A rapid quantitative PCR-based assay for testing antiviral agents against human adenoviruses demonstrates type specific differences in ribavirin activity.

Antiviral Research 72 (2006) 34–41
A rapid quantitative PCR-based assay fo
against human adenoviruses demon
differences in ribavirin
adi
novir
20 M
Abstract
Human ad fatal
patients. De easy
interpret and g the
HAdV DNA e ass
of ribavirin infec
Several H fecti
of species C 99 11
for example HAdV-31 (EC50 56�M, EC99 > 500�M). Differential ribavirin sensitivity of HAdV types may contribute to the variable outcome of
ribavirin therapy. Rapid screening of antiviral agents with the rapid qPCR-based assay against a multitude of HAdV serotypes may also facilitate
development of future antiviral agents.
© 2006 Elsevier B.V. All rights reserved.
Keywords: A
1. Introdu
The six
Mastadeno
associated
(Swenson e
ity and mo
et al., 2001
patients HA
example, n
with symp
organs (pn
atitis, ence
2002; La R
∗ Correspon
Hochschule H
+49 511 5324
E-mail ad
1 Present a
Service, Kobl
0166-3542/$
doi:10.1016/jdenovirus; Ribavirin; Phenotypic assay; Quantitative PCR; Immunosuppression
ction
species (A–F) of human adenoviruses (HAdV, genus
virus, family Adenoviridae) with their 51 types are
with a variety of diseases affecting all organ systems
t al., 2003). HAdV are an important cause of morbid-
rtality in the immunocompromised host (Bordigoni
; Carrigan, 1997; Chakrabarti et al., 2002). In these
dV can lead to severe organ infections such as, for
ephritis or to life threatening disseminated disease
toms of septic shock and involvement of multiple
eumonia, gastroenteritis, hemorrhagic cystitis, hep-
phalitis) (Baldwin et al., 2000; Chakrabarti et al.,
osa et al., 2001; Lion et al., 2003). The incidence
ding author at: Institut fu¨r Virologie, OE5230, Medizinische
annover, Carl-Neuberg-Str. 1, D-30625 Hannover, Germany. Tel.:
311; fax: +49 511 5328736.
dress: ahei@virologie.mh-hannover.de (A. Heim).
ddress: Central Institute of the Federal Armed Forces Medical
enz, Germany.
of severe HAdV diseases is increasing with the growing num-
ber of immunocompromised patients. Children after allogeneic
bone marrow transplantation are at high risk for disseminated
disease with fatality rates as high as 50–80% (Bordigoni et al.,
2001; Gavin and Katz, 2002; Seidemann et al., 2004; Venard
et al., 2000). At present, there is no effective antiviral therapy
available which could be considered as the gold standard to treat
HAdV diseases. Several antiviral drugs have been used clinically
with varying clinical outcome (Bordigoni et al., 2001; Gavin and
Katz, 2002; Kojaoghlanian et al., 2003; La Rosa et al., 2001;
Seidemann et al., 2004), but controlled clinical studies have not
yet been performed due to multiple organizational problems.
Even determining the in vitro sensitivity of HAdV to antivi-
ral agents is not an easy task because HAdV CPE reduction
assays are difficult to interpret and may take more than 1 week
with some slowly replicating HAdV serotypes (Mentel et al.,
1997; Naesens et al., 2005). Therefore, we developed a phe-
notypic assay for testing the antiviral activity during the first
round of replication using HAdV DNA content as an objective
readout. As a prerequisite for this assay, the HAdV DNA repli-
cation kinetics of multiple HAdV serotypes was determined by
– see front matter © 2006 Elsevier B.V. All rights reserved.
.antiviral.2006.03.009Ru¨diger Stock 1, Gabi Harste, Ijad M
Institute for Virology, German National Reference Laboratory for Ade
Received 20 January 2006; accepted
enovirus (HAdV) infections are increasingly frequent and potentially
termining the in vitro sensitivity of HAdV to antiviral agents is not an
may take more than 1 week. We developed a phenotypic assay for testin
concentration as an objective readout within 30 h. After evaluating th
against different HAdV serotypes because clinical response of HAdV
AdV prototypes (1, 2, 5, 11, 31, 34, 48) associated with disseminated in
were more sensitive to ribavirin (HAdV-2 and -5: EC50 < 10�M, ECr testing antiviral agents
strates type specific
activity
sch, Albert Heim ∗
uses, Hannover Medical School, Germany
arch 2006
as a disseminated disease in highly immunocompromised
task because HAdV CPE reduction assays are difficult to
antiviral activity during the first round of replication using
ay with cidofovir, we focused on determining the antiviral
tions towards ribavirin treatment varied considerably.
ons and clinical isolates were tested. Predominating HAdV
1 and 104�M, respectively) than HAdV of other species,

R. Stock et al. / Antiviral Research 72 (2006) 34–41 35
a highly specific, quantitative PCR protocol (qPCR) with a Taq-
Man probe capable to detect all 51 HAdV types (Heim et al.,
2003). Fro
promise to
which inte
against any
which do n
cidofovir a
and various
In contr
avirin has
viruses (Sid
out any mo
an “atypica
tus” for the
case report
pressed pat
of ribavirin
Emovon et
2000), but
(Bordigoni
serotypes m
51 HAdV s
al., 2003)
bilities of
2005). For
are strongl
mised host
-C5), HAd
(Seideman
A31 was a
compromis
2004; Vena
species D (
cient, termi
Therefore,
cally signifi
Signific
Drug sensi
cient patien
of the same
2. Materia
2.1. Cell c
Cell cu
and HeLa
ised condi
used betw
trypsin/ED
medium (D
10 U penici
antiviral as
seeded on s
day before
application
2.2. Antiviral agents
aviri
was
. Th
olut
ovid
disso
for
irus
an
5, B
tion,
osup
HAd
nce
cells
as h
d tha
eared
for
ined
h (H
by
uant
tota
by a
dete
his
valid
cen
irus
flue
in ca
licity
y, ce
m. A
ere
tract
, Ge
(Hei
apid
rder
eLa
as
um w
resh
d 50
espem this data, an assay was established which holds
determine the antiviral activity of antiviral agents,
ract with the early steps of the replication cycle,
HAdV serotype or clinical isolate including these
ot exhibit a clear CPE. This assay was evaluated with
nd HAdV-C1, and more extensively with ribavirin
HAdV serotypes.
ast to other nucleoside analogue antiviral drugs, rib-
a wide-spectrum activity against DNA and RNA
well et al., 1972). Due to its chemical structure with-
dification of the ribose moiety, it is frequently called
l” nucleoside analogue. It holds an “orphan drug sta-
treatment of HAdV infection (Blasi, 2003). Several
s, and small, non-controlled studies in immunosup-
ients with HAdV disease indicated beneficial effects
(Arav-Boger et al., 2000; Chakrabarti et al., 2002;
al., 2003; Gavin and Katz, 2002; Miyamura et al.,
treatment failures were also reported occasionally
et al., 2001; La Rosa et al., 2001). Different HAdV
ay vary widely in their sensitivity to ribavirin as the
erotypes are genetically highly diverse (Swenson et
and CPE-based assays indicated different suscepti-
HAdV reference strains to ribavirin (Morfin et al.,
tunately, only a limited number of HAdV serotypes
y associated with disease in the immunocompro-
: HAdV-1, -2, and -5 of species C (HAdV-C1, -C2,
V-B11, and the closely related HAdV-B34 and -B35
n et al., 2004; Swenson et al., 2003). Recently, HAdV-
lso described as a significant pathogen in immuno-
ed patients (Bordigoni et al., 2001; Seidemann et al.,
rd et al., 2000). Previously, human adenoviruses of
e.g. HAdV-D48) were found in highly immunodefi-
nally ill AIDS patients (Schnurr and Dondero, 1993).
we focused on evaluating our assay with these clini-
cant HAdV serotypes.
ant differences in ribavirin sensitivity were observed.
tivity of clinical HAdV isolates from immunodefi-
ts were found to be similar to HAdV prototype strains
serotype.
ls and methods
ulture
lture experiments were performed with A549
cells, which were propagated under standard-
tions (37 ◦C, 5% carbondioxide). The cells were
een the 20th and 60th passage and harvested by
TA. Dulbecco’s modification of minimal essential
MEM) supplemented with 10% fetal bovine serum,
llin G/ml, and 10�g streptomycin/ml was used. For
says and HAdV DNA replication kinetics, cells were
ix-well plates at 106 cells/well in quadruplicates one
each experiment (infection with HAdV and ribavirin
).
Rib
mide)
many)
stock s
was pr
It was
−20 ◦C
2.3. V
Hum
C2, C
Collec
immun
eases (
Refere
HeLa
Virus w
ing an
was cl
−20 ◦C
determ
Muenc
well as
2.4. Q
The
sured
probe
gene. T
sively
the con
2.5. V
Con
A549
multip
quentl
mediu
wells w
was ex
Hilden
qPCR
2.6. R
In o
cells (H
HAdV
inocul
PBS. F
200, an
fovir, rn (1-�-d-ribofuranosyl-1,2,4-triazole-3-carboxa-
provided by ICN Pharmaceutics (Frankfurt, Ger-
e dry substance was dissolved as 10 mM aqueous
ion and stored at −20 ◦C for further use. Cidofovir
ed by Pharmacia & Upjohn (Kalamazoo, MI, USA).
lved as 10 mM aequous stock solution and stored at
further use.
stock solutions
adenovirus (HAdV) prototype viruses of types C1,
11, A31, B34, and D48 (American Type Culture
Manassas, VA) and adenoviruses isolated from
pressed patients suffering from disseminated dis-
V types C1 K, C2 J, and A31 W; German Adenovirus
Centre) were propagated on A549 cells (as well in
in case of HAdV-A31) in 75 cm2 cell culture flasks.
arvested at >95% cytopathic effect (CPE) by freez-
wing the cell culture flasks tree times. Supernatant
by centrifugation (3000 × g for 5 min) and stored at
further use. The concentration of HAdV stocks was
both by CCID50 method as proposed by Reed and
ierholzer and Killington, 1996) using A549 cells as
quantitative PCR (Heim et al., 2003).
itative HAdV-PCR
l intracellular HAdV DNA concentration was mea-
quantitative real-time PCR (qPCR) with a Taqman
cting a conserved region at the 5′ end of the hexon
qPCR is generic for all HAdV types and was exten-
ated for virus load diagnostics (interassay S.D. for
tration in the range of 12–16%) (Heim et al., 2003).
DNA-replication kinetics
nt monolayers of A549 cells in 16 wells (HeLa and
se of HAdV-A31) were infected with HAdV at a
of infection (moi) of 10 CCID50/cell for 1 h. Subse-
lls were washed with PBS and incubated in fresh
fter 3, 12, 24, and 48 h post infection (p.i.), four
harvested with trypsin/EDTA and total cellular DNA
ed using the QiaAmp DNA Blood Mini Kit (Qiagen,
rmany). Adenovirus DNA load was measured using
m et al., 2003).
qPCR-based antiviral activity assay
to test activity of antiviral agents, 24 wells with A549
and A549 in case of HAdV-A31) were infected with
described above. After 1 h of incubation the virus
as removed and the cell layers were washed with
medium containing antiviral agents (0, 10, 50, 100,
0�M ribavirin or 0, 1, 5, 25, 100, and 500�M cido-
ctively) was added to each of four wells. The medium

36 R. Stock et al. / Antiviral Research 72 (2006) 34–41
was removed 24 h p.i. (24 and 48 h in case of HAdV-A31) and the
cell layer was harvested by trypsin/EDTA. Total cellular DNA
was extracted using the QiaAmp DNA Blood Mini Kit (Qiagen).
The concentration of adenoviral DNA load was determined by
qPCR (Heim et al., 2003).
2.7. CPE-based antiviral activity assay
A549 cells were seeded in 96-well plates at 1 × 104 cells/well
and incubated for 24 h. Subsequently medium was aspirated and
100�l fresh medium containing antiviral agents and 104 TCID50
HAdV-C1 was added. Different concentrations of antiviral
agents (cidofovir 0, 0.5, 1, 2, 4, 8, 16, and 32�M; ribavirin
0, 5, 10, 20, 40, 80, 160, and 320�M, each in 12 wells) were
used. Cell culture media were changed at day 2 p.i. with fresh
antiviral agents added to the medium. On day 5 p.i. cultures were
inspected visually for CPE in a blinded fashion.
2.8. Cell-proliferation and viability assays
Reduction of cell proliferation by ribavirin was quanti-
tatively evaluated by a XTT reduction assay (EZ4U assay,
Biomedica
1 × 104 A5
and incuba
tions of rib
1000�M,
was added
2 h at 37 ◦C
a standard
two-fold se
ibrator.
2.9. Statist
For each
of total intr
tion of the relative standard deviation. Statistical significance of
the HAdV DNA concentration increase was determined by the
Student’s t-test. Inhibitory concentration 50%, 90%, and 99%
(EC50, EC90, and EC99) of ribavirin and cidofovir were cal-
culated using the median effect plot (Chou and Talalay, 1984).
Fifty percent of cytotoxic concentrations (CC50) of ribavirin and
cidofovir were calculated by non-linear regression. Confidence
intervals of EC and CC values were calculated by the Monte
Carlo method.
3. Results
3.1. Virus DNA-replication kinetics
In order to determine the interval required for HAdV DNA
replication after infection of cells, viral DNA replication kinet-
ics for each serotype were evaluated after infecting A549 cells
at a high moi (10 CCID50/cell). This approach was chosen to
achieve an almost synchronous infection of the cells. The intra-
cellular HAdV DNA concentration was determined after virus
adsorption and internalization at 3 h p.i. as a basic value. Viral
replication kinetics of all tested HAdV prototypes and clinical
s are
AdV
t 12
f the
cas
(Ta
ted i
late
diss
linic
mila
dV-A
ay b
dV-
a ce
Table 1
HAdV DNA tratio
compared to i
e (HA
HAdV-C1 0.07)
HAdV-C1 K 0.70)
HAdV-C2 0.12)
HAdV-C2 J 0.11)
HAdV-C5 0.06)
HAdV-B11 0.07)
HAdV-A31 0.11)
.HAdV-A31W 0.1)**
HAdV-A31 (H 0.55)
HAdV-A31W 0.14)
HAdV-B34 0.08)
HAdV-D48 0.27)
Statistical sig termi
* p < 0.05.
** p < 0.01.
*** p < 0.001., Vienna, Austria) (Wutzler et al., 2002). Briefly,
49 cells/well were seeded in 96-well cell culture trays
ted for a period of 4 days with various concentra-
avirin or cidofovir (0, 15, 31, 62, 125, 250, 500, and
each in 8 wells) in DMEM with 10% FBS. EZ4U
to cell culture media and cells were incubated for
. Absorbance was read at a wavelength of 450 nm in
photometer (reference wavelength 620 nm). A serial
rial dilution of 1 × 105 A549 cells was used as a cal-
ical analyses
series of HAdV DNA replication kinetics, variability
acellular HAdV DNA was estimated by determina-
isolate
and H
cant a
Most o
only in
slower
suspec
ical iso
severe
This c
tern si
of HA
A31 m
the HA
in HeL
replication kinetics displayed an increase of intracellular HAdV DNA concen
ntracellular HAdV DNA concentration 3 h p.i.
log increase (HADV DNA) 12 h p.i. log increas
0.13 (S.D. 0.13) 3.38 (S.D.
1.57 (S.D. 0.16)*** 4.21 (S.D.
0.28 (S.D. 0.11)** 2.93 (S.D.
0.65 (S.D. 0.11)*** 2.99 (S.D.
0.17 (S.D. 0.06)** 2.89 (S.D.
0.68 (S.D. 0.07)*** 3.03 (S.D.
0.54 (S.D. 0.11)*** 0.66 (S.D.
0.58 (S.D. 0.13)** 1.37 (S.D.
eLa cells) 0.58 (S.D. 0.02)* 0.78 (S.D.
(HeLa cells) 0.38 (S.D. 0.10)*** 0.64 (S.D.
1.09 (S.D. 0.07)*** 2.94 (S.D.
0.23 (S.D. 0.29) 3.29 (S.D.
nificance of the HAdV DNA concentration increase compared to 3 h p.i. was depresented in Table 1. With exception of HAdV-C1
-D48, increase of HAdV DNA was already signifi-
h p.i. but increase of HAdV DNA was still <1 log.
HAdV DNA replicated between 12 h and 24 h p.i.;
e of HAdV-A31 DNA replication was considerably
ble 1). Because an attenuation phenotype may be
n case of the laboratory strain of HAdV-A31, a clin-
from an immunosuppressed patient suffering from a
eminated infection (HAdV-A31 W) was also tested.
al isolate also exhibited a slow DNA replication pat-
r to the results obtained with the laboratory strain
31 (Table 1). As slow DNA replication of HAdV-
e cell type-dependent, DNA replication kinetics of
A31 prototype and HAdV-A31 W were also tested
lls. Replication kinetics of both viruses were almost
n in A549 cells (except where indicated) at 12, 24, and 48 h p.i.
DV DNA) 24 h p.i. log increase (HADV DNA) 48 h p.i.
*** 4.26 (S.D. 0.04)***
*** 5.43 (S.D. 0.06)***
*** 4.66 (S.D. 0.07)***
*** 4.40 (S.D. 0.04)***
*** 4.37 (S.D. 0.02)***
*** 4.08 (S.D. 0.11)***
*** 2.23 (S.D. 0.07)***
* 2.32 (S.D. 0.33)***
3.33 (S.D. 0.07)***
*** 3.33 (S.D. 0.76)***
*** 2.90 (S.D. 0.09)***
*** 3.43 (S.D. 0.06)***
ned by the Student’s t-test.

R. Stock et al. / Antiviral Research 72 (2006) 34–41 37
identical in HeLa cells compared to A549 cells (Table 1). These
results indicate that slow DNA replication seems to be a char-
acteristic o
Clinical
evaluated i
in capabili
laboratory
laboratory
comparable
3.2. Evalu
As man
side analog
tion, we de
cation over
DNA repli
tested (wit
of more th
fore, calcu
and 99% r
was feasibl
because HA
(e.g. 1010 c
Seidemann
magnitudes
values may
a first step
established
cidofovir a
DNA conce
the EC50 fo
and EC99 w
and 18.8�
step antivir
in our qPC
replication
(Fig. 1A). R
a concentra
and EC99 v
103�M (C
(Table 2).
For com
strated acti
EC50 value
tively), as c
tification.
3.3. Antivi
of HAdV
Table 2
on DNA re
Ribavirin h
species C (
the qPCR-b
of species
nhibi
�M r
0.4�
ribav
77�
tion
, th
ing
-C1). Antiviral activity of ribavirin against HAdV-B11
ghtly lower compared to HAdV-C species, but SI was still
able 2). For HAdV-A31, it was not feasible to determine
l activity of ribavirin against HAdV prototypes and clinical isolates in
lls (except where indicated)
EC50 (�M) EC90 (�M) EC99 (�M)
rototype
-C1 <10 27 (15–54) 104 (52–231)
-C2 <10 30 (23–42) 111 (79–159)
-C5 <10 28 (11–86) 104 (35–392)
-B11 <10 38 (14–131) 200 (63–883)
-A31 56 (19–277) 192 (55–1146) nda
-A31 (HeLa) 32 (12–127) 359 (98–2987) nda
-B34 16 (7–40) 143 (54–475) >500
-D48 24 (9–84) 102 (33–448) 495 (133–2777)
isolates
-C1 K <10 41 (24–75) 229 (121–474)
-C2 J 13 (8–20) 42 (26–72) 155 (88–293)
-A31 W 17 (5–85) 196 (41–1718) nda
-A31 W (HeLa) 34 (15–90) 446 (157–1596) nda
n concentrations for 50%, 90%, and 99% inhibition of viral DNA replica-
50, EC90, and EC99). Ninety-five percent confidence intervals indicated
ets.
determined due to insufficient increase of HAdV in untreated cultures.f HAdV-A31.
isolates HAdV-C1 K and HAdV-C2 J were also
n order to determine whether there is a difference
ty of DNA replication between clinical isolates and
strains of species HAdV-C. Similar to HAdV-A31,
strains and clinical isolates of species HAdV-C had
DNA replication kinetics (Table 1).
ation of qPCR-based antiviral activity testing
y antiviral agents are nucleotide analogue or nucleo-
ue drugs which interfere with viral genome replica-
cided to determine their effects on HAdV DNA repli-
a 24 h period post infection when most of the HAdV
cation takes place (Table 1). For all HAdV types
h exception of HAdV-A31) a HAdV DNA increase
an 2 log was observed at 24 h p.i. (Table 1). There-
lation of EC50, EC90, and EC99 values (50%, 90%,
eduction of HAdV DNA replication, respectively)
e. This may be an advantageous feature of the assay
dV titers in disseminated disease are extremely high
opies/ml) (Claas et al., 2005; Heim et al., 2003;
et al., 2004) and reduction of virus loads by several
may be required for clinical effects. Therefore, EC99
be more helpful for predicting efficacy in vivo. As
of validation of our qPCR-based protocol, the well-
antiviral activity of the nucleotide analogue drug
gainst HAdV-C1 was tested. Reduction of HAdV
ntration was clearly concentration-dependent. Thus,
r cidofovir was determined to be <1�M, the EC90
ere 3.1�M (95% confidence interval (CI) 2–6�M)
M (CI 95% 10–44�M), respectively. As a second
al activity of ribavirin against HAdV-C1 was tested
R-based protocol. For example, HAdV-C1 DNA
was inhibited by 100�M ribavirin at about 99.3%
ibavirin also reduced increase of HAdV-C1 DNA in
tion-dependent manner (Fig. 1A), and EC50, EC90,
alues of <10�M, 27.3�M (CI 95% 15–54�M) and
I 95% 52–230�M) were calculated from this data
parison, a CPE-based antiviral assay also demon-
vity of cidofovir and ribavirin against HAdV-1 but
s were about 10-fold higher (11 and 64�M, respec-
ompared to values determined by HAdV DNA quan-
ral activity of ribavirin against various serotypes
summarizes the in vitro inhibitory effect of ribavirin
plication of HAdV prototypes and clinical isolates.
ad significant activity against all tested HAdV of
HAdV-C1, -C1 K, -C2, -C2 J, -C5) as determined by
ased assay (Table 2). Mean EC50, EC90, and EC99
HAdV-C were 9.3�M (S.D. 1.8), 33.7�M (S.D. 7)
Fig. 1. I
and 500
and 14
ing of
662–9
replica
parison
indicat
HAdV
was sli
>80 (T
Table 2
Antivira
A549 ce
HAdV p
HAdV
HAdV
HAdV
HAdV
HAdV
HAdV
HAdV
HAdV
Clinical
HAdV
HAdV
HAdV
HAdV
Ribaviri
tion (EC
in brack
a Nottion of HAdV DNA replication at 24 h p.i. by 0, 10, 50, 100, 200,
ibavirin. (A) HAdV-C1 and (B) HAdV-B34.
M (S.D. 53.7). Comparison with cytotoxicity test-
irin on A549 cells (CC50 value of 802�M (CI 95%
M)) indicated a specific inhibition of HAdV DNA
not related to general cytotoxicity (SI > 80). For com-
e CC50 values of cidofovir were even >1000�M
a highly selective antiviral action (SI > 1336 for

38 R. Stock et al. / Antiviral Research 72 (2006) 34–41
an EC99 value at 24 h p.i., because the increase of HAdV-A31
DNA during the first 24 h p.i. was less than 2 log (Table 1).
EC50 and E
cal isolate
activity aga
respectivel
the standar
48 h p.i. Th
avirin again
EC90 202.5
clinical iso
EC90 104�
compared
that the low
was cell ty
A549 cells
of A549 ce
for the HA
W (Table 2
HAdV-B34
HAdV-A31
tively.
We did
other than H
that cidofo
serotypes t
B11, -B14,
2005).
4. Discuss
Develop
antiviral ag
different se
study. CPE
infected ce
to 3 weeks
serotype an
al., 2005; N
down testin
clinical iso
ovirus CPE
visually fro
laboratory
cent focus
the antivira
et al., 1996
number of
required se
We develo
DNA repli
advantage
tion of all
screening o
of HAdV r
protocol m
convention
As a first step of validation of our qPCR-based protocol,
the antiviral activity of the nucleotide analogue drug cidofovir
t HA
ined
ted a
ion a
pre
tivit
ined
1991
n det
tated
2 cel
repo
y of
is of
2005
ed in
us in
fied
nfect
(Na
activ
moi
strat
C50
DN
ve fo
ly o
easu
trati
ilarl
var
oi,
2005
y of
n dif
A m
C s
05),
y eve
re th
eter
Naes
tive
al ac
peci
ed b
cido
repli
PE-b
n the
s dir
the
nown
in se
naloC90 values of the prototype HAdV-31 and a clini-
(HAdV-A31W) indicated that ribavirin had only low
inst HAdV-A31 resulting in SI values of 14 and 47,
y. In order to calculate an EC99 value for HAdV-A31,
d protocol was modified, and samples were taken at
ese results confirmed that the antiviral activity of rib-
st HAdV-A31 (EC50 20.8�M (CI 95% 10–48�M),
�M (CI 95% 84–596�M), EC99 > 500�M) and the
late HAdV-A31 W (EC50 8.1�M (CI 95% 2–90�M),
M (CI 95% 16–2718�M), EC99 > 500�M) was low
to HAdV-C species (Table 2). As it was suspected
antiviral activity of ribavirin against HAdV-A31
pe-related because of its slow DNA replication in
, experiments were repeated with HeLa cells instead
lls. However, almost identical results were obtained
dV-A31 prototype and the clinical isolate HAdV-31
). Furthermore, antiviral activity of ribavirin against
and -D48 was in the same range as activity against
(Table 2), resulting in SI values of 50 and 33, respec-
not test the activity of cidofovir against HAdV types
AdV-C1 because previous studies had demonstrated
vir had almost identical activity against all HAdV
ested (HAdV-C1, -C2, -E4, -C5, -B7, -D8, -D9, -
-D19, and -F41) (Gordon et al., 1991; Morfin et al.,
ion
ment of a rapid and objective in vitro assay for testing
ents against the 51 types of HAdV, which may exhibit
nsitivity to an antiviral agent was the first step of our
with rounding and grapelike clustering of swollen
lls need to develop at least 48 h but may require up
, depending on the amount of virus (moi), HAdV
d cell line used (Kojaoghlanian et al., 2003; Morfin et
aesens et al., 2005; Swenson et al., 2003). This slows
g of antiviral agents and almost precludes testing of
lates for therapy decision making. Moreover, aden-
is sometimes hard to identify and hard to distinguish
m non-specific cell alterations even by experienced
personal (Mentel et al., 1996). Therefore, a fluores-
reduction assay has already been developed to test
l activity of several drugs against HAdV (Mentel
). This protocol made it much easier to identify the
HAdV positive cells culture. However, this assay still
veral days of time and specially trained lab personal.
ped a protocol, which uses the inhibition of HAdV
cation as an objective readout. This protocol takes
of a generic qPCR system which permits quantifica-
51 HAdV serotypes (Heim et al., 2003). Moreover,
f new antiviral drugs that interact with the early steps
eplication up to DNA replication by a qPCR-based
ay be much faster and more objective compared to
al CPE-based assays.
agains
determ
indica
reduct
ably to
The ac
determ
et al.,
antige
study s
(HEp-
it was
tiplicit
analys
et al.,
was us
chrono
quanti
ever, i
effects
lower
a low
demon
low E
HAdV
sensiti
probab
than m
concen
Sim
HAdV
line, m
et al.,
activit
data o
1972).
HAdV
al., 20
activit
the mo
fovir d
2005;
less ac
antivir
that a s
observ
As for
DNA
by a C
used i
inhibit
nately,
is unk
ribavir
oside adV-C1 was tested. Thus, the EC50 for cidofovir was
to be 0.6�M (CI 95% 0.4–0.9). Although this value
higher activity than our results with a classical CPE
ssay (EC50 11�M), both results compared favor-
viously published data that varied also considerably.
y of cidofovir against HAdV-C1 in A549 cells as
by plaque reduction assay (EC50 0.6�M) (Gordon
) was more than 10-fold higher as determined by
ection (EC50 31�M) (Morfin et al., 2005). The latter
also some variability depending on the cell line used
ls, EC50 67�M; PLC cells, EC50 10�M). Moreover,
rted that other factors such, as for example, mul-
infection (moi) and time between inoculation and
antiviral effects may influence EC50 results (Naesens
). A high multiplicity of infection (10 CCID50/cell)
our qPCR-based protocol to achieve an almost syn-
fection of cells. Therefore, antiviral activity can be
during the first round of HAdV replication. How-
ion with a high moi may lead to decreased antiviral
esens et al., 2005) and consequently to a virtually
ity of an antiviral agent compared to an assay using
. Nevertheless, antiviral activity of cidofovir was
ed in our qPCR-based protocol with comparatively
values. This indicates that direct measurement of
A replication, the putative target of cidofovir, is more
r detecting the antiviral activity of cidofovir (and
ther nucleotide/nucleoside analogue antiviral agents)
ring indirect parameters such as CPE or antigen
on.
y to cidofovir, antiviral activity of ribavirin against
ied considerably, probably also depending on cell
and incubation time (Morfin et al., 2005; Naesens
; Sidwell et al., 1972). In the first paper, antiviral
ribavirin against HAdV was reported but no detailed
ferent HAdV types were presented (Sidwell et al.,
ore recent study showed activity of ribavirin against
pecies but not other species of HAdV (Morfin et
whereas another recent study did not detect antiviral
n against HAdV-C2 (Naesens et al., 2005). Regarding
an 1 log differences between EC50 values of cido-
mined by slightly different protocols (Morfin et al.,
ens et al., 2005), it is plausible that with a probably
substance such as ribavirin contradictory results on
tivity may be generated. However, it is remarkable
es-specific activity of ribavirin against HAdV-C was
oth by Morfin et al. (2005) and in this study (Table 2).
fovir, antiviral activity of ribavirin against HAdV-C1
cation (EC50 <10�M) was higher than determined
ased assays (EC50 62�M) although a lower moi was
latter assay. This result may suggest that ribavirin
ectly or indirectly HAdV DNA replication. Unfortu-
mode of antiviral action of ribavirin against HAdV
. A direct inhibition of HAdV DNA polymerase by
ems to be quite unlikely because ribavirin is nucle-
gue and not a deoxynucleoside analogue. However,

R. Stock et al. / Antiviral Research 72 (2006) 34–41 39
this mode of action is not impossible because precise substrate
specificity of HAdV DNA polymerase is unknown.
Recentl
the RNA po
al., 2000).
ribavirin le
tant for ant
(Crotty et
showing th
sackievirus
Moreover,
cific intera
et al., 1997
with primi
(Temperley
scripts of H
1979; Sche
sion of HA
replication
teins or typ
to ribavirin
types to rib
nase by rib
(Streeter e
sensitivity
clude that I
antiviral ac
et al. (2003
induced int
activity.
Another
activity of
serotypes s
avirin agai
may explai
treated wit
ical studies
avirin or c
the life-thr
ited (Baldw
1997; Chak
Rosa et al.
Neverthele
of potentia
promised p
2002; Koja
et al., 2004
(Bordigoni
that treatm
mised imm
antiviral tr
be surmise
massively
fering from
reach conc
(Heim et a
Seidemann
Therefore, a prerequisite for a successful treatment seems to
be a significant inhibition of HAdV replication by at least 1 log
or s
y is
ur qP
ty of
(for
eter
emin
alue
resu
ot on
indi
00).
ides
om p
stud
as HA
of c
in se
ical
clin
st, re
ained
g th
, the
inst
CID
e of H
here
HA
CID
vity
ined
houg
dV d
rmac
lasm
of h
ched
(Las
0.2�
n et a
inica
ons.
daily
M (
s of
cant
ytic
(Ler
ulati
ack 5
r, 19
on A
77�
f ribay, it was demonstrated that ribavirin directly interacts
lymerase of poliovirus as a false substrate (Crotty et
In case of poliovirus, an increased mutation rate by
ading to error catastrophe seems to be more impor-
iviral effect than direct inhibition of the polymerase
al., 2000). This may also explain previous results
at ribavirin is more active against persistent Cox-
infections than acute infections (Heim et al., 1997).
ribavirin resistant mutants emerge indicating the spe-
ction of ribavirin with poliovirus polymerase (Heim
; Vignuzzi et al., 2005). Ribavirin may also interfere
ng of HAdV DNA replication by terminal protein
and Hay, 1992) or inhibit capping of early tran-
AdV as described for other viruses (Goswami et al.,
idel and Stollar, 1991) and thus interfere with expres-
dV early proteins necessary for efficient HAdV DNA
. HAdV type specific expression levels of early pro-
e-related susceptibility of HAdV DNA polymerase
could explain the variable sensitivity of HAdV
avirin. By contrast, inhibition of IMP dehydroge-
avirin resulting in decreased cellular dGTP pools
t al., 1973) can probably not explain type-related
of different HAdV types. However, this does not pre-
MP dehydrogenase inhibition may contribute to the
tivity of ribavirin against HAdV. Recently, Zhang
) suggested, that ribavirin interferes with the virus-
erferon signaling pathway and intenses the interferon
topic of this study was to establish the in vitro
ribavirin against several clinically important HAdV
trains and clinical isolates. Different activity of rib-
nst HAdV types (Table 2) (Morfin et al., 2005)
n the different outcomes of HAdV infected patients
h ribavirin. So far, controlled and randomized clin-
on antiviral treatment of HAdV disease with rib-
idofovir are not available and treatment options for
eatening disseminated HAdV disease are very lim-
in et al., 2000; Bordigoni et al., 2001; Carrigan,
rabarti et al., 2002; Kojaoghlanian et al., 2003; La
, 2001; Ljungman, 2004; Seidemann et al., 2004).
ss, smaller studies and case reports were suggestive
lly beneficial effects of ribavirin in immunocom-
atients (Arav-Boger et al., 2000; Gavin and Katz,
oghlanian et al., 2003; Ljungman, 2004; Schleuning
) but ribavirin therapy was not always successful.
et al., 2001; Lankester et al., 2004). It was suspected
ent failures may be related to massively compro-
une status, and the importance of host factors in
eatment of disseminated adenovirus infections can
d. HAdV DNA loads in clinical blood samples of
immunocompromised bone marrow recipients suf-
life-threatening disseminated HAdV disease may
entrations of >1 × 1010 HAdV DNA copies/ml blood
l., 2003; Lankester et al., 2004; Lion et al., 2003;
et al., 2004).
(90%)
activit
ever, o
capaci
pletely
param
in diss
EC99 v
These
may n
of the
al., 20
Bes
lates fr
in this
range
typing
ribavir
of clin
ing of
contra
be obt
isolatin
rapidly
qPCR
with C
ferenc
1.7). T
of 104
of 10 C
sensiti
determ
Alt
of HA
on pha
peak p
values
be rea
avirin
were 1
(Laski
port cl
infecti
twice
11.07�
tration
signifi
hemol
avirin
accum
cells l
Conno
effects
662–9
icity oeveral logs (for example 99% or 99.9%). Antiviral
usually indicated by calculating EC50 values. How-
CR experiments were also designed to evaluate the
ribavirin to inhibit HAdV replication almost com-
example by 99%) by determining EC99 values. This
seems to be a better predictor for antiviral efficacy
ated HAdV infections with high virus loads, since
s of different HAdV types varied widely (Table 2).
lts suggest that clinical failure of ribavirin therapy
ly depend on host factors but also on the sensitivity
vidual HAdV type causing the infection (Venard et
prototype stains of HAdV types, clinical HAdV iso-
atients with disseminated HAdV disease were tested
y. EC50, EC90, and EC99 values were in the same
dV prototypes (Table 2). These results suggest that
linical HAdV isolates may be sufficient to predict
nsitivity. Nevertheless, ribavirin sensitivity testing
isolates seems to be a sensible strategy because typ-
ical isolates usually requires more than 1 week. By
sults on ribavirin sensitivity of clinical isolates can
by the qPCR-based protocol as early as 30 h after
e virus. For providing sensitivity testing results more
clinical isolate inoculum should be quantified by
ead of the CCID50 method. Comparison of qPCR
50 results of clinical isolates revealed a mean dif-
AdV DNA (copies/ml) to CCID50/ml of 3 log (S.D.
fore, we suggest an inoculum with a concentration
dV DNA copies/cell, approximately equal to a moi
50/cell. However, threshold EC50 values describing
or resistance of clinical HAdV isolates have to be
in clinical treatment studies.
h controlled clinical studies for ribavirin treatment
isease are not available, there is ample clinical data
okinetics, tolerated doses and side effects. Ribavirin
a concentrations (160.8�M) were even above EC99
ighly sensitive HAdV-C viruses (Table 2) and can
after a single dose of 2400 mg intravenous rib-
kin et al., 1987). Trough levels 8 h after infusion
M, still above EC50 values of HAdV-C (Table 2)
l., 1987). These pharmacokinetic results clearly sup-
l studies with i.v. ribavirin for disseminated HAdV-C
With long term oral application of 400 mg ribavirin
, mean peak concentration of 6.61�M (S.D. 0.5) to
S.D. 2.3) were reached, approximately EC50 concen-
species C HADV (Table 2) (Lertora et al., 1991). No
toxicities related to ribavirin were apparent, except
anemia which did not require discontinuation of rib-
tora et al., 1991). Hemolytic anemia is caused by
on of ribavirin in red blood cells because red blood
′ nucleotidase and alkaline phosphatase (Page and
90). In vitro, ribavirin had no or only minor cytotoxic
549 cells (CC50 802�M, 95% confidence interval
M). This result is in accordance with the low cytotox-
virin in HeLa and Vero cells, as determined by 50%

40 R. Stock et al. / Antiviral Research 72 (2006) 34–41
inhibition of plating efficiency (4000�M) (Kirsi et al., 1984) and
CC50 values for HEp-2 cells (400�M) and PLC cells (2200�M)
(Morfin et
observed i
30�M) (N
a significan
siderably h
than anticip
In conc
qPCR-base
ity of cidof
ribavirin ag
established
ing of new
in diagnost
future sens
to design c
HAdV.
Acknowled
We than
F. Kennedy
for help wi
Reference
Arav-Boger, R
Clearance
transplant
Baldwin, A.,
Goulden,
2000. Out
tion follow
1333–133
Blasi, F., 2003
Bordigoni, P.,
adenoviru
cell transp
Carrigan, D.R
Am. J. Me
Chakrabarti, S
per, P.E., M
allogeneic
graft man
1619–162
Chou, T.C., T
the combi
Regul. 22
Claas, E.C., S
Lankester
real-time
transplant
Crotty, S., Ma
Cameron,
is an RNA
Emovon, O.E
P.K., Chav
adenoviru
successfu
venous im
Gavin, P.J., K
ovirus dis
Gordon, Y.J., Romanowski, E., Araullo-Cruz, T., Seaberg, L., Erzurum, S., Tol-
man, R., De Clercq, E., 1991. Inhibitory effect of (S)-HPMPC, (S)-HPMPA,
and 2′-nor-cyclic GMP on clinical ocular adenoviral isolates is serotype-
nden
i, B.
broad
hem.
., Ebn
detec
28–2
., Gr
dolf,
infec
an na
zer, J.
y, B.W
s, Lon
J., Mc
., 198
ribavi
lania
rus in
171.
, A.M
d, I.,
noviru
. Infe
er, A.
ius, R
ma vi
. Infe
O.L.,
J.D.,
cquir
J.J.,
., Agr
toler
unod
, Bau
, S., F
003.
ic bo
ase. B
an, P.,
d hos
R., K
bitory
ogues
R., M
s redu
h. 59,
ra, K
Saito,
oder
orrha
rs ra
F., Du
k, P., B
005.
nden
, L.,
arini,
cleos
–101
, Con
eatedal., 2005). However, a higher cytotoxic activity was
n human embryonic lung fibroblasts (HEL) (CC50
aesens et al., 2005). The latter CC50 value indicated
t cytotoxicity of ribavirin, which seems to be con-
igher than in all other cell lines and also much higher
ated by side effects of clinical studies.
lusion, rapid and objective results generated by a
d antiviral activity assay confirmed the high activ-
ovir and indicated the serotype-dependent activity of
ainst the most predominant HAdV of species C. The
qPCR-based protocol may also speed up the screen-
antiviral agents against HAdV. Virus load testing
ic blood samples, typing of clinical isolates, and in
itivity testing of clinical HAdV isolates may also help
linical studies for evaluating antiviral agents against
gment
k Jens Hainmueller, MPA (Harvard University, John
School, Department of Government, Boston, MA)
th biostatistics.
s
., Echavarria, M., Forman, M., Charache, P., Persaud, D., 2000.
of adenoviral hepatitis with ribavirin therapy in a pediatric liver
recipient. Pediatr. Infect. Dis. J. 19, 1097–1100.
Kingman, H., Darville, M., Foot, A.B., Grier, D., Cornish, J.M.,
N., Oakhill, A., Pamphilon, D.H., Steward, C.G., Marks, D.I.,
come and clinical course of 100 patients with adenovirus infec-
ing bone marrow transplantation. Bone Marrow Transplant. 26,
8.
. Ribavirin. It. J. Chest Dis. 57, 31–42.
Carret, A.S., Venard, V., Witz, F., Le Faou, A., 2001. Treatment of
s infections in patients undergoing allogeneic hematopoietic stem
lantation. Clin. Infect. Dis. 32, 1290–1297.
., 1997. Adenovirus infections in immunocompromised patients.
d. 102, 71–74.
., Mautner, V., Osman, H., Collingham, K.E., Fegan, C.D., Klap-
oss, P.A., Milligan, D.W., 2002. Adenovirus infections following
stem cell transplantation: incidence and outcome in relation to
ipulation, immunosuppression, and immune recovery. Blood 100,
7.
alalay, P., 1984. Quantitative analysis of dose-effect relationships:
ned effects of multiple drugs or enzyme inhibitors. Adv. Enzyme
, 27–55.
chilham, M.W., de Brouwer, C.S., Hubacek, P., Echavarria, M.,
, A.C., van Tol, M.J., Kroes, A.C., 2005. Internally controlled
PCR monitoring of adenovirus DNA load in serum or plasma of
recipients. J. Clin. Microbiol. 43, 1738–1744.
ag, D., Arnold, J.J., Zhong, W., Lau, J.Y., Hong, Z., Andino, R.,
C.E., 2000. The broad-spectrum antiviral ribonucleoside ribavirin
virus mutagen. Nat. Med. 6, 1375–1379.
., Lin, A., Howell, D.N., Afzal, F., Baillie, M., Rogers, J., Baliga,
in, K., Nickeleit, V., Rajagapalan, P.R., Self, S., 2003. Refractory
s infection after simultaneous kidney-pancreas transplantation:
l treatment with intravenous ribavirin and pooled human intra-
munoglobulin. Nephrol. Dial. Transplant. 18, 2436–2438.
atz, B.Z., 2002. Intravenous ribavirin treatment for severe aden-
ease in immunocompromised children. Pediatrics 110, e9.
depe
Goswam
The
Bioc
Heim, A
tive
70, 2
Heim, A
Kan
state
hum
Hierhol
Mah
Pres
Kirsi, J.
R.K
and
Kojaogh
novi
155–
La Rosa
Raa
Ade
Clin
Lankest
Bred
plas
Clin
Laskin,
nor,
for a
555.
Lertora,
R.B
term
imm
Lion, T.
uner
H., 2
gene
dise
Ljungm
mise
Mentel,
Inhi
anal
Mentel,
focu
Met
Miyamu
M.,
T., K
hem
dono
548.
Morfin,
lace
D., 2
depe
Naesens
Balz
of nu
1010
Page, T.
nuclt in vitro. Antiviral Res. 16, 11–16.
B., Borek, E., Sharma, O.K., Fujitaki, J., Smith, R.A., 1979.
-spectrum antiviral agent ribavirin inhibits capping of mRNA.
Biophys. Res. Commun. 89, 830–836.
et, C., Harste, G., Pring-Akerblom, P., 2003. Rapid and quantita-
tion of human adenovirus DNA by real-time PCR. J. Med. Virol.
39.
umbach, I., Pring-Akerblom, P., Stille-Siegener, M., Muller, G.,
R., Figulla, H.R., 1997. Inhibition of coxsackievirus B3 carrier
tion of cultured human myocardial fibroblasts by ribavirin and
tural interferon-alpha. Antiviral Res. 34, 101–111.
C., Killington, R.A., 1996. Virus isolation and quantitation. In:
.J., Kangro, H.O. (Eds.), Virology Methods Manual. Academic
don, pp. 25–46.
Kernan, P.A., Burns III, N.J., North, J.A., Murray, B.K., Robins,
4. Broad-spectrum synergistic antiviral activity of selenazofurin
rin. Antimicrob. Agents Chemother. 26, 466–475.
n, T., Flomenberg, P., Horwitz, M.S., 2003. The impact of ade-
fection on the immunocompromised host. Rev. Med. Virol. 13,
., Champlin, R.E., Mirza, N., Gajewski, J., Giralt, S., Rolston, K.V.,
Jacobson, K., Kontoyiannis, D., Elting, L., Whimbey, E., 2001.
s infections in adult recipients of blood and marrow transplants.
ct. Dis. 32, 871–876.
C., Heemskerk, B., Claas, E.C., Schilham, M.W., Beersma, M.F.,
.G., van Tol, M.J., Kroes, A.C., 2004. Effect of ribavirin on the
ral DNA load in patients with disseminating adenovirus infection.
ct. Dis. 38, 1521–1525.
Longstreth, J.A., Hart, C.C., Scavuzzo, D., Kalman, C.M., Con-
Roberts, R.B., 1987. Ribavirin disposition in high-risk patients
ed immunodeficiency syndrome. Clin. Pharmacol. Ther. 41, 546–
Rege, A.B., Lacour, J.T., Ferencz, N., George, W.J., VanDyke,
awal, K.C., Hyslop Jr., N.E., 1991. Pharmacokinetics and long-
ance to ribavirin in asymptomatic patients infected with human
eficiency virus. Clin. Pharmacol. Ther. 50, 442–449.
mgartinger, R., Watzinger, F., Matthes-Martin, S., Suda, M., Pre-
utterknecht, B., Lawitschka, A., Peters, C., Potschger, U., Gadner,
Molecular monitoring of adenovirus in peripheral blood after allo-
ne marrow transplantation permits early diagnosis of disseminated
lood 102, 1114–1120.
2004. Treatment of adenovirus infections in the immunocompro-
t. Eur. J. Clin. Microbiol. Infect. Dis. 23, 583–588.
inder, M., Wegner, U., von Janta-Lipinski, M., Matthes, E., 1997.
activity of 3′-fluoro-2′-deoxythymidine and related nucleoside
against adenoviruses in vitro. Antiviral Res. 34, 113–119.
atthes, E., von Janta-Lipinski, M., Wegner, U., 1996. Fluorescent
ction assay for the screening of antiadenoviral agents. J. Virol.
99–104.
., Hamaguchi, M., Taji, H., Kanie, T., Kohno, A., Tanimoto,
H., Kojima, S., Matsuyama, T., Kitaori, K., Nagafuji, K., Sato,
a, Y., 2000. Successful ribavirin therapy for severe adenovirus
gic cystitis after allogeneic marrow transplant from close HLA
ther than distant donors. Bone Marrow Transplant. 25, 545–
puis-Girod, S., Mundweiler, S., Falcon, D., Carrington, D., Sed-
ierings, M., Cetkovsky, P., Kroes, A.C., van Tol, M.J., Thouvenot,
In vitro susceptibility of adenovirus to antiviral drugs is species-
t. Antiviral Ther. 10, 225–229.
Lenaerts, L., Andrei, G., Snoeck, R., Van Beers, D., Holy, A.,
J., De Clercq, E., 2005. Antiadenovirus activities of several classes
ide and nucleotide analogues. Antimicrob. Agents Chemother. 49,
6.
nor, J.D., 1990. The metabolism of ribavirin in erythrocytes and
cells. Int. J. Biochem. 22, 379–383.

R. Stock et al. / Antiviral Research 72 (2006) 34–41 41
Scheidel, L.M., Stollar, V., 1991. Mutations that confer resistance to mycophe-
nolic acid and ribavirin on Sindbis virus map to the nonstructural protein
nsP1. Virology 181, 490–499.
Schleuning, M., Buxbaum-Conradi, H., Jager, G., Kolb, H.J., 2004. Intravenous
ribavirin for eradication of respiratory syncytial virus (RSV) and adenovirus
isolates from the respiratory and/or gastrointestinal tract in recipients of
allogeneic hematopoietic stem cell transplants. Hematol. J. 5, 135–144.
Schnurr, D., Dondero, M.E., 1993. Two new candidate adenovirus serotypes.
Intervirology 36, 79–83.
Seidemann, K., Heim, A., Pfister, E.D., Ko¨ditz, H., Beilken, A., Sander, A.,
Melter, M., Sykora, K.W., Sasse, M., Wessel, A., 2004. Monitoring of aden-
ovirus infection in pediatric transplant recipients by quantitative PCR: report
of six cases and review of the literature. Am. J. Transplant. 4, 2102–2108.
Sidwell, R.W., Huffman, J.H., Khare, G.P., Allen, L.B., Witkowski, J.T.,
Robins, R.K., 1972. Broad-spectrum antiviral activity of Virazole: 1-�-d-
ribofuranosyl-1,2,4-triazole-3-carboxamide. Science 177, 705–706.
Streeter, D.G., Witkowski, J.T., Khare, G.P., Sidwell, R.W., Bauer, R.J., Robins,
R.K., Simon, L.N., 1973. Mechanism of action of 1-�-d-ribofuranosyl-1,2,4-
triazole-3-carboxamide (Virazole), a new broad-spectrum antiviral agent.
Proc. Natl. Acad. Sci. U.S.A. 70, 1174–1178.
Swenson, P.D., Wadell, G., Allart, A., Hierholzer, J.C., Murray, P.R., Baron, E.J.,
Pfaller, M., Jorgensen, J., Yolken, R., 2003. Adenoviruses. In: Murray, P.R.,
et al. (Eds.), Clinical Microbiology. American Society for Microbiology,
Washington, DC, pp. 1404–1417.
Temperley, S.M., Hay, R.T., 1992. Recognition of the adenovirus type 2 origin
of DNA replication by the virally encoded DNA polymerase and preterminal
proteins. EMBO J. 11, 761–768.
Venard, V., Carret, A., Corsaro, D., Bordigoni, P., Le Faou, A., 2000. Genotyping
of adenoviruses isolated in an outbreak in a bone marrow transplant unit
shows that diverse strains are involved. J. Hosp. Infect. 44, 71–74.
Vignuzzi, M., Stone, J.K., Andino, R., 2005. Ribavirin and lethal mutagenesis of
poliovirus: molecular mechanisms, resistance and biological implications.
Virus Res. 107, 173–181.
Wutzler, P., Sauerbrei, A., Klocking, R., Brogmann, B., Reimer, K., 2002.
Virucidal activity and cytotoxicity of the liposomal formulation of povidone-
iodine. Antiviral Res. 54, 89–97.
Zhang, Y., Jamaluddin, M., Wang, S., Tian, B., Garofalo, R.P., Casola, A.,
Brasier, A.R., 2003. Ribavirin treatment up-regulates antiviral gene expres-
sion via the interferon-stimulated response element in respiratory syncytial
virus-infected epithelial cells. J. Virol. 77, 5933–5947.