Chemistry & Biology, Vol. 12, 303–311, March, 2005, ©2005 Elsevier Ltd All rights reserved. DOI 10.1016/j.chembiol.2005.01.006
The Adamantane-Derived Bananins Are Potent
Inhibitors of the Helicase Activities
and Replication of SARS Coronavirus
Julian A. Tanner,1,6Bo-Jian Zheng,2,6Jie Zhou,2
Rory M. Watt,1,3Jie-Qing Jiang,1Kin-Ling Wong,2
Yong-Ping Lin,2Lin-Yu Lu,1Ming-Liang He,4
Hsiang-Fu Kung,4Andreas J. Kesel,5,*
and Jian-Dong Huang1,*
1Department of Biochemistry
2Department of Microbiology
3Department of Chemistry and
Open Laboratory of Chemical Biology
University of Hong Kong, Pokfulam
Hong Kong, China
4Center for Emerging Infectious Diseases
Faculty of Medicine
The Chinese University of Hong Kong
Hong Kong, China
ment of Parkinson’s disease . Other drugs include
rimantadine (against influenza A) , tromantadine
(against herpes simplex virus),  and memantine
(N-methyl-D-aspartate [NMDA] receptor antagonist) .
In order to add potential cytoprotective functionality ,
oligo-oxa-adamantanes have recently been conjugated
to vitamin B6 (pyridoxal) to create a new class of ada-
mantanes—the bananins (Figure 1) . We have synthe-
sized six bananin derivatives—bananin (BAN), iodoba-
nanin (IBN), adeninobananin (ADN), vanillinbananin (VBN),
eubananin (EUB), and ansabananin (ABN). The synthe-
sis of bananin was described , the synthesis of IBN
and ADN will be reported elsewhere (A.J.K., submitted),
and the final three are described in this report.
Severe acute respiratory syndrome (SARS) is caused
by infection with the SARS coronavirus (SCV) [10–12].
SCV was rapidly sequenced following its identification
[13, 14], leading to the recognition of a number of pos-
sible drug targets. Although treatment with ribavirin and
corticosteroids has been shown to have a slight posi-
tive effect , side-effects  and lack of activity of
ribavirin in cell culture  highlight the need for more
effective treatments. Most recent anti-SCV drug devel-
opment has targeted the viral main protease, also
called the 3CL protease, and following an initial crystal
structure , a structure together with inhibitors was
solved . Other classes of boronic-acid-based prote-
ase inhibitors have also been identified , and large-
scale screens have identified SCV inhibitors effective
against the protease . We and collaborators have
also recently identified inhibitors of the SCV helicase,
the protease and spike-mediated viral entry by a chem-
ical genetics approach . In this paper we focus
specifically on the SCV NTPase/helicase. Drugs target-
ing viral helicases have had marked success in animal
models of herpes simplex virus [23, 24], and there has
been progress in targeting the hepatatis C virus heli-
case . We have previously cloned, purified, and per-
formed an initial biochemical characterization of the
SCV helicase ; our results showed that the helicase
exhibited strict 5# to 3# polarity, consistent with con-
temporary reports from another group . It has also
been shown recently that the SCV helicase possesses
an RNA 5#-triphosphatase activity that may be involved
in capping viral RNA .
The SCV helicase consists of three major domains—
a putative N-terminal metal binding domain (MBD), a
hinge domain, and an NTPase/helicase domain. It is
clear from previous work on another member of the Ni-
dovirales order, equine arterivirus (EAV), that the metal
binding domain is essential to viral viability [29, 30]—
this is also likely to be the case for closely related SCV.
In this paper, we investigate the effects of the bananin
series of compounds on both the ATPase and helicase
activities of the SCV helicase, and also test bananin
against SCV in cell culture. We find that bananin and
three of its derivatives are potent inhibitors of both the
ATPase and helicase activities of the SCV helicase, and
that bananin inhibits SCV replication at a concentration
significantly below that at which it is toxic to the cell.
Bananins are a class of antiviral compounds with a
unique structural signature incorporating a trioxa-
adamantane moiety covalently bound to a pyridoxal
derivative. Six members of this class of compounds:
bananin, iodobananin, vanillinbananin, ansabananin,
eubananin, and adeninobananin were synthesized and
tested as inhibitors of the SARS Coronavirus (SCV)
helicase. Bananin, iodobananin, vanillinbananin, and
eubananin were effective inhibitors of the ATPase ac-
tivity of the SCV helicase with IC50values in the range
0.5–3 ?M. A similar trend, though at slightly higher
inhibitor concentrations, was observed for inhibition
of the helicase activities, using a FRET-based fluores-
cent assay. In a cell culture system of SCV, bananin
exhibited an EC50of less than 10 ?M and a CC50of
over 300 ?M. Kinetics of inhibition are consistent with
bananin inhibiting an intracellular process or pro-
cesses involved in SCV replication.
decanes, are structurally unusual compounds where
four cyclohexane rings are fused to each other in a par-
ticularly strain-free, all chair conformation. Oligo-oxa-
adamantanes contain oxygens in place of methylene
linkages within the structure and play a variety of roles
in nature; from the infamous neurotoxin from the puffer
fish tora fugu, tetrodotoxin (TTX) (Figure 1) , to plant
natural product steroids such as daigremontianin .
Not all adamantane derivatives are extremely toxic, and
many synthetic derivatives such as amantadine are
used clinically (Figure 1) . Amantadine is used as an
antiviral agent , and as a muscle relaxant in the treat-
6These authors contributed equally to this work.
Chemistry & Biology
Figure 1. Chemical Stuctures of the Six Syn-
thetic Bananin Derivatives (Bananin, Iodoba-
nanin, Adeninobananin, Vanillinbananin, Eu-
bananin, and Ansabananin), the Anti-Influenza
and Anti-Parkinson’s Drug Amantadine, and
the Fugu Fish Toxin Tetrodotoxin
Further cell culture studies suggest that bananin inhib-
its intracellular activities mechanistically involved with
key viral processes, as opposed to the viral entry step.
These results are consistent with this class of drugs
targeting the SCV helicase within the cell.
our knowledge, an array of completely new adaman-
tane derivatives, which may be easily diversified by re-
acting various aromatic aldehydes with phloroglucinol.
Inhibition of SCV Enzymatic Activity
We have previously developed a colorimetric assay to
measure the NTPase activity of the SCV helicase in 96-
well plates, in a high throughput format [26, 32, 33]. In
this discontinuous colorimetric assay using malachite
green and ammonium molybdate, released phosphate
is quantified after a 5 min reaction period, observed at
a wavelength of 630 nm. Oligo-dT24was included in the
assay at a saturating concentration of 200 nM, to mimic
the nucleic acid-stimulated NTPase activity of the SCV
helicase. Potential inhibitors of the ATPase reaction
would be expected to reduce the amount of phosphate
released during the reaction, reflected in a decrease in
the measured absorbance at 630 nm.
We first checked whether the bananin compounds
were able to inhibit the dT24-stimulated ATPase activity
of the SCV helicase. Controls were carried out to en-
sure that the bananin compounds themselves did not
affect the phosphate measurement assay. Reactions
were carried out in the presence of various concentra-
tions of the six bananin derivatives and the results were
plotted and fitted to a simple model (Figure 2A). We
also checked that under our reaction conditions, we
were making measurements within the linear region
Our results showed that the parent compound ba-
nanin inhibited the ATPase activity of the SCV helicase
with anATPaseIC50value of 2.3 ?M (IC50values shown
in Figure 4C). Iodobananin and vanillinbananin exhib-
ited the strongest inhibition, withATPaseIC50values of
0.54 and 0.68 ?M, respectively. Inhibition by vanillinba-
nanin indicates that the presence of a six-membered
nitrogen heterocycle is not absolutely essential for in-
hibitory activity. Eubananin showed similar inhibitory
activity to bananin itself with anATPaseIC50of 2.8 ?M.
Interestingly, ansabananin was a weak inhibitor, with an
ATPaseIC50of 51 ?M, while adeninobananin did not show
any inhibitory activity at all. These results suggest that
Synthesis of Bananin and Its Derivatives
The bananins were synthesized by the reaction of
phloroglucinol (most likely in its triketo tautomeric form)
with aromatic aldehydes, catalyzed by hydrochloric
acid or sodium hydroxide in aqueous solution . Gen-
erally, acidic catalysis was used due to the degradation
of pyridoxal under highly basic conditions. Alkaline ca-
talysis was used for reaction with aromatic aldehydes
such as vanillin. Bananin synthesis is driven by the cre-
ation of the highly symmetric trioxa-adamantane-triol
(TAT) cage system. The prototypical compound of the
TAT series, the vitamin B6-derived bananin (BAN) or
2,8,9-trioxaadamantane-3,5,7-triol, can be iodinated
with subsequent oxidation to iodobananin (6#-iodoba-
nanin 5#-carboxylic acid, IBN). The iodine in IBN can be
replaced by various substituents as exemplified by the
synthesis of adeninobananin (6#-adeninobananin 5#-
carboxylic acid hydrochloride, ADN) using an activated
adenine nucleobase derivative. Interestingly, BAN is
susceptible to Michael addition with the natural pro-
duct eugenol, isolated from the essential oil of cloves
(Syzygium aromaticum). This NaOH-catalyzed addition
leads to eugenolbananin (eubananin, EUB), which can
be transformed by cyclic hemiketal condensation into
the ansa-compound ansabananin (ABN), inspired by
ansamycins such as rifamycin and geldanamycin .
In the aromatic aldehyde series, vanillin was reacted
with phloroglucinol to yield vanillinbananin (VBN) or
tane-3,5,7-triol. It is expected that numerous naturally
occurring aldehydes can be introduced to form the cor-
responding TATs with phloroglucinol in 3.33 M aqueous
NaOH. The bananin group of compounds represents, to
Bananins As Inhibitors of the SARS Coronavirus
are the double reciprocal Lineweaver Burke plots for
the data in Figures 3A and 3C. In both cases, as the
Vmaxwas significantly decreased in the presence of the
inhibitor, but the KMchanged little, this indicated that
bananin was acting as a noncompetitive inhibitor of the
ATPase activity of the SCV helicase with respect to
both ATP and nucleic acid. This suggests that bananin
inhibits by binding at a site distinct from the ATP and
nucleic acid binding sites.
Building on this foundation, we next tested the anti-
helicase activities of these compounds. We used a
newly developed fluorimetric assay based on the very
strong fluorescence resonance energy transfer (FRET)
from the fluorophore Cy3 to the quencher Black Hole
Quencher 2 (BHQ2). A similar approach has been out-
lined very recently in assaying the hepatitis C virus
(HCV) helicase, a 3# to 5# helicase . However, as we
have recently shown that the SCV helicase holds strict
5# to 3# directionality , we designed a system with
a 5#-oligo(dT) overhang. The principle behind this new
assay is outlined in Figure 4A. There is a Cy3 fluoro-
phore at the 3# end of one of the oligomers of the du-
plex, in close proximity to a BHQ2 quencher at the 5#
end of the other oligomer. When the two oligomers are
in very close proximity (i.e., when the two oligomers
are annealed), then the Cy3 fluorescence is strongly
quenched by the FRET effect. After the duplex has
been unwound by the SCV helicase, then the Cy3 fluo-
rescence is no longer quenched, and a dramatic
increase in the fluorescence may be observed. To en-
sure that the primers do not reanneal, a second capture
primer is included in the reaction. This is identical to the
BHQ2 primer but does not contain the BHQ2 quenching
group, therefore the annealing process has little effect
on the fluorescence of Cy3. We optimized reaction con-
ditions to ensure that all measurements were carried
out in the linear region (Figure S2). As a one minute
time-point was in the linear region, we then probed the
effect of the presence of various concentrations of the
bananin inhibitors and compared them to control reac-
tions where no inhibitor was added (Figure 4B). Addi-
tional controls were carried out to verify that the ba-
nanin compounds did not fluoresce themselves at the
wavelength at which we were reading. Data from these
experiments were fitted to the logistic equation to ob-
overallhelicaseIC50values appeared slightly weaker (i.e,.
larger) than theATPaseIC50values, it was observed that
the inhibitors followed the same general trends as
those observed for the ATPase data. Bananin, iodoba-
nanin, vanillinbananin, and eubananin were effective in-
hibitors of helicase activity, while ansabanin and aden-
inobananin barely inhibit the reaction.
We also performed a final control to check whether
bananin acted as a general helicase inhibitor or not. We
cloned and purified the E. coli DnaB helicase, which is
a well characterized helicase with 5# to 3# polarity of
unwinding . The purity of DnaB may be observed
by SDS-PAGE in Figure S3A. We found that 250 µM
bananin did not inhibit DnaB in our FRET-based assay
(Figure S3B). These results suggest that bananin does
not act as a general helicase inhibitor.
Figure 2. Inhibition of SCV Helicase ATPase Activity by the Six Dif-
ferent Bananin Derivatives and Inhibition of the Nonstimulated
ATPase Activity by Bananin
(A) Inhibition of dT24-stimulated ATPase activity. A colorimetric as-
say was used to measure phosphate release due to ATP to ADP
hydrolysis after a 5 min period. Points shown are the average of
triplicate experiments and the error bars represent the standard
distribution. Data were fitted with the logistic equation to calculate
each IC50. (B) Inhibition of nonstimulated ATPase activity. The same
assay was used to measure inhibition of the reaction in the ab-
sence of dT24by bananin.
helicaseIC50 values (Figure 4C). Although the
bulky side groups on the six-membered ring of this
class of compounds reduce their inhibitory activity
against the SCV helicase.
We also checked whether bananin would inhibit the
unstimulated basal ATPase activity in the absence of
dT24of the SCV helicase (Figure 2B). It is clear that
bananin is not an effective inhibitor of the unstimulated
ATPase activity, although slight inhibition was observed
at 100 µM.
We then tested the mechanism of inhibition of the
ATPase activity by bananin, and checked competition
with respect to both ATP (Figures 3A and 3B) and with
respect to dT24(Figures 3C and 3D). Figures 3B and 3D
Chemistry & Biology
Figure 3. Bananin Is a Noncompetitive Inhibitor with Respect to Both ATP and to Nucleic Acid
Points are an average of triplicate experiments. (A) ATPase activity was measured under varying ATP concentrations in the presence and
absence of 2.3 µM bananin. (B) Lineweaver Burke plot of inhibition data from (A). Solid circles represent absence of inhibitor while open
circles represent presence of 2.3 µM bananin. Dotted lines represent 95% confidence of fit of straight lines. (C) ATPase activity was measured
under varying dT24concentrations in the presence and absence of 2.3 µM bananin. (D) Lineweaver Burke plot of inhibition data from (C).
Solid circles again represent absence of inhibitor while open circles represent presence of 2.3 µM bananin. Dotted lines represent 95%
confidence of fit of straight lines.
Inhibition of SARS Coronavirus Replication
The potency of these inhibitors against the SCV heli-
case enzymatic activities prompted us to investigate
their ability to inhibit SCV replication in a cell culture
system. We chose to test bananin itself, it being the
most representative of the class and the parent com-
pound. SCV has previously been established in fetal
rhesus kidney-4 cells (FRhK-4) in our laboratories .
SCV infection typically presents clear cytopathic ef-
fects (CPEs): the cells appear inflamed with “ridged”
cell membranes when infected with the virus. Visual in-
spection of cell cultures infected with SCV in the pres-
ence of 50 ?M bananin, revealed that CPEs were dis-
tinctly reduced relative to those of a control infection
(results not shown). However, levels of CPEs were diffi-
cult to quantify accurately, and so an alternative pro-
cedure was pursued.
To quantify the antiviral activity of bananin, we mea-
sured the viral titre under different inhibitor concentra-
tions (Figure 5). The infectivity of the virus in the cul-
tures in the presence and absence of bananin was
measured by a standard TCID50protocol using serial
dilution of the cell culture supernatant , and com-
pared to virus controls where no drug was added.
Briefly, cell culture supernatant at various time points
after infection was serially diluted and fresh FRhK-4
cells were infected with the serial dilutions. Cytopathic
effects were observed three days after infection with
the serial dilutions, thereby allowing measurement of
the viral titre. In this study, drugs were added either one
hour before or one hour after the infection with a 0.03
multiplicity of infection (MOI) of the virus. In FRhK cells,
the generation time of the SCV replication has been
shown to be 17–19 hr (our unpublished data). There-
fore, the readings at 24 hr are effectively after a single
generation. It can be seen from this data that at a con-
centration of 10 ?M bananin, the viral titre was reduced
by almost 50% after 24 hr (Figure 5), and the drug was
most effective when added one hour after infection
compared to one hour before. When drug was added
before, it was not removed on addition of the virus and
was present for the rest of the experiment. After 48 hr,
the difference between addition of drug before and af-
ter infection became more pronounced, and the viral
titre had dropped below 35% of the control in the ab-
sence of inhibitor (Figure 5). At a concentration of 50
?M, these effects were more pronounced; when the
drug was added one hour after infection, the viral titre
was below 10% of an untreated control infection after
24 hr. At 100 ?M bananin, the viral titre was almost zero
Bananins As Inhibitors of the SARS Coronavirus
Figure 5. The Antiviral Activities of Bananin Measured by 50% Tis-
sue Culture Infective Dose (TCID50)
Bananin was added to cultures at the concentration indicated
either one hour before (white bars) or one hour after (hatched bars)
infection with the virus. The TCID50was measured either 24 hr or
48 hr after infection by a standard serial dilution protocol and com-
pared to a control where the culture had been infected in the ab-
sence of inhibitor.
in the cultures containing bananin, even at a concentra-
tion of 100 ?M (Figure 6). However, after 12 hr, it was
clearly observed that the drug was having an inhibitory
effect, even at a concentration of 10 ?M, and this effect
increased with time up to 48 hr. This data also indicated
that adding the drug after viral infection was consider-
ably more effective than when it was added before (Fig-
We measured the toxicity of bananin using a stan-
dard MTT assay. Fitting the logistic equation to the data
Figure 4. Principle of the FRET-Based Helicase Assay, Inhibition of
SCV Helicase Helicase Activity by the Six Different Bananin Deriva-
tives, and Summary of Enzymatic Inhibition Data
(A) Schematic showing the principles behind the FRET-based fluo-
rimetric assay of helicase activity.
(B) Inhibition of the helicase activity of the SCV helicase in the pres-
ence of various concentrations of the six bananin derivatives.
Points shown are the average of triplicate experiments. Data were
fitted with the logistic equation to calculate each IC50. Error bars
represent the standard deviation of triplicate measurements.
(C) Table showing the IC50values for inhibition of both ATPase and
helicase activities of the SCV helicase.
after 24 hr and 48 hr. These results suggest first that
bananin is an effective inhibitor with an EC50of below
10 µM (when measured 48 hr post-infection and when
drug was added one hour after virus infection). Second,
due to the increased efficacy of the drug when added
post infection, these results suggest that bananin does
not inhibit the entry step, but inhibits a later step of the
infection cycle after the virus has penetrated the cell.
As a further assay, we used quantitative real time
PCR to measure the relative quantities of viral RNA
(specified by primers targeted to the SCV S-gene) com-
pared to cellular RNA (specified by primers targeted to
the gene β-actin). β-actin is expressed stably at basal
level in FRhK-4 cells as determined by Q-RT-PCR and
is therefore a good control for possible bias in the ex-
periment. Again, the drug was added to various con-
centrations either one hour before, or one hour after
infection, and compared to a control where no drug
was added (Figure 6). Bananin was maintained at the
same concentration throughout the experiment. The ki-
netics of infection were examined by making Q-RT-PCR
measurements 1 hr, 6 hr, 12 hr, 24 hr, and 48 hr post-
infection. Up to 6 hr post infection, there was little dif-
ference in SCV S-gene levels between the control and
Figure 6. Kinetics of the Antiviral Activity of Bananin Measured by
Quantitative Real Time PCR
Bananin was added to cultures at the concentration indicated
(square: 10 ?M, up triangle: 50 ?M, down triangle: 100 ?M) either
1 hr before (D-V represents drug then virus, filled symbols) or 1 hr
after (V-D represents virus then drug, open symbols) infection with
the virus. Cellular and viral RNA levels were measured by quantita-
tive real time PCR over a 48 hr period, using primers complemen-
tary to the β-actin and SCV Spike protein genes, respectively, and
are displayed using a logarithmic scale. A control in the absence
of inhibitor was also carried out (crosses). Error bars represent the
standard deviation of triplicate measurements.
Chemistry & Biology
nonspecific aggregator. The ATPase inhibition results
were consistent with those from the helicase assays:
bananin, vanillinbananin, eubananin, and iodobananin
were the best inhibitors of DNA-unwinding, while ansa-
banin and adeninobananin were poor inhibitors. Gen-
erally, the helicase inhibition activities measured for
each compound, as determined by IC50values, were
less than the corresponding ATPase inhibition activi-
ties, but the trends remained consistent. This appears
to be a characteristic common to many helicase inhibi-
tors [24, 25].
Bananin, which acted as an effective inhibitor in both
enzymatic assays and is the prototypical member of
this class of compounds, was tested in a cell culture
system of the virus. Bananin exhibited antiviral effects
at concentrations significantly below those causing cell
toxicity (EC50< 10 ?M, CC50= 390 ?M). Experiments
measuring the viral titer in two different situations, one
where the drug was added one hour before viral infec-
tion, and one where the drug was added one hour after
viral infection, suggested that bananin did not inhibit
viral entry but inhibited some key intracellular pro-
cesses involved in viral replication or pathogenesis.
These experiments were further confirmed by kinetic
experiments measuring the relative quantity of viral
RNA compared to cellular RNA over 48 hr. The drug had
little effect on RNA levels during the early stages of the
life cycle (approximately 0–6 hr post infection), indicat-
ing that the amount of the virus which has entered the
cells was similar in the presence or absence of the
drug. But the viral RNA levels were very different during
the later stages (from approximately 12–48 hr), sug-
gesting that viral transcription/replication was inhibited
by bananin. It is interesting to note the marked reduc-
tion in efficacy when bananin is added prior to viral in-
fection, despite its continued presence post infection.
First, this would suggest that the compound does not
inhibit a viral entry process. Second, it suggests that
bananin may be affecting other cellular pathways that
in the absence of viral infection may resist the protec-
tive effects of the drug. The SCV helicase is one pos-
sible target within the cell, although at this stage we
cannot exclude the possibility that bananin may be in-
hibiting via other pathway(s).
Figure 7. Toxicity of Bananin to FRhK-4 Cells As Measured Using
the MTT Assay
Cell viability was measured after 48 hr in the presence of the indi-
cated concentrations of bananin by the MTT assay. Toxin repre-
sents amanitin at 30 ?g/ml. Error bars represent the standard devi-
ation of five measurements.
indicated that 48 hr post infection, bananin exhibited a
cytotoxic concentration causing 50% cell mortality
(CC50) of 390 ?M (Figure 7). As the EC50for bananin
added post infection at 48 hr was less than 10 ?M (Fig-
ure 5B), this result implies a specificity index (CC50/
EC50) of over 39.
Here we describe the synthesis and inhibitory effects
against enzymatic activities of the SARS coronavirus
helicase for several structurally unusual trioxa-ada-
mantane derivatives, trivially referred to as bananins.
We also show that bananin exhibits significant anti-SCV
activity in cell culture, through the inhibition of a pro-
cess occurring after viral entry into the cell. We have
developed an extremely convenient and quick method
for testing both ATPase activities colorimetrically, and
helicase activities fluorimetrically through a type of
FRET assay. This combination of assays may be adapted
easily for high-throughput screening of compound li-
braries against both NTPase and DNA-unwinding enzy-
matic activities, and avoids the use of radioactive32P,
which is commonly used in many traditional helicase
ATPase assays revealed that iodobananin and vanil-
linbananin were the most effective SCV helicase inhibi-
tors, withATPaseIC50values of 0.54 ?M and 0.68 ?M,
respectively. Bananin (ATPaseIC50= 2.3 ?M) and euba-
nanin (ATPaseIC50= 2.8 ?M) were also reasonable inhibi-
tors, but ansabananin and adeninobananin, which con-
tain bulky appendages on the pyridoxal system, showed
little if any inhibition. Bananin acted as a noncompeti-
tive inhibitor with respect to both ATP and nucleic acid,
suggesting this class of inhibitors binds at a site dis-
tinct from the ATP and nucleic acid binding sites. Far
weaker inhibition of the unstimulated ATPase activity
was observed, showing that bananin does not act as a
Adamantane derivatives have been used clinically for
many years as antiviral treatments and as muscle re-
laxants. Here, we have demonstrated that a class of
pyridoxal-conjugated trioxaadamantanes, the bana-
nins, inhibit both the ATPase and helicase activities
of the SARS coronavirus helicase. Testing a number
of bananin derivatives, we have shown that it is im-
portant to reduce steric hindrance around the pyri-
doxal ring for effective inhibition of the SCV helicase.
Furthermore, bananin was shown to be an effective
antiviral drug in a cell culture of the virus. The mode
of viral inhibition supports the hypothesis that the
SCV helicase is a target of these compounds. Given
the paucity of drugs shown to be effective in treating
this recently emerged disease, the bananins repre-
Bananins As Inhibitors of the SARS Coronavirus
C14H16O8(M = 312.27 g/mol). The structure was established by a
COSY combination of1H-NMR and13C-NMR spectroscopy, sup-
plemented by UV/VIS spectrophotometry.
sent a class of compounds with significant therapeu-
tic potential against SARS.
Cloning and Purification of the SCV Helicase
The SCV helicase domain (nsp13-pp1ab, accession number NP_
828870, originally denoted as nsp10) was cloned and purified as
previously described .
Synthesis of Eugenolbananin (Eubananin, EUB)
from Bananin and Eugenol
A mass of 4.41 g bananin  (M = 327.29 g/mol) (13.4743 mmol)
and 3.00 ml eugenol (3.20 g) (M = 164.20 g/mol) (ρ420= 1.0664
g/ml) (19.4836 mmol) were suspended in 30 ml of water. Then, 6.00
g of sodium hydroxide (NaOH) pearls (0.15 mol) were added in
small portions, then heated until all material had dissolved. Then
15.0 ml of 10 M hydrochloric acid (0.15 mol HCl) was added in small
portions. The mixture was kept at 4°C for 12 hr and a precipitate
recovered by filtration and dessication. Yield: 5.33 g (81%) reddish-
brown powder 1-[6-[(2RS)-1-(4-hydroxy-3-methoxyphenyl)-2-pro-
9-trioxaadamantane-3,5,7-triol (eubananin, EUB) C24H29NO10(M =
491.49 g/mol). The structure was established by a COSY combina-
tion of1H-NMR and13C-NMR spectroscopy.
Cloning and Purification of E. coli DnaB
The DnaB helicase was amplified by PCR from E. coli genomic DNA
using 5#-GGCGAATTCATGGCAGGAAATAAAACCCTTCAAC-3# and
5#-TAATATCTCGAGTCATTCGTCGTCGTACTGCGGCCC-3#. The PCR
product was gel purified then EcoRI/XhoI ligated into pET28a to
form plasmid DnaB-pET28a. The plasmid was maintained in strain
DH10B and transformed into strain BL21-DE3 for expression. A 5
ml LB culture containing 50 µg/ml kanamycin was grown overnight,
then 1 ml of the overnight culture was added to 500 ml LB contain-
ing 50 µg/ml kanamycin. The culture was induced with 0.5 mM
IPTG at AU = 0.4, then grown further for three hours at 37°C. The
cells were collected by centrifugation and sonication was used to
split soluble and insoluble fractions. DnaB was observed by SDS-
PAGE to be mainly present in the insoluble fraction, and further
purification was from the insoluble fraction. The insoluble fraction
was washed three times with 30 ml 50 mM TRIS-HCl (pH 7.4). The
pellet was then dissolved in 15 ml 6 M GuCl / 50 mM TRIS-HCl
(pH 7.4)/20 mM imidazole, and any insoluble material removed by
centrifugation. The protein was refolded by injecting 15 ml solution
through a fine-bore needle into 135 ml of rapidly vortexing 50 mM
TRIS-HCl (pH 6.8) / 5 mM MgCl2/ 20% glycerol / 1% triton / 10 mM
β-mercaptoethanol on ice. Very little precipitate was seen in this
procedure; any precipitate was removed by centrifugation. The
protein was then passed onto a 5 ml Ni-NTA column, washed with
50 ml of 50 mM TRIS-HCl (pH 8.5) / 40 mM imidazole, before being
eluted with 50 mM TRIS-HCl (pH 8.5) / 200 mM imidazole. β-mer-
captoethanol was then added to 10 mM, glycerol added to 20%,
and the protein was stored at −20°C.
ATPase assays were performed using a phosphomolybdate-mal-
achite green assay described previously . Reaction conditions
were 50 mM TRIS-HCl (pH 6.8), 5 mM MgCl2, 200 nM dT24(for
stimulated reactions), 0.1 mg/ml BSA, 3.2 ng SCV helicase (for
stimulated) or 32 ng SCV helicase (for nonstimulated) in a 50 µl
reaction volume for 5 min in a 96-well plate. Reaction was stopped
by addition of EDTA to 50 mM, then AM/MG reagent and trisodium
citrate were added as described to measure phosphate released in
the reaction . Titration of ATPase activity with inhibitors in the
presence of fixed concentrations of polynucleotide and ATP was
described by a modified logistic equation .
“One Pot Synthesis” of AZTRION
Masses of 18.44 g vanillin (M = 152.14 g/mol) (121.20 mmol) and
15.29 g phloroglucinol (M = 126.11 g/mol) (121.24 mmol) were sus-
pended in 73 ml of water. Then, 29.27 g of solid sodium hydroxide
pearls (NaOH) (M = 40.00 g/mol) was added in small portions. Im-
mediately after the solidification, 100 ml of water was added and
the suspension was shaken vigorously for 5 min until the develop-
ment of heat ceased and the mixture turned crystalline. Then a
mass of 22.28 g 1,3,5,7-tetraazatricyclo[220.127.116.11,7]decane (methen-
amine, urotropin, hexamethylenetetramine) (M = 140.19 g/mol)
(158.93 mmol), 173 ml of water, and 39.93 g NaOH was added and
the mixture was refluxed for 20 min before cooling at −18°C for 6
hr. Yield: 22.16 g (66%) fine reddish-orange needles disodium 6,10-
hydrate [1-azaadamantane-4,6,10-trione bis(sodium hydroxide) ad-
duct monohydrate, AZTRION] C9H11NO5Na2× H2O (M = 277.18
g/mol). The structure was established by a COSY combination of
1H-NMR and13C-NMR spectroscopy, supplemented by IR spectro-
Synthesis of the Ansa Compound Ansabananin (ABN)
from Eubananin and AZTRION
A mass of 1.17 g eubananin (M = 491.49 g/mol) (2.3805 mmol) was
dissolved in 40.0 ml of 0.5 M sodium hydroxide aqueous solution.
A mass of 1.21 g AZTRION (M = 277.18 g/mol) (4.3654 mmol) was
added. The black solution was treated dropwise with 3.50 ml of 10
M hydrochloric acid. Afterward, 2.00 g of NaOH pearls (M = 40.00
g/mol) (50 mmol NaOH) was added. The black solution was kept at
4°C in an open crystallization dish for two days. After that time, the
dark crystalline mass was harvested and pressed between filter
papers. The material was recrystallized from 30.0 ml 2.0 M NaOH
aqueous solution. The black solution was kept at 4°C in an open
crystallization dish for two days. Yield: 2.43 g (99%) tan crystals
4##-diyldioxy]-1-azaadamantane-4-one sodium hydroxide adduct
pentadecahydrate (ansabananin, ABN) C33H36N2O13Na2× NaOH ×
15 H2O (M = 1024.86 g/mol). The structure was established by a
COSY combination of1H-NMR and13C-NMR spectroscopy.
A([L]) = 1 −
FRET-Based Helicase Assays
We used a protocol modified from that described , using oligo-
mers suitable for a 5# to 3# helicase. Two oligomers were synthe-
sized and purified by HPLC: DT20Cy3 (5#-TTTTTTTTTTTTTTTTTT
TTCGAGCACCGCTGCGGCTGCACC(Cy3)-3#), and ReleaseBHQ (5#-
(BHQ2)GGTGCAGCCGCAGCGGTGCTCG-3#) (Proligo). The two
oligomers were annealed by mixing a 1:1.2 ratio of DT20Cy3:Re-
leaseBHQ at a concentration of 8.2 ?M (of DT20Cy3) in 10 mM
TRIS-HCl (pH 8.5), heating to 90°C, then cooling slowly to 40°C
over one hour. The reaction was carried out in a 1 ml volume of 5 nM
DT20Cy3:ReleaseBHQ, 10 nM Release oligomer (5#-GGTGCAGC
CGCAGCGGTGCTCG-3#), 0.5 mM ATP, 0.1 mg/ml BSA, 2 nM SCV
helicase, 5 mM MgCl2, and 50 mM TRIS-HCl (pH 6.8) at 25°C for 1
min. The change in fluorescence (excitation 550 nm, emission 570
nm) after 1 min was used to monitor the extent of unwinding of the
duplex. The DnaB FRET assay was carried out with 10 µg DnaB
under the same conditions.
Preparation of Vanillinbananin (VBN) from Vanillin
Masses of 25.20 g vanillin (M = 152.14 g/mol) (165.64 mmol) and
20.88 g phloroglucinol (M = 126.11 g/mol) (165.57 mmol) were dis-
solved in 300 ml of water. Then 40.00 g of solid sodium hydroxide
(NaOH) (M = 40.00 g/mol) (1 mol NaOH) was added in small por-
tions, and, afterward, the mixture was titrated with 100 ml of 10 M
hydrochloric acid (1 mol HCl). It was subsequently cooled at 4°C
for 6 hr. Yield: 45.17 g (87%) yellow powder 1-(4-hydroxy-3-meth-
oxyphenyl)-2,8,9-trioxaadamantane-3,5,7-triol or 1-(4-hydroxy-3-
Cell Culture and Determination of Cytopathic Effects
Fetal rhesus kidney (FRhK-4) cells were plated on a 96-well plate
(2000 cells per well) under minimum essential medium containing
Chemistry & Biology
5% (v/v) fetal bovine serum, 1% (w/v) sodium pyruvate, 100 U/ml
penicillin, 0.1 mg/ml streptomycin and were cultured at 37°C in 5%
CO2. To test anti-SCV activities, FRhK-4 cultures were treated with
a range of different drug concentrations one hour before or after
infection with 0.03 MOI of SCV (strain GZ50). Either 24, 36, or 48 hr
post infection, cytopathic effects (CPE) were observed by phase-
contrast microscopy. The uninfected cells appeared smooth while
infected cells showed prominent ridge-like structures along the
membranes. Viral reproduction in the infected cells was quantified
by virus titration, as described below.
2. Wagner, H., Fischer, M., and Lotter, H. (1985). New bufadienol-
ides from Kalanchoe daigremontiana Hamet et Perr. (Crassula-
ceae). Z. Naturforsch Teil B 40, 1226–1227.
3. Davies, W.L., Grunert, R.R., Haff, R.F., McGahen, J.W., Neu-
mayer, E.M., Paulshock, M., Watts, J.C., Wood, T.R., Hermann,
E.C., and Hoffmann, C.E. (1964). Antiviral activity of 1-Ada-
mantanamine (Amantadine). Science 144, 862–863.
4. Schwab, R.S., England, A.C., Jr., Poskanzer, D.C., and Young,
R.R. (1969). Amantadine in the treatment of Parkinson’s dis-
ease. JAMA 208, 1168–1170.
5. Wintermeyer, S.M., and Nahata, M.C. (1995). Rimantadine: a
clinical perspective. Ann. Pharmacother. 29, 299–310.
6. Rosenthal, K.S., Sokol, M.S., Ingram, R.L., Subramanian, R.,
and Fort, R.C. (1982). Tromantadine: inhibitor of early and late
events in herpes simplex virus replication. Antimicrob. Agents
Chemother. 22, 1031–1036.
7. Kornhuber, J., Weller, M., Schoppmeyer, K., and Riederer, P.
(1994). Amantadine and memantine are NMDA receptor antag-
onists with neuroprotective properties. J. Neural Transm.
Suppl. 43, 91–104.
8. Kesel, A.J., Sonnenbichler, I., Polborn, K., Gurtler, L., Klinkert,
W.E., Modolell, M., Nussler, A.K., and Oberthur, W. (1999). A
new antioxidative vitamin B6 analogue modulates pathophysi-
ological cell proliferation and damage. Bioorg. Med. Chem. 7,
9. Kesel, A.J. (2003). A system of protein target sequences for
anti-RNA-viral chemotherapy by a vitamin B6-derived zinc-
chelating trioxa-adamantane-triol. Bioorg. Med. Chem. 11,
10. Peiris, J.S., Lai, S.T., Poon, L.L., Guan, Y., Yam, L.Y., Lim, W.,
Nicholls, J., Yee, W.K., Yan, W.W., Cheung, M.T., et al. (2003).
Coronavirus as a possible cause of severe acute respiratory
syndrome. Lancet 361, 1319–1325.
11. Ksiazek, T.G., Erdman, D., Goldsmith, C.S., Zaki, S.R., Peret, T.,
Emery, S., Tong, S., Urbani, C., Comer, J.A., Lim, W., et al.
(2003). A novel coronavirus associated with severe acute respi-
ratory syndrome. N. Engl. J. Med. 348, 1953–1966.
12. Drosten, C., Gunther, S., Preiser, W., van der Werf, S., Brodt,
H.R., Becker, S., Rabenau, H., Panning, M., Kolesnikova, L.,
Fouchier, R.A., et al. (2003). Identification of a novel coro-
navirus in patients with severe acute respiratory syndrome. N.
Engl. J. Med. 348, 1967–1976.
13. Rota, P.A., Oberste, M.S., Monroe, S.S., Nix, W.A., Campagnoli,
R., Icenogle, J.P., Penaranda, S., Bankamp, B., Maher, K.,
Chen, M.H., et al. (2003). Characterization of a novel coro-
navirus associated with severe acute respiratory syndrome.
Science 300, 1394–1399.
14. Marra, M.A., Jones, S.J., Astell, C.R., Holt, R.A., Brooks-Wil-
son, A., Butterfield, Y.S., Khattra, J., Asano, J.K., Barber, S.A.,
Chan, S.Y., et al. (2003). The Genome sequence of the SARS-
associated coronavirus. Science 300, 1399–1404.
15. Lau, A.C., So, L.K., Miu, F.P., Yung, R.W., Poon, E., Cheung,
T.M., and Yam, L.Y. (2004). Outcome of coronavirus-associated
severe acute respiratory syndrome using a standard treatment
protocol. Respirology 9, 173–183.
16. Booth, C.M., Matukas, L.M., Tomlinson, G.A., Rachlis, A.R.,
Rose, D.B., Dwosh, H.A., Walmsley, S.L., Mazzulli, T., Aven-
dano, M., Derkach, P., et al. (2003). Clinical features and short-
term outcomes of 144 patients with SARS in the greater To-
ronto area. JAMA 289, 2801–2809.
17. Cinatl, J., Morgenstern, B., Bauer, G., Chandra, P., Rabenau,
H., and Doerr, H.W. (2003). Glycyrrhizin, an active component
of liquorice roots, and replication of SARS-associated coro-
navirus. Lancet 361, 2045–2046.
18. Anand, K., Ziebuhr, J., Wadhwani, P., Mesters, J.R., and Hilgen-
feld, R. (2003). Coronavirus main proteinase (3CLpro) structure:
basis for design of anti-SARS drugs. Science 300, 1763–1767.
19. Yang, H., Yang, M., Ding, Y., Liu, Y., Lou, Z., Zhou, Z., Sun, L.,
Mo, L., Ye, S., Pang, H., et al. (2003). The crystal structures of
severe acute respiratory syndrome virus main protease and its
complex with an inhibitor. Proc. Natl. Acad. Sci. USA 100,
20. Bacha, U., Barrila, J., Velazquez-Campoy, A., Leavitt, S.A., and
Freire, E. (2004). Identification of novel inhibitors of the SARS
coronavirus main protease 3CLpro. Biochemistry 43, 4906–
Inhibition of SCV Reproduction
FRhK-4 cell cultures were infected with SCV one hour before or
after being treated with various concentrations of drug. Following
incubation for 24 and 48 hr, viable SCV production was measured
by back titration of the culture media supernatant using a TCID50
(50% tissue culture infectious dose) protocol . Briefly, the su-
pernatant was serially 10-fold diluted with fresh cell culture media
(MEM) and inoculated into FRhK-4 cells in 96-well plates. The virus
titre was determined by observation of cytopathic effects (CPE) in
FRhK-4 cells after 3 days of culture [36, 38].
Cells were washed twice with PBS, and total RNA was extracted
using an RNeasy Mini kit (Qiagen, Germany) in accordance with the
manufacturer’s instructions. Reverse-transcription was performed
using random hexamers with the ThermoScript RT system (Invitro-
gene, CA). Intracellular viral RNA was quantified using quantitative
RT-PCR (Q-RT-PCR) [36, 38, 39], using the forward primer 5#-GCT
TAG GCC CTT TGA GAG AGA CA-3# and the reverse primer 5#-
GCC AAT GCC AGT AGT GGT GTA AA-3# (final concentration 200
nM), the fluorescent probe 5#-CCT GAT GGC AAA CCT TGC AC-3#
and phosphate probe 5#-(LC640)CAC CTG CTC TTA ATT GTT ATT
GGC C-3# (final concentration 800 nM). The real-time quantification
was carried out using LC Faststart DNA Master Hyb Probes and
LightCycler (Roche Diagnostics, USA). PCR conditions employed
were 95°C for 10 min and then 50 cycles at 95°C for 10 s, 60°C for
5 s, 72°C for 5 s and 40°C for 30 s. The increase in PCR products
was monitored for each amplification cycle by measuring the
increase in fluorescence caused by the binding of SYBR Green I to
double-stranded DNA. The crossing point values were determined
for each sample and specificity of the amplicons was measured by
melting curve analysis and visualized by agarose gel electrophore-
sis. A ten-fold serial dilution of plasmid ranging from 1.5 pg/ml to
1.5 × 106pg/ml were used as standard and the gene β-actin was
used as an endogenous control to normalize for intersample varia-
tions in the amount of total RNA.
Determination of Drug Cytotoxic Concentration
Concentration was measured using a standard methylthiazolyldi-
phenyl-tetrazolium bromide (MTT) assay. The CC50was determined
by fitting data to the logistic equation, as described above; amani-
tin at 30 ?g/ml was used as a toxic control.
Supplemental Data for this article is available online at http://
This work was supported by the Hong Kong Health, Welfare and
Food Bureau under grants from the Research Fund for the Control
of Infectious Diseases. R.M.W. was supported by the Area of Excel-
lence Scheme of the University Grants Committee.
Received: August 8, 2004
Revised: December 16, 2004
Accepted: January 12, 2005
Published: March 25, 2005
1. Woodward, R.B. (1964). The structure of tetrodotoxin. Pure
Appl. Chem. 9, 49–74.
Bananins As Inhibitors of the SARS Coronavirus
21. Wu, C.Y., Jan, J.T., Ma, S.H., Kuo, C.J., Juan, H.F., Cheng, Y.S.,
Hsu, H.H., Huang, H.C., Wu, D., Brik, A., et al. (2004). Small
molecules targeting severe acute respiratory syndrome human
coronavirus. Proc. Natl. Acad. Sci. USA 101, 10012–10017.
22. Kao, R.Y., Tsui, W.H., Lee, T.S., Tanner, J.A., Watt, R.M., Huang,
J.D., Hu, L., Chen, G., Chen, Z., Zhang, L., et al. (2004). Identifi-
cation of novel small-molecule inhibitors of severe acute respi-
ratory syndrome-associated coronavirus by chemical genetics.
Chem. Biol. 11, 1293–1299.
23. Kleymann, G., Fischer, R., Betz, U.A., Hendrix, M., Bender, W.,
Schneider, U., Handke, G., Eckenberg, P., Hewlett, G., Pevzner,
V., et al. (2002). New helicase-primase inhibitors as drug candi-
dates for the treatment of herpes simplex disease. Nat. Med.
24. Crute, J.J., Grygon, C.A., Hargrave, K.D., Simoneau, B.,
Faucher, A.M., Bolger, G., Kibler, P., Liuzzi, M., and Cordingley,
M.G. (2002). Herpes simplex virus helicase-primase inhibitors
are active in animal models of human disease. Nat. Med. 8,
25. Borowski, P., Schalinski, S., and Schmitz, H. (2002). Nucleotide
triphosphatase/helicase of hepatitis C virus as a target for anti-
viral therapy. Antiviral Res. 55, 397–412.
26. Tanner, J.A., Watt, R.M., Chai, Y.B., Lu, L.Y., Lin, M.C., Peiris,
J.S., Poon, L.L., Kung, H.F., and Huang, J.D. (2003). The severe
acute respiratory syndrome (SARS) coronavirus NTPase/heli-
case belongs to a distinct class of 5# to 3# viral helicases. J.
Biol. Chem. 278, 39578–39582.
27. Thiel, V., Ivanov, K.A., Putics, A., Hertzig, T., Schelle, B., Bayer,
S., Weissbrich, B., Snijder, E.J., Rabenau, H., Doerr, H.W., et al.
(2003). Mechanisms and enzymes involved in SARS coro-
navirus genome expression. J. Gen. Virol. 84, 2305–2315.
28. Ivanov, K.A., Thiel, V., Dobbe, J.C., van der Meer, Y., Snijder,
E.J., and Ziebuhr, J. (2004). Multiple enzymatic activities asso-
ciated with severe acute respiratory syndrome coronavirus he-
licase. J. Virol. 78, 5619–5632.
29. van Dinten, L.C., van Tol, H., Gorbalenya, A.E., and Snijder, E.J.
(2000). The predicted metal-binding region of the arterivirus
helicase protein is involved in subgenomic mRNA synthesis,
genome replication, and virion biogenesis. J. Virol. 74, 5213–
30. van Marle, G., van Dinten, L.C., Spaan, W.J., Luytjes, W., and
Snijder, E.J. (1999). Characterization of an equine arteritis virus
replicase mutant defective in subgenomic mRNA synthesis. J.
Virol. 73, 5274–5281.
31. Uehara, Y. (2003). Natural product origins of Hsp90 inhibitors.
Curr. Cancer Drug Targets 3, 325–330.
32. Wardell, A.D., Errington, W., Ciaramella, G., Merson, J., and
McGarvey, M.J. (1999). Characterization and mutational analy-
sis of the helicase and NTPase activities of hepatitis C virus
full-length NS3 protein. J. Gen. Virol. 80, 701–709.
33. Baykov, A.A., Evtushenko, O.A., and Avaeva, S.M. (1988). A
malachite green procedure for orthophosphate determination
and its use in alkaline phosphatase-based enzyme immuno-
assay. Anal. Biochem. 171, 266–270.
34. Boguszewska-Chachulska, A.M., Krawczyk, M., Stankiewicz,
A., Gozdek, A., Haenni, A.L., and Strokovskaya, L. (2004). Di-
rect fluorometric measurement of hepatitis C virus helicase ac-
tivity. FEBS Lett. 567, 253–258.
35. Biswas, S.B., Chen, P.H., and Biswas, E.E. (1994). Structure
and function of Escherichia coli DnaB protein: role of the
N-terminal domain in helicase activity. Biochemistry 33,
36. He, M.L., Zheng, B., Peng, Y., Peiris, J.S., Poon, L.L., Yuen,
K.Y., Lin, M.C., Kung, H.F., and Guan, Y. (2003). Inhibition of
SARS-associated coronavirus infection and replication by RNA
interference. JAMA 290, 2665–2666.
37. Porter, D.J. (1998). Inhibition of the hepatitis C virus helicase-
associated ATPase activity by the combination of ADP, NaF,
MgCl2, and poly(rU). Two ADP binding sites on the enzyme-
nucleic acid complex. J. Biol. Chem. 273, 7390–7396.
38. Zheng, B.J. (2004). Prophylactic and therapeutic effects of
small interfering RNA targeting SARS-Coronavirus. Antivir.
Ther. 9, 365–374.
39. He, M.L., Wu, J., Chen, Y., Lin, M.C., Lau, G.K., and Kung, H.F.
(2002). A new and sensitive method for the quantification of
HBV cccDNA by real-time PCR. Biochem. Biophys. Res. Com-
mun. 295, 1102–1107.