Benzohydroxamic acids as potent and selective anti-HCV agents

Article abstract of DOI:10.1016/j.bmcl.2013.08.081

A diverse collection of 40 derivatives of benzohydroxamic acid (BHAs) of various structural groups were synthesized and tested against hepatitis C virus (HCV) in full-genome replicon assay. Some of these compounds demonstrated an exceptional activity, suppressing viral replication at sub-micromolar concentrations. The compounds were inactive against key viral enzymes NS3, and NS5B in vitro assays, suggesting host cell inhibition target(s). The testing results were consistent with metal coordination by the BHAs hydroxamic group in complex with a target(s). Remarkably, this class of compounds did not suppress poliomyelitis virus (PV) propagation in RD cells indicating a specific antiviral activity of BHAs against HCV.

Full text of DOI:10.1016/j.bmcl.2013.08.081

Bioorganic & Medicinal Chemistry Letters xxx (2013) xxx–xxx
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
Benzohydroxamic acids as potent and selective anti-HCV agents
a
b
a
Maxim V. Kozlov a, , Alla A. Kleymenova , Lyudmila I. Romanova , Konstantin A. Konduktorov ,
⇑
Olga A. Smirnova a, Vladimir S. Prasolov a, Sergey N. Kochetkov a
a Engelhardt Institute of Molecular Biology, Russian Academy of Sciences, Vavilov Str. 32, 119991 Moscow, Russia
b Chumakov Institute of Poliomyelitis, Russian Academy of Medical Sciences, 142782 Moscow Region, Russia
a r t i c l e i n f o
a b s t r a c t
Article history:
A diverse collection of 40 derivatives of benzohydroxamic acid (BHAs) of various structural groups were
synthesized and tested against hepatitis C virus (HCV) in full-genome replicon assay. Some of these com-
pounds demonstrated an exceptional activity, suppressing viral replication at sub-micromolar concentra-
tions. The compounds were inactive against key viral enzymes NS3, and NS5B in vitro assays, suggesting
host cell inhibition target(s). The testing results were consistent with metal coordination by the BHAs
hydroxamic group in complex with a target(s). Remarkably, this class of compounds did not suppress
poliomyelitis virus (PV) propagation in RD cells indicating a specific antiviral activity of BHAs against
HCV.
Received 25 June 2013
Revised 14 August 2013
Accepted 18 August 2013
Available online xxxx
Keywords:
Benzohydroxamic acids
Hepatitis C virus
Poliomyelitis virus
Metal-depending enzymes
Ó 2013 Elsevier Ltd. All rights reserved.
HCV is a major cause of chronic hepatitis, which affects three to
four million people worldwide annually. Chronic hepatitis C is
associated with a high risk for development of liver cirrhosis and
hepatocellular carcinoma.1 The generally accepted HCV therapy
is treatment with a combination of interferon-alpha-2b and ribavi-
rin. However, this treatment promotes a sustained viral response
in only about 50% of the patients.2 Therefore, new efficient antiviral
therapies are in great demand.
an in vitro assay. Furthermore, benzohydroxamic acid (BHA, com-
pound 3), while structurally much less related to DKA, possessed
significant antiviral activity as well (Fig. 1). It was established that
compounds 1–3 were inactive against HCV protease/helicase (NS3)
in vitro assays and did not interfere with stability of another viral
protein (NS5a) that contains zinc finger domain.8 These findings
suggest an host cell inhibition target for BHA derivatives and grant
further detailed investigation on their inhibitory mechanism. In
the present research, we report the synthesis and structure–activ-
ity relationship (SAR) studies on HCV inhibition in a full-genome
replicon assay for benzohydroxamic acid (BHA) derivatives. Some
of the new compounds displayed exceptional antiviral activity
and low toxicity for mammalian cells, which makes them promis-
ing candidates for drug development.
All tested hydroxamic acids were synthesized by four different
methods depending on steric hindrance of carboxylic groups as
well as on reactive ability of corresponding methylated hydroxyl-
amines. The use of free acids and CDI as condensing agent9,10
proved to be the most convenient route to BHAs with unhindered
carbonyl group (method a). Compound 33 was obtained by treat-
ment of corresponding methyl ester of 5-nitrosalicylic acid with
hydroxylamine in aqueous DMSO under weakly basic conditions
(method b). Methyl esters of benzoic acids derivatives with bulky
groups in ortho-position readily reacted with hydroxylamine in
methanol in the presence of double molar excess of potassium
hydroxide yielding the desired hydroxamates (method c).11 The
chloroanhydrides of corresponding acids were useful for acylation
of O- and N-methylated hydroxylamines as well as for preparation
of extremely sterically hindered benzohydroxamate 37 (method
d). All benzamidoxime derivatives were prepared from related
HCV is a positive strand RNA virus whose genome consists of
9600 base pairs encoding 3 structural and 7 non-structural pro-
teins.3 Two key viral enzymes NS3 protease and NS5B polymerase
remain the most popular targets for design of new anti-HCV
drugs;4 however, rapid resistance development to the agents tar-
geting these enzymes justifies the search and characterization of
new proteins involved in viral life cycle, that could be affected by
small molecules.
In our previous work5 we have proposed a binding mode for
a,c-diketo acid inhibitors (DKA, Fig. 1) in the active center of
RNA-dependent RNA polymerase of HCV (NS5b). According to the
model, DKA present an additional Mg2+ (or Mn2+) ion into the ac-
tive site of the enzyme, thereby interfering with nucleotidyl trans-
fer catalysis.5 Z-imide form6 of salicylic hydroxamic acid (SHA,
compound 1) and ortho-anisic hydroxamic acid (ortho-OCH3 BHA,
compound 2) seemed to mimic ‘three-oxygen’ metal-chelating sys-
tem of DKA in enol form7 consistently with high anti-HCV activity
in full-genome HCV replicon assay (Fig. 1). However, these com-
pounds failed to inhibit polymerase activity of NS5b as tested in
⇑
Corresponding author. Tel.: +7 (499) 135 0590; fax: +7 (499) 135 1405.
E-mail address: kozlovmv@hotmail.com (M.V. Kozlov).
0960-894X/$ - see front matter Ó 2013 Elsevier Ltd. All rights reserved.
http://dx.doi.org/10.1016/j.bmcl.2013.08.081
Please cite this article in press as: Kozlov, M. V.; et al. Bioorg. Med. Chem. Lett. (2013), http://dx.doi.org/10.1016/j.bmcl.2013.08.081

2
M. V. Kozlov et al. / Bioorg. Med. Chem. Lett. xxx (2013) xxx–xxx
Me2+
Me2+
O
H3C
Me2+
O
Me2+
O
H
O
O
O
O
O
O
O
O
N
N
N
O
3
BHA ( )
DKA
SHA (1)
ortho-OCH3 BHA (2)
EC50 = 15 µM
CC50 = 550 µM
EC50 = 12 µM
CC50 = 630 µM
EC50 = 15 µM
CC50 = 210 µM
Figure 1. Molecular structures of DKA, SHA, ortho-OCH3 BHA and BHA complexes with Me2+ ion (DKA is shown in enol form, SHA, ortho-OCH3 BHA and BHA in Z-imide form)
and values of anti HCV activity and cytotoxicity for Huh7 cells.
Table 1
Anti-HCV activity and cytotoxicity of methylated benzohydroxamates, benzamidoximes and benzamides
R1
R2
Compd
R1
R2
EC50
(
lM)
CC50
400
>800
>800
(lM)
No.
R1
R2
EC50
(lM)
CC50(lM)
4d
5c
–H
–OH
–OCH3
65
>300
>300
10e
11e
–H
–OH
–OCH3
270
140
>300
>800
500
>800
O
O
NOH
6a
12e
OH
N
NH2
NH2
CH3
7d
8d
–H
–OH
–OCH3
>300
>300
>300
>800
>800
>800
CH3
O
O
13
14
–OH
>300
>300
>800
>800
9a
–OCH3
N
H
a,c,d,eMethod of preparation.
nitriles through treatment with aqueous EtOH solutions of hydrox-
ylamine under neutral conditions (method e).12 More than one
stage syntheses of BHAs 5, 8, 30, 32, 38 and 40 were carried out
as shown in Supplementary data.
attributed to prolongation the life span of the reporter protein by
hydroxamates or to enhanced degradation of the HCV RNA.
To test the specificity of the BHAs antiviral activity in regard to
various viruses we analyzed the effect of the representative com-
pounds ortho-OCH3 BHA (active) and analog 5 (not active) on polio-
virus (PV) infection of RD cells. The rational for this choice was that
PV (as well as HCV) is single stranded RNA viruses with a genome
of positive polarity. However, despite of similarity in genome orga-
nization and in replication cycle, these species differ drastically in
many other biological aspects and virulence.13
The control treatment of non-infected RD cells with BHAs was
essentially similar to that of Huh7 cells, except higher inhibitors
concentration (twofold) and prolonged incubation time (additional
7 h) were used. Additional incubation time was equal to that of the
PV reproduction cycle. No visible difference in the density of the
cell monolayers compared to that of non-drug treated control cells
was observed (Fig. 3A, C and E). There were also no overt morpho-
logical signs of cell death or changes in cell morphology as deter-
mined by fluorescence microscopy after staining of the nuclear
DNA with Hoechst33342 (Fig. 3B, D and F). The above observations
were consistent with CC50 values obtained by MTT assay for ortho-
To address the role of hydroxamic moiety in anti-HCV activity
of BHAs compounds their modified analogs (compounds 4–14)
were prepared and tested in replicon assay (Table 1). All used com-
pounds of this group representing derivatives of benzoic, salicylic
and ortho-anisic acids can be divided into four groups based on
the structure of carbonyl group radicals: N-hydroxy-N-methyla-
mides 4-6, N-methoxyamides 7–9, amidoximes 10–12 as well as
amides 13 and 14. In comparison with dianionic bidentate ligand
that hydroxamic residue represents, the chelating ability of the
mentioned amide ligands is impaired because of reduction of their
negative charge and (or) the number of metal-coordinating groups.
As seen from Table 1, only compounds 4, 10 and 11 bearing
monoanionic bidentate carbonyl residues showed detectable anti-
viral activity, whereas all remaining compounds were inactive.
Remarkably, the presence of an extra coordination center (oxygen
atom in ortho-position) in SHA derivatives 5, 8 and 11 did not sup-
press the negative effect of hydroxamic group substitutions on
antiviral activity. These results point to a very precise mode of me-
tal coordination by the BHAs hydroxamic group in complex with a
target(s).
OCH3 BHA (330lM) and compound 5 (>800 lM).
In the next set of experiments naive and drug-treated RD cells
were infected with poliovirus (PV) at a multiplicity of infection
(MOI) of ꢀ20 or ꢀ600 plaque forming units (PFU)/cell. Virus and
drug induced alterations of the cell monolayer were examined
right after infection or after 7 h, the time period of the PV repro-
duction cycle. As expected, PV infection resulted in destruction of
RD cell monolayer and in nuclear alterations of productively of
the infected cells, as determined by light microscopy of the cells
and fluorescence microscopy of Hoechst stained nuclei (data not
Next SHA (compound 1) and ortho-OCH3 BHA (compound 2)
were tested in independent quantitative RT-PCR based assay to
find how the luciferase activity in replicon assay correlated with
the amount of viral RNA (Fig. 2). Compared to the EC50 values of
SHA and ortho-OCH3 BHA obtained by replicon assay (15 and
12
lM) the EC50 values as measured by qRT-PCR assay were some-
what lower (8 and 7
lM). This relatively small difference might be
Please cite this article in press as: Kozlov, M. V.; et al. Bioorg. Med. Chem. Lett. (2013), http://dx.doi.org/10.1016/j.bmcl.2013.08.081

M. V. Kozlov et al. / Bioorg. Med. Chem. Lett. xxx (2013) xxx–xxx
3
100
luciferase assay
qRT-PCR assay
100
50
0
luciferase assay
qRT-PCR assay
50
0
0,1
1
10
100
0,1
1
10
100
[SHA](μM)
[ortho -OCH3 BHA](μM)
Figure 2. Antiviral activity of BHAs on a HCV replicon carrying a reporter gene. Cells persistently carrying the replicon were treated with 0, 1, 5, 25 and 125
lM SHA and
ortho-OCH3 BHA for 72 h. The inhibition of translation and replication was determined by measuring the firefly luciferase activity (continuous line) and the amount of
replicon RNA (dashed line), respectively.
Figure 3. Light transmission (A, C and E) and fluorescence (B, B and F) images of RD cells. Cell cultures were incubated for 72 h in the absence of inhibitors (A/B) or for 72 h at
[125 lM] and then for 7 h at [250 lM] of ortho-OCH3 BHA (C/D) and compound 5 (E/F).
shown). Qualitatively similar results were obtained with PV-in-
fected cells treated with ortho-OCH3 BHA and compound 5 (data
not shown). The amount of PV progeny produced from drug treated
and non-treated cells were quantified by PFU assays. A negligible
difference in PV reproduction between control and drug-treated
cells was detected for RD cells that were pretreated for 72 h with
125 lM ortho-OCH3 BHA (Table 2). This result very likely reflected
rather latent cytotoxic action than antiviral activity of this com-
pound. The lack of significant antiviral activity on PV in our assay
argues for a specific anti-HCV action of ortho-OCH3 BHA if com-
pared with suppression of HCV proliferation (Fig. 1).
In next SAR studies, fifteen BHAs carrying one of five types of
substituents (CH3, OCH3, Cl, Br and NO2) in ortho-, meta- or para-
position were tested in replicon assay, allowing determination of
the basic structural features of BHAs crucial for inhibition of HCV
replicon propagation (Table 3). The collected data might be sum-
marized as follows: (i) ortho-substituents increased activity and
selectivity of antiviral action relatively to BHA with one exception
of CH3-group; (ii) para-substituents mainly enhanced activity but
not selectivity of antiviral action relatively to BHA; (iii) meta-sub-
stituents did not change selectivity relatively to BHA and had gen-
erally negative effect on antiviral activity as compared to ortho-
and para-substituents.
A follow-up testing included four more ortho-substituted deriv-
atives. As seen from Table 4, inhibitory properties of fluoro-ana-
logue 29 were close to those for benzohydroxamic acid (Table 3).
One of the two hydroxamates bearing bulky ortho-substituent
(compound 30) was almost inactive, while another one (compound
31) displayed significant anti-HCV activity and fair selectivity in
good agreement with antiviral properties of the related com-
pounds, ortho-CH3 (compound 15) and ortho-OCH3 BHA (com-
pound 2) correspondingly (Table 3). These data also suggested an
Table 2
Effect of BHAs on poliovirus growth in RD cells
Multiplicity of
infection (PFU/cell) presence
Virus yield, PFU/ml  10À9, by 7 h post infection in the
No drug
ortho-OCH3 BHA
Compound 5
A
B
A
B
20
600
1.2
1.8
0.64
0.70
1.1
1.3
1.2
1.6
1.4
1.8
(A) RD cells were pretreated for 72 h with the respective compound at 125
then incubated at 250 M for 7 h post infection; (B) compound was added at a final
concentration of 250 M only for 7 h post infection.
lM, and
l
l
Please cite this article in press as: Kozlov, M. V.; et al. Bioorg. Med. Chem. Lett. (2013), http://dx.doi.org/10.1016/j.bmcl.2013.08.081

4
M. V. Kozlov et al. / Bioorg. Med. Chem. Lett. xxx (2013) xxx–xxx
Table 3
Table 5
Selectivity of anti-HCV action of monosubstituted benzohydroxamates
Selectivity of anti-HCV action of two substituted benzohydroxamates
R1
O
O
NHOH
NHOH
R2
R
Compd
R
Isomer
EC50
(lM)
CC50
(l
M)
SI
14
8
53
165
90
55
Compd
R1
Position-R2
EC50
(
l
M)
CC50 (lM)
SI
33b
34a
35a
36a
37d
38c
39c
40c
–OH
5-NO2
5-NO2
4-CH3
5-CH3
6-OCH3
4-CH3
4-Cl
160
3
15
10
50
1.5
0.80
0.50
>1000
350
900
800
>800
140
>6
120
60
80
>16
93
3
H
—
ortho-
15
85
12
2.0
3.0
1.0
210
650
630
330
270
55
15c
2a
–CH3
–OCH3
–Cl
–Br
–NO2
–OCH3
–OCH3
–OCH3
–OCH3
–Cl
16c
17c
18c
19a
20a
21a
22a
23a
–CH3
–OCH3
–Cl
–Br
–NO2
meta-
para-
20
70
10
12
5.0
340
570
120
310
160
17
8
12
26
32
–Cl
–CH3
140
120
175
240
4-Br
a,c,d,eMethod of preparation.
24a
25a
26a
27a
28a
–CH3
–OCH3
–Cl
–Br
–NO2
8.0
12
3.0
2.0
3.0
110
180
55
25
70
14
15
18
13
23
compound 39. Surprisingly, combination of CH3- and Br-groups
in position 2 and 4, correspondingly, strongly (170-fold) enhanced
the inhibitory activity of compound 40 compared to ortho-CH3 BHA
(compound 15), thus elevating SI value to 240 (Table 5).
Reported results demonstrated that benzohydroxamates sup-
press HCV full-genome replicon propagation in Huh7 cells and al-
lowed us to identify major structural features of BHAs crucial for
their antiviral activity. This class of compounds efficiently (at low
micromolar range) suppressed viral proliferation in dose-depen-
dent mode as judged by reduction of luciferase reporter signal
and viral RNA content. The antiviral action of BHAs displayed selec-
tivity and specificity with respect to HCV replication. The sug-
gested inhibitory mechanism was consistent with metal-
mediated binding of BHAs to a target through chelation. Our preli-
minary data point to matrix metalloproteinase-2 and histone
deacetylase 6 as possible cellular enzymes responsible for antiviral
BHAs action. The next challenge is further testing of this hypothe-
sis and elucidation of the exact mechanism through which the BHA
compounds affect HCV proliferation.
a,cMethod of preparation.
Table 4
Selectivity of anti HCV action of ortho-substituted benzohydroxamates
Compd Structure
R
EC50
M)
CC50
M)
SI
(l
(l
29a
–F
15
170
10
300
>1000
380
20
>6
38
R
O
30c
31c
–CH2OCH3
–
O(CH2)3CH3
NHOH
CONHOH
O
32a
CH2
–O(CH2)2O–
35
>800
>23
2
a,cMethod of preparation.
Acknowledgments
The work was supported by the Russian Foundation for Basic
Research, project no. 11-04-01202, and the program of the Presid-
ium of the Russian Academy of Sciences ‘Molecular and Cell Biol-
ogy’. We thank Professor Ralf Bartenschlager at University of
Heidelberg, Germany, for providing the Huh7 FL-Neo replicon cells,
Professor Arkady Mustaev at University of Medicine and Dentistry
of New Jersey, USA, for helpful discussion in manuscript
preparation.
importance of the presence of oxygen atom connected to phenyl
ring in ortho-position for selective antiviral activity. Notably, fusing
of the two residues of ortho-OCH3 BHA through alkyl C–C bond did
not abolish antiviral activity of the resulting compound 32.
To study a possible synergistic effect of the second substitutions
in the phenyl ring on ortho-BHAs antiviral properties (suggested by
the properties of corresponding mono-substituted compounds)
disubstituted benzohydroxamates 33–40 were prepared and
tested in the replicon assay. As seen from Table 5, hydroxamate
34 retained low cytotoxicity and high antiviral potency of parent
ortho-OCH3 and meta-NO2 BHAs (Table 3, compounds 2 and 23).
Contrary to the expectations, 5-nitro-salicylhydroxamate 33 dis-
Supplementary data
Supplementary data associated with this article can be found, in
the online version, at http://dx.doi.org/10.1016/j.bmcl.2013.
08.081.
played enhanced EC50 value (160 lM) most likely due to negative
effect of hydroxyl group ionization (facilitated by the nitro group)
on the cellular uptake.
References and notes
Introduction of an additional methyl group in positions 4 or 5 of
aromatic ring and ortho-OCH3 and ortho-Cl BHAs (compounds 35,
36 and 38) had a predictable (deduced from consideration of the
properties of corresponding mono-substituted compounds) little
effect on antiviral activity. Compound 37 bearing two metoxy
groups in positions 2 and 6 displayed reduced anti-HCV activity,
probably, due to the influence of second substituent on the com-
pound fitness in the binding site. Minor additive effect of 2,4-di-
chloro substitution on antiviral properties was detected for
1. Seeff, L. B. Hepatology 2002, 36, 35.
2. Scott, L. J.; Perry, C. M. Drugs 2002, 62, 507.
3. Bartenschlager, R.; Frese, M.; Pietschmann, T. Adv. Virus Res. 2004, 63, 71.
4. De Francesco, R.; Tomei, L.; Altamura, S.; Summa, V.; Migliaccio, G. Antiviral Res.
2003, 58, 1.
5. Kozlov, M. V.; Polyakov, K. M.; Filippova, S. E.; Evstifeev, V. V.; Lyudva, G. S.;
Kochetkov, S. N. Biochemistry (Moscow) 2009, 74, 834.
6. Garcıa, B.; Ibeas, S.; Leal, J. M.; Secco, F.; Venturini, M.; Senent, M. L.; Nino, A.;
Munoz, C. Inorg. Chem. 2005, 44, 2908.
7. Summa, V.; Petrocchi, A.; Pace, P.; Matassa, V. G.; De Francesco, R.; Altamura, S.;
Tomei, L.; Koch, U.; Neuner, P. J. Med. Chem. 2004, 47, 14.
Please cite this article in press as: Kozlov, M. V.; et al. Bioorg. Med. Chem. Lett. (2013), http://dx.doi.org/10.1016/j.bmcl.2013.08.081

M. V. Kozlov et al. / Bioorg. Med. Chem. Lett. xxx (2013) xxx–xxx
5
8. Tellinghuisen, T. L.; Marcotrigiano, J.; Gorbalenya, A. E.; Rice, C. M. J. Biol. Chem.
11. Hauser, C. R.; Renfrow, W. B. Org. Synth Coll. 1943, 2, 67.
2004, 279, 48576.
12. Srivastava, R. M.; Pereira, M. C.; Faustino, W. W. M.; Coutinho, K.; dos Anjos, J.
V.; de Melo, S. J. Monatsh. Chem. 2009, 140, 1319.
9. Usachova, N.; Leitis, G.; Jirgensons, A.; Kalvinsh, I. Synth. Commun. 2010, 40,
927.
13. De Jesus, N. H. Virol. J. 2007, 4, 70.
10. Kozlov, M. V.; Kleymenova, A. A.; Konduktorov, K. A.; Kochetkov, S. N. Russ. J.
Bioorg. Chem. 2013, 39, 101.
Please cite this article in press as: Kozlov, M. V.; et al. Bioorg. Med. Chem. Lett. (2013), http://dx.doi.org/10.1016/j.bmcl.2013.08.081