Detection of inhibition of bovine viral diarrhea virus by aromatic cationic molecules

Article abstract of DOI:10.1128/AAC.47.7.2223-2230.2003

Bovine viral diarrhea virus (BVDV) is an economically significant pathogen of cattle and a problematic contaminant in the laboratory. BVDV is often used as an in vitro model for hepatitis C virus during drug discovery efforts. Aromatic dicationic molecule

Full text of DOI:10.1128/AAC.47.7.2223-2230.2003

ANTIMICROBIAL AGENTS AND CHEMOTHERAPY, July 2003, p. 2223–2230
0066-4804/03/$08.00ϩ0 DOI: 10.1128/AAC.47.7.2223–2230.2003
Copyright © 2003, American Society for Microbiology. All Rights Reserved.
Vol. 47, No. 7
Detection of Inhibition of Bovine Viral Diarrhea Virus
by Aromatic Cationic Molecules
M. Daniel Givens,1* Christine C. Dykstra,1 Kenny V. Brock,1 David A. Stringfellow,1
Arvind Kumar,2 Chad E. Stephens,2 Hakan Goker,2 and David W. Boykin2
Department of Pathobiology, College of Veterinary Medicine, Auburn University, Auburn, Alabama 36849,1 and
Department of Chemistry, Georgia State University, Atlanta, Georgia 303032
Received 8 October 2002/Returned for modification 12 December 2002/Accepted 24 March 2003
Bovine viral diarrhea virus (BVDV) is an economically significant pathogen of cattle and a problematic
contaminant in the laboratory. BVDV is often used as an in vitro model for hepatitis C virus during drug
discovery efforts. Aromatic dicationic molecules have exhibited inhibitory activity against several RNA viruses.
Thus, the purpose of this research was to develop and apply a method for screening the aromatic cationic
compounds for in vitro cytotoxicity and activity against a noncytopathic strain of BVDV. The screening method
evaluated the concentration of BVDV in medium and cell lysates after 72 h of cell culture in the presence of
either a 25 or 5 ␮M concentration of the test compound. Five of 93 screened compounds were selected for
further determination of inhibitory (90 and 50%) and cytotoxic (50 and 10%) concentration endpoints. The
screening method identified compounds that exhibited inhibition of BVDV at nanomolar concentrations while
exhibiting no cytotoxicity at 25 ␮M concentrations. The leading compounds require further investigation to
determine their mechanism of action, in vivo activity, and specific activity against hepatitis C virus.
Bovine viral diarrhea virus (BVDV), which is the prototype
of the Pestivirus genus of the Flaviviridae family, causes early
embryonic death, abortion, teratogenesis, respiratory prob-
lems, chronic wasting syndrome, and immune system dysfunc-
tion in cattle throughout the world. Acute infections of immu-
nocompetent cattle with different strains of BVDV have
caused mortality rates of 17 to 32% (8, 15, 31). Vaccines have
provided inconsistent protection against infection with BVDV
(36).
In addition to being a pathogen of cattle, BVDV can be a
problematic contaminant in the laboratory. Two biotypes of
BVDV (noncytopathic and cytopathic), which are based on
their effects in cell cultures, are recognized (3). Noncytopathic
biotypes of BVDV have been identified in commercially avail-
able lots of fetal bovine serum despite testing by the manufac-
turer (6, 45). Consequently, BVDV has been identified in com-
mercially available bovine, canine, feline, and primate cell lines
(16, 19), human viral vaccines for measles-mumps-rubella (18,
21), and interferons for human use (20). Viral contamination
of biologicals for human use may be the reason BVDV-specific
antibodies have been found in some samples of human serum
(17). No antiviral pharmaceuticals are currently available for
controlling BVDV in the laboratory or on the farm.
ability to propagate HCV efficiently by cell culture has caused
researchers to adopt BVDV as a viral model for HCV (2).
Because of the relatedness of BVDV and HCV, research with
the inhibitors of BVDV replication is useful both for the con-
trol of animal diseases and for efforts to discover drugs effec-
tive against HCV. Aromatic cationic molecules have exhibited
inhibitory activity against respiratory syncytial virus (14), rota-
virus (39), and human immunodeficiency virus (24). Therefore,
the purpose of this research was to develop and apply a
method of screening aromatic cationic molecules for in vitro
toxicity and activity against BVDV. These cationic molecules
were selected for study in view of their recognition of specific
RNA sequences in other RNA viruses (27).
MATERIALS AND METHODS
Compounds. All of the compounds screened in this study were synthesized in
the laboratory of one of the authors (D.W.B.). Stock solutions of 5 or 10 mM
were made in sterile distilled water or dimethyl sulfoxide (DMSO) and stored at
Ϫ80°C until use. While screening results for 41 compounds are described, 52
additional compounds were screened. These additional compounds exhibited
cytotoxicity and/or lack of antiviral effect. Information concerning these com-
pounds is available upon request. The synthesis and physical properties of the
following compounds have been described in the references indicated: com-
pound DB18 (12); compounds DB60, DB75, and DB99 (11); compound DB289
(7); compounds DB293, DB294, and DB302 (22); compound DB501 (41); com-
pounds DB606, DB771, and DB772 (26); and compounds DB701, DB711,
DB748, DB762, and DB763 (35). Brief descriptions of the synthesis and physical
properties of DB779, DB458, DB456, and DB459 are given here. The descrip-
tions of the synthesis of the remaining compounds will be published separately.
Melting points (mp) were determined in open capillary tubes with a Mel-
Temperature 3.0 capillary mp apparatus and are reported uncorrected. 1H and
13C nuclear magnetic resonance (NMR) spectra were recorded on a Varian
Unity ϩ 300 or a Varian VRX 400 instrument. Coupling constants are reported
in hertz. Mass spectra were recorded on a VG Instruments 70-SE spectrometer
at the Georgia Institute of Technology, Atlanta. Elemental analyses were per-
formed by Atlantic Microlab, Norcross, Ga. All of the chemicals and solvents
were purchased from Aldrich Chemical Co., Fisher Scientific, or Acros Organics.
2,5-Bis(3-ethoxy-4-guanidinophenyl)furan dihydrochloride (DB779). 2-Nitro-
5-bromophenetole (64% yield; mp, 78 to 79°C [ethanol-water]) was produced by
Hepatitis C virus (HCV), a member of the Hepacivirus genus
of the family Flaviviridae, is a major cause of human liver
disease throughout the world. The World Health Organization
estimates that 170 million people are chronically infected with
HCV (2). The organization of the HCV genome encoding the
proteins for viral replication is very similar to that of BVDV
(with the exception of the 5Ј-terminal protease) (4). The in-
* Corresponding author. Mailing address: 127 Sugg Laboratory, De-
partment of Pathobiology, College of Veterinary Medicine, Auburn
University, Auburn, AL 36849. Phone: (334) 844-4952. Fax: (334)
844-4955. E-mail: givenmd@vetmed.auburn.edu.
2223

2224
GIVENS ET AL.
ANTIMICROB. AGENTS CHEMOTHER.
the reaction of 3,4-dinitrobromobenzene with sodium ethoxide in ethanol (37).
Coupling of the bromo compound with 2,5-bis(tributylstannyl)furan gave, after
recrystallization from N,N-dimethylformamide–methanol, 2,5-bis(3-ethoxy-4-ni-
antibiotics (100 U of penicillin G/ml, 100 ␮g of streptomycin/ml, and 0.25 ␮g of
amphotericin B/ml) (MEM-eq). Virus was harvested by freezing and thawing and
was stored as aliquots in cryovials at Ϫ80°C until needed. The 50% cell culture
infective doses (CCID50) of virus per milliliter were determined by the statistical
method of Reed and Muench (33). Antiviral testing was performed in MEM-eq.
Virus inhibition test. Initially, 93 compounds were screened at 25 and/or 5 ␮M
for antiviral effects and cytotoxicity. Compounds were selected for further testing
based on their inhibition of viral replication without visible cytotoxicity in cell
culture. For five selected compounds (DB456, DB459, DB606, DB771, and
DB772), antiviral activity and cytotoxicity were evaluated in twofold dilutions of
the compound at concentrations of 25 to 0.007 ␮M to determine the inhibitory
concentrations at 90% (IC90) and at 50% (IC50).
trophenyl)furan as a yellow-orange fluffy solid (75% yield; mp, 192 to 194°C). 1
H
NMR (DMSO-d6): 1.38 (t, 6H), 4.34 (q, 4H), 7.51 (s, 2H), 7.59 (dd, J ϭ 8.4, 1.8,
2H), 7.69 (d, J ϭ 1.8 Hz, 2H), 7.97 (d, J ϭ 8.7, 2H). Analysis calculated for
C
20H18N2O7 (398.36): C, 60.30; H, 4.55; N, 7.03. Found: C, 60.34; H, 4.58; N,
6.93.
Hydrogenation with Pd on C gave, after crystallization from methanol-water,
2,5-bis(4-amino-3-ethoxyphenyl)furan as a light green and tan solid (85% yield).
1H NMR (DMSO-d6): 1.36 (t, 6H), 4.07 (q, 4H), 4.85 (br s, 4H), 6.63 to 6.68 (m,
4H), 7.10 (m, 4H). From the diamine, the title bis-guanidine was prepared as a
light green hygroscopic solid (76% yield for a two-step procedure). 1H NMR
(DMSO-d6): 1.38 (q, 6H), 4.21 (q, 4H), 7.21 (d, 2H), 7.27 (dd, J ϭ 8.1, 2.1, 2H),
7.42 (br s, 8H), 7.44 to 7.49 (m, 4H), 9.40 (br s, 2NH). Mass spectrum (electro-
spray): m/e 423.3 (60% yield; Mϩ Ϫ 2HCl). Analysis calculated for C22H26N6O3
⅐ 2HCl ⅐ 0.5H2O (504.41): C, 52.38; H, 5.79; N, 16.67. Found: C, 52.25; H, 5.81;
N, 16.52.
MDBK cells (4 ϫ 104 to 8 ϫ 105/2.0-cm2-surface-area well) that were free of
BVDV were incubated in 200 ␮l of medium containing the test compound (or no
compound for the negative control) for 15 min before inoculation with BVDV
(strain SD-1) at a multiplicity of infection of 0.5. The cells were cultured with this
inoculum for 1 h at 38.5°C with 5% CO2 in humidified air. After this 1-h
incubation, culture medium was removed and the cells were washed with 1 ml of
Dulbecco’s phosphate-buffered saline (PBS) without CaCl2 and MgCl2. Imme-
diately after the cells were washed, 1 ml of antiviral test medium that contained
the test compound (or maintained as the negative control) was placed on the
inoculated monolayer. After 72 h of incubation at 38.5°C with 5% CO2 in
humidified air, the cell monolayers were observed for cytotoxic effect with an
inverted culture microscope at a magnification of ϫ400. Culture medium was
then removed from the cell monolayer and stored at Ϫ80°C for later viral assay.
One milliliter of fresh medium containing no test compound was added to the
cell monolayers prior to freezing at Ϫ80°C and thawing to lyse MDBK cells for
the purpose of releasing any intracellular BVDV.
Virus isolation and detection. BVDV was detected and quantified from cell
culture medium and cell lysate samples by endpoint dilution immunoperoxidase
monolayer assays (1). Samples were assayed in triplicate in a 96-well culture plate
by adding 50 ␮l of MEM-eq containing approximately 2.5 ϫ 103 MDBK cells to
25-␮l aliquots of sample diluted in 75 ␮l of MEM-eq. One-hundred-microliter
aliquots of serial dilutions (10Ϫ1 to 10Ϫ7) of all of the samples were also assayed
for BVDV in triplicate. Plates were incubated for 72 h at 38.5°C in a humidified
atmosphere of 5% CO2 and air before the immunoperoxidase labeling technique
was performed.
After incubation for 72 h and subsequent fixation with 20% acetone, poten-
tially infected cells were incubated with monoclonal antibody D89 (38, 43), which
is specific for E2 (gp53), a major envelope glycoprotein of BVDV (44), and
monoclonal antibody 20.10.6, which is specific for NS3 (p80) or unprocessed NS2/3
(p125), a conserved nonstructural protein (10). After the cells were washed with
PBS and Tween 20 to remove unbound antibodies, peroxidase-conjugated rabbit
anti-mouse immunoglobulin G (Jackson Immuno Research Lab, West Grove,
Pa.) was added. After a short incubation period, unbound peroxidase-conjugated
antibody was removed by washing with PBS and Tween 20. Finally, the enzyme
substrate, aminoethyl carbazole (Zymed Laboratories, South San Francisco,
Calif.), which produces a reddish-brown color when oxidized by horseradish
peroxidase, was added. Color change was visualized under light microscopy and
compared to those of the known positive and negative controls on each plate.
Toxicity evaluation. For five selected compounds (DB456, DB459, DB606,
DB771, and DB772), the viability of drug-treated cell cultures was quantitated by
using the tetrazolium-based compound XTT (2,3-bis[2-methoxy-4-nitro-5-sulfo-
phenyl]-2H-tetrazolium-5-carboxanilide) (42) in twofold dilutions of the com-
pound at concentrations from 100 to 1.56 ␮M to determine the 50 and 10%
cytotoxic concentrations (CC50 and CC10).
A cell suspension of MDBK cells was divided into aliquots in antiviral test
medium in 96-well plates to give a final cell count per well of approximately 104.
The plates were incubated at 37°C for 1 to 2 h to allow for cell attachment.
Following visual confirmation of cell attachment, dilutions of selected com-
pounds were added to the appropriate experimental wells to give a final well
concentration of 100, 50, 25, 12.5, 6.25, 3.12, 1.56, or 0 ␮M. The plates were
incubated at 37°C for 24 h. Each plate contained multiple controls of untreated
cells and medium without cells. Each compound was assayed for toxicity at each
concentration in six replicates. Prior to addition of the XTT reagent, the antiviral
test medium in the 96-well plate was removed and replaced with 100 ␮l of
medium lacking phenol red/well. The XTT reagent was then added (25 ␮l/well),
and the plate was incubated at 37°C for 20 min. The optical density (OD) of each
2-[5(6)-{N-Isopropylamidino}-2-benzimidazoyl]-5-(4-nitrophenyl)furan
(DB458). A mixture of 5-(4-nitrophenyl)furfural (0.434 g, 0.002 mol), 4-N-iso-
propylamidino-1,2-phenylenediamine hydrochloride hydrate (0.493 g, 0.002
mol), and 1,4-benzoquinone (0.216 g, 0.002 mol) in 40 ml of ethanol (under
nitrogen) was heated at reflux for 6 h. The volume of the reaction mixture was
reduced to about 15 ml under reduced pressure, the mixture was cooled, and the
resultant solid was collected by filtration to yield 0.66 g (80%) of the monohy-
drochloride salt. The monohydrochloride salt was dissolved in 100 ml of ethanol
and acidified with HCl-saturated ethanol, and after cooling in an ice bath, the
resultant solid was filtered, washed with ether, and dried for 24 h in a vacuum
oven at 75°C to yield 0.7 g (91%) at an mp of Ͼ300°C. 1H NMR (DMSO-d6/
D2O): 8.26 (d, J ϭ 8.8, 2H), 8.11 (d, J ϭ 8.8 Hz, 2H), 8.01 (d, J ϭ 1.2, 1H), 7.77
(d, J ϭ 8.8, 1H), 7.59 (dd, J ϭ1.2, 8.8, 1H), 7.50 (d, J ϭ 7.6, 1H), 7.42 (d, J ϭ 7.6,
1H), 4.04 (septet, J ϭ 6.8, 1H), 1.3 (d, J ϭ 6.8, 6H). 13C NMR (DMSO-d6): 162.7,
153.8, 147.2, 145.2, 144.8, 140.7, 138.2, 135.2, 125.4, 124.7, 124.0, 123.5, 116.3,
115.9, 115.3, 112.6, 45.6, 21.4. Fast atom bombardment mass spectrum
(FABMS): m/e 376 (Mϩ ϩ 1). Analysis calculated for C21H19N5O3 ⅐ 2HCl ⅐
2.0H2O: C, 49.71; H, 5.16; N, 13.80. Found: C, 49.65; H, 5.11; N, 13.50.
2-[5(6)-{2-Imidazolinyl}-2-benzimidazoyl]-5-(4-aminophenyl)furan (DB456).
The monohydrochloride salt of the nitro analog described above (0.5 g, 0.0013
mol) and 0.2 g of 10% Pd or C in 130 ml of methanol were subjected to
hydrogenation at 50 lb/in2 for 4 h. The catalyst was removed by filtration over
diatomaceous earth and by washing with warm methanol. The solvent volume
was reduced to approximately half under reduced pressure. The flask containing
the solution was placed in an ice bath and saturated with HCl gas. The mixture
was stirred at room temperature for 4 h and treated with dry ether, and the solid
was collected by filtration. The solid was dried under vacuum at 75°C for 24 h to
yield 0.55 g (86%) at an mp of Ͼ300°C. 1H NMR (DMSO-d6/D2O): 8.24 (d, J ϭ
1.2, 1H), 7.88 (d, J ϭ 8.0, 2H), 7.80 (s, 2H), 7.51 (d, J ϭ 3.6, 1H), 7.21 (d, J ϭ 8.4,
2H), 7.10 (dd, J ϭ 1.2, 3.6, 1H), 4.0 (s, 4H). 13C NMR (DMSO-d6/D2O): 165.8,
156.4, 145.8, 142.0, 140.9, 137.9, 126.2, 123.7, 121.0, 117.0, 116.8, 115.3, 108.5,
44.6. FABMS: m/e 344 (Mϩ ϩ 1). Analysis calculated for C20H17N5O ⅐ 3HCl ⅐
2.1H2O: C, 48.96; H, 4.97; N, 14.27. Found: C, 48.58; H, 4.32; N, 14.27.
2-[5(6)-N-Isopropylamidino-2-benzimidazoyl]-5-(4-aminophenyl)furan (DB
459). The monohydrochloride salt of the nitro analog described above (0.411 g,
0.001 mol) and 0.3 g of 10% Pd or C in 120 ml of methanol were subjected to
hydrogenation at 50 lb/in2 for 4 h. The catalyst was removed by filtration over
Filteraid. The solvent volume was reduced to approximately half under reduced
pressure. The flask containing the solution was placed in an ice bath and satu-
rated with HCl gas. The mixture was stirred at room temperature for 4 h and
treated with dry ether, and the solid was collected by filtration. The solid was
dried under vacuum at 80°C for 24 h to yield 0.41 g (87%) at an mp of Ͼ300°C.
1H NMR (DMSO-d6/D2O): 8.04 (d, J ϭ 1.6, 1H), 7.91 (d, J ϭ 8.4, 2H), 7.80 (d,
J ϭ 8.4, 1H), 7.64 (dd, J ϭ 1.6, 8.4, 1H), 7.60 (d, J ϭ 4.0, 1H), 7.24 (d, J ϭ 8.4,
2H), 7.14 (d, J ϭ 4.0, 1H), 4.05 (septet, J ϭ 6.4, 1H), 1.3 (d, J ϭ 6.4, 6H). 13C
NMR (DMSO-d6): 162.4, 156.8, 144.4, 140.9, 138.8, 137.6, 135.0, 126.3, 125.4,
124.6, 124.1, 121.1, 118.0, 115.6, 114.9, 108.6, 45.6, 21.3. FABMS, m/e 360 (Mϩ
ϩ 1). Analysis calculated for C21H21N5O3 ⅐ 3HCl: C, 53.80; H, 5.15; N, 14.93.
Found: C, 54.22; H, 4.75; N, 15.05.
Test organism and medium. A genotype I, noncytopathic strain of BVDV
(SD-1) was used for the determination of viral inhibition (13). Stock virus,
initially isolated from the serum of a persistently infected cow, was propagated in
BVDV-free Madin Darby bovine kidney (MDBK) cells cultured in minimum
essential medium with Earle’s salts supplemented with 10% (vol/vol) equine
serum, 0.75 mg of sodium bicarbonate/ml, 0.29 mg of L-glutamine/ml, and
well was then read with
a SPECTRAFluor Plus microplate fluorometer
(TECAN) at 620 and 450 nm by using the 405/450/492/620-nm absorbance filter.
The amount of reduced XTT was calculated by subtracting the sample’s OD at
620 nm (OD620; turbidity) from its OD450 (OD450 Ϫ OD620 ϭ OD due to XTT

VOL. 47, 2003
DETECTION OF INHIBITION OF BVDV
TABLE 1. Structures and in vitro activities of cationic diaryl 5-membered ring compounds
2225
BVDV remaining (% of controlb) at indicated
concn of compound
Molecule ata:
W
Compound no.
25 ␮M
Cell lysates
5 ␮M
X
R
Y
Medium
Medium
Cell lysates
c
DB18
S
H
Am
Im
Am
Am
O-MeAm
OCH3
Am
—
—
—
—
—
180
31
—
—
DB60
O
O
O
O
O
O
O
H
Im
—
Ͻ0.001
0.6
180
—
—
3.2
DB75
DB99
DB289
DB606
DB777
DB784
H
Am
Am
O-MeAm
Im
NHAm
Ͻ0.001
0.3
8
CH3
H
32
18
Ͻ0.001
31
18
32
180
Ͻ0.001
10
H
—
H
OCH3
OCH3
—
H
NH(CANH)-2-pyridyl
1.8
32
a Am, C(ANH)NH2; O-MeAm, C(ANOMe)NH2.
b Percentage of control ϭ (CCID50 of BVDV in compound sample)/(CCID50 of BVDV in control sample lacking compound).
c —, cytotoxic effect of compound observed after 72 h.
reduction). This final OD value was calculated for the cell and medium controls,
as well as for the compound assays.
RESULTS
Statistical calculations. The CCID50 of BVDV in cell culture medium and cell
Screening assays. During screening assays at both 25 and
5 ␮M concentrations, 4 of 93 molecules were visibly toxic to
MDBK cells in the exponential growth phase (Tables 1 through
5). At a concentration of only 25 ␮M, eight additional mole-
cules were visibly toxic to the MDBK cells. Screening assays
identified 5 of 93 aromatic cationic molecules that were select-
ed for determination of CC50, CC10, IC90, and IC50 for BVDV
in cell culture (Fig. 1).
Analysis of structure-activity relationships: cationic diaryl
five-membered ring compounds. As a group, the cationic diaryl
five-membered ring compounds were cytotoxic in this assay
(Table 1) compared to other groups of compounds. The inclu-
sion of oxygen rather than sulfur within the central five-mem-
bered ring was noted to decrease cytotoxicity (compare results
lysate samples were determined by the statistical method of Reed and Muench
(33). BVDV in cell culture medium and cell lysate samples was evaluated by
comparison to equivalent samples from temporal control cultures in which no
compound was added before or after inoculation [percentage of control ϭ
(CCID50 of BVDV in the compound sample)/(CCID50 of BVDV in the control
sample lacking the compound)]. The IC90 and IC50 of selected compounds were
calculated with JMP software by least-squares regression techniques using the
logarithm of the percentage of the control for each sample (34).
The toxicity of each compound at each concentration was calculated as a percent-
age of the untreated cell control by first subtracting the average final OD of the
medium control wells from the final OD values of the cell controls and drug assays.
The drug assay values were then compared to the untreated cell control average
value for each plate and evaluated as percentage of control, with 100% being
equivalent to the cell control value [(drug assay values/average cell control value) ϫ
100 ϭ percentage of untreated cell control]. The CC50 and CC10 of selected com-
pounds were calculated with JMP software by least-squares regression techniques
using the square of the percentage of control of each sample (34).
TABLE 2. Structures and in vitro activities of dicationic guanidino compounds
Molecule at:
BVDV remaining (% of controlb) at indicated concn of compound
25 ␮M 5 ␮M
Compound no.
X
R
R1
Medium
Cell lysates
Medium
Cell lysates
DB701
DB711
DB748
DB752
DB762
DB763
DB778
DB779
O
O
O
S
O
O
O
O
CF3
OCH3
CH3
CH3
OEt
H
H
H
CH3
H
H
OCH3
H
OEt
0.2
7
0.6
320
NT
NT
NT
NT
120
2
NT
NT
NT
NT
900
9
31
320
—
21
Ͻ0.001
1.8
Ͻ0.001
—
28
Ͻ0.001
3
Ͻ0.001
Oi-Pr
H
31
56
1.8
0.6
a NT, not tested.
b See Table 1, footnotes b and c, for details.

2226
GIVENS ET AL.
ANTIMICROB. AGENTS CHEMOTHER.
TABLE 3. Structures and in vitro activities of cationic benzimidazole-substituted aryl furansa
Molecule atb:
BVDV remaining (% of controld) at indicated concn of compound
25 ␮M 5 ␮M
Compound no.
X
Y
Medium
Cell lysates
Medium
Cell lysates
DB293
DB294
DB302
DB456
DB457
DB458
DB459
DB501
DB771
DB772
DB805
Am
i-PrAm
Im
NH2
NO2
NO2
NH2
NH2
Am
Am
i-PrAm
Im
Im
Im
i-PrAm
i-PrAm
Am
10
—
1,000
0.1
0.003
—
Ͻ0.001
27
0.2
—
57
57
100
0.2
NT
NT
100
47
Ͻ0.001
Ͻ0.001
—
100
56
32
56
Ͻ0.001
NT
NT
10
100
Ͻ0.001
Ͻ0.001
—
2
0.006
—
0.002
6
Ͻ0.001
Ͻ0.001
—
H
H
Ͻ0.001
Ͻ0.001
—
Im
NO2
c-PrAm
a For structure of Im, see Table 1.
b Am, C(ANH)NH2, i-PrAm, C(ANH)NH-i-Pr; c-PrAm, C(ANH)NH-c-Pr.
c NT, not tested.
d See Table 1, footnotes b and c, for details.
for DB18 and DB75), while the addition of methyl substituents
to the central five-membered ring decreased antiviral efficacy.
From this chemical group, DB606 was selected for further
antiviral characterization.
Dicationic guanidino compounds. As a group, the dicationic
guanidino compounds also exhibited a lack of toxicity when
oxygen was substituted for sulfur within the central five-mem-
bered ring (Table 2). While two compounds within this group,
DB763 and DB779, exhibited significant antiviral effects, they
were not selected for further testing due to the identification of
compounds in other chemical groups that were more effica-
cious at the 5 ␮M concentration.
Cationic benzimidazole-substituted aryl furans. Within this
chemical group, several compounds exhibited effective inhibi-
tion of viral growth at 5 ␮M concentrations without evidence
of cytotoxic effect at 25 ␮M concentrations (Table 3). Com-
pounds DB456, DB459, DB771, and DB772 were selected for
further antiviral characterization.
Cationic aryl benzimidazole compounds. This group of com-
pounds exhibited limited antiviral effects at 5 ␮M concentra-
TABLE 4. Structures and in vitro activities of cationic aryl benzimidazole compounds
Molecule ata:
BVDV remaining (% of controlc) at indicated concn of compound
Compound no.
25 ␮M
5 ␮M
X
Y
Medium
Cell lysates
Medium
Cell lysates
DB682
DB683
DB684
DB685
DB759
DB787
DB788
DB789
DB790
DB791
DB792
DB793
DB794
DB795
DB796
DB797
5-(p-NH2C6H4)furan-2yl
5-(p-NH2C6H4)furan-2yl
5-(p-NH2C6H4)furan-2yl
5-(p-NO2C6H4)furan-2yl
p-MeOC6H4CH2CH2
p-MeOC6H4CH2CH2
p-MeOC6H4CH2CH2
p-MeOC6H4CH2CH2
p-EtC6H4CH2CH2
Am
—
—
0.007
—
—
NT
NT
NT
NT
NT
NT
NT
NT
NT
NT
NT
—
—
0.03
—
—
NT
NT
NT
NT
NT
NT
NT
NT
NT
NT
NT
—
8
0.8
0.8
21
—
100
31
180
280
10
Im
i-PrAm
THP
Im
Am
320
180
100
56
i-PrAm
c-pentyl
Am
100
18
100
10
p-EtC6H4CH2CH2
Im
83
p-EtC6H4CH2CH2
i-PrAm
c-pentyl
Am
56
18
p-EtC6H4CH2CH2
56
57
p-FC6H4CH2CH2
180
32
10
p-FC6H4CH2CH2
Im
i-PrAm
c-pentyl
6
6
p-FC6H4CH2CH2
100
56
p-FC6H4CH2CH2
18
a Am ϭ C(ANH)NH2, i-PrAm ϭ C(ANH)NH-i-Pr. For structure of Im, see Table 1.
b NT, not tested.
c See Table 1, footnotes b and c, for details.

VOL. 47, 2003
DETECTION OF INHIBITION OF BVDV
2227
TABLE 5. Structures and in vitro activities of cationic 4H-1-benzopyran-4-one substituted benzimidazoles
Molecule ata:
Y
BVDV remaining (% of controlb) at indicated concn of compound
25 ␮M 5 ␮M
Compound no.
X
L
Medium
Cell lysates
Medium
Cell lysates
DB636
DB637
DB638
DB640
DB737
DB738
Ph
Ph
H
H
OCH3
H
Am
i-PrAm
Am
i-PrAm
i-PrAm
i-PrAm
Nil
Nil
36
360
280
910
560
160
910
28
20
31
650
120
2,400
65
90
91
160
280
160
28
1,4-phenylene
1,4-phenylene
1,4-phenylene
2,5-dithienyl
100
2,400
2,400
36
a Am ϭ C(ANH)NH2, i-PrAm ϭ C(ANH)NH-i-Pr.
b See Table 1, footnotes b and c, for details.
tions (Table 4). Only agents with significant antiviral effects at
5 ␮M were screened for effects at the 25 ␮M concentration. Of
those agents screened at the higher concentration, four of five
compounds exhibited cytotoxicity.
hibition of the viral life cycle. This screening method allowed
the timely detection of compounds that exhibited a lack of in
vitro cytotoxicity but that effectively inhibited the viral life
cycle.
Cationic 4H-1-benzopyran-4-one-substituted benzimid-
azoles. The cationic 4H-1-benzopyran-4-one-substituted benz-
imidazoles exhibited no cytotoxicity and no significant antiviral
effects (Table 5).
The screening method described in this report is an appro-
priate precursor to in vivo testing of the leading compounds for
prevention or treatment of BVDV infections. Regardless of
their therapeutic effects in vivo, compounds that exhibit non-
toxic inhibition of the viral life cycle might be used to reduce
the risks of contamination of vaccines and biologics produced
in cell culture (45). If the leading compounds are effective
against BVDV in vivo, their use during critical time periods
might reduce the need to use antibiotics for treatment of the
bacterial infections that are sequelae of the pulmonary tissue
damage produced by BVDV.
The compounds selected by this screening method, particu-
larly DB771 and DB772, are strong candidates for further
testing to determine their efficacy against HCV. It is notewor-
thy that the most effective compounds identified by this screen-
ing method are monocations, which are significantly more po-
tent than related dicationic molecules. Due to the inability to
grow HCV effectively in cell culture, current investigations of
potential therapeutic agents involve in vitro subgenomic HCV
replicon assays with selected compounds advanced into a chim-
panzee animal model or a transgenic mouse model for HCV
infection (28). The limited availability and significant expense
of the chimpanzee animal model and the transgenic mouse
model require that in vitro screening be performed to identify
the leading compounds. Subgenomic HCV replicon assays con-
sist of transfecting permissive cells with a subset of the viral
RNA which can be translated into nonstructural proteins that
replicate the positive-sense viral RNA (5, 29). Thus, the rep-
licon assay is limited to detection of antiviral agents that inhibit
binding of the 5Јnontranslated region of the HCV genome to
cellular ribosomes, translation of the viral RNA, or replication
of the viral RNA within the cytoplasm. The screening method
described in this research has the potential to identify com-
pounds that exhibit a mechanism of action beyond the scope of
the subgenomic replicon assay. As the RNA-dependent RNA
polymerase of HCV lacks proofreading ability and exhibits a
Compounds selected for IC90, IC50, CC50, and CC10 end-
point determinations. Five agents were evaluated at multiple
dilutions for determination of CC50 and CC10 by the XTT
assay and for determination of IC90 and IC50 for BVDV in cell
culture (Fig. 1). Of these agents, DB606 exhibited the greatest
cytotoxicity at the 25 ␮M concentration. The IC90 results
ranged from 15.9 (DB456) to 0.018 (DB772) ␮M concentra-
tions. The IC50 results ranged from 13.6 (DB456) to 0.014
(DB772) ␮M concentrations. Agent DB772 exhibited IC90s at
18 and 20 nM concentrations, but did not exhibit toxicity to
transformed cells in exponential growth phase at a concentra-
tion of 25 ␮M (Fig. 1).
DISCUSSION
As BVDV is an economically significant pathogen of cattle
(23), a problematic contaminant in the laboratory (18, 21), and
a viral model for HCV to facilitate in vitro drug discovery
efforts (46), the goal of this research was to screen aromatic
cationic molecules for in vitro toxicity and activity against a
noncytopathic strain of BVDV. The use of a noncytopathic
rather than a cytopathic strain of BVDV permitted evaluation
of the cytotoxic effects of chemical compounds without the
confounding cytopathic effects of the virus. This choice of viral
biotypes did increase the labor necessary to assess antiviral
efficacy, as immunoperoxidase labeling with BVDV-specific
antibodies was required for viral detection and quantitation.
As cleavage of the nonstructural NS2/3 viral protein is the
source of cytopathogenicity (25), compounds that prevent this
cleavage could prevent cytopathic effects without altering viral
attachment, infection, replication, packaging, or release. Thus,
the use of a noncytopathic strain of BVDV prevented the
inhibition of NS2/3 cleavage from being misinterpreted as in-

2228
GIVENS ET AL.
ANTIMICROB. AGENTS CHEMOTHER.
FIG. 1. Curve fit analysis for cytotoxic concentrations and BVDV inhibitory concentrations in cell culture of selected aromatic cationic
molecules.

VOL. 47, 2003
DETECTION OF INHIBITION OF BVDV
2229
mutation rate of approximately 5 ϫ 10Ϫ3/site per year (9), a
number of antiviral agents directed at multiple targets will be
required to reduce or eliminate HCV quasispecies that have
developed resistance to other drugs (5). If the replicon assay is
used as the sole determinant in selecting HCV antiviral can-
didates, then compounds with antiviral targets involving viral
attachment and entry into cells, packaging of viral RNA within
the virion, or viral release will be regrettably overlooked. We
believe that the screening methods described in this research
complement the subgenomic HCV replicon assay and provide
utility for the identification of compounds or mechanisms of
action that might effectively inhibit HCV infection and repli-
cation.
11. Das, B. P., and D. W. Boykin. 1976. Synthesis and antiprotozoal activity of
2,5-bis(4-guanylphenyl)furans. J. Med. Chem. 20:531–536.
12. Das, B. P., and D. W. Boykin. 1977. Synthesis and antiprotozoal activity of
2,5-bis(4-guanylphenyl)thiophenes and -pyrroles. J. Med. Chem. 20:1219–
1221.
13. Deng, R., and K. V. Brock. 1992. Molecular cloning and nucleotide sequence
of a pestivirus genome, noncytopathic bovine viral diarrhea virus strain SD-1.
Virology 191:867–879.
14. Dubovi, E. J., J. D. Geratz, and R. R. Tidwell. 1980. Inhibition of respiratory
syncytial virus by bis(5-amidino-2-benzimidazolyl)methane. Virology 103:
502–509.
15. Ellis, J. A., K. H. West, V. S. Cortese, S. L. Myers, S. Carman, K. M. Martin,
and D. M. Haines. 1998. Lesions and distribution of viral antigen following
an experimental infection of young seronegative calves with virulent bovine
virus diarrhea virus-type II. Can. J. Vet. Res. 62:161–169.
16. Fernelius, A. L., G. Lambert, and G. J. Hemness. 1969. Bovine viral diarrhea
virus-host cell interactions: adaptation and growth of virus in cell lines.
Am. J. Vet. Res. 30:1561–1572.
The mechanisms of action of the antiviral agents identified
in this research are yet to be determined. Since related com-
pounds have been identified as protease inhibitors (14, 39) and
helicase inhibitors (32) and have been shown to be effective in
binding to nucleic acids (24, 40), multiple mechanisms of ac-
tion might be employed by the nontoxic antiviral compounds
reported here. Mechanisms of action of other anti-BVDV
agents include inhibition of viral RNA-dependent RNA poly-
merase (2) and of ␣-glucosidase, which results in the misfold-
ing of viral proteins (30). Further research will likely identify
additional targets for compounds that inhibit replication of
pestiviruses and HCV.
In conclusion, a method of in vitro screening with a noncy-
topathic strain of BVDV enabled the identification of aromatic
cationic molecules that inhibit the BVDV life cycle. These
leading compounds, which inhibit BVDV at nanomolar con-
centrations and exhibit limited cytotoxicity at 25 ␮M, merit
further investigation to determine their mechanisms of action,
in vivo efficacies, and specific activities against HCV.
17. Giangaspero, M., G. Vacirca, M. Buettner, G. Wolf, E. Vanopdenbosch, and
G. Muyldermans. 1993. Serological and antigenical findings indicating pes-
tivirus in man. Arch. Virol. Suppl. 7:53–62.
18. Giangaspero, M., G. Vacirca, R. Harasawa, M. Buttner, A. Panuccio, C. De
Giuli Morghen, A. Zanetti, A. Belloli, and A. Verhulst. 2001. Genotypes of
pestivirus RNA detected in live virus vaccines for human use. J. Vet. Med.
Sci. 63:723–733.
19. Harasawa, R., and H. Mizusawa. 1995. Demonstration and genotyping of
pestivirus RNA from mammalian cell lines. Microbiol. Immunol. 39:979–
985.
20. Harasawa, R., and T. Sasaki. 1995. Sequence analysis of the 5Ј untranslated
region of pestivirus RNA demonstrated in interferons for human use. Bio-
logicals 23:263–269.
21. Harasawa, R., and T. Tomiyama. 1994. Evidence of pestivirus RNA in
human virus vaccines. J. Clin. Microbiol. 32:1604–1605.
22. Hopkins, K. T., W. D. Wilson, B. C. Bender, D. R. McCurdy, J. E. Hall, R. R.
Tidwell, A. Kumar, M. Bajic, and D. W. Boykin. 1998. Extended aromatic
furan amidino derivatives as anti-Pneumocystis carinii agents. J. Med. Chem.
41:3872–3878.
23. Houe, H. 1999. Epidemiological features and economical importance of
bovine virus diarrhoea virus (BVDV) infections. Vet. Microbiol. 64:89–107.
24. Kumar, A., R. A. Rhodes, J. Spychala, W. D. Wilson, D. W. Boykin, R. R.
Tidwell, C. C. Dykstra, J. E. Hall, S. K. Jones, and R. F. Schinazi. 1995.
Synthesis of dicationic diarylpyridines as nucleic-acid binding agents. Eur.
J. Med. Chem. 30:99–106.
25. Kummerer, B. M., N. Tautz, P. Becher, H. Thiel, and G. Meyers. 2000. The
genetic basis for cytopathogenicity of pestiviruses. Vet. Microbiol. 77:117–128.
26. Lansiaux, A., L. Dassonneville, M. Facompre, A. Kumar, C. E. Stephens, M.
Bajic, F. Tanious, W. D. Wilson, D. W. Boykin, and C. Bailly. 2002. Distri-
bution of furamidine analogues in tumor cells: influence of the number of
positive charges. J. Med. Chem. 45:1994–2002.
ACKNOWLEDGMENTS
This work was supported by an NIH development grant (NIAID
AI-01728-01) and Immtech International, Inc., Vernon Hills, Ill.
27. Li, K., M. Fernandez-Saiz, C. T. Rigl, A. Kumar, K. G. Ragunathan, A. W.
McConnaughie, D. W. Boykin, H. J. Schneider, and W. D. Wilson. 1997.
Design and analysis of molecular motifs for specific recognition of RNA.
Bioorg. Med. Chem. 5:1157–1172.
28. Locarnini, S. A., and A. Bartholomeusz. 2002. Advances in hepatitis C: what
is coming in the next 5 years? J. Gastroenterol. Hepatol. 17:442–447.
29. Lohmann, V., F. Korner, J. Koch, U. Herian, L. Theilmann, and R. Barten-
schlager. 1999. Replication of subgenomic hepatitis C virus RNAs in a
hepatoma cell line. Science 285:110–113.
REFERENCES
1. Afshar, A., G. C. Dulac, C. Dubuc, and T. H. Howard. 1991. Comparative
evaluation of the fluorescent antibody test and microtiter immunoperoxidase
assay for detection of bovine viral diarrhea virus from bull semen. Can. J.
Vet. Res. 55:91–93.
2. Baginski, S. G., D. C. Pevear, M. Seipel, S. C. Sun, C. A. Benetatos, S. K.
Chunduru, C. M. Rice, and M. S. Collett. 2000. Mechanism of action of a
pestivirus antiviral compound. Proc. Natl. Acad. Sci. USA 97:7981–7986.
3. Baker, J. C. 1987. Bovine viral diarrhea virus: a review. J. Am. Vet. Med.
Assoc. 190:1449–1458.
30. Mehta, A., N. Zitzmann, P. M. Rudd, T. M. Block, and R. A. Dwek. 1998.
Alpha-glucosidase inhibitors as potential broad based anti-viral agents.
FEBS Lett. 430:17–22.
31. Pellerin, C., J. van den Hurk, J. Lecomte, and P. Tijssen. 1994. Identification
of a new group of bovine viral diarrhea virus strains associated with severe
outbreaks and high mortalities. Virology 203:260–268.
4. Behrens, S.-E., C. W. Grassmann, H.-J. Thiel, G. Meyers, and N. Tautz.
1998. Characterization of an autonomous subgenomic pestivirus RNA rep-
licon. J. Virol. 72:2364–2372.
32. Phoon, C. W., P. Yon Ng, A. E. Ting, S. L. Yeo, and M. M. Sim. 2001.
Biological evaluation of hepatitis C virus helicase inhibitors. Bioorg. Med.
Chem. Lett. 11:1647–1650.
33. Reed, L. J., and H. Muench. 1938. A simple method of estimating fifty
percent end points. Am. J. Hyg. 27:493–497.
5. Blight, K. J., A. A. Kolykhalov, and C. M. Rice. 2000. Efficient initiation of
HCV RNA replication in cell culture. Science 290:1972–1974.
6. Bolin, S. R., P. J. Matthews, and J. F. Ridpath. 1991. Methods for detection
and frequency of contamination of fetal calf serum with bovine viral diarrhea
virus and antibodies against bovine viral diarrhea virus. J. Vet. Diagn. In-
vestig. 3:199–203.
7. Boykin, D. W., A. Kumar, B. K. Bender, J. E. Hall, and R. R. Tidwell. 1996.
Anti-Pneumocystis activity of bis-amidoximes and bis-O-alkylamidoximes
prodrugs. Bioorg. Med. Chem. Lett. 6:3017–3020.
8. Carman, S., T. van Dreumel, J. Ridpath, M. Hazlett, D. Alves, E. Dubovi, R.
Tremblay, S. Bolin, A. Godkin, and N. Anderson. 1998. Severe acute bovine
viral diarrhea in Ontario, 1993–1995. J. Vet. Diagn. Investig. 10:27–35.
9. Christie, J. M., H. Chapel, R. W. Chapman, and W. M. Rosenberg. 1999.
Immune selection and genetic sequence variation in core and envelope
regions of hepatitis C virus. Hepatology 30:1037–1044.
10. Corapi, W. V., R. O. Donis, and E. J. Dubovi. 1990. Characterization of a
panel of monoclonal antibodies and their use in the study of the antigenic
diversity of bovine viral diarrhea virus. Am. J. Vet. Res. 51:1388–1394.
34. Sall, J., and A. Lehman (ed.). 1996. Fitting curves through points: regression,
p. 171–193. In JMP start statistics. Duxbury Press, Belmont, Calif.
35. Stephens, C. E., F. Tanious, S. Kim, W. D. Wilson, W. A. Schell, J. R. Perfect,
S. G. Franzblau, and D. W. Boykin. 2001. Diguanidino and “reversed”
diamidino 2,5-diarylfurans as antimicrobial agents. J. Med. Chem. 44:1741–
1748.
36. van Oirschot, J. T., C. J. Bruschke, and P. A. van Rijn. 1999. Vaccination of
cattle against bovine viral diarrhoea. Vet. Microbiol. 64:169–183.
37. Varenikova, S. F., A. A. Nekhoroshev, and D. P. Sevbo. 1985. Methoxylation
of 3,4-dinitrochloro(bromo)benzene. J. Org. Chem. USSR 21:1651–1652.
38. Vickers, M. L., and H. C. Minocha. 1990. Diagnosis of bovine viral diarrhea
virus infection using monoclonal antibodies. J. Vet. Diagn. Investig. 2:300–
302.

2230
GIVENS ET AL.
ANTIMICROB. AGENTS CHEMOTHER.
39. Vonderfecht, S. L., R. L. Miskuff, S. B. Wee, S. Sato, R. R. Tidwell, J. D.
Geratz, and R. H. Yolken. 1988. Protease inhibitors suppress the in vitro and
in vivo replication of rotavirus. J. Clin. Investig. 82:2011–2016.
40. Wang, L., C. Bailly, A. Kumar, D. Ding, M. Bajic, D. W. Boykin, and W. D.
Wilson. 2000. Specific molecular recognition of mixed nucleic acid se-
quences: an aromatic dication that binds in the DNA minor groove as a
dimer. Proc. Natl. Acad. Sci. USA 97:12–16.
41. Wang, L., C. Carrasco, A. Kumar, C. E. Stephens, C. Bailly, D. W. Boykin,
and W. D. Wilson. 2001. Evaluation of the influence of compound structure
on stacked-dimer formation in the DNA minor groove. Biochemistry 40:
2511–2521.
43. Xue, W., F. Blecha, and H. C. Minocha. 1990. Antigenic variations in bovine
viral diarrhea viruses detected by monoclonal antibodies. J. Clin. Microbiol.
28:1688–1693.
44. Xue, W., S. Zhang, and H. C. Minocha. 1997. Characterization of a putative
receptor protein for bovine viral diarrhea virus. Vet. Microbiol. 57:105–118.
45. Yanagi, M., J. Bukh, S. U. Emerson, and R. H. Purcell. 1996. Contamination
of commercially available fetal bovine sera with bovine viral diarrhea virus
genomes: implications for the study of hepatitis C virus in cell cultures.
J. Infect. Dis. 174:1324–1327.
46. Zitzmann, N., A. S. Mehta, S. Carrouee, T. D. Butters, F. M. Platt, J.
McCauley, B. S. Blumberg, R. A. Dwek, and T. M. Block. 1999. Imino sugars
inhibit the formation and secretion of bovine viral diarrhea virus, a pestivirus
model of hepatitis C virus: implications for the development of broad spec-
trum anti-hepatitis virus agents. Proc. Natl. Acad. Sci. USA 96:11878–11882.
42. Weislow, O. S., R. Kiser, D. L. Fine, J. Bader, R. H. Shoemaker, and M. R.
Boyd. 1989. New soluble-formazan assay for HIV-1 cytopathic effects: ap-
plication to high-flux screening of synthetic and natural products for AIDS-
antiviral activity. J. Natl. Cancer Inst. 81:577–586.