Cocaine Potentiates the Switch between Latency
and Replication of Epstein–Barr Virus in Raji Cells
Paolo Di Francesco,*,1Antonella Lisi,† Sabrina Rieti,† Vanessa Manni,†
Settimio Grimaldi,† and Enrico Garaci*
*Department of Experimental Medicine and Biochemical Sciences, Microbiology, University of Rome Tor Vergata,
Rome, Italy; and †Institute of Experimental Medicine, CNR, Rome, Italy
Received August 21, 1999
T his paper shows that cocaine amplifies E pstein–
Barr virus (E BV) reactivation in R aji cells. Its effect on
early viral protein synthesis was maximal when it was
(T PA) plus n-butyrate, but nil when added alone. T he
enhancing effect of cocaine on early replicative stages
of latent E BV was associated with an increase of Ca2?
mobilization induced by the drug and with an induc-
tion of cellular protein phosphorylation in chemicals
and cocaine-treated R aji cells. Cocaine also acted syn-
ergistically with T PA and n-butyrate to induce Z
E pstein–Barr replication activator (ZE BR A), a nuclear
phosphoprotein responsible for the activation of early
viral gene expression. T hese findings provide the first
evidence that cocaine may represent an important co-
factor in the reactivation of early stages of latent E BV
© 1999 Academic Press
Recent works have shown that cocaine has a direct
effect on the susceptibility toviral infection of a patho-
gen’s target cells. Peterson et al. (1, 2) found that
cocaine potentiates human immunodeficiency virus
type 1 (HIV-1) replication in mitogen- or cytomegalo-
virus-activated human peripheral blood mononuclear
cells (PBMC). Bagasra and Pomerantz (3) also found
that PBMC treated with cocainehad increased levels of
HIV-1 replication after an acute infection in vitro. In
addition, the rate of HIV-1 production in alveolar mac-
rophages from crack abusers is increased two- tothree-
fold when compared toinfected cells from non-smoking
subjects (4). We have recently demonstrated that co-
caine increases the replication of a cytoplasmic RNA
virus, i.e., parainfluenza-1 Sendai virus, and this
change in susceptibility to infection is directly related
to the alteration in intracellular redox status induced
by cocaine (5). Whether cocaine can also affect the
induction of the replicative cycle of viral latent infec-
tions is at the present unknown.
Epstein–Barr virus (EBV) is a human herpes virus
etiologically linked to chronic active virus infection,
infectious mononucleosis and closely associated with
human malignancies, i.e., Burkitt’s lymphoma, naso-
pharyngeal carcinoma and Hodgkin’s disease (6, 7).
EBV can infect human B lymphocytes: cells became
immortalized and the viral genome persists in infected
cells in a latent state (8). A variety of chemical and
biological inducing agents added in vitro to the cells
can trigger viral reactivation. These include phorbol
esters, n-butyrate, calcium modulators, nitric oxide,
anti-immunoglobulins and the superinfection with the
lytic strain P3HR-1 of EBV (9–11). The capacity to
activatereplication results initially in theexpression of
viral gene BZLF1, which encodes a protein termed Z
Epstein–Barr replication activator (ZEBRA). This pro-
tein is responsible for activation of the viral early gene
expression and ultimately viral replication (12, 13).
To investigate the effect of cocaine on EBV reactiva-
tion, we utilized a virus-carrying nonproducer human
Burkitt’s lymphoma cell line (Raji cells). Our data
showed that cocaine potentiates the early stages of the
viral replicative cycle induced in Raji cells by treatment
with 12-O-tetradecanoyl phorbol-13-acetate (TPA) and
n-butyrate. Its effect is associated with induction of in-
tracellular Ca2?mobilization and cellular protein phos-
phorylation and with enhanced expression of ZEBRA.
Scotland) supplemented with 10% fetal calf serum (Gibco Lab.).
B-lymphoid cell line that expresses CD21.
Raji cells (5 ? 106/ml) were rinsed with
Locke’s solution (154 mM NaCl, 5.6 mM KCl, 3.6 mM NaHCO3, 2.3
mM CaCl2, 5.6 mM Glucose, 5 mM Hepes, pH 7.4) containing 10 mM
MgCl2. After washing cells were suspended in 100 (Locke’s solution
containing 0.4 mM MgCl2and 0.3 mCi/ml
Raji cells were grown in RPMI 1640 (Gibco Lab.,
1To whom correspondence should be addressed. E-mail: difra@
Biochemical and Biophysical Research Communications 264, 33–36 (1999)
Article ID bbrc.1999.1447, available online at http://www.idealibrary.com on
Copyright © 1999 by Academic Press
All rights of reproduction in any form reserved.
ersham) 1 h on a rocker at room temperature. Cells were then
treated with cocaine and chemicals and lysed with isoelectrofocusing
buffer (ethylene glycol 33%, NP-40 2.5%, ampholine pH range 3-10
8%, SDS 0.05%, DTT 0.2 mM).
focusing gel (1.8 ? 75 mm) containing a 3% total carrier ampholytes
mixture composed of 75% pH 3.5-9.5 25%, pH 5.0-8.0, 9 M Urea,
0.5% Nonidet P-40, 1.6% 3-(3-cholamidopropyl)-dimethylammonio-1-
propane sulphonate and 5% acrylamide were run for 1500 Vh. The
second dimension was carried out on 6-14% SDS polyacrylamide gra-
dient gel. Incorporation of
and, in two-dimensional gels, quantitated with a phosphorimages den-
Thefirst dimension, nonequilibriumisoelectric
32P-label was revealed by autoradiography
Immunoblotting for detection of early antigens and ZEBRA.
Cells, suspended in SDS sample buffer at a concentration of 1 ? 106
cells/ml, were sonicated and boiled prior toelectrophoresis through a
10% polyacrylamide gel. After electrophoresis, gel were transferred
tonitrocellulose over night and blocked with 5% milk buffer (5% non
fat dry milk, 0.03% sodium azide in PBS) for 2 h at 37°C. Then
nitrocellulose was incubated with 1:50 dilution of anti-ZEBRA or
1:200 dilution of anti EA specific monoclonal antibodies in milk
buffer for 18 h at 4°C, washed with Tween buffer (0.05 mM Tris-HCl,
pH 7.4, 0.15M NaCl, 5 mM EDTA, 0.05% Ten Tween 20) and reacted
with a 1:100 dilution of anti-mouse IgG biotynilated for 1 h at 4°C,
washed with Ten Tween buffer and reacted with a 1:100 dilution
streptavidin horseradish peroxidase coniugate, than developed with
the peroxidase colorimetric methods.
complete medium, starting at a concentration of 5 ? 105/ml. The cells
were washed, air-dried, fixed in cold acetone for 10 min, and tested by
indirected immunofluorescence for the presence of EBV early antigens.
An EA?serum(titer 1:1280) froma nasopharyngeal carcinoma patients
and an EA-serum from a healthy donor were used for the staining at a
dilution of 1:20. After 45 min of incubation, the slides were washed
three timeswithPBS, dried,
isothiocyanate-conjugatedgoat antiserumtohuman immunoglobulin G
(Cappel, West Chester, PA) at a dilution of 1:15. After further incuba-
tion for 45 min the slides were washed and examined by fluorescence
microscopy at least 500 cells were counted in each assay.
Raji cells were cultured for 72 h in
and stainedwith fluorescein-
Detection of intracellular calcium.
Krebs-Ringer solution were mixed with a stock solution of 10 mM
Quin 2-AM in dimethyl sulfoxide toa final concentration of 100 mM,
incubated for 20 min at 37°C, diluted tenfold, and further incubated
for 40 min. Finally the cells were washed and resuspended in Krebs-
Ringer (containing Ca2?) or in Krebs-Ringer [Ca]2?-free solution con-
taining 1 mM EGTA. Fluorescence was determined with a recording
spectrofluorimeter (Perkin-Elmer 650-40) (emission, 490 nm; excita-
tion, 339 nm). The relation of fluorescence at 492 nm with the
concentration of free calcium is determined from the equation:
[Ca]2?? Kd(F ? Fmin)/(Fmax? F), where Kdis 1.15 ? 10?7M, F is the
fluorescence of the intracellular indicator, Fmaxis designal detected
after lysis of the cells with Triton X-100 (0.05%), and Fmin is the
signal after lysis with Triton X-100 and addition of EGTA (4 mM).
Cells suspended in Ca2?-free
EBV reactivation is marked by renewed synthesis of
early gene products such as the EBV early antigens
(EAs). In order to study the effect of cocaine on EBV-
EAs expression, Raji cells were treated for 72 h with
different cocaine concentrations (10 and 100 ?g/ml)
alone or in combination with chemicals, such as TPA
(30 ng/ml) and n-butyrate (3 mM). Figure 1 shows that
cocaine alone is unable to induce the expression of
early antigens tested both by immunoblotting assay
(panel A) and by indirect immunofluorescence (panel
B). On the contrary, the density of immunoblot lanes 4,
5 and 6 in Fig. 1A showed that the treatment of co-
caine, in cells activated with TPA plus n-butyrate, in-
creases the expression of EBV-EAs respect tochemical
treatment alone. This enhancing effect was also con-
firmed by a significant induction of the percentage of
EAs-positive Raji cells treated with the combination of
drug and chemicals (Fig. 1B). Cell viability and cellular
F IG. 1.
with cocaine (10 and 100 ?g/ml) and chemicals (TPA, 30 ng/ml and n-butyrate, 3 mM). EAs were identified on the basis of their molecular
weights and in relation to the position of marker proteins (left of the gel). (B) EBV-EAs detected by indirect immunofluorescence. Raji cells
were treated for 72 h with cocaine (100 ?g/ml) and chemicals, as in A. Data points represent means ? SE of three different experiments, in
triplicate. TPA/n-but/Coc differed significantly from TPA/n-but (P ? 0.05 using Student’s t test).
(A) Effect of cocaine on EBV early antigens (EAs) expression detected by immunoblotting assay. Raji cells were treated for 72 h
Vol. 264, No. 1, 1999BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
protein synthesis were not affected by cocaine under
the conditions used (data not shown).
Activation of latent EBV is a result of a variety of early
intracellular signals (10). Our main objective was to de-
termine whether cocaine was interfering with the mobi-
lization of cytosolic free calcium, namely the most impor-
tant component of the reactivation stimulus (9, 14). Raji
cells weresuppliedwith cocaine(100 ?g/ml) and/or chem-
icals. As seen in Fig. 2A, the free calcium concentration
increasedupon exposuretothedrug, whilethetreatment
with TPA and n-butyrate had no effect on the Ca2?mo-
bilization. Additionally, incubation of the cells with co-
caine plus TPA and n-butyrate resulted in the same in-
duction when compared to phorbol ester alone. The
resultsobtainedmonitoring thefreeintracellular calcium
using the calcium fluorophore quin-2 in our system show
that cocaine alone is inducing the influx of extracellular
calcium, as alsodemonstrated in nervous and lymphatic
systems (15, 16).
Since protein phosphorylation plays a key role in
EBV reactivation (17), wealsoinvestigated theeffect of
cocaine on protein kinase activity. An autoradiograph
from Raji cells of the phosphorylated species resolved
in two-dimensional gels was shown in Fig. 2B (control
panel). Cell exposure to cocaine (100 ?g/ml) or chemi-
cals for 5 min resulted in a modulation in the incorpo-
are more phosphorylated (see spots indicated by ar-
rows in the panels). Interestingly, TPA and n-butyrate
acted additionally with cocaine to induce phosphoryla-
tion in Raji cells. A densitometric analysis revealed
that the expression of proteins was increased by from
30% to 70% respect to chemicals or drug alone.
EBV encodes a phosphoprotein, termed ZEBRA,
which enables the virus to switch from a latent to a
lytic lifecycle(13). After activation by phosphorylation,
ZEBRA binds to its own promoter, which is also the
EBV promoter. The results in Fig. 2C clearly show that
cocaine can trigger ZEBRA induction in Raji cells in
synergism with TPA/n-butyrate treatment.
32P in phosphoproteins: some of these proteins
The results presented here indicated that treatment
of Raji cells with a combination of cocaine and tumor
promoters enhanced the reactivation of EBV. In this
cell line the spontaneous induction of the EBV replica-
tive cycle normally either does not occur (latency) or
occurs at a low rate. The latency can be overcame in
vitro by the combination of various chemical-inducing
F IG. 2.
mM Ca2?. Arrow indicates the addition of cocaine (100 ?g/ml) and chemicals, respectively. The tracings shown were obtained from an
experiment representative of four. (B) Effect of cocaine on total protein phosphorylation. Raji cells were incubated with
for 1 h and then treated with cocaine (100 ?g/ml) and chemicals for 5 min. Cell homogenates were separated by charge in the first dimension
and in SDS-PAGE in thesecond dimension. Autoradiograph of thegels shows lebelled spots (arrows) corresponding toproteins with themajor
72 h. ZEBRA expression was detected with immunoblotting assay and identified on the basis of their molecular weights, in relation to the
position of marker proteins (left of the gel).
(A) Effect of cocaine on rise of Ca2?in quin 2-AM loaded Raji cells. Cells were suspended in Krebs-Ringer solution containing 1
32P incorporation. (C) Effect of cocaine on ZEBRA expression. Raji cells were treated with cocaine (100 ?g/ml) and chemicals for
Vol. 264, No. 1, 1999 BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
agents such as TPA and n-butyrate (9, 10). However,
our findings providethefirst evidencethat cocainemay
represent a pharmacological stimulus in the reactiva-
tion of latent EBV infection.
Although our results support the conclusion that the
combination of cocaine and chemicals is more effective
than treatment with either cocaine alone, we found a
variation in the Ca2?concentration induced only by co-
caine. Cocaine treatment of Raji cells induces in fact a
risein intracellular Ca2?frombasal level to180–200 nM.
We have previously shown that calcium ionophores (i.e.,
ionomycin) induce the replication of EBV in target cells,
and this event is mediated by activation of cellular pro-
tein kinases (9). While the increase in Ca2?induced by
cocainetreatment may becorrelatedin Raji cells with the
activation of calcium-dependent protein kinases, as
shown by the increase of
phoproteins, thesamereported increaseby cocainetreat-
ment seems tobe not sufficient toinduce EBV-EAs. This
is not surprising since it is known that the intracellular
calcium must be at ?M levels tohave detectable produc-
tion of viral antigens (9). On the other hand, TPA also
induces protein kinase activities that play an important
role in reactivating latent EBV (17, 18). Cocaine could
thus cooperatewith chemical treatment tomodulatepro-
tein phosphorylation, leading to an additional effect on
early viral antigen expression.
It is noted that cocaine generally by itself does not
activate cellular responses but, rather, the effect of the
drug is synergistic with other activating signals. For ex-
ample, the drug was incapable of stimulating HIV-1 rep-
lication in nonactivated PBMC, whereas marked aug-
mentation of viral replication was observed when
cocaine-treated PBMC were stimulated with mitogens
(1). Similarly, cocaine only potentiates HIV-1 replication
in cytomegalovirus-activatedPBMC (2). In our paper this
combined effect is particularly evident in theinduction of
ZEBRA.In fact cocaineisunableper setoinduceZEBRA,
but it acts synergistically with chemicals toupregulation
of theEBV replication activator. Thereis however a need
toknow more about the evidently complex mechanism(s)
through which cocaineinfluences viral replication. In this
regard, drug’s effect on early antigen expression in cells
stimulated by ligation of the B cell receptor, i.e. the more
likely scenario for in vivo activation, or in cells that are
spontaneously entering the lytic cycle is currently the
subject of further investigation.
In summary, our data suggested a potential physio-
logical role for cocaine on EBV reactivation. In partic-
ular, cocaine can be considered an important co-factor
in the activation of EBV latency. Cocaine seems to be
necessary, but not sufficient to induce the early stages
of EBV replication. The hypothesis that drugs of abuse
act as co-factors in increasing the severity of infections
is supported by several evidences from in vivo and in
vitro investigations (4, 19–23). However, the study of
co-factors already represents a promising line of re-
32P incorporation in some phos-
search aimed at understanding the pathogenesis of
viral diseases and hence their therapeutic approach.
Grant support: National Research Council (CNR, Contracts
PS04.97.02455 and Target Project on Biotechnology 97.01093. PF49)
and IstitutoSuperiore Prevenzione e Sicurezza sul Lavoro(ISPELS-
DIPIA, Contract 86/97-B).
1. Peterson, P. K., Gekker, G., Chao, C. C., Schut, R., Molitor, T. W.,
and Balfour, H. H. (1991) J . Immunol. 146, 81–84.
2. Peterson, PK., Gekker, G., Chao, C. C., Schut, R., Verhoef, J .,
Edelman, C. K., Erice, A., and Balfour, H. H. (1992) J . Immunol.
3. Bagasra, O., and Pomerantz, R. J . (1993) J . Infect. Dis. 168,
4. Baldwin, G. C., Roth, M. D., and Tashkin, D. P. (1998) J . Neu-
roimmunol. 83, 133–138.
5. Palamara, A. T., Di Francesco, P., Ciriolo, M., Lafavia, E., Ro-
tilio, G., and Garaci, E. (1996) Biochem. Biophys. Res. Commun.
6. Maeda, A., Wakiguchi, H., Yokoyama, W., Hisakawa, H., To-
moda, T., and Kurashige, T. (1999) J . Infect. Dis. 179, 1012–
7. Kallin, B., Luka, J ., and Klein, G. (1979) J . Virol. 32, 710–716.
8. Liebowitz, D., and Kieff, E. (1993) in The Human Herpesviruses
(Roizman, B., Whitley, R. J ., and Lopez, C., Eds.), pp. 107–172.
Raven Press, New York.
9. Faggioni, A., Zompetta, C., Grimaldi, S., Barile, G., Frati, L., and
Lazdins, J . (1986) Science 232, 1554–1556.
10. Takada, K., and Ono, Y. (1989) J . Virol. 63, 445–449.
11. Mannick, J . B., Asano, K., Izumi, K. Kieff, E., and Stamler, J . S.
(1994) Cell 79, 1137–1146.
12. Countryman, J ., J enson, H., Seibel, R., Wolf, H., and Miller G.
(1987) J . Virol. 61, 3672–3679.
13. Ragoczy, T., Heston, L., and Miller G. (1998) J . Virol. 72, 7978–
14. Mosialos, G., Hanissian, S. H., J awahar, S., Vara, L., Kieff, E.,
and Chatila, T. A. (1994) J . Virol. 68, 1697–1705.
15. Pierc, R. C., Quick, E. A., Reeder, D. C., Morgan, Z. R., and
Kalivas, P. W. (1998) J . Pharmacol. Exp. Ther. 286, 1171–1176.
16. Matsui, K., Friedman, H., and Klein, T. W. (1992) Int. J . Immu-
nopharmacol. 14, 1213–1220.
17. Aquino, A., Lisi, A., Pozzi, D., Ravagnan, G., and Grimaldi, S.
(1993) Biochem. Biophys. Res. Commun. 196, 794–802.
18. Satoh T., Hoshikawa Y., Satoh Y., Kurata T., and Sairenji T.
(1999) Virus Genes 18, 57–64.
19. Fiala, A. M., Gan, X. H., Newton, T., Chiappelli, F., Shapshak,
P., Kermani, V., Kung, M. A., Diagne, A., Martinez, O., Way, D.,
Weinand, M., Witte, M., and Graves, M. (1996) Adv. Exp. Med.
Biol. 402, 145–156.
20. Carr, D. J . J ., and Serou, M. (1995) Immunopharmacology 31,
21. Di Francesco, P., Gaziano, R., Casalinuovo, I. A., Belogi, L.,
Palamara, A. T., Favalli, C., and Garaci, E. (1994) Clin. Exp.
Immunol. 97, 347–354.
22. Nyland, S. B., Specter, S., Im-Sin, J ., and Ugen, K. E. (1998) Adv.
Exp. Med. Biol. 437, 91–100.
23. Chuang, L. F., Killam, K. F., and Chuang, R. Y. (1993) Biochem.
Biophys. Res. Commun. 195, 1165–1173.
Vol. 264, No. 1, 1999BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS