Amphotericin-B-mediated reactivation of latent HIV-1 infection
J. Jonesa, B.R. Kosloffb, E.N. Benvenisteb, G.M. Shawa,c, O. Kutscha,*
aDepartment of Medicine, The University of Alabama at Birmingham, Birmingham, AL, United States
bDepartment of Cell Biology, The University of Alabama at Birmingham, Birmingham, AL, United States
cThe Howard Hughes Medical Institute, Birmingham, AL, United States
Received 24 June 2004; returned to author for revision 3 September 2004; accepted 6 October 2004
Available online 2 November 2004
To date, attempts to eliminate HIV-1 infection from its latent reservoirs, a prerequisite for the development of a curative treatment strategy
for HIV-1 infection, have been unsuccessful. We demonstrate that the FDA approved antifungal agent amphotericin B efficiently reactivates
HIV-1 infection in THP89GFP cells, a model of HIV-1 latency in macrophages. Although amphotericin B does not directly reactivate latent
HIV-1 infection in T cells (e.g., J89GFP), amphotericin-B-stimulated macrophages (THP89GFP cells or primary macrophages) when
cocultured with J89GFP cells can induce HIV-1 reactivation in these cells in trans. Because of the close proximity of antigen presenting
macrophages and Tcells in the primary lymphoid organs, such interaction between antigen presenting macrophages and Tcells are frequent,
and it seems reasonable to assume that trans-reactivation strategies hold promise to also reactivate latent HIV-1 infection in vivo.
D 2004 Elsevier Inc. All rights reserved.
Keywords: HIV-1; Latency; Reactivation; Amphotericin B; Reporter cell lines; THP89GFP; J89GFP
While efficiently suppressing HIV-1 replication, highly
active antiretroviral therapy (HAART) cannot eradicate the
virus from infected patients (Chun et al., 1999a, 2000;
Davey et al., 1999; Wong et al., 1997). Present HIV-1
therapy would thus require a life-long treatment, which is
hampered by drug side effects, development of drug
resistant virus strains, and high treatment costs. In this
setting, the development of a curative treatment strategy
with the goal to eradicate HIV-1 from the infected patients is
a major challenge. An essential step towards this goal is the
depletion of the viral reservoirs in latently infected cells.
According to the present understanding, HIV-1 latency
develops primarily in memory T cells and macrophages
(Chun et al., 1997a, 1997b, 1998; Crowe and Sonza, 2000;
Lebargy et al., 1994; Mikovits et al., 1992; Sonza and
Crowe, 2001) (reviewed in (Crowe et al., 2003). In vitro,
reactivation of latent HIV-1 infection in both cell types can
be achieved by a variety of physiological stimuli, such as
TNF-a, IL-1h, IL-2 IL-6, IFN-g, or CD154 (Biswas et al.,
1992; Butera et al., 1991; Duh et al., 1989; Folks et al.,
1987; Kutsch et al., 2003; Poli et al., 1994). Chemical
compounds that reactivate latent HIV-1 infection, among
others, are the phorbol esters PMA (Folks et al., 1988) and
prostratin (Korin et al., 2002; Kulkosky et al., 2001), and the
histone deacetylase inhibitor trichostatin A (TSA) (Quivy et
al., 2002; Sheridan et al., 1997; Van Lint et al., 1996) and
sodium butyrate (Laughlin et al., 1993, 1995). In addition,
certain activating antibodies can mediate HIV-1 reactivation
(anti-CD3) (Tong-Starkesen et al., 1989; Wong et al., 1997).
Although these stimuli efficiently reactivate latent HIV-1
infection in vitro, severe side effects prevent their usage in
vivo. To date, only IL-2 and anti-CD3 antibodies have been
used in small clinical trials. However, initial studies in
patients have shown no meaningful decay of the viral latent
reservoir but significant adverse side effects (Chun et al.,
1999b; Dybul et al., 2002; Kovacs et al., 2000, 2001;
Kulkosky et al., 2002).
0042-6822/$ - see front matter D 2004 Elsevier Inc. All rights reserved.
* Corresponding author. Division of Hematology/Oncology, University
of Alabama at Birmingham, KAUL 840, 720 20th Street South,
Birmingham, AL, 35294. Fax: +1 205 934 1580.
E-mail address: email@example.com (O. Kutsch).
Virology 331 (2005) 106–116
To screen for and analyze the effect of pharmaceutical
compounds with HIV-1 reactivating properties, we have
recently established a cell-based reporter system that allows
direct quantification of HIV-1 reactivation on a single cell
basis, using EGFP as a read-out (Kutsch et al., 2002). The
system consists of the monocytic THP89GFP cells and two
T cell lines, J89GFP and JNLGFP, which contain an
integrated but transcriptionally silent copy of molecular
clones of HIV-1 89.6 and HIV-1 NL4-3, respectively,
engineered to express EGFP.
In this study, we used these cell lines to study the effect of
HIV-1 infection. Amphotericin B is the treatment of choice
for a variety of mycoses and is used in the treatment offungal
infections in AIDS patients (Coker et al., 1993; Hsieh et al.,
vitro, amphotericin B and its derivate MS-8209 have been
reported to induce TNF-a expression in macrophages and to
increase HIV-1 expression in acutely infected macrophages
(Chia and Pollack, 1989; Clayette et al., 2000; Tokuda et al.,
1993). Other reports demonstrated that amphotericin B and
its derivate MS-8209 have anti-HIV-1 properties, inhibiting
viral replication in Tcells in in vitro experiments (Cefai et al.,
1991; Hansen et al., 1990; Konopka et al., 1999; Otake et al.,
1989; Pleskoff et al., 1995, 1996; Pontani et al., 1989a,
1989b; Schaffner et al., 1986).
Here, we report the ability of amphotericin B to reactivate
latent HIV-1 infection. Low doses of amphotericin B are
sufficient to directly reactivate latent HIV-1 infection in the
monocytic cell line THP89GFP. Although amphotericin B
did not directly reactivate latent HIV-1 infection in T cells
(J89GFP cells), cocultures of amphotericin-B-stimulated
THP89GFP cells or primary macrophages with J89GFP cells
were able to trigger HIV-1 reactivation in J89GFP in trans.
In vivo, such a btrans-reactivationQ approach would
exploit the frequent immunological interaction of T cells
and macrophages during antigen presentation. Specific
activation of macrophages could directly reactivate latent
HIV-1 in macrophages, which upon cell-to-cell contact
trigger reactivation of latent HIV-1 in T cells.
This approach, here exemplified by the use of amphoter-
icin B, adds to the strategies that should be further
investigated as an alternative approach to eradicate the pool
of latent HIV-1 from infected patients.
Amphotericin-B-mediated reactivation of latent HIV-1
It is an accepted concept that specific activation of cells
harboring latent HIV-1 can result in reactivation of HIV-1
infection. Unfortunately, presently no FDA approved drugs
are described that could be used for this purpose. As
amphotericin B, a potent approved antifungal drug, has
previously been described to activate macrophages, we
thought to determine the effect of amphotericin B on latent
HIV-1 infection in THP89GFP cells, a model for latently
infected macrophages, and the latently infected T cell lines
J89GFP and JNLGFP.
Treatment of the latently infected Tcell lines J89GFP and
JNLGFP with amphotericin B did not result in reactivation
of latent HIV-1 infection. However, HIV-1 reactivation in
THP89GFP cells was seen at amphotericin B concentrations
as low as 0.3 Ag/ml. Near-complete reactivation of HIV-1
infection in THP89GFP cells on a population level was
achieved at an amphotericin B concentration of 3 Ag/ml. At
a concentration of 10 Ag/ml of amphotericin B, levels of
HIV-1 expression measured as EGFP mean channel
fluorescence were comparable to those seen following
stimulation with TNF-a (10 ng/ml) (Fig. 1). As seen
following stimulation with TNF-a, amphotericin-B-medi-
ated HIV-1 reactivation was accompanied by a pronounced
increase in the cell surface expression of ICAM-1, but in
contrast to stimulation with IFN-g, no induction of MHC
class II expression was seen, suggesting that amphotericin B
exerts its function mostly by activating the NF-nB pathway
Augmentation of amphotericin B triggered HIV-1
reactivation in THP89GFP cells by autocrine secreted
It has been previously reported that stimulation of macro-
phages with amphotericin B results in TNF-a secretion,
which in turn could contribute to HIV-1 reactivation (Chia
and Pollack, 1989; Tokuda et al., 1993). TNF-a secretion
be controlled by the ERK MAPK pathway (Andersson and
Sundler, 2000; Pearson et al., 2001; Suttles et al., 1999). To
investigate the possible contribution of TNF-a secretion to
amphotericin-B-mediated HIV-1 reactivation in THP89GFP
cells, we thus pretreated the cells for 1 h with various
concentrations of the specific ERK inhibitor U0126 (0.01–1
AM), or the inactive control compound UO124 (1 AM), and
then stimulated the cells with amphotericin B (10 Ag/ml).
Levels of TNF-a in the culture supernatants were determined
by ELISA and compared to levels of HIV-1 expression in
THP89GFP cells, as measured by flow cytometry. 50 pg/ml
TNF-a were detected in the supernatants of unstimulated
THP89GFP cells. Forty-eight hours after stimulation, TNF-a
levels in cultures treated with amphotericin B alone were
increased to 300 pg/ml. Addition of increasing amounts of
U0126 (0.1–1.0 AM) gradually inhibited TNF-a secretion to
baseline levels but had no effect on EGFP fluorescence as a
marker of HIV-1 expression, indicating that the presence of
TNF-a in the supernatants is mostly augmenting HIV-1
expression but is not necessary to trigger HIV-1 reactivation.
Higher concentrations of U0126 also affected the levels of
HIV-1 reactivation. As concentrations of U0126 up to 10 AM
did not alter the viability of the cells, this suggests that the
J. Jones et al. / Virology 331 (2005) 106–116
Tong-Starkesen, S.E., Luciw, P.A., Peterlin, B.M., 1989. Signaling through
T lymphocyte surface proteins, TCR/CD3 and CD28, activates the HIV-
1 long terminal repeat. J. Immunol. 142 (2), 702–707.
Van Lint, C., Emiliani, S., Ott, M., Verdin, E., 1996. Transcriptional
activation and chromatin remodeling of the HIV-1 promoter in response
to histone acetylation. EMBO J. 15 (5), 1112–1120.
Voss, T.G., Fermin, C.D., Levy, J.A., Vigh, S., Choi, B., Garry, R.F., 1996a.
Alteration of intracellular potassium and sodium concentrations
correlates with induction of cytopathic effects by human immunodefi-
ciency virus. J. Virol. 70 (8), 5447–5454.
Voss, T.G., Gatti, P.J., Fermin, C.D., Garry, R.F., 1996b. Reduction of
human immunodeficiency virus production and cytopathic effects
by inhibitors of the Na+/K+/2Cl-cotransporter. Virology 219 (1),
Wong, J.K., Hezareh, M., Gunthard, H.F., Havlir, D.V., Ignacio, C.C.,
Spina, C.A., Richman, D.D., 1997. Recovery of replication-competent
HIV despite prolonged suppression of plasma viremia. Science 278
Zhang, L., Chung, C., Hu, B.S., He, T., Guo, Y., Kim, A.J., Skulsky, E., Jin,
X., Hurley, A., Ramratnam, B., Markowitz, M., Ho, D.D., 2000.
Genetic characterization of rebounding HIV-1 after cessation of highly
active antiretroviral therapy. J. Clin. Invest. 106 (7), 839–845.
J. Jones et al. / Virology 331 (2005) 106–116