β-Lactones as a new class of cysteine proteinase inhibitors: inhibition of hepatitis A virus 3C proteinase by N-Cbz-serine β-Lactone

Article abstract of DOI:10.1021/ol990148r

Formula presented N-Benzyloxycarbonyl-L-serine β-lactone (1) is shown to irreversibly inactivate the 3C cysteine proteinase of hepatitis A virus (HAV) with k_inact = 0.70 min^-1, K_I = 1.84 x 10^-4 M and k_inact/K_I = 3800 M^-1 min^-1 at an enzyme concentration of 0.1 μM. Mass spectrometric and HMQC NMR studies using ¹³C-labeled 1 show that the active site cysteine (Cys-172) thiol of the HAV 3C proteinase attacks the β-position (i.e. C-4) of the oxetanone ring, thereby leading to ring opening and alkylation of the sulfur. In contrast, the enantiomer of this β-lactone, 2, is a reversible competitive inhibitor (K_i = 1.50 x 10^-6 M) at similar enzyme concentrations. The β-lactone motif represents a new class of inhibitors of cysteine proteinases.

Full text of DOI:10.1021/ol990148r

ORGANIC
LETTERS
1
999
Vol. 1, No. 5
03-806
â-Lactones as a New Class of Cysteine
Proteinase Inhibitors: Inhibition of
Hepatitis A Virus 3C Proteinase by
N-Cbz-serine â-Lactone
8
Manjinder S. Lall, Constantine Karvellas, and John C. Vederas*,†
Department of Chemistry, UniVersity of Alberta, Edmonton, Alberta, Canada T6G 2G2
john.Vederas@ualberta.ca
Received July 1, 1999
ABSTRACT
N-Benzyloxycarbonyl-L-serine â-lactone (1) is shown to irreversibly inactivate the 3C cysteine proteinase of hepatitis A virus (HAV) with kinact
-
1
-4
-1
-1
)
Ι Ι
0.70 min , K ) 1.84 × 10 M and kinact/K ) 3800 M min at an enzyme concentration of 0.1 µM. Mass spectrometric and HMQC NMR
13
studies using C-labeled 1 show that the active site cysteine (Cys-172) thiol of the HAV 3C proteinase attacks the â-position (i.e. C-4) of the
oxetanone ring, thereby leading to ring opening and alkylation of the sulfur. In contrast, the enantiomer of this â-lactone, 2, is a reversible
competitive inhibitor (K
-
6
i
) 1.50 × 10 M) at similar enzyme concentrations. The â-lactone motif represents a new class of inhibitors of
cysteine proteinases.
6
Considerable effort is devoted to the study and inhibition of
cysteine proteinases because they are targets for development
of therapeutic agents for many diseases, including viral
infections, parasitic ailments, arthritis, cancer, and osteoporo-
ruses.3,5 Crystal structures have been reported for HAV 3C,
7
8
HRV 3C, and poliovirus 3C proteins and show that
structurally these enzymes resemble trypsin-like serine
2
,3
proteinases. Enzyme inhibition studies on HAV 3C and
HRV 3C proteinases have employed a large variety of
compounds as potential therapeutic leads, including peptide
1
sis. The 3C cysteine proteinases of picornaviruses attract
extensive attention because they are essential for maturation
and replication of pathogens such as hepatitis A virus (HAV),
human rhinovirus (HRV) (a causative agent for common
9
,10
11
12
aldehydes,
peptide fluoromethyl ketones, â-lactams,
1
3
14
isatins, homophthalimides, vinylogous esters and sul-
cold), poliovirus, foot and mouth disease virus, and encepha-
lomyocarditis virus.1-3 This family of viruses has a positive
(4) Schultheiss, T.; Sommergruber, W.; Kusov, Y.; Gauss-M u¨ ller, V. J.
Virology 1995, 69, 1727-1733.
strand RNA genome and generates a single polyprotein that
undergoes multiple proteolytic cleavages to produce the
mature viral proteins. In HAV, these cleavages are initiated
(
5) Dougherty, W. G.; Semler, B. L. Microbiol. ReV. 1993, 57, 781-
822.
(6) Bergmann, E. M.; Mosimann, S. C.; Chernaia, M. M.; Malcolm, B.
A.; James, M. N. G. J. Virol. 1997, 71, 2436-2448.
(7) Matthews, D. A.; Smith, W. W.; Ferre, R. A.; Condon, B.; Budahazi,
G.; Sisson, W.; Villafranca, J. E.; Janson, C. A.; McElroy, H. E.; Gribskov,
C. L.; Worland, S. Cell 1994, 77, 761-771.
(8) Mosimann, S. C.; Cherney, M. M.; Sia, S.; Plotch, S.; James, M. N.
G. J. Mol. Biol. 1997, 273, 1032-1047.
(9) Malcolm, B. A.; Lowe, C.; Shechosky, S.; McKay, R. T.; Yang, C.
C.; Shah, V. J.; Simon, R. J.; Vederas, J. C.; Santi, D. V. Biochemistry
1995, 34, 8172-8179.
(10) Webber, S. E.; Okano, K.; Little, T. L.; Reich, S. H.; Xin, Y.;
Fuhrman, S. A.; Matthews, D. A.; Love, R. A.; Hendrickson, T. F.; Patick,
A. K.; Meador, J. W., III; Ferre, R. A.; Brown, E. L.; Ford, C. E.; Binford,
S. L.; Worland, S. T. J. Med. Chem. 1998, 41, 2786-2805.
4
by the 3C proteinase, an enzyme present in all picornavi-
†
Tel: (780) 492-5475. FAX: (780) 492-8231.
(1) For reviews, see: (a) Roose, J. P.; Van Noorden, C. J. F. J. Lab.
Clin. Med. 1995, 125, 433-441. (b) Otto, H.-H.; Schirmeister, T. Chem.
ReV. 1997, 97, 133-171. (c) Thompson, S. K. Bioorg. Med. Chem. 1999,
7
, vii and following articles in issue.
(
2) Malcolm, B. A. Protein Sci. 1995, 4, 1439-1445.
(
3) (a) Bergmann, E. M.; James, M. N. G. In Proteinases of Infectious
Agents; Dunn, B., Ed.; Academic Press: San Diego, CA, 1999; pp 139-
1
1
63. (b) Gorbalenya, A. E.; Snijder, E. J. Perspect. Drug DiscoV. Design
996, 6, 64-86.
1
0.1021/ol990148r CCC: $18.00 © 1999 American Chemical Society
Published on Web 08/12/1999

fones,1
5-17
halomethyl carbonyls,18,19 and azapeptide com-
either the carbonyl or at the â-position25 suggests that
cysteine proteinases could be irreversibly inactivated by
â-lactones having correct substituents and stereochemistry
(Figure 1). It seemed that N-Cbz-L-serine-â-lactone (1) would
pounds.1
8,20,21
Some of these substances show in vitro
11,13-17
antiviral activity in cell culture;
however, toxicity can
be a problem for molecules that are generally reactive with
13
thiols. We now report that a motif not previously descibed
for cysteine proteinase inhibition, namely the â-lactone
22,23
functionality
in the enantiomeric serine derivatives 1 and
2, can display good irreversible and reversible inhibition,
respectively, of the HAV 3C enzyme.
â-Lactones occur naturally in a variety of organisms, and
24
many possess potent biological activity. The ability of thiols
to open the four-membered ring by nucleophilic attack at
(11) Morris, T. S.; Frormann, S.; Shechosky, S.; Lowe, C.; Lall, M. S.;
Gauss-M u¨ ller, V.; Purcell, R. H.; Emerson, S. U.; Vederas, J. C.; Malcolm,
B. A. Bioorg. Med. Chem. 1997, 5, 797-807.
Figure 1. Possible modes of nucleophilic attack on N-benzyloxy-
carbonyl (Cbz) serine â-lactones.
(
12) Skiles, J. W.; McNeil, D. Tetrahedron Lett. 1990, 31, 7277-7280.
(13) Webber, S. E.; Tikhe, J.; Worland, S. T.; Fuhrman, S. A.;
Hendrickson, T. F.; Matthews, D. A.; Love, R. A.; Patick, A. K.; Meador,
J. W.; Ferre, R. A.; Brown, E. L.; DeLisle, D. A.; Ford, C. E.; Binford, S.
L. J. Med. Chem. 1996, 39, 5072-5082.
be a reasonable initial target for inhibition of the viral 3C
proteinases because of its facile preparation by Mitsunobu
cyclization of N-Cbz-L-serine, its simple scaffold which
(14) Wang, Q. M.; Johnson, R. B.; Jungheim, L. N.; Cohen, J. D.;
Villarreal, E. C. Antimicrob. Agents Chemother. 1998, 42, 916-920.
(15) Dragovich, P. S.; Webber, S. E.; Babine, R. E.; Fuhrman, S. A.;
25
Patick, A. K.; Matthews D. A.; Lee, C. A.; Reich, S. H.; Prins, T. J.;
Marakovits, J. T.; Littlefield, E. S.; Zhuo, R.; Tikhe, J.; Ford, C. E.; Wallace,
M. B.; Meador, J. W., III; Ferre, R. A.; Brown, E. L.; Binford, S. L.; Harr,
J. E. V.; DeLisle, D. M.; Worland, S. T. J. Med. Chem. 1998, 41, 2806-
permits structural variation for subsequent structure-activity
2
studies, and its benzyl group which may mimic the P ′
phenylalanine side chain in HAV 3C substrates.1
9,26
A
2
818.
16) Kong, J.; Venkatraman, S.; Furness, K.; Nimkar, S.; Shepherd, T.
A.; Wang, Q. M.; Aub e´ , J.; Hanzlik, R. P. J. Med. Chem. 1998, 41, 2579-
587.
17) Dragovich, P. S.; Prins, T. J.; Zhuo, R.; Webber, S. E.; Marakovits,
(
possible concern is the susceptibility of 1 to hydrolysis since
R-amino-â-lactones bearing no â-substituent display low
2
25
(
stability in basic aqueous media. However, the half-life for
hydrolysis of 1 in phosphate buffer at pH 7.5 is 76 min,
which is sufficiently long for enzyme inhibition studies.
J. T.; Fuhrman, S. A.; Patick, A. K.; Matthews D. A.; Lee, C. A.; Ford, C.
E.; Burke, B. J.; Rejto, P. A.; Hendrickson, T. F.; Tuntland, T.; Brown, E.
L.; Meador, J. W., III; Ferre, R. A.; Harr, J. E. V.; Worland, S. T. J. Med.
Chem. 1999, 42, 1213-1224.
1
Despite the absence of a P glutamine side chain important
(18) Sham, H. L.; Rosenbrook, W.; Kati, W.; Betebenner, D. A.;
for substrate recognition, 1 is a good time-dependent
irreVersible inhibitor of HAV 3C proteinase (kinact ) 0.70
Wideburg, N. E.; Saldivar, A.; Plattner, J. J.; Norbeck, D. W. J. Chem.
Soc., Perkin Trans. 1 1995, 1081-1082.
(19) McKendrick, J. E.; Frormann, S.; Luo, C.; Semchuk, P.; Vederas,
-1
-4
-1
-1
min , K
Ι
) 1.84 × 10 M, kinact/K
Ι
) 3800 M min ) at
J. C.; Malcolm, B. A. Int. J. Mass Spectrom. 1998, 176, 113-124.
20) Venkatraman, S.; Kong, J.; Nimkar, S.; Wang, Q. M.; Aub e´ , J.;
Hanzlik, R. P. Bioorg. Med. Chem. Lett. 1999, 9, 577-580.
21) Yanting, H.; Malcolm, B. A.; Vederas, J. C. Bioorg. Med. Chem.
998, 7, 607-619.
22) Tetrahydrolipstatin (Orlistat) inactivates human pancreatic lipase by
2
7
(
an enzyme concentration of 0.1 µM. Interestingly the
enantiomer 2 is a competitive reVersible inhibitor of HAV
C proteinase (K
(
-
6
3
i
) 1.50 × 10 M) at a similar enzyme
1
2
8,29
(
concentration.
The possibility that compound 2 may in
acylating a serine hydroxyl of the enzyme: (a) L u¨ thi-Peng, Q.; M a¨ rki, H.
P.; Hadv a´ ry, P. FEBS Lett. 1992, 299, 111-115. (b) Hadv a´ ry, P.; Sidler,
W.; Meister, W.; Vetter, W.; Wolfer, H. J. Biol. Chem. 1991, 266, 2021-
fact be a time-dependent inhibitor but that this was not seen
under the assay conditions could be eliminated because
studies at other pH conditions (e.g. pH 6) and with varying
concentrations of inactivator 2 also display simple competive
behavior. Clearly the HAV 3C active site shows different
modes of binding for enantiomers 1 and 2, with only the
former leading to permanent covalent modification of the
active site (see below). Furthermore, the inhibitory properties
of 1 and 2 were not affected by short exposure to 10-fold
molar excess concentrations of dithiothreitol, suggesting that
â-lactones of this type could be specific enzyme inhibitors
that would not react inadvertently with ubiquitous biological
thiols (e.g. glutathione).
2
027. (c) Other â-lactones may also acylate related lipases: Kocienski, P.
J.; Pelotier, B.; Pons, J.-M.; Prideaux, H. J. Chem. Soc., Perkin Trans. 1
1
998, 1373-1382 and references therein. (d) â-Lactones such as the
antibiotic F-244 (1233A) may acylate a cysteine residue of HMG CoA
synthase: Romo, D.; Harrison, P. H. M.; Jenkins, S. I.; Riddoch, R. W.;
Park, K.; Yang, H. W.; Zhao, C.; Wright, G. D. Bioorg. Med. Chem. 1998,
6
, 1255-1272. Mayer, R. J.; Louis-Flamberg, P.; Elliott, J. D.; Fisher, M.;
Leber, J. Biochem. Biophys. Res. Commun. 1990, 169, 610-616.
23) Omuralide, the â-lactone derived from lactacystin, has been shown
(
to inhibit the 20S proteosome by O-acylation of a threonine residue: (a)
Corey, E. J.; Li, W. D. Z. Chem. Pharm. Bull. 1999, 47, 1-10. (b) Corey,
E. J.; Li, W. D. Z.; Nagamitsu, T.; Fenteany, G. Tetrahedron 1999, 55,
3
305-3316.
(
24) For reviews of natural â-lactones, see: (a) Lowe, C.; Vederas, J.
C. Org. Prep. Proced. Int. 1995, 27, 305-346. (b) Pommier, A.; Pons, J.
M. Synthesis 1995, 7, 729-744. (c) For a review of syntheses of â-lactones,
see: Yang, H. W.; Romo, D. Tetrahedron 1999, 55, 6403-6434.
(25) For a discussion of reactions and leading references, see: (a) Pansare,
(28) Segel, I. H. In Enzyme Kinetics; John Wiley & Sons: New York,
1975; pp 125-135.
S. V.; Huyer, G.; Arnold, L. D.; Vederas, J. C. Org. Synth. 1991, 70, 1-9.
(
1
b) Pansare, S. V.; Arnold, L. D.; Vederas, J. C. Org. Synth. 1991, 70,
0-17. (c) Arnold, L. D.; Kalantar, T. H.; Vederas, J. C. J. Am. Chem.
(29) The wild-type HAV 3C has 219 amino acids with a molecular weight
of 24 kilodaltons, exists as an active monomer, and has a noncatalytic
external cysteine (Cys24) as well as the active site Cys172. Our studies
employed the recombinant C24S enzyme which has catalytic properties
indistinguishable from the wild type (see ref 21). Ideal peptide substrates
mimic the 2B/2C junction of the large precursor polyprotein with glutamine
preferred at the P1 residue: the kcat of this enzyme is typically about 1.8
Soc. 1985, 107, 7105-7109. (d) Arnold, L. D.; May, R. G.; Vederas, J. C.
J. Am. Chem. Soc. 1988, 110, 2237-2241.
(26) (a) Jewell, D. A.; Swietnicki, W.; Dunn, B. M.; Malcolm, B. A.
Biochemistry 1992, 31, 7862-7869.
27) Kinetic constant were obtained as described in the following: Kitz,
R.; Wilson, I. B. J. Biol. Chem. 1962, 237, 3245-3249.
(
-
1
s
with a Km of 2.1 mM at pH 7.5; see ref 2.
804
Org. Lett., Vol. 1, No. 5, 1999

To probe the importance of the â-lactone ring for HAV
C inhibition, serine, homoserine, and γ-lactone analogues
3
were prepared (Table 1). The acyclic derivatives 3, 4, 7, and
Table 1. Serine and Homoserine Derivatives as Inhibitors of
HAV 3C Proteinase
entry
X
Y
R-confign
IC50 (µM)a
1
2
3
4
5
6
7
8
OsCO
OsCO
S
R
S
R
S
R
S
R
35
6
OH
OH
CO2Li
CO2Li
.100
.100
.100
.100
.100
.100
H2COsCO
H2COsCO
CH2OH
CH2OH
CO2Li
CO2Li
a
IC50 conditions: 0.1 µM HAV 3C, 10 µM Dabsyl-GLRTQSFS-Edans,
2
mM EDTA, 0.1 mg/mL BSA, 100 mM K3PO4 at pH 7.5, 1% DMF.
8
at a concentration of 100 µM show no significant inhibition
of HAV 3C proteinase. However, the lack of inhibition could
potentially be due to the anionic charge of the carboxylate
at pH 7.5. Hence, γ-lactones 5 and 6 were prepared and tested
against HAV 3C proteinase. γ-Lactones are well-known to
undergo ring opening as a result of nucleophilic attack at
the carbonyl, but they can react with nucleophiles at the
3
0
γ-position. However, 5 and 6 at 100 µM concentration
failed to show any inhibition of this enzyme.
The kinetic observations with 1 are consistent with a rapid
covalent inactivation of HAV 3C proteinase. To confirm this,
electrospray mass spectrometry was used to examine the
enzyme after treatment with 1. The spectra of the uncom-
plexed enzyme and the HAV 3C-1 complex are shown in
Figure 2. The mass difference between the enzyme-inhibitor
complex and the uninhibited enzyme (219 Da) is within
experimental error of the calculated mass of the inhibitor
Figure 2. Electrospray mass spectra of HAV 3C proteinase (top
spectra: A ) 23880 Da) and HAV 3C-1 enzyme inhibitor complex
(
bottom spectra: A ) 23882 Da; B ) 24101 Da) obtained on a
Micromass ZabSpec Hybrid Sector-TOF instrument.
(
221 Da). In parallel with these studies, a control dialysis
the corresponding commercially available labeled L-serine.
Model thioether and thioester compounds were generated to
assist in ascertaining whether enzymatic thiolate attack on 1
proceeds at the â-position to give a thioether or at the
experiment was done on the HAV 3C-1 complex and
uninhibited HAV 3C. After dialysis for 8 h at 4 °C, the
uninhibited enzyme retained activity, but the HAV 3C-1
complex showed no recovery of proteinase activity, thereby
suggesting that 1 may be covalently attached rather than
tightly held in a noncovalent complex.
To determine the type of adduct formed between 1 and
the HAV 3C enzyme by NMR spectroscopy, 1 was synthe-
sized with 13C (99% isotopic purity) at the â-position from
2
2,31
carbonyl of the â-lactone to form a thioester.
Since it is
known that serine or threonine residues in proteins can act
as nucleophiles to open â-lactones by attack at carbonyl to
2
3,32,33
give acylated products,
the chemical shifts of model
31
serine ester and ether derivatives were also examined. The
(
31) Model compound, (â-carbon chemical shift): N-(benzyloxycar-
13 13
bonyl)-[3- C]-L-serine â-lactone, 1(â- C), (68 ppm); N-(benzyloxycar-
bonyl)-S-methyl-L-cysteine 10, (37 ppm); N-(benzyloxycarbonyl)-L-serine
ethyl thioester 11, (64 ppm); N-(benzyloxycarbonyl)-L-serine methyl ester
12, (62 ppm); N-(benzyloxycarbonyl)-O-methyl-D-serine 13, (73 ppm). An
Inova 600 Varian NMR spectrometer was used to obtain HMQC spectra.
(32) Kim, D. H.; Ryoo, J. J. Bioorg. Med. Chem. Lett. 1995, 5, 1287-
1292.
(
30) (a) For an example of water attack at the γ-position of homoserine
lactone, see: Schwab, J. M.; Ray, T.; Ho, C.-K. J. Am. Chem. Soc. 1989,
11, 1057-1063. (b) For examples of thiol attack at the γ-position of
γ-lactones, see: Vaughan, W. R.; Baumann, J. B. J. Org. Chem. 1962, 27,
1
7
39-741 and references therein. (c) For general discussions of ring opening
of γ-lactones, see: Aldridge, D. C.; Nicholson, S.; Taylor, P. J. J. Chem.
Soc., Perkin Trans. 2 1997, 1929-1938. Huisgen, R.; Ott, H. Tetrahedron
(33) Corey, E. J.; Li, W.; Nagamitsu, T. Angew. Chem., Int. Ed. Engl.
1998, 37, 1676-1679.
1
959, 6, 253-267.
Org. Lett., Vol. 1, No. 5, 1999
805

heteronuclear multiple quantum coherence (HMQC) spectra
structure-activity relationships for â-lactone inhibition of
viral cysteine proteinases are in progress.
13
(Supporting Information) of 1(â- C) shows a cross-peak at
68 ppm for the labeled methylene carbon; upon the addition
of HAV 3C proteinase this signal disappears and a new peak
appears at 40 ppm on the carbon chemical shift axis. This
signal for the enzyme inhibitor complex, i.e., HAV 3C-1(â-
Acknowledgment. The authors thank Drs. Bruce A.
Malcolm (Schering Plough) and Michael N. James and Ernst
Bergmann (Biochemistry Department, University of Alberta)
for helpful discussions and suggestions. Drs. Bruce A.
Malcolm and Scott C. Wishart (Faculty of Pharmacy,
University of Alberta) are gratefully acknowledged for
supplying recombinant HAV 3C proteinase. These investiga-
tions were supported by the U.S. National Institutes of Health
1
3
C), clearly demonstrates formation of a thioether and thus
alkylation of the active site cysteine by attack at the â-carbon
of 1. The observed chemical shift is in good agreement with
that of the â-carbon of N-(benzyloxycarbonyl)-S-methyl-L-
cysteine (37 ppm).
In summary, HAV 3C cysteine proteinase is inactivated
by â-lactone 1 via nucleophilic ring opening of the oxetanone
ring by the cysteine thiolate at the â-carbon. In contrast, the
enantiomer 2 is not an inactivator, but rather binds com-
petitively in the active site more strongly than 1, presumably
because its â-methylene carbon is not correctly placed with
respect to the required trajectory for an incoming cysteine
thiolate nucleophile. Preliminary results (data not shown)
with HRV-14 3C proteinase, an enzyme that has similar
(
Grant AΙ 38249), the Natural Sciences and Engineering
Research Council of Canada, and the Alberta Heritage
Foundation for Medical Research (scholarship to M. S. L.).
Supporting Information Available: Experimental pro-
cedures for chemical transformations, mass spectrometry of
HAV 3C-1 inhibitor complex, HMQC spectroscopy of HAV
13
3
C-1(â- C) inhibitor complex, enzymatic analysis, and X-ray
structural information for 2. This material is available free
of charge via the Internet at http://pubs.acs.org.
26
substrate specificity to HAV 3C and is a potential thera-
peutic target for the common cold, show that â-lactones 1
and 2 have comparable potency. Additional studies on
OL990148R
806
Org. Lett., Vol. 1, No. 5, 1999