Biaryl guanidine inhibitors of in vitro HCV-IRES activity

Article abstract of DOI:10.1016/j.bmcl.2004.07.066

An SAR study on a high-throughput screening lead for HCV-IRES translation inhibition led to the identification of low μM inhibitors.

Full text of DOI:10.1016/j.bmcl.2004.07.066

Bioorganic & Medicinal Chemistry Letters 14 (2004) 5139–5143
Biaryl guanidine inhibitors of in vitro HCV-IRES activity
Elizabeth A. Jefferson,* Punit P. Seth, Dale E. Robinson, Dana K. Winter, Alycia Miyaji,
Stephen A. Osgood, Eric E. Swayze and Lisa M. Risen
Ibis Therapeutics, A Division of Isis Pharmaceuticals, Inc., 2292 Faraday Avenue, Carlsbad, CA 92008, USA
Received 6 May 2004; revised 28 July 2004; accepted 28 July 2004
Available online 20 August 2004
Abstract—A structure–activity relationship analysis was carried out on a high-throughput small molecule screening lead for HCV-
IRES translation inhibition. The study led to the identification of a guanidine-based structure with low lM inhibitory activity.
Ó 2004 Elsevier Ltd. All rights reserved.
Over 170 million people are infected with hepatitis C
virus (HCV), representing 1–3% of the world popula-
tion.1,2 Persistent HCV infections can lead to liver disor-
ders such as cirrhosis and hepatocellular carcinomas
that eventually require organ transplantation.3 Current
therapies, based on combinations of pegylated interfer-
ons and ribavirin, provide a sustained response in only
a portion of the treated patients and side effects can be
quite severe.4 Since there is still no effective, well-toler-
ated treatment for HCV infection, alternative novel
therapies are desperately needed.5
inhibitory activity to exploit the full potential of this
target.
A library of ꢀ180,000 compounds, biased towards
structured RNA targets, was evaluated for HCV-IRES
transcription/translation inhibitory activity. Inhibition
of IRES-mediated translation was detected by use of a
dicistronic plasmid, as described by Honda et al.14 In
the dual reporter assay the dicistronic RNA transcript
is expressed containing sequences encoding the reporters
separated by a functional HCV-IRES sequence. The
translation of the upstream Renilla luciferase reading
frame is initiated by the cap-dependent mechanism while
translation of the downstream firefly luciferase reading
frame is controlled by the IRES.
A new approach to combat HCV infection would be a
small molecule inhibitor of the HCV internal ribosome
entry site (IRES).6 The IRES, a highly structured non-
translated domain of 340–342 bases located at the 50 ter-
minus of the HCV genome,7 allows HCV to bypass the
normal cap-dependent cellular pathway for protein syn-
thesis. The IRESÕs unique function and highly conserved
sequence make it an interesting target for selective antiv-
iral therapy.5,8 Furthermore, an HCV-IRES inhibitor is
especially attractive as it would intervene early in the
life-cycle of the virus. The IRES has been targeted by
oligonucleotides9 and artificial ribozymes,10 but little
progress has been made in the development of small
molecule inhibitors.11–13 New small molecule lead struc-
tures need to be identified and optimized for HCV-IRES
From the high-throughput screening exercise, biaryl
guanidines 1 and 2 were identified as inhibitors of
HCV-IRES translation, with some selectivity (ꢀ2.5-
fold) over cap-dependent translation (Fig. 1).15 Guan-
idine 2 (protonated guanidinium group at physiological
Keywords: HCV-IRES; HCV; Inhibitor; RNA.
*
Corresponding author. Tel.: +1 760 603 2533; fax: +1 760 603
4654; e-mail: ejeffers@isisph.com
Figure 1. Inhibitors of the HCV-IRES identified from high-through-
put screening (left). Components of the SAR study on screening lead 2
(right).
Undergraduate Medicinal Chemistry COOP student (University of
Ottawa, Canada).
0960-894X/$ - see front matter Ó 2004 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2004.07.066

5140
E. A. Jefferson et al. / Bioorg. Med. Chem. Lett. 14 (2004) 5139–5143
pH) was selected as the starting point for an exploratory
structure–activity relationship (SAR) study (Fig. 1).
Table 1 summarizes the SAR of the initial set of com-
pounds in this study. Extension of the carbon linker to
three methylene units (compound 3), decreased activity
ꢀ15-fold, compared to screening lead 2, with the two
carbon linker. The guanidine NHÕs are crucial for activ-
ity as N-methylation removes the inhibitory activity for
guanidines 4 and 5. Replacement of the guanidine core
with an urea (6)19 or a thiourea moiety (7)20 also abol-
ishes activity. When the pyrimidine nitrogens are re-
placed with carbons, activity is also lost (analogs 8
and 9) and removing a single pyrimidine methyl substi-
tuent resulted in a 2.8-fold loss in potency for guanidine
10. These modifications highlight the importance of the
guanidine H-bonding motif for inhibitory activity. Evi-
dently, biaryl guanidine 2 provided the best starting
structure for further SAR analysis.
The biaryl guanidines were prepared using either solid
or solution-phase methodologies. In the solid phase
approach, the pyrimidine isothiocyanate, prepared by
treating 2-amino-4,6-dimethylpyrimidine with di-2-pyr-
idyl thionocarbonate,16 was reacted with ArgoGelÒ-
Rink-NH2 resin to generate a resin bound thiourea,
according to Scheme 1.17 The thiourea, activated with
1-ethyl-3-(30-dimethylaminopropyl) carbodiimide hydr-
ochloride (EDC), was then treated with an aryl amine
to afford the guanidine derivative, after resin cleavage.
The appropriate amines were obtained from commercial
sources or prepared using standard protocols, as de-
scribed in the literature.18 In the solution-phase ap-
proach, the amino pyrimidine was treated with either
Fmoc- or benzylcarbonyl-protected isothiocyanate to
give the thiourea derivative, according to Scheme 2.
The thiourea was then reacted with an amine to afford
the guanidine after deprotection. All structures were
purified by reversed-phase HPLC to >95% homogeneity
(isolated as AcOH salts in most cases). Compounds gave
Table 2 summarizes the effect of a variety of aryl
substituents on HCV-IRES inhibitory activity. The
order of potency of the 4-substituted analogs is
H > OMe ꢀ Me ꢀ CO2Me ꢀ F ꢀ Cl > NO2 ꢀ OH ꢀ Br
ꢀ NH2 > Ph ꢀ OEt ꢀ OCF3 ꢀ CF3 > CH(Me)2 > CO2H,
C(Me)3. There is no clear SAR trend based on the
electronic nature of the substituents. Steric bulk at the
4-position reduced activity significantly as 4-C(Me)3
1
spectral data (LC–MS and H NMR) consistent with
their assigned structures.
Scheme 1. Reagents and conditions: (a) DPT (di-2-pyridyl thionocarbonate), DCM, reflux, 5h; (b) ArgoGelÒ-Rink-NH2, rt, 12h; (c) 10equiv EDC,
20equiv DIPEA, DCM, rt, 30min; (d) 10equiv R1NH2, 40°C, 20h; (e) 35% TFA, DCM, 2h, rt, 55–60% over three steps based upon initial resin
substitution (HPLC-purified).
Scheme 2. Reagents and conditions: (a) X-NCS, X = Fmoc or COPh, DCM; (b) 2equiv R1NH2, 2equiv EDC, 4equiv DIPEA, DCM; (c) X = Fmoc,
PS-trisamine resin, DMF; (d) X = COPh, 10% HCl, dioxane, 55–60% over three steps (HPLC-purified).
Table 1. Initial SAR for screening lead 2
R2
N
R1
N
Y
R3
n
X
Z
R4
Compd
n
X
R1
R2
Y
Z
R3
R4
HCV-IRES IC50 (lM)
Selectivitya
1
2
3
4
5
6
7
8
9
10
1
2
3
2
2
2
2
2
2
2
NH
NH
NH
NH
NH
OH
S
H
H
H
Me
H
H
H
H
H
H
H
H
H
H
Me
N
H
H
H
H
N
N
N
N
N
N
N
N
C
N
N
N
N
N
Me
N
C
Me
Me
Me
Me
Me
Me
Me
Me
Me
Me
76
9.6
2.6
2.4
1.3
––
150
>200
>200
––
––
Me
>200
Me
Me
Me
H
Me
Me
Me
Me
>200
>200
>200
46
––
––
––
3.5
NH
NH
NH
C
N
N
a Selectivity = (non-IRES IC50/HCV-IRES IC50). Selectivity values are not reported for compounds with non-IRES IC50Õs > 200lM.

E. A. Jefferson et al. / Bioorg. Med. Chem. Lett. 14 (2004) 5139–5143
5141
Table 2. Aryl SAR
Table 3. Heterocycle SAR
H
N
H
N
H
H
N
N
N
N
X
NH
N
NH
N
X
Compd
X
H
HCV-IRES
IC50 (lM)
Selectivitya
Compd
X
HCV-IRES
IC50 (lM)
Selectivitya
2
9.6
25
2.4
4.1
2
Ph
9.5
39
2.4
2.4
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
4-OMe
4-OEt
4-Ph
82
80
>2.4
>2.5
3.0
N
38
39
40
4-OH
4-Me
49
34
3.8
4-OCF3
4-CO2H
4-NO2
4-C(Me)3
4-CH(Me)2
4-CF3
92
>200
54
>2.2
––
32
55
4.0
N
1.5
>200
104
82
––
>3.6
>1.9
2.2
N
4-NH2
4-CO2Me
4-F
61
29
2.5
3.4
O
41
42
97
38
>2.0
1.3
23
24
4.1
2.3
4-Cl
4-Br
50
15
2.5
1.8
S
2-Me
2-Cl
16
13
1.8
1.8
2-OMe
3-Cl
3-Me
O
12
23
1.8
2.8
43
44
25
23
2.8
1.7
2,5-Me
2,4-Me
2,4,6-OMe
2,4-OMe
2,5-OMe
3,5-OMe
35
23
5.7
3.3
NH
N
100
68
>2.0
2.6
N
94
26
>2.1
>7.7
a Selectivity = (non-IRES IC50/HCV-IRES IC50). Selectivity values are
45
23
3.7
not reported for compounds with non-IRES IC50Õs > 200lM.
a Selectivity = (non-IRES IC50/HCV-IRES IC50). Selectivity values are
not reported for compounds with non-IRES IC50Õs > 200lM.
and –CH(Me)2 substitutions were not tolerated. Substi-
tution was slightly better tolerated at the 2 and 3 posi-
tions (2-Me > 3-Me > 4-Me and 2-Cl ꢀ 3-Cl > 4-Cl).
The ionizable 4-carboxy substitution abolishes the activ-
ity of biaryl guanidine 17. The aryl ring was substi-
tuted for a variety of heterocycles, according to Table
3. The order of potency of the heterocyclic modifications
Table 4. Linker SAR
X
H
N
H
N
N
are
2-benzofuran ꢀ 2-benzimidazole ꢀ 3-quinoline >
NH
N
3-pyridine ꢀ 2-pyridine ꢀ 2-thiophene > 4-pyridine > 2-
furan. The fused heterocyclic substitutions, with better
p-stacking potential, were slightly more potent than
the others. Since neither the aryl nor the heterocyclic
substitutions (Tables 2 and 3) gave better HCV-IRES
inhibitory activity over screening lead 2, the phenyl
group was retained for the linker and pyrimidine SAR.
Compd
X
H
HCV-IRES
IC50 (lM)
Selectivitya
2
9.5
35
2.4
5.7
Me
Me
OH
OH
Ph
46
47
21
75
9.7
75
1.9
>2.7
2.2
A few substitutions were made on the linker between
the phenyl group and the guanidine core, according to
Table 4. Methyl substitution decreased inhibitory activ-
ity and enantiomer 47 was 1.7-fold more potent than
enantiomer 46. For hydroxy substitution at the same
position, inhibitory activity was retained for enantiomer
49 and ꢀ8-fold reduced for enantiomer 48. The potency
difference between enantiomers 48 and 49 may suggest
the accessibility of a somewhat structured binding site.
48
49
50b
>2.7
a Selectivity = (non-IRES IC50/HCV-IRES IC50).
b Racemic mixture.

5142
E. A. Jefferson et al. / Bioorg. Med. Chem. Lett. 14 (2004) 5139–5143
Table 5. Pyrimidine SAR
(See Scheme 3) were prepared using standard literature
protocols.21 Guanidines 55 and 56 were prepared using
straightforward procedures, according to Scheme 3.
H
N
H
N
N
R1
R2
NH
N
In the pyrimidine SAR set, replacing methyl with meth-
oxy substituents in guanidine 51 resulted in a 3.6-fold
loss in activity. Activity was only slightly reduced for
52 with the ethyl substitution. Substitution was tolerated
at the 4-position (R2) of the pyrimidine and the order of
potency is Me > H ꢀ Br ꢀ OMe > OH.
R3
Compd R1
R2
R3
HCV-IRES Selectivitya
IC50 (lM)
2
Me
OMe
Et
H
Me
OMe 35
9.6
2.4
5.0
2.2
1.3
2.2
3.6
3.4
2.0
51
52
53
54
55
56
57
H
H
Et
13
4.2
8.1
19
Me
Me
Me
Me
Me
Me
Br
Me
Me
Me
Me
Since substitution was tolerated at the 4-position of the
pyrimidine ring, we chose to introduce a dimethylami-
nopropoxy side-chain, a potential RNA-binding motif,
at this position.22,23 Guanidine 57 was prepared accord-
ing to Scheme 4 in a straightforward manner24 and was
found to be 4.5-fold more potent than screening lead 2.
Unfortunately, an improvement in selectivity did not
accompany the enhanced inhibitory activity for 57.
OH
OMe
9.9
2.1
O(CH2)3– Me
N(Me)2
a Selectivity = (non-IRES IC50/HCV-IRES IC50).
Phenyl substitution on the linker resulted in a ꢀ8-fold
reduction in HCV-IRES activity of racemic biaryl
guanidine 50, compared to parent structure 2 (Table 4).
In summary, the SAR study led to the identification of
guanidine 57 as a 2lM inhibitor of HCV-IRES transla-
tion. Overall, the screening lead structure 2 was rather
resistant to chemical optimization as attempts to iden-
tify additional binding contacts with numerous H-bond-
ing and p-stacking functionalities did not improve
activity. A modest 4.5-fold improvement in HCV-IRES
inhibitory activity was observed for biaryl guanidine 57
with the cationic dimethylaminopropoxy side-chain sub-
stitution. Future SAR studies can be focused on improv-
For the exploratory pyrimidine SAR study (Table 5), a
limited number of amino pyrimidine precursors were
commercially available. Compounds 51–53 were pre-
pared according to Scheme 1, with the appropriate
amino pyrimidines. The amino pyrimidine precursor
required for the synthesis of 51 was commercially avail-
able, while precursors for guanidines 52, 53 and 54
Scheme 3. Reagents and conditions: (a) Br2, AcOH, 0°C, 50%; (b) Ba(OH)2, Cu–bronze powder, H2O, 180°C, 20h, 22%; (c) Fmoc-NCS, acetone,
12h; (d) EDC, DIPEA, 0°C, DCM, 15min; (e) Ph–(CH2)2–NH2, rt, 12h; (f) PS-trisamine resin, DMF, 12h, rt; (g) MeOH, PPh3, diisopropyl
diazodicarboxylate (DIAD), THF, 12h, rt, 25% over five steps.
Scheme 4. (a) Reagents and conditions: (a) PhCONCS, acetone–DCM, 1:1, 12h; (b) EDC, DIPEA, DCM, 0°C, 15min; (c) Ph(CH)2–NH2, 12h, rt;
(d) Br–(CH2)3–OH, acetone, K2CO3, 12h, reflux; (e) MsCl, TEA, DMAP, DCM, 0°C then 12h, rt; (f) 2M (CH3)2–NH in THF, 40°C, 12h; (g) 10%
HCl/dioxane, 12h, reflux.

E. A. Jefferson et al. / Bioorg. Med. Chem. Lett. 14 (2004) 5139–5143
5143
12. Ikeda, M. S.; Tsuai, I.; Zuao, I.; Ryan, H.; Iyan, S.; Kai,
Y.; Kako, Y.; Tsukada, I.; Yanagisawa, M. JP8268890,
1996.
13. Kai, Y. S.; Kaku, Y.; Tsukada, I.; Yanagisawa, M.
JP7069899, 1996.
ing the selectivity and potency of new lead structure 57.
In general, the pharmacophoric information presented
in this paper is useful for the design and optimization
of HCV-IRES small molecule inhibitors.
14. Honda, M.; Kaneko, S.; Matsushita, E.; Kobayashi, K.;
Abell, G. A.; Lemon, S. M. Gastroenterology 2000, 118,
152–162.
Acknowledgements
15. Compounds 1 and 2 as well as a representative set of
active compounds in the SAR analysis showed no inhibi-
tion of message translation from either reporter gene. IC50
values were determined from dose–response curves using
10 concentrations per compound. Reported values are the
average of two–three determinations.
We thank Professor Stanley M. Lemon (The University
of Texas Medical Branch at Galveston) for helpful
suggestions.
References and notes
16. Kim, S.; Yi, K. Y. Tetrahedron Lett. 1985, 26, 1661–
1664.
1. Di Bisceglie, A. M. Lancet 1998, 351, 351–355.
2. Choo, Q. L.; Kuo, G.; Weiner, A. J.; Overby, L. R.;
Bradley, D. W.; Houghton, M. Science 1989, 244,
359–362.
17. Li, M.; Wilson, L. J.; Portlock, D. E. Tetrahedron Lett.
2001, 42, 2273–2275.
18. Amundsen, L. H.; Nelson, L. S. J. Am. Chem. Soc. 1951,
73, 242–244.
3. Berenguer, M.; Wright, T. L. Antiviral Ther. 1998, 3,
125–136.
4. Di Bisceglie, A. M.; McHutchison, J.; Rice, C. M.
Hepatology 2002, 35, 224–231.
5. Tan, S.-L.; Pause, A.; Shi, Y.; Sonenberg, N. Nat. Rev.
Drug Discovery 2002, 1, 867–881.
6. Gallego, J.; Varani, G. Biochem. Soc. Trans. 2002, 30,
140–146.
19. Synthesis of compound 6: Hutchins, S. M.; Chapman, K.
T. Tetrahedron Lett. 1994, 35, 4055–4058.
20. Synthesis of compound 7: Gomez, L.; Gellibert, F.;
Wagner, A.; Mioskowski, C. J. Comb. Chem. 2000, 2,
75–79.
21. Wendelin, W.; Harler, A. Monatsh. Chem. 1975, 106,
1479–1493.
22. Maier, M. A.; Leeds, J. M.; Balow, G.; Springer, R. H.;
Bharadwaj, R.; Manoharan, M. Biochemistry 2002, 41,
1323–1327.
7. Major, M. E.; Feinstone, S. M. Hepatology 1997, 25,
1527–1538.
8. Jubin, R. Curr. Opin. Mol. Ther. 2001, 3, 278–287.
9. Zhang, H.; Hanecak, R.; Brown-Driver, V.; Azad, R.;
Conklin, B.; Fox, M. C.; Anderson, K. P. Antimicrob.
Agents Chemother. 1999, 43, 347–353.
10. Ryu, K. J.; Lee, S. W. J. Biochem. Mol. Biol. 2003, 36,
538–544.
23. Wyatt, P. G.; Bethell, R. C.; Cammack, N.; Charon, D.;
Dodic, N.; Dumaitre, B.; Evans, D. N.; Green, D. V. S.;
Hopewell, P. L.; Humber, D. C.; Lamont, R. B.; Orr, D.
C.; Plested, S. J.; Ryan, D. M.; Sollis, S. L.; Storer, R.;
Weingarten, G. G. J. Med. Chem. 1995, 38, 1657–1665.
1
24. Structural data for 57: H NMR (CD3OD, 300MHz): d
11. Wang, W.; Preville, P.; Morin, N.; Mounir, S.; Cai, W.;
Siddiqui, M. A. Bioorg. Med. Chem. Lett. 2000, 10,
1151–1154.
7.21 (m, 5H), 3.77 (t, 2H, J = 6.2), 3.58 (t, 2H, J = 6.8),
2.90 (t, 2H, J = 6.2), 2.56 (m, 2H), 1.91 (m, 2H), 1.80 (s,
6H); MS (ES+) 371.1 (MH+).