Structure-based design of a benzodiazepine scaffold yields a potent allosteric inhibitor of hepatitis C NS5B RNA polymerase

Article abstract of DOI:10.1021/jm9005548

HCV NS5B polymerase, an essential and virus-specific enzyme, is an important target for drug discovery. Using structure-based design, we optimized a 1,5-benzodiazepine NS5B polymerase inhibitor chemotype into a new sulfone-containing scaffold. The design

Full text of DOI:10.1021/jm9005548

J. Med. Chem. 2009, 52, 4099–4102 4099
DOI: 10.1021/jm9005548
lead to improved therapeutic outcomes for HCV patients in
the future.2
Structure-Based Design of a Benzodiazepine
Scaffold Yields a Potent Allosteric Inhibitor of
Hepatitis C NS5B RNA Polymerase§
Inhibition of the essential, virus-specific nonstructural pro-
tein 5B RNA-dependent RNA polymerase of HCV (HCV
NS5B) with small molecules is a compelling target for drug
discovery, and several HCV NS5B inhibitors are presently
undergoing clinical evaluation.2a,5 The current state of the art
suggests that targeting the active site (nucleoside inhibitors) or
one of four distinct allosteric binding sites (non-nucleoside
inhibitors) of this enzyme represents a viable approach to a
new mode of HCV chemotherapy.6 We have recently de-
scribed a new class of 1,5-benzodiazepine (BZD, Figure 1)3
inhibitors of HCV NS5B that bind to an allosteric site near
the interface of the palm and thumb subdomains. Other
chemotypes known to bind at or near this site include
benzothiadiazines,7 pyrrolidines,8 acrylic acids,9 rhoda-
nines,10 isothiazoles,11 benzofurans,12 and proline sulfona-
mides.4,13 Using structure-based design, we have extracted
interaction information from the publicly available 3D struc-
ture of an NS5B-inhibitor complex with an unrelated chemo-
type and applied this to our BZD series.
‡
‡
ꢀ
Koen Vandyck, Maxwell D. Cummings, Origene
Nyanguile,‡ Carlo W. Boutton,‡ Sandrine Vendeville,‡
David McGowan,‡ Benoit Devogelaere,‡ Katie Amssoms,‡
Stefaan Last,‡ Klara Rombauts, ‡ Abdellah Tahri,‡ Pedro
Lory,‡ Lili Hu,‡ Derek A. Beauchamp,† Kenny Simmen,‡
and Pierre Raboisson*,‡
‡Tibotec BVBA, Generaal de Wittelaan L 11B 3, B-2800 Mechelen,
Belgium, and †Johnson & Johnson Pharmaceutical Research &
Development, Raritan, New Jersey 08869
Received April 30, 2009
Abstract: HCV NS5B polymerase, an essential and virus-specific
enzyme, is an important target for drug discovery. Using structure-
based design, we optimized a 1,5-benzodiazepine NS5B polymerase
inhibitor chemotype into a new sulfone-containing scaffold. The
design yielded potent inhibitor (S)-4c (KD = 0.79 nM), which has
∼20-fold greater affinity for NS5B than its carbonyl analogue (R)-2c.
˚
We determined the 2.75 A resolution crystal structure of the
NS5B-(R)-1b complex (PDB code 3GNV), and observed that
the carbonyl O of bound (R)-1b forms an intermolecular
hydrogen bond with Tyr448:N (Figure 2). Superimposition of
the published structure of the NS5B complex of 3 (PDB code
2GC8),4 an inhibitor chemotype distinct from that of 1, revealed
that one sulfonamide oxygen of bound 3makes a similar contact
with Tyr448:N (Figure 2). In an adjacent region of the binding
site, a carboxylate O of bound 3 forms another intermolecular
hydrogen bond, in this case with Gly449:N (Figure 2).
Our optimization effort had by this time progressed from
the chloro series to the closely related fluoro series, with larger
N-acyl substituents (cf. 1b and 2c in Figure 1), and we had
established that the Cl-to-F substitution did not appear to
affect the inhibitor binding mode.3,14 Thus, 2c is derived from
It is estimated that 170 million people, 3% of the world’s
population, are infected with hepatitis C virus (HCVa), a
condition that poses a high risk of ultimately leading to liver
cirrhosis and hepatocellular carcinoma.1 The current treat-
ment of HCV infection, involving a combination of pegylated
interferon-R and ribavirin, is associated with numerous side
effects and yields at best a 50% sustained virological response
(SVR) for genotype 1 infected patients.2 Ongoing research
efforts anticipate that a combination of small-molecule drugs
targeting different essential viral proteins, in an approach
similar to the current standard of care for HIV therapy, will
Figure 1. 1,5-Benzodiazepine and proline sulfonamide inhibitors
of HCV NS5B.3,4
*To whom correspondence should be addressed. Phone: ++32
15461823. Fax: +32-(0)15-40-12-57. E-mail: pierreraboisson@aol.com,
praboisson@its.jnj.com.
§ The atomic coordinates and structure factors (PDB codes 3GNV
and 3GNW for NS5B-(R)-1b and NS5B-(S)-4c, respectively) have been
deposited in the Protein Data Bank, Research Collaboratory for Struc-
tural Bioinformatics, Rutgers University, New Brunswick, NJ (http://
www.rcsb.org/).
a Abbreviations: HCV, hepatitis C virus; SVR, sustained virological
response; BZD, 1,5-benzodiazepine; TFA, trifluoroacetic acid.
˚
Figure 2. Overlay of the 2.75 A resolution crystal structure of BZD
(R)-1b (green, by atom) bound to HCV NS5B (white, by atom) with
that of proline sulfonamide 3 (purple, by atom), indicating the
common hydrogen bond with Tyr448:N and the additional hydro-
gen bond of 3 with Gly449:N.
r
2009 American Chemical Society
Published on Web 06/09/2009
pubs.acs.org/jmc

4100 Journal of Medicinal Chemistry, 2009, Vol. 52, No. 14
Scheme 1. Preperation of Sulfone-BZD Compounds 4a,ca
Vandyck et al.
a Reagents: (a) 1 equiv TFA, DMF, 50 °C, 6 h; (b) 0.9 equiv of NaHCO3, 4-benzyloxy-2-fluorobenzaldehyde (9), 50 °C, 12 h; overall yield for 4a (a and
b) of 18%; (c) 6-methyl-2-pyridinecarbonyl chloride, DIPEA, THF, room temp, 12 h; (d) LiOH, H2O/THF, room temp, 3 h, combined yield of rac-4c
(c and d) of 82%; (e) Chiralpak AD, hexane/2-propanol/MeOH (7 N NH3) 60/30/10.
1b.3 On the basis of our structural observations, we hypothe-
Table 1. NS5B Polymerase Affinity (KD), Inhibition (IC50), and Re-
plicon Activity (EC50) of BZD Inhibitors
sized that replacement of the carbonyl of 2c by a sulfone
compd
K
D ( μM)
IC50 ( μM)
EC50 ( μM)
moiety would yield a benzodiazepine inhibitor (4c) that forms
an additional intermolecular hydrogen bond, maintaining the
interaction with Tyr448:N and gaining the Gly449:N contact,
thus improving binding affinity. Small-molecule conforma-
tional analysis supported the conclusion that the bound
conformation observed for 1b (and predicted for 2c) was
equally accessible to 4c (see below), and calculated interaction
energies suggested improved affinity for the sulfone.14
(R)-1b
(R)-2c
rac-4c
(R)-4c
(S)-4c
nd
0.093
0.074
0.12
1.2
0.015
nd
nd
0.40
0.052
4.8
6.7
0.026
0.00079
0.029
Scheme 2. Preparation of Ketosulfone 5a,14,15
a Reagents: (a) thiourea, EtOH, reflux, 16 h, 95%; (b) NaOH, H2O,
90 °C, 4 h, 70%; (c) NaOMe, 2,2-dimethyloxirane, MeOH, room temp,
16 h; 95%; (d) formic acid, 80 °C, 1 h, 34%; (e) Oxone, CH2Cl2, H2O,
room temp, 4 h, 36%.
The synthetic route for this previously unreported tricyclic
benzodiazepine system is outlined in Scheme 1. Condensation
of ketosulfone 5, obtained in five steps from commercial 2,3-
dichloropropene (Scheme 2)14,15 with 2,3-diaminophenol 6
was performed in DMF in the presence of 1 equiv of TFA,
yielding a mixture of 7 and 8, which was then converted into a
mixture of 4a and 10 by reaction with aldehyde 9. This
synthetic sequence differs from that of the corresponding
carbonyl BZD (e.g., 1);3 the original Dean-Stark conditions
were modified, for the enamines 7 and 8, to a lower reaction
temperature (50 °C) in the presence of TFA to avoid the
formation of side products. Interestingly, TFA also catalyzed
the formation of the isomeric mixture of sulfone-BZD (4a/10,
∼6/4), allowingomissionofthe intermediate isolationof7and
8. After separation of regioisomers 4a and 10, scaffold 4a,
obtained in 18% isolated yield after a one-pot, two-step
procedure, was converted to the racemate 4c, which was then
separated into enantiomers (S)-4c and (R)-4c by chiral chro-
matography. The absolute configuration of (S)-4c was estab-
lished by single crystal X-ray crystallography (see below).14
The rationally designed sulfone (S)-4c exhibits a marked
19-fold increase in NS5B binding affinity over the parent
˚
Figure 3. The 2.4 A resolution crystal structure of (S)-4c bound to
HCV NS5B polymerase.
carbonyl compound (R)-2c (Table 1, KD values determined by
surface plasmon resonance14). Importantly, most of the
potency enhancement translates to antiviral activity, based
on the observed 14-fold improvement in replicon activity
(Table 1). The contrasting 3-fold improvement in IC50 is
attributed to the low sensitivity of the enzyme assay, which
often precludes discrimination of highly active compounds by
this method. The greater discrepancy between replicon activ-
ityandbindingaffinityfor(S)-4ccomparedto(R)-2c(Table1)
is also fully consistent with the increased polar surface area of
the designed compound.
˚
We determined the 2.4 A resolution crystal structure of the
NS5B-(S)-4c complex (PDB code 3GNW). Consistent with
our structure-based design rationale, the observed binding
contacts include the predicted intermolecular hydrogen bonds
for the sulfone moiety, the designed contact with Gly449 and
the “conserved” (from the original carbonyl analogue (R)-2c)

Letter
Journal of Medicinal Chemistry, 2009, Vol. 52, No. 14 4101
References
(1) World Health Organization (WHO). Hepatitis C. Fact Sheet No.
164. Revised October 2000. http://www.who.int/mediacentre/
factsheets/fs164/en/ (2000).
(2) (a) Manns, M. P.; Foster, G. R.; Rockstroh, J . K.; Zeuzem, S.;
Zoulim, F.; Houghton, M. The way forward in HCV treatment;
finding the right path. Nat. Rev. Drug Discovery 2007, 6, 991–1000.
(b) De Clercq, E. The design of drugs for HIV and HCV. Nat. Rev.
Drug Discovery 2007, 6, 1001–1018.
(3) (a) Nyanguile, O.; Pauwels, F.; Van den Broeck, W.; Boutton, C.
W.; Quirynen, L.; Ivens, T.; van der Helm, L.; Vandercruyssen, G.;
Mostmans, W.; Delouvroy, F.; Dehertogh, P.; Cummings, M. D.;
Bonfanti, J.-F.; Simmen, K. A.; Raboisson, P. 1,5-Benzodiaze-
pines, a novel class of hepatitis C virus polymerase nonnucleoside
inhibitors. Antimicrob. Agents Chemother. 2008, 52, 4420–4431. (b)
McGowan, D.; Nyanguile, O.; Cummings, M. D.; Vendeville, S.;
Vandyck, K.; Van den Broeck, W.; Boutton, C. W.; Quirynen, L.;
Amssoms, K.; Bonfanti, J.-F.; Last, S.; Rombauts, K.; Lin, T.-I.;
Tahri, A.; Hu, L.; Delouvroy, F.; Surleraux, D.; Lory, P.; Kinder-
man, N.; Pille, G.; Simmen, K.; Raboisson, P. 1,5-Benzodiazepine
inhibitors of HCV NS5B polymerase. Bioorg. Med. Chem. Lett.
2009, 19, 2492–2496.
Figure 4. Overlay of the small molecule crystal structure (pink) and
the NS5B-bound structure (color by atom) of (S)-4c.
(4) Gopalsamy, A.; Chopra, R.; Lim, K.; Ciszewski, G.; Shi, M.;
Curran, K. J.; Sukits, S. F.; Svenson, K.; Bard, J.; Ellingboe, J.
W.; Agarwal, A.; Krishnamurthy, G.; Howe, A. Y. M.; Orlowski,
M.; Feld, B.; O’Connell, J.; Mansour, T. S Discovery of proline
sulfonamides as potent and selective hepatitis C virus NS5b poly-
merase inhibitors. Evidence for a new NS5b polymerase binding
site. J. Med. Chem. 2006, 49, 3052–3055.
(5) (a) Liu-Young, G.; Kozal, M. J. Hepatitis C protease and poly-
merase inhibitors in development. AIDS Patient Care STDs 2008,
22 (6), 449–457. (b) Beaulieu, P. L. Non-nucleoside inhibitors of the
HCV NS5B polymerase: progress in the discovery and develop-
ment of novel agents for the treatment of HCV infections. Curr.
Opin. Invest. Drugs 2007, 8 (8), 614–634.
(6) Kwong, A. D.; McNair, L.; Jacobson, I.; George, S. Recent
progress in the development of selected hepatitis C virus NS3.4A
protease and NS5B polymerase inhibitors. Curr. Opin. Pharmacol.
2008, 8, 1–10.
(7) Dhanak, D.; Duffy, K. J.; Johnston, V. K.; Lin-Goerke, J.;
Darcy, M.; Shaw, A. N.; Gu, B.; Silverman, C.; Gates, A. T.;
Nonnemacher, M. R.; Earnshaw, D. L.; Casper, D. J.; Kaura, A.;
Baker, A.; Greenwood, C.; Gutshall, L. L.; Maley, D.; DelVecchio,
A.; Macarron, R.; Hofmann, G. A.; Alnoah, Z.; Cheng, H.-Y.;
Chan, G.; Khandekar, S.; Keenan, R. M.; Sarisky, R. T. Identifi-
cation and biological characterization of heterocyclic inhibitors of
the hepatitis C virus RNA-dependent RNA polymerase. J. Biol.
Chem. 2002, 277, 38322–38327.
interaction with Tyr448:N (Figure 3). The designed sulfone
group captures the intermolecular hydrogen bonds of the sulfone
and carboxylate moieties of bound 3. In all other respects, and
as expected, the binding mode of (S)-4c (PDB code 3GNW) is
essentially identical to that of the parent carbonyl (R)-1b
(PDB code 3GNV).16
Small-molecule conformational analysis indicated that the
benzodiazepine scaffold is quite rigid, and much of the
calculated conformational diversity stems from the relatively
flexible benzyloxy group.14 These studies suggest that, with
the exception of the relatively flexible benzyloxy moiety, the
required binding conformation for (R)-2c and (S)-4c is close
to energetically optimal. The benzyloxy group is predicted to
prefer a fully extended conformation rather than the more
compact conformation observed to bind to NS5B. The small-
molecule crystal structure of (S)-4c is also consistent with a
preference of the benzyloxy group for a fully extended con-
formation over the more compact conformation seen in the
NS5B complex (Figure 4).14 Overall, we conclude that the
NS5B-bound conformation of (S)-4c and related compounds
in these chemical series is relatively favorable and readily
accessible. Binding to NS5B requires adoption of a subopti-
mal conformation for the flexible benzyloxy moiety of the
inhibitor.
(8) Burton, G.; Ku, T. W.; Carr, T. J.; Kiesow, T.; Sarisky, R. T.;
Lin-Goerke, J.; Baker, A.; Earnshaw, D. L.; Hofmann, G. A.;
Keenan, R. M.; Dhanak, D. Identification of small molecule
inhibitors of the hepatitis C virus RNA-dependent RNA polymer-
ase from a pyrrolidine combinatorial mixture. Bioorg. Med. Chem.
Lett. 2005, 15, 1553–1556.
In summary, a structure-based design approach generated
a novel sulfone-BZD chemotype that enabled further pro-
gress in HCV drug discovery, yielding specific inhibitors of
HCV NS5B polymerase with low nanomolar potencies in
biochemical and cell-based assays. Crystallography validated
our design process, clearly underscoring the value of public
and proprietary 3D structural information in the drug dis-
covery process. Further characterization of molecules in this
series is ongoing, and these results will be reported in due
course.
(9) 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.
Inhibitors of HCV NS5B polymerase. Part 1: Evaluation of the
southern region of (2Z)-2-(benzoylamino)-3-(5-phenyl-2-furyl)ac-
rylic acid. Bioorg. Med. Chem. Lett. 2005, 15, 2481–2486.
(10) Powers, J. P.; Piper, D. E.; Li, Y.; Mayorga, V.; Anzola, J.; Chen, J.
M.; Jaen, J. C.; Lee, G.; Liu, J.; Peterson; Tonn, G. R.; Ye, Q.;
Walker, N. P. C.; Wang, Z. SAR and mode of action of novel non-
nucleoside inhibitors of hepatitis C NS5b RNA polymerase. J.
Med. Chem. 2006, 49, 1034–1046.
(11) Yan, S.; Appleby, T.; Gunic, E.; Shim, J. H.; Tasu, T.; Kim, H.;
Rong, F.; Chen, H.; Hamatake, R.; Wu, J. Z.; Hong, Z.; Yao, N.
Isothiazoles as active-site inhibitors of HCV NS5B polymerase.
Bioorg. Med. Chem. Lett. 2007, 17, 28–33.
(12) Howe, A. Y. M.; Cheng, H.; Johann, S.; Mullen, S.; Chunduru, S.
K.; Young, D. C.; Bard, J.; Chopra, R.; Krishnamurthy, G.;
Mansour, T.; O’Connell, J. Molecular mechanism of hepatitis C
virus replicon variants with reduced susceptibility to a benzofuran
inhibitor, HCV-796. Antimicrob. Agents Chemother. 2008, 52,
3327–333.
Acknowledgment. The authors thank Natalie Kindermans
(Tibotec) for analytical SFC, Jef Proost and Hilde Vanbaelen
(J&JPRD, Beerse) for preparative SFC, and Hendrik De
Bondt (Tibotec) for PDB file submission. Protein crystal-
lography was by Sabine Hoppner and Christine Wenzkowski
(Proteros).
(13) Koch, U; Narjes, F. Recent progress in the development of
inhibitors of the hepatitis C virus RNA-dependent RNA polymer-
ase. Curr. Top. Med. Chem. 2007, 7, 1302–1329.
(14) See Supporting Information.
(15) (a) Lansbury, P. T.; Nienhouse, E. J.; Scharf, D. J.; Hilfiker, F. R.
General approach to cycloalkanone synthesis. Intramolecular
Supporting Information Available: Experimental and com-
putational details and crystallographic data for (S)-4c. This
material is available free of charge via the Internet at http://
pubs.acs.org.

4102 Journal of Medicinal Chemistry, 2009, Vol. 52, No. 14
Vandyck et al.
alkylation of 2-chloro-1-olefins. J. Am. Chem. Soc. 1970, 92, 5649–
5657. (b) Lansbury, P. T.; Scharf, D. J. A facile entry into 3-thianone
and 3-piperidone ring systems. J. Am. Chem. Soc. 1968, 90, 536–7.
(16) The impact on affinity and activity of the larger R1 group of (S)-4c
compared to that of (R)-1b is accounted for in the design “parent”
(R)-2c (Figure 1, Table 1). On the basis of an extensive set of 3D
complex structures in this series, both published3 and unpublished,
and related computational studies,14 we have determined that the
nature of the R1 group (Figure 1) does not affect the binding mode
observed for the remainder of the inhibitor structure.