Inhibition of interleukin-1β converting enzyme (ICE or caspase 1) by aspartyl acyloxyalkyl ketones and aspartyl amidooxyalkyl ketones

Article abstract of DOI:10.1016/j.bmcl.2010.07.031

A series of acyloxyalkyl and amidooxyalkyl ketones appended to a carbobenzyloxy aspartic acid core have been prepared. The most potent of these new inhibitors was 4i with a K_i of 0.5 μM. These two series provide an improved understanding of the binding requirements for the hydrophobic prime side of ICE.

Full text of DOI:10.1016/j.bmcl.2010.07.031

Bioorganic & Medicinal Chemistry Letters 20 (2010) 5089–5094
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
Inhibition of interleukin-1b converting enzyme (ICE or caspase 1) by aspartyl
acyloxyalkyl ketones and aspartyl amidooxyalkyl ketones
a
a
a
a
a
Paul Galatsis a, , Bradley Caprathe , Dennis Downing , John Gilmore , William Harter , Sheryl Hays ,
Catherine Kostlan a, Kristin Linn a, Elizabeth Lunney a, Kim Para a, Anthony Thomas a, Joseph Warmus a,
Hamish Allen b, Kenneth Brady b, Robert Talanian b, Nigel Walker b
*
a Pfizer Global Research and Development, 2800 Plymouth Road, Ann Arbor, MI 48105, United States
b BASF Bioresearch Corporation, 100 Research Drive, Worcester, MA 01605, United States
a r t i c l e i n f o
a b s t r a c t
Article history:
Received 26 March 2010
Revised 6 July 2010
Accepted 8 July 2010
Available online 11 July 2010
A series of acyloxyalkyl and amidooxyalkyl ketones appended to a carbobenzyloxy aspartic acid core have
been prepared. The most potent of these new inhibitors was 4i with a Ki of 0.5 lM. These two series pro-
vide an improved understanding of the binding requirements for the hydrophobic prime side of ICE.
Ó 2010 Elsevier Ltd. All rights reserved.
Keywords:
ICE inhibitor
Prime side
Hydrophobic pocket
Interleukin IL-1b (IL-1b) is a pro-inflammatory cytokine that
has been implicated in a number of pathophysiological states.1
Thus inhibitors of ICE, the key enzyme that converts the inactive
pro-IL-1b to the biologically active form, would be of pharmacolog-
ical value.2 Structurally, ICE inhibitors can be dissected into three
components: a core fragment, a P4-P2 peptidomimetic head-frag-
ment, and a prime side tail-fragment (Fig. 1). SAR campaigns but-
tressed by X-ray crystallography3 have revealed a key requirement
for Asp as the core fragment.4 While multiple scaffolds have been
disclosed for the peptidomimetic fragment, the prime side tail
has received limited attention. Most of these moieties have been
functionally designed as leaving groups, thus not taking full advan-
tage of all the available protein interactions.
The simplest prime side group is the corresponding Asp-CHO,
typically found in its masked acylal form, as exemplified by the
clinical candidates pralnacasan5 and VX-7656 (Fig. 2). Other varia-
tions found for the prime side moiety include halomethyl Asp ke-
tones (AcYVAD-CMK or MX-11227), halophenyl derivatives (Win-
676948 or IDN-75689), or heterocyclic derivatives (PTP10).
for intramolecular H-bonding effects that were able to rigidify the
backbone. Our analog 2 exhibited comparable potency (Ki = 4 nM)
to 1. These appear to be the only reports directed at fully accessing
the prime side binding pocket. From analysis of the X-ray crystal
structure of ICE and consistent with the observed SAR, it is evident
that the prime side pocket is hydrophobic in nature, since it is sur-
rounded by hydrophobic amino acid residues (Ile176, Pro177, and
Ile239).
CO2H
H
N
O
H
N
O
N
H
N
H
O
O
O
1
HO
CO2H
N
CO2H
O
H
N
O
H
N
O
O
O
S
N
H
N
H
O
H
O
The first reported study of ligands projecting into the prime side
looked at the effect of methylene spacers on the binding of a phe-
nyl group.11 The authors identified the peptidomimetic AcY-
VAD(CH2)5Ph, 1, as a potent ICE inhibitor (Ki = 18.5 nM). Later,
our group12 reported sulfonamide derivatives having the potential
2
HO
While providing valuable information related to the prime side
interactions, these reports left a great deal of questions unan-
swered. We wished to fill this knowledge gap in order to better de-
sign ICE inhibitors. To rapidly assess prime side SAR, a strategy of
optimizing this fragment on a truncated P side moiety was envi-
sioned. This was accomplished by utilizing a carbobenzyloxy aspar-
* Corresponding author. Address: MS 8220-4225, Pfizer Global R&D, Eastern
Point Road, Groton, CT 06340, United States. Tel.: +1 860 686 9379.
E-mail address: paul.galatsis@pfizer.com (P. Galatsis).
0960-894X/$ - see front matter Ó 2010 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2010.07.031

5090
P. Galatsis et al. / Bioorg. Med. Chem. Lett. 20 (2010) 5089–5094
Table 1
CO2H
prime side
Comparison of binding affinity for select compoundsa
P4 P2
N
H
Entry
Compound
CO2H
Ki (lM)
O
core (Asp)
Cbz
Cbz
Cbz
Cbz
Cbz
Cbz
Cbz
Cbz
5
119
N
H
O
Figure 1. General structural features of an ICE inhibitor.
O
CO2H
O
O
O
4a
4b
4c
4d
4e
4f
8.3
H2N
N
H
O
O
O
N
N
O
Cl
O
H
N
CO2H
O
N
O
O
N
H
O
4.5
N
OEt
O
O
N
H
O
O
N
H
N
H
O
OEt
VX-765
CO2H
pralnacasan
O
Cl
Cl
CO2H
16
H
N
O
N
H
O
O
O
MX-1122
N
H
F
CO2H
O
O
O
O
O
O
O
S
1.7
1.1
5.6
6.0
CO2H
N
H
O
H
O
O
Cl
O
Cbz
N
Win-67694
N
N
O
CO2H
H
O
H
F
O
Cl
CO2H
F
F
CF3
N
H
O
CO2H
F
O
O
H
N
O
N
N
R1
CO2H
N
H
O
N
H
N
H
O
H
N
O
Cl
O
O
R2
N
H
O
PTP
O
IDN-7568
CO2H
Figure 2. Representative ICE inhibitors illustrating the typical structural motifs
found on the prime side.
4g
N
H
O
O
tic acid core to construct a series of acyloxyalkyl and amidooxyalkyl
ketones. The advantage of using this single amino acid substrate
was that it increased the ease of synthesis and simplified the anal-
ysis of the final products since, for example, di- and tripeptide acy-
loxyalkyl ketones can lead to bimodal inhibitors.13 This strategic
approach was completed by appending the optimized prime side
fragment to the full length P side to afford ICE inhibitors with im-
proved potency.
Synthesis of the aspartyl acyloxyalkyl ketones 4 followed the
protocol outlined in Scheme 1.14 The known15 bromoketone 3
could be converted into the desired esters 4 by bromide displace-
ment with the corresponding carboxylic acid followed by depro-
tection of the aspartyl b-acid with TFA. Esters 4 were tested as
inhibitors of ICE using a recombinant form of the enzyme and
monitored by fluorescence (Table 1).13 Ketone 5 was synthesized,
based on our strategic design, as a reference standard to be analo-
gous with 1.
With compounds 4 in-hand, it was quickly observed that simple
acyloxyalkyl ketones, such as 4a, demonstrated improved activity
(14-fold) compared to 5. Phenylalkyl derivatives 4b and 4c probe
the depth of the prime side pocket and showed optimal binding
with two methylene spacers. The tolerability of heteroatoms in
the linker was examined with compounds 4d–f. While a nitrogen
atom (4f) is tolerated, oxygen (4d) and sulfur (4e) show a modest
CO2H
O
O
O
4h
4i
1.3
0.5
Cbz
Cbz
Cbz
N
H
O
O
CO2H
N
H
O
O
CO2H
O
N
H
O
4j
1.0
1.0
O
CO2H
O
Cbz
4k
N
H
O
O
CO2H
O
O
O
O
Cbz
Cbz
Cbz
Cbz
4l
4m
4n
4.4
1.3
2.0
8.5
N
H
O
O
CO2H
N
H
O
O
CO2H
N
H
O
O
CO2H
CO2H
1. RCO2H
KF
O
CO2tBu
Cbz
Cbz
N
H
O
R
2. CF3CO2H
N
H
Br
O
O
4o
N
H
O
4
O
3
Scheme 1. Preparation of aspartyl acyloxyalkyl ketones.

P. Galatsis et al. / Bioorg. Med. Chem. Lett. 20 (2010) 5089–5094
5091
Table 1 (continued)
Entry
Compound
Ki (lM)
CO2H
O
4p
4q
0.8
1.1
Cbz
N
H
O
O
CO2H
N
O
O
O
O
Cbz
Cbz
Cbz
Cbz
N
H
O
O
CO2H
NH
Figure 3. X-ray structure of 1 (orange) and 4i (green) in ICE active site.
4r
4s
4t
0.6
0.8
1.2
N
H
O
O
specifically, the naphthyl group binds in the hydrophobic region
formed by the Ile176, Pro177, and Ile239 side chains. The naphthyl
group is oriented with its second phenyl ring binding edge-wise to-
wards Gly238 which is located in a beta strand.
CO2H
O
S
N
H
O
O
Interestingly, the potent diphenyl derivative 4p docked with its
ester group oriented essentially 90° to the ester group of 4i (Fig. 4).
One prime side phenyl ring of 4p intersects almost perpendicularly
to the second ring of the naphthyl group in 4i, but still formed
hydrophobic contacts with Ile176 and Ile239. The second phenyl
group in 4p orients proximal to the Asp288 loop in the C-terminus
region of the p20 domain. Since this second phenyl group resulted
in more than a sevenfold enhancement in potency, relative to 4b,
this intriguing binding pose highlights an additional binding sub-
site that could be targeted by prime side groups.
CO2H
N
H
O
O
CO2H
O
Cbz
4u
2.6
N
H
O
S
O
N
a
For a description of the assay see Ref. 13. Ki values are the geometric mean of
two or more experiments.
Complementary to these studies and as part of our ICE inhibitor
project, a diverse combinatorial library was generated to fully ex-
plore the chemical space associated with the prime side pocket
of the protease active site. From this array, hydroxamic acid deriv-
ative 6 was determined to be an active compound. This hit was se-
lected for further optimization to probe the nature of the
hydrophobic pocket and explore the effect of the hydroxamic acid
functional group on ICE inhibition. Could the hydroxamate func-
tional group be isosteric with the ester group in this series of ICE
inhibitors?
2.6- and 4-fold, respectively, improvement in binding relative to
4b. A 2-naphthyl moiety (4g) modification of 4b shows the addi-
tional phenyl was tolerated. However, a 4.6-fold improvement in
binding was observed by shortening the chain by a single methy-
lene group (4h). Compound 4i maintains the shorter linker but
now changes the orientation of the naphthyl moiety by linking
through the 1-position. This modification provided a 12-fold in-
crease in potency relative to 4g. Inserting an oxygen atom into
the linker, as in 4j, was tolerated with only a slight twofold de-
crease in potency relative to 4i. In an attempt to restrict the confor-
mational space the naphthyl fragment could occupy, the
methylene spacer in 4i was methylated to afford 4k, which was
equipotent with 4j. While multiple compounds were prepared,
compounds 4l–o summarizes the naphthyl moiety substituent ef-
fects. Substitution at the 4-position (4m) and 5-position (4n) were
better tolerated than at the 2-position (4l) or 8-position (4o). Add-
ing a phenyl substituent to the linker chain of 4b generated 4p that
proved to be equipotent to 4i. Compound 4q is the result of addi-
tion of a pyridyl fragment to linker of 4b. This resulted in a slight
erosion of potency compared to 4i. Compounds 4r–u provided
additional examples that showed the prime side could tolerate
additional polarity. The indole moiety of 4r proved to be the most
potent, followed by benzofuran 4s, benzothiophene 4t, and benzo-
thiazole 4u.
O
CO2H
O
H
N
O
N
H
O
H
O
6
CO2Me
In order to determine which attributes of this compound imparted
it with such potency, an X-ray crystal structure of 6 bound in the
active site of the enzyme was determined (Fig. 5). Analysis of the
These results strongly suggest that the principal feature needed
for optimal activity of the prime side ester is the juxtaposition of
the aryl group within the ICE prime side pocket. In order to study
this prime side interaction more fully, we obtained X-ray crystal
structures of 4i. This juxtaposition is displayed in Figure 3, which
shows the binding pose of 1 relative to ester 4i. The Asp b-carboxy
binds into the oxy-anion hole. The Cbz group of 4i reaches into the
S2 pocket, whereas 1 extends to fully interact with the P side. On
the prime side, the phenyl group of 1 and the naphthyl ring of 4i
occupy very similar positions in the hydrophobic pocket. This
was accomplished by the linker taking similar trajectories. More
Figure 4. Docked overlay of 4i (white) and 4p (orange).

5092
P. Galatsis et al. / Bioorg. Med. Chem. Lett. 20 (2010) 5089–5094
Table 2
ICE enzyme inhibition dataa
Entry
Structure
Ki (lM)
CO2H
Cbz
Cbz
5
119
N
H
O
O
CO2H
H
N
N
H
O
6
15.0
H
O
CO2Me
CO2H
Cbz
Cbz
Cbz
Cbz
N
7a
7b
7c
7d
87.2
N
H
O
O
O
O
CO2H
H
H
N
9.20
N
H
O
H
O
O
CO2H
N
17.5
Figure 5. X-ray binding pose for 6.
N
H
O
H
O
O
CO2H
structural data indicated that 6 made interactions, with respect to
the P1 position, as seen with other inhibitors. That is, the b-carboxyl
Asp group was bound to the oxy-anion hole formed by the Arg341,
Arg179, and Gln283 residues. Additional interactions observed were
the carbonyl of the Asp moiety H-bonding with His237 and the
amide nitrogen H-bonding to the backbone Ser339 carbonyl. A
unique feature of 6, captured by this structure, is the additional
H-bond with Gly238 residue from the carbonyl of the hydroxamic
acid moiety. The crystal structure also revealed that the carbome-
thoxy of the hydroxamic acid moiety was able to reach and presum-
ably make contacts in the S2 pocket, along with the Cbz protecting
group (Fig. 5).
N
6.77
N
H
O
O
O
CO2H
7e
7f
Cbz
Cbz
Cbz
Cbz
Cbz
Cbz
N
14.0
21.0
66.5
27.5
84.0
N
H
O
O
CO2H
H
N
N
H
O
O
O
O
O
O
O
CO2H
H
N
7g
7h
7i
Having made a preliminary attempt to rationalize which struc-
tural features of the hydroxamic acid moiety were utilized for its
binding, the next step was to vary these groups to gain greater in-
sight into these interactions. Hydroxamates 7 implement the same
strategy used to examine the ester SAR. These inhibitors main-
tained the truncated core fragment and the prime side moiety
was optimized for potency.
N
H
O
O
CO2H
H
N
N
H
O
O
CO2H
H
N
The hydroxamates 7 could be synthesized14 in a straightfor-
ward manner (Scheme 2) from the readily available aspartic acid
N
H
O
O
derivative 315 using
a variation of the chemistry that was
CO2H
H
N
employed in the preparation of the esters 4. Displacement of the
bromide now by appropriately substituted hydroxamic acids, fol-
lowed by deprotection with TFA afforded the desired target com-
pounds 7.
The series of compounds 7 were assayed for ICE inhibitory
activity and the results are summarized in Table 2. In addition to
6, compound 5 reprised its role as a reference compound as a way
to calibrate our strategic approach and relate these compounds with
esters 4. Thus, compound 6 was found to be eightfold more potent
7j
298
N
H
O
O
O
CO2H
O
7k
1.50
Cbz
N
N
H
O
O
O
a
For a full description of the assays see Ref. 8. Values are the geometric mean of
two or more experiments.
1. KF
R2
than 5. Compound 7a contains a minimally functionalized hydroxa-
mate fragment which proved to be 5.8-fold less potent than 6 but
equipotent with 5. Thus, the hydroxamate functional group was
making more productive interactions when compared to 5 and
different from those with 4. Compounds 7b and 7c provide some in-
sight into the substituents required for improved binding interac-
tions. These two compounds lack the pendant carbomethoxy
group found in 6 but exhibited approximately equivalent potency.
N
R1
CO2H
R2
N
CO2tBu
HO
R1
O
Cbz
Cbz
N
H
O
N
H
Br
O
2. TFA
O
O
7
3
Scheme 2. Preparation of aspartyl amidooxyalkyl ketones.

P. Galatsis et al. / Bioorg. Med. Chem. Lett. 20 (2010) 5089–5094
5093
CO2tBu
and 7e as the six-membered ring was made aromatic. Compounds
7f–j were designed to probe the hydrophobic characteristics of the
prime side region. Implementation of these moieties on the ester
variation proved to be a very successful approach for achieving
greater potency but thisSAR did not translateto the hydroxamicacid
variation. Indeed, on the contrary, potency was negatively impacted.
Interestingly, improvements to potency (80-fold relative to 1) were
recovered with the oxa[2.2.1]bicyclo 7k. Presumably, the convex
nature of this group fills the hydrophobic pocket more effectively.
Having identified prime side fragments that greatly increased
binding efficiency relative to the reference compound 5, the time
to close the loop on the validation of our strategy arrived. Several
of the optimized prime side groups with demonstrated potency
improvements were combined with peptide P side fragments to
generate fully derivatized ICE inhibitors that were then evaluated
for overall improvement of binding efficacy. The commercially
available, orthogonally protected aspartic acid derivative 8 was
transformed using robust chemistry into the desired prime side
analogs 11 and 12 (Scheme 3).14 Elongation of 8 using standard
peptide chemistry afforded the Cbz-capped tripeptide acid 9 which
was converted to the corresponding bromoketone 10. Deprotection
CO2tBu
1. Cbz-VA-OH
~HCl
H2N
HOBt, EDCI, NMM
Cbz-VA
N
H
CO2H
CO2Me
2. NaOH, EtOH, H2O
9
8
1. R1CO2H, KF
or
R1CONR2OH, KF
CO2tBu
1. IBCF, NMM, THF
2. CH2N2
3. HBr, HOAc
Cbz-VA
N
H
Br
O
2. CF3CO2H
10
CO2H
CO2H
O
R2
N
Cbz-VA
or
Cbz-VA
R1
N
H
O
R1
N
H
O
O
O
O
11
12
Scheme 3. Preparation of tripeptide acyloxyalkyl ketones.
Clearly, the carbomethoxy group was not contributing in a produc-
tive manner, to the binding interactions, thus giving rise to the
observed results. No increases in potency were observed with 7d
Table 3
Comparison of in vitroa and whole cellb inhibitory potency
Entry
Structure
Ki (l
M)
IC50
(
lM)
PBMC (lM)
CO2H
O
H
N
O
H
N
O
N
H
N
H
O
O
1
0.011
0.078
0.9
HO
CO2H
H
N
O
O
Cbz
N
13
0.134
0.976
0.002
4.9
0.6
N
H
H
O
O
O
CO2H
H
N
O
11a
0.003c
Cbz
N
N
H
O
H
O
CO2H
H
N
O
O
O
O
11b
11c
0.001c
0.002c
0.005
0.003
4.3
1.4
Cbz
N
N
H
O
H
O
O
O
CO2H
H
N
Cbz
N
N
H
O
H
O
O
CO2H
H
N
O
H
Cbz
N
N
N
H
O
12a
0.007
0.026
4.8
H
H
O
O
CO2Me
O
CO2H
H
N
O
O
Cbz
N
N
12b
12c
0.002
0.001
0.003
0.003
3.0
2.4
N
H
O
H
O
O
O
O
CO2H
H
N
H
Cbz
N
N
N
H
O
H
H
O
a
b
c
For a description of the assay see Ref. 13. Ki values are the geometric mean of two or more experiments.
For a description of the assay see Ref. 16.
Mechanistic change to irreversible inhibition (bimodal kinetics).

5094
P. Galatsis et al. / Bioorg. Med. Chem. Lett. 20 (2010) 5089–5094
of the Asp b-carboxylic acid followed bromide displacement with
the appropriate carboxylic or hydroxamic acid (vide supra) to gen-
erate the desired targets 11 and 12.
Supplementary data
Supplementary data (coordinate files (pdb) containing the ac-
tive site residues with the bound ligand for Figs. 3 and 5) associ-
ated with this article can be found, in the online version, at
doi:10.1016/j.bmcl.2010.07.031.
With these compounds in-hand, their potency for ICE inhibi-
tion was determined and the data is presented in Table 3. Com-
pound 13 shows that the P4 substituent of 1 is more efficient
at binding compared to the Cbz group present in 12. The ester
analogs 11 and hydroxamates 12 are significantly more potent
than the reference compound 13. The analysis for these inhibitors
was complicated by the observation that these more potent pep-
tide derivatives of esters 11 could now serve as leaving groups.
This mechanistic switch is not unknown, since di- and tripeptide
acyloxyalkyl ketones can lead to bimodal inhibitors as previously
described.13
Comparison of these compounds in a functional assay using hu-
man peripheral blood mononuclear cells (PBMC) shows a much
different result. In this functional assay, esters 11b and 11c are
approximately equipotent with reference compound 13. In con-
trast 11a is about 10-fold more potent in the PBMC assay than
the other compounds. This poor correlation between enzymatic
inhibition and cell-based activity has been noted previously4a
and is problematic for the prediction and discovery of an ICE inhib-
itor with therapeutic potential. Since ICE is an intracellular en-
zyme, the loss of activity in going from the enzyme assay to the
cell-based assay may, in part, be a result of transport problems
across the cellular membrane. This explanation, however, seems
far too simplistic when structurally similar compounds such as
11a and 11b show divergent functional activity. A satisfactory
explanation of this poor correlation awaits future study.
References and notes
1. (a) Thornberry, N. A.; Bull, H. G.; Calaycay, J. R.; Chapman, K. T.; Howard, A. D.;
Kostura, M. J.; Miller, D. K.; Molineaux, S. M.; Weidner, J. R.; Aunin, 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, 768; (b) Thornberry, N. A. Br. Med. Bull. 1996, 53, 478; (c) Bird,
S.; Zou, J.; Wang, T.; Munday, B.; Cunningham, C.; Secombers, C. J. Cytokine
Growth Factor Rev. 2002, 13, 483; (d) Joosten, L. A. B.; van den Berg, W. B. Drug
Discov. Today 2006, 3, 169.
2. (a) Ator, M. A.; Dolle, R. E. Curr. Pharm. Des. 1995, 1, 191; (b) Giegel, D. A.; Kostlan,
C. R. Annu. Rep. Med. Chem. 1998, 33, 183; (c) Talanian, R. V.; Brady, K. D.; Cryns, V.
L. J. Med. Chem. 2000, 43, 3351; (d) Ashwell, S. Expert Opin. Ther. Patents 2001, 11,
1593; (e) Randle, J. C. R.; Harding, M. W. H.; Ku, G.; Schonharting, M.; Kurrle, R.
Expert Opin. Investig. Drugs 2001, 10, 1207; (f) Braddock, M.; Quinn, A. Nat. Rev.
Drug Discov. 2004, 3, 1; (g) Dinarello, C. A. Curr. Opin. Pharmacol. 2004, 4, 378; (h)
Le, G. T. E.; Abbenante, G. Curr. Med. Chem. 2005, 12, 2963; (i) Fischer, U.; Schulze-
Osthoff, K. Pharmacol. Rev. 2005, 57, 187.
3. (a) Wilson, K. P.; Black, J. F.; Thomson, 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; (b) Walker, N. P. C.; Talanian, R. V.; Brady, K. D.; Dang, L.
C.; Bump, N. J.; Ferenz, C. R.; Franklin, S.; Ghayur, T.; Hackett, M. C.; Hammill, L.
D.; Herzog, L.; Hugunin, M.; Houy, W.; Mankovick, J. A.; McGuiness, L.;
Orlewicz, E.; Paskind, M.; Pratt, C. A.; Reis, P.; Summani, A.; Terranova, M.;
Welch, J. P.; Moller, A.; Tracey, D. E.; Kamen, R.; Wong, W. W. Cell 1994, 78, 343.
4. (a) Thornberry, N. A.; Molineaux, S. M. Protein Sci. 1994, 4, 3; (b) Rano, T. A.;
Timkey, T.; Peterson, E. P.; Rotonda, J.; Nicholson, D. W.; Becker, J. W.;
Champman, K. T.; Thornberry, N. A. Chem. Biol. 1997, 4, 149.
Having examined the inhibitory activity of the esters 11, our
focus was directed towards the hydroxamic acid terminated pepti-
domimetics 12 (Table 3). Compound 12a, based on the original
library hit 6, was found to be almost 40-fold more potent, by
IC50, than reference compound 13. Of the three hydroxamates pre-
sented in Table 3, 12a proved to be the weakest. Compounds 12b
and 12c were found to be about eightfold more potent. Addition-
ally, hydroxamates 12 stand in contrast to esters 11. The former
did not display the change in mechanism, from reversible to irre-
versible enzyme inhibition, observed with the later despite them
both having comparable potency.
In summary, we have designed and synthesized a series of ICE
inhibitors with a variety of esters and hydroxamates that interact
with the prime side amino acid residues of the enzyme. These
inhibitors were designed and optimized using a protected aspartic
acid core and then the most potent groups were appended to a lar-
ger tripeptide producing low nanomolar ICE inhibitors. The results
demonstrate that, while potent inhibitors can be generated, one
must be cognizant of the nature of the binding interaction. This
is important to ensure the mechanism of inhibition does not
change. In sharp contrast to the ester variation, the hydroxamates
appear to have a narrower SAR trend. These two series have re-
sulted in an improved understanding of the binding requirements
of the lipophilic pocket on the prime side of the enzyme and this
information can now be utilized in the design of future ICE
inhibitors.
5. (a) Rudolphi, K.; Gerwin, N.; Verzijl, N.; van der Kraan, P.; van den Berg, W.
Osteoarthritis Cartilage 2003, 11, 738; (b) Bauer, C.; Loher, F.; Dauer, M.; Mayer,
C.; Lehr, H. A.; Schonharting, M.; Hallwachs, R.; Endres, S.; Eigler, A. Dig. Dis. Sci.
2007, 52, 1642.
6. Wannamaker, W.; Davies, R.; Namchuk, M.; Pollard, J.; Ford, P.; Ku, G.; Decker,
C.; Charifson, P.; Weber, P.; Germann, U. A.; Kuida, K.; Randle, J. C. R. J.
Pharmacol. Exp. Ther. 2007, 321, 509.
7. Cai, S. X.; Guan, L.; Jia, S.; Wang, Y.; Yang, W.; Tseng, B.; Drewe, J. Bioorg. Med.
Chem. Lett. 2004, 14, 5295.
8. Miller, B. E.; Krasney, P. A.; Gauvin, D. M.; Holbrook, K. B.; Koonz, D. J.;
Abruzzese, R. V.; Miller, R. E.; Pagani, K. A.; Dolle, R. E.; Ator, M. A.; Gilman, S. C.
J. Immunol. 1995, 154, 1331.
9. Ueno, H.; Kawai, M.; Shimokawa, H.; Hirota, M.; Ohmi, M.; Sudo, R.; Ohta, A.;
Arano, Y.; Hattori, K.; Ohmi, T.; Kato, N.; Kojima, M.; Ueno, Y.; Yamamoto, M.;
Moriguchi, Y.; Eda, H.; Masubuchi, K. Bioorg. Med. Chem. Lett. 2009, 19, 199.
10. Kulkarni, S. S.; Kulkarni, V. M. J. Med. Chem. 1999, 42, 373.
11. Mjalli, A. M. M.; Chapman, K. T.; MacCross, M.; Thornberry, N. A. Bioorg. Med.
Chem. Lett. 1993, 3, 2689.
12. Harter, W. G.; Albrecht, H.; Brady, K.; Caprathe, B.; Dunbar, J.; Gilmore, J.; Hays,
S.; Kostlan, K. R.; Lunney, B.; Walker, N. Bioorg. Med. Chem. Lett. 2004, 14, 809.
13. Brady, K. D. Biochemistry 1998, 37, 8508.
14. All compounds were characterized by 1H NMR, MS and were >98% pure by
either CHN analysis or analytical HPLC.
15. Albeck, A.; Estreicher, G. Tetrahedron 1997, 53, 5325.
16. Human peripheral blood mononuclear cells (PBMC) were isolated from
heparinized blood by centrifugation over a Ficoll cushion and then washed
three times with PBS. The PBMC’s were suspended in a medium containing RPMI
1640 with glutamine, penicillin, streptomycin, and 2% human AB serum and then
plated at 106 cells per well in a 96 well flat bottom plate. PBMC’s were stimulated
overnight with 10 ng/mL of LPS in the presence or absence of inhibitor. The
medium was harvested and the level of mature IL-1b was determined using an
ELISA kit from R&D Systems. Compound inhibition (IC50 values) was assessed by
determining the concentration of agent which reduced IL-1b levels by 50%.