Hepatitis C viral IRES inhibition by phenazine and phenazine-like molecules

Article abstract of DOI:10.1016/S0960-894X(00)00217-1

An in vitro assay based on the expression of Fluci reporter gene under the translational control of HCV IRES was used to evaluate and screen compound libraries. A structure-activity relationship study on a phenazine hit was conducted. Our data suggest that an intact phenazine or phenazine-like core with two distal polar substitutions is crucial for potency. (C) 2000 Elsevier Science Ltd. All rights reserved.

Full text of DOI:10.1016/S0960-894X(00)00217-1

Bioorganic & Medicinal Chemistry Letters 10 (2000) 1151±1154
Hepatitis C Viral IRES Inhibition by Phenazine and
Phenazine-Like Molecules
Wuyi Wang, a,* Patrice Preville, a Nicolas Morin, a Samir Mounir, a
Weizhong Cai b and M. Arshad Siddiqui a
a
Â
BioChem Pharma Inc., 275 Armand-Frappier Blvd., Laval, Quebec, Canada H7V 4A7
bOSI Pharmaceuticals Inc., 106 Charles Lindbergh Blvd., Uniondale, NY 11553, USA
Received 16 December 1999; accepted 4 March 2000
AbstractÐAn in vitro assay based on the expression of Fluci reporter gene under the translational control of HCV IRES was used
to evaluate and screen compound libraries. A structure±activity relationship study on a phenazine hit was conducted. Our data
suggest that an intact phenazine or phenazine-like core with two distal polar substitutions is crucial for potency. # 2000 Elsevier
Science Ltd. All rights reserved.
It is estimated that approximately 3% of the world's
population is infected with hepatitis C virus (HCV), a
member of the Flaviviridae. As HCV is a major cause of
post-transfusion and community acquired non-A non-B
hepatitis, it poses a great threat to the public health. The
acute symptoms of this infection include jaundice and
fatigue. More importantly, this infection persists in >80%
of the infected population. The chronic infection often
leads to liver cirrhosis which is associated with a high
risk of the development of hepatocellular carcinoma.1
intervention in the viral life cycle. Several ¯uorenones
and xanthones have shown inhibitory properties in vitro
against HCV IRES-dependent translation.7 For a
broader search of low molecular weight anti-IRES
agents, we have adopted a strategy which allows quick
hit identi®cation, validation and subsequent optimiza-
tion. A library of 132,000 compounds and fungal extracts
were screened using an in vitro assay8 in which a mole-
cular construct (CAT-IRES-Fluci) was expressed in
Huh7 cells under a CMV promoter. The HC40 cell line
Luciferase activity would therefore report IRES-depen-
dent translation. Compounds showing inhibition in this
assay were further evaluated in cell toxicity assay9 to
establish the toxicity pro®le of the inhibitors. As a result
of such screenings, several chemical classes emerged,
showing activity in Luciferase assay with only low level of
toxicity. In this communication, we describe our investi-
gations on phenazine series as novel anti IRES agents.
The hit molecule 1 (also known as neutral red) exhibited
potent anti-IRES activity which was con®rmed by repe-
ated assays. Chemically, the planar structure stretched to
allow extended nitrogen±arene conjugation and, as a
result, was highly colored. The two distal amino sub-
stitutions together with the ring nitrogen atoms could
potentially form a unique hydrogen bonding array. In
view of the strong staining e€ect as an undesired feature
with compound 1, a structural modi®cation became
necessary.
Much e€ort is being directed towards the understanding
of HCV replication and the disease associated with it. In
particular, research has been focused on the molecular
biology of viral protein functions.2 It has been shown
that the non-structural protein NS3, which requires
NS4A cofactor3c,d for enhanced activity, encodes pro-
tease functions for polyprotein processing hence is likely
to play a key role for downstream regulation of viral
replication.3,4 One of such protein products, (i.e., NS5B),
is found to encode functions as RNA-dependent-RNA
polymerase responsible for viral RNA replication.5
Alternatively, HCV RNA genome has been the subject
for the investigation. Studies6 show that HCV internal
ribosome entry site (IRES), the `landing pad' contained
within the genomic 50 non-coding region, is required for
translation initiation. The unique viral secondary RNA
structure would o€er a potentially valid target for early
In order to evaluate the necessity of the ring system
de®ned by structure 1 for IRES activity, a number of
molecules lacking a central ring were tested. The results
*Corresponding author. Tel.: +1-450-978-7788; fax: +1-450-978-
7777; e-mail: wangwu@biochempharma.com
0960-894X/00/$ - see front matter # 2000 Elsevier Science Ltd. All rights reserved.
PII: S0960-894X(00)00217-1

1152
W. Wang et al. / Bioorg. Med. Chem. Lett. 10 (2000) 1151±1154
of some representative examples are summarized in
Table 1. The crucial role played by the central ring was
evident, as all `open' analogues (2±4) displayed lower, or
no activity relative to the parent molecule. It was con-
ceivable that the angle formed by the central linking
group (NH or CO) and two ¯anking aryls was now
widened.10 The observation that 2 retained residual
activity appeared to also indicate the importance of both
alkylated amine and bridging nitrogen, as 3 or 4 lacking
either one of these features were completely inactive.
Other compounds listed in Table 2 were all acquired
from commercial sources. Chemical integrity of all analo-
gues was secured by H NMR, mass spectroscopy (for
synthetic analogues) and HPLC analysis prior to biological
evaluations.11
1
As shown in Table 2, the anti-IRES activity was very
sensitive to structural variations. Phenazines 16±20,
while maintaining same or similar core structure like 1,
all displayed reduced activity. It was apparent that the
concurrent involvement of multiple functions in 1 in the
binding was mandatory for good activity. For example,
merely changing dimethylamine function to a hydroxyl
group as seen in 16 resulted in a signi®cant drop in
activity. Similarly the replacement of amine function in
1 with iodide abolished the activity. On the other hand,
the activity was a€ected to a lesser degree with arylation
of the central nitrogen (as seen in 18). All phenazine
analogues showed respectable toxicity margin in MTT
assay in HC 40 cell lines.9
Subsequently, our investigation was focused on intact
phenazine or phenazine-like cores with structural changes
limited to only ring size and substitutions. We expected
that the modi®ed ring structures would be more amenable
to chemical manipulations needed for achieving optimal
activity.
Among modi®cations of ring substituents, the 7-amino
function was chosen initially for replacement by a halo-
gen function via Sandmeyer reaction (Scheme 1). This
step should provide analogue(s) lacking one amino
function for direct comparison with the parent struc-
ture. Chemically, the halogen function could be used as
a handle to introduce a variety of groups. The standard
Sandmeyer reaction carried out in aqueous medium at
0 ꢀC a€orded formation of diazonium ion which gave,
upon treatment with potassium iodide, compound 5 as
the only product.11
Interestingly, phenazine-like compounds 19 and 20 car-
rying permanent charges were comparatively active,
although notably more toxic than their phenazine coun-
terparts. The attempt to adjust the ring size by using
structurally similar carbazole and benzoxazepinone tem-
plates (14, 15 and 21) resulted in complete loss of activity.
The interruption of conjugation rather than size change
appeared to be one of the causes as compound 22 hav-
ing standard 6-6-6 ring arrangement but lacking the
conjugation was, too, inactive.
Synthesis of a novel ring system was exempli®ed by ver-
satile, drug-like dibenzoxazapinones. Upon condensation
of o-hydroxyanilines with o-chloroaryl acids,12 amides 10±
11 were obtained. Cyclization, followed by catalytic
hydrogenolysis, a€orded diamino-benzoxazepinone 14 as
well as monoamino-benzoxazepinones 15 in good
yields.12
Although the measurement of Fluci signal in HC40 cells
largely re¯ected the level of IRES activity, non speci®c
transcriptional and translational modulation could not
be excluded. A CMV-Fluci assay in HCF cells (pooled
Huh-7 cells stably transfected with CMV-Fluci) was
Table 1. Anti-IRES activity of compounds with truncated central
ring
Compounds
Structures
Activity Fluci (HC40)
IC50 (mg/mL)
1
0.02
2
2.19
3
4
>1 0
>10
Scheme 1. (a) SOCl2/THF/DIPEA; (b) from 11, pyridine, 90 ꢀC; from
12 and 13, NaOH/DMF; (c) H2/Pd.

W. Wang et al. / Bioorg. Med. Chem. Lett. 10 (2000) 1151±1154
1153
Table 2. Anti-IRES activity of phenazine and phenazine-like analo-
gues
therefore employed as a secondary assay. In this system
the measured Fluci signal would still be under the CMV
transcriptional promoter control but not under IRES
translational control. As seen in Table 2, there appeared
some level of separation in the two Fluci assays for
several compounds, with the ratio of HC40 versus HCF
in the rank of 1 (10) >16, 17, 19 (2±3) >18, 20 (1±<1).
Compound
Structure
Fluci IC50
(HC40) (HCF)/IC50 (HC40)
IC50
MTT
(mg/mL)
(HC40)
CC50
(mg/mL)
1
2
0.02
10
>10
As a crucial RNA genomic structure required by virus
for initiation of translation, HCV IRES would appear
to provide a unique target for pharmaceutical interven-
tion. Our dual luci testing system has served e€ectively
for the selection of promising compounds for further
downstream studies. In the absence of suitable reference
compounds, multiple assays are often needed in order to
reach reliable conclusions. The additional evaluation in
vivo13 on compound 1 appeared to lend support for the
validity of the dual luci assay strategy.
2.19
3.74
0.8
2.3
16
>2.7
>10
In conclusion, our search for inhibitors of HCV IRES
target has led to the identi®cation of a novel phenazine
series. The preliminary SAR study has established that
the phenazine core and polar substituents such as
amines at positions 2 and 8 are crucial for potent activ-
ity, possibly due to the possession of unique hydrogen
bonding array. The reduced activity observed with the
`open' analogues 2±4 were indicative of change of the
array as a result of, at least partially, geometry changes.
The results also suggested that a pyrazine moiety is
important but not essential for activity as phenazine-
like molecules (19 and 20) lacking a pyrazine arrange-
ment were also comparatively active although more
toxic. While alteration of 1 has resulted in drop in
activity, good toxicity pro®le and moderate selectivity
retained in 16 and 17 suggest that there is room for
further improvement of the structures.
1
5
4.03
2.5
Ð
>10
>1 0
>1 0
18
ꢁ1
1
>10
19
20
1.00
2.86
3
<1
Acknowledgements
1.7
4.67
We thank Drs R. Rando and R. Storer for useful sugges-
tions, and Dr. T. Bowlin, Dr. A. Hegwy, J. Schwedes, E.
Maier, Drs P. Nguyen-Ba and B. Bachand for their input.
21
14
>1 0
Ð
Ð
>1 0
>10
References and Notes
1. (a) World Health Organization. Hepatitis C: global pre-
valence. Weekly Epidemiological Record 1997, 10, 65. (b)
Alter, M. J.; Margolis, H. S.; Krawczynsk, K.; Judson, F. N.;
Mares, A.; Alexander, W. J.; Hu, P. Y.; Gerber, M. A.; Sam-
pliner, R. E.; Meeks, E. L.; Beach, M. J. N. Engl. Med. 1992,
327, 1899.
ꢂ10
2. (a) Bartenschlager, R. Intervirology 1997, 40, 378. (b) Oka-
moto, H.; Miyakawa, Y.; Mayumi, M. Viral Hepatitis 1997, 3,
51.
15
22
ꢂ10
ꢂ10
Ð
Ð
ꢂ10
ꢂ10
3. (a) Gallinari, P.; Brennan, D.; Nardi, C.; Brunetti, M.;
Tomei, L.; Steinkuler, C.; Francesco, R.-D. J. Virol. 1998, 72,
6758. (b) Bartenschlager, R.; Ahlborn-Laake, L.; Mous, J.;
Jacobsen, H. J. Virol., 1993, 67, 3835. (c) Lin, C.; Thomson, J.
A.; Rice, C. M. J. Virol. 1995, 69, 14373. (d) Kim, J. L.;
Morgenstern, K. A.; Lin, C.; Fox, T.; Dwyer, M. D.; Landro,
J. A.; Chambers, S. P.; Markland, W.; Lepre, C.; O'Malley, E.
T.; Harbeson, S. L.; Rice, C. M.; Murcko, M. A.; Caron, P.
R.; Thomson, J. A. Cell 1996, 87, 343.

1154
W. Wang et al. / Bioorg. Med. Chem. Lett. 10 (2000) 1151±1154
4. For NS3 helicase functions: (a) Kim, J. L.; Morgenstern, K.
A.; Grith, J. P.; Dwyer, M. D.; Thomson, J. A.; Murcko, M.
A.; Lin, C.; P. Caron. Structure 1998, 6, 89. (b) Cho, H.-S.;
Ha, N.-C.; Kan, L.-W.; Chung, K. M.; Back, S. H.; Jang, S.
K.; Oh, B.-H. J. Biol. Chem. 1998, 273, 15045.
10. A molecular mechanics calculation using HyperChem
program suggests, in addition to the widened angle, a slight
pyramidal con®guration for the central nitrogen atom in
structures 2 and 3.
11. The synthetic analogues have been characterized by the NMR
analysis (spectra recorded on Varian 300 MHz or Bruker
400 MHz spectrometer), MS analysis (using a Finnigan MAT
LCQ system) and HPLC analysis (using 0.01% TFA bu€ered
water/acetonitrile system on a reverse phase C18 VYDAC col-
umn with UV detector set at 215 nm wave length). For compound
5, 1H NMR (CDCl3): d, 2.68 (s, 3H), 3.22 (s, 6H), 7.04 (d,
J=2.7 Hz, 1H), 7.60 (dd, J1=9.5 Hz, J2=3 Hz, 1H), 7.96 (s, 1H),
7.97 (d, J=9.5 Hz, 1H), 8.68 (s, 1H); MS+, 164.2; HPLC purity:
>87%. For compound 14, 1H NMR (DMSO): 10.26 (s, 1H), 7.65
(d, J=3.0 Hz, 1H), 7.31 (d, J=3.0 Hz, 1H), 6.84 (d, J=8.5 Hz,
1H), 6.25±6.20 (m, 2H), 5.36 (s, 2H), 5.09 (s, 2H), MH+: 242. For
compound 15, 1H NMR (DMSO): 10.27 (s, 1H), 8.41 (dd,
J=5.0 Hz, J=2.0 Hz, 1H), 8.17 (dd, J=7.5 Hz, J=2.0 Hz, 1H),
7.38 (dd, J=7.5 Hz, J=4.5 Hz, 1H), 6.78 (d, J=8.5 Hz, 1H), 6.47
(d, J=2.5 Hz, 1H), 6.35 (dd, J=8.5 Hz, J=2.5 Hz, 1H), 5.20 (s,
2H); MH+: 227. All commercial compounds showed purity
>90% by HPLC analysis.
12. Klunder, J. M.; Hargrave, K. D.; West, M.; Cullen, C.;
Pal, K.; Behnke, M. L.; Kapadia, R. S.; McNeil, D. J.; Wu, J.
C.; Chow, G. C.; Adams, J. J. Med. Chem. 1992, 35, 1887.
13. Zauberman, A.; Nussbaum, O.; Ilan, E.; Eren, R.; Ben-
Mosh, O.; Arazi, Y.; Berre, S.; Lubin, I.; Shouval, D.; Galun,
E.; Reiser, Y.; Dagan, S. 6th International Symposium on
Hepatitis C and Related Virus. Molecular Virology and Patha-
genesis, Bethesda, MD, USA, 6±9 June, 1999, Abstract # 4.3.
5. Behrens, S.-E.; Tomei, L.; De Francesco, R. EMBO 1996,
15, 12±22.
6. (a) Fukushi, S.; Kurihara, C.; Ishiyama, N.; Hoshino, F. B.;
Oya, A.; Katayama, K. J. Virol. 1997, 71, 1662. (b) Lu, H.-H.;
Wimmer, E. Proc. Natl. Acad. Sci. USA 1996, 93, 1412.
7. (a) Ikeda, M.; Sakai, S.; Tsuai, I.; Zuao, I.; Ryan, H.; Iyan,
S.; Kai, Y.; Kako, Y.; Tsukada, I.; Yanagisawa, M.
JP8268890, 1996. (b) Kai, Y.; Sasho, M.; Kaku, Y.; Tsukada,
I.; Yanagisawa, M. JP7069899, 1996.
8. (a) Wang, C.; Sarnow, P.; Siddiqui, A. J. Virol. 1994, 68,
7301. (b) Fluci assays in HC 40 and HCF cells: the HC 40 and
HCF cell lines were derived from hepatocellular carcinoma
cell line Huh-7. The cell line was transfected with the dicis-
tronic construct pCMV-CAT-IRES-Fluci for HC 40 and with
pCMV-Fluci construct for HCF. The following conditions
were used for both assays: 24 h prior to the assay, cells were
seeded at 1.5Â104 cells/well in a 96-well plate. The compounds
were then added at various concentrations in a ®nal DMSO
concentration of 0.2% and incubate 24 h at 37 ꢀC. Following
the incubation the cells were washed twice in PBS. 50 mL of
Luciferin lysis bu€er is added and kept for 10 min in the dark.
Cells were then re-suspended and 80 mL was taken on a white
96-well plate for Luciferase activity.
9. Plumb, A. J.; Milroy, R.; Kaye, S. B. Cancer Res. 1989, 49,
4435.