An M2-V27A channel blocker demonstrates potent in?vitro and in?vivo antiviral activities against amantadine-sensitive and -resistant influenza A viruses

Article abstract of DOI:10.1016/j.antiviral.2017.01.006

Adamantanes such as amantadine (1) and rimantadine (2) are FDA-approved anti-influenza drugs that act by inhibiting the wild-type M2 proton channel from influenza A viruses, thereby inhibiting the uncoating of the virus. Although adamantanes have been suc

Full text of DOI:10.1016/j.antiviral.2017.01.006

Accepted Manuscript
An M2-V27A channel blocker demonstrates potent in vitro and in vivo antiviral
activities against amantadine-sensitive and -resistant influenza A viruses
Yanmei Hu, Rami Musharrafieh, Chunlong Ma, Jiantao Zhang, Donald F. Smee,
William F. DeGrado, Jun Wang
PII:
S0166-3542(16)30522-8
10.1016/j.antiviral.2017.01.006
AVR 3983
DOI:
Reference:
To appear in: Antiviral Research
Received Date: 16 September 2016
Revised Date: 5 January 2017
Accepted Date: 9 January 2017
Please cite this article as: Hu, Y., Musharrafieh, R., Ma, C., Zhang, J., Smee, D.F., DeGrado, W.F.,
Wang, J., An M2-V27A channel blocker demonstrates potent in vitro and in vivo antiviral activities
against amantadine-sensitive and -resistant influenza A viruses, Antiviral Research (2017), doi: 10.1016/
j.antiviral.2017.01.006.
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to
our customers we are providing this early version of the manuscript. The manuscript will undergo
copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please
note that during the production process errors may be discovered which could affect the content, and all
legal disclaimers that apply to the journal pertain.

ACCEPTED MANUSCRIPT
Graphical Abstract
+
NH3 Cl-
A/Solomon Islands/3/2006 (M2-WT) (H1N1) Amantadine-sensitive
70% survival protection at 100 mg/kg/day
A/WSN/33 (M2-N31S+V27A) (H1N1) Amantadine-resistant
87.5% survival protection at 100mg/kg/day
3

ACCEPTED MANUSCRIPT
An M2-V27A channel blocker demonstrates potent in vitro and in vivo antiviral activities
against amantadine-sensitive and -resistant influenza A viruses
Yanmei Hu1, #, Rami Musharrafieh2, #, Chunlong Ma3, Jiantao Zhang1, Donald F. Smee4, William
F. DeGrado5 and Jun Wang1,3*
1Department of Pharmacology and Toxicology, College of Pharmacy, The University of Arizona,
Tucson, Arizona 85721, United States
2Department of Chemistry and Biochemistry, The University of Arizona, Tucson, Arizona 85721,
United States
3BIO5 Institute, The University of Arizona, Tucson, Arizona 85721, United States
4Institute for Antiviral Research, Department of Animal, Dairy and Veterinary Sciences, Utah
State University, Logan, Utah 84322, United States
5Department of Pharmaceutical Chemistry, School of Pharmacy, University of California, San
Francisco, California 94158, United States
#Yanmei Hu and Rami Musharrafieh contributed equally to this work.
*Corresponding author: Jun Wang, Ph.D., Room 423, 1657 E. Helen St., BIO5 Institute, Tucson,
Arizona 85721, United States. Phone: 520-626-1366; Email: junwang@pharmacy.arizona.edu

ACCEPTED MANUSCRIPT
Abstract
Adamantanes such as amantadine (1) and rimantadine (2) are FDA-approved anti-influenza
drugs that act by inhibiting the wild-type M2 proton channel from influenza A viruses, thereby
inhibiting the uncoating of the virus. Although adamantanes have been successfully used for
more than four decades, their efficacy was curtailed by emerging drug resistance. Among the
limited number of M2 mutants that confer amantadine resistance, the M2-V27A mutant was
found to be the predominant mutant under drug selection pressure, thereby representing a high
profile antiviral drug target. Guided by molecular dynamics simulations, we previously designed
first-in-class M2-V27A inhibitors. One of the potent lead compounds, spiroadamantane amine
(3), inhibits both the M2-WT and M2-V27A mutant with IC50 values of 18.7 and 0.3 µM,
respectively, in in vitro electrophysiological assays. Encouraged by these findings, in this study
we further examine the in vitro and in vivo antiviral activity of compound 3 in inhibiting both
amantadine-sensitive and -resistant influenza A viruses. Compound 3 not only had single to sub-
micromolar EC50 values against M2-WT- and M2-V27A-containing influenza A viruses in
antiviral assays, but also rescued mice from lethal viral infection by either M2-WT- or M2-
V27A-containing influenza A viruses. In addition, we report the design of two analogs of
compound 3, and one was found to have improved in vitro antiviral activity over compound 3.
Collectively, this study represents the first report demonstrating the in vivo antiviral efficacy of
inhibitors targeting M2 mutants. The results suggest that inhibitors targeting drug-resistant M2
mutants are promising antiviral drug candidates worthy of further development.

ACCEPTED MANUSCRIPT
Keywords
Influenza virus, M2 proton channel, V27A, drug resistance, spiroadamantane, antiviral

ACCEPTED MANUSCRIPT
1. Introduction
Influenza virus infections are a persistent public health concern. Despite the existence of
influenza vaccines and antiviral drugs, influenza viruses claim 250,000–500,000 lives per year
worldwide (Thompson et al., 2010). In the unpredictable event of influenza pandemics, the death
toll is normally several orders higher (Monto and Webster, 2013). Although influenza vaccines
are generally effective in preventing influenza infection, they are less effective for the elderly,
children, and immunocompromised persons (Houser and Subbarao, 2015). As a result, they are
the most vulnerable populations for influenza virus infection. Moreover, it normally takes at least
six months to produce and manufacture sufficient quantities of influenza vaccines for public use;
thus influenza vaccines are unlikely to be an option to confine the first wave of an emerging
influenza pandemic (Houser and Subbarao, 2015). In contrast, small molecule antiviral drugs can
be stockpiled and readily distributed in bulk quantities; thus they are not an alternative, but rather
an indispensable complement to vaccines. Currently there are two classes of anti-influenza drugs.
M2 inhibitors such as amantadine (1) and rimantadine (2) inhibit the viral M2 proton channel
and are viral entry inhibitors. Neuraminidase inhibitors such as oseltamivir, zanamivir, and
peramivir inhibit the viral neuraminidase and are viral egress inhibitors. The emergence of drug-
resistant influenza strains now necessitates the development of the next generation of antiviral
drugs. Resistance to adamantanes is so widely spread that the CDC no longer recommends the
use of adamantanes in prophylaxis and the treatment of influenza infection. This leaves
neuraminidase inhibitors as the last FDA-approved option. The alarming fact is that oseltamivir-
resistant strains have been continuously reported among human and avian influenza viruses
(Govorkova et al., 2013; Hurt et al., 2009; Hurt, 2014). In fact, the 2008–2009 seasonal influenza
viruses with the H275Y NA mutation were completely resistant to the only orally bioavailable
drug, oseltamivir (Hurt, 2014; Renaud et al., 2011). Therefore, the potential of resistance

ACCEPTED MANUSCRIPT
evolution, coupled with the lack of an alternative antiviral drug with a distinct mechanism of
action, calls for the development of the next generation of antiviral drugs (Lee and Yen, 2012;
Loregian et al., 2014).
With the aim of developing novel anti-influenza drugs, we chose the M2 proton channel as
the drug target because M2 is one of the most conserved viral proteins (Dong et al., 2015; Wang,
2016; Wang et al., 2015; Wang et al., 2011c). Although many M2 mutants emerged under drug
selection pressure in cell cultures (Abed et al., 2005; Brown et al., 2010), animals (Herlocher et
al., 2003), and human patients (Hayden and Hay, 1992; Iwahashi et al., 2001), only a few of
them acquired the fitness to transmit among humans (Dong et al., 2015; Furuse et al., 2009a;
Furuse et al., 2009b). Genetic studies have shown that the predominant M2 mutants circulating
among humans are S31N, V27A, and L26F (Dong et al., 2015). Among this limited number of
mutants, V27A was shown to be the predominant mutation under drug selection pressure (Furuse
et al., 2009a; Furuse et al., 2009b). We therefore are interested in designing the second
generation of M2 channel blockers by targeting the V27A mutant (Balannik et al., 2009; Rey-
Carrizo et al., 2013; Wang et al., 2011a; Wang et al., 2011b).
M2 is a homo-tetrameric proton channel that is embedded in the viral membrane (Hong and
DeGrado, 2012). The proton conductance function is encoded by the transmembrane domain
spanning residues 25–46. The V27 side-chain faces the centroid of the channel and forms the
first gate (Pielak and Chou, 2010). H37 and W41 are conserved residues that are critical for
proton conductance and selectivity (Balannik et al., 2010). Guided by molecular dynamics
simulations, we previously generated a homology model of the V27A mutant (Wang et al.,
2011a). The model showed an enlarged channel cavity at the N-terminus compared with M2-WT.
This prediction was later confirmed by the NMR investigation of M2-V27A (Pielak and Chou,

ACCEPTED MANUSCRIPT
2010) as well as molecular dynamics simulations from another group (Yi et al., 2008). On the
basis of this observation, we designed a 3D-shaped spiroadamantane amine molecule (3) that has
improved shape complementarity to the V27A mutant compared to amantadine (Wang et al.,
2011a). Compound 3 was found to have potent channel blockage against V27A mutant in
electrophysiological assays, with an IC50 of 0.3 ± 0.1 µM. It also inhibits the M2-WT channel
with an IC50 of 18.7 ± 1.4 µM. The mechanism of this dual inhibition phenomenon was revealed
by molecular dynamics simulations. It suggested that compound 3 binds deeper in the M2-WT
channel and that there is only one layer of water between the ammonium from compound 3 and
the H37 (Wang et al., 2011a). In the case of the M2-V27A mutant channel, compound 3 shifted
up toward the N-terminus, and there are two layers of water between the ammonium from
compound 3 and the H37.
Although a few compounds were reported to inhibit the M2-V27A mutant channel (Rey-
Carrizo et al., 2014; Rey-Carrizo et al., 2013; Wang et al., 2011a), there has been no report on
their antiviral activity against the M2-V27A-containing influenza A virus. Given the promising
channel blockage efficacy of compound 3 in inhibiting both the M2-WT and the M2-V27A
mutant, we took a step further to evaluate the in vitro and in vivo antiviral efficacy of compound
3. Specifically, the antiviral activity of compound 3 against M2-WT and the M2-V27A-
containing influenza A viruses was tested in plaque assays. For in vivo studies, the non-toxic
dose of compound 3 was determined, then the effect of compound 3 on body weight and survival
rate of mice infected with a lethal dose of influenza A viruses was observed. Furthermore, we
designed and synthesized two analogs of compound 3 with the aim of further improving the
antiviral efficacy.

ACCEPTED MANUSCRIPT
2. Materials and methods
2.1. Compounds
Compound 3 was synthesized as previously reported (Wang et al., 2011a). Novel synthesis
procedures and compound characterization by NMR and MS can be found in Fig. 5 and the
Supplementary Data. Amantadine was purchased from Sigma (St. Louis, MO). Oseltamivir (as
Tamiflu®) was obtained from a local pharmacy as 75 mg capsules. The compounds were
prepared in sterile water for oral gavage (p.o.) administration. For oseltamivir, this required
using one or more entire capsules, since excipients were also present.
2.2. Viruses
M2-WT-expressing A/Udorn/72 (H3N2), A/WSN/33 (M2-N31S) (H1N1), A/Solomon
Islands/3/2006 (H1N1), and M2-V27A-expressing A/WSN/33 (M2-N31S + V27A) (H1N1) were
used to infect MDCK cells (American Type Culture Collection, Manassas, VA) in the presence
or absence of compounds to evaluate their antiviral activity. Influenza A/WSN/33 (M2-N31S)
(H1N1) and A/WSN/33 (M2-N31S + V27A) (H1N1) viruses were generated by reverse genetics
of the A/WSN/33 (H1N1) virus. The M genes of the resulting recombinant viruses were
sequenced and the sequences were confirmed as expected. The virus was later titrated in BALB/c
mice for lethality. Influenza A/Udorn/72 (H3N2) was obtained from Dr. Robert Lamb’s
laboratory at the Northwestern University. Influenza A/Solomon Islands/3/2006 (H1N1) virus
was obtained from the Centers for Disease Control and Prevention (Atlanta, GA). The virus was
passaged seven times in mice to enhance its virulence, and later titrated in BALB/c mice for
lethality. Approximately a 100% lethal dose of each H1N1 virus was selected for use in this
strain of mouse.

ACCEPTED MANUSCRIPT
2.3. Plaque assays
Plaque assay were performed according to previous reports (Ma et al., 2016a; Ma et al.,
2016b). M2-WT-expressing A/Udorn/72 (H3N2), A/WSN/33 (M2-N31S) (H1N1), A/Solomon
Islands/3/2006 (H1N1), and M2-V27A-expressing A/WSN/33 (M2-N31S + V27A) (H1N1) were
used to infect MDCK cells in the presence or absence of compounds to evaluate their antiviral
activity. Briefly, a confluent monolayer of MDCK cells was incubated with ∼100-pfu virus
samples in DMEM with 0.5% bovine serum albumin for 1 h at 4 °C, then 37 °C for 1 h. The
inoculums were removed, and the cells were washed with phosphate-buffered saline (PBS). The
cells were then overlaid with DMEM containing 1.2% Avicel microcrystalline cellulose (FMC
BioPolymer, Philadelphia, PA) and N-acetyl trypsin (2.0 µg/ml). To examine the effect of the
compounds on plaque formation, the overlay media were supplemented with compounds at
testing concentrations. At day 2 after infection, the monolayers were fixed and stained with
crystal violet dye solution (0.2% crystal violet, 20% methanol). Plaque area was quantified using
ImageJ (Guzman et al., 2014) and the fifty percent virus-inhibitory concentration (EC50) values
were subsequently determined.
2.4. Cytotoxicity assays
Evaluation of the cytotoxicity of compounds was carried out using the neutral red uptake
assay (Repetto et al., 2008). Briefly, 80,000 cells/ml MDCK cells in DMEM medium, which was
supplemented with 10% FBS and 100 U/ml of Penicillin-Streptomycin solution, were dispensed
into clear 96-well cell culture plates (Corning Cat #: CLS3362) at 100 µL/well. Twenty-four
hours later, the growth medium was removed and washed with 100 µL PBS buffer; then 200 µL
fresh DMEM (no FBS) medium containing serial diluted compounds was added to each well.

ACCEPTED MANUSCRIPT
After incubation for 24 h at 37 °C with 5% CO2 in a CO2 incubator, the medium was removed
and replaced with 100 µl DMEM medium containing 40 mg/ml neutral red for 4 h at 37 °C. The
amount of retained neutral red was determined by absorbance at 540 nm using a Multiskan FC
Microplate Photometer (Fisher Scientific). The 50% cytotoxic concentration (CC50) values were
calculated from best-fit dose response curves with variable slope in Prism 5.
2.5. Electrophysiological assays
All compounds were initially tested in a TEVC assay using Xenopus laevis frog oocytes
microinjected with RNA expressing either the WT or the V27A mutant of A/Udorn/72-M2, as
described in a previous report (Balannik et al., 2009). The potency of the inhibitors was
expressed as the percent inhibition of M2 current observed after 2 min of incubation with 100
µM compounds.
2.6. Animals
Female 18–20 g BALB/c mice were obtained from Charles River Laboratories (Wilmington,
MA) for the antiviral study and for toxicity determinations. The animals were maintained on
standard rodent chow and tap water ad libitum. They were housed 5 per cage in corn cob bedding
and quarantined for 48 h prior to use. Mice were block randomized so that approximately the
same average initial body weight was in each group. Before and during the infection and
treatment, mice are observed for general well-being (activity, physical appearance, and unusual
behaviors following treatment). This was in addition to monitoring body weight.
2.7. Experiment design of animal studies

ACCEPTED MANUSCRIPT
BALB/c mice were anesthetized by intraperitoneal injection of ketamine/xylazine (50/5
mg/kg), and the animals were infected intranasally with a 90-µl suspension of influenza virus
(day 0). The virus challenge was approximately 3–4 50% mouse lethal infectious doses, equating
to about 1-2 x 104 50% cell culture infectious doses (CCID50) of virus per mouse. Oral
treatments were given twice a day (at 12 h intervals) for 5 days starting 30 min before virus
challenge. Parameters for assessing the infection were survival, mean day of death, body weight
changes, and lung infection parameters (hemorrhage score, weight, and virus titer). Animals
were weighed every other day through day 21 of the infection. Mice whose body weights fell
below 30% of initial weight were humanely euthanized and deaths were noted. During
infections, there were 10 compound-treated and 20 placebos per group held for determining
survival. Five uninfected mice per group were treated with compound 3 or placebo were
assessed with 10 normal controls (uninfected, untreated) for drug toxicity in one experiment.
To assess lung infection parameters, lungs from sacrificed animals (5 mice/group, held
for this purpose) were harvested. Lung hemorrhage score was assessed in a blinded manner by
visual inspection for color changes from pink to plum. This occurs regionally in the lungs, not by
a gradual color change of the whole lung. Hemorrhage scores ranged from 0 (normal) to 4 (total
lung showing plum color) in 0.5 unit increments averaged over the whole lung. The scoring
range correlates to 0 to 100% lung involvement. The lungs were weighed and then homogenized
in cell culture medium, followed by freezing at −80 °C. Later, thawed lungs were serially diluted
and titrated by the endpoint dilution method in 96-well plates of MDCK cells (Smee et al., 2008),
using 4 microwells per dilution, and by diluting homogenates out to a 108 dilution. Cultures were
incubated at 37 oC for 6 days and wells scored for presence or absence of viral cytopathic effect

ACCEPTED MANUSCRIPT
in order to determine the 50% viral titer endpoints. Virus titers were reported as log10 cell culture
infectious doses per gram of lung tissue (log10 CCID50/g).
2.8. Statistical analysis of animal data
Kaplan–Meier plots for multiple group comparisons were analyzed by the Mantel–Cox log-
rank test to determine statistical significance. Subsequently, pairwise comparisons were made by
the Gehan–Breslow–Wilcoxon test. Percent survival results were analyzed by the two-tailed
Fishers exact test. Mean day of death and lung hemorrhage data were not normally distributed.
Analysis of these parameters was made by Kruskal-Wallis test followed by pairwise comparisons
analyzed by the two-tailed Mann-Whitney U-test. Mean lung weight and lung viral titer data
were normally distributed, and were analyzed by one-way analysis of variance (ANOVA)
followed by Tukey–Kramer multiple comparisons test. All comparisons were made between
drug-treated and placebo groups, and were analyzed using InStat® and Prism® software
(GraphPad Software, San Diego, CA).
2.9. Ethics regulation of laboratory animals
The animal experiments were conducted in accordance with the approval of the Institutional
Animal Care and Use Committee of Utah State University. They were performed in the
AAALAC-accredited Laboratory Animal Research Center of Utah State University. The U. S.
Government (National Institutes of Health) approval (PHS Assurance No. A3801-01) and
conduct of the work were in accordance with the National Institutes of Health Guide for the Care
and Use of Laboratory Animals.
3. Results

ACCEPTED MANUSCRIPT
3.1. In vitro antiviral efficacy of compound 3
As compound 3 was confirmed already to be a potent blocker of both M2-WT and M2-V27A
proton channels (Table 1) (Wang et al., 2011a), we then wanted to confirm its antiviral efficacy
against both M2-WT- and M2-V27A-containing influenza A viruses. The A/Udorn/72 (H3N2)
virus and the A/WSN/33 (M2-N31S + V27A) (H1N1) virus, which contain M2-WT and M2-
V27A, respectively, were chosen to assess the drug sensitivity of compound 3. The A/WSN/33
(M2-N31S + V27A) (H1N1) virus was generated by reverse genetics. The asparagine residue in
position 31 was mutated to serine, and the A/WSN/33-N31S (H1N1) was amantadine-sensitive
(Wang et al., 2013), similar to the M2-WT. In addition, valine in position 27 was mutated to
alanine. As shown in Fig. 1A, compound 3 had a similar antiviral activity as amantadine (1) in
inhibiting the M2-WT-containing virus A/Udorn/72 (H3N2), with an EC50 value of 0.3 ± 0.1 µM.
When tested against another M2-WT-containing virus A/WSN/33 (M2-N31S) (H1N1) virus,
both amantadine (1) and compound 3 showed potent antiviral activity with EC50 values of 0.2 ±
0.01 µM, and 0.1 ± 0.01 µM, respectively (Table S1). In contrast, the A/WSN/33 (M2-N31S +
V27A) (H1N1) virus was nearly completely resistant to amantadine (1) (Fig. 1B), and no plaque
reduction was observed up to 1 µM. Further increasing the concentration of amantadine (1) to 50
µM also failed to significantly inhibit the replication of the A/WSN/33 (M2-N31S + V27A)
(H1N1) virus (plaque reduction less than 10 %). For this reason, oseltamivir carboxylate (4) was
used as a positive control, and it showed complete inhibition at 200 nM. Gratifyingly, compound
3 showed dose-dependent inhibition of the M2-V27A mutant–containing A/WSN/33 (M2-N31S
+ V27A) (H1N1) virus with an EC50 of 1.8 ± 0.2 µM.
To assess the genetic barrier to resistance for compound 3, serial viral passage experiments
were performed. Specifically, the M2-WT-containing virus A/Udorn/72 (H3N2) was amplified

ACCEPTED MANUSCRIPT
for several rounds in the presence of increasing concentrations of compound 3, and the antiviral
EC50 value was determined at each passage. At passage 1, compound 3 was applied at 300 nM
and the EC50 was 0.9 ± 0.1 µM, which confers a 3-fold increase. At passage 2, compound 3 was
applied at 600 nM and the EC50 was more than 10 µM, indicating complete resistance. In
parallel, amantadine (1) was also subjected to the serial viral passage experiments. At passage 1,
amantadine (1) was applied at 300 nM and the EC50 was more than 10 µM, indicating complete
resistance. Thus, compound 3 appears to have a similar genetic barrier to resistance as
amantadine.
In summary, compound 3 not only had potent channel blockage activity against the
A/Udorn/72 M2-WT and M2-V27A channels with IC50 values of 18.7 ± 1.4 µM and 0.3 ± 0.1
µM, respectively, it also had potent antiviral activity against both M2-WT and M2-V27A
mutant–containing influenza A viruses.
3.2. Assaying the in vivo non-toxic dose of compound 3
The in vivo non-toxic dose of compound 3 was assessed with uninfected BALB/c mice.
Compound 3 was prepared in sterile water and given to mice by oral gavage (p.o.) administration.
The compound was given twice a day (at 12 h intervals) for five days at either 30 mg/kg/day or
100 mg/kg/day. As shown in Fig. 2, a dose of compound 3 up to 100 mg/kg/day was well
tolerated. Although mice treated with compound 3 exhibited a 3% body weight difference
compared to placebos and normal control animals on day 3, this difference was not statistically
significant (P>0.05). For cumulative toxicity to occur, body weights should have been at nadir
on day 5 (which was 12 hours after the final treatment). Instead weight gain was seen in all
groups between days 3 and 5. No apparent differences in physical activity or behavior were

ACCEPTED MANUSCRIPT
noted among groups during the treatment period or thereafter. On the basis of this result, the
maximum dose was set as 100 mg/kg/day for the following efficacy studies.
3.3. In vivo antiviral efficacy of compound 3 in treatment of an M2-WT-containing
influenza A/Solomon Islands/3/2006 (H1N1) virus infection in mice
As the amantadine-sensitive A/Udorn/72 (H3N2) virus was not mouse-adapted and 100%
lethality could not be achieved, another strain of M2-WT-containing influenza A virus,
A/Solomon Islands/3/2006 (H1N1), was chosen for the in vivo efficacy study. A/Solomon
Islands/3/2006 (H1N1) became lethal to mice after seven serial passages in the lungs of infected
animals (Murray et al., 2012). To confirm that the mouse-adapted A/Solomon Islands/3/2006
(H1N1) virus was sensitive to amantadine (1) and compound 3, plaque assays were performed.
The EC50 values of amantadine (1) and compound 3 in inhibiting the M2-WT-containing
A/Solomon Islands/3/2006 (H1N1) virus were 0.1 ± 0.01 µM and 0.1 ± 0.02 µM (Table S1),
respectively. Thus the choice of A/Solomon Islands/3/2006 (H1N1) for the in vivo studies was
justified. As shown in Fig. 3A, no mice in the placebo group survived beyond day 8 after
infection with A/Solomon Islands/3/2006 (H1N1). Oral treatment of compound 3 at a doses of
40 and 100 mg/kg/day or above significantly improved the survival rate to 70% (Fig. 3A, Table
2). In comparison, amantadine (1) was more potent than compound 3, with all doses providing
some protection, and doses of 20, 40, and 100 mg/kg/day providing 100% protection (Fig. 3C,
Table 2).
Body weight changes during the infection are shown in Figs. 3B and 3D. Oral treatment with
compound 3 was effective in reducing the extent of weight loss at the 40 and 100 mg/kg/day
doses (Fig. 3B). In a similar manner, amantadine (1) treatments of 20, 40, and 100 mg/kg/day

ACCEPTED MANUSCRIPT
reduced weight loss during the acute infection (Fig. 3D). Each compound performed in a dose-
responsive manner.
The results of this experiment established that compound 3 possessed in vivo antiviral
activity against a M2-WT-containing influenza A virus, A/Solomon Islands/3/2006 (H1N1).
Compound 3 was effective in reducing mortality and weight loss at 40 and 100 mg/kg/day when
given by oral gavage route; however, its potency was less than that of amantadine (1).
3.4. In vivo antiviral efficacy of compound 3 in treatment of a M2-V27A-containing
influenza A/WSN/33 (M2-N31S + V27A) (H1N1) virus infection in mice
As shown in Fig. 4A and Table 3, compound 3 at 100 mg/kg/day was highly protective but
lower doses were not effective. The positive control drug oseltamivir (4) was 100% protective at
a dose of 10 mg/kg/day. Amantadine (1) was not effective in preventing death, as A/WSN/33
(M2-N31S + V27A) (H1N1) was amantadine-resistant in vitro (Fig. 1). High significance was
seen in groups treated with compound 3 (100 mg/kg/day) and oseltamivir (4). A considerably
less but statistically significant benefit (primarily in delaying the time to death) was afforded by
treatment with compound 3 (30 mg/kg/day, 10.3 days MDD) or with amantadine (1) (100
mg/kg/day, 11.1 days MDD). The results of antiviral treatment on lung infection parameters
during this same infection are shown in Table 3. On day 6 of the infection the mean lung scores
in the compound 3 100 mg/kg/day group were normal, as were lungs from the oseltamivir (4)
group (Table 3). Certain other treatments with compound 3 or amantadine (1) also reduced lung
hemorrhage scores significantly. Significant reductions in lung weight were found in a few of the
compound 3 and amantadine (1) groups, but were not as great as in the oseltamivir (4) group.
Lung virus titers were determined from animals sacrificed on day 6. Compound 3 at 100

ACCEPTED MANUSCRIPT
mg/kg/day and oseltamivir (4) significantly reduced virus titers compared to placebo.
Amantadine (1) at 100 mg/kg/day also reduced virus titers, but the results did not achieve
statistical significance (P>0.05).
Body weights during the infection are depicted in Figs. 4B and 4C. Minimal weight loss
occurred in the oseltamivir-treated group, followed by the compound 3 (100 mg/kg/day) (Fig. 4B)
and the amantadine (1) (100 mg/kg/day) (Fig. 4C) groups. Rebound in body weight occurred
faster in the compound 3 (100 mg/kg/day) group than in the comparable amantadine group.
In conclusion, the 100-mg/kg/day dose of compound 3 was effective in terms of survival
outcome, protection from body weight loss, and improvement in lung hemorrhage score. In
contrast, amantadine (1) showed minimum effectiveness at 100 mg/kg/day in treating the M2-
V27A virus infection.
3.5. Design, synthesis, and in vitro activity of analogs of compound 3.
Encouraged by the promising in vitro and in vivo antiviral efficacy of compound 3, we then
designed two analogs of compound 3 (Fig. 5) with the aim to further improve antiviral efficacy.
Compound 7 is the guanidine analog of 3, and it was synthesized through a two-step synthesis
procedure with an overall yield of 68%. Compound 16 is the dithiane analog of compound 3.
Synthesis of compound 16 started from 2-amino-1,3-propanediol (8). The amine was first
protected with benzyl chloroformate (CbzCl) to give the intermediate 10 with 85% yield, then
the alcohol in 10 was converted to thioester 12 through the Mitsunobu reaction with 72% yield.
The ester 12 was hydrolyzed and the dithiol intermediate 13 was condensed with 2-adamantone
14 to give the protected dithiane 15 in 65% overall yield. Finally, the Cbz protecting group was
cleaved by Pd/C and ammonium formate, and the hydrochloride salt of the amine (16) was
formed by treatment of HCl in methanol. The overall yield for the last two steps was 78%.

ACCEPTED MANUSCRIPT
The channel blockage activity of the newly synthesized compounds 7 and 16 was assayed
against both M2-WT and M2-V27A proton channels using the two-electrode voltage clamp
(TEVC) assays (Table 4). For the M2-WT channel inhibition, compound 7 had a similar potency
as compound 3, reaching 94.5 ± 2.0 % current inhibition at 100 µM. Compound 16 had reduced
channel blockage with 64.0 ± 1.2 % at 100 µM. For M2-V27A channel inhibition, all compounds
nearly completely inhibited the current conductance at 100 µM, and the IC50 values for
compounds 3, 7, and 16 in inhibiting the A/Udorn/72 (M2-V27A) were 0.3 ± 0.1 µM, 0.5 ± 0.2
µM, and 0.4 ± 0.1 µM, respectively (Table 4). Collectively, the two analogs, 7 and 16, had
confirmed channel blockage activity against both M2-WT and M2-V27A mutant channels.
When tested against the M2-S31N channel, none of the compounds (3, 7, and 16) showed more
than 12 % channel inhibition at 100 µM. These results were expected since compounds 3, 7, and
16 were designed to target the M2-V27A mutant, not the M2-S31N mutant.
We then tested the antiviral activity of compounds 7 and 16 against both M2-WT-containing
A/Udorn/72 (M2-WT) (H3N2) virus and the M2-V27A-containing A/WSN/33 (M2-N31S +
V27A) (H1N1) virus in plaque assays (Fig. 6). Compounds 7 and 16 each had potent antiviral
activity against the M2-WT-containing A/Udorn/72 (H3N2) virus, with EC50 values of 6.4 ± 0.4
µM and 0.07 ± 0.01 µM, respectively. In contrast, only compound 16 had potent antiviral
activity against the M2-V27A-containing A/WSN/33 (M2-N31S + V27A) (H1N1) virus, with an
EC50 value of 1.0 ± 0.1 µM. Compound 7 had no effect on the replication of the M2-V27A-
containing A/WSN/33 (M2-N31S + V27A) (H1N1) virus up to 30 µM, despite its potent channel
blockage, as confirmed by the TEVC assay (Table 4). The cytotoxicity of compounds 3, 7, and
16 were tested against the MDCK cells using the neutral red method, and the two analogs, 7 and
16, were less cytotoxic than compound 3 (Table 5).

ACCEPTED MANUSCRIPT
As shown, the correlations between the channel blockage results and the antiviral efficacy
were not linear. Compound 16 had reduced M2-WT channel blockage compared to compound 3
(64.0 ± 1.2 % versus 89.1 ± 0.4 %); however, the antiviral activity of compound 16 in inhibiting
the M2-WT-containing A/Udorn/72 (H3N2) virus was four-fold greater than compound 3 (EC50
= 0.07 ± 0.01 µM versus 0.3 ± 0.1 µM). Similarly, compound 7 had very poor antiviral (EC50 >
30 µM) potency against M2-V27A-containing A/WSN/33 (M2-N31S + V27A) (H1N1) virus
despite its potent channel blockage (93.3 ± 1.3 % at 100 µM). The discrepancy promoted us to
take a further look at the drug binding kinetics. It is known that the adamantane class of drugs
bind with slow-on and very slow-off rates to the M2 channel, and the rate of recovery varies
significantly between individual compounds (Balannik et al., 2009; Ma et al., 2016b; Tu et al.,
1996; Wang et al., 1993). Examination of the current traces from electrophysiological
recordings revealed interesting findings (Fig. 7). In these experiments, the pH is first dropped to
5.5 to allow measurement of channel conduction. Once a steady-state value is approached, the
compound is added, leading to rapid inhibition. After the full extent of inhibition is achieved
free drug is removed by perfusion with drug-free pH 8.5 buffer (the high pH avoids membrane
leakage from prolonged exposure to low pH). After a delay of 1 min, the buffer is again returned
to pH 5.5, and the extent of reversal of channel block assessed from the fractional change in
proton flux relative to the value prior to drug treatment. In the case of compound 3, as shown by
the second activation trace (Fig. 7A), the M2-V27A channel remained partially occluded even
with perfusion with compound-free solution (pH = 8.5). The channel conductance current at the
second activation event was 50.4 ± 7.5% of the maximum current. In contrast, the M2-V27A
channel inhibition by compound 7 was nearly completely reversible (Fig. 7B), reaching 93.9 ±
1.7% of the maximum current at the second activation event. This observation provides an

ACCEPTED MANUSCRIPT
attractive explanation for the lack of antiviral activity of compound 7. The dithiane compound 16,
on the other hand, showed essentially irreversible inhibition of the M2-V27A channel on this
time scale, and only 12.2 ± 1.3 % current was recovered at the second activation event (Fig. 7C).
Taken together, it became apparent that the kinetics of drug inhibition is a critical parameter for
channel inhibition: a very slow rate of reversal appears to correlate with potent antiviral efficacy,
while more rapidly reversible channel blockers were less active. As such, caution has to be taken
when predicting the compound’s antiviral activity based on its percent channel blockage in
electrophysiological assays.
4. Discussion and conclusions
M2-V27A is one of the most frequent amantadine-resistant mutations among influenza A
viruses that are circulating among humans and avian species (Dong et al., 2015; Durrant et al.,
2015). More concerning is the fact that M2-V27A mutant is directly correlated with drug
selection pressure (Furuse et al., 2009b). Although the most frequent amantadine-resistant
mutant M2-S31N remains partially sensitive to amantadine (Wang et al., 2013), the M2-V27A
mutant was completely resistant to amantadine (Balannik et al., 2009), which makes it a
challenging drug target. Nevertheless, propelled by an integrated approach involving molecular
dynamics simulations, NMR, electrophysiology, and medicinal chemistry, we were able to
design the first-in-class M2-V27A inhibitors (Wang et al., 2011a; Wang et al., 2011b). In this
report, we further evaluated the in vitro and in vivo antiviral activity of the M2-V27A inhibitors.
The representative compound, 3, not only had potent in vitro antiviral activity, but also potent in
vivo antiviral activity against both M2-WT- and M2-V27A-containing influenza A viruses. This
proof-of-concept study demonstrates that drug-resistant M2 mutants are valid antiviral drug
targets and should be further pursued.

ACCEPTED MANUSCRIPT
During the course of drug discovery targeting the M2 proton channels, the percentage of
channel blockage or dose-dependent channel blockage (IC50 value) of compounds was normally
used to predict their antiviral activity. It is generally assumed that compounds with higher
channel blockage activity will display potent antiviral activity. However, as has been
demonstrated by compound 7 in inhibiting the M2-V27A channel, this assumption appears
incorrect. The results from this study emphasize the role of drug binding kinetics: slow-
reversible or irreversible channel blockers correlate with potent antiviral activity, while
reversible channel blockers showed weak or no antiviral activity. With such knowledge in mind,
priority in lead optimization should be given to compounds with slow or irreversible binding
kinetics during the following drug discovery development.
Acknowledgements
This work was supported by the startup funds and the NIH AI 119187 from the University of
Arizona to J.W. and NIH GM056423 to W.F.D. R.M. was supported by the NIH
training grant T32 GM008804. Funding for D.F.S. and for the animal studies was provided by
the Respiratory Diseases Branch, DMID, NIAID, NIH under contract HHSN272201000039I
We thank David Bishop for proofreading and editing the manuscript.
.

ACCEPTED MANUSCRIPT
Figure legends:
Fig. 1. Antiviral efficacy of amantadine (1) and compound 3 in inhibiting M2-WT- and M2-
V27A-containing influenza A viruses. (A) Plaque assay results of amantadine (1) and
compound 3 in inhibiting M2-WT-containing A/Udorn/72 (H3N2) virus. The EC50 values for
amantadine (1) and compound 3 were 0.3 ± 0.1 µM and 0.3 ± 0.1 µM, respectively. (B) Plaque
assay results of amantadine (1) and compound 3 in inhibiting M2-V27A-containing A/WSN/33
(M2-N31S + V27A) (H1N1) virus. Oseltamivir carboxylate (4) was used as a control. The EC50
value for compound 3 was 1.8 ± 0.2 µM.
Fig. 2. Effects of compound 3 on body weight in uninfected mice. Oral gavage treatments
were administered twice a day for 5 days starting on day 0. The normal control (uninfected,
untreated) group contained 10 mice. The other groups contained 5 mice each. Mean values ±
SEM are shown at each time point. This assessment was performed in parallel with the infection
experiment reported in Figs. 3 and 4.
Fig. 3. Survival curves and body weight changes for compound 3 (A and B), amantadine (1)
(C and D), and oseltamivir (4) in p.o. treatment of an influenza A/Solomon Islands/3/2006
(H1N1) virus infection in mice. (A) Survival curves for compound 3. (B) Body weight changes
for compound 3. (C) Survival curves for amantadine (1). (D) Body weight changes for
amantadine (1). Oral gavage treatments were given twice a day for 5 days starting 30 min before
virus exposure at day 0. *** P<0.001 compared to placebo, determined by the Mantel-Cox log-
rank test with pairwise comparisons made by the Gehan-Breslow-Wilcoxon test. Groups of 10
compound-treated mice and 20 placebos were used. Groups of 10 compound-treated mice and

ACCEPTED MANUSCRIPT
20 placebos were used. Mean values are shown at each time point. Error bars are not shown
because animals were weighed as groups instead of individually in this experiment.
Fig. 4. Survival curves and body weight changes for compound 3, amantadine (1), and
oseltamivir (4) on an influenza A/WSN/33 (M2-N31S + V27A) (H1N1) virus infection in
mice. (A) Survival curves for compound 3, amantadine (1), and oseltamivir (4). (B) Body weight
changes for compound 3. (C) Body weight change for amantadine (1). Oral gavage treatments
were administered twice a day for 5 days starting 30 min pre-infection on day 0. ** P<0.01, ***
P<0.001, compared to placebo, determined by the Mantel-Cox log-rank test with pairwise
comparisons made by the Gehan-Breslow-Wilcoxon test. Groups of 10 compound-treated mice
and 20 placebos were used. Groups of 10 compound-treated mice and 20 placebos were used.
Mean values ± SEM are shown at each time point.
Fig. 5. Synthesis procedures of compound 3 analogs. (A) Synthesis of the guanidine analog 7.
(B) Synthesis of the dithiane analog 16.
Fig. 6. Antiviral efficacy of compounds 7 and 16 in inhibiting M2-WT and M2-V27A-
containing influenza A viruses. (A) Plaque assay results of compounds 7 and 16 in inhibiting
M2-WT containing A/Udorn/72 (H3N2) virus. The EC50 values for 7 and 16 were 6.4 ± 0.4 µM
and 0.07 ± 0.01 µM, respectively. (B) Plaque assay results of compounds 7 and 16 in inhibiting
M2-V27A-containing A/WSN/33 (M2-N31S + V27A) (H1N1) virus. Compound 7 was not
active and compound 16 had an EC50 of 1.0 ± 0.1 µM. Oseltamivir carboxylate (4) was used as a
control.

ACCEPTED MANUSCRIPT
Fig. 7. Electrophysiological assays of compounds 3, 7, and 16 in inhibiting the current
conductance of the A/Udorn/72 (M2-V27A) proton channel. (A) M2-V27A channel
conductance traces with and without compound 3. (B) M2-V27A channel conductance traces
with and without compound 7. (C) M2-V27A channel conductance trace with and without
compound 16.

ACCEPTED MANUSCRIPT
References
Abed, Y., Goyette, N., Boivin, G., 2005. Generation and characterization of recombinant
influenza A (H1N1) viruses harboring amantadine resistance mutations. Antimicrob. Agents
Chemother. 49, 556-559.
Balannik, V., Carnevale, V., Fiorin, G., Levine, B., Lamb, R., Klein, M., DeGrado, W., Pinto, L.,
2010. Functional Studies and Modeling of Pore-Lining Residue Mutants of the Influenza A Virus
M2 Ion Channel. Biochemistry 49, 696-708.
Balannik, V., Wang, J., Ohigashi, Y., Jing, X., Magavern, E., Lamb, R.A., DeGrado, W.F., Pinto,
L.H., 2009. Design and Pharmacological Characterization of Inhibitors of Amantadine-Resistant
Mutants of the M2 Ion Channel of Influenza A Virus. Biochemistry 48, 11872-11882.
Brown, A.N., McSharry, J.J., Weng, Q., Driebe, E.M., Engelthaler, D.M., Sheff, K., Keim, P.S.,
Nguyen, J., Drusano, G.L., 2010. In Vitro System for Modeling Influenza A Virus Resistance
under Drug Pressure. Antimicrob. Agents Chemother. 54, 3442-3450.
Dong, G., Peng, C., Luo, J., Wang, C., Han, L., Wu, B., Ji, G., He, H., 2015. Adamantane-
Resistant Influenza A Viruses in the World (1902–2013): Frequency and Distribution of M2
Gene Mutations. PloS one 10, e0119115.
Durrant, M.G., Eggett, D.L., Busath, D.D., 2015. Investigation of a recent rise of dual
amantadine-resistance mutations in the influenza A M2 sequence. BMC Genet. 16, S3.
Furuse, Y., Suzuki, A., Kamigaki, T., Oshitani, H., 2009a. Evolution of the M gene of the
influenza A virus in different host species: large-scale sequence analysis. Virol. J. 6, 67.
Furuse, Y., Suzuki, A., Oshitani, H., 2009b. Large-scale sequence analysis of M gene of
influenza A viruses from different species: mechanisms for emergence and spread of amantadine
resistance. Antimicrob. Agents Chemother. 53, 4457-4463.
Govorkova, E.A., Baranovich, T., Seiler, P., Armstrong, J., Burnham, A., Guan, Y., Peiris, M.,
Webby, R.J., Webster, R.G., 2013. Antiviral resistance among highly pathogenic influenza A
(H5N1) viruses isolated worldwide in 2002-2012 shows need for continued monitoring. Antiviral
Res. 98, 297-304.
Guzman, C., Bagga, M., Kaur, A., Westermarck, J., Abankwa, D., 2014. ColonyArea: an ImageJ
plugin to automatically quantify colony formation in clonogenic assays. PloS one 9, e92444.
Hayden, F.G., Hay, A.J., 1992. Emergence and transmission of influenza A viruses resistant to
amantadine and rimantadine. Curr. Top. Microbiol. Immunol. 176, 119-130.
Herlocher, M.L., Truscon, R., Fenton, R., Klimov, A., Elias, S., Ohmit, S.E., Monto, A.S., 2003.
Assessment of development of resistance to antivirals in the ferret model of influenza virus
infection. J. Infect. Dis. 188, 1355-1361.
Hong, M., DeGrado, W.F., 2012. Structural basis for proton conduction and inhibition by the
influenza M2 protein. Protein Sci. 21, 1620-1633.
Houser, K., Subbarao, K., 2015. Influenza Vaccines: Challenges and Solutions. Cell Host
Microbe 17, 295-300.
Hurt, A., Holien, J., Parker, M., Barr, I., 2009. Oseltamivir Resistance and the H274Y
Neuraminidase Mutation in Seasonal, Pandemic and Highly Pathogenic Influenza Viruses. Drugs
69, 2523-2531.
Hurt, A.C., 2014. The epidemiology and spread of drug resistant human influenza viruses. Curr.
Opin. Virol. 8, 22-29.

ACCEPTED MANUSCRIPT
Iwahashi, J., Tsuji, K., Ishibashi, T., Kajiwara, J., Imamura, Y., Mori, R., Hara, K., Kashiwagi,
T., Ohtsu, Y., Hamada, N., Maeda, H., Toyoda, M., Toyoda, T., 2001. Isolation of amantadine-
resistant influenza a viruses (H3N2) from patients following administration of amantadine in
Japan. J. Clin. Microbiol. 39, 1652-1653.
Lee, S.M., Yen, H.L., 2012. Targeting the host or the virus: Current and novel concepts for
antiviral approaches against influenza virus infection. Antiviral Res. 96, 391-404.
Loregian, A., Mercorelli, B., Nannetti, G., Compagnin, C., Palù, G., 2014. Antiviral strategies
against influenza virus: towards new therapeutic approaches. Cell. Mol. Life Sci., 1-25.
Ma, C., Li, F., Musharrafieh, R.G., Wang, J., 2016a. Discovery of cyclosporine A and its analogs
as broad-spectrum anti-influenza drugs with a high in vitro genetic barrier of drug resistance.
Antiviral Res. 133, 62-72.
Ma, C., Zhang, J., Wang, J., 2016b. Pharmacological Characterization of the Spectrum of
Antiviral Activity and Genetic Barrier to Drug Resistance of M2-S31N Channel Blockers. Mol.
Pharmacol. 90, 188-198.
Monto, A.S., Webster, R.G., 2013. Influenza pandemics: History and lessons learned, Textbook
of Influenza. John Wiley & Sons, Ltd, pp. 20-34.
Murray, J.L., McDonald, N.J., Sheng, J., Shaw, M.W., Hodge, T.W., Rubin, D.H., O'Brien, W.A.,
Smee, D.F., 2012. Inhibition of influenza A virus replication by antagonism of a PI3K-AKT-
mTOR pathway member identified by gene-trap insertional mutagenesis. Antivir. Chem.
Chemother. 22, 205-215.
Pielak, R.M., Chou, J.J., 2010. Solution NMR structure of the V27A drug resistant mutant of
influenza A M2 channel. Biochem. Biophys. Res. Commun. 401, 58-63.
Renaud, C., Kuypers, J., Englund, J.A., 2011. Emerging oseltamivir resistance in seasonal and
pandemic influenza A/H1N1. J. Clin. Virol. 52, 70-78.
Repetto, G., del Peso, A., Zurita, J.L., 2008. Neutral red uptake assay for the estimation of cell
viability/cytotoxicity. Nat. Protoc. 3, 1125-1131.
Rey-Carrizo, M., Barniol-Xicota, M., Ma, C., Frigolé-Vivas, M., Torres, E., Naesens, L., Llabrés,
S., Juárez-Jiménez, J., Luque, F.J., DeGrado, W.F., Lamb, R.A., Pinto, L.H., Vázquez, S., 2014.
Easily Accessible Polycyclic Amines that Inhibit the Wild-Type and Amantadine-Resistant
Mutants of the M2 Channel of Influenza A Virus. J. Med. Chem. 57, 5738-5747.
Rey-Carrizo, M., Torres, E., Ma, C., Barniol-Xicota, M., Wang, J., Wu, Y., Naesens, L.,
DeGrado, W., Lamb, R., Pinto, L., Vazquez, S., 2013. 3-
Azatetracyclo[5.2.1.1(5'8).0(1,5)]undecane Derivatives: From Wild-Type Inhibitors of the M2
Ion Channel of Influenza A Virus to Derivatives with Potent Activity against the V27A Mutant. J.
Med. Chem. 56, 9265-9274.
Smee, D.F., Bailey, K.W., Wong, M.H., O'Keefe, B.R., Gustafson, K.R., Mishin, V.P., Gubareva,
L.V., 2008. Treatment of influenza A (H1N1) virus infections in mice and ferrets with
cyanovirin-N. Antiviral Res. 80, 266-271.
Thompson, M.G., Shay, D.K., Zhou, H., Bridges, C.B., Cheng, P.Y., Burns, E., Bresee, J.S., Cox,
N.J., 2010. Estimates of Deaths Associated With Seasonal Influenza-United States, 1976-2007
(Reprinted from MMWR, vol 59, pg 1057-1062, 2010). JAMA 304, 1778-1780.
Tu, Q., Pinto, L.H., Luo, G., Shaughnessy, M.A., Mullaney, D., Kurtz, S., Krystal, M., Lamb,
R.A., 1996. Characterization of inhibition of M2 ion channel activity by BL-1743, an inhibitor of
influenza A virus. J. Virol. 70, 4246-4252.
Wang, C., Takeuchi, K., Pinto, L.H., Lamb, R.A., 1993. Ion channel activity of influenza A virus
M2 protein: characterization of the amantadine block. J. Virol. 67, 5585-5594.

ACCEPTED MANUSCRIPT
Wang, J., 2016. M2 as a target to combat influenza drug resistance: what does the evidence say?
Future Virol. 11, 1-4.
Wang, J., Li, F., Ma, C., 2015. Recent progress in designing inhibitors that target the drug-
resistant M2 proton channels from the influenza A viruses. Biopolymers 104, 291-309.
Wang, J., Ma, C., Fiorin, G., Carnevale, V., Wang, T., Hu, F., Lamb, R.A., Pinto, L.H., Hong,
M., Kein, M.L., DeGrado, W.F., 2011a. Molecular Dynamics Simulation Directed Rational
Design of Inhibitors Targeting Drug-Resistant Mutants of Influenza A Virus M2. J. Am. Chem.
Soc. 133, 12834-12841.
Wang, J., Ma, C., Wu, Y., Lamb, R.A., Pinto, L.H., DeGrado, W.F., 2011b. Exploring
Organosilane Amines as Potent Inhibitors and Structural Probes of Influenza A Virus M2 Proton
Channel. J. Am. Chem. Soc. 133, 13844-13847.
Wang, J., Qiu, J.X., Soto, C., DeGrado, W.F., 2011c. Structural and dynamic mechanisms for the
function and inhibition of the M2 proton channel from influenza A virus. Curr. Opin. Struct. Biol.
21, 68-80.
Wang, J.Z., Ma, C.L., Wang, J., Jo, H., Canturk, B., Fiorin, G., Pinto, L.H., Lamb, R.A., Klein,
M.L., DeGrado, W.F., 2013. Discovery of Novel Dual Inhibitors of the Wild-Type and the Most
Prevalent Drug-Resistant Mutant, S31N, of the M2 Proton Channel from Influenza A Virus. J.
Med. Chem. 56, 2804-2812.
Yi, M., Cross, T.A., Zhou, H.-X., 2008. A Secondary Gate As a Mechanism for Inhibition of the
M2 Proton Channel by Amantadine. J. Phys. Chem. B 112, 7977-7979.

ACCEPTED MANUSCRIPT
Table 1. Channel blockage activity of amantadine (1), rimantadine (2), and
spiroadamantane amine (3) in inhibiting the A/Udorn/72 (M2-WT) and A/Udorn/72 (M2-
V27A) proton channels.
Electrophysiological Assay
IC50 (µM)
M2-WT
M2-V27A
15.7 ± 1.2
> 100
10.8 ± 0.2
18.7 ± 1.4
> 100
0.3 ± 0.1

ACCEPTED MANUSCRIPT
Table 2. Effects of compound 3 and amantadine (1) on an influenza A/Solomon
Islands/3/2006 (H1N1) virus infection in mice. Oral treatments were administrated twice a day
(at 12 h intervals) for 5 d starting 30 min before virus exposure on day 0.
Compound
(mg/kg/day)
3 (100)
3 (40)
3 (20)
3 (10)
Survivors/Totala
MDDb ± SD
7/10***
7/10***
1/10
0/10
0/10
10.3 ± 3.1
15.0 ± 6.0*
11.9 ± 4.6***
7.7 ± 1.2
7.0 ± 1.1
-
3 (5)
Amantadine (1) (100) 10/10***
Amantadine (1) (40)
Amantadine (1) (20)
Amantadine (1) (10)
Amantadine (1) (5)
Placebo p.o.
10/10***
10/10***
6/10***
5/10**
-
-
13.0 ± 6.4
9.4 ± 1.9*
7.1 ± 0.8
0/20
a Statistically evaluated by two-tailed Fisher’s exact test.
b Mean day of death of mice that died prior to day 22, with pairwise comparisons statistically
evaluated by two-tailed Mann-Whitney U-Test.
*P<0.05, ***P<0.001 compared to placebo.