Synthesis and evaluation of thiazepines as interleukin-1β converting enzyme (ICE) inhibitors

Article abstract of DOI:10.1016/j.bmcl.2006.07.016

A series of monocyclic thiazepine inhibitors of interleukin-1β converting enzyme (ICE) were synthesized in eight steps from commercially available intermediates. In vitro biological evaluation showed the thiazepines to be moderately potent ICE inhibitors,

Full text of DOI:10.1016/j.bmcl.2006.07.016

Bioorganic & Medicinal Chemistry Letters 16 (2006) 4728–4732
Synthesis and evaluation of thiazepines as interleukin-1b converting
enzyme (ICE) inhibitorsq
Christopher D. Ellis,* Kofi A. Oppong, Michael C. Laufersweiler, Steven V. O’Neil,
David L. Soper, Yili Wang, John A. Wos, Amy N. Fancher, Wei Lu,
Maureen K. Suchanek, Richard L. Wang, Biswanath De and Thomas P. Demuth, Jr.
Procter and Gamble Pharmaceuticals Incorporated, 8700 Mason-Montgomery Road, Mason, OH 45040, USA
Received 5 May 2006; revised 28 June 2006; accepted 5 July 2006
Available online 25 July 2006
Abstract—A series of monocyclic thiazepine inhibitors of interleukin-1b converting enzyme (ICE) were synthesized in eight steps
from commercially available intermediates. In vitro biological evaluation showed the thiazepines to be moderately potent ICE inhib-
itors, with the most active compound exhibiting an IC50 value of 30 nM in an enzyme inhibition assay. Compounds of this class
possessed good selectivity against the related enzymes caspase-3 and caspase-8.
ꢀ 2006 Elsevier Ltd. All rights reserved.
The cytokine interleukin-1b (IL-1b) has been shown to
play an important role in a variety of disease processes,
S3
S2
O
including inflammation, septic shock, wound healing,
and arthritis.2 The IL-1 receptor antagonist Kineret
O
H
N
O
O
TM
HN
S4
N
has been approved for use in the treatment of rheuma-
toid arthritis (RA), indicating that modulation of IL-
1b activity provides a clinical benefit. IL-1b is produced
as an inactive pro-form 31 kDa precursor (pro-IL-1b),
which is cleaved to the biologically active 17.5 kDa cyto-
kine by the enzyme interleukin-1b converting enzyme
(ICE, caspase-1).3,4 The enzyme is a member of the cas-
pase (cysteine aspartate specific protease) family of cys-
teine proteases, and as such requires Asp116–Ala117
recognition from the substrate for activation and cleav-
age to the mature cytokine. As a result, ICE potentially
offers an attractive target for inhibition of IL-1b produc-
tion for a variety of disease indications.
H
S1
O
O
N
H
OH
HO
1
H-bonds to ICE
P3
O
P2
N
N
O
O
P4
O
N
P1
N
O
H
O
N
H
OR
2, R = Et
3, R = H
To date, 15 mammalian caspases with varying biological
functions and substrate specificities have been reported.5
Using positional scanning libraries, the tetrapeptide
fragment Ac-YVADCHO (1, Fig. 1) was identified as
a potent and selective inhibitor (Ki ꢀ 5 nM) of ICE.6
Most small molecule inhibitors of ICE are peptidomi-
Figure 1. Selected ICE Inhibitors.
metic compounds7 based on Ac-YVADCHO, which is
reported to block the release of IL-1b from human
whole blood cells with an IC50 of 4 lM. The major focus
in ICE peptidomimetics has been to restrict the Val–Ala
portion in the tetrapeptide, maintaining the important
interaction of the peptide backbone which results in
properly orienting the important P1 and P4 recognition
regions of the molecule in the enzyme pocket.3 One
Keywords: ICE; Caspase-1; Thiazepine; Peptidomimetic; Interleukin-
1b.
q
Ref. 1.
*
Corresponding author. E-mail: ellis.cd@pg.com
0960-894X/$ - see front matter ꢀ 2006 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2006.07.016

C. D. Ellis et al. / Bioorg. Med. Chem. Lett. 16 (2006) 4728–4732
4729
notable example is Pralnacasanꢁ (2), the prodrug form
of a reversible ICE inhibitor which constricts the
P2–P3 region of the molecule as a pyridazinodiazepine
and was progressed into late-stage clinical trials.8 The
free drug form of Pralnacasanꢁ is 3.
In this paper, we report the synthesis and evaluation of
thiazepines as monocyclic peptidomimetic ICE inhibi-
tors. We theorized that we could achieve the required
conformation to maintain the interaction at P1 using
an aspartate group and at P4 with aromatic groups to
mimic Tyr. The sulfur of the thiazepine could point into
the S3 pocket to occupy space similar to the Val of
Ac-YVADCHO (see Fig. 2). In addition, the monocyclic
thiazepines should be more readily synthesized than
bicyclic scaffolds such as Pralnacasanꢁ. The thiazepine
scaffold was studied in conjunction with a number of
other easily prepared scaffolds, including diazocans, pyr-
rolopyrimidinones, cycloalkyl carboxamides, and
caprolactams.9
Figure 4. Molecular model of thiazepine 12e.
AcYVAD-CHO structure. The monocyclic thiazepine
scaffold can potentially orient the P1 and P4 substituents
well for interaction with the enzyme similar to that of
Ac-YVADCHO while maintaining the desired H-bond
interactions of the peptidomimetic backbone.
Figures 3 and 4 compare the published crystal structure
of Ac-YVADCHO bound to ICE3 and a model of our
thiazepine based on the crystal structure of a previously
reported ACE inhibitor.10 Based on the single crystal
X-ray structure of the reported thiazolidinone ring, a
model of inhibitor 12e was constructed. The rotatable
bonds of the P1 and P4 moieties of the model were ori-
ented in low-energy conformations that overlapped with
the corresponding P1 and P4 substituents in the
AcYVAD-CHO structure from the crystal structure.
Finally, this model was docked into the active site of
the ICE protein crystal structure in place of the
In order to test our hypothesis, the thiazepine scaffold
was synthesized. L-Cysteine methyl ester hydrochloride
and 2-chloroethylamine hydrochloride (Scheme 1) were
reacted using literature procedures to afford the
cyclized thiazepine scaffold 4,11 which was then pro-
tected using (Boc)2O and Et3N in methanol yielding 5
in 70–80% yield. Alkylation with ethyl bromoacetate
in THF using KHMDS as the base produced 6 in
yields of 50–70%. Boc-deprotection to 711 followed by
acylation with an aryl acid chloride or aryl acid affor-
ded 8,12 with 35–70% yields over two steps. Following
saponification of 8, the resulting acid 9 was coupled
with the aspartate trap 1013 using EDCI and HOBT,
to provide compound 11 (40–70%, two steps). At this
point in the synthesis, proton NMR showed the
presence of only two compounds (due to the racemic
acetal). This provided evidence that little or no
racemization took place in the earlier KHMDS
deprotonation. Hydrolysis of the ethyl acetal gave the
final sulfide compounds 1214 in 50–80% yield. Oxida-
tion of the ring sulfur of 11 using m-CPBA afforded 13
(50–65%), which was hydrolyzed yielding the final
sulfone compounds 1415 in 35–50% yield.
P3
P2
S
O
O
P4
N
O
N
P1
O
H
O
N
H
OH
12e
Figure 2. Thiazepine ICE Inhibitor.
The thiazepine compounds showed generally good
potency in the ICE inhibition assay,16 ranging from 30
to 350 nM (see Table 1). They were generally P100-fold
selective for ICE over caspase-8 and were all inactive
against caspase-3. The analogs with fused bicyclic P4
substituents (12c–e) were more active than the monocy-
clic aromatics (12a–b). In the enzyme assay, we observed
a 4- to 5-fold decrease in the activity when moving from
the sulfides to the sulfones (for example, compare com-
pounds 12e and 14e). The compounds were also tested in
the THP-1 whole cell assay17 measuring the inhibition of
IL-1b production caused by exposure to the ICE inhib-
itor. In the THP-1 assay, most of the compounds had
IC50 values which were 5- to 15- fold higher than the
corresponding ICE enzyme IC50.
Figure 3. X-ray structure of AcYVAD-CHO complexed to ICE.

4730
C. D. Ellis et al. / Bioorg. Med. Chem. Lett. 16 (2006) 4728–4732
SH
O
S
S
S
a
b
c
Cl
O
+
NH2•HCl
OMe
d
NH
NH
N
HCl•H2N
H2N
e
BocHN
BocHN
S
OEt
O
O
O
4
5
6
S
S
f
O
N
O
O
O
O
O
N
N
H2N
N
H
N
OEt
O
OEt
OH
Ar
O
Ar
H
O
7
O
O
9
8
S
S
O
O
O
g
h
N
N
O
O
N
H
N
H
N
N
H
Ar
Ar
O
O
O
O
H
O
O
OEt
OH
O
11
12
O
O
N
H
i
OEt
O
10
S
O
O
S
O
O
O
h
O
N
O
O
N
H
N
N
H
O
Ar
N
H
O
N
H
Ar
OH
OEt
13
14
Scheme 1. Reagents and conditions: (a) Et3N, MeOH, reflux; (b) (Boc)2O, Et3N, MeOH; (c) ethyl bromoacetate, KHMDS, THF; (d) TFA CH2Cl2;
(e) ArC(O)Cl, Et3N, THF or ArCO2H, DIC, DMAP, DMF; (f) LiOH Æ H2O, THF/H2O; (g) 10, (Ph3P)4Pd, 1,3-dimethylbarbituric acid, HOBT,
EDCI, CH2Cl2; (h) TFA, CH3CN/H2O; (i) mCPBA, NaHCO3, CH2Cl2.
Table 1. In vitro data for the thiazepine series
Comp
Ar
IC50 (nM)a
Casp-3
IC50 (nM)b
THP-1
ICE
Casp-8
12a
12b
12c
12d
12e
14a
14b
14c
14d
14e
3
Ph
3-CF3Ph
350
150
>10,000
>10,000
>10,000
>10,000
>10,000
>10,000
Not tested
>10,000
>10,000
>10,000
1300
5200
>10,000
9500
1700
1000
640
2-Benzo[b]thiophene
1-Isoquinolyl
2-Naphthyl
Ph
80
70
7000
9400
730
480
30
2300
Not tested
320
>10,000
Not tested
8500
3200
3400
1700
1400
1100
190
3-CF3Ph
2-Benzo[b]thiophene
1-Isoquinolyl
2-Naphthyl
Pralnacasanꢁ18
130
150
5000
>10,000
40
3.6
a Enzyme IC50 results are expressed as 30% or less.
b Whole cell THP IC50 results vary approximately 2-fold.
In summary, a series of monocyclic thiazepines was pre-
pared as a new class of ICE inhibitors. These com-
pounds offer the advantage of a short (eight step)
synthesis while also reducing the number of chiral cen-
ters in the molecule. The thiazepines were moderately
potent ICE inhibitors possessing good selectivity against
caspase-3 and -8. This approach demonstrated that the
monocyclic scaffold maintained the key hydrogen-bond-
ing interactions of previously reported inhibitors and
presented the P1 and P4 substituents in the necessary
conformation. Although the better performing thiaze-
pines, such as compound 12e, were less active than the
clinical candidate Pralnacasanꢁ, we have shown the
potential utility of monocycles as scaffolds for effective
ICE inhibition. Alternative monocyclic scaffolds may
provide opportunities for the design of more competitive
ICE inhibitors.
References and notes
1. Partially presented at the 29th National Medicinal Chem-
istry Symposium, Madison, WI, June 2004. Ellis, C. D.;
Oppong, K. A.; Laufersweiler, M. C.; O’Neil, S. V.; Soper,
D. L.; Wang, Y.; Wos, J. A.; Fancher, A. N.; Lu, W.;
Suchanek, M. K.; Wang, R. L; De, B. Poster # 110.
2. For a review on IL-1 biology, see: Dinarello, C. A.; Wolff,
S. M. N. Eng. J. Med. 1993, 328, 106.
3. Wilson, K. P.; Black, J.; Thompson, J. A.; Kim, E. E.;
Griffith, J. P.; Navia, M. A.; Murcko, M. A.; Chambers, S.
P.; Aldape, J. P.; Raybuck, S. A.; Livingston, D. J. Nature
1994, 370, 270.
4. (a) Black, R. A.; Kronheim, S. R.; Cantrell, M.; Deeley,
M. C.; March, C. J.; Prickett, K. S.; Wignall, J.; Conlon,
P. J.; Cosman, D.; Hopp, T. P.; Mochizuki, D. Y. J. Biol.
Chem. 1988, 263, 9437; (b) Kostura, M. J.; Tocci, M. J.;
Limjuco, G.; Chin, J.; Cameron, P.; Hillman, A. G.;

C. D. Ellis et al. / Bioorg. Med. Chem. Lett. 16 (2006) 4728–4732
4731
Chartrain, N. A.; Schmidt, J. A. Proc. Natl. Acad. Sci.
USA 1989, 86, 5227.
13. Robidoux, A. L. C; Wilson, J. D.; Dieterich, P.; Storer,
N.; Leonardi, S. World Patent WO 00/10979, 2000.
5. Eckhart, L.; Ballaun, C.; Uthman, A.; Kittel, C.; Stic-
henwirth, M.; Buchberger, M.; Fischer, H.; Sipos, W.;
Tschachler, E. J. Biol. Chem. 2005, 280, 35077.
14. Preparation of N-((R)-4-(2-((S)-2-hydroxy-5-oxo-tetrahy-
drofuran-3-ylamino)-2- oxoethyl)-5-oxo-1,4-thiazepan-6-
yl)-2-naphthamide (12e).
A solution of 8e (0.20 g,
6. (a) Thornberry, N. A.; Bull, H. G.; Calaycay, J. R.;
Chapman, K. T.; Howard, A. D.; Kostura, M. J.; Miller,
D. K.; Molieaux, S. M.; Weidner, J. R.; Aunins, J.;
Elliston, K. O.; Ayala, J. M.; Casano, F. J.; Chin, J.; Ding,
G. J.-F.; Egger, L. A.; Gaffney, E. P.; Limjuco, G.; Palyha,
O. C.; Raju, S. M.; Rolando, A. M.; Salley, J. P.; Yamin,
T.-T.; Lee, T. D.; Shively, J. E.; MacCross, M.; Mumford,
R. A.; Schmidt, J. A.; Tocci, M. J. Nature 1992, 356, 786;
(b) Thornberry, N. A.; Peterson, E. P.; Zhao, J. J.;
Howard, A. D.; Griffin, P. R.; Chapman, K. T. Biochem-
istry 1992, 33, 3934.
7. For reviews of ICE inhibitors, see: (a) Talanian, R. V.;
Brady, K. D.; Cryns, V. L. J. Med. Chem. 2000, 43, 3351;
(b) Ashwell, S. Expert Opin. Ther. Patents 2001, 11, 1593.
8. (a) Dolle, R. E.; Prasad, C. V. C.; Prouty, C. P.; Salvino, J.
M.; Awad, M. M. A.; Schmidt, S. J.; Hoyer, D.; Ross, T.
M.; Graybill, T. L.; Speier, G. J.; Uhl, J.; Miller, B. E.;
Helaszek, C. T.; Ator, M. A. J. Med. Chem. 1997, 40,
1941; (b) Randall, J. C. R. Abstracts of Papers, 228th ACS
National Meeting, Philadelphia, PA, August 2004, MEDI-
200.
9. (a) Oppong, K. A.; Ellis, C. D.; Laufersweiler, M. C.;
O’Neil, S. V.; Wang, Y.; Soper, D. L.; Baize, M. W.; Wos,
J. A.; De, B.; Bosch, G. K.; Fancher, A. N.; Lu, W.;
Suchanek, M. K.; Wang, R. L.; Demuth, T. P., Jr. Bioorg.
Med. Chem. Lett. 2005, 15, 4291; (b) Laufersweiler, M. C.;
Wang, Y.; Soper, D. L.; Suchanek, M. K.; Fancher, A. N.;
Wang, R. L.; Oppong, K. A.; Ellis, C. D.; Baize, M. W.;
O’Neil, S. V.; Wos, J. A.; Demuth, T. P., Jr. Bioorg. Med.
Chem. Lett. 2005, 15, 4322; (c) Soper, D. L.; Sheville, J.;
O’Neil, S. V.; Wang, Y.; Laufersweiler, M. C.; Oppong, K.
A.; Wos, J. A.; Ellis, C. D.; Fancher, A. N.; Lu, W.;
Suchanek, M. K.; Wang, R. L.; De, B.; Demuth, T. P., Jr.
Bioorg. Med. Chem. Lett. 2006, 16, 4233; (d) Wang, Y.;
O’Neil, S. V.; Wos, J. A.; Oppong, K. A.; Laufersweiler,
M. C.; Soper, D. L.; Ellis, C. D.; Baize, M. W.; Bosch, G.
K.; Fancher, A. N.; Lu, W.; Suchanek, M. K.; Wang, R.
L.; Schwecke, W. P.; Cruze, C. A.; Buchalova, M.; Belkin,
M.; De, B.; Demuth, T. P., Jr., 2006, manuscript in
preparation.
10. Yanagisawa, H.; Ishihara, S.; Ando, A.; Kanazaki, T.;
Miyamoto, S.; Kike, H.; Iijima, Y.; Oizumi, K.; Matsush-
ita, Y.; Hata, T. J. Med. Chem. 1987, 30, 1984.
11. (a) Ahmed, S. A.; Esaki, N.; Tanaka, H.; Soda, K. FEBS
Lett. 1984, 174, 76; (b) Das, J.; Robl, J. A.; Reid, J. A.; Sun,
C.-Q.; Misra, R. N.; Brown, B. R.; Ryono, D. E.; Asaad, M.
M.; Bird, J. E.; Trippodo, N. C.; Petrillo, E. W.; Karanew-
sky, D. S. Bioorg. Med. Chem. Lett. 1994, 4, 2193.
12. Preparation of (R)-ethyl 2-(6-(2-naphthamido)-5-oxo-1,4-
thiazepan-4-yl)acetate (8e). To a solution of 7 (0.23 g,
0.99 mmol) in 5 ml THF at 0 ꢂC, Et3N (0.21 ml, 1.5 mmol)
was added, followed by 2-naphthoyl chloride (0.21 g,
1.1 mmol). The reaction mixture was allowed to warm to
rt and stirred for 1 h. EtOH was added to quench the
remaining acid chloride and the mixture evaporated in
vacuo. The crude mixture was purified by flash chroma-
tography (1:1 EtOAc:hexanes) to yield 0.20 g (52%) of 8e
as a white solid. 1H NMR (CDCl3): d 8.39 (s, 1H), 7.98 (m,
1H), 7.96 (d, J = 2.4 Hz, 1H), 7.94 (d, J = 1.2 Hz, 1H),
7.90 (m, 1H), 7.59 (m, 2H), 5.39 (m, 1H), 4.37 (d,
J = 17.7 Hz, 1H), 4.22 (m, 2H), 4.17 (d, J = 17.1 Hz, 1H),
3.75 (dd, J = 4.8, 17.4 Hz, 1H), 3.01 (m, 2H), 2.66 (dd,
J = 5.4, 14.4 Hz, 1H), 1.56 (s, 2H), 1.35 (t, J = 7.2 Hz, 3H).
ESI-MS 387.1 (M + H)+.
0.51 mmol) in THF (6 mL) and water (2 mL) was treated
with LiOH Æ H2O (43 mg, 1.0 mmol) and stirred for 1 h at
rt. The solution was acidified (pH3) with 1 N HCl and
extracted with Et2O, washed with saturated brine, dried
(MgSO4), and concentrated. The crude acid (9e) was
isolated as
a white solid, 0.17 g (91%), MS 359.1
(M + H)+. Catalytic Pd(PPh3)4 was added to a solution
of 10 (0.32 g, 1.4 mmol) and 1,3-dimethylbarbituric acid
(0.41 g, 2.6 mmol) in 5 mL CH2Cl2 at rt. The solution
was stirred at rt for 15 min and then the crude carboxylic
acid (9e) prepared above was added as a solution in 5 mL
CH2Cl2 and 0.5 mL DMF followed by HOBt (0.19 g,
1.4 mmol) and EDCI (0.26 g, 1.3 mmol). The solution
was stirred for 2 h, washed with saturated NaHCO3 and
brine, dried (MgSO4), and concentrated. After purifica-
tion by preparative reverse-phase HPLC, 11e was recov-
ered as a pale yellow solid (0.13 g, 60% yield), MS 486.2
(M + H)+. 11e (45 mg, 0.093 mmol) was hydrolyzed with
TFA in CH3CN/H2O. Purification by preparative reverse
phase HPLC yielded 12e as a white solid, 19 mg (45%).
1H NMR (CD3OD): d 8.47 (s, 1H), 8.01 (q, 1H), 7.97 (m,
3H), 7.61 (m, 2H), 5.42 (d, J = 8.4 Hz, 1H), 4.64 (m, 1H),
4.36 (m, 1H), 4.20 (m, 3H), 3.67 (d, J = 15.3 Hz, 1H),
2.99 (m, 3H), 2.70 (m, 2H), 2.53 (m, 1H). ESI-MS 458.1
(M + H)+.
15. Preparation of N-((R)-4-(2-((S)-2-hydroxy-5-oxo-tetra-
hydrofuran-3-ylamino)-2-
oxoethyl)-1,1-dioxido-5-oxo-
1,4-thiazepan-6-yl)-2-naphthamide (14e). To a solution
of 11e (83 mg, 0.17 mmol) in 5 mL CH2Cl2, NaHCO3
(40 mg, 0.48 mmol), and mCPBA (86 mg, 0.35 mmol)
were added. After stirring at rt for 2 h, the solids
were filtered off and the solution concentrated. After
purification by preparative reverse-phase HPLC, 13e
was recovered as an off-white solid (85 mg, 96%
yield), MS 518.2 (M + H)+. 13e was hydrolyzed with
TFA in CH3CN/H2O. Purification by preparative
reverse phase HPLC yielded 14e as a white solid,
33 mg (41%). 1H NMR (CD3OD): d 8.45 (s, 1H),
7.95 (m, 1H), 7.93 (s, 3H), 7.59 (m, 2H), 5.55 (d,
J = 10.2 Hz, 1H), 4.65 (dd, J = 3.9, 9.0 Hz, 1H), 4.52
(dd, J = 5.1, 16.2 Hz, 1H), 4.39 (m, 2H), 4.08 (dd,
J = 3.0, 16.5 Hz, 1H), 3.79 (m, 3H), 3.50 (dd, J = 2.4,
13.8 Hz, 1H), 3.28 (m, 1H), 2.71 (m, 1H), 2.53 (m,
1H). ESI-MS 490.1 (M + H)+.
16. The isolated enzyme (ICE, caspase-3, and caspase-8)
assays were performed in a 96-well format using fluor-
ogenic substrates, enzymes, and control peptide inhibi-
tors purchased from BioMol Research Laboratories
(Plymouth Meeting, PA). The assay was conducted
according to the manufacturer’s instructions. Enzyme
inhibition was monitored over 30 min at 37 ꢂC by
measuring fluorescence using a BMG Fluostar plate
reader (excitation filter 390 nm, emission filter 460 nm).
IC50 values were calculated based on the equation
IC50 = [I]/(Vo/Vi) À 1, where Vi is the initial velocity of
substrate cleavage in the presence of inhibitor at
concentration [I], and Vo is the initial velocity in the
absence of inhibitor.
17. A suspension of human monocytic cells (THP-1, ATCC
strain TIB202, 2 · 106/ml in RPMI 1640 medium from
Gibco BRL) was plated in 96-well plates, incubated with
or without compounds (administered as solutions in
DMSO, such that test concentrations ranged from 1 nM
to 10 lM) for 15 min, and then stimulated with LPS

4732
C. D. Ellis et al. / Bioorg. Med. Chem. Lett. 16 (2006) 4728–4732
(1 lg/ml) for 4 h. Cells were centrifuged and the condi-
18. The values for the free drug form of Pralnacasanꢁ
(compound 3b) in our assays were comparable to those
reported in similar bioassays; see Rudolphi, K.; Gerwin,
N.; Verzijl, N.; van der Kraan, P.; van den Berg, W.
OsteoArthritis Cartilage 2003, 11, 738.
tioned media were collected to quantify the release of
IL-1b by an ELISA measurement according to the
manufacturer’s instructions (R&D Systems, catalog num-
ber DLB50) or stored at À20 ꢂC for future use.