Isocyanides as influenza A virus subtype H5N1 wild-type M2 channel inhibitors

Article abstract of DOI:10.1002/cmdc.201500318

Basic bulky amines such as amantadine are well-characterized M2 channel blockers, useful for treating influenza. Herein we report our surprising findings that charge-neutral, bulky isocyanides exhibit activities similar to - or even higher than - that of amantadine. We also demonstrate that these isocyanides have potent growth inhibitory activity against the H5N1 virus. The -NH₂ to -N≡C group replacement within current anti-influenza drugs was found to give compounds with high activities at low-micromolar concentrations. For example, a tenfold improvement in potency was observed for 1-isocyanoadamantane (27), with an EC₅₀ value of 0.487 μm against amantadine-sensitive H5N1 virus as determined by both MTT and plaque-reduction assays, without showing cytotoxicity. Furthermore, the isocyanide analogues synthesized in this study did not inhibit the V27A or S31N mutant M2 ion channels, according to electrophysiology experiments, and did not exhibit activity against amantadine-resistant virus strains. Charge-neutral bulky isocyanides were found to exhibit antiviral activities similar to - or even higher than - that of amantadine. Moreover, we demonstrated that these isocyanides have potent growth inhibitory activity against the wild-type H5N1 virus. The NH₂ to N≡C group replacement within current anti-influenza drugs was found to result in compounds with low-micromolar activities.

Full text of DOI:10.1002/cmdc.201500318

DOI: 10.1002/cmdc.201500318
Full Papers
Isocyanides as Influenza A Virus Subtype H5N1 Wild-Type
M2 Channel Inhibitors
Shuwen Wu,[a, b] Jing Huang,[a] Sabrina Gazzarrini,[c] Si He,[a] Lihua Chen,[a] Jun Li,[a] Li Xing,[d]
Chufang Li,[e] Ling Chen,[e] Constantinos G. Neochoritis,[f] George P. Liao,[f] Haibing Zhou,[a]
Alexander Dçmling,[f] Anna Moroni,[c] and Wei Wang*[a]
Basic bulky amines such as amantadine are well-characterized
M2 channel blockers, useful for treating influenza. Herein we
report our surprising findings that charge-neutral, bulky isocya-
nides exhibit activities similar to—or even higher than—that of
amantadine. We also demonstrate that these isocyanides have
provement in potency was observed for 1-isocyanoadaman-
tane (27), with an EC50 value of 0.487 mm against amantadine-
sensitive H5N1 virus as determined by both MTT and plaque-
reduction assays, without showing cytotoxicity. Furthermore,
the isocyanide analogues synthesized in this study did not in-
potent growth inhibitory activity against the H5N1 virus. The À hibit the V27A or S31N mutant M2 ion channels, according to
NH2 to ÀNꢀC group replacement within current anti-influenza
drugs was found to give compounds with high activities at
low-micromolar concentrations. For example, a tenfold im-
electrophysiology experiments, and did not exhibit activity
against amantadine-resistant virus strains.
Introduction
Influenza virus infections remain one of the largest global
threats to both human health and the poultry industry.[1] Re-
cently, influenza has become an important public health con-
cern following the 2003–2009 potential avian influenza pan-
demic,[2] which was caused by a strain of highly pathogenic
type A, subtype H5N1 influenza virus.[1a] Currently, anti-influen-
za drug research is focused on three major targets: hemagglu-
tinin (HA), neuraminidase (NA), and the M2 proton channel.[3]
Of the three, antiviral drugs targeting NA (oseltamivir and za-
nanmivir) and M2 (amantadine and rimantadine) are currently
on the market for influenza treatment.[3a]
Amantadine (1) was the first compound to be used for anti-
influenza treatment in 1966,[4] and most research efforts have
focused on cyclic or acyclic aliphatic amine structures in order
to improve potency.[3a] Various amantadine derivatives have
been studied, including spiro structures,[5] benzyl-substituted
amines,[6] heterocyclic-linked amines,[7] or contracted rings.[8] Ri-
mantadine (5), an amantadine analogue, was successfully intro-
duced in 1994. Other amine-based compounds such as non-
adamantane-based M2 inhibitors, spiropiperidine,[9] organosi-
lane amines[10] and polycyclic amines[11] have also been investi-
gated in attempts to inhibit the influenza virus. Replacement
of the amino group with polar or basic groups, such as hy-
droxy, guanidine, amidine and hydroxylamine groups (Figure 1,
2–4), led to the development of compounds with moderate
viral inhibition relative to the original amine.[11b,12] These stud-
ies have confirmed that the amino group is the most impor-
tant polar functional group in M2 inhibitors (Figure 2, 5–8),
taking into account both the potency and drug-like physio-
chemical properties.[11a,13] After the screening of a focused
amine library, (1R,2R,3R,5S)-(À)-isopinocampheylamine (7) was
[a] Dr. S. Wu,+ J. Huang,+ S. He, L. Chen, Dr. J. Li, Prof. H. Zhou, Prof. W. Wang
Key Laboratory of Combinatorial Biosynthesis & Drug Discovery
Ministry of Education, and School of Pharmaceutical Sciences
Wuhan University, Wuhan 430071 (P.R. China)
E-mail: waw6@whu.edu.cn
[b] Dr. S. Wu+
State Key Laboratory of Virology of China
Wuhan University, Wuhan 430071 (P.R. China)
[c] Dr. S. Gazzarrini,+ Prof. A. Moroni
Department of Biosciences, University of Milan
and Institute of Biophysics (IBF) Milan, National Research Council (CNR)
Via Celoria 26, 20133 Milan (Italy)
[d] Dr. L. Xing
Worldwide Research & Development, Pfizer Inc.
200 Cambridge Park Drive, Cambridge, MA 02421 (USA)
[e] Dr. C. Li, Prof. L. Chen
Guangzhou Institute of Biomedicine & Health, Chinese Academy of Sciences
Guangzhou, Guangdong 510530 (P.R. China)
[f] Dr. C. G. Neochoritis, G. P. Liao, Prof. A. Dçmling
Department of Drug Design, University of Groningen
Antonius Deusinglaan 1, 9713 AV Groningen (The Netherlands)
[+] These authors contributed equally to this work.
Supporting information for this article is available on the WWW under
http://dx.doi.org/10.1002/cmdc.201500318.
Figure 1. Adamantanes with various polar groups from past studies.
ChemMedChem 2015, 10, 1837 – 1845
1837
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Papers
much better cell penetration, considering their lower clogP,
and may not exhibit the neurotoxicity associated with amines.
Isocyanide groups have also been discovered in some natural
compounds; however, few of these have shown biological ac-
tivity.[18] To the best of our knowledge, there is only one report
of biological (antifouling) activity for the natural isocyanide
compound 13 (Figure 3).[19]
The isocyanide group is not mentioned or listed in Rapid
Elimination of Swill (REOS)[20] or other lists of inappropriate re-
active functional groups.[21] Nevertheless, it is excluded from
drug library design as a reactive functional group, similar to
isocyanate.[22] In this study, the amino groups of six com-
pounds were replaced by isocyanide functional groups, and
the resulting compounds exhibit anti-H5N1 activity. Further-
more, the isocyanide group was not observed to have signifi-
cant toxicity, which is consistent with previous studies.[23] This
study aims to prove that the isocyanide functional group
could be used in drug discovery research.
Figure 2. Typical A/M2 inhibitors with amine functional groups.
Results and Discussion
found to be active.[14] Additionally, spiropiperidine 8 was one
of the amines without the adamantyl hydrophobic group that
showed superior activity over amantadine.[12]
Chemistry
Formamides and isocyanides were synthesized according to
known procedures (Scheme 1; see Procedures A and B in the
Experimental Section below).[23] Several of the isocyanide com-
The ammonium group mimicking a water cluster might sta-
bilize a diffusing hydronium ion; however, this mechanism
needs to be confirmed with more examples.[15] The necessity
of the amino group was also investigated during the develop-
ment of wild-type (WT) and mutant M2 (V27A) dual inhibitors,
such as the amantadine spiro derivative 9.[15] To develop inhibi-
tors against the mutant S31N M2 protein, the amine of the
amantadine group was linked with a hydroxybenzyl or isoxa-
zole moiety, yielding the adamantyl secondary amines 10 and
11.[6,7] Polycyclic amine 12 was discovered to be active against
the WT, V27A, and L26F mutant M2 channels.[11a] The amino
group may play an essential role in the inhibition of the wild-
type as well as the three M2 channel mutants.[15]
Scheme 1. Synthesis of formamides and isocyanides from amines.
pounds tested here were widely used in reactions, but their
anti-influenza activities have not been reported. We prepared
the corresponding isocyanides from three types of amines
(Figure 4). Type I amines are highly active in inhibiting the in-
fluenza virus and consist of amantadine (1), rimantadine (5),
(1R,2R,3R,5S)-(À)-isopinocampheylamine (7), 2-adamantylamine
(14), and 2,4,4-trimethylpentan-2-amine (15), the latter of
which has been proven to be the smallest type A M2 channel
(A/M2) inhibitor.[10]
The isocyanide group is a widely used reactive reagent in
multicomponent reactions and is frequently used in drug dis-
covery research.[16] Isocyanides can be represented either in
the zwitterionic or carbene form to illustrate their electronic
structures. The carbene structure may explain the nucleophilic
and electrophilic character of isocyanides, while the zwitterion-
ic structure is consistent with their dipole character with the
positive charge on the nitrogen and the negative charge on
the carbon (Figure 3).[17] The advantages of isocyanides are
their weak basic properties and increased hydrophobicity rela-
tive to the other nitrogen-containing functional groups, such
as amines and amidines. Moreover, isocyanides appear to have
Type II amines are aliphatic or aromatic amines, which were
chosen according to the size of their hydrophobic regions rela-
tive to amantadine.[12] The amines used in this study included
tert-butylamine 16, 4-tert-butylcyclohexylamine 17, 4-(tert-
butyl)aniline 18, 4-(tert-butyl)-2,6-dimethylaniline 19, 3-amino-
1-adamantanol 20, and bornylamine 21. Finally, type III amines
are amino acid ester derivatives. Both the formamide and the
corresponding isocyanide compounds were prepared for all
three types of amines. To the best of our knowledge, no bio-
logical activity has been reported in previous studies for these
isocyanides.
Figure 3. Typical isocyanide electronic representation (left) and an anti-
fouling isocyanide 13 (right).
ChemMedChem 2015, 10, 1837 – 1845
www.chemmedchem.org
1838
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Papers
known to bind to the M2 pro-
tein,[25] and its isocyanide deriva-
tive 32 showed an EC50 value of
9.8 mm. The amantadine deriva-
tives tert-octylamine 15 and tert-
butylamine 16 were also tested
in the antiviral assay: in the A/
M2 TEVC assay, tert-octylamine
15 exhibited activity similar to
that of amantadine (87% inhibi-
tion at 100 mm).[12] Both the for-
mamide and isocyanide ana-
logues of octylamine were pre-
pared, and the measured EC50
value of 3.8 mm for tert-octyliso-
cyanide 33 was similar to that of
amantadine.
On the other hand, the small-
er fragment tert-butylisocyanide
exhibited very low activity, with
EC50 >500 mm. Because amino
groups were considered to be
Figure 4. Three types of amine scaffolds were used in this study: type 1: known active M2 inhibitors, type II:
analogues of compounds from type I, type III: amino acid esters.
Antiviral activity
the most important pharmacophore in anti-influenza drug dis-
covery, 70 amines with different linear and branched alkyl or
aromatic groups were screened for activity. Five compounds
turned out to be potent in the virus inhibition assays.[14] In ad-
dition to compound 7, 4-tert-butylcyclohexylamine 17 (cis,trans
mixture) was also active. However, the activities of the forma-
mide 34 and isocyanide 35 derivatives of 4-tert-butylcyclohex-
ylamine 17 were lower than their amine analogue. Although 4-
(tert-butyl)aniline 18 and 4-(tert-butyl)-2,6-dimethylaniline 19
were not active against the influenza virus, their formamide
and isocyanide derivatives were still tested. Only isocyanide of
4-(tert-butyl)-2,6-dimethylaniline compound 36 exhibited anti-
viral activity, and its activity was similar to that of amantadine.
With an extra hydroxy group introduced in amantadine, the
formamide and isocyanide analogues of 3-amino-1-adamanta-
nol 20 proved to be inactive in our assays. Amongst the com-
pounds of the type II group, isocyanide 36 exhibited activity
even though its amine starting material was not active against
the influenza virus. Neither formamides nor isocyanides de-
rived from the type III amino esters 22–25 were active in our
tests.
The activity against influenza A/chicken/Hubei/327/2004
(H5N1) virus and cytotoxicity of the compounds were tested
using 3-[4,5-dimethylthiazol-2-yl]-2,5-diphenyltetrazolium bro-
mide (MTT) assays in MDCK cells (Table 1). In addition, the two-
electrode voltage clamp (TEVC) assay was applied for test com-
pound–M2 protein complexes to determine their pore block-
ing activity (Table 1). All tested compounds (including isocya-
nides) did not show cytotoxicity, with CC50 values greater than
100 mm. Amantadine formamide 26 and isocyanide 27 exhibit-
ed IC50 values of 0.38 and 0.48 mm, respectively, which are ap-
proximately tenfold lower than the IC50 value of 3.9 mm for
amantadine. Amantadine formamide 26 was tested in an early
structure–activity relationship (SAR) report and only exhibited
very weak activity (AVI50: 28 mgkgÀ1) in the in vivo mouse test
(relative to the amantadine AVI50 value of 4.6 mgkgÀ1).[24] The
weak in vivo inhibition could be attributed to poor bioavaila-
bility and/or rapid clearance of the formamide bond. Because
the formamides and isocyanides were found to be well tolerat-
ed in cell-based assays, these functional groups were chosen
as the amino group bioisosteres; 15 formamides and 15 iso-
cyanides were synthesized from 15 amines (Figure 4), and
a total of 12 compounds (four formamides and eight isocya-
nides) exhibited activity in our assay. All the data for the active
compounds are listed in Table 1.
To determine if the target of the isocyanides and forma-
mides is the M2 protein, the inhibitory activity of the com-
pounds was tested on A/M2 channels expressed in Xenopus
oocytes using the TEVC technique. All the inhibitors were
tested at a concentration of 100 mm. None of the tested com-
pounds proved to be more potent than amantadine in the
TEVC test. The inhibitory activities of formamides 26 (78.6%)
and 28 (81.8%) were stronger than their corresponding isocya-
nides 27 (68.2%) and 29 (42.1%); the activities were confirmed
by the IC50 values of 26 and 27 (26 mm and 35 mm); the IC50 of
29 was over 100 mm. This result suggests that isocyanide 29 in-
hibits the virus through another mechanism than M2 blockade.
Additionally, most of the other tested compounds exhibited
modest inhibition with the exception of 33, which appeared to
Rimantadine is another established and commercial drug,
and its activity is superior to that of amantadine. The activity
of rimantadine formamide was much lower than that of aman-
tadine formamide, but the activity of rimantadine isocyanide
(EC50: 0.238 mm) was similar to the amantadine isocyanide.
(1R,2R,3R,5S)-(À)-Isopinocampheylamine (7) was reported to be
active against amantadine-sensitive viruses from a primary
amine library screening;[14] the formamide 30 and isocyanide
31 derivatives of compound 7 exhibited respective EC50 values
of 2.76 and 7.36 mm our MTT assays. 2-Adamantylamine 14 is
ChemMedChem 2015, 10, 1837 – 1845
www.chemmedchem.org
1839
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Papers
Table 1. H5N1 inhibitory activities of the active formamide and isocya-
Table 1. (Continued)
nide compounds.
Compd
EC50 [mm][a]
2.14Æ0.64
A/M2 [%][b]
Compd
EC50 [mm][a]
3.97Æ1.32
A/M2 [%][b]
1
91Æ2.1
36
37
35.0Æ1.9
26
0.385Æ0.212
0.487Æ0.158
78.6Æ3.2
68.2Æ7.4
90.4Æ28.5
ND
27
28
[a] Values are derived from the results of an MTT assay performed with
MDCK cells infected with H5N1 virus (influenza strain A/chicken/Hubei/
327/2004); data are the mean ÆSEM of at least three independent ex-
periments (CC50 values were also derived from these assays, and were all
>100 mm).[b] Percent inhibition of wild-type A/M2 at 100 mm.
6.28Æ2.62
81.8Æ1.9
have low blockade of the M2 channel; its activity could there-
fore also be attributed to a mechanism other than A/M2 inhibi-
tion. Coincidentally, a novel antiviral mechanism has recently
been proposed for 2-aminoadamantanes.[26] None of the tested
compounds exhibited strong blockade of the mutant S31N
and V27A A/M2 proteins. Furthermore, the compounds were
also not active against the amantadine-resistant A/WS/33
H1N1 virus strain according to the in vitro assays. Finally,
a plaque-reduction assay was performed to confirm the activity
of compound 29 using the H5N1 virus. Indeed, it caused
a clear decrease in plaque size at a concentration of 1.2 mm
29
30
0.24Æ0.07
2.76Æ1.36
42.1Æ4.2
34.3Æ4.0
31
7.36Æ2.55
37.6Æ1.9
32
33
9.8Æ3.23
ND
3.88Æ2.37
6Æ2.5
34
29.5Æ8.52
32.6Æ1.7
35
125Æ42.6
32.9Æ1.7
Figure 5. Plaque-reduction assay with the H5N1 virus. MDCK cells were in-
fected with influenza virus (H5N1) in the presence of test compounds aman-
tadine (1) and compound 29 at four concentrations as indicated. Plaques
were visualized by crystal violet staining.
ChemMedChem 2015, 10, 1837 – 1845
www.chemmedchem.org
1840
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Papers
(Figure 5). The antiviral activity of the amine derivatives was
maintained or improved upon replacement of the amino
groups with formamide or isocyanide groups. Rimantadine iso-
cyanide 29 was the most potent in the H5N1 inhibition assay,
whereas rimantadine formamide 28 was one-sixth less active
relative to amantadine formamide 26. The amantadine and ri-
mantadine isocyanides showed stronger activity than the other
isocyanides. To understand how isocyanides may have a differ-
ence in M2 activity, isocyanides were compared with the re-
ported active amines by molecular topology. The measured ac-
tivities were consistent with a previous SAR study on aminoa-
damantane derivatives and are discussed below.
In an early pioneering SAR study of amantadines, researchers
used animal models to assess antiviral activity.[24,27] However,
SAR data were only listed without any rational explanations.
Years after the biological target of amantadine was proposed
to be the M2 protein based on analysis of mutations,[28] many
new amantadine analogues were tested against various influ-
enza virus strains. The most active derivatives, such as amanta-
dine and rimantadine, were also found to be active against
H2N2.[5d] From that in vivo activity data, we could conclude
that the amino group can only tolerate a two-bond distance
from the carbon atom of adamantane (a bulky hydrophobic
group), whereas amines with a distance of over four bond
lengths connecting the adamantane appear to have decreased
activity.[4,5c,d] Furthermore, the size of amantadine was also in-
vestigated by SAR studies. Whereas 3-methyladamantan-1-
amine, with a hydrophobic group radius of 3.6 Š was active
(the radius was calculated based on the van der Waals radius
of the atoms and the bond length using MarvinSpace Marvin
14.9.22, 2014, ChemAxon), 3,5,7-trimethyladamantan-1-amine,
with a radius of 4.2 Š, was not active.[24] According to the re-
ported SAR data and the size of the M2 channel, drug activity
is thus dependent on the size of the hydrophobic moiety and
the length of the linker that connects the moiety with the
amine. This conclusion is continuously confirmed by the re-
cently reported active amines.[8,29]
Figure 6. Three orientations of amine compounds have been proposed in
the literature: the N atom of the amine is illustrated as a blue sphere, and
the pink shadow represents the M2 channel. The amantadine amine 1 (left)
points downward to the C terminus, whereas the polycyclic amine 38 is di-
rected toward the side (middle),[8] and the adamantyl secondary amine 11
(right) points toward the N terminus (with S31N M2 channel). Image pro-
duced by Mole2 and PyMOL.[31b]
ate activity against H1N1 was reported.[29] Azapropellane amine
39 has respective dimensions of 4.9 and 5.8 Š in the horizontal
and perpendicular directions, and the size of both sides are
longer than 4.2 Š (the radius of M2 channel) (Figure 6). The
possibility that the amine may point to the N terminus is not
ruled out when the amine is attached to other substituents.[6,7]
For isoxazole analogue 11, the adamantane moiety of the com-
pound is bound to the spacious pore between Asn31 and
Gly34. When blocking S31N, the aromatic group could face up
toward the N terminus based on the solution NMR structure
data and modeling analysis.[31]
The channel size limitation was also demonstrated by com-
pound 38; its M2 blocking activity in TEVC experiments has
been confirmed;[8] however, larger azapropellanes did not
block the A/M2 proton channel even though the compounds
exhibited activity against the H1N1 strain.[29] Thus, the M2 ac-
tivity seems to be very sensitive to the size of the hydrophobic
moiety.
Activities of isocyanide compounds were also studied based
on structural information. Considering the structural informa-
tion from 2KQT, a large space remains between the amine ni-
trogen atom and the pseudo-center of the four His37 residues.
The distance between these constituents is 7.6 Š. From the
published SAR results, the M2 channel can only tolerate amino
groups with limited lengths. The radius analysis showed that
there is an extra channel of ~5 Š for the amantadine amine
before the radius of the channel is significantly reduced (“extra
space” shown in Figure 7). In many studies, water molecules
are present in the top space of the His37 tetramer network.
Therefore, a remaining length of ~3 Š is present if one water
molecule is included, which is sufficient space for an extra
bond or atom. The water molecules were precisely determined
to be located on the His47 tetramers by high-resolution X-ray
analysis (PDB code: 3LBW).[32] The location of the water mole-
cules on the His47 tetramers could increase the antiviral activi-
ty of the isocyanide relative to the corresponding amine by al-
The pore binding site was considered to be a pharmacologi-
cally relevant drug binding site supported by X-ray crystal
(PDB codes: 3C9J, 2LJC) and solid-state NMR (PDB: 2KQT)
structures.[30] In the determined wild-type structures, the amine
points to the C terminus of the channel. For amine 38, the
amino group is oriented toward the walls of the channel and
not in the top (N-terminal) or bottom (C-terminal) directions
according to the molecular dynamics (MD) results.[8] This result
can be explained by the size of the scaffold, as it has a perpen-
dicular dimension of 4.0 Š and a horizontal length of 7.1 Š,
which allows the molecule to fit in the channel in the perpen-
dicular direction. For comparison, azapropellane amine 39 did
not inhibit the A/M2 channel well, despite the fact that moder-
ChemMedChem 2015, 10, 1837 – 1845
www.chemmedchem.org
1841
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Papers
isocyanide 36 was active against
the H5N1 influenza virus, where-
as its corresponding amine 19
was not. Our results suggest that
isocyanides can be used as
a
novel replacement in the
design of proton channel inhibi-
tors.
Experimental Section
Chemistry
General: Unless stated otherwise,
reagents and solvents used in the
preparation of the following com-
pounds were purchased from com-
mercial sources (J&K, Aladdin,
Adamas-Beta) and used as re-
ceived. All NMR spectra were re-
corded at 400 MHz on a Bruker In-
struments NMR unless otherwise
stated. Chemical shifts are ex-
pressed as d units using tetrame-
thylsilane as the external standard
(s: singlet, d: doublet, t: triplet, q:
quartet, m: multiplet, br: broad
Figure 7. Channel radius distribution of pores in the protein–inhibitor complex; red brackets refer to the extra
space for potential inhibitors.
peak). TLC analysis was performed using Aldrich 254 nm plates
(60 Š, 250 mm) and visualized using UV, PMA, KMnO4, and ninhy-
drin stains. The purities of test compounds are measured over
95% by qNMR protocol.
lowing the isocyanide to fit in the channel space when the
amine is directed toward the C terminus. No structural infor-
mation is available from previous studies regarding the interac-
tions between isocyanide or formamide analogues and the M2
protein. Therefore, the mechanism that allows isocyanides to
block the channel is similar to the mechanism of amantadine.
The activities of the various isocyanide derivatives are dis-
cussed based on molecular modeling and docking analysis.
Molecular docking was performed to determine the binding
details of the channel–compound complex (Supporting Infor-
mation). The formamide group of 26 and isocyanide group of
27 are oriented toward the viral interior, fitting at the portion
of the channel. Compounds 26 and 27 can fit in the channel
well, but 29 could not be placed in the M2 channel adequate-
ly, with the NꢀC group pointing toward the C terminus. This
could explain the lack of M2 activity of compound 29.
Procedure A. Synthesis of formamides: Formic acid (7.0 equiv)
and acetic anhydride (6.0 equiv) were allowed to mix for 1 h at RT,
after which the appropriate amine (1.0 equiv) was added dropwise
to the mixture at 08C. The ice bath was removed, and the reaction
was stirred at RT until the amine was consumed completely as de-
termined by TLC. The solution was diluted with CH2Cl2 (100 mL),
washed with water (100 mL), the aqueous phase was then extract-
ed with CH2Cl2 (2”100 mL), and the combined organic phases
were separated, dried (Na2SO4), filtered, and concentrated to give
the desired product without purification.
Procedure B. Synthesis of isocyanides: A stirred solution of for-
mamide (1.0 equiv) in CH2Cl2 (50 mL) at 08C was treated with Et3N
(5.0 equiv), followed by POCl3 (1.0 equiv) dropwise over 15 min.
The reaction was allowed to warm to RT while stirring. After
10 min, the reaction was stopped by the addition of water (50 mL),
and the organic layer was extracted with CH2Cl2 (2”50 mL). The
combined organic extracts were dried (Na2SO4), filtered and con-
centrated to give the crude product. Purification by flash column
chromatography (PE/EtOAc) gave the desired isocyanide.
Replacement of an isocyanide group with other inactive
amine-containing moieties such as branched aliphatic amines,
aromatic amines, or amino acid esters, did not promote antivi-
ral activity. Therefore, the conclusion that the isocyanide group
does not exhibit antiviral activity on its own can be drawn.
N-((3S,5S,7S)-Adamantan-1-yl)formamide (26): Following proce-
dure A, (3S,5S,7S)-adamantan-1-amine (1 g, 6.61 mmol) gave 26 as
an off-white solid (1.12 g, 95%): H NMR (400 MHz, [D6]DMSO): d=
Conclusions
1
Replacement of the amino group with functional groups such
as formamides and isocyanides have shown activity in cell-
based viral inhibition and TEVC assays. The present work clear-
ly demonstrates that the isocyanide group can be used to re-
place the amino group in anti-influenza compounds without
causing cytotoxicity. Additionally, amantadine isocyanide 27
exhibited stronger activity than its parent amine. In one case,
7.82 (d, J=1.8 Hz, 1H), 2.01 (d, J=11.3 Hz, 6H), 1.91 (s, 6H),
1.75 ppm (d, J=1.7 Hz, 3H); 13C NMR (100 MHz, CDCl3, one major
rotamer): d=163.9, 52.1, 47.7, 46.5, 46.2, 39.3, 38.4, 35.2, 28.0, 23.3,
20.7 ppm; GC–MS (EI): m/z=179.1 [M]+; HRMS (ESI): m/z [M+H]+
calcd for C11H17NO: 180.1388, found: 180.1376; Purity: >98%.
1-Isocyanoadamantane (27): Following procedure B, N-((3S,5S,7S)-
adamantan-1-yl)formamide (3 g, 16.73 mmol) gave 27 as an off-
ChemMedChem 2015, 10, 1837 – 1845
www.chemmedchem.org
1842
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Papers
white solid (2.02 g, 78%): 1H NMR (400 MHz, CDCl3): d=2.09 (s,
3H), 2.02 (s, 6H), 1.68 ppm (d, J=14.3 Hz, 6H); 13C NMR (100 MHz,
CDCl3): d=151.6, 54.2, 43.5 (3C), 35.5 (3C), 28.7 (3C) ppm; MS (EI):
m/z=161.1 [M]+; HRMS (ESI): m/z [M+Na]+ calcd for C11H15N:
184.1102, found: 184.1124; Purity: >97%.
N-(4-(tert-Butyl)cyclohexyl)formamide (34): Following procedure
A, N-(4-(tert-butyl)cyclohexyl)formamide (5 g, 32.19 mmol) gave 34
as a yellow oil (5.78 g, 98%): H NMR (400 MHz, CDCl3): d=1.97 (d,
J=20.5 Hz, 2H), 1.83 (dd, J=29.3, 12.4 Hz, 2H), 1.65 (d, J=11.9 Hz,
1H), 1.53 (t, J=11.8 Hz, 1H), 1.10 (m, J=32.9, 8.5 Hz, 4H), 0.85–
0.86 ppm (m, 9H); 13C NMR (100 MHz, CDCl3, one major rotamer):
d=160.6, 47.6, 43.1, 33.4 (2C), 30.5 (3C), 27.5 (2C), 21.7 ppm; MS
(EI): m/z=183.1 [M]+; HRMS (ESI): m/z [M+H]+ calcd for C11H21NO:
184.1701, found: 184.1686; Purity: >96%.
1
N-((3S,5S,7S)-Adamantan-1-yl)formamide (28): Following proce-
dure A,
N-(1-((3R,5R,7R)-adamantan-1-yl)ethyl)
amine
(3 g,
13.91 mmol) gave 28 as an off-white oil (1.02 g, 87%): 1H NMR
(400 MHz, [D6]DMSO): d=7.82 (d, J=1.8 Hz, 1H), 2.01 (d, J=
11.3 Hz, 6H), 1.91 (s, 6H), 1.75 ppm (d, J=1.7 Hz, 3H); 13C NMR
(100 MHz, CDCl3, one major rotamer): d=160.8, 51.8, 36.9 (3C),
36.8 (3C), 35.6, 28.2 (3C), 14.5 ppm; MS (EI): m/z=207.1 [M]+;
HRMS (ESI): m/z [M+H]+ calcd for C13H21NO: 208.1701, found:
208.1684; Purity: >99%.
1-(tert-Butyl)-4-isocyanocyclohexane (35): Following procedure B,
N-(4-(tert-butyl)cyclohexyl)formamide (2 g, 10.91 mmol) gave 35 as
a
yellow oil (0.97 g, 54%): 1H NMR of 1:1 trans/cis mixture
(400 MHz, CDCl3): d=2.22 (d, J=12.4 Hz, 1H), 2.01 (d, J=10.9 Hz,
1H), 1.81 (d, J=10.4 Hz, 1H), 1.67 (d, J=10.7 Hz, 1H), 1.53 (d, J=
11.7 Hz, 1H), 1.44 (t, J=11.4 Hz, 2H), 1.25 (s, 1H), 0.98 (d, J=
10.2 Hz, 2H), 0.90–0.82 ppm (m, 9H, cis/trans=1:1); 13C NMR
(100 MHz, CDCl3, 1:1 trans/cis mixture): d=162.6, 52.3 and 50.4 (cis
and trans mixture), 47.4 and 46.6 (cis and trans mixture), 34.1, 32.5
and 32.3 (cis and trans mixture), 31.4, 27.4 (3C), 25.5, 21.2
(2C) ppm; MS (EI): m/z=164.7 [M]+; HRMS (ESI): m/z [M+H]+ calcd
for C11H20N: 166.1596, found: 166.1586; Purity: >95%.
(3R,5R,7R)-1-(1-Isocyanoethyl)adamantine (29): Following proce-
dure B, N-((3S,5S,7S)-adamantan-1-yl)formamide (1.5 g, 7.24 mmol)
1
gave 29 as a white solid (1.02 g, 74%): H NMR (400 MHz, CDCl3):
d=3.29–3.17 (m, 1H), 1.73 (d, J=12.3 Hz, 4H), 1.68–1.58 (m, 7H),
1.55 (s, 4H), 1.33–1.22 ppm (m, 3H); 13C NMR (100 MHz, CDCl3): d=
154.6, 60.6, 37.9 (3C), 36.7 (3C), 35.0, 28.1 (3C), 15.1 ppm; MS (EI):
m/z=189.10 [M]+; Purity: >97%.
5-(tert-Butyl)-2-isocyano-1,3-dimethylbenzene (36): Following
procedure B,
N-(4-(tert-butyl)-2,6-dimethylphenyl)formamide
N-((1R,2S,3R,5S)-2,6,6-Trimethylbicyclo[3.1.1]heptan-3-yl)forma-
mide (30): Following procedure A, (1R,2S,3R,5S)-2,6,6-trimethylbicy-
clo[3.1.1]heptan-3-amine (1 g, 6.52 mmol) gave 30 as a clear color-
less oil (1.14 g, 96%): 1H NMR (400 MHz, CDCl3): d=8.08 (dd, J=
35.5, 23.6 Hz, 1H), 4.42–4.21 (m, 1H), 2.66–2.55 (m, 1H), 2.43 (dt,
J=6.0, 5.6 Hz, 1H), 2.04–1.92 (m, 2H), 1.77 (ddt, J=22.7, 14.0,
7.2 Hz, 3H), 1.59–1.50 (m, 1H), 1.23 (t, J=4.3 Hz, 4H), 1.03 (d, J=
11.1 Hz, 4H), 0.86 ppm (dd, J=9.9, 3.1 Hz, 1H); 13C NMR (100 MHz,
CDCl3, one major rotamer): d=163.9, 52.1, 47.7, 46.5, 46.2, 38.4,
35.2, 30.0, 23.3, 20.7, 19.8 ppm; MS (EI): m/z=182.0 [M+H]+;
HRMS (ESI): m/z [M+H]+ calcd for C11H19NO: 182.1545, found:
182.1530; Purity: >95%.
(200 mg, 0.97 mmol) gave 36 as a white solid (90 mg, 34%):
1H NMR (400 MHz, CDCl3): d=7.02 (s, 2H), 2.34 (s, 6H), 1.21 ppm (s,
9H); 13C NMR (101 MHz, CDCl3): d=164.4, 152.0, 134.3 (2C), 124.8
(3C), 34.6, 31.2 (3C), 19.2 (2C) ppm; MS (EI): m/z=205.1 [M+H2O]+;
HRMS (ESI) m/z [M+H]+ calcd for C13H17N: 188.1439, found:
188.1448; Purity: >95%.
(1R,3R,5R,7S)-3-Isocyanoadamantan-1-ol (37): Following proce-
dure B, N-((1R,3S,5R,7S)-3-hydroxyadamantan-1-yl)formamide (2 g,
10.25 mmol) gave 37 as a white solid (1.02 g, 56%): 1H NMR
(400 MHz, CDCl3): d=7.96 (s, 1H), 2.38 (d, J=28.9 Hz, 4H), 2.14–
1.90 (m, 8H), 1.72–1.50 ppm (m, 2H); 13C NMR (101 MHz, CDCl3):
d=159.8, 79.8, 55.7, 46.8, 42.3 (2C), 39.7 (2C), 34.0, 30.0 (2C) ppm;
MS (EI): m/z=177.1 [M]+; HRMS (ESI): m/z [M+H]+ calcd for
C11H15NO: 178.1232, found: 178.1217; Purity: >96%.
(1R,2S,3R,5S)-3-Isocyano-2,6,6-trimethylbicyclo[3.1.1]heptane
(31): Following procedure B, N-((1R,2S,3R,5S)-2,6,6-trimethylbicy-
clo[3.1.1]heptan-3-yl)formamide (1 g, 5.52 mmol) gave 31 as
1
a yellow oil (0.71 g, 74%): H NMR (400 MHz, CDCl3): d=3.82 (ddd,
J=8.2, 4.6, 2.2 Hz, 1H), 2.63–2.54 (m, 1H), 2.52–2.41 (m, 1H), 2.32
(ddd, J=9.4, 6.1, 2.4 Hz, 1H), 2.18 (s, 1H), 2.08 (dtd, J=14.1, 5.7,
2.7 Hz, 1H), 1.99 (td, J=5.6, 2.7 Hz, 1H), 1.88–1.81 (m, 1H), 1.26–
1.17 (m, 6H), 0.98–0.88 ppm (m, 3H).13C NMR (100 MHz, CDCl3): d=
153.3, 47.1, 45.9, 40.9, 38.2, 36.3, 34.3, 27.8, 23.3, 20.5 ppm; GC–MS
(CI): m/z=162.1 [MÀH]À; HRMS (ESI): m/z [M+H]+ calcd for
C11H17N: 164.1439, found: 164.1506; Purity: >95%.
Biological methods
Two-electrode voltage clamp analysis: A/Udorn/72 cDNA, encoding
the A/M2 protein, was cloned into a pSUPER vector (a modified
version of pGEM-HE). A/M2 V27A and A/M2 S31N single mutants
were generated by QuicChange site-directed mutagenesis kit (Sta-
tagene). cRNAs were transcribed in vitro using T7 RNA polymerase
(Promega) as described previously.[33] Oocytes were prepared ac-
cording to standard methods and injected with 50 mL water or
50 nL cRNA (0.8 gLÀ1) and incubated in ND96 solution (96 mm
NaCl, 2 mm KCl, 1.8 mm CaCl2 1 mm MgCl2, 5 mm HEPES, pH 7.5
with NaOH 5 mm sodium pyruvate, 50 mgmLÀ1 gentamicin). Meas-
urements were performed 2–5 days after injection. Currents were
recorded by two-electrode voltage clamp, using the Gene-Clamp
500 amplifier (Axon Instruments) under control of pCLAMP7. Oo-
cytes were perfused at room temperature at a rate of 2 mLminÀ1
in Barth’s solution containing 88 mm NaCl, 1 mm KCl, 2.4 mm
NaHCO3, 0.3 mm NaNO3, 0.71 mm CaCl2, 0.82 mm MgCl2, and
15 mm HEPES for pH 8.5, or 15 mm MES for pH 5.5. Currents were
recorded at À20 mV. Compounds were dissolved in DMSO (0.5%)
and applied at pH 5.5 at 100 mm when the inward currents reach
maximum. The compounds were applied for 2 min and the cur-
rents before and after application of drugs were compared. Data
were analyzed using ORIGIN7.
(1R,3R,5R,7R)-2-Isocyanoadamantane (32): Following procedure B,
(adamantan-2-yl)formamide (1 g, 5.59 mmol) gave 32 as a white
solid (0.54 g, 60%): 1H NMR (400 MHz, CDCl3): d=3.81 (s, 1H),
2.22–2.08 (m, 4H), 1.95–1.82 (m, 4H), 1.80–1.57 ppm (m, 6H);
13C NMR (101 MHz, CDCl3): d=154.1, 59.1, 37.1 (2C), 32.7 (2C), 31.2
(3C), 26.6 (2C) ppm; MS (EI): m/z=161.1 [M]+; HRMS (ESI): m/z
[M+Na]+ calcd for C11H15N: 184.1102, found: 184.1104; Purity:
>95%.
2-Isocyano-2,4,4-trimethylpentane (33): Following procedure B,
N-(2,4,4-trimethylpentan-2-yl)formamide (1 g, 6.37 mmol) gave 33
1
as a yellow oil (0.556 g, 63%): H NMR (400 MHz, DMSO): d=1.60–
1.57 (m, 2H), 1.46–1.41 (m, 6H), 1.04 ppm (s, 9H); 13C NMR
(101 MHz, CDCl3): d=152.9, 56.9, 53.8, 31.8 (2C), 31.6 (3C),
31.0 ppm; HRMS (ESI): m/z [M+H]+ calcd for C9H18N: 140.1439,
found: 140.1443; Purity: >96%.
ChemMedChem 2015, 10, 1837 – 1845
www.chemmedchem.org
1843
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Papers
Das, J. Med. Chem. 2012, 55, 6263–6277; d) F. Li, C. Ma, J. Wang, Curr.
Med. Chem. 2015, 22, 1361–1382.
[4] W. L. Davies, R. R. Grunert, R. F. Haff, J. W. McGahen, E. M. Neumayer, M.
Paulshock, J. C. Watts, T. R. Wood, E. C. Hermann, C. E. Hoffmann, Science
1964, 144, 862–863.
Antiviral assay: Monolayers of MDCK cells were seeded in 12-well
culture plates and incubated overnight. The cells were washed
with phosphate-buffered saline (PBS) twice before being infected
with 1 mL per well of a given dilution of the H5N1 virus (influenza
strain A/chicken/Hubei/327/2004) in the maintenance medium
without trypsin at a multiplicity of infection (MOI) of 0.01. After in-
cubation at 378C in 5% CO2 for 2 h, the medium was removed,
and the plates were washed with PBS for three times, after which
1 mL per well of 1.5% agarose overlay (1 mL) containing Eagle’s
minimal essential medium (MEM), trypsin (2 mgmLÀ1), and various
concentrations of drugs or test compounds were added. Plates
were incubated at 378C in 5% CO2 for a further 3–4 days to check
for plaque formation. The agars were then carefully removed, and
the monolayer of cells was fixed with 10% formalin, stained with
crystal violet (0.5%), and the plaques were counted. The percent-
age of plaque inhibition relative to the infected control (no drug)
plates was determined for each drug concentration, and the 50%
effective concentration (EC50) was calculated by the Reed–Muench
method. All data are obtained for at least three independent ex-
periments.
[5] a) K. Lundahl, J. Schut, J. L. M. A. Schlatmann, G. B. Paerels, A. Peters, J.
Med. Chem. 1972, 15, 129–132; b) R. Van Hes, A. Smit, T. Kralt, A. Peters,
J. Med. Chem. 1972, 15, 132–136; c) N. Kolocouris, G. B. Foscolos, A. Ko-
locouris, P. Marakos, N. Pouli, G. Fytas, S. Ikeda, E. De Clercq, J. Med.
Chem. 1994, 37, 2896–2902; d) N. Kolocouris, A. Kolocouris, G. B. Fosco-
los, G. Fytas, J. Neyts, E. Padalko, J. Balzarini, R. Snoeck, G. Andrei, E. De
Clercq, J. Med. Chem. 1996, 39, 3307–3318; e) G. Zoidis, C. Fytas, I. Pa-
panastasiou, G. B. Foscolos, G. Fytas, E. Padalko, E. De Clercq, L. Nae-
sens, J. Neyts, N. Kolocouris, Bioorg. Med. Chem. 2006, 14, 3341–3348;
f) N. Kolocouris, G. Zoidis, G. B. Foscolos, G. Fytas, S. R. Prathalingham,
J. M. Kelly, L. Naesens, E. De Clercq, Bioorg. Med. Chem. Lett. 2007, 17,
4358–4362; g) G. Stamatiou, G. B. Foscolos, G. Fytas, A. Kolocouris, N.
Kolocouris, C. Pannecouque, M. Witvrouw, E. Padalko, J. Neyts, E. D.
Clercq, Bioorg. Med. Chem. 2003, 11, 5485–5492.
[6] J. Wang, C. Ma, J. Wang, H. Jo, B. Canturk, G. Fiorin, L. H. Pinto, R. A.
Lamb, M. L. Klein, W. F. DeGrado, J. Med. Chem. 2013, 56, 2804–2812.
[7] a) J. Wang, Y. Wu, C. Ma, G. Fiorin, J. Wang, L. H. Pinto, R. A. Lamb, M. L.
Klein, W. F. DeGrado, Proc. Natl. Acad. Sci. USA 2013, 110, 1315–1320;
b) W. Zhang, J. Xu, F. Liu, C. Li, Y. Jie, S. Chen, Z. Li, J. Liu, L. Chen, G.
Zhou, Chin. J. Chem. 2010, 28, 1417–1423.
Plaque-reduction assay: MDCK cells were infected with influenza
virus strain A/chicken/Hubei/327/2004 (H5N1) at 0.01 MOI for 2 h
at 378C. After viral adsorption, the cell monolayer was covered
with overlay medium containing compounds and further cultured
at 378C in 5% CO2 for 48–72 h. The overlay medium was then re-
moved, the cell monolayer was fixed with 10% formalin, stained
with 0.5% crystal violet, and plaques were counted.
[8] M. D. Duque, C. Ma, E. Torres, J. Wang, L. Naesens, J. Juµrez-JimØnez, P.
Camps, F. J. Luque, W. F. DeGrado, R. A. Lamb, L. H. Pinto, S. Vµzquez, J.
Med. Chem. 2011, 54, 2646–2657.
[9] J. Wang, S. D. Cady, V. Balannik, L. H. Pinto, W. F. DeGrado, M. Hong, J.
Am. Chem. Soc. 2009, 131, 8066–8076.
[10] J. Wang, C. Ma, Y. Wu, R. A. Lamb, L. H. Pinto, W. F. DeGrado, J. Am.
Chem. Soc. 2011, 133, 13844–13847.
[11] a) M. Rey-Carrizo, M. Barniol-Xicota, C. Ma, M. FrigolØ-Vivas, E. Torres, L.
Naesens, S. LlabrØs, J. Juµrez-JimØnez, F. J. Luque, W. F. DeGrado, R. A.
Lamb, L. H. Pinto, S. Vµzquez, J. Med. Chem. 2014, 57, 5738–5747; b) M.
Rey-Carrizo, E. Torres, C. Ma, M. Barniol-Xicota, J. Wang, Y. Wu, L. Nae-
sens, W. F. DeGrado, R. A. Lamb, L. H. Pinto, S. Vµzquez, J. Med. Chem.
2013, 56, 9265–9274; c) M. Rey-Carrizo, S. Gazzarrini, S. LlabrØs, M. Fri-
golØ-Vivas, J. Juµrez-JimØnez, M. Font-Bardia, L. Naesens, A. Moroni, F. J.
Luque, S. Vµzquez, Eur. J. Med. Chem. 2015, 96, 318–329.
Molecular modeling
Tetrameric M2 transmembrane bundle of the influenza A virus
(PDB code: 3LBW) was used as model structure. Molecular docking
was done using AutoDock 3.0.5. The channel analysis was finished
using Mole2 and PyMOL. The measurement of hydrophobic groups
was carried out with ChemAxon.
[12] J. Wang, C. Ma, V. Balannik, L. H. Pinto, R. A. Lamb, W. F. DeGrado, ACS
Med. Chem. Lett. 2011, 2, 307–312.
[13] J. Wang, F. Li, C. Ma, Pept. Sci. 2015, 104, 291–309.
[14] W. Hu, S. Zeng, C. Li, Y. Jie, Z. Li, L. Chen, J. Med. Chem. 2010, 53, 3831–
3834.
Acknowledgements
[15] J. Wang, C. Ma, G. Fiorin, V. Carnevale, T. Wang, F. Hu, R. A. Lamb, L. H.
Pinto, M. Hong, M. L. Klein, W. F. DeGrado, J. Am. Chem. Soc. 2011, 133,
12834–12841.
[16] A. Dçmling, W. Wang, K. Wang, Chem. Rev. 2012, 112, 3083–3135.
[17] R. Ramozzi, N. Cheron, B. Braida, P. C. Hiberty, P. Fleurat-Lessard, New J.
Chem. 2012, 36, 1137–1140.
[18] a) M. J. Garson, J. S. Simpson, Nat. Prod. Rep. 2004, 21, 164–179; b) P. J.
Scheuer, Acc. Chem. Res. 1992, 25, 433–439; c) M. S. Edenborough, R. B.
Herbert, Nat. Prod. Rep. 1988, 5, 229–245.
This project was supported by the National Natural Science Foun-
dation of China (21202125, 31300060), the Hubei Province Natu-
ral Science Foundation (2014CFB241), the Scientific Research
Foundation for the Returned Overseas Chinese Scholars, and the
Fundamental Research Funds for the Central Universities.
Keywords: amantadines
· antiviral agents · influenza ·
[19] a) Y. Nogata, Y. Kitano, E. Yoshimura, K. Shinshima, I. Sakaguchi, Biofoul-
ing 2004, 20, 87–91; b) Y.-F. Zhang, Y. Kitano, Y. Nogata, Y. Zhang, P.-Y.
Qian, PLoS One 2012, 7, e45442; c) Y. Nogata, Y. Kitano in Antifouling
Compounds, Vol. 42 (Eds.: N. Fusetani, A. Clare), Springer, Heidelberg,
2006, pp. 87–104; d) Y. Kitano, C. Akima, E. Yoshimura, Y. Nogata, Bio-
fouling 2011, 27, 201–205.
[20] W. P. Walters, M. Namchuk, Nat. Rev. Drug Discovery 2003, 2, 259–266.
[21] J. B. Baell, G. A. Holloway, J. Med. Chem. 2010, 53, 2719–2740.
[22] G. Schneider, K.-H. Baringhaus in De novo Molecular Design, Wiley-VCH,
2013, pp. 1–55.
[23] I. Ugi, U. Fetzer, U. Eholzer, H. Knupfer, K. Offermann, Angew. Chem. Int.
Ed. Engl. 1965, 4, 472–484; Angew. Chem. 1965, 77, 492–504.
[24] P. E. Aldrich, E. C. Hermann, W. E. Meier, M. Paulshock, W. W. Prichard,
J. A. Synder, J. C. Watts, J. Med. Chem. 1971, 14, 535–543.
[25] A. Kolocouris, P. Spearpoint, S. R. Martin, A. J. Hay, M. López-Querol, F. X.
Sureda, E. Padalko, J. Neyts, E. De Clercq, Bioorg. Med. Chem. Lett. 2008,
18, 6156–6160.
inhibitors · isocyanides · M2 channels
[1] a) K. S. Li, Y. Guan, J. Wang, G. J. D. Smith, K. M. Xu, L. Duan, A. P. Rahard-
jo, P. Puthavathana, C. Buranathai, T. D. Nguyen, A. T. S. Estoepangestie,
A. Chaisingh, P. Auewarakul, H. T. Long, N. T. H. Hanh, R. J. Webby,
L. L. M. Poon, H. Chen, K. F. Shortridge, K. Y. Yuen, R. G. Webster, J. S. M.
Peiris, Nature 2004, 430, 209–213; b) S. Tong, X. Zhu, Y. Li, M. Shi, J.
Zhang, M. Bourgeois, H. Yang, X. Chen, S. Recuenco, J. Gomez, L.-M.
Chen, A. Johnson, Y. Tao, C. Dreyfus, W. Yu, R. McBride, P. J. Carney, A. T.
Gilbert, J. Chang, Z. Guo, C. T. Davis, J. C. Paulson, J. Stevens, C. E. Rup-
precht, E. C. Holmes, I. A. Wilson, R. O. Donis, PLoS Pathog. 2013, 9,
e1003657.
[2] Y. Poovorawan, S. Pyungporn, S. Prachayangprecha, J. Makkoch, Pathog.
Global Health 2013, 107, 217–223.
[3] a) E. De Clercq, Nat. Rev. Drug Discovery 2006, 5, 1015–1025; b) J. Du,
T. A. Cross, H.-X. Zhou, Drug Discovery Today 2012, 17, 1111–1120; c) K.
ChemMedChem 2015, 10, 1837 – 1845
www.chemmedchem.org
1844
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Papers
[26] A. Kolocouris, C. Tzitzoglaki, F. B. Johnson, R. Zell, A. K. Wright, T. A.
Cross, I. Tietjen, D. Fedida, D. D. Busath, J. Med. Chem. 2014, 57, 4629–
4639.
17987–17995; b) D. Sehnal, R. Svobodova Varekova, K. Berka, L. Pravda,
V. Navratilova, P. Banas, C.-M. Ionescu, M. Otyepka, J. Koca, J. Cheminf.
2013, 5, 39.
[27] J. G. Whitney, W. A. Gregory, J. C. Kauer, J. R. Roland, J. A. Snyder, R. E.
Benson, E. C. Hermann, J. Med. Chem. 1970, 13, 254–260.
[28] a) A. J. Hay, A. J. Wolstenholme, J. J. Skehel, M. H. Smith, EMBO J. 1985,
4, 3021–3024; b) C. Wang, K. Takeuchi, L. H. Pinto, R. A. Lamb, J. Virol.
1993, 67, 5585–5594.
[32] R. Acharya, V. Carnevale, G. Fiorin, B. G. Levine, A. L. Polishchuk, V. Balan-
nik, I. Samish, R. A. Lamb, L. H. Pinto, W. F. DeGrado, M. L. Klein, Proc.
Natl. Acad. Sci. USA 2010, 107, 15075–15080.
[33] B. Plugge, S. Gazzarrini, M. Nelson, R. Cerana, J. L. Van Etten, C. Derst, D.
DiFrancesco, A. Moroni, G. Thiel, Science 2000, 287, 1641–1644.
[29] E. Torres, R. Leiva, S. Gazzarrini, M. Rey-Carrizo, M. FrigolØ-Vivas, A.
Moroni, L. Naesens, S. Vµzquez, ACS Med. Chem. Lett. 2014, 5, 831–836.
[30] R. M. Pielak, K. Oxenoid, J. J. Chou, Structure 2011, 19, 1655–1663.
[31] a) Y. Wu, B. Canturk, H. Jo, C. Ma, E. Gianti, M. L. Klein, L. H. Pinto, R. A.
Lamb, G. Fiorin, J. Wang, W. F. DeGrado, J. Am. Chem. Soc. 2014, 136,
Received: July 21, 2015
Published online on September 7, 2015
ChemMedChem 2015, 10, 1837 – 1845
www.chemmedchem.org
1845
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim