SAR by MS: Discovery of a new class of RNA-binding small molecules for the hepatitis C virus: Internal ribosome entry site IIA subdomain

Article abstract of DOI:10.1021/jm050815o

A new class of small molecules that bind the HCV RNA IRES IIA subdomain with sub-micromolar affinity is reported. The benzimidazole 'hit' 1 with a KD -100 μM to a 29-mer RNA model of Domain IIA was identified from a 180000-member library using mass spectr

Full text of DOI:10.1021/jm050815o

J. Med. Chem. 2005, 48, 7099-7102
7099
SAR by MS: Discovery of a New Class of
RNA-Binding Small Molecules for the
Hepatitis C Virus: Internal Ribosome
Entry Site IIA Subdomain
Punit P. Seth,* Alycia Miyaji, Elizabeth A. Jefferson,
Kristin A. Sannes-Lowery, Stephen A. Osgood,
Stephanie S. Propp, Ray Ranken, Christian Massire,
Rangarajan Sampath, David J. Ecker,
Eric E. Swayze, and Richard H. Griffey
Ibis Therapeutics, A Division of Isis Pharmaceuticals Inc.,
1891 Rutherford Road, Carlsbad, California 92008
Received August 16, 2005
Abstract: A new class of small molecules that bind the HCV
RNA IRES IIA subdomain with sub-micromolar affinity is
reported. The benzimidazole ‘hit’ 1 with a KD ∼100 µM to a
29-mer RNA model of Domain IIA was identified from a
180000-member library using mass spectrometry-based screen-
ing methods. Further MS-assisted SAR (structure-activity
relationships) studies afforded benzimidazole derivatives with
sub-micromolar binding affinity for the IIA RNA construct. The
optimized benzimidazoles demonstrated activity in a cellular
replicon assay at concentrations comparable to their KD for
the RNA target.
It is estimated that hepatitis C virus (HCV) has
infected more than 170 million people worldwide, nearly
five times more than human immunodeficiency virus
(HIV).1 These infections frequently become chronic and
often lead to liver cirrhosis or cancer. Current therapies,
either interferon (IFN)-R monotherapy or a combination
of IFN-R with ribavarin, are effective at treating only a
subgroup of HCV-infected patients and are frequently
associated with serious side effects. As a result, new
therapies for the treatment of HCV are highly desir-
able.2,3
HCV is a small, enveloped, positive-stranded RNA
virus from the Flaviviridae family.4 The HCV genome
is composed of a highly conserved 340 nucleotide 5′-UTR
(untranslated region), a single open reading frame
(ORF) of ∼9000 nucleotides, and a short 3′-UTR. The
5′-UTR contains an internal ribosome entry site (IRES),
which mediates the initiation of viral-RNA translation
in a cap-independent manner.5 Even though the HCV-
IRES presents an exciting opportunity for developing
novel HCV therapeutics,6 it remains a largely unex-
plored drug target for small molecules.3 With the
exception of complex RNA-binding natural antibiotics
and organic cations, discovery of new classes of small
molecules that bind structured RNA targets with speci-
ficity and selectivity has been difficult.7-11 The paucity
of RNA-focused medicinal chemistry efforts has also
contributed to a poor understanding of the general
principles that govern the recognition of RNA by small
molecules.7 In this report, we describe the use of SAR
by MS12 methods, to discover a new class of small
molecules with high affinity for the HCV-IRES IIA
subdomain.
Figure 1. Secondary structure of HCV-IRES and 29-mer IIA
screening RNA.
HCV-IRES is known to be important for IRES depend-
ent translation, as well as for HCV RNA replication.
Cryoelectron microscopy mapping of the HCV-IRES
bound to the 40S ribosomal subunit shows that stem II
induces a conformation change in the 40S ribosomal
subunit and positions the single-stranded coding RNA
into the decoding site.13 Mutation studies of the different
bulge regions of domain II have also demonstrated the
importance of stem II for HCV replication and transla-
tion in a cell culture system.14,15
Mass spectrometry-based high throughput screen-
ing16-19 of our compound collection against the IIA 29-
mer (in the presence of a 33-mer structured RNA
control) led to the identification of the benzimidazole
“hit” 1 (Table 1) displaying modest affinity and selectiv-
ity for the target RNA.20 Our initial SAR plan for
optimizing benzimidazole 1 involved investigating the
importance of the dimethylamino headgroup on the
tether at N1 as well as introducing electron-withdraw-
ing and donating groups on the benzimidazole ring
(∼150 analogues prepared).21
The dimethylamino headgroup on the tether at N1
was found to be critical for RNA-binding as removal of
this group resulted in a significant loss of binding
affinity (Table 1, 2a). Replacement of the dimethylamino
group with other cationic headgroups such as pyrroli-
dino, diethylamino, and morpholino (2b-d) resulted in
a slight to substantial loss of binding affinity, whereas
We started with the IIA subdomain as the focus for
high throughput screening (HTS), using a 29-mer RNA
construct as the target (Figure 1). The stem II of the
10.1021/jm050815o CCC: $30.25 © 2005 American Chemical Society
Published on Web 10/19/2005

7100 Journal of Medicinal Chemistry, 2005, Vol. 48, No. 23
Table 1. N1 Head group SAR for Selected Analogues
Letters
compd
R
KD 29-mer (µM)
MS selectivity
1
Me2N
H
1-pyrrolidino
Et2N
4-morpholino
(n-butyl)2N
100
>200
125
3
1
3
3
1
1
2a
2b
2c
2d
2e
130
>200
>200
Table 2. Alkoxy Group SAR for Selected Analogues
Figure 2. Analogues used to probe specificity of binding
contacts for benzimidazole 3.
compd
R
KD 29-mer (µM) MS selectivity
2f
2g
2h
2i
2j
3
6-OMe
4-OMe
5-OMe
7-OMe
6-OBn
6-(Me2NCH2CH2CH2O)
6-(Me2NCH2CH2O)
6-[(Me)2N(CH2)4O]
6-(Me2CHCH2CH2O)
40
>200
>200
50
6
1
1
4
6
99
10
15
11
7
4
8
7
5
6
48
6
the bulky dibutyl group led to a total loss of binding
(2e). Reducing the length of the tether between the
headgroup and the benzimidazole ring also resulted in
a 2-fold loss of binding affinity (data not shown). From
the above results it was concluded that the dimethyl-
aminopropyl chain at N1 was optimal for RNA binding.
Introduction of a variety of substituents at the C5 and
C6 positions of the benzimidazole nucleus resulted in a
substantial decrease in binding affinity (data not shown).
The only substituent that resulted in a slight improve-
ment of binding affinity was a methoxy group at C6
(Table 2, 2f). To further probe the effect of the methoxy
group on RNA-binding, benzimidazoles 2g-i with meth-
oxy groups at C4, C5, and C7 were prepared and eval-
uated. While the introduction of the methoxy group at
C7 was tolerated, the introduction of this substituent
at C4 and C5 did not have any beneficial effect on
binding. Presumably, the binding pocket in the RNA is
not spacious enough to accommodate substitution at the
C4 and C5 positions of the benzimidazole nucleus.
To further evaluate the effect of alkoxy substitution
at the C6 position, a number of benzimidazoles substi-
tuted at this position were prepared and evaluated for
binding in the MS assay. Introduction of a benzyl group
(2j) as well as a variety of other alkyl chains on the
oxygen atom at C6 did not provide any improvement in
binding affinity (data not shown). Interestingly, intro-
duction of a 3-(dimethylamino)propyloxy side chain at
C6 in compound 3 provided a roughly 10-fold increase
in binding affinity and a 5-fold increase in selectivity
for the 29-mer IIA target. Replacement of the propyl
linker between the cationic headgroup and the oxygen
atom with an ethyl or butyl linker provided analogues
with similar binding affinities for the RNA but reduced
Figure 3. Strategy for preparing conformationally restricted
benzimidazoles.
selectivity (4 and 5). In contrast, removal of the charge
by substitution of the nitrogen atom in side chain with
carbon (6) resulted in a loss of binding affinity.
To eliminate the possibility that the increase in
binding affinity was a result of nonspecific interaction
of the cationic side-chains with the RNA, benzimidazoles
7, 8, and 9 (Figure 2) were prepared and screened in
the MS assay. Gratifyingly, none of the above com-
pounds showed any significant binding to the IIA target
(KD > 100 µM), indicating that the 2-amino group and
the cationic side chains at N1 and C6 were all forming
specific contacts with the RNA.
In the next round of optimization, we identified a
number of motifs as potential replacements for the
dimethylamino headgroup at the C6 position. However,
the motifs that provided the most dramatic increases
in binding affinity tended to be polycationic in nature
and resulted in compounds that were deemed too polar
for potential therapeutic applications (data not shown).
An alternate strategy to optimize binding of benzimid-
azole 3 to the IIA target constrained the flexible side
chains22 at N1 and C6 into the benzimidazole nucleus
to provide compounds 10, 11, 12, and 13 (Figure 3).

Letters
Journal of Medicinal Chemistry, 2005, Vol. 48, No. 23 7101
Scheme 1a
a Reagents and conditions: (a) NaH, Diethyl malonate, THF, -78 °C - rt; (b) LAH, THF, -78 °C - rt (70% over two steps); (c) acetyl
chloride, CH2Cl2, Et3N, DMAP; (d) fuming HNO3, 0 °C, 30 min.; (e) 17, CH2Cl2, CaCO3, rt, 16 h; (f) dry DMSO, MeOH, K2CO2, 50 °C,
54 h (25% over four steps); (g) Pd/C, H2 balloon, MeOH; (h) BzNCS, DIPEA, CH2Cl2; (i) EDC, CH2Cl2 (57% over three steps); (j) 10% HCl,
dioxane, reflux, 8 h; (k) CH2CO2Cl, Et3N, DMAP, CH2Cl2; (l) 40% Me2NH/H2O, DMF, 40 °C.
The synthetic route used to prepare the double-
constrained benzimidazole 13 is described in Scheme
1. Reaction of benzyl bromide 14 with the sodium salt
of diethyl malonate followed by reduction of the alky-
lated diester with LAH provided diol 15 (70% over two
steps). Protection of diol 15 with acetyl chloride followed
by nitration with fuming nitric acid provided compound
16. Selective SNAr displacement of the o-flourine with
amine 17 provided nitroaniline 18. A one-pot deprotec-
tion of the acetyl groups followed by intramolecular
cyclization of the released diol provided benzopyran 19
(25% yield from 15). Further elaboration of benzopyran
19 to the protected benzimidazole 20 was carried out
as described previously.23 Deprotection of the benzoyl
protecting group in 20 was accomplished by refluxing
in 10% aq HCl/dioxane. Treatment of the crude mixture
above with methanesulfonyl chloride resulted in cy-
Figure 4. Mapping data for benzimidazole 13 to 40-mer IIA
clization to the double-constrained benzimidazole nucleus
as well as mesylation of the primary hydroxyl group on
the benzopyran ring. Subsequent displacement with
40% aq dimethylamine provided the desired benzimid-
azole 13 after purification by reverse phase preparative
HPLC.
screening RNA. Protection from enzymatic digestion by RNase
A in the presence of 13 is observed in the lower bulge region
of the stem IIa structure. Structure I was proposed based on
chemical and enzymatic probing,26,27 and Structure II was
proposed based on genetic analysis25 and subsequently identi-
fied by NMR studies.
While this work was in progress, an NMR structure
of domain II was published.24 This structure supported
a fold of stem II25 different from the previously proposed
structure26,27 and called into question the relevance of
our original binding data. To address this, we designed
a construct which contains residues 49-67 on the 5′-
side and 100-114 on the 3′-side of the HCV 5′-UTR.
This 40-mer RNA (Figure 4) contains the key stem IIa
region, yet allows either of the folds proposed in the
literature. We found that compound 3 bound to the 40-
mer RNA with a KD of 17 µM, only slightly weaker than
the 10 µM observed for the 29-mer. Further studies are
in progress to characterize the nature of the structures
of the 29-mer and 40-mer stem IIa screening constructs.
All the constrained benzimidazoles showed signifi-
cantly improved binding to the 40-mer RNA model of
IIA (Table 3). Constraining the N1 side chain (10)
provided a roughly 3-fold increase in binding affinity
relative to benzimidazole 3. The effect of constraining
Table 3. MS Binding, Replicon Activity and Cellular Toxicity
of Selected Benzimidazoles
MS KD 40-mer
(µM)
replicon EC50
MTT (Huh-7) CC50
compd
(µM)
(µM)
1
>100
17
NT
37
>100
>100
>100
>100
>100
>100
3
10
11
12
13
3.5
29
1.7
0.86
0.72
1.5
3.9
5.4
the C6 side chain as a benzofuran (11) or benzopyran
(12) ring was more spectacular and produced a 10-fold
or a 20-fold increase respectively, in binding affinity
relative to benzimidazole 3. The doubly constrained
benzimidazole 13 (racemic mixture of cis and trans
diastereomers) proved to have the highest affinity in
this series with a measured KD of 0.72 µM. This was a
25-fold increase in binding affinity relative to benzim-

7102 Journal of Medicinal Chemistry, 2005, Vol. 48, No. 23
Letters
(8) Sucheck, S. J.; Wong, C.-H. RNA as a target for small molecules.
Curr. Opin. Chem. Biol. 2000, 4, 678-686.
idazole 3 and 140-fold overall increase in binding affinity
relative to the initial hit 1.28
(9) Wilson, W. D.; Li, K. Targeting RNA with small molecules. Curr.
Med. Chem. 2000, 7, 73-98.
The binding site of benzimidazole 13 on the 40-mer
RNA was further investigated by a RNA footprinting
experiment using RNase A digestion (Figure 4). The 40-
mer RNA was incubated with RNase A without any
ligand and the cleavage sites were identified. In the
presence of benzimidazole 13, the backbone phosphates
in the internal loop were strongly protected (>70%
reduction in RNA cleavage) from enzymatic cleavage at
the sites shown in Figure 4. This experiment provides
additional evidence that the benzimidazoles are binding
specifically in a pocket created by the three-dimensional
architecture of the RNA.
The optimized benzimidazoles were finally tested for
activity in an HCV-replicon assay29 where they reduced
HCV RNA levels, as measured by RT-PCR, at low
micromolar concentrations. Notably, the SAR trends
seen in the replicon assay were similar to those observed
in the MS assay. The slightly better activity in the
cellular replicon assay seen with benzimidazole 11 could
be attributed to enhanced cellular penetration proper-
ties of this analogue. All the benzimidazole compounds
also showed minimal toxicity (CC50 > 100 µM) against
Huh-7 cells in an MTT assay.
(10) Chow, C. S.; Bogdan, F. M. A Structural Basis for RNA-Ligand
Interactions. Chem. Rev. 1997, 97, 1489-1513.
(11) Tor, Y. Targeting RNA with small molecules. ChemBioChem
2003, 4, 998-1007.
(12) Swayze, E. E.; Jefferson, E. A.; Sannes-Lowery, K. A.; Blyn, L.
B.; Risen, L. M.; Arakawa, S.; Osgood, S. A.; Hofstadler, S. A.;
Griffey, R. H. SAR by MS: A Ligand Based Technique for Drug
Lead Discovery Against Structured RNA Targets. J. Med. Chem.
2002, 45, 3816-3819.
(13) Spahn, C. M.; Kieft, J. S.; Grassucci, R. A.; Penczek, P. A.; Zhou,
K.; Doudna, J. A.; Frank, J. Hepatitis C virus IRES RNA-induced
changes in the conformation of the 40s ribosomal subunit.
Science 2001, 291, 1959-1962.
(14) Odreman-Macchioli, F.; Baralle, F. E.; Buratti, E. Mutational
Analysis of the Different Bulge Regions of Hepatitis C Virus
Domain II and Their Influence on Internal Ribosome Entry Site
Translational Ability. J. Biol. Chem. 2001, 276, 41648-41655.
(15) Kalliampakou, K. I.; Psaridi-Linardaki, L.; Mavromara, P.
Mutational analysis of the apical region of domain II of the HCV
IRES. FEBS Lett. 2002, 511, 79-84.
(16) Hofstadler, S. A.; Sannes-Lowery, K. A.; Crooke, S. T.; Ecker,
D. J.; Sasmor, H.; Manalili, S.; Griffey, R. H. Multiplexed
Screening of Neutral Mass-Tagged RNA Targets against Ligand
Libraries with Electrospray Ionization FTICR MS: A Paradigm
for High-Throughput Affinity Screening. Anal. Chem. 1999, 71,
3436-3440.
(17) Sannes-Lowery, K. A.; Drader, J. J.; Griffey, R. H.; Hofstadler,
S. A. High-performance mass spectrometry as a drug discovery
tool: a high-throughput screening assay to identify RNA-binding
ligands. Proc. SPIE-Int. Soc. Opt. Eng. 2001, 4264, 27-36.
(18) Sannes-Lowery, K. A.; Drader, J. J.; Griffey, R. H.; Hofstadler,
S. A. Fourier transform ion cyclotron resonance mass spectrom-
etry as a high throughput affinity screen to identify RNA binding
ligands. TrAC, Trends Anal. Chem. 2000, 19, 481-491.
(19) Sannes-Lowery, K. A.; Griffey, R. H.; Hofstadler, S. A. Measuring
Dissociation Constants of RNA and Aminoglycoside Antibiotics
by Electrospray Ionization Mass Spectrometry. Anal. Biochem.
2000, 280, 264-271.
(20) In our assay a hit is described as a ligand with a dissociation
constant of 100 µM or less and a selectivity of 3 or greater for
the target RNA. Selectivity is defined as the ratio of ligand
dissociation constant for the 33-mer vs the 29-mer RNA.
(21) Seth, P. P.; Jefferson, E. A.; Griffey, R. H.; Swayze, E. E.
Benzimidazoles and analogs thereof as antivirals. PCT Int. Appl.
WO 2004050035, 2004.
In conclusion, using mass spectrometry screening as
a tool to guide SAR, we have developed a new class of
small molecules with high affinity for the HCV-IRES
IIA subdomain. The optimized benzimidazoles reduced
viral RNA in a cellular replicon assay at concentrations
comparable to the binding constants observed in the MS
assay. Further investigation into the mechanism by
which the benzimidazoles elicit their biological activity
is in progress and will be reported in due course.
Acknowledgment. We thank Dr. Stanley Crooke for
many helpful discussions and Dr. Erich Koller and Dr.
Larry Blynn for assistance with the replicon assay.
(22) Hart, P. A.; Rich, D. H. Stereochemical aspects of drug action I:
conformational restriction, steric hindrance, and hydrophobic
collapse. In Practice of Medicinal Chemistry; Wermuth, C. G.,
Ed.; Academic Press: San Diego, 1996; pp 393-412.
(23) Seth, P. P.; Robinson, D. E.; Jefferson, E. A.; Swayze, E. E.
Efficient solution phase synthesis of 2-(N-acyl)-aminobenzimid-
azoles. Tetrahedron Lett. 2002, 43, 7303-7306.
(24) Lukavsky, P. J.; Kim, I.; Otto, G. A.; Puglisi, J. D. Structure of
HCV IRES domain II determined by NMR. Nat. Struct. Biol.
2003, 10, 1033-1038.
Supporting Information Available: Detailed experi-
mental procedures for the preparation of compounds 11, 12,
and 13, along with a description of procedures used for the
cellular assays, the MS binding assay, and MS mapping
experiment for benzimidazole 13. This material is available
free of charge via the Internet at http://pubs.acs.org.
(25) Zhao, W. D.; Wimmer, E. Genetic analysis of a poliovirus/
hepatitis C virus chimera: new structure for domain II of the
internal ribosomal entry site of hepatitis C virus. J. Virol. 2001,
75, 3719-3730.
(26) Kieft, J. S.; Zhou, K.; Jubin, R.; Murray, M. G.; Lau, J. Y.;
Doudna, J. A. The Hepatitis C Virus Internal Ribosome Entry
Site Adopts an Ion- dependent Tertiary Fold. J. Mol. Biol. 1999,
292, 513-529.
(27) Honda, M.; Beard, M. R.; Ping, L. H.; Lemon, S. M. A phyloge-
netically conserved stem-loop structure at the 5′ border of the
internal ribosome entry site of hepatitis C virus is required for
cap-independent viral translation. J. Virol. 1999, 73, 1165-1174.
(28) The optimized benzimidazoles typically displayed 30-50-fold
selectivity for the IIA target over other RNA subdomains used
in the screening assay.
(29) Yi, M.; Bodola, F.; Lemon, S. M. Subgenomic hepatitis C virus
replicons inducing expression of a secreted enzymatic reporter
protein. Virology 2002, 304, 197-210.
References
(1) Lauer, G. M.; Walker, B. D. Hepatitis C Virus Infection. N. Engl.
J. Med. 2001, 345, 41-52.
(2) Strader, D. B.; Wright, T.; Thomas, D. L.; Seeff, L. B. Diagnosis,
Management, and Treatment of Hepatitis C. Hepatology 2004,
39, 1147-1171.
(3) Tan, S.-L.; Pause, A.; Shi, Y.; Sonenberg, N. Hepatitis C
therapeutics: current status and emerging strategies. Nat. Rev.
Drug Discovery 2002, 1, 867-881.
(4) Major, M. E.; Feinstone, S. M. The Molecular Virology of
Hepatitis C. Hepatology 1997, 25, 1527-1538.
(5) Kieft, J. S.; Zhou, K.; Jubin, R.; Doudna, J. A. Mechanism of
ribosome recruitment by hepatitis C IRES RNA. RNA 2001, 7,
194-206.
(6) Gallego, J.; Varani, G. The hepatitis C virus internal ribosome-
entry site: a new target for antiviral research. Biochem. Soc.
Trans. 2002, 30, 140-146.
(7) Gallego, J.; Varani, G. Targeting RNA with Small-Molecule
Drugs: Therapeutic Promise and Chemical Challenges. Acc.
Chem. Res. 2001, 34, 836-843.
JM050815O