Pyrrolidine-5,5-trans-lactams. 1. Synthesis and incorporation into inhibitors of hepatitis C virus NS3/4A protease.

Article abstract of DOI:10.1021/ol027013x

[reaction: see text] In this, the first of two letters, we outline the use of the pyrrolidine-5,5-trans-lactam template to design small, neutral, mechanism-based inhibitors of hepatitis C NS3/4A protease. The hitherto unreported reaction of the acyl imini

Full text of DOI:10.1021/ol027013x

ORGANIC
LETTERS
2002
Vol. 4, No. 25
4475-4478
Pyrrolidine-5,5-trans-lactams. 1.
Synthesis and Incorporation into
Inhibitors of Hepatitis C Virus NS3/4A
Protease
David M. Andrews,* Seb J. Carey, Helene Chaignot, Barry A. Coomber,
Norman M. Gray, S. Lucy Hind, Paul S. Jones, Gail Mills, J. Ed Robinson, and
Martin J. Slater
GlaxoSmithKline Medicines Research Centre, Gunnels Wood Road,
SteVenage, SG1 2NY, UK
daVid.m.andrews@gsk.com
Received October 2, 2002
ABSTRACT
In this, the first of two letters, we outline the use of the pyrrolidine-5,5-trans-lactam template to design small, neutral, mechanism-based
inhibitors of hepatitis C NS3/4A protease. The hitherto unreported reaction of the acyl iminium ion precursor 4 with dialkyl-substituted silyl
ketene acetals (e.g., 8b) is described. Compound 12b, with a spirocyclobutyl P1 substituent and a cyclopropylacyl substituent on the lactam
nitrogen, has a kobs/I of 400 M-1 s-1 and demonstrates activity in a replicon cell-based surrogate HCV assay.
Hepatitis C virus (HCV) infects chronically an estimated 3%
of the global human population,1 often leading to cirrhosis,
hepatocellular carcinoma, and liver failure in later life.2 It
has been estimated that of those currently infected, 20% and
4% are likely to develop liver cirrhosis and liver cancer,
respectively, in the next decade.3 Current therapies are based
upon interferon-R, alone or in combination with ribavirin.
Although sustained response rates are markedly improved
using combination therapies, at least 50% of patients fail to
show a sustained response. Additionally, current therapies
have the disadvantage of frequent and severe side-effects.4
The development of new therapies to treat HCV infection
effectively is thus of paramount importance, and is currently
an intensive area of research.5
HCV is a small, enveloped virus, the genome of which is
a 9.5 kb single-stranded RNA that encodes for a single
polyprotein of 3010-3030 amino acids. This polyprotein is
processed by cellular signal peptidases to produce the
structural viral proteins (C, E1, E2, p7), whereas viral
proteases (NS2, NS3) are responsible for the production of
mature nonstructural replicative proteins. The multifunctional
70 kD NS3 protein is the most extensively studied viral
protein.6 The amino terminal third of the protein is a trypsin-
(1) World Health Organization Weekly Epidemiological Record 1997,
72, 65.
(2) (a) Gerber, M. A. J. Hepatol. 1993, 17 (suppl 3), 108. (b) Alter, M.
J. Hepatology 1997, 26, 62. (c) Hoofnagle, J. H. Hepatology 1997, 26, 15S.
(d) Kwong, A. D. AntiViral Res. 1998, 40, 1. (e) Hoofnagle, J. H. Digestion
1998, 59, 563.
(4) (a) Reichard, O.; Scvarcz, R.; Weiland, O. Hepatology 1997, 26, 108S.
(b) Davies, G. L.; et al. N. Eng. J. Med. 1998, 339, 1493. (c) McHutchinson,
J. G.; et al. N. Eng. J. Med. 1998, 339, 1485.
(5) May Wang, Q.; Du, M. X.; Hockman, M. A.; Johnson, R. B.; Sun,
X.-L. Drugs Future 2000, 25, 933.
(3) (a) Di Bisceglie, A. M. Lancet 1998, 351, 351. (b) Brown, J. L. Lancet
1998, 351, 83.
(6) (a) Clarke, B. E. J. Gen. Virol. 1997, 78, 2397. (b) Major, M. E.;
Feinstone, S. H. Hepatology 1997, 25, 1527.
10.1021/ol027013x CCC: $22.00 © 2002 American Chemical Society
Published on Web 11/19/2002

like serine protease that cleaves the NS3-4A, NS4A-NS4B,
NS4B-NS5A, and NS5A-NS5B junctions. Although iso-
lated NS3 is enzymatically active, it forms a heterodimer
with the NS4A cofactor, which is believed to be the
physiologically most relevant form of the enzyme.7 It has
been reported that when appropriate mutations were intro-
duced into the NS3 protease region of the HCV genome,
the infectivity of these RNAs in chimpanzees was abolished.8
NS3 protease is thus an essential viral function and should
prove to be an excellent target for the development of novel
anti-HCV agents.
accessibility. We hypothesised that the stability issue could
be overcome by preparing the R,R-disubstituted trans-lactam
templates 3a-c (Figure 2), which would also lead to a
A number of peptidic and nonpeptidic inhibitors of the
NS3/4A serine protease have been reported.9 The majority
of the former have been competitive inhibitors designed from
peptide substrates or cleavage products, while nonpeptidic
molecules have emerged through random screening and have
displayed noncompetitive mechanisms of action. As a starting
point for the design of inhibitors we sought to take advantage
of the pyrrolidine-5,5-trans-lactam template, suitable sub-
stitution of which allows access to the S1, S1′, and S3-S4
specificity pockets10 and traps the active site serine hydroxyl
group by acylation (Figure 1).
Figure 2. R,R-Disubstituted pyrrolidine-5,5-trans-lactams.
considerable simplification of the synthetic challenge posed
by the template, since it would remove a difficult to introduce
chiral center. Our earlier results had demonstrated that the
S1 pocket was capable of accommodating an ethyl side chain
in either the R- or â-configuration. Modeling studies (based
upon crystal structures of ethyl trans-lactams soaked into
NS3 protease) suggested that the S1 pocket should be capable
of accommodating an R,R-dimethyl or spirocyclobutyl-
substituted trans-lactam. Concise syntheses of 3a-c and their
elaboration into potent, low molecular weight, nonpeptidic
NS3/4A protease inhibitors are reported below.
Dimethyl compound 3a can be accessed by the direct
alkylation methodology of Borthwick et al.,16 but we chose
to utilize the more flexible acyliminium ion methodology
developed by Macdonald et al.12,17 (Scheme 1).
Key intermediate 4 was prepared in a chirally pure form
from L-methionine by reported methods.12 Silyl ketene acetals
8b and 8c were obtained by modification of the method of
Ainsworth et al.18 and subjected to Lewis acid-mediated
coupling with acyliminium ion precursor 4 to yield fully
protected amino esters 5a-c. Trifluoroacetamide hydrolysis
was performed on unpurified 5a-c since a facile acid/base
extraction was found to be sufficient to permit isolation of
the amino esters 6a-c in excellent yield and purity. In all
cases, ring closure with tBuMgCl to form the trans-lactam
proceeded in excellent yield.
Our previous studies had shown that a simple Boc-Valine
substituent at R2 could furnish compounds displaying excel-
lent activity in combination with small, electron-withdrawing
substituents on the lactam nitrogen (see Figure 2). Our earlier
studies had demonstrated that substituting the lactam with a
methanesulfonyl group generated a highly electrophilic series
of compounds susceptible to rapid plasma hydrolysis.
Although R,R-disubstitution on the lactam ring improved
stability, hydrolysis still proceeded at an unacceptable rate
Figure 1. Ethyl pyrrolidine-5,5-trans-lactam template: potential
interactions with protease subsites.
The trans-lactam template has been found to be widely
applicable to a number of serine proteases. These templates,
developed as thrombin,11 human neutrophil elastase (HNE),12
and human cytomegalovirus (HCMV) inhibitors,13 are active
intracellularly, stable in plasma, and orally active in vivo.
The first report concerning the application of this template
to HCV NS3/4A protease described R- and â-ethyl-
substituted trans-lactams.14 The first generation inhibitors
(e.g., 1) suffered from modest hydrolytic stability in plasma
(mediated by nonspecific esterases/hydrolases, analogous to
experience in the elastase area15) and limited synthetic
(7) Kim, J. L.; et al. Cell 1996, 8, 344.
(8) Kolykhalov, A. A.; Mihalik, K.; Feinstone, S. M.; Rice, C. M. J.
Virol. 2000, 74, 2046.
(15) (a) Macdonald, S. J. F.; et al. Bioorg. Med. Chem. Lett. 2001, 11,
895 and references cited therein. (b) Finke, P. E.; et al. J. Med. Chem.
1995, 38, 2449.
(9) Steinkuhler, C.; Koch, U.; Narjes, F.; Matassa, V. G. Curr. Med.
Chem. 2001, 8, 919.
(16) (a) Borthwick, A. D.; et al. Tetrahedron Lett. 2001, 42, 6933. (b)
Borthwick, A. D.; et al. J. Med. Chem. 2002, 45, 1.
(17) For recent reviews, see: (a) Speckamp, W. N.; Moolenaar, M. J.
Tetrahedron 2000, 56, 3817. (b) Hiemstra, H.; Speckamp, W. N. In
ComprehensiVe Organic Synthesis; Trost, B. M., Fleming, I., Eds; Perga-
mon: Oxford, 1991; Vol. 2, pp 1047-1082.
(10) For a description of protease nomenclature, see: Schechter, I.;
Berger, A. Biochem. Biophys. Res. Commun. 1967, 27, 157.
(11) Pass, M.; et al. Bioorg. Med. Chem. Lett. 1999, 9, 1657.
(12) MacDonald, S. J. F.; et al. J. Org. Chem. 1999, 64, 5166.
(13) Borthwick, A. D.; et al. J. Med Chem. 2000, 43, 4452.
(14) Slater, M. J.; et al. Bioorg. Med. Chem. Lett. In press.
(18) Ainsworth, C.; Chen, F.; Kuo, Y. N. J. Organomet. Chem. 1972,
46, 59.
4476
Org. Lett., Vol. 4, No. 25, 2002

Scheme 1. Synthesis of R,R-Disubstituted
Pyrrolidine-5,5-trans-lactam Core Template
Scheme 2. Elaboration of Core Template
(data not shown). In parallel with our structure-based design
efforts, we also sought to capitalize on experience gained
by our co-workers working on the HNE and HCMV protease
targets, and to this end, representative compound collections
from both programs were screened in a medium-throughput
manner. One compound highlighted as a result of this
exercise was an R-methyl-substituted trans-lactam 9,19 bear-
ing a cyclopropylacyl lactam substituent (Figure 3). We
hydrogen over 10% Pd/C in the presence of 1 equiv of HCl
proceeded smoothly to provide amines 11a-c as hydrochlo-
ride salts. HATU-mediated coupling was employed to
introduce the Boc-Valine substituent to provide 12a-c in
excellent yields. Compounds 13a-c were analogously
prepared using methanesulfonyl chloride in place of cyclo-
propane carbonyl chloride.
Compounds were initially assayed in a chromogenic assay,
determining an IC50 following 4 h of enzyme/inhibitor
preincubation, and this enabled the compounds to be crudely
ranked (Table 1). It is apparent that R,R-dimethyl and
spirocyclopentyl are roughly equivalent in potency, being
approximately 10-fold less potent than the spirocyclobutyl
series. The enhanced potency of the latter could be a
consequence of ring strain (reactivity) or a more optimal fit
in S1. In the case of the spirocyclopentyl substituent (and
for dimethyl), the side-chain internal bond angle at the spiro
junction is tetrahedral, whereas in the case of the spirocy-
clobutyl, the angle is near 90°.20 This may contribute to
increased lactam ring strain in the latter, which may result
in enhanced reactivity. Future work (based upon the meth-
odology of Sykes et al.21) will investigate this point.
Alternatively, the improved potency of spirocyclobutyl is a
consequence of a better fit in S1.
Figure 3. Cyclopropyl substituted pyrrolidine-5,5-trans-lactam.
considered that this compound was probably a genuine “hit”
since, although the single R-methyl substituent fills S1
suboptimally, the small lactam substituent should be well
accommodated in the S1′ pocket of NS3 protease. Thus,
compounds 12 and 13 were selected as final targets.
Deprotonation of 7a-c with lithium hexamethyldisilazide
followed by addition of cyclopropane carbonyl chloride
yielded 10a-c. Removal of the benzyl carbamate with
It is also clear that although substituting the lactam
nitrogen with a methanesulfonyl moiety should render the
compounds much more electrophilic than the more sterically
(20) Wilson, A. G.; Goldhamer, D. J. Chem. Educ. 1963, 40, 504.
(21) Sykes, N. O.; Macdonald, S. J. F.; Page, M. I. J. Med. Chem. 2002,
45, 2850.
(19) Borthwick, A. D.; et al. Bioorg. Med. Chem. Lett. 2002, 12, 1719.
Org. Lett., Vol. 4, No. 25, 2002
4477

Table 1. HCV NS3/4A Protease Inhibitory Activity and Plasma Stability of Pyrrolidine-5,5-trans-lactams
R1′
)
HCV protease
IC50 (uM)22
R1′
)
HCV protease
IC50 (uM)
HCV protease
human plasma stability
(% turnover after 4 h)24
23
R )
methanesulfonyl
cyclopropylacyl
K
obs/I (M-1 s-1
)
R-ethyl
1
30
2
9
12a
12b
12c
8.9
97
3
0.51
2.5
64
42
21
28
39
R-methyl
dimethyl
cyclobutyl
cyclopentyl
13a
13b
13c
34
4.4
31
31
400
36
hindered cyclopropylacyl compounds, it is the latter that are
better recognized by the protease. The 10-fold enhancement
in potency demonstrated by this modification arises from
the fact that the planar bond angle of the acyl group secures
better binding than the tetrahedral methanesulfonyl. In
addition, it is suggested that the cyclopropyl moiety fills the
S1′ pocket more completely than the methyl group of the
methanesulfonyl.
as 12b, with kobs/I ca. 400 M-1 s-1 are sufficiently potent to
show replicon25 potency in the single figure micromolar range
(see the following letter in this issue).
In the following paper, we describe our efforts to capitalize
on the superior binding of the spirocyclobutyl trans-lactam
template by further investigation of the requirements of the
lactam substituent as well as the optimization of the amino
acid substituent on the pyrrolidine nitrogen.
Using similar methods to those reported previously, 2 was
synthesized and used as the reference compound to facilitate
comparison of the plasma stability of 12a-c. Compounds
were incubated for 4 h in the presence of human plasma,
and the percentage remaining was analyzed. It was particu-
larly gratifying to note that our hypothesis that it should be
possible to increase potency with concomitant improvement
in stability had been realized (cf. 12b with 2, Table 1).
Acknowledgment. We thank Drs. Berwyn Clarke and
George Hardy for support and encouragement; Drs. Graham
Baker, Sue Bethell, and Malcolm Ellis for provision of NS3
protease protein and initial assay systems; Dr. Martin Johnson
for synthesis of chiral reference materials; Derek Evans for
provision of intermediates; and Drs. Tony Pateman and
Angela Patikis for human plasma stability data.
Supporting Information Available: Detailed experi-
mental procedures for synthesis of representative compounds
and characterization data for test compounds. This material
is available free of charge via the Internet at http://pubs.acs.org.
Once a range of spirocyclobutyl trans-lactam analogues
had been prepared, it was very evident that recording IC50s
following 4 h of preincubation was not capable of delivering
a robust SAR. In the chromogenic assay, the enzyme
concentration was approximately 0.2 uM, so the more potent
analogues were acting as active site titrants. For this reason,
the format of the assay was changed to employ a fluorogenic
substrate and the assay run in continuous readout mode to
generate kobs/I data - a more robust, kinetic measure of
potency. Under these modified conditions, SAR trends were
much more apparent and this assay forms the basis upon
which all future results will be described.
OL027013X
(22) Chromogenic assay: enzyme NS3 protease domain only, 0.2 µM
final concentration; 4A concentration, 10 µM; substrate (Ac-EDVVPC-
pNA) final concentration, 1.5mM; substrate Km ) 1.85 mM; absorbance
read at 405 nm. Test compounds were preincubated with enzyme for 4 h
before the addition of substrate.
(23) Fluorogenic assay: enzyme NS3-4A full-length, 20 nM final
concentration; substrate (aminobenzoyl-E-D-V-V-P-C-S-M-S-Y(3-NO2)-
NH2) 25 µM final concentration (signal increase at Em420 nm (Ex320 nm));
K
obs/I values were obtained at three concentrations (approximately 5-fold
apart); the Kobs/I value was calculated by dividing the Kobs value by the
molar concentration used. Units of measurement are M-1 s-1
In conclusion, we have demonstrated that placing a
spirocyclobutyl substituent R to the lactam carbonyl, in
combination with cyclopropylacyl substitution on the lactam
itself, increases the potency of pyrrolidine-5,5-trans-lactams
by approximately 2 orders of magnitude compared to the
previously reported compounds. In addition, the disubstituted
compounds show markedly enhanced synthetic accessibility
and stability to human plasma hydrolysis. Compounds such
.
(24) Each compound was incubated in fresh human plasma at a
concentration of 25 uM and a temperature of 37 °C; an aliquot was
withdrawn and deproteinated with acetonitrile for 4 h. Samples were assayed
by LC-MS on an API-300 using APCI source and single-ion monitoring.
Results were expressed as percentage turnover.
(25) (a) Lohmann, V.; Korner, F.; Koch, J.; Herian, U.; Theilmann, L.;
Bartenschlager, R. Science 1999, 285 (5424), 110. (b) Blight, K. J.;
Kolykhalov, A. A.; Rice, C. M. Science 2000, 290 (5498), 1972. (c)
Bartenschlager, R.; Lohmann, V. AntiViral Res. 2001, 52, 1.
4478
Org. Lett., Vol. 4, No. 25, 2002