ANTIMICROBIAL AGENTS AND CHEMOTHERAPY, June 2011, p. 2696–2703
Copyright © 2011, American Society for Microbiology. All Rights Reserved.
Vol. 55, No. 6
The HIV Protease Inhibitor Nelfinavir Inhibits Kaposi’s
Sarcoma-Associated Herpesvirus Replication In Vitro?
Soren Gantt,1,7Jacquelyn Carlsson,7Minako Ikoma,3Eliora Gachelet,2Matthew Gray,3
Adam P. Geballe,2,4,8,9Lawrence Corey,3,4,9,10Corey Casper,4,5,6,9,10,11
Michael Lagunoff,2and Jeffrey Vieira3*
Departments of Pediatrics,1Microbiology,2Laboratory Medicine,3Medicine,4Global Health,5and Epidemiology,6University of
Washington, Seattle Children’s Hospital,7and Human Biology,8Clinical Research,9Vaccine and Infectious Diseases,10and
Public Health Sciences11Divisions, Fred Hutchinson Cancer Research Center, Seattle, Washington
Received 22 September 2010/Returned for modification 27 October 2010/Accepted 4 March 2011
Kaposi’s sarcoma (KS) is the most common HIV-associated cancer worldwide and is associated with high
levels of morbidity and mortality in some regions. Antiretroviral (ARV) combination regimens have had mixed
results for KS progression and resolution. Anecdotal case reports suggest that protease inhibitors (PIs) may
have effects against KS that are independent of their effect on HIV infection. As such, we evaluated whether PIs
or other ARVs directly inhibit replication of Kaposi’s sarcoma-associated herpesvirus (KSHV), the gamma-
herpesvirus that causes KS. Among a broad panel of ARVs tested, only the PI nelfinavir consistently displayed
potent inhibitory activity against KSHV in vitro as demonstrated by an efficient quantitative assay for infectious
KSHV using a recombinant virus, rKSHV.294, which expresses the secreted alkaline phosphatase. This
inhibitory activity of nelfinavir against KSHV replication was confirmed using virus derived from a second
primary effusion lymphoma cell line. Nelfinavir was similarly found to inhibit in vitro replication of an
alphaherpesvirus (herpes simplex virus) and a betaherpesvirus (human cytomegalovirus). No activity was
observed with nelfinavir against vaccinia virus or adenovirus. Nelfinavir may provide unique benefits for the
prevention or treatment of HIV-associated KS and potentially other human herpesviruses by direct inhibition
Kaposi’s sarcoma-associated herpesvirus (KSHV; also called
human herpesvirus 8 [HHV-8]) is a member of the gamma-
herpesvirus family and causes Kaposi’s sarcoma (KS). KS is the
most common cancer in HIV-infected people worldwide (1,
42). In many parts of sub-Saharan Africa, where KSHV infec-
tion is highly prevalent, KS has become the most common
cancer in the general population (43). Even where antiretro-
viral (ARV) treatment and cancer chemotherapy are available,
complete resolution of KS is achieved in only ?50% of cases
(40), highlighting the need for novel KS prevention and treat-
KSHV replication is central to the pathophysiology of KS.
The detection of KSHV in the peripheral blood is strongly
correlated with the development of KS (20, 53), and the pres-
ence of lytic KSHV appears to be required for the maintenance
of KS tumors (24). The loss of immune control over KSHV
replication due to HIV/AIDS or immunosuppressive medica-
tions appears to be the dominant risk factor for development
of KS (11, 38, 41, 54). Given the importance of KSHV repli-
cation in the clinical manifestations of KS, it stands to reason
that inhibition of KSHV replication could be an important
component of strategies to prevent or treat KS. Data from
cohort studies in the early HIV epidemic have shown that the
administration of ganciclovir to HIV-infected patients for the
treatment of human cytomegalovirus (HCMV) retinitis re-
sulted in lower rates of KS (34, 36). Subsequent studies found
that the herpesvirus polymerase antagonists ganciclovir (and
its prodrug valganciclovir), cidofovir, and foscarnet have activ-
ity against KSHV both in vitro (28, 35, 39) and in vivo (8, 9),
potentially explaining the ability of these drugs to prevent KS.
Additional data from HIV cohorts in the early epidemic also
suggested that specific components of ARV regimens might
impact the incidence and resolution of KS. Treatment of HIV
with high-dose zidovudine monotherapy resulted in a reduced
incidence of KS in some, but not all, trials (27). ARV combi-
nations that contain HIV protease inhibitors (PIs) may be
superior to those without PIs for treatment of patients with KS
(2, 21, 30, 45). The efficacy of ARVs in the treatment and
prevention of KS has largely been attributed to their ability to
suppress HIV replication and improve immune reconstitution.
However, few data support the idea that PIs are more effective
in these two areas than are other ARV regimens; in fact, the
effects of PIs on KS were often independent of their effect on
HIV. Moreover, recent research has shown that PIs have an-
tiangiogenic and antitumor properties (44, 47).
There is precedent that ARVs may affect herpesvirus repli-
cation: zidovudine and stavudine have been shown to be sub-
strates for the KSHV thymidine kinase (ORF21) (32) and
therefore could directly inhibit KSHV replication, and ARVs
have been shown to significantly reduce the detection of rep-
licating KSHV in the oropharynx of HIV-infected men (10).
To date, no comprehensive studies to test whether ARVs are
able to inhibit KSHV viral production have been conducted
(28, 48). We used a novel in vitro assay, based on a recombi-
* Corresponding author. Mailing address: Division of Virology, De-
partment of Laboratory Medicine, University of Washington, Box
358070, 1959 NE Pacific Street, Seattle, WA 98109-8070. Phone: (206)
732-6107. Fax: (206) 732-6109. E-mail: firstname.lastname@example.org.
?Published ahead of print on 14 March 2011.
nant virus expressing the secreted alkaline phosphatase
(SeAP), to evaluate a broad panel of ARVs, including PIs,
zidovudine, and stavudine, for their ability to inhibit KSHV
MATERIALS AND METHODS
Cells. All cells were maintained at 37°C in a humidified 5% CO2atmosphere.
Human fibroblasts (HF) and Vero cells were cultivated in Dulbecco’s modified
Eagle’s medium (DMEM; Gibco) containing 10% fetal bovine serum (FBS) and
100 units per ml penicillin G and 100 ?g per ml streptomycin (Pen-Strep), as
previously described (7, 14, 52). Telomerase-immortalized microvascular endo-
thelial (TIME) cells (50) were maintained in EGM-2 MV medium (Lonza)
supplemented with a bullet kit containing FBS; vascular endothelial growth
factor; basic fibroblast growth factor; insulin-like growth factor 1; epidermal
growth factor; and hydrocortisone, ascorbic acid, gentamicin, and amphotericin
B (29). A549 cells were a gift from Tim Rose (Seattle Children’s Hospital) and
cultivated in F-12K medium (Gibco) with 10% FBS and Pen-Strep. Vero and 293
cells stably transfected with the tetracycline-controlled transactivator tTA2
(Clontech, Mountain View, CA) were maintained under selection with hygro-
mycin (100 ?g/ml).
Viruses. KSHV inocula used to infect TIME cells were obtained from BCBL-1
cells as previously described (29). The recombinant virus rKSHV.294 was con-
structed by inserting the secreted alkaline phosphatase (SeAP) gene (4) into
pTRE-Tight (Clontech, Mountain View, CA). The Tet-responsive element
(TRE)–SeAP construct was isolated as a XhoI fragment and inserted into a XhoI
site in pQ152 (51) adjacent to the green fluorescent protein (GFP) gene to create
pQ294. Vero cells containing latent rKSHV.219 (derived from JSC-1 cells )
were transfected with pQ294 so that the GFP/red fluorescent protein (RFP)/
puromycin (Puro) construct in rKSHV.219 could be replaced by the GFP/SeAP/
Neo construct of pQ294 by homologous recombination, and 2 days posttrans-
fection, the cells were grown with G418 selection (500 ?g/ml). Once the cultures
were selected with G418, the virus in the cells was activated by infection with
BacK50 and treatment with sodium butyrate, as described previously (52). Three
days postactivation, cell-free supernatant from the induced cultures was used to
infect fresh Vero cells at a low multiplicity of infection (MOI), and the infected
cells were selected with G418. Once these cultures were confluent with G418, the
activation of virus and infection of fresh Vero cells were repeated. Colonies that
formed under G418 selection were cloned, expanded, and checked for the loss of
puromycin resistance and for the absence of RFP expression upon lytic activa-
tion. Virus from puromycin-sensitive, RFP-negative cultures was then analyzed
by PCR for purity and correct insertion site, for the identification of a correct
rKSHV.294 isolate. Herpes simplex virus type 1 (HSV-1) (strain F) was a gift
from Keith Jerome (Fred Hutchinson Cancer Research Center). HCMV Towne
strain was propagated in HF. A LacZ-recombinant vaccinia virus (VV) was
constructed by transfecting pSC11 (12) into cells infected with VV Copenhagen
strain VC2 and plaque purifying a ?-galactosidase-expressing recombinant virus.
Wild-type adenovirus type 5 (Ad5) was a gift from A. Dusty Miller (Fred
Hutchinson Cancer Research Center).
PCR. The following primers were used for PCR analysis of rKSHV.294 viral
DNA, 5?-3?: ORF56f, GATACTGGGAGCAAAGTGTG; SeAPr, TGTCCTTC
TTCTGCCCTTTGAGAATCCTGG; ORF57u, TTGCCAAACCCCATGGCA
GAGTG; SeAPf, TGCTGCTGCTGCTGCTGCTGGGCCT; GFPf, TGACCA
CCTTGACCTAC; GFPr, CTCAGGTAGTGGTTGTC; Neor, TAGCCGGAT
CAAGCGTATG; and K9, TTGCGGCGAGGTGCAGTAATTTC.
PCR was carried out in 30-?l reaction mixtures using 1 ?l of virus DNA in 1?
PCR buffer (Invitrogen) with 2.5 mM MgCl2, 200 ?M (each) deoxynucleoside
triphosphate (dNTP), 6 pmol of each primer, and 1.5 units of AmpliTaq DNA
polymerase (Applied Biosystems). Reactions were performed for 5 min at 94°C,
followed by 35 cycles of 30 s at 94°C, 30 s at 54°C, and 6 min at 72°C, followed
by a final extension step for an additional 6 min at 72°C.
Viral DNA hybridization. Total genomic DNA was purified from rKSHV.294
recombinant or JSC-1 wild-type virus-infected cells by standard methods (46).
Twenty micrograms of each DNA sample was digested overnight at 37°C with
either AflII-HF or SspI-HF restriction endonucleases according to the manufac-
turer’s recommendations (New England BioLabs, Ipswich, MA). Gel electro-
phoresis, alkaline transfer, Southern blot hybridization, and stringency washes
were performed as described by Sambrook and Russell (46). Prehybridizations
and hybridizations were carried out in ULTRAhyb ultrasensitive hybridization
buffer (AM8670; Applied Biosystems). To generate biotin-labeled probes, PCRs
were carried out as described above except that the starting concentration of
dTTP was reduced from 200 to 120 ?M, and reaction mixtures were supple-
mented with 80 ?M biotin-16-dUTP (Biotium, Hayward, CA). Five primer sets
were used to generate 5 separate KS-specific probes that, together, correspond to
the entire wild-type 4,774-bp BamHI KSHV fragment used to construct
rKSHV.294. One other primer pair was used to generate a 791-bp biotin-labeled
probe corresponding to the neomycin resistance gene, present only within
rKSHV.294. Synthesis of individual labeled PCR products was confirmed by gel
electrophoresis, and positive detection of biotin incorporation was verified by dot
blot analyses of the individual probes after removal of unincorporated label.
Hybridization of biotinylated probes to the DNA blots was detected using per-
oxidase-labeled streptavidin (catalog no. 474-3000; KPL, Gaithersburg, MD) at
a 1:500 dilution in 1? PBS-0.05% Tween 20, followed by chemiluminescent
substrate (SuperSignal West Pico chemiluminescent substrate, product no.
43080; Thermo Scientific, Rockford, IL) and exposure of X-ray film (CL-
XPosure film, product no. 34092; Pierce) according to the manufacturers’ rec-
The primers used to generate the KSHV probes were as follows: KS1F,
GTCGGTGTCATGACAAACTG; KS1R, CGACGAAGATAGCACGATAC;
KS2F, GCATCATGGATCGCAATGAG; KS2R, CACCGTCAATTATCATG
CTC; KS3F, GAGCATGATAATTGACGGTG; KS3R, TTGCGGCGAGGTG
CAGTAATTTC; KS4F, TTGCCAAACCCCATGGCAGAGTG; KS4R, CTCG
TTTAAAGGCACCAG; KS5F, CTGGTGCCTTTAAACGAG; and KS5R, AT
GGCTAGATTTCGCACCAG. Those primers used to generate the Neo probe
were neoF, CAAGATGGATTGCACGCAG, and neoR2, CCCGCTCAGAAG
Drugs. All antiretroviral drugs were obtained through the AIDS Research and
Reference Reagent Program, Division of AIDS, NIAID, NIH. Atazanavir,
amprenavir, lopinavir, nelfinavir, ritonavir, saquinavir, abacavir, didanosine, zi-
dovudine, efavirenz, and nevirapine were solubilized in dimethyl sulfoxide
(DMSO). Indinavir, lamivudine, stavudine, tenofovir, acyclovir (Sigma-Aldrich),
and ganciclovir (Sigma-Aldrich) were solubilized in water.
Cytotoxicity testing. Cytotoxicity for each drug was determined in parallel with
antiviral activity using the same cell lines and conditions. All cell types were
incubated in the presence of serially diluted drug or solvent under conditions
used for assessment of antiviral activity, after which cytotoxicity was measured by
the TOX-1 MTT [3-(4,5-dimethyl-2-thiazolyl)-2,5-diphenyl-2H-tetrazolium bro-
mide] assay (Sigma-Aldrich) and verified by measurement of ATP concentration
(CelltiterGlo; Promega), according to the manufacturer’s instructions. Assays
were performed using triplicate treatments and tested in ?2 independent exper-
iments. For the antiviral activity experiments reported below, drugs were used
only at concentrations that did not result in detectable cytotoxicity by either
KSHV yield reduction screening using the SeAP-recombinant virus. Vero cells
infected with rKSHV.294 were grown to 60 to 80% confluence in 96-well plates.
Baculovirus expressing replication and transcription activator (RTA; ORF50)
was added to wells for 30 min to induce lytic replication (52). Medium containing
BacK50 was removed and replaced with medium containing 1.5 mM sodium
butyrate (52) as well as serially diluted drug or solvent controls (equivalent
concentration of DMSO or water). All treatments were performed in triplicate
wells. After incubation for 72 h, the medium was collected to harvest the virus
produced in each treatment well. Harvested virus was transferred to Vero-tTA
cells at 60 to 80% confluence, which activates the TRE promoter and allows
expression of SeAP in cells infected with rKSHV.294. After 48 h, medium was
collected from wells containing Vero-Tta cells and transferred to an empty
96-well plate to measure SeAP activity. Plates containing medium were incu-
bated at 65°C for 30 min to inactivate endogenous alkaline phosphatase. Then,
3 ?M 4-methylumbelliferyl phosphate substrate (MUP; Sigma-Aldrich) was
added, and SeAP activity was measured by fluorescence measurements (excita-
tion, 360 nm; emission, 450 nm). For initial experiments, the yield reduction of
rKSHV.294 measured by SeAP activity was confirmed by manually counting the
number of GFP-expressing cells by fluorescent microscopy.
BCBL-1-derived KSHV yield reduction using immunofluorescence. TIME
cells at 60 to 80% confluence in 6-well plates were infected with KSHV for 3 h
as previously described, and the inoculum was removed (37). Cells were then
infected with adenovirus expressing RTA (a generous gift from Don Ganem,
University of California—San Francisco) in the presence of 1 ?g/ml poly-L-lysine
for 2 h in order to activate KSHV lytic replication, after which cells were washed
and incubated for 48 h to allow for viral production in medium containing either
drug or solvent control in the presence of 1.5 mM sodium butyrate. All treat-
ments were performed in triplicate. The medium was harvested, centrifuged at
300 ? g for 10 min to pellet cellular debris, and then used to infect fresh TIME
cells at 60 to 80% confluence in chamber slides (LabTek; Nunc) for 3 h. The
inoculum was then replaced with fresh medium, and the cells were incubated for
48 h. Nuclear staining using 4?,6-diamidino-2-phenylindole (DAPI) and immu-
VOL. 55, 2011 NELFINAVIR INHIBITS KSHV REPLICATION 2697
nofluorescent staining of LANA (ORF73) were performed as described previ-
ously (29), and the proportion of infected cells derived from each treatment was
measured by manual counting of LANA expression among 200 cells per treat-
HCMV, HSV-1, and Ad5 yield reduction and quantification of PFU. One-step
growth experiments were performed for HCMV, HSV-1, and Ad5, each at a
multiplicity of infection (MOI) of 1 PFU/cell. All treatments were performed in
duplicate or triplicate, and results were confirmed by ?2 independent experi-
ments. For HCMV, HF were incubated with virus for 1 h, and then virus was
removed and replaced with medium containing serial dilutions of drug or solvent
control for 3 days. Medium was removed, cellular debris was pelleted, and the
concentration of virus produced on HF under each condition was measured by
standard titration of PFU per ml after 5 to 7 days, until plaque formation was
evident. Cells were fixed with 0.5% formaldehyde and stained with 0.1% crystal
violet (Sigma-Aldrich) in phosphate-buffered saline, and plaques were counted
manually. HSV-1 was similarly tested in both HF and Vero cells using a 1-hour
infection, after which the inoculum was removed and cells were washed before
addition of complete medium containing drug or solvent control. After 24 h,
supernatants and cells were frozen at ?70°C, thawed 3 times to release intra-
cellular virus, and diluted 4:1 in 10% skim milk. Titration of HSV-1 was per-
formed with the addition of pooled human serum (2.5 ?g/ml) after infection to
reduce secondary infection of cells and allow the formation of discrete plaques
at ?48 h. Fixation and staining were performed as described for HCMV. Ad5
was tested in A549 cells using a 16-hour infection, after which the inoculum was
removed and cells were washed before addition of medium containing drug or
solvent control. After 72 h, supernatants and cells were frozen at ?70°C, thawed
3 times to release intracellular virus, and diluted 4:1 in 10% skim milk. Titration
of Ad5 was performed with an overlay of 0.44% agarose in medium. After 3 days,
overlay was added to feed cells, and titers were read at 5 days, with fixation and
staining performed as described for HCMV.
VV yield reduction and determination of ?-galactosidase activity. HF or Vero
cells were infected with the LacZ-recombinant VV at an MOI of 0.1 for 1 h at
37°C, after which the inoculum was removed and replaced with complete me-
dium containing serial dilutions of drug or solvent control for 18 h at 37°C.
Supernatants were discarded and replaced with 0.44 mM 4-methylumbelliferyl-
?-D-galactopyranoside (MUG; Sigma-Aldrich) in order to measure ?-galactosi-
dase activity in infected cells as previously described (15).
Determination of effective drug concentrations. The results of each replicate
treatment from yield reduction experiments were converted to percent inhibition
compared to negative controls. The mean percent inhibition of each set of
replicates was plotted against the concentration of drug using Excel version 11.5
(Microsoft Corp.). The 50% and 90% effective concentrations (EC50and EC90;
the drug concentrations required to inhibit viral replication by 50% and 90%,
respectively) were determined graphically, by identifying the concentrations at
which a line intercepted the curves at the relevant percent inhibition (28, 39).
Recombinant KSHV expressing SeAP. To facilitate the
quantitation of infectious KSHV, we developed a recombinant
virus, rKSHV.294, that expresses the SeAP (4) from a Tet-
responsive element (TRE) promoter under the control of the
Tet-controlled transactivator (tTA). rKSHV.294 was con-
structed from rKSHV.219 (52) by replacing the RFP/GFP/
Puro insert with an SeAP/GFP/Neo cassette (Fig. 1A). DNA
hybridization analysis of restriction enzyme digestions of
rKSHV.294 and JSC-1 genomic DNA resulted in the predicted
pattern of DNA fragments for rKSHV.294, demonstrating that
the insert is in the correct position (Fig. 1B). The structure of
rKSHV.294 was also examined by PCR analysis and resulted in
amplicons of the predicted sizes (Fig. 1A and C). By inducing
lytic replication with BacK50 and sodium butyrate, 105to 106
infectious units of rKSHV.294 per ml was produced from Vero
cells, similarly to rKSHV.219 (52). In additional functional
testing, we have also found that rKSHV.294 can infect primary
keratinocytes, establish latency, and produce infectious virus
during keratinocyte differentiation (J. Kim and J. Vieira, un-
published data). Preparations of rKSHV.294 produced from
cells lacking tTA contain essentially no SeAP activity. SeAP
expression from rKSHV.294 was greatly increased upon infec-
tion of 293-tTA cells compared with that after infection of 293
cells that do not express tTA (Fig. 1D). Because rKSHV.294
also expresses the GFP, we were able to confirm that the
number of KSHV-infected cells determined by counting GFP-
positive cells is directly correlated with measurement of SeAP
activity in the same cultures (Fig. 1E).
Nelfinavir inhibits KSHV replication in vitro. A broad panel
of ARVs, including nucleoside and nonnucleoside reverse
transcriptase inhibitors and PIs, was tested for inhibitory ac-
tivity using the SeAP-recombinant KSHV, rKSHV.294, in
Vero cells (Fig. 2). Ganciclovir, which has been reported to
inhibit KSHV replication in vitro (28, 35, 39), was used as a
positive control. The EC50of ganciclovir was approximately 27
?M (Fig. 2 and Table 1), which was somewhat greater than
that described in previous reports (0.96 to 8.9 ?M) using dif-
ferent methods (28, 35, 39, 58). Nelfinavir was found to have an
EC50of 7.4 ?M (standard deviation, ?0.7) by the rKSHV.294
assay (Fig. 2 and Table 1), showing it to be 3.5 times more
potent than ganciclovir. No cytotoxicity due to nelfinavir could
be detected using nelfinavir concentrations of ?20 ?M (data
not shown), as has been reported elsewhere (13, 25). Although
zidovudine, stavudine, lopinavir, and ritonavir displayed mini-
mal reductions (10 to 20%) in KSHV replication at their max-
imum noncytotoxic concentrations, no ARVs tested other than
nelfinavir showed potent or reproducible activity against
KSHV (Fig. 2).
The effect of nelfinavir on KSHV replication was further
evaluated using virus produced by a primary effusion lym-
phoma cell line to infect TIME cells, which are derived from
human vascular endothelium and support continuous passage
of KSHV (29). Using this assay, nelfinavir again showed strong
activity in the low-micromolar range (EC50? 2.0 ?M) (Fig. 3
and Table 1), confirming the effect of nelfinavir on KSHV
replication in another model system. Ganciclovir was also
slightly more active in this assay, with an EC50of approxi-
mately 3.1 ?M (Fig. 3).
Nelfinavir inhibits replication of HCMV and HSV-1 but not
Ad5 or VV. In order to determine whether nelfinavir’s inhibi-
tion of KSHV is virus specific, we tested nelfinavir for activity
against the related viruses HMCV and HSV-1, as well as two
unrelated DNA viruses, VV and Ad5. As measured by the
yield of PFU, nelfinavir inhibited replication of HCMV in HF,
and of HSV-1 in HF and Vero cells, with an EC50in the
low-micromolar range for each experiment (Table 1). In con-
trast, nelfinavir did not show any effect on Ad5 replication in
A549 cells as detected by the same plaque-based methods at
concentrations up to 10 ?M (Table 1). Additionally, we did not
observe any activity against VV in HF or Vero cells at 10 ?M
using a LacZ recombinant virus assay, further suggesting a
specific mechanism of action for nelfinavir on HHV replication
in these cell types.
Previously, there have not been methods conducive to high-
throughput quantitation of infectious KSHV. The assay de-
scribed here using rKSHV.294 entails the measurement of
infectious virus in a 96-well plate format by determination of
2698 GANTT ET AL.ANTIMICROB. AGENTS CHEMOTHER.
SeAP activity in media collected from infected cultures. This
recombinant virus allows for a sensitive quantitative assay with
low background, which is compatible with colorimetric, lumi-
nescent, and fluorescent substrates (4, 19, 55). In rKSHV.294,
the SeAP gene is expressed using a tetracycline-regulated sys-
tem (23) with the gene under the control of the TRE-Tight
promoter (Clontech, Mountain View, CA), which is activated
by the tTA protein. In cells lacking tTA, there is minimal SeAP
expression from rKSHV.294, but upon the infection of cells
containing the tTA, there is strong SeAP expression. The tet-
racycline-regulated system was used so that rKSHV.294 could
be produced from cells without the tTA protein, and therefore,
no SeAP that would require removal by virus purification, or
extensive washing of cells infected with the virus, before infec-
tion assays could be done is produced. It is straightforward to
generate tTA-expressing cells to use as targets in infection
studies. In addition to the Vero and 293 cells expressing tTA,
we have generated HaCaT, DU145, and other cell types for
various studies. This system can have utility in drug studies, the
examination of cellular and viral components involved in virus
production, the study of neutralizing antibodies to KSHV, and
the process of KSHV infection. A SeAP-based method
has previously been used with recombinant rhesus monkey
rhadinovirus, another gammaherpesvirus, in antibody neutrali-
zation studies (5). Here we demonstrated the suitability of
rKSHV.294 for screening drugs for antiviral activity against
We found that of a large panel of ARVs tested using the
rKSHV.294 assay, only nelfinavir strongly inhibited KSHV rep-
lication. Inhibitory activity was greater using the TIME cell
assay than with the SeAP assay for both nelfinavir and ganci-
clovir, suggesting that it may be useful to compare or validate
results by multiple systems, as was done here. Of note, nelfi-
navir was reported by Sgadari et al. (48) not to inhibit KSHV
replication, although unlike in this study, production of infec-
tious virus was not directly measured. Nelfinavir also inhibited
replication of HSV and HCMV and thus displayed broad ac-
tivity against alpha-, beta-, and gammaherpesviruses. Replica-
tion of unrelated DNA viruses was not affected by nelfinavir,
suggesting that this drug targets a viral or host cell function
FIG. 1. Construction and quantitation of rKSHV.294. (A) Schematic diagram of rKSHV.294 showing the insertion site in the KSHV genome;
the BamHI sites flanking the 4.8-kb segment of the KSHV genome used are indicated. The relative positions of the SeAP, the GFP, and the Neo
elements are shown with their respective promoters but not to scale. Beneath the 294 construct are shown the expected PCR products (a to e) with
the primers used for analysis of viral DNA. (B) Hybridization analysis of rKSHV.294 and JSC-1 viral DNA. (Left) KSHV probe using the 4.8-kb
BamHI fragment used to construct the virus. (Right) Neo probe. Lane 1, rKSHV.294 ? AflII, predicted fragments of 15.9, 5.7, and 4.8 kb. The
4.8-kb fragment contains the SeAP/GFP/Neo insert with only 170 bp of KSHV DNA, which accounts for the weak band. Lane 2, JSC-1 ? AflII,
predicted fragments of 15.9 and 5.8 kb. Lane 3, rKSHV.294 ? SspI, predicted fragments of 8 and 6.2 kb. Lane 4, JSC-1 ? SspI, predicted fragments
of 6.2 and 3.3 kb. Lane 5, rKSHV.294 ? AflII, predicted fragment of 4.8 kb. Lane 6, JSC-1 ? AflII. Lane 7, rKSHV.294 DNA ? SspI, predicted
fragment of 8 kb. Lane 8, JSC-1 ? SspI. DNA markers in kb are on the left side. (C) Ethidium bromide-stained agarose gel following
electrophoresis of the PCR products resulting with rKSHV.294 viral DNA with the indicated primers described in Materials and Methods. Lane
a, ORF56f (outside BamHI site) and SeAPr; lane b, ORF57u and SeAPr; lane c, SeAPf and GFPf; lane d, GFPr and Neor; lane e, Neof and K9.
(D) rKSHV.294 was tested for tTA control of SeAP expression by the infection of 293 cells or of 293 cells expressing tTA at an MOI of ?0.01.
Two days postinfection, 0.1 ml of cell-free medium was assayed for SeAP. (E) Graph comparing GFP-positive cells and SeAP assay fluorescence
resulting from infection of 293-tTA cells with 2-fold dilutions of rKSHV.294 with GFP-positive cells and SeAP determination, done at 2 days
VOL. 55, 2011 NELFINAVIR INHIBITS KSHV REPLICATION2699
specific to HHVs. Given that nelfinavir, like other PIs, acts on
the HIV aspartyl protease, it is unlikely that nelfinavir acts by
inhibition of the HHV protease, as HHVs have not been
shown to encode an aspartyl protease and the HHV serine
protease that is required for capsid maturation has no appar-
ent homology to the HIV protease or other aspartyl proteases
(49). Furthermore, none of the other members of the PI class
that we evaluated reproducibly inhibited KSHV replication
despite their potent activity against the HIV aspartyl protease.
Interestingly, however, nelfinavir and other PIs were predicted
FIG. 2. Nelfinavir but not other ARVs potently inhibits replication of recombinant KSHV in vitro using a high-throughput assay. ARVs were
screened for activity against KSHV replication in Vero cells using a virus that encodes a SeAP reporter and constitutively expresses GFP. Infected
cells were induced to undergo lytic replication and treated with drug or solvent controls in triplicate for 72 h, after which medium was harvested.
Production of infectious virus was quantified by infecting new cells and determining the SeAP activity in the medium after 48 h. Panel A shows
the mean percent inhibition of viral replication compared with negative controls (which correspond to 0% inhibition), based on the results of 3
independent experiments. Error bars show the standard error for each treatment. GCV, ganciclovir; NFV, nelfinavir; LPV, lopinavir; RTV,
ritonavir; SQV, saquinavir; APV, amprenavir; IDV, indinavir; ATV, atazanavir; AZT, zidovudine; d4T, stavudine; 3TC, lamivudine; ddI,
didanosine; ABC, abacavir; TDF, tenofovir; NVP, nevirapine; EFV, efavirenz. The maximum noncytotoxic concentration of each ARV is shown,
as well as lower concentrations for nelfinavir and ganciclovir, which also showed inhibitory activity. The quantity of infectious virus produced was
also evaluated by fluorescence microscopy; representative examples are shown in panel B. The top row shows infected cells expressing GFP; the
bottom row shows the cells in the same field by light microscopy.
2700 GANTT ET AL.ANTIMICROB. AGENTS CHEMOTHER.
to bind to the HCMV protease based on computational struc-
tural modeling studies (26). While it is possible that nelfinavir
acts on a novel viral target, we speculate, because of its mul-
tiple cellular effects, that nelfinavir inhibits HHV replication
through modulation of a necessary host cell function.
PIs have numerous secondary effects on human cells that
differ markedly between members of the class and which may
be dose and cell type dependent (reviewed in reference 17).
Among PIs, nelfinavir appears to be the most potent inhibitor
of signaling through the phosphatidylinositol 3-kinase (PI3K)/
Akt and signal transducer and activator of transcription 3
(STAT3) pathways, which mediate cardinal cellular functions,
including glucose metabolism, protein synthesis, proliferation,
and survival (22, 56, 57). Due to these and other cellular
effects, nelfinavir may be the PI that holds the greatest promise
for use as an antitumor drug, and it is currently being evaluated
against several solid tumors in clinical trials (17, 22). Nelfinavir
was active against HHVs in vitro at concentrations similar to
those achieved in plasma with oral dosing used for treatment of
HIV infection (area under the concentration-time curve, ?20
?M ? h?1) (3, 33), which suggests that nelfinavir may be useful
for treatment of HHV infections at standard doses. Of note,
the maximum tolerated dose of nelfinavir was not determined
during initial development, but it may be established as a part
of ongoing cancer treatment trials (22).
Nelfinavir has been widely used for HIV infection with an
excellent safety profile and could be rapidly and economically
repositioned for the prevention and treatment of HHV-related
diseases. Nelfinavir may have particular benefit for use in pa-
tients with the AIDS-defining malignancy KS, and perhaps
Epstein-Barr virus (EBV)-associated non-Hodgkin lympho-
mas, given its potential inhibitory activity against HIV, HHVs,
and tumor cells. In addition, nelfinavir might reduce viral shed-
ding and ulcers in HIV-infected persons with genital HSV,
potentially reducing transmission of HIV (18). Nelfinavir or
analogs also represent a novel therapeutic option for HHV
infections in HIV-negative patients, such as HCMV infection
TABLE 1. Nelfinavir inhibits replication of human herpesviruses but not other DNA viruses in vitro
Virus Cell typeDrug
Effective concn (?M)a
HSV-1 (strain F)
HSV-1 (strain F)
Vaccinia virus (LacZ-VC2)
Vaccinia virus (LacZ-VC2)
7.4 ? 0.7
27 ? 4.5
2.0 ? 0.8
3.1 ? 0.3
4.4 ? 0.6
5.3 ? 1.5
5.4 ? 1.1
4.7 ? 1.3
10.7 ? 0.5
7.8 ? 1.1
8.7 ? 1.9
8.5 ? 1.5
aMean values ? standard deviations are derived from ?2 independent experiments, with each treatment performed in duplicate or triplicate.
bConcentrations greater than those indicated exhibited cytotoxic effects and therefore were not tested.
FIG. 3. Nelfinavir inhibits replication of KSHV in a human endo-
thelial cell line. KSHV obtained from BCBL-1 cells was used to infect
TIME cells and induced to undergo lytic replication in the presence of
drug or solvent controls. After 48 h, medium was harvested and used
to infect new TIME cells. Cell nuclei were then stained using DAPI,
and infected cells were fluorescently labeled with antibody against
LANA. The production of infectious virus was quantified by counting
the proportion of cells in each treatment demonstrating the specific
punctate nuclear staining pattern of LANA expression. Shown are the
mean results of 3 experiments, in which infected cells were treated with
nelfinavir (NFV; triangles), ganciclovir (GCV; squares), or solvent
controls. The percent inhibition is shown compared to solvent controls.
Error bars show the standard errors.
VOL. 55, 2011NELFINAVIR INHIBITS KSHV REPLICATION2701
in solid organ and stem cell transplant recipients, for which less
toxic drugs and drugs active against ganciclovir-resistant infec-
tions are urgently needed (discussed in references 6 and 31).
Further studies are required to evaluate the efficacy of nelfi-
navir for treating HHV-related diseases and to determine its
mechanism of action for the development of novel antiherpes
We thank Stephanie Child and Frew Meshesha for technical assis-
tance; Tim Rose, Keith Jerome, and Dusty Miller for sharing reagents;
and Lisa Frenkel and Anna Wald for insightful discussions.
This work was supported by the National Institutes of Health grants
KL2RR025015 (S.G.), P30AI027757 (S.G. and M.L.), UL1RR025014
(S.G.), AI26672 (A.P.G.), R01CA138165 (C.C.), R01CA111204 (J.V.),
and R01DE016809 (J.V.).
No authors have conflicts of interest to declare.
1. Antman, K., and Y. Chang. 2000. Kaposi’s sarcoma. N. Engl. J. Med. 342:
2. Bani-Sadr, F., S. Fournier, and J. M. Molina. 2003. Relapse of Kaposi’s
sarcoma in HIV-infected patients switching from a protease inhibitor to a
non-nucleoside reverse transcriptase inhibitor-based highly active antiretro-
viral therapy regimen. AIDS 17:1580–1581.
3. Bardsley-Elliot, A., and G. L. Plosker. 2000. Nelfinavir: an update on its use
in HIV infection. Drugs 59:581–620.
4. Berger, J., J. Hauber, R. Hauber, R. Geiger, and B. R. Cullen. 1988. Secreted
placental alkaline phosphatase: a powerful new quantitative indicator of
gene expression in eukaryotic cells. Gene 66:1–10.
5. Bilello, J. P., J. S. Morgan, B. Damania, S. M. Lang, and R. C. Desrosiers.
2006. A genetic system for rhesus monkey rhadinovirus: use of recombinant
virus to quantitate antibody-mediated neutralization. J. Virol. 80:1549–1562.
6. Boeckh, M., and P. Ljungman. 2009. How we treat cytomegalovirus in
hematopoietic cell transplant recipients. Blood 113:5711–5719.
7. Carroll, P. A., E. Brazeau, and M. Lagunoff. 2004. Kaposi’s sarcoma-asso-
ciated herpesvirus infection of blood endothelial cells induces lymphatic
differentiation. Virology 328:7–18.
8. Casper, C., et al. 2008. Valganciclovir for suppression of human herpesvi-
rus-8 replication: a randomized, double-blind, placebo-controlled, crossover
trial. J. Infect. Dis. 198:23–30.
9. Casper, C., W. G. Nichols, M. L. Huang, L. Corey, and A. Wald. 2004.
Remission of HHV-8 and HIV-associated multicentric Castleman disease
with ganciclovir treatment. Blood 103:1632–1634.
10. Casper, C., et al. 2004. HIV infection and human herpesvirus-8 oral shed-
ding among men who have sex with men. J. Acquir. Immune Defic. Syndr.
11. Cattelan, A. M., et al. 2001. Acquired immunodeficiency syndrome-related
Kaposi’s sarcoma regression after highly active antiretroviral therapy: bio-
logic correlates of clinical outcome. J. Natl. Cancer Inst. Monogr. 28:44–49.
12. Chakrabarti, S., K. Brechling, and B. Moss. 1985. Vaccinia virus expression
vector: coexpression of beta-galactosidase provides visual screening of re-
combinant virus plaques. Mol. Cell. Biol. 5:3403–3409.
13. Chandler, B., et al. 2003. The effects of protease inhibitors and nonnucleo-
side reverse transcriptase inhibitors on p-glycoprotein expression in periph-
eral blood mononuclear cells in vitro. J. Acquir. Immune Defic. Syndr.
14. Child, S. J., M. Hakki, K. L. De Niro, and A. P. Geballe. 2004. Evasion of
cellular antiviral responses by human cytomegalovirus TRS1 and IRS1. J. Vi-
15. Child, S. J., L. K. Hanson, C. E. Brown, D. M. Janzen, and A. P. Geballe.
2006. Double-stranded RNA binding by a heterodimeric complex of murine
cytomegalovirus m142 and m143 proteins. J. Virol. 80:10173–10180.
16. Chow, W. A., S. Guo, and F. Valdes-Albini. 2006. Nelfinavir induces liposar-
coma apoptosis and cell cycle arrest by upregulating sterol regulatory ele-
ment binding protein-1. Anticancer Drugs 17:891–903.
17. Chow, W. A., C. Jiang, and M. Guan. 2009. Anti-HIV drugs for cancer
therapeutics: back to the future? Lancet Oncol. 10:61–71.
18. Corey, L., A. Wald, C. L. Celum, and T. C. Quinn. 2004. The effects of herpes
simplex virus-2 on HIV-1 acquisition and transmission: a review of two
overlapping epidemics. J. Acquir. Immune Defic. Syndr. 35:435–445.
19. Cullen, B. R. 2000. Utility of the secreted placental alkaline phosphatase
reporter enzyme. Methods Enzymol. 326:159–164.
20. Engels, E. A., et al. 2003. Detection and quantification of Kaposi’s sarcoma-
associated herpesvirus to predict AIDS-associated Kaposi’s sarcoma. AIDS
21. Gill, J., et al. 2002. Prospective study of the effects of antiretroviral therapy
on Kaposi sarcoma-associated herpesvirus infection in patients with and
without Kaposi sarcoma. J. Acquir. Immune Defic. Syndr. 31:384–390.
22. Gills, J. J., et al. 2007. Nelfinavir, a lead HIV protease inhibitor, is a
broad-spectrum, anticancer agent that induces endoplasmic reticulum stress,
autophagy, and apoptosis in vitro and in vivo. Clin. Cancer Res. 13:5183–
23. Gossen, M., and H. Bujard. 1992. Tight control of gene expression in mam-
malian cells by tetracycline-responsive promoters. Proc. Natl. Acad. Sci.
U. S. A. 89:5547–5551.
24. Grundhoff, A., and D. Ganem. 2004. Inefficient establishment of KSHV
latency suggests an additional role for continued lytic replication in Kaposi
sarcoma pathogenesis. J. Clin. Invest. 113:124–136.
25. Gupta, A. K., G. J. Cerniglia, R. Mick, W. G. McKenna, and R. J. Muschel.
2005. HIV protease inhibitors block Akt signaling and radiosensitize tumor
cells both in vitro and in vivo. Cancer Res. 65:8256–8265.
26. Jenwitheesuk, E., and R. Samudrala. 2005. Virtual screening of HIV-1
protease inhibitors against human cytomegalovirus protease using docking
and molecular dynamics. AIDS 19:529–531.
27. Joffe, M. M., et al. 1997. Effect of treatment with zidovudine on subsequent
incidence of Kaposi’s sarcoma. Clin. Infect. Dis. 25:1125–1133.
28. Kedes, D. H., and D. Ganem. 1997. Sensitivity of Kaposi’s sarcoma-associ-
ated herpesvirus replication to antiviral drugs. Implications for potential
therapy. J. Clin. Invest. 99:2082–2086.
29. Lagunoff, M., et al. 2002. De novo infection and serial transmission of
Kaposi’s sarcoma-associated herpesvirus in cultured endothelial cells. J. Vi-
30. Leitch, H., M. Trudeau, and J. P. Routy. 2003. Effect of protease inhibitor-
based highly active antiretroviral therapy on survival in HIV-associated ad-
vanced Kaposi’s sarcoma patients treated with chemotherapy. HIV Clin.
31. Limaye, A. P. 2002. Ganciclovir-resistant cytomegalovirus in organ trans-
plant recipients. Clin. Infect. Dis. 35:866–872.
32. Lock, M. J., N. Thorley, J. Teo, and V. C. Emery. 2002. Azidodeoxythymidine
and didehydrodeoxythymidine as inhibitors and substrates of the human
herpesvirus 8 thymidine kinase. J. Antimicrob. Chemother. 49:359–366.
33. Markowitz, M., et al. 1998. A preliminary evaluation of nelfinavir mesylate,
an inhibitor of human immunodeficiency virus (HIV)-1 protease, to treat
HIV infection. J. Infect. Dis. 177:1533–1540.
34. Martin, D. F., et al. 1999. Oral ganciclovir for patients with cytomegalovirus
retinitis treated with a ganciclovir implant. Roche Ganciclovir Study Group.
N. Engl. J. Med. 340:1063–1070.
35. Medveczky, M. M., E. Horvath, T. Lund, and P. G. Medveczky. 1997. In vitro
antiviral drug sensitivity of the Kaposi’s sarcoma-associated herpesvirus.
36. Mocroft, A., et al. 1996. Anti-herpesvirus treatment and risk of Kaposi’s
sarcoma in HIV infection. Royal Free/Chelsea and Westminster Hospitals
Collaborative Group. AIDS 10:1101–1105.
37. Morris, V. A., A. S. Punjabi, and M. Lagunoff. 2008. Activation of Akt
through gp130 receptor signaling is required for Kaposi’s sarcoma-associated
herpesvirus-induced lymphatic reprogramming of endothelial cells. J. Virol.
38. Nagy, S., R. Gyulai, L. Kemeny, P. Szenohradszky, and A. Dobozy. 2000.
Iatrogenic Kaposi’s sarcoma: HHV8 positivity persists but the tumors regress
almost completely without immunosuppressive therapy. Transplantation 69:
39. Neyts, J., and E. De Clercq. 1997. Antiviral drug susceptibility of human
herpesvirus 8. Antimicrob. Agents Chemother. 41:2754–2756.
40. Nguyen, H. Q., et al. 2008. Persistent Kaposi sarcoma in the era of highly
active antiretroviral therapy: characterizing the predictors of clinical re-
sponse. AIDS 22:937–945.
41. Paparizos, V. A., K. P. Kyriakis, V. Papastamopoulos, M. Hadjivassiliou,
and N. G. Stavrianeas. 2002. Response of AIDS-associated Kaposi sarcoma
to highly active antiretroviral therapy alone. J. Acquir. Immune Defic. Syndr.
42. Parkin, D. M., S. L. Whelan, J. Ferlay, L. Teppo, and D. B. Thomas (ed.).
2002. Cancer incidence in five continents, vol. VIII. IARC scientific publi-
cation no. 155. International Agency for Research on Cancer, Lyon, France.
43. Parkin, D. M., H. Wabinga, S. Nambooze, and F. Wabwire-Mangen. 1999.
AIDS-related cancers in Africa: maturation of the epidemic in Uganda.
44. Pore, N., A. K. Gupta, G. J. Cerniglia, and A. Maity. 2006. HIV protease
inhibitors decrease VEGF/HIF-1alpha expression and angiogenesis in glio-
blastoma cells. Neoplasia 8:889–895.
45. Rey, D., et al. 2001. Efavirenz as a substitute for protease inhibitors in
HIV-1-infected patients with undetectable plasma viral load on HAART: a
median follow-up of 64 weeks. J. Acquir. Immune Defic. Syndr. 27:459–462.
46. Sambrook, J., and D. W. Russell. 2001. Molecular cloning: a laboratory
manual, 3rd ed. Cold Spring Harbor Laboratory Press, Cold Spring Har-
47. Sgadari, C., et al. 2002. HIV protease inhibitors are potent anti-angiogenic
molecules and promote regression of Kaposi sarcoma. Nat. Med. 8:225–232.
48. Sgadari, C., P. Monini, G. Barillari, and B. Ensoli. 2003. Use of HIV
2702GANTT ET AL.ANTIMICROB. AGENTS CHEMOTHER.
protease inhibitors to block Kaposi’s sarcoma and tumour growth. Lancet Download full-text
49. Unal, A., et al. 1997. The protease and the assembly protein of Kaposi’s
sarcoma-associated herpesvirus (human herpesvirus 8). J. Virol. 71:7030–
50. Venetsanakos, E., et al. 2002. Induction of tubulogenesis in telomerase-
immortalized human microvascular endothelial cells by glioblastoma cells.
Exp. Cell Res. 273:21–33.
51. Vieira, J., P. O’Hearn, L. Kimball, B. Chandran, and L. Corey. 2001. Acti-
vation of Kaposi’s sarcoma-associated herpesvirus (human herpesvirus 8)
lytic replication by human cytomegalovirus. J. Virol. 75:1378–1386.
52. Vieira, J., and P. M. O’Hearn. 2004. Use of the red fluorescent protein as a
marker of Kaposi’s sarcoma-associated herpesvirus lytic gene expression.
53. Whitby, D., et al. 1995. Detection of Kaposi sarcoma associated herpesvirus
in peripheral blood of HIV-infected individuals and progression to Kaposi’s
sarcoma. Lancet 346:799–802.
54. Wijnveen, A. C., H. Persson, S. Bjorck, and I. Blohme. 1987. Disseminated
Kaposi’s sarcoma—full regression after withdrawal of immunosuppressive
therapy: report of a case. Transplant. Proc. 19:3735–3736.
55. Yang, T. T., P. Sinai, P. A. Kitts, and S. R. Kain. 1997. Quantification of gene
expression with a secreted alkaline phosphatase reporter system. Biotech-
56. Yang, Y., et al. 2006. NFV, an HIV-1 protease inhibitor, induces growth
arrest, reduced Akt signalling, apoptosis and docetaxel sensitisation in
NSCLC cell lines. Br. J. Cancer 95:1653–1662.
57. Yang, Y., et al. 2005. HIV-1 protease inhibitor induces growth arrest and
apoptosis of human prostate cancer LNCaP cells in vitro and in vivo in
conjunction with blockade of androgen receptor STAT3 and AKT signaling.
Cancer Sci. 96:425–433.
58. Zhu, W., et al. 2005. Potent antiviral activity of north-methanocarbathymi-
dine against Kaposi’s sarcoma-associated herpesvirus. Antimicrob. Agents
VOL. 55, 2011 NELFINAVIR INHIBITS KSHV REPLICATION2703