HIV-induced giant cell (syncytium) formation, and direct virus inactivation. ..... Baba M, Schols D, Pauwels R, Nakaskhima H, and De Clercq E: Sulfated ...
AIDS RESEARCH AND HUMAN RETROVIRUSES Volume 13, Number 2, 1997 Mary Ann Lieber!, Inc.
Cell
Type-Dependent Effect of Sodium Valproate on Human Immunodeficiency Virus Type 1 Replication in Vitro
MYRIAM
WITVROUW, JEAN-CLAUDE SCHMIT, BARBARA VAN REMOORTEL, DIRK DAELEMANS, JOSÉ A. ESTÉ, A.-M. VANDAMME, JAN DESMYTER, and ERIK DE CLERCQ
ABSTRACT Sodium valproate (VPA), a simple branched-chain fatty acid that has anticonvulsant activity and is used in the treatment of many forms of epilepsy, has been reported to stimulate human immunodeficiency virus (HIV) type 1 replication in acutely infected CEM and chronically infected Ul cells (Chemico-Biological Interactions 1994;91:111-121). When attempting to reproduce and extend these findings, we confirmed that VPA is able to stimulate HIV-l(IIIB) replication in acutely infected CEM and C8166 T lymphocytic cell lines and chronically infected ACH-2 and U937/IHB/LAI cells in a concentration-dependent manner. The stimulatory effect of VPA on HIV replication in CEM cells was not increased by pretreatment of the cells with VPA for 24 hr before infection. However, we could not detect any stimulatory effect of VPA on HIV-l(IIIB) replication in acutely infected peripheral blood mononuclear cells (PBMCs), MT-4, MT-2, HUT-78, and MOLT-4 (clone 8) cells and in chronically infected HUT-78/IIIB/LAI cells. The stimulatory effect by VPA under certain conditions (see above) may be ascribed to an enhanced HIV transcription, as VPA was found to enhance the HIV long terminal repeat (LTR)-directed expression of /3-galactosidase in transiently transfected HLtat, P4, and COS7 cells. VPA did not enhance /3-galactoside expression mediated by the cytomegalovirus (CMV) promoter. VPA did not affect HIV-induced syncytium formation. Nor had VPA any direct inactivating effect on HIV.
INTRODUCTION
Slargely
(VPA), a simple, branched-chain fatty acid with a broad-spectrum anticonvulsant activity, is used in the treatment of primary and secondary forms of epilepsy.1 Seizures are common in HIV-1-infected patients, either due to opportunistic infections (i.e., toxoplasmosis) or cancers (i.e., lymphoma) or to HIV-related lesions of the brain (HIV encephalopathy). These conditions frequently require anticonvulsant therapy and VPA may be considered as a possible therapeutic choice. However, Simon et air2 have reported that VPA stimulates the replication of human immunodeficiency virus type 1 (HIV-1) in acutely infected lymphocytic CEM cells and chronically infected monocytic Ul cells. In this study, we attempted to reproduce and extend these findings by investigating the effect of VPA on HIV replication in acutely infected CEM, C8166, PBMC, MT-4, MT-2, HUT-78, and MOLT-4 (clone 8) cells, HIV replication in chronically infected ACH-2, U937, and HUT-78 cells, HIV long terminal repeat (LTR)- and cytomegalovirus (CMV) promoter-directed gene expression, odium valproate
,
.
Rega Institute for Medical Research,
.
.
HIV-induced giant cell (syncytium) formation, and direct virus inactivation. Treatment of epilepsy or seizures in HIV-positive individuals by VPA would not be recommended, if VPA indeed stimulates HIV replication,
MATERIALS AND METHODS
Compounds ,.
,
.,
,,
valproate (VPA) was provided by Recherche, Montpellier, France). Sodium
,,,„.
_
_
„
^
P. Gros
(Sanofi-
..
MT-4 cells,3 MT-2 cells,4 CEM
cells,5 MOLT-4 cells (clone
8),6 C8166 cells,7-8 ACH-2 cells,9-10 U937 (ATCC CRL 1593)
cells, and HUT-78 cells11 were grown and maintained in RPMI 1640 medium; and HLtat cells,12"14 HeLa-CD4-LTR-/3gal (P4) cells,15 and COS7 cells16 were maintained in Dulbecco modified Eagle's medium (4500 mg/liter glucose). Both media were
Katholieke Universiteit Leuven, B-3000 Leuven,
187
Belgium.
188
WITVROUW ET AL.
supplemented with 10% heat-inactivated fetal calf serum, 2 mM L-glutamine, 0.1 % sodium bicarbonate, and 20 p,g of gentamicin per milliliter. U937/IIIB/LAI and HUT-78/IIIB/LAI cells were obtained after infection of U937 and HUT-78 cells, respectively, with HIV-1 strain IIIB/LAI.17 HLtat cells contain stably integrated copies of the HIV-1 LTR promoter linked to the synthetic first exon of the tat gene. This cell line was generated by cotransfection of HeLa cells with pSV2neo, which provides neomycin resistance, and pL3tat, which contains the HIV-1 LTR promoter, the synthetic first tat exon, and the simian virus 40 polyadenylation signal. ACH-2 and HLtat cells were obtained through the AIDS Research and Reference Reagent Program (Division of AIDS, National Institute of Allergy and Infectious Disease, Berthesda, MD) and were contributed by T. Folks and B. K. Felber and G. N. Pavlakis, respectively. C8166 cells were obtained from the Medical Research Council (MRC) Reagent Program (London, United Kingdom) and were contributed by G. Farrar. HeLa-CD4-LTR-/3gal cells were kindly provided by P. Chameau (Unité d'Oncologie Virale, Institut Pasteur, Paris, France). COS7 cells were received from F. Wuytack (Laboratory of Physiology, Gasthuisberg, Leuven, Belgium). Peripheral blood mononuclear cells (PBMCs) were isolated from HIV-seronegative donor buffy coats using a Ficoll gradient (Lymphoprep; Nycomed, Norway), stimulated for 3 days in phytohemagglutinin (PHA-P, 2 /¿g/ml; Sigma, Bornem, Belgiumj-containing medium, washed, and resuspended in RPMI 1640 medium supplemented with 2 mM L-glutamine gentamicin (50 /xg/ml), 15% heat-inactivated fetal calf serum, recombinant human interleukin 2 (IL-2) (10 U/ml) (Boehringer GmbH, Mannheim, Germany), and Polybrene (2 /xg/ml).
Plasmid The plasmid pHIVlacZ,1819 containing the lacZ gene driven by the HIV-1 LTR promoter, was obtained from J. J. Maio through the AIDS Research and Reference Reagent Program. pCTAT, containing the tat gene, was obtained from the National Institutes of Health. pCMV/3 was first described by MacGregor and
Caskey.20
Viruses HIV-1 strain IIIB/LAI17 was kindly provided by R. C. Gallo. A stock was obtained from the culture supernatant of an HIV1-infected cell line (MT-4/IIIB/LAI). In a similar way a viral stock was grown on PBMCs, titrated for infectivity, and used at an infectious dose of 1000 (50% cell culture infective dose) per milliliter (multiplicity of infection [MOI] 0.001) for PB MC
experiments. Determination of the HIV replication
effect of sodium valproate
In MT-4 cells the in vitro effect of VPA on HIV-1 replication was determined by measuring the virus-induced cytopathicity. Briefly, MT-4 cells were suspended at 3 X 105 cells/ml, and 100 pA of the cell suspension was placed into each well of flat-bottom 96-well microtiter trays containing 100 pA of various concentrations of the test compound. Cells were infected with HIV-1 at 100-fold the CCID50 per milliliter. After 3, 4, and 5 days of incubation at 37°C, the number of viable cells was determined by the trypan blue exclusion method. The supernatant of C8166, MT-4, and HUT-78/IIIB/LAI cells incubated during 3, 4, and 6 days, respectively, in the presence of various concentrations of VPA was used to determine virus yield by titration on MT-4 cells using the Reed and Muench end-point dilution method.21 The effects of VPA on the replication of HIV-1 strain IIIB/LAI in acutely infected CEM, C8166, PB MC, MT-4, MT2, HUT-78, MOLT-4, and persistently HIV-1 infected U937/IIIB/LAI, HUT-78/IIIB/LAI, and ACH-2 cells were monitored by the detection of HIV-1 p24 core antigen and HIV-1 reverse transcriptase (RT) activity in the cell culture supernatant. Cells were plated in the presence of VPA concentrations ranging from 0.03 to 2.7 mM, and infected with HIV-1 strain IIIB/LAI at 100 CCID50/ml (MOI 0.003), except for PBMCs, which were infected at 1000 CCID50/ml. At 2 to 6 days after plating the cells (Table 1), p24 antigen was detected by an enzyme-linked immunosorbent assay (ELISA) (Du Pont,
Table 1. Different Cell Types Tested for Stimulation of HIV-1 Viral Replication by Sodium Valproate"
Cell line MT-4 MT-2 C8166 CEM MOLT-4 HUT-78 HUT-78/IIIB/LAI ACH-2 U937/IIIB/LAI PBMC
CCso (mM) 0.9 0.2 0.5 2.4 1.0 0.3 0.3 0.6 0.6 0.5
on
Stimulation of HIV-1 replication
No.
of cells/ml
Mold)
(X 105)
1 1 3.4 6 1 1 1 16 3
1.5 1.0 1.0 2.5 1.0 1.0 1.0 1.0 2.0 10.0
1
Day of sampling 3 4 3 6 5 or
5
2 3 4 5
"Assay conditions (number of cells per milliliter at the day of infection, day of sampling for p24 antigen and RT activity assays) and 50% cytotoxic concentration of VPA, estimated after 5 days of incubation on mockinfected cells by the trypan blue exclusion method. Cell lines HUT-78/IIIB/LAI, U937/IIIB/LAI, and ACH-2, chronically infected with HIV-1, were carefully washed in RPMI 1640 medium on day 0 to eliminate p24 antigen or RT activity and were then followed in the same way as the acutely infected cell lines.
189
EFFECT OF VPA ON HIV REPLICATION
Dreieich, Germany). Reverse transcriptase activity was measured using the Quant-T-RT assay system (Amersham Life Sciences, Gent, Belgium) based on the scintillation proximity assay (SPA) principle. Briefly, 10 pA of the culture supernatant was incubated for 4 hr at 37°C in 90 pA of reaction mix containing 1 pA of [3H]dTTP ([/ne%M',2'-3H]thymidine 5'triphosphate) in 25 mM Tris-HCl (pH 8), 5 mM MgCl2, 5 mM dithiothreitol (DDT), 19 pA of 5.25 X assay buffer (260 mM Tris-HCl [pH 8], 420 mM KC1, 52.5 mM DTT, 52.5 mM MgCl2, 13 mM EGTA, 0.26% [w/w] Nonidet P-40), 10 pA of primer/template (DNA/RNA bound to streptavidin-coated SPA beads in 10 mM Tris-HCl [pH 8], 1 mM MgCl2, and 0.05% [w/w] sodium azide), and 60 pA of distilled water. After adding 200 pA of stop solution (0.12 M EDTA, pH 8), the samples were diluted in 500 pA of TBE buffer containing 1 % sodium dodecylsulfate (SDS) to inactivate the virus particles completely. The samples were counted twice in a Packard 1900TR scintillation analyzer. Each run contained the appropriate negative and positive controls. Samples were considered positive if the counts per minute (cpm) were at least twice the value of the negative control. All results are expressed in counts per minute
per milliliter of supernatant.
Cytotoxicity
assays
Cytotoxicity of VPA was determined in MT-4, MT-2, C8166, CEM, MOLT-4, HUT-78, HUT-78/IIIB/LAI, ACH-2, U937/IIIB/LAI, and peripheral blood mononuclear cells by
measuring the viabilities of mock-infected cells after 5 days of incubation, using the trypan blue exclusion method.
ß-Galactosidase
trans-activation assay
HLtat and P4 cells
were
DNA, and COS7 cells
transfected with pHIVlacZ plasmid cotransfected with pHIVlacZ and
were
pCTAT plasmid DNA by electroporation (EASYJECT one electroporator; Eurogentec, Gent, Belgium). Immediately after transfection, cells
were
incubated for 24 hr in the presence of
varying concentrations of VPA. Cell extracts were prepared as described previously22-23 and the ß-galactosidase (/3-Gal) activity in the cell extracts was quantitated by a colorimetric assay following two methods as described by Sambrook et al.24 and Witvrouw al.22 Protein concentrations in the cell extracts were determined by using an automated Bradford method (Bioet
Rad, Hercules, CA). Giant cell
(syncytium) formation
MT-4 cells
method.21
using
the Reed and Muench
end-point
dilution
RESULTS C8166 and CEM cells were infected with HIV-1 strain IIIB/LAI in the presence of VPA concentrations ranging from 0.1 to 2.7 mM. The HIV-1 p24 core antigen content in the cell culture medium at 3 and 6 days after infection, respectively, increased in a concentration-dependent fashion with increasing VPA concentrations (Fig. 1). At 2.7 mM, VPA gave a 3.4- and 2.6-fold increase in p24 antigen and in the production of infectious virus particles, respectively, in C8166 cells. The results for HIV-1 RT activity (data not shown) completely correlated with the p24 antigen and virus yield results. CEM cells treated with 0.9 mM VPA, a nontoxic concentration for the CEM cells, showed a p24 antigen level which was sixfold higher than in untreated cells (data not shown). We studied the effect of preincubation of CEM cells with VPA. CEM cells were pretreated for 24 hr with various concentrations of VPA before infection with HIV-1 strain IIIB/LAI. When monitored 6 days after infection, the stimulatory effect of VPA on HIV replication in CEM cells was not higher when the cells had been pretreated with VPA, as compared to cells that had not been (pre)treated
(data
not
shown).
We also studied the effect of VPA on HIV expression in chronically HIV-1-infected ACH-2 and monocyte-derived U937 cells. VPA induced HIV expression in persistently infected ACH-2 and U937 cells. The stimulatory effect of VPA in ACH-2 cells became evident from a concentration of 0.1 mM and was maximal at 2.7 mM. Three days after treatment with 2.7 mM VPA, virion production of ACH-2 cells was markedly enhanced as indicated by p24 antigen levels (Fig. 2) that were 16-fold higher than in the culture supernatant of untreated cells. The same conclusion was reached when measuring HIV-1 RT activity (data not shown). U937/IIIB/LAI cells treated with 0.3
500000 -, _
300
450000 250
S 400000
ill
3 350000
II
c
J« assay
MOLT-4 (clone 8) cells (1.8 X 106 cells/ml) were cultured with persistently HIV-1-infected HUT-78 cells (2 X 105 cells/ml) in microtiter tray wells containing various concentrations of VPA. A few hours after cocultivation of MOLT-4 cells with chronically infected HUT-78/IIIB/LAI cells giant cells (syncytia) started to develop. After a 6- and 24-hr cocultivation period, the number of giant cells was recorded microscopically, as described previously.25
Virus inactivation assay To test if the effect of VPA on HIV is virucidal, HIV-1 strain IIIB/LAI virus particles were incubated for 20 min with VPA concentrations ranging from 0.08 to 10 mM at room temperature. Virus titers (4.6 log!o CCID50) were determined in
300000-
|
250000
*
200000
PI
150000 100000
50000
-
-
0
0
0.1 0.3 0.9 Concentration of VPA (mM)
FIG. 1. Stimulatory effect of VPA on HIV-1 replication in C8166 cells, as measured by p24 antigen and virus yield production in the culture supernatant after 3 days. The effect of VPA is concentration dependent, within the concentration range of 0.1 to 2.7 mM. Values of p24 antigen are medians for tests done in triplicate. The stimulation of viral replication was confirmed in an RT activity assay. Viability of VPA-treated cells was assessed in parallel. All results were confirmed in at least two
independent tests.
190
WITVROUW ET AL.
being the least sensitive to the cytopathic effect of VPA CC50 values of 0.2 and 2.4 mM, respectively (Table 1).
cells
16000
at 14000 12000
DISCUSSION
-
10000"a
¡
8000
s
3a.
6000
LJ
4000 2000 0
m 0
0.1
0.3
0.9
Concentration of VPA (mM)
FIG. 2.
Stimulatory effect of VPA on HIV-1 expression in chronically infected ACH-2 cells, as measured by p24 antigen production in the culture supernatant after 4 days. The effect of VPA is concentration dependent within the concentration range of 0.1 to 2.7 mM. Values of p24 antigen are medians of tests done in triplicate. Viability of VPA-treated cells was assessed in parallel. All results were confirmed in at least two independent tests.
mM VPA showed a p24 antigen level and RT activity that were threefold higher than in untreated cells (data not shown). We also studied the effect of VPA on HIV replication in acutely infected PBMCs (Fig. 3), and in MT-4, MT-2, HUT78, MOLT-4, and persistently infected HUT-78/IIIB/LAI cells. In none of these cell systems was a stimulatory effect of VPA in the concentration range of 0.1 to 2.7 mM observed. In PBMCs, no stimulatory effect of VPA in the concentration range of 0.1 to 0.9 mM was observed. In further experiments, we showed that VPA enhances the Tat-dependent (Fig. 4A) and Tat-independent (Fig. 4B) HIV long terminal repeat (LTR)-direced expression of /3-galactosidase. In HLtat and P4 cells transiently transfected by a construct containing the lacZ gene under LTR control, the production of /3-galactosidase, detected by a fluorometric assay, was increased by 3.1- and 3.6-fold, respectively, at a concentration of 4 mM VPA (Fig. 4A and B). Similar results were obtained in COS7 cells (data not shown). On the contrary, VPA did not show any effect on the CMV promoter-directed expression of ß-galactosidase in HLtat cells (Fig. 4A). VPA was toxic for HLtat, COS7, and P4 cells at a 50% cytotoxic concentration (CC50) of approximately 10 mM (Fig. 4A and B). VPA did not show any microscopically detectable effect on HIV-1-induced syncytium (giant cell) formation after 6 or 24 hr of cocultivation (data not shown). When evaluated in a direct virus inactivation assay, the compound did not influence the viral titer (4.6 log 10 CCID50) under the conditions used (data not
shown).
VPA was toxic for all cell lines tested. Cells were ordered by their sensitivity to the toxic effect of VPA as follows: MTHUT-78/IIIB/LAI > PBMC C8166 > 2 > HUT-78 ACH2 = U937/IIIB/LAI > MT-4 > MOLT-4 > CEM (Figs. 1-3 and Table 1), MT-2 cells being the most sensitive and CEM =
=
In this study we have shown that valproic acid (VPA), an anticonvulsant drug, is able to stimulate HIV-1 replication in acutely infected CEM and C8166 T lymphocytic cell lines and in chronically infected ACH-2 and U937/IIIB/LAI cells in a concentration-dependent manner. The stimulatory effect was not higher whether the cells had been pretreated with VPA or not. Activation of HIV-1 replication by VPA in these cells was achieved at concentrations that can be reached in the plasma of VPA-treated patients (0.3-0.7 mM). We have shown that the stimulation of HIV replication may be mediated by activation of the HIV LTR promoter sequences, using HLtat and P4 cells transiently transfected with a construct containing a reporter gene under control of the HIV LTR promoter. On the other hand, VPA did not show any effect on CMV promoter-directed gene expression. From these data we conclude that the effect of VPA on HIV replication is specific and mediated by the HIV LTR promoter. No stimulatory effect of VPA was observed on the replication of HIV in acutely infected PBMC, MT-4, MT-2, HUT-78, or MOLT-4 cells, and chronically infected HUT-78/IIIB/LAI cells. Moreover, VPA did not show any microscopically detectable effect on syncytium formation between uninfected CD4+ MOLT-4 cells and chronically infected HUT78/IIIB/LAI cells. Nor did it prove directly virucidal in an HIV inactivation assay. It is tempting to speculate on the mechanism(s) by which VPA activates the HIV LTR promoter and stimulates HIV repli-
0
0.1
0.3
Concentration of VPA (mM)
FIG. 3. Absence of stimulation of VPA on HIV-1 replication in PBMCs, as measured by RT activity in culture supernatant after 5 days. Viablity of VPA-treated cells was assessed in parallel. The decrease of RT activity with increasing VPA concentrations can be explained by drug-induced cell toxicity. Values of RT activity are medians for tests done in triplicate. Absence of stimulation has been confirmed in three independent experiments in PBMCs derived from three different donors. Viability of VPA-treated cells was assessed in parallel.
191
EFFECT OF VPA ON HIV REPLICATION
0
0.16
0.8
0.16
4
Concentration of VPA (mM)
0.8
Ï *
I
5
