Isothiazoles as active-site inhibitors of HCV NS5B polymerase

Article abstract of DOI:10.1016/j.bmcl.2006.10.002

Isothiazole analogs were discovered as a novel class of active-site inhibitors of HCV NS5B polymerase. The best compound has an IC₅₀ of 200 nM and EC₅₀ of 100 nM, which is a significant improvement over the starting inhibitor (1). The X-ray complex structure of 1 with HCV NS5B was obtained at a resolution of 2.2 A, revealing that the inhibitor is covalently linked with Cys 366 of the 'primer-grip'. Furthermore, it makes considerable contacts with the C-terminus, β-loop, and more importantly, to the active-site of the enzyme. The uniqueness of this binding mode offers a new insight for the rational design of novel inhibitors for HCV NS5B polymerase.

Full text of DOI:10.1016/j.bmcl.2006.10.002

Bioorganic & Medicinal Chemistry Letters 17 (2007) 28–33
Isothiazoles as active-site inhibitors of HCV NS5B polymerase
Shunqi Yan,* Todd Appleby, Esmir Gunic, Jae Hoon Shim, Tania Tasu, Hongwoo Kim,
Frank Rong, Huaming Chen, Robert Hamatake, Jim Z. Wu, Zhi Hong and Nanhua Yao*
Valeant Pharmaceutical Research and Development, 3300 Hyland Ave., Costa Mesa, CA 92626, USA
Received 13 September 2006; revised 1 October 2006; accepted 3 October 2006
Available online 5 October 2006
Abstract—Isothiazole analogs were discovered as a novel class of active-site inhibitors of HCV NS5B polymerase. The best com-
pound has an IC50 of 200 nM and EC50 of 100 nM, which is a significant improvement over the starting inhibitor (1). The X-ray
˚
complex structure of 1 with HCV NS5B was obtained at a resolution of 2.2 A, revealing that the inhibitor is covalently linked with
Cys 366 of the ‘primer-grip’. Furthermore, it makes considerable contacts with the C-terminus, b-loop, and more importantly, to the
active-site of the enzyme. The uniqueness of this binding mode offers a new insight for the rational design of novel inhibitors for
HCV NS5B polymerase.
Ó 2006 Elsevier Ltd. All rights reserved.
Infection by hepatitis C virus (HCV) is a global health
crisis. An estimated more than 170 million individuals
worldwide have been chronically infected, with 3–4 mil-
lion new infections believed to occur each year.1 Many
infected individuals will develop liver damage with an
increased risk of progression to fibrosis, cirrhosis, and
liver cancer.2 The current standard HCV therapy ap-
proved by FDA shows a sustained viral response for
about half of the patients.3 Hence, there is an increasing
demand for new anti-HCV therapies to fulfill this unmet
medical need.
focused on finding other mechanistically unique scaf-
folds. One such effort resulted in the discovery of com-
pound 1 which was subsequently found to be a direct
active-site inhibitor. This compound has an IC50 of
5.9 lM against BK strain of NS5B. Furthermore, it
has an EC50 of 2.0 lM in a cell-based genotype 1b rep-
licon system, and a CC50 of greater than 100 lM. Com-
pound 1 thus provides us with a good lead for
optimization due to its small molecular weight, reason-
able enzymatic and replicon potencies, as well as its min-
imal cell toxicity. Herein, we describe our design,
synthesis, and SAR of isothiazole derivatives as HCV
NS5B polymerase inhibitors, together with an X-ray
crystallographic structural study.
NS5B polymerase is one of the virus-encoded proteins in
the HCV genome and has been demonstrated to be
essential for viral replication and transcription.4 Medic-
inal chemistry pursuits of this polymerase for anti-HCV
drug-discovery have led to the identification of numer-
ous classes of structurally diversified inhibitors.5–13
One of the reported NS5B inhibitors, NM283, is cur-
rently in phase II clinical trials,13 thus validating NS5B
as a target for the discovery of anti-HCV drugs.
CN
H
OH
N
F3C
S
N
CF3
1
The desired compounds were synthesized according to
Scheme 1. In general, anilines (2) were obtained from
commercial sources or prepared by reduction of the ni-
tro compounds. Reaction of aniline (2) with thiophos-
gene afforded isothiocyanate (3), which was then
reacted with 2-cyanoacetimide to form thioamides (4),
which were oxidized to the desired isothiazoles (5).
We have previously described a series of allosteric inhib-
itors for HCV NS5B.5,6 Our continued endeavors were
Keywords: HCV; NS5B; NS5B inhibitor; HCV NS5B polymerase;
HCV NS5B inhibitors; Isothiazole; NS5B active-site inhibitor;
Covalent inhibitor.
Intermediate 7, the key reagent for the synthesis of the
mono-substituted sulfonamide of analog 5, was pre-
pared in three steps from 6 (Scheme 2). Once the aniline
*
Corresponding authors. Tel.: +1 714 545 0100; fax: +1 714 668
3142; e-mail addresses: syan@valeant.com; nyao@valeant.com
0960-894X/$ - see front matter Ó 2006 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2006.10.002

S. Yan et al. / Bioorg. Med. Chem. Lett. 17 (2007) 28–33
29
CN
NO2
NO2
NO2
NH2
c, d
a, b
O2N
NH2
e, f, g
NH2
NH
NCS
R1
O
R1
c
b
R1
O
S
O
O
S
O
S
SO2Cl H2N
O2N
NH2
Cl
R2
NH-Ad
R2
R2
NH-Ad
12
13
14
15
2
3
4
d
Scheme 4. Reagents and conditions: (a) NaNO2, HCl, HOAc, 0 °C,
25 min; (b) SO2(gas), CuCl(I), HOAc, HCl, 0 °C ! rt, 40 min 56%; (c)
Ad-NH2, pyridine, THF, DMAP(cat.), rt, 16 h, 67%; (d) (NH4)2S,
EtOH, reflux, 1 h, 84%; (e) same as (a); (f) CuCl, HCl, 60 °C, 1 h, 72%;
(g) Fe/NH4Cl, EtOH(aq, 70%), 92 °C, 1 h, 84%.
CN
N
-NO2
-NH2
H
OH
N
a
R1
S
R2
5
Scheme 1. Reagents and conditions: (a) Fe/NH4Cl, EtOH(aq, 70%),
90 °C, 2 h, 70–90%; (b) CSCl2, CHCl3, rt, 2 h, 70–95%; (c) 2-
cyanoacetimide, DMF, KOH, rt, 16–40 h, 65–87%, (d) Br2, EtOAc,
rt, 2.5 h, 70–90%.
The compounds prepared above were then evaluated
against a recombinant HCV NS5B polymerase using
the enzymatic assay described previously.17 The IC50
values are shown in Tables 1–3.
As shown in Table 1, all of the compounds bearing elec-
tron-withdrawing group(s) on the phenyl ring have IC50
values in the nano- to low micro-molar range. Replace-
ment of –CF3 by an electron-donating group eliminated
the activity. For example, when R1 and/or R2 were re-
placed by Me, or –OMe, the resulting compounds,
16–18, were inactive with IC50 values greater than
100 lM, indicating the importance of the electron-with-
drawal groups. Introduction of a slightly less electron-
withdrawing, but a slightly bulkier, group (22 and 23),
resulted in about 3–4-fold boost of the IC50, probably
because of the stronger hydrogen-bonding ability of
–CO2R than –CF3. Interestingly, when CF3 (1) was
substituted with F or Cl, the resulting compounds (19–
21) demonstrated significantly enhanced potencies. This
is reflected by the observations that IC50 values of 19–21
are all in the nano-molar range. Particularly, 20 and 21
achieved IC50 values of 300 and 200 nM, respectively,
which are nearly 20–30-fold more potent than the origi-
nal compound (1) (5.9 lM). Such an improvement per-
haps originates from a combined effect of bulkiness
and electron-withdrawing capability. F and Cl atoms
are less bulky than CF3, and they are also slightly less
electron-withdrawing as revealed by their smaller values
of Hammett constant (r)—a parameter measuring the
electron-withdrawing ability.18 For example, F or Cl
has similar r to –CO2R, +0.34 (F), +0.37 (Cl) vs
+0.36 (CO2Me), but less than –CF3 (+0.43).
NH2
NO2
a, b, c
O
S
O
SO3Na
HN
R1
6
7
Scheme 2. Reagents and conditions: (a) SOCl2, DMF, 0 °C ! rt, 3 h,
19%; (b) R1–NH2, TEA, DMAP, THF, rt, overnight, 40%; (c) Fe/
NH4Cl, EtOH(aq, 70%), 90 °C, 3 h, 80%.
(7) was acquired, it was readily converted to the desired
analog 5 according to Scheme 1.
For R1 = CF3, and R2 = SO2NH-adamantane (ꢀAd),
the related analog 5 was synthesized through an inter-
mediate 11 (Scheme 3). Commercially available diamine
(8) was acylated to a mono-amine (9), which was further
converted into sulfonyl chloride (10) via diazotization
according to an established procedure.14,15 Coupling of
10 with Ad-NH2, followed by de-protection, yielded
the desired intermediate (11), which was ultimately con-
verted into the corresponding analog 5 using procedures
described in Scheme 1.
The aniline intermediate 15, required for the synthesis of
analog 5 with R1 = Cl, and R2 = SO2NH-Ad, was gener-
ated according to Scheme 4. Di-nitro-aniline 12 was
converted into sulfonyl chloride 13,14,15 followed by
coupling prior to reduction of one of the two nitro
groups by (NH4)2S,16 to give 14. Another diazotization
to convert –NH2 to –Cl, followed by reduction, afforded
15, which set the stage to synthesize the related analog 5
following the procedures in Scheme 1.
Intrigued by the substantial outcome of electron-with-
drawing groups on the potency, we decided to design
and synthesize compounds containing substituents with
greater r values than CF3. Mono-substituted com-
pounds carrying –SO2NH2 (24), –SO2Me (25), or
–SO2Bu (26), all with r ꢁ +0.60, are slightly weaker
inhibitors, with IC50 values of 9.9, 9.7 and 8.3 lM,
respectively. Compound 27, similarly mono-substituted,
but further linked with a more bulky adamantane (Ad)
group, had a potency increase of 7-fold versus 24 and
4-fold versus 1. The improved IC50 of 27 indicates a
potentially favorable hydrophobic interaction between
the Ad group and the enzyme. Moreover, modeling
suggested that this same Ad group overlaps in part with
an inhibitor described by Pfefferkorn et al.,10,19
prompting us to theorize that its combination with
CF3 or Cl of 1 or 21 might offer superior IC50s. As a
NH2
NHAc
NH2
NHAc
b, c
d, e
a
O
F3C
O
S
F3C
SO2Cl
F3C
NH2 F3C
NH2
NH
8
9
10
11
Ad
Scheme 3. Reagents and conditions: (a) AcCl, pyridine, rt, 24 h,
33%; (b) NaNO2, HCl, HOAc, 0 °C, 30 min; (c) SO2(gas), CuCl(I),
HOAc, HCl, 0 °C ! rt, 20 min, 58%; (d) R–NH2, pyridine, THF,
DMAP(cat.), rt, 16 h, 82%; (e) i—NaOH(aq)/dioxane, 80 °C, 3 h;
ii—HCl(aq, 1 N), rt, 79%.

30
S. Yan et al. / Bioorg. Med. Chem. Lett. 17 (2007) 28–33
Table 1. IC50 values of 3,5-substituted-isothiazole derivatives 5 (in lM)
CN
H
N
OH
R1
N
S
R2
5
Compound
R1
R2
IC50
Compound
R1
R2
IC50
1
16
17
18
19
20
21
22
CF3
Me
CF3
H
5.9
100
100
100
0.9
0.3
0.2
2.0
23
24
25
26
27
28
29
30
CO2H
CF3
H
1.5
9.7
9.7
8.3
1.4
3.5
1.8
0.3
SO2NH2
SO2Me
SO2Bu
Me
OMe
H
Me
H
H
H
F
SO2NH-Ad
SO2NH-Ad
SO2NH-Ad
CF3
H
F
Cl
F
Cl
CF3
Cl
CO2Me
CF3
Br
Table 2. IC50 values of 2,5-substituted-isothiazoloes (lM)
CN
N
R1
H
OH
N
S
R2
Compound
R1
R2
H
IC50
1.3
Compound
R1
R2
Br
IC50
0.3
31
35
O
O
O
32
33
F
0.5
1.0
36
37
–CO2Me
Cl
Cl
4.7
1.8
O
Cl
O
O
O
O
34
Cl
6.4
38
Br
1.7
O
Table 3. IC50 values of other isothiazoles for HCV NS5B (lM)
CN
N
H
OH
N
Ar
S
Compound
Ar
IC50
0.7
Compound
R1
IC50
1.2
Cl
39
41
42
Cl
I
N
S
40
43
1.2
44
F
F
NC
H
OH
N
Br
O N
IC50: 33 lM
Br

S. Yan et al. / Bioorg. Med. Chem. Lett. 17 (2007) 28–33
31
result, compounds, 28 and 29, were designed and pre-
pared. However, they did not achieve much IC50 advan-
tage, but nevertheless were more potent than the
original hit (1) (Table 1).
disulfide bond (Fig. 2). In the same region, the bis-CF3
phenyl moiety is surrounded by a hydrophobic pocket
formed by Pro 197, Asn 316, Ser 367, Leu 384, Met
414, Tyr 415, and Tyr 448 (Fig. 1). Both Cys 366 and
Ser 367 are located in the so-called primer-grip,21 while
Asn 316 and Tyr 448 are in the catalytic site and b-loop,
respectively. This binding is similar to one of the inhib-
itors described by Powers et al.12 What’s more, it ap-
pears that a rotated open-conformation of 1, (E)-
configuration, would fit more suitably into the X-ray
electron-density map than its natal ring-opened form,
(Z)-configuration (Fig. 2). Without excluding other pos-
sible mechanisms, a plausible one is proposed in the
notes.22 Such a rotation projects the nitrile group –CN
closely toward the GDD motif (Gly 317-Asp 318-Asp
319) of the NS5B active site, allowing direct, yet weak,
hydrogen-bonding interactions between –CN and –NH
of backbone Gly 317 and carboxylate of Asp 318
(Fig. 1). Finally, the amide –CONH2 group of the
bound inhibitor is engaged in an extensive hydrogen-
bonding network with Ser 556. For example, one of
the H atoms of –CONH2 hydrogen bonds with both
the –OH group of the side chain of Ser 556 directly
We next turned our attention to the 2,5-bis-substitu-
tions. An initial effort driven by the structural diversity
exploration identified 31 with an IC50 value of 1.3 lM.
Table 1 show that R2 favored a halogen which was
maintained and we thus focused on the substitutions
of R1 in this series. A small focused-set of diversified
R1 groups with the considerations of the size, connectiv-
ity, and electronegativity was then selected and a few
examples of such compounds are shown in Table 2. It
can be seen that all of them have better or similar
IC50s versus 1, with 32 and 35 having excellent IC50 val-
ues of 500 and 300 nM, respectively, yet weaker than
3,5-F2 (20) or 3,5-Cl2 (21). Nonetheless, these results
demonstrate that a bulky hydrophobic group at the 2-
position is well tolerated.
Other substitutions, such as 2,4-dichloro (39), 4-fluoro
(40), and 4-iodo (41), offered decent (Table 3), but re-
duced, potency vs 3,5-Cl2 (or F2) (20 and 21) (Table
1). A compound with a hetero-ring (42) was virtually
inactive (44 lM). Interestingly, a bioisosteric isoxazole
replacement of the isothiazole resulted in a nearly inac-
tive compound (43) (33 lM), highlighting the signifi-
cance of isothiazole ring for attaining high potency
against HCV NS5B polymerase.
˚
(the distance of H ꢂ ꢂ ꢂ OH: 2.8 A), and the –NH of
backbone bridged via a water molecular. The same
backbone –NH group is also involved in the hydro-
gen-bonding interaction with Tyr 448, which is in the
In order to understand the mechanism of inhibition of
isothiazole analogs, the X-ray complex structure of 1
soaked into CD21 NS5B (deletion of the C-terminal 21
amino acids) crystals was obtained at a resolution of
˚
2.2 A. The inhibitor 1 was centered primarily within
the putative RNA-binding site which is otherwise sup-
posedly occupied by the primer (Fig. 1).20 The elec-
tron-density map shows electron-density that is fully
continuous between the sulfur atoms of 1 and Cys
366. In addition, the size of the electron-density cage
that contains these two atoms constrains the atomic
spacing between these two sulfur atoms to a covalent
bond distance. This thus suggests the formation of a
Figure 2. Electron-density map of ring-open form of 1 covalently
linked with Cys 366 via S–S bond from X-ray structure at a resolution
˚
of 2.2 A.
Figure 1. Binding mode of 1. Left: the connolly surface shown near the active site in green and red colors for clarity; the space filling in yellow is from
X-ray complex structure of 1 with NS5B; RNA, template, NTP and Mn2+ were modeled in based on Ref. 20. Right: same X-ray structure of 1 in
complex with NS5B polymerase.

32
S. Yan et al. / Bioorg. Med. Chem. Lett. 17 (2007) 28–33
b-loop of NS5B (Fig. 1). Furthermore, while HCV NS5B
contains 19 cystein residues, only Cys 366 is observed to
be covalently modified, suggestive of the high selectivity
and mechanism-based nature of the inhibitor 1.
In conclusion, isothiazole analogs were discovered to
be a novel class of active-site inhibitors for HCV
NS5B polymerase. The best compound identified has
an IC50 of 200 nM and EC50 of 100 nM, which are
30- and 20-fold better, respectively, over the original
inhibitor (1). An X-ray complex structure of 1 with
NS5B was obtained from a soaked crystal at a
Such a distinctive binding mode of 1 offers a clear in-
sight into the mechanism of inhibition for isothiazole
analogs (Fig. 1). Briefly, the incoming inhibitor (1)
pre-positions itself precisely into the primer site in prox-
imity to Cys 366 to enable the reaction to take place.
Consequently, the resulting ring-open structure is cova-
lent-linked with the ‘primer-grip’. Furthermore, it makes
considerable contacts with the C-terminus, b-loop, and
more importantly the active-site of the enzyme. The
primer-grip has formerly been shown to play a large role
in initiation of viral RNA synthesis.10,12,23 The biologi-
cal function of the C-terminus is related to the regula-
tion of polymerase activity.24 Collectively, this binding
mode of 1, and its related analogs, can be anticipated
to act upon the HCV NS5B polymerase by one or sever-
al of the following mechanisms alone or cooperatively:
(1) preventing proper positioning of the template, (2)
disrupting the suitable entry path of initiating rNTP
substrate for de novo initiation, and (3) locking the
C-terminus into an inactive conformation.
˚
resolution of 2.2 A. The structure revealed that the
inhibitor binds in the active-site and is covalently
linked with Cys 366 of the ‘primer-grip’. Furthermore,
it makes considerable contacts with the C-terminus,
b-loop, and more importantly the active-site of the en-
zyme. The uniqueness of this binding mode offers a
new insight for the rational design of novel inhibitors
of HCV NS5B polymerase.26
Acknowledgments
We thank Vesna Stojiljkovic for compound purifica-
tions; Vicky Lai, Weidong Zhong, Shannon Dempsey,
Heli Walker, and Daniel Bellows for performing biolog-
ical assays.
References and notes
The covalent binding of the inhibitor revealed by the
X-ray crystal structure explains several features of the
SAR. Thus, compound 43, with an isoxazole replacing
the isothiazole ring, cannot form a disulfide bond with
Cys 366 to achieve the same binding mode as 1 and is
therefore nearly inactive. The electronic properties of
the substitutions on the phenyl ring appear to affect
the IC50 by tuning the electrophilicity of the sulfur atom
on isothiazole ring. A stronger electron-withdrawing
group would likely make the sulfur more electrophilic
and thus more accessible to nucleophilic attack by Cys
366. This effect, along with other features such as hydro-
gen-binding ability, hydrophobicity, and an appropriate
molecular shape, may be essential for the desired bind-
ing mode, which in return results in the high potency
as observed from the SAR results.
1. World Health Organization: Hepatitis C—Global Preva-
lence (update). Wkly Epidemiol. Rec. 2000, 75, 18.
2. Memon, M. I.; Memon, M. A. J. Viral Hepat. 2002, 9, 84.
3. Manns, M. P.; McHutchison, J. G.; Gordon, S. C.; Rustgi,
V. K.; Shiffman, M.; Reindollar, R.; Goodman, Z. D.;
Koury, K.; Ling, M.; Albrecht, J. K. Lancet 2001, 358, 958.
4. Kolykhalov, A. A.; Mihalik, K.; Feinstone, S. M.; Rice, C.
M. J. Virol. 2000, 74, 2046.
5. Yan, S.; Appleby, T.; Larson, G.; Wu, J. W.; Hamatake,
R.; Hong, Z.; Yao, N. Bioorg. Med. Chem. Lett. 2006, 16,
5888.
6. Yan, S.; Larson, G.; Appleby, T.; Ding, Y.; Hamatake,
R.; Wu, J. W.; Hong, Z.; Yao, N. Bioorg. Med. Chem.
Lett. 2006. doi:10.1016/j.bmcl.2006.09.095.
7. Chan, L.; Reddy, T. J.; Proulx, M.; Das, S. K.; Pereira, O.;
Wang, W.; Siddiqui, A.; Yannopoulos, C. G.; Poisson, C.;
Turcotte, N.; Drouin, A.; Alaoui-Ismaili, M. H.; Bethell,
R.; Hamel, M.; L’Heureux, L.; Bilimoria, D.; Nguyen-Ba,
N. J. Med. Chem. 2003, 46, 1283.
8. Love, R. A.; Parge, H. E.; Yu, X.; Hickey, M. J.; Diehl, W.;
Gao, J.; Wriggers, H.; Ekker, A.; Wang, L.; Thomson, J. A.;
Dragovich, P. S.; Fuhrman, S. A. J. Virol. 2003, 77, 7575.
9. Harper, S.; Avolio, S.; Pacini, B.; Di Filippo, M.;
Altamura, S.; Tomei, L.; Paonessa, G.; Di Marco, S.;
Carfi, A.; Giuliano, C.; Padron, J.; Bonelli, F.; Migliaccio,
G.; De Francesco, R.; Laufer, R.; Rowley, M.; Narjes, F.
J. Med. Chem. 2005, 48, 4547.
With a better understanding of the inhibition mecha-
nism, it then becomes important to see whether the
cell-based replicon activity (EC50) would match up with
the related enzymatic potency (IC50).25 The EC50s of a
few potent compounds are shown in Table 4. As expect-
ed, EC50s correlated well with the corresponding IC50s.
Moreover, all compounds have marginal or no cell tox-
icity. The 3,5-Cl2 compound 21 represents the most
active one, exhibiting an IC50, EC50, and CC50 of 200,
100 nM and 52 lM, respectively.
10. Pfefferkorn, J. A.; Greene, M. L.; Nugent, R. A.; Gross,
R. J.; Mitchell, M. A.; Finzel, B. C.; Harris, M. S.; Wells,
P. A.; Shelly, J. A.; Anstadt, R. A.; Kilkuskie, R. E.;
Kopta, L. A.; Schwende, F. J. Bioorg. Med. Chem. Lett.
2005, 15, 2481.
Table 4. EC50 from replicon assay for the selected compounds
Compound
IC50 (lM)
EC50 (lM)
CC50 (lM)
11. Beaulieu, P. L.; Gillard, J.; Bykowski, D.; Brochu, C.;
Dansereau, N.; Duceppe, J.-S.; Hache, B.; Jakalian, A.,
et al. Bioorg. Med. Chem. Lett. 2006, 16, 4987.
12. Powers, J. P.; Piper, D. E.; Li, Y.; Mayorga, V.; Anzola,
J.; Chen, J. M.; Jaen, J. C.; Lee, G.; Liu, J.; Peterson, M.
G.; Tonn, G. R.; Ye, Q.; Walker, N. P.; Wang, Z. J. Med.
Chem. 2006, 49, 1034.
1
19
20
21
32
41
5.9
0.9
0.3
0.2
0.5
1.2
2.0
2.9
0.22
0.1
3.4
1.1
100
202
145
52
42
81

S. Yan et al. / Bioorg. Med. Chem. Lett. 17 (2007) 28–33
33
13. (a) Pierra, C.; Benzaria, S.; Amador, A.; Moussa, A.;
Mathieu, S.; Storer, R.; Gosselin, G. Nucleosides Nucle-
otides Nucleic Acids 2005, 24, 767; (b) Valopicitabine
(NM283): www.idenix.com.
14. Hoffman, R. V. Org. Synth. 1981, 60, 121.
15. Hoffman, R.V. Org. Synth. 1990, Coll. Vol. 7, 508.
16. Li, X.; Chu, S.; Feher, V. A.; Khalili, M.; Nie, Z.;
Margosiak, S.; Nikulin, V.; Levin, J.; Sprankle, K. G.;
Tedder, M. E.; Almassy, R.; Appelt, K.; Yager, K. M.
J. Med. Chem. 2003, 46, 5663.
17. Shim, J.; Larson, G.; Lai, V.; Naim, S.; Wu, J. Z. Antiviral
Res. 2003, 58, 243.
18. Hansch, C.; Leo, A. J. Exploring QSAR—Fundamentals
and Applications in Chemistry and Biology; American
Chemical Society: Washington, DC, 1995.
rotated (E)-configuration rather than its natal (Z) form.
One of the plausible mechanisms is given below. Initially,
1 pre-positions itself in the active site to allow for
favorable contacts with the enzyme. A proton shifts to
the lone pair N of isothiazole ring from the –SH of Cys
366 to form B and makes the thio strongly nucleophilic,
which in return attacks the S atom of the isothiazole ring
to form lactim C, prior to its natal amide tautomer D: (Z)-
configuration. (Leung-Toung, R. et al. Bioorg. Med.
Chem. 2003, 11, 5529). Double bond then shifts to
conjugate with phenyl to form imine (E), followed by
the rotation of the single bond in order to make more
favorable hydrogen bond interaction, then shift of double
bond back resulting in the final state of binding of the
rotated (E)-configuration.
Ser 556
Ser 556
O
H
O
H
N
N
H
H
O
N
O
N
H
H
F3C
F3C
NH
S
N
S
O
O
O
O
CF3
CF3
Asp 319
Asp 319
S
SH
B
1
Cys 366
Cys 366
Ser 556
O
Ser 556
H
O
H
N
N
H
N
H
O
2
O
N
H
F3C
F3C
NH
S
NH
O
S
O
O
O
CF3
S
Asp 319
CF3
S
Asp 319
C
Cys 366
Cys 366
D: "natal (Z)-configuration"
Ser 556
O
Ser 556
H
O
H
N
H
H
N
O
S
tautomerization
rotate
O
N
H
N
F3C
NH
S
2
O
O
F3C
N
O
O
CF3
S
Asp 319
CF3
S
Asp 319
E
Cys 366
Cys 366
F: "rotated (E)-configuration"
19. GOLD-v2.1 www.ccdc.cam.ac.uk, The Cambridge Crys-
tallographic Data Centre, UK.
20. Bressanelli, S.; Tomei, L.; Roussel, A.; Incitti, I.; Vitale, R.
L.; Mathieu, M.; De Francesco, R.; Rey, F. A. Proc. Natl.
Acad. Sci. U.S.A. 1999, 96, 13034.
21. Jacobo-Molina, A.; Ding, J.; Nanni, R. G.; Clark, A. D.,
Jr.; Lu, X.; Tantillo, C.; Williams, R. L.; Kamer, G.;
Ferris, A. L.; Clark, P., et al. Proc. Natl. Acad. Sci.
U.S.A. 1993, 90, 6320.
22. We postulate that NS5B coordinates the change of the
ring-open conformations of bound 1 so that it adopts a
23. Hong, Z.; Cameron, C. E.; Walker, M. P.; Castro,
C.; Yao, N.; Lau, J. Y.; Zhong, W. Virology 2001,
285, 6.
24. Ago, H.; Adachi, T.; Yoshida, A.; Yamamoto, M.;
Habuka, N.; Yatsunami, K.; Miyano, M. Structure
1999, 7, 1417.
25. Lohmann, V.; Korner, F.; Koch, J. O.; Herian, U.;
Theilmann, L.; Bartenschlager, R. Science 1999, 285, 110.
26. PDB: the coordinates discussed in the paper has been
deposited in the PDB database (www.rcsb.org) with the
PDB code: 2IJN.