Identification of potent and novel small-Molecule inhibitors of caspase-3

Article abstract of DOI:10.1016/j.bmcl.2003.08.024

The design and synthesis of a series of novel, reversible, small molecule inhibitors of caspase-3 are described.

Full text of DOI:10.1016/j.bmcl.2003.08.024

Bioorganic & Medicinal Chemistry Letters 13 (2003) 3651–3655
Identification of Potent and Novel Small-Molecule Inhibitors of
Caspase-3
Darin A. Allen,* Phuongly Pham, Ingrid C. Choong, Bruce Fahr, Matthew T. Burdett,
Willard Lew, Warren L. DeLano, Eric M. Gordon, Joni W. Lam, Tom O’Brien
and Dennis Leey
Sunesis Pharmaceuticals, Inc., 341 Oyster Point Boulevard, South San Francisco, CA 94080, USA
Received 28 May 2003; revised 13 August 2003; accepted 13 August 2003
Abstract—The design and synthesis of a series of novel, reversible, small molecule inhibitors of caspase-3 are described.
# 2003 Elsevier Ltd. All rights reserved.
Caspases (cysteinyl aspartate-specific proteases) have
been implicated in various disease states involving dys-
regulated apoptosis such as myocardial infarction,
stroke, traumatic brain injury, and sepsis. Caspase-3 lies
at a key junction in the apoptotic cascade, mediating
apoptosis from both the intrinsic and extrinsic activa-
tion pathways. Therefore, caspase-3 inhibition may
prove to be a valuable therapeutic approach to treating
these diseases.1 As part of our ongoing efforts to dis-
cover small molecule caspase-3 inhibitors through use of
Tethering2 with extenders, we identified compound 1 as
a low micromolar (Ki=4.0 mM) inhibitor.3 Molecular
modeling of 1 bound to the caspase-3 active site sug-
gested that the molecule interacts with the protein via
the catalytic cysteine (Cys163) as well as with the S1 and
S4 binding subsites of caspase-3, and reproduces the
binding mode observed in crystal structures of earlier
analogues.3 Compounds derived from the DEVD tetra-
peptide sequence have been shown to be potent inhibitors
of Caspase-3, and the P2 isopropyl side chain of the
valine residue is important for inhibitory activity.4
Because the proposed model of 1 bypasses the S2
recognition site, we reasoned that the inhibitory potency
of 1 could be increased by incorporation of a P2 binding
element. The present report describes one of our approa-
ches in developing SAR around this series of compounds
by the systematic examination of P2 and P4 substituents,
scaffold rigidification and aldehyde warhead replacement.
Chemistry
For synthetic ease, we initially prepared the compounds
as diastereomeric mixtures at P2. Syntheses of alkyl
analogues 2–7 is described in Scheme 1. The P1 amino acid
and warhead portions of the molecule were installed using
*Corresponding author. Tel.:+1-650-266-3651; fax:+1-650-266-3501; e-mail: dallen@sunesis.com
yPresent address: Glaxo SmithKline Pharmaceuticals, 709 Swedeland Rd., King of Prussia, PA 19406, USA.
0960-894X/$ - see front matter # 2003 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2003.08.024

3652
D. A. Allen et al. / Bioorg. Med. Chem. Lett. 13 (2003) 3651–3655
Scheme 3. (a) H2SO4/Mg2SO4/CH2CH2/t-BuOH; (b) NaH/DMF; (c)
1:4 TFA/CH2Cl2; (d) (i) Oxalyl chloride/cat. DMF/EtOAc; (ii) n-
BuLi/(S)-(À)-4-benzyl-2-oxazolidinone; (iii) flash chromatography
(25/75 EtOAc/hexane); (e) LiOH/H2O2/THF/H2O; (f) N2H4/EtOH/
reflux; (g) Fmoc-OSu/NaHCO3/dioxane/H2O; (h) H2N-Asp(O-t-Bu)-
CHNNHCONH- resin/EDC/DMF; (i) 1:4 piperidine/DMF; (j) 5-
chlorosulfonylsalicylic acid/DIEA/CH2Cl2; (k) 5:1:1:0.25 THF/AcOH/
MeCHO/TFA; (l) 1:3 TFA/CH2Cl2.
Scheme 1. (a) DMF di-t-butyl acetal/benzene/reflux;7 (b) NaN3/
DMSO; (c) LDA/THF/À78 ꢀC; (d) alkyl halide; (e) H2/Pd/C; (f)
Fmoc-OSu/NaHCO3/dioxane/H2O; (g) 1:3 TFA/CH2Cl2; (h) H2N–
Asp(O-t-Bu)–CHNNHCONH– resin/EDC/DMF; (i) 1:4 piperidine/
DMF; (j) 5-chlorosulfonyl salicylic acid/DIEA/CH2Cl2; (k) 5:1:1:0.25
THF/AcOH/MeCHO/TFA.
Scheme 2. (a) DMF di-t-butyl acetal/benzene/reflux or H2SO4/
Mg2SO4/CH2CH2/t-BuOH8; (b) NaH/DMF or LDA/THF; (c) for
P=Phth; (i) N2H4/EtOH/reflux; (ii) Fmoc-OSu/NaHCO3/dioxane/
H2O; (iii) 1:3 TFA/CH2Cl2; for P=Boc2, (i) 1:3 TFA/CH2Cl2; (ii)
Fmoc-OSu/NaHCO3/dioxane/H2O; (d) H2N-Asp(O-t-Bu)–CHNNH-
CONH–resin/EDC/DMF; (e) 1:4 piperidine/DMF; (f) 5-chloro-
sulfonylsalicylic acid/DIEA/CH2Cl2; (g) 5:1:1:0.25 THF/AcOH/
MeCHO/TFA; (h) 1:3 TFA/CH2Cl2.
Scheme 4. (a) For R=Ph, PhBr/NaHDMS/Pd(OAc)2/ligand/THF;
for R=Pr; (i) NaH/allyl bromide/DMF; (b) Cl2CH(OMe)/SnCl4/
CH2Cl2;13 (c) HONH2/EtOH; (d) Zn/AcOH; (e) NaOH/MeOH; (f)
Fmoc-OSu/NaHCO3/dioxane/H2O; (g) for R=Pr only H2/Pd/C/
EtOAc; (h) H2N-Asp(O-t-Bu)–CHNNHCONH–resin/EDC/DMF; (i)
1:4 piperidine/DMF; (j) 5-chlorosulfonylsalicylic acid/DIEA/CH2Cl2;
(k) 5:1:1:0.25 THF/AcOH/MeCHO/TFA; (l) 1:3 TFA/CH2Cl2.
solid phase synthesis. A resin-bound semicarbazide5 was
treated with Fmoc-Asp(Ot-Bu)-CHO6 to give the sup-
port-bound semicarbazone. Fmoc removal (piperidine/
DMF) was then followed by amide formation with the
desired acid. A second Fmoc deprotection followed by
treatment with 5-chlorosulfonylsalicylic acid provided
the resin-bound sulfonamide. Cleavage from the resin
gave the aldehyde and removal of the Asp t-butyl ester
gave the final products 2–7 which were purified via
reverse-phase preparatory HPLC.
Syntheses of sulfonyl chlorides needed for compounds
22–29 were accomplished either from direct chloro-
sulfonylation with ClSO3H or from the treatment of the
requisite aryl diazonium salt with SO2 and a copper
catalyst according to literature procedure.9 The thio-
methyl10 and oxazolyl ketones11 of Asp were coupled
with the requisite acid to give compounds 31–33.
The syntheses of thiophene analogues 34–36 are descri-
bed in Scheme 4. Alkylation of methyl 2-thiopheneace-
tate with allyl bromide or palladium catalyzed
arylation12 provided substituted acetates 37 and 38,
which were fully elaborated as shown.
Derivatives 8–20 were synthesized as described in
Scheme 2. Alkylation of the requisite ester with either
bis-Boc or phthalimide protected amino butyl bromide
furnished protected amino acids 21, which were fully
elaborated in a similar manner as described above.
Results and Discussion
Since direct chromatographic separation of the diaster-
eomers of 9 was not possible, the enantiomers of an
advanced intermediate, derivatized as acyloxazolidinone
diastereomers, were separated via column chromato-
graphy (Scheme 3). Cleaving the chiral auxiliary and
subsequent elaboration gave the final products 9a and
Our initial series of compounds were designed to
explore the effect of a P2 binding element by installing
substituents alpha to the amide carbonyl of 1 (Table 1).
In general, alkyl substitutions with increasing steric bulk
resulted in greater caspase-3 inhibitory activity relative
to 1. An aromatic substituent gave the largest increase
in potency. For example, phenyl analogue 9 displayed a
40-fold gain in activity. Analogues with simple changes
to the aromatic ring had activity similar to 9, with
compounds 14 and 19 being the most potent of the ser-
1
9b. Unfortunately analysis via H NMR indicated that
each diastereomer was contaminated with 20% of the
other diastereomer, indicating that partial epimerization
had occured in one of the later synthetic steps.

D. A. Allen et al. / Bioorg. Med. Chem. Lett. 13 (2003) 3651–3655
3653
ies. As previously mentioned, all analogues were syn-
thesized as diastereomeric mixtures at P2 for synthetic
convenience. However, once general SAR trends were
established we went back and examined the importance
of stereochemistry at the P2 position. Molecular model-
ing of 9 (Fig. 1) indicated that both diastereomers were
capable of orienting the phenyl ring into the S2 subsite
of caspase-3. To this end, the two diastereomers 9a and
9b were prepared and isolated as 80:20 mixtures and
evaluated for caspase-3 inhibitory activity (Table 2).
Diastereomer 9a appeared to be 5-fold more active than
diastereomer 9b, showing the importance of stereo-
chemistry at the P2 positon. However, we did not attempt
to determine the absolute configuration of 9a or 9b.
ing results summarized in Table 3. Among the
analogues evaluated, only oxadiazolone 27 exhibited
inhibitory activity comparable to that of salicylic acid
analogue 9. The decreased potency of the corresponding
N-methyl oxadiazolone 28 appears to indicate the
importance of an acidic proton in retaining good inhi-
bitory activity.16 We also briefly examined the replace-
ment of the sulfonamide in 9 with an amide. The 4-fold
loss in potency of amide 29 indicates the importance of
the ability of the sulfonamide linkage to adopt a cis
Table 3. In vitro caspase-3 inhibition assay results for compounds
22–29
We previously had determined that the hydroxyl and
the carboxylic acid group of the salicylic acid moiety are
involved in key binding interactions with amino acid
residues of the caspase-3 S4 binding subsite.3 We selec-
ted compound 9 as a starting point for additional ana-
logue synthesis to investigate SAR around the P4 group.
A number of analogues containing several known car-
boxylic acid bioisosteres15 were synthesized with bind-
Compd
R=
R0=
X=
Ki(app), mM
9
COOH
H
H
COOH
OH
H
OH
H
SO2
SO2
SO2
SO2
0.1
5.8
2.0
6.8
22
23
24
Table 1. In vitro caspase-3 inhibition assay results for compounds 1–
20. The effectiveness of compounds against the activity of human
recombinant caspase-3 was measured by collecting kinetic data over a
20-min fluorescence-based assay using Ac-DEVD-AFC as substrate.
All Ki(apparent) values were calculated using the following equation:
Ki=IC50/(1+[substrate]/Km)3
25
26
H
H
SO2
SO2
1.4
0.5
27
H
SO2
0.1
28
29
H
SO2
CO
1.8
0.4
Compd
R=
Ki(app), mM
COOH
OH
1
2
3
4
5
6
7
8
H
Me
Et
n-Pr
i-Pr
i-Bu
4.0
6.4
1.8
0.8
0.7
0.6
4.4
1.3
0.1
0.1
0.2
0.2
0.2
0.07
0.2
0.2
0.1
0.6
0.07
0.1
Table 4. In vitro caspase-3 inhibition assay results for compounds
30–33
PhCH2
Cyclopentyl
Ph
9
10
11
12
13
14
15
16
17
18
19
20
2-FPh
3-FPh
4-FPh
2-ClPh
3-ClPh
4-ClPh
2-MePh
3-MePh
4-MePh
2-Thienyl
3-Thienyl
Compd
R=
Ki(app), mM
9
30
H
–CH2SCH3
0.1
0.6
31
32
0.3
3.3
Table 2. In vitro caspase-3 inhibition assay results for the enriched
diastereomeric mixtures of 9
Compd
Diastereomer ratio (a/b)
Ki(app), mM
9
9a
9b
50:50
80:20
20:80
0.1
0.04
0.2
33
1.0

3654
D. A. Allen et al. / Bioorg. Med. Chem. Lett. 13 (2003) 3651–3655
Table 5. In vitro caspase-3 inhibition assay results for compounds
34–36
Compd
R=
Ki(app), mM
34
35
36
H
n-Pr
Ph
4.0
0.3
0.3
ing the P2 and P4 binding elements as seen in 9 (Fig. 2). The
inhibitory activities for thiophene analogues 34–36 are
shown in Table 5. Simply rigidifying the flexible chain
resulted in a compound (34) equipotent to 1. Adding a
P2 binding element gave a 10-fold increase in potency,
yet the compounds were not better inhibitors than 9.
These results suggest that although the strategy of rigi-
difying the flexible linker of 9 is valid, a thiophene ring
does not appear to allow the ideal binding geometry
between the salicylic acid, a large P2 substituent and the
aspartyl side chain acid that is seen in 9.
Figure 1. Theoretical binding conformation of the (S,S) diastereomer
of 9 in the caspase-3 active site..14
In summary, we have discovered a novel series of
reversible,17 potent in vitro caspase-3 inhibitors through
the use of our extended tethering technology. Beginning
with an early lead, compound 1, we developed SAR by
systematically examining both P2 and P4 binding ele-
ments, aldehyde warhead replacements and scaffold rigi-
dification. From our work described in this paper, we feel
there is potential for further optimization of this series and
subsequent work will be reported in due course.
Figure 2. Theoretical binding conformation of the (S,S) diastereomer
of 35 in the caspase-3 active site.
conformation in orienting the salicylic acid group in the
caspase-3 S4 binding subsite.
Acknowledgements
We would like to thank Stuart Lam, Thomas Webb and
Alex Hsi for their assistance with preparatory HPLC
purifications.
Due to potential therapeutic liabilities of an aldehyde
warhead, we decided to investigate warheads based
upon less reactive ketones. Both thiomethyl and oxazo-
lyl ketones have been shown to be suitable warheads in
peptidomimetic based caspase inhibitors.9,10 We pre-
pared several analogues of compound 9 with the results
shown in Table 4. Incorporation of either thiomethyl or
oxazolyl ketones onto 9 yielded inhibitors with reduced
potency relative to the parent aldehyde. Among the
ketone analogues in this series, the thiomethyl ketones
30 and 31 exhibited the best potency, indicating the
reactivity of the carbonyl is more important than
References and Notes
1. (a) McBride, C. B.; McPhail, L. T.; Steeves, J. D. Emerg.
Therap. Targets 1999, 3, 391. (b) Earnshaw, W. C.; Martins,
L. M.; Kaufman, S. H. Annu. Rev. Biochem. 1999, 68, 383. (c)
Salvesen, G. S.; Denault, J. B. Chem. Rev. 2002, 102, 4489.
2. Tethering is a service mark of Sunesis Pharmaceuticals for
its fragment-based drug discovery.
0
potential binding contacts in the S1 subsite.
3. (a) Choong, I. C.; Lew, W.; Lee, D.; Pham, P.; Burdett,
M. T.; Lam, J. W.; Wiesmann, C.; Luong, T. N.; Fahr, B.;
DeLano, W. L.; McDowell, R. S.; Allen, D. A.; Erlanson,
D. E.; Gordon, E. M.; O’Brien, T. J. Med. Chem. 2002, 45,
5005. (b) Erlanson, D. E.; Lam, J. W.; Wiesmann, C.; Luong,
T. N.; Simmons, R. L.; DeLano, W. L.; Choong, I. C.; Bur-
dett, M. T.; Flanagan, W. M.; Lee, D.; Gordon, E. M.;
O’Brien, T. Nat. Biotechnol. 2003, 308.
We then turned our attention to replacing the flexible
scaffold of 9 with a more rigid framework. We reasoned
that replacement of three of the methylene groups of the
main chain by an aromatic ring could improve potency
by eliminating several rotational degrees of freedom.
We chose the 2,5-disubstituted thiophene system to test
this hypothesis since molecular modeling suggested this
system should best approximate the geometry for present-
4. Cai, S. X.; Drewe, J.; Wang, Y.; Weber, E. US Patent
6,153,591, 2000.

D. A. Allen et al. / Bioorg. Med. Chem. Lett. 13 (2003) 3651–3655
3655
5. Batchelor, J. B.; Drewe, J.; Wang, Y.; Weber, E.; Bebbing-
ton, D.; Bemis, G. W.; Fridman, W. H.; Gillespie, R. J.;
Golec, J. M. C.; Lauffer, D. J.; Livingston, D. J.; Matharu,
S. S.; Mullican, M. D.; Murcko, M. A.; Murdoch, R.; Zelle,
R. E. US Patent 6,204,261 B1, 2001.
11. Bemis, G. W.; Golec, J. M. C.; Lauffer, D. J.; Mullican,
M. D.; Murcko, M. A.; Livingston, D. J. US Patent 6,103,711,
2000.
12. Moradi, W. A.; Buchwald, S. L. J. Am. Chem. Soc. 2001,
123, 7996.
6. Karanewsky, D. S.; Kalish, V. J.; Robinson, E. D.; Ullman,
B. R. WO0023421, 2000.
7. Widmer, U. Synthesis 1983, 135.
8. Wright, S. W.; Hageman, D. L.; Wright, A. S.; McClure,
L. D. Tetrahedron Lett. 1997, 38, 7345.
9. Jones, T. R.; Varney, M. D.; Webber, S. E.; Lewis, K. K.;
Marzoni, G. P.; Palmer, C. L.; Kathardekar, V.; Welsh, K. M.;
Webber, S.; Matthews, D. A.; Appelt, K.; Smith, W. W.; Jan-
son, C. A.; Villafranca, J. E.; Bacquet, R. J.; Howland, E. F.;
Booth, C. L. J.; Herrmann, S. M.; Ward, R. W.; White, J.;
Moomaw, E. W.; Bartlett, C. A.; Morse, C. A. J. Med. Chem.
1996, 39, 904.
13. Garvey, E. P.; Oplinger, J. A.; Tanoury, G. J.; Sherman,
P. A.; Fowler, M.; Marshal, S.; Harmon, M. F.; Paith, J. E.;
Furfine, E. S. J. Biol. Chem. 1994, 269, 26669.
14. Graphics produced using PyMol v. 0.86; DeLano Scien-
tific LLC; San Carlos, CA, USA; http://www.pymol.org
15. Kohara, Y.; Kubo, K.; Imamiya, E.; Wada, T.; Inada, Y.;
Naka, T. J. Med. Chem. 1996, 39, 5228.
16. The pKa of 3-phenyl-1,2,4-oxadiazol-5(4H)-one has been
reported as 5.06; see: Ponzio, G.; Zanardi-Lamberti, B. Gazz.
Chim. Ital. 1923, 53, 821.
17. Representative analogues were tested for reversible inhi-
bition by incubation of compound with Caspase-3 followed by
analysis by MS.
10. Mjalli, A. M.; Chapman, K. T.; MacCoss, M.; Thorn-
berry, N. A. Bioorg. Med. Chem. Lett. 1993, 3, 2689.