Jurkat E6.1 cells were treated
at various times for up to 2 h with the pV compound bpV[pic].
Cleared cell lysates were first incubated with rabbit anti-IB
antibodies overnight at 4 °C with intermittent mixing. Thereafter,
protein G-Agarose beads (Santa Cruz Biotechnology, Inc., Santa Cruz,
CA) were added and incubation at 4 °C with intermittent mixing was
pursued for 3 h. Immune complexes were washed three times in lysis
buffer without Nonidet P-40 before analysis by immunoblot assay.
1G5 cells were either left
untreated or were incubated for different periods at 37 °C with the
peroxovanadium compounds (10 µM). The incubation of 1G5
cells with the various stimulating agents was terminated by the
addition of ice-cold PBS, and nuclear extracts were prepared according
to the microscale preparation protocol (34). In brief, sedimented cells
were resuspended in 400 µl of cold buffer A (10 mM HEPES,
pH 7.9, 1.5 mM MgCl2, 10 mM KCl,
0.5 mM dithiothreitol, and 0.2 mM
phenylmethylsulfonyl fluoride). After 10 min on ice, the lysate was
vortexed for 10 s, and samples were centrifuged for 10 s at
12,000 × g. The supernatant fraction was discarded,
and the cell pellet was resuspended in 100 µl of cold buffer B (20 mM HEPES-KOH, pH 7.9, 25% glycerol, 420 mM
NaCl, 1.5 mM MgCl2, 0.2 mM EDTA,
0.5 mM dithiothreitol, and 0.2 mM
phenylmethylsulfonyl fluoride) and incubated on ice for 20 min.
Cellular debris was removed by centrifugation at 12,000 × g for 2 min at 4 °C, and the supernatant fraction was
stored at 70 °C until used.
Electrophoretic
mobility shift assay was performed with 10 µg of nuclear extracts.
Protein concentrations were determined by the bicinchoninic assay with
a commercial protein assay reagent (Pierce). Nuclear extracts were
incubated for 30 min at 23 °C in 15 µl of buffer C (100 mM HEPES, pH 7.9, 40% glycerol, 10% Ficoll, 250 mM KCl, 10 mM dithiothreitol, 5 mM
EDTA, 250 mM NaCl, 2 µg of poly(dI-dC), 10 µg of
nuclease-free bovine serum albumin fraction V) containing 0.8 ng of
32P-5-end-labeled double-stranded DNA (dsDNA)
oligonucleotide. Double-stranded DNA (100 ng) was labeled with
[
-32P]ATP and T4 polynucleotide kinase in a kinase
buffer (New England Biolabs, Beverly, MA). This mixture was incubated
for 30 min at 37 °C, and the reaction was stopped with 5 µl of 0.2 M EDTA. The labeled oligonucleotide was extracted with
phenol/chloroform and passed through a G-50 spin column. The dsDNA
oligonucleotide, which was used as a probe or as a competitor,
contained the consensus NF-
B-binding site corresponding to the
sequence 5
-ATGTGAGGGGACTTTCCCAGGC-3
. A dsDNA oligonucleotide
containing a mutated NF-
B-binding site (in bold and underlined) was
also used for competition
(5
-ATGTGAGCTCACTTTCCCAGGC-3
). Oligonucleotides
were purchased from Santa Cruz Biotechnology, Inc. (Santa Cruz, CA).
DNA·NF-
B complexes were resolved from free labeled DNA by
electrophoresis in native 4% (w/v) polyacrylamide gels containing 50 mM Tris-HCl, pH 8.5, 200 mM glycine, and 1 mM EDTA. The gels were subsequently dried and
autoradiographed. Cold competitor assays were carried out by adding a
100-fold molar excess of homologous unlabeled dsDNA NF-
B
oligonucleotide simultaneously with the labeled probe. Supershift
assays were performed by preincubation of nuclear extracts with 1 µl
of either preimmune serum (control) or specific antibodies in the
presence of all the components of the binding reaction described above
for 15 min at room temperature prior to the addition of the probe.
Antibodies used are specific either to the p50 or p65 subunits of the
NF-
B complex and were kindly provided by Dr. Nancy Rice (NCI,
Frederick, MD).
Reverse transcriptase activity was monitored as reported previously (33). Briefly, ACH-2, J1.1, OM10.1, and U1 cells were seeded at a density of 105 cells per well in a final volume of 200 µl in a 96-well plate and treated with PMA (20 ng/ml) or the pV compounds (10 µM) for 24, 48, and 72 h. Enzymatic activity was measured with 50 µl of clarified supernatant fluid to which 10 µl of a solution A (5 mM dithiothreitol, 50 mM KCl, 0.05% Triton X-100) and 40 µl of a solution B (5 mM MgCl2, 0.5 M EGTA, 0.04 mg of poly(rA)-oligo(dT)12-18, 3 mCi of [3H]TTP (40-70 Ci/mmol)) had been added. After incubation for 1 h at 37 °C, samples were precipitated prior to filtration onto glass fiber filters by using a cell harvester system (Tomtec). The filters were dried, and radioactivity was measured in a liquid scintillation counter (1205/1204 BS Beta-plate; Wallac Oy, Turku, Finland). The assays were performed in triplicate.
Statistical AnalysesStatistically significant differences between groups were performed with the analysis of variance module of SAS software (version 6.07, SAS Institute, Cary, NC) using the Fisher least significant difference test. p values <0.05 were considered to be statistically significant (p values are given in the figure legends). All data are presented as mean ± S.D.
Phosphorylation events have been shown to be important in several steps of the HIV-1 life cycle. This is probably associated with the fact that viral transcription is highly dependent on the intracellular environment (35). We were interested in testing whether PTP could directly modulate HIV-1 LTR activity. To achieve such a goal, the standard PTP inhibitor sodium vanadate tyrosine was used along with new pV compounds (30). In our studies, we have used several structurally defined pV compounds, each containing an oxo ligand, two peroxoanions, and an ancillary ligand located in the inner coordination sphere of vanadate.
Peroxovanadium Compounds Strongly Activate HIV-1 LTR Activity in T CellsTo assess whether PTP inhibitors could modulate HIV-1 LTR
activity, pV compounds bpV[HOpic], bpV[bipy], and bpV[pic] were incubated for 24 h with 1G5 cells. This cell line is derived from the Jurkat E6.1 cell line and contains an integrated construct made of
the luciferase reporter gene under the control of the HIV-1LAI LTR. Sodium orthovanadate
(Na3VO4), a commonly used PTP inhibitor, was
similarly tested in this series of experiments. As shown in Fig.
1, a weak increase in HIV-1-LTR-driven luciferase activity (2-fold) was obtained with 25 µM of
Na3VO4 (Fig. 1, panel A). The
mitogenic agent PHA was used as a positive control and was shown, as
expected, to induce HIV-1 LTR activity (4.6-fold). However, the three
pV derivatives used gave a significant increase in HIV-1 LTR-driven
luciferase activity at a concentration of 10 µM (Fig. 1,
panels B-D). When positioned in order of importance, fold
increments induced by the PTP inhibitors were bpV[bipy] (3.2-fold), bpV[HOpic] (5.2-fold), and bpV[pic] (8.1-fold). Although fold increase showed some degree of variation between experiments (data not
shown), the order and the proportional difference between the fold
increase induced by the three PTP inhibitors in the 1G5 cell line were
constant. When lower concentrations of the pV compounds were used (0.1 and 1 µM), no drastic HIV-1 LTR stimulation could be
observed (Fig. 1). MTS assay for cell viability was performed in
parallel and indicated no cytotoxic or cytostatic effect from the pV
derivatives at 10 µM concentration (data not shown).
Thus, these results indicate that pV compounds (bpV[bipy],
bpV[HOpic], and bpV[pic]) drastically increase HIV-1 LTR activity
and to a much greater extent than Na3VO4.
Induction of HIV-1 LTR Activity by pV Derivatives Correlates with an Increase in Total Intracellular Phosphotyrosine Levels
To
determine whether the difference in HIV-1 LTR activity correlated with
the strength of the pV compounds to increase phosphotyrosine content,
intracellular FACS analysis was performed with the anti-phosphotyrosine 4G10 monoclonal antibody. At concentrations lower than 10 µM, no significant changes in the state of intracellular
tyrosine phosphorylation in pV-treated 1G5 cells were apparent (data
not shown). However, upon the addition of 10 µM of
bpV[bipy], bpV[HOpic] and bpV[pic], enhancement in
phosphotyrosine contents was observed in 1G5 cells with mean
fluorescence values (indicative of the number of molecules per single
cell shown on a logarithmic scale) ranging from 2.7 to 8.37 (Fig.
2, panels A-C). To test whether a
correlation existed between the increase in intracellular tyrosine phosphorylation and HIV-1 LTR activity, regression was performed between the luciferase activity measurements and FACS analysis mean
value for each inhibitor. As seen in Fig. 2, panel D, a
correlation between these two parameters could be clearly observed with
a high significant value (r2 = 0.992). Thus,
these results demonstrate that pV-mediated enhancement in the level of
tyrosine-phosphorylated proteins is directly proportional to the
observed up-regulation of HIV-1 LTR activity.
The Increase in HIV-1 LTR Activity by the pV Compounds Is Neither Cell-specific Nor HIV-1 LTR Strain-specific
To determine whether
the induction of HIV-1 LTR activity by the pV compounds could be
observed in other cell lines, transient transfection experiments were
performed in several T cell lines using an HIV-1 LTR-driven reporter
luciferase construct. After addition of any of the three pV derivatives
at a concentration of 10 µM, transfected
Jurkat-tat cells showed a statistically significant
induction of HIV-1 LTR-dependent reporter gene expression (p < 0.01) (Fig. 3, panel
A). In these experiments, PHA was not found to be a good inducer
of HIV-1 LTR activity as it was in the case of 1G5 cells because
Jurkat-tat cells are negative for surface expression of the
T cell receptor as monitored by FACS analysis (data not shown). Thus,
PMA was used as a positive control and found to be a potent inducer of
HIV-1 LTR activity (Fig. 3, panel A). We determined that a
fraction of the pV-mediated stimulation of HIV-1 LTR transcription in
Jurkat-tat cells was attributable to an increase in the
level of endogenous trans-activating viral TAT protein as assessed by
intracellular FACS analysis (data not shown). A statistically
significant induction of HIV-1 LTR-driven luciferase activity
(p < 0.01) was also seen following treatment with the
three different PTP inhibitors of another T cell line (HPB-ALL) (Fig.
3, panel B). In addition, treatment with pV derivatives of
M311, a Molt-4 derivative stably transfected with an HIV-1 LTR-CAT
construct, or Molt-4 clone 8 transiently transfected with pLTRX-LUC
led to a weak but statistically significant induction of HIV-1 LTR
activity (1.6- to 1.7-fold over basal level) (data not shown). The
parental Jurkat E6.1 cell line was shown to respond to a somewhat
weaker extent than its stably transfected derivative 1G5 and
Jurkat-tat cells. For example, a 1.6-, 1.8-, and 2.2-fold increase of HIV-1 LTR activity were detected following treatment of
transiently transfected Jurkat E6.1 cells with bpV[bipy],
bpV[HOpic], and bpV[pic], respectively (Fig. 3, panel
C). It should be noted that modulatory effects on HIV-1
LTR-dependent gene expression were until then all evaluated
following a 24-h incubation period with pV compounds. A greater
enhancement in HIV-1 LTR activity (3-8-fold) was seen when reporter
gene expression was monitored at an earlier time post-treatment with pV
derivatives (8 h) (Fig. 3, panel D). Similar levels of
induction in transiently transfected Jurkat E6.1 cells were observed
with another HIV-1 pLTR-LUC construct (data not shown), which harbors
an HIV-1 LTR promoter from a different genetic background (strain HXB2;
molecular clone pLTR-LUC, see below). This demonstrates that
pV-mediated activation of HIV-1 LTR activity is not restricted to a
certain cell line nor to a particular HIV-1 LTR strain.
Induction of HIV-1 LTR Activity by pV Derivatives Is Partially Dependent on Intact NF-
The NF-B
transcription factor is a very potent factor and is often implicated in
the induction of HIV-1 LTR activity (3). We were interested in testing
the involvement of NF-
B in the observed pV-mediated increase of
HIV-1 LTR activity. Jurkat E6.1 cells were transfected with reporter
gene plasmids possessing the full-length HIV-1 LTR promoter (pLTR-LUC)
and the same HIV-1 LTR with mutated NF-
B binding sites
(pm
BLTR-LUC) and then incubated with either PHA, TNF-
, or the pV
compounds for 8 h. As shown in Fig. 4, panel
A, when pLTR-LUC was transfected in Jurkat E6.1, all activators,
including the pV compounds, gave a marked increase in luciferase
activity. As expected, no activation of HIV-1 LTR-driven gene activity
was seen following treatment with PHA and TNF-
in Jurkat E6.1 cells
transfected with pm
BLTR-LUC (Fig. 4, panel B). The
increase in HIV-1 LTR-driven luciferase activity following treatment
with PTP inhibitors was greatly diminished in cells transfected
with the NF-
B-mutated molecular construct but not totally abolished
(Fig. 4, panel B). Indeed, a 3.8-, 3.4-, and 3.4-fold
increase in HIV-1 LTR-dependent gene activity was seen following treatment with bpV[HOpic], bpV[bipy], and
bpV[pic], respectively. These results suggest that induction
of HIV-1 LTR activity by pV derivatives is partly
NF-
B-dependent.
pV Compounds Induce the Nuclear Translocation of NF-
To
further confirm the involvement of the transcription factor NF-B in
the increase of the HIV-1 LTR activity by pV compounds, DNA mobility
shift assays were carried out. First, we stimulated 1G5 cells for 1, 6, 12, and 24 h with the pV derivatives bpV[bipy] and bpV[pic].
Results demonstrated the presence of a specific band following
treatment of 1G5 cells for 1 h, which was no longer observed after
6 h of treatment (data not shown). Further experiments revealed
that all three PTP inhibitors at 1 h showed the presence of this
signal (Fig. 5, panel A, lanes 3-5).
Competition assays with 100-fold excess of cold wild-type NF-
B
binding site showed complete disappearance of the signal for all
nuclear extracts from pV-treated 1G5 cells (lanes 6-8),
whereas 100-fold excess of cold mutated NF-
B oligonucleotide had no
effect (data not shown). Data from this experiment clearly indicate
that the binding was specific for NF-
B.
Complexes binding to NF-B sequences may constitute a variety of
hetero- or homodimers. These dimers can be composed of many distinct
subunits, which are all members of the mammalian NF-
B/rel homology
domain family of transcription factors. To clearly identify the
pV-induced NF-
B complex, nuclear extracts from 1G5 treated with 10 µM bpV[HOpic] were incubated with rabbit antibodies
against the NF-
B subunits p50 or p65 (Fig. 5, panel B).
Upon addition of the polyclonal antibodies, the binding complex had
completely disappeared (lanes 2 and 3), whereas a
preimmune serum did not greatly affect the NF-
B complex (lane
4). From these results, we conclude that translocation of the
NF-
B p50·p65 heterodimer is induced by the pV compounds and is
responsible for the NF-
B-dependent activation of the
regulatory elements of HIV-1.
NF-B nuclear translocation is mainly mediated by the
degradation of the repressor I
B
, which sequesters the complex in
the cytoplasm (36-38). This degradation is known to be almost
completely dependent on the phosphorylation of the two serine residues
32 and 36, although tyrosine phosphorylation had been reported
previously to act similarly (8, 39). Because tyrosine phosphorylation is the targeted intracellular event in the presently studied HIV-1 LTR
induction pathway, we wanted to verify if serine phosphorylation of
I
B
was involved in the pV-induced effect. To achieve this, we
used a dominant negative version of I
B
mutated to alanine on both
serine 32 and 36 residues. The encoded protein is hence unable to be
serine-phosphorylated but retains its ability to bind to the NF-
B
complex. In addition, a plasmid containing only the NF-
B HIV-1 LTR
binding sites upstream of a TATA box driving the expression of the
luciferase gene (p
B-TATA-LUC) was used in this set of experiments to
exclusively study pV-induced activation of NF-
B. Jurkat E6.1 cells
were transfected with a 2 to 1 ratio of the p
B-TATA-LUC and
pCMV-I
B
plasmids, the latter being either the wild-type
(pCMV-I
B
wild-type) or the dominant form of the repressor
(pCMV-I
B
serine 32/36). It should be noted that the p
B-TATA-LUC plasmid resulted in a greater level of induction of
luciferase activity (40-50-fold increase) upon addition of the pV
compounds as compared with those found with the pLTRX-LUC construct.
This could be attributed to either the deletion of the HIV-1 LTR
negative regulatory elements or to a better positioning of the NF-
B
binding sites for its trans-activating potential. Although transfection
of the wild-type I
B
had little effect on NF-
B-induced reporter
gene activity, when the mutated version of the I
B
repressor was
used, incubation of the cells with all three pV compounds showed an
almost complete inhibition of NF-
B-mediated luciferase activity
(Fig. 6). TNF-
, a positive control, was also similarly affected when I
B
S32A/36A was added. These results indicate that pV-mediated activation and nuclear translocation of
NF-
B is resulting from the serine phosphorylation of I
B
on
residues 32 and 36.
No Changes in Tyrosine Phosphorylation of I
Since it has been previously reported that treatment
of Jurkat E6.1 cells with pervanadate induced NF-B translocation via tyrosine phosphorylation of I
B
with no concomitant degradation (40), we next wanted to evaluate whether a similar phenomenon was
occurring with pV compounds. Results shown in Fig. 7 are
indicative of no change in phosphotyrosine level of I
B
throughout
the time course using the anti-phosphotyrosine 4G10 or PY20 antibodies (panel A and data not shown). In comparison, at 30 and 60 min post-treatment with bpV[pic], a decrease in band intensity of the
I
B
protein was observable, which can likely be attributed to
degradation of I
B
(Fig. 7, panel B). Finally, total
cell extracts from Jurkat E6.1 cells treated with bpV[pic] for
similar amounts of time as shown in panels A and
B were analyzed with an anti-phosphotyrosine antibody and
demonstrated an increase in overall phosphotyrosine levels starting at
the 5 min time point (Fig. 7, panel C). Our results thus
suggest that pV molecules are triggering NF-
B translocation by a
degradation process of the I
B
repressor which is dependent on
intracellular tyrosine phosphorylation levels but with no concomitant
increase in tyrosine phosphorylation of I
B
itself.
pV Compounds Stimulate Viral Production in T Cell Lines Latently Infected with HIV-1
Since pV compounds are able to increase HIV-1
LTR activity, we next wanted to test their potency in the context of an
entire viral genome. This is based on the observation that HIV-1
replication is a complex process regulated by proteins of viral and
cellular origins that have been shown to work in cis and trans. As
presented in Fig. 8, treatment of J1.1, a cell line
latently infected with HIV-1, with all three inhibitors induced HIV-1
replication over time as determined by an increase in reverse
transcriptase activity. PMA was used as a positive control and also
induced HIV-1 replication (data not shown). Another T cell line
containing latently integrated HIV-1 DNA, ACH-2, gave similar induction
of reverse transcriptase activity when two of the three inhibitors
(bpV[HOpic] and bpV[pic]) were tested (data not shown). In
conclusion, pV compounds can also stimulate HIV-1 replication in
latently infected T cells.
pV Derivatives Increase HIV-1 LTR Activity and Virion Production in Monocytoid Cells
Monocyte-derived macrophages are considered to
be the most frequently identified hosts of HIV-1 in tissues of infected
individuals (41). We therefore tested the effect of pV molecules on
U38, a U937-derived cell line stably transfected with an HIV-1 LTR-CAT construct. After addition of the three inhibitors to this cell line,
CAT activity was drastically increased (up to 18-fold) as compared with
untreated cells (Fig. 9, panel A). A more
modest but significant increase in CAT activity was also seen with PMA (data not shown). To further test the induction of HIV-1 LTR activity in cells of monocytoid origin, U1 cell line, which contains two integrated full-length proviral DNA copies, was incubated with the PTP
inhibitors. Again, a time-dependent increase in reverse transcriptase activity was observed for all pV compounds tested (Fig.
9, panel B). Similar results were obtained when the latently infected monocytoid OM10.1 cell line was tested with the inhibitors bpV[HOpic] and bpV[pic] (data not shown). The data presented above hence indicate that pV molecules can increase HIV-1 LTR activity and
virus replication in a wide range of cell types including monocytoid
cell lines.
HIV-1 replication is controlled by several different cytokines and
mitogens. Although several reports have focused on the agents
stimulating or repressing LTR activity, little information regarding
intracellular pathways that act on HIV-1 LTR regulation is available.
Most information is centered around the NF-B pathway, which is
dependent on the degradation of the I
B
repressor induced by the
phosphorylation of 2 serine residues (8). Although central to many
different pathways, tyrosine phosphorylation events per se
have not been directly investigated in HIV-1 regulation. In this study,
we have thus analyzed the role of tyrosine phosphorylation on HIV-1 LTR
regulation. pV compounds, previously characterized as potent PTP
inhibitors, were used in this study to increase intracellular
phosphotyrosine levels.
Our results first demonstrated that pV compounds were effective in
inducing HIV-1 LTR activity in 1G5 T lymphoid cells and to a much
higher extent than sodium orthovanadate, which is consistent with the
reported observation showing that pV molecules are more potent PTP
inhibitors than sodium orthovanadate (30). Our results are also in
agreement with a previous study that reported an activation of HIV-1
LTR activity in 1G5 cells following treatment of vanadate and hydrogen
peroxide (42), a combination known to form pervanadate molecules when
intermixed. However, our findings are different from a previous study
that has shown that pervanadate and phenylarsine oxide, two PTP
inhibitors, can abrogate TNF--mediated NF-
B translocation (43).
This discrepancy might be explained by the targeting of different PTP
in the two different pathways analyzed. Indeed, the different pV
compounds used in the present work vary in terms of their ancillary
ligand which is attached to the vanadate core. This may give a
different specificity to these molecules in terms of the PTP affected.
This notion is exemplified by previous results showing that pervanadate
and vanadate demonstrate distinct potency and substrate specificity
(44). However, it should be noted that our data showed a strong
correlation between the level of intracellular tyrosine phosphorylation
and the induction of HIV-1 LTR activity. This most likely suggests that
a similar or a set of similar PTP are targeted by the pV derivatives
but to a different extent.
We have determined that pV compounds induced HIV-1 LTR activity in several lymphoblastoid T and monocytoid cells. The pV-mediated increases in HIV-1 LTR transcription were greater in stably transfected cells than in transient transfection experiments. Although cell type might explain some variation, it is likely that the number of cells carrying the molecular HIV-1 construct, as well as differences in the number of copies introduced per cell, might be responsible for the observed variations in the response to pV derivatives. The effects of pV molecules in transiently transfected Jurkat E6.1 cells were measured after 8 h of treatment, which permitted higher levels of enhancement as compared with the 24-h treatment period. However, a more significant increase in HIV-1 LTR transcription was seen following treatment for 24 h of HPB-ALL with the pV compounds when compared with Jurkat E6.1 (Fig. 3, panel B), indicating variation in responsiveness to pV among the different T cell lines.
Based on DNA mobility shift assays and transient transfection
experiments with HIV-1 LTR constructs bearing mutated NF-B binding
sites, our results suggest that part of the pV-induced effect on HIV-1
LTR activity is mediated via the activation and nuclear translocation
of NF-
B. This is similar to most HIV-1 LTR activators which
stimulate LTR activity via this transcription factor. Although several
forms of NF-
B dimers exist, our supershift assays are indicative of
p50·p65 heterodimer being the bound component of the activated
complex. This dimer is in fact the most important activator of all
formed heterodimers related to the rel family (45). These
results are in agreement with the fact that several pathways inducing
NF-
B translocation have been known to act through an increase in
tyrosine phosphorylation. For example, it has been shown that
overexpression of p59fyn in T cell lines could stimulate HIV-1
LTR activity by NF-
B-like DNA binding proteins (9). We have further
analyzed the pathway by which NF-
B translocation is induced by pV
inhibitors. It is known that the I
B
repressor sequesters the
NF-
B complex in the cytoplasm by an interaction with the nuclear
localization signal region. Upon proper activation, serine
phosphorylation of residues 32 and 36 occurs on the I
B
repressor
by a kinase whose activity is dependent on ubiquitination (46). Thus,
phosphorylation targets I
B
repressor to the proteasome via a set
of entrained ubiquitination events (8, 47). On the other hand, it has recently been reported that tyrosine phosphorylation of I
B
in hypoxia-treated Jurkat cells could lead to activation of NF-
B (39).
Our results indicate that pV compounds induce
NF-
B-dependent activation of HIV-1 LTR activity via
serine phosphorylation events since a dominant negative version mutated
at serine 32 and 36 abrogated the pV-mediated effect on HIV-1
transcription. These results are again similar to NF-
B translocation
induced by several T cell activation signals (i.e. TNF-
,
PMA, lipopolysaccharide) which all act through serine phosphorylation
of I
B
resulting in its concomitant degradation (48).
Our Western analyses confirmed the importance played by serine
phosphorylation since degradation was apparent in Jurkat E6.1 cells treated with the pV compound bpV[pic] with no indication of
stimulated tyrosine phosphorylation of IB
. Our observations are
thus in sharp contrast with a recently described work which demonstrated that pervanadate can lead to translocation of NF-
B via
tyrosine phosphorylation of I
B
with no concomitant proteolytic degradation of I
B
(40). This discrepancy might be related to the
possibility that pV compounds might, as postulated above, target
different PTP. In agreement with this hypothesis, we found that H7, a
protein kinase C inhibitor, could drastically reduce pV-induced
activation of HIV-1 LTR activity (data not shown), while such a
treatment had no effect on NF-
B translocation mediated by
pervanadate (40).
The participation of phosphatase(s) in the activation of NF-B was
previously provided by a study demonstrating that the Ser/Thr PTP
inhibitor okadaic acid is also a potent activator of NF-
B (49). Our
results represent the first demonstration that specific inhibitors of
PTP lead to NF-
B activation via a site-specific serine
phosphorylation of I
B
. Thus, our studies imply that the serine
phosphorylation state of I
B
is controlled by the dynamic balance
between PTK and PTP activities. It is also indicative of the presence
of a common kinase(s) activated by several different signaling pathways
that is ultimately participating to the regulation and function of
NF-
B. PTP could be a very important regulator of NF-
B
translocation keeping the activating pathway at a resting state. This
is reminiscent of the implication of certain PTP, such as CD45 and
PTP-1C, in restraining other activated pathways by terminating tyrosine
phosphorylation signaling (50, 51). Thus, the present data highlight a
potential role for PTP as negative regulatory elements in NF-
B
activation.
pV inhibitors were also found to stimulate HIV-1 production in four
different cell lines latently infected with HIV-1 which is indicative
of a common signaling pathway for the induction of NF-B
translocation in several different cell lines that ultimately leads to
HIV-1 production regardless of the cause of viral latency (52, 53). It
has been shown that the frequency of cells carrying transcriptionally
active HIV-1 proviral DNA in infected individuals is at least 1 to 2 orders of magnitude lower than the total amount of infected cells (54),
indicating that the majority of infected cells exist in a state of
transcriptional latency. Thus, we can postulate that PTP are implicated
to some extent in the regulation of HIV-1 expression by keeping the
virus under some form of latency.
We have demonstrated that the pV compounds induce the HIV-1 LTR
activity through two distinct pathways, one which is
NF-B-dependent and one which is NF-
B-independent. The
NF-
B-independent pathway remains to be elucidated, but preliminary
results from our laboratory suggest that this mode of activation is not
exclusive to the HIV-1 LTR (data not shown). A similar bimodal
NF-
B-dependent and -independent type of HIV-1 LTR
induction has also been reported with okadaic acid (49). It is
interesting to note that this inhibitor acts on serine/threonine
phosphatases and might thus trigger a cascade similar, at some point,
to that initiated by pV compounds.
The most important question resulting from our findings is the
identification of the targeted PTP and participating intracellular second messengers. Although several different PTPs have been described to date, the integral membrane PTP CD45 seems a good candidate for many
reasons. This phosphatase is the most abundant leukocyte cell surface
glycoprotein and is present in all hematopoietic cell types (15). It is
known to be important in T cell activation via the activation of the
protein tyrosine kinases p56lck and p59fyn (15, 55).
Treatment with pervanadate was demonstrated to inhibit most of the CD45
PTP activity (16) and to affect CD45 interaction with other proteins
including p56lck (56). More importantly, in CD45 negative cell
lines, HIV-1 LTR activity is greater than in the wild-type counterpart
due to constitutive translocation of NF-B (57). Studies are
currently in progress to investigate the role played by CD45 in
pV-mediated activation of HIV-1 LTR transcription.
Our results suggest for the first time that a still undefined protein
tyrosine phosphatase(s) is a key element in the signal transduction
pathway leading to IB
serine phosphorylation and NF-
B nuclear
translocation. In light of our results, we hypothesize two possible
modes of action by which pV compounds could lead to I
B
serine
phosphorylation. First, it is conceivable that an I
B
-specific
serine/threonine kinase, part of the presumed multisubunit kinase
complex (46), might be activated by tyrosine phosphorylation which, in
turn, would be controlled by a pV-sensitive PTP. Second, pV compounds
might be inactivating a dual-specificity phosphatase resulting in an
increase of the half-life of I
B
serine phosphorylation. This
eventuality is based on the postulate that I
B
phosphorylation
level is controlled by an equilibrium between PTK and PTP
activities.
In conclusion, these pV derivatives may allow us to further
characterize the signaling events that contribute to activation of
NF-B following serine phosphorylation of I
B
. Moreover, pV compounds may also permit us to acquire a better understanding of the
intracellular factor(s) that keep the integrated HIV-1 genome under a
certain latent state. Identification of PTP involved in serine
phosphorylation of I
B
by pV molecules may provide new
opportunities to develop therapeutic strategies aimed at inhibiting dysregulated NF-
B functions seen in certain diseases.
We thank N. Rice for polyclonal anti-p65 and
-p50 antibodies, A. Weiss for HPB-ALL cells, O. Schwartz for pLTRX-LUC,
K. L. Calame for pLTR-LUC and pmLTR-LUC, V. Horejsi for
anti-P-Tyr-02 antibodies, and W. C. Greene for rabbit anti-I
B
antibodies, p
B-TATA-LUC, pCMV-I
B
, and pCMV-I
B
S32A/36A.
Several items were obtained from the NIH AIDS Research and Reference
Reagent Program: 1G5, ACH-2, J1.1, Jurkat-tat, M311, Molt-4
clone 8, OM10.1, U1, and U38. We are grateful to Dr. Maurice Dufour and
Sylvie Perron for technical assistance in flow cytometry studies and reverse transcriptase assays, respectively.