Novel β-lactam derivatives: Potent and selective inhibitors of the chymotrypsin-like activity of the human 20S proteasome

Article abstract of DOI:10.1016/j.bmcl.2006.10.047

A series of β-lactam derivatives has been designed and synthesized to inhibit the chymotrypsin-like activity of the human 20S proteasome. The most potent compounds of this new structural class of β-subunit selective 20S proteasome inhibitors exhibit IC

Full text of DOI:10.1016/j.bmcl.2006.10.047

Bioorganic & Medicinal Chemistry Letters 17 (2007) 358–362
Novel b-lactam derivatives: Potent and selective inhibitors
of the chymotrypsin-like activity of the human 20S proteasome
*
´
´
Patricia Imbach, Marc Lang, Carlos Garcıa-Echeverrıa, Vito Guagnano,
Maria Noorani, Johannes Roesel, Francis Bitsch, Grety Rihs and Pascal Furet*
Novartis Institutes for BioMedical Research, WKL-136.4.25, CH-4002 Basel, Switzerland
Received 11 September 2006; revised 19 October 2006; accepted 20 October 2006
Available online 24 October 2006
Abstract—A series of b-lactam derivatives has been designed and synthesized to inhibit the chymotrypsin-like activity of the human
20S proteasome. The most potent compounds of this new structural class of b-subunit selective 20S proteasome inhibitors exhibit
IC50 values in the low-nanomolar range and show good selectivity over the trypsin-like and post-glutamyl-peptide hydrolytic activ-
ities of the enzyme.
ꢀ 2006 Elsevier Ltd. All rights reserved.
The proteasome is a multicatalytic protease complex in-
volved in the ubiquitin (Ub)-dependent degradation of
proteins that are components of critical intracellular reg-
ulatory cascades (e.g., mitotic cycle, cell growth and via-
bility, antigen presentation or inflammatory response).1
The proteolytic activity of this N-terminal threonine
hydrolase occurs in a 700-kDa cylindrical-shaped core
structure known as the 20S proteasome. This catalytic
core consists of four stacked rings arrayed in an
a7b7b7a7 manner, and exhibits at least three distinct pep-
tidase activities: the chymotrypsin-like, trypsin-like, and
post-glutamyl-peptide hydrolytic activities.2 Inhibitors
of the 20S proteosome have been explored for use as
anti-inflammatory agents and for the treatment of can-
cer and autoimmune diseases. Our specific target in
the search for novel cytotoxic and antiproliferative
agents is the chymotrypsin-like activity of the 20S
proteasome. Modulation of this enzymatic activity by
b-subunit-specific inhibitors may convey an antitumor
effect through induction of cell cycle arrest and apopto-
sis in tumor cells.3,4 Parallel to our efforts to develop
noncovalent inhibitors of the 20S proteasome,5–8 we
have also explored the possibility of designing inhibitors
able to form an adduct with the hydroxyl group of the
catalytic threonine residue. We report herein, the design,
synthesis, and biological evaluation of a new class of 20S
proteasome inhibitors that bear a C-terminal substituted
b-lactam as a reactive group. These compounds exploit
key interactions identified during the optimization of
our noncovalent inhibitors, and exhibit high selectivity
for the chymotrypsin-like activity of the 20S
proteasome.
The first compound in this new series (compound 8) was
prepared as outlined in Scheme 1. A key step in our ini-
tial synthetic strategy was the formation of the b-lactam.
This was accomplished by reaction of the dianion-eno-
late of the Cbz-protected a-amino acid ester 1 with the
secondary cyanomethylamine (compound 2).9,10 The lat-
ter serves as the precursor of an in situ generated imine.
The resolution of the racemic, protected b-lactam (com-
pound 3) was performed by HPLC using a semi-pre-
parative chiral column.11
A
linear sequence of
standard coupling and de-protecting reactions led to
precursor 7. The target compound 8 was obtained by
oxidative de-protection of compound 7 using potassium
peroxodisulfate in buffered media.12 The absolute ste-
reochemistry of the b-lactam ring in compound 8 and
derivatives thereof (Table 1) was assigned from the X-
ray structure of the synthetic precursor 7 (Fig. 1).13
The introduction of methoxy groups in the central phen-
yl ring of compound 8, motivated by their beneficial ef-
fect at this position in our previous series of noncovalent
inhibitors, required to modify our initial synthetic
strategy. The oxidative de-protection of the b-lactam
Keywords: Proteasome inhibitors; b-Lactams; Covalent inhibitors;
Antiproliferative agents.
*
Corresponding authors. Tel.: +41 61 696 62 56; fax: +41 61 696 62 46
(P.I). Tel.: +41 61 696 79 90; fax: +41 61 696 15 67 (P.F); e-mail
addresses: patricia.imbach@novartis.com; pascal.furet@novartis.
com
0960-894X/$ - see front matter ꢀ 2006 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2006.10.047

P. Imbach et al. / Bioorg. Med. Chem. Lett. 17 (2007) 358–362
359
H
Cbz
N
H2N
O
a
b,c
Cbz
O
N
O
O
H
N
N
O
O
O
O
H
N
N
O
O
1
2
3
4
O
O
d,e
f,g
h
O
H2N
N
N
N
O
H2N
H
O
N
H
H
N
O
O
O
O
5
6
O
O
i
O
O
O
O
O
O
N
N
N
N
N
H
N
O
H
H
H
H
N
NH
H
O
O
O
8
7
Scheme 1. Synthesis of compound 8. Reagents and conditions: (a) LHMDS, ꢀ70 ꢁC ! rt, CH3CN; (b) HPLC separation, CHIRALCEL OD1118;
(c) H2, Pd/C 10% in CH3OH, rt; (d) Na-Cbz-L-Val-OH, TPTU, DIEA, DMF, rt; (e) Pd/C 10% in CH3OH, rt; (f) Na-Cbz-L-Phe-OH, TPTU, DIEA,
DMF, rt; (g) Pd/C 10% in CH3OH, rt; (h) (3-phenoxyphenyl)-acetic acid, TPTU, DIEA, DMF, rt; (i) K2S2O8/Na2HPO4 in CH3CN/H2O.
The ability of compounds 8–17 (Table 1) to inhibit the
proteolytic activities of the human 20S proteasome
was determined in vitro using fluorogenic peptides as
substrates.15 The rates of hydrolysis were monitored
by the fluorescence increase and the initial linear por-
tions of the curves were used to calculate the IC50 values
(Tables 1 and 2). Covalent interaction of compound 8
with the 20S proteasome protein was assessed by mass
spectrometry.16 The antiproliferative activity of com-
pounds 8, 10, and 14 (Table 2) was determined using
the human breast carcinoma cell line MDA-MB-435.17
Figure 1. X-ray crystal structure of compound 7 (acetone solvate).
ORTEP plot with displacement ellipsoids represented at the 30%
A substantial number of 20S proteasome inhibitors de-
scribed today exert their biological activity by forming
a covalent bond with the side chain of the catalytic
N-terminal threonine residue. These covalent, active
site-directed, inhibitors belong to the following classes
probability level. The configuration of the chiral center in the b-lactam
ring is R.
of compounds: epoxyketones, boronic acids, a-ketoa-
mides, a-ketoaldehydes, b-lactones, and vinyl sulf-
ones.3a,b As an alternative to the preceding reactive
groups, we decided to explore the possibility of using a
chiral b-lactam group to target the catalytic residue of
the 20S proteasome. b-Lactams have been successfully
used as war head groups in the design of covalent inhib-
itors of serine proteases.18 They exert their inhibitory ac-
tion on this class of enzymes by acylating the
nucleophilic catalytic serine residue. We reasoned that
a similar mechanism of action could be exploited to
inhibit the 20S proteasome.19 To explore this new strat-
egy, we decided to use the molecular scaffold of a class
nitrogen at the last step in Scheme 1 was not tolerated
when other methoxy groups were present in the mole-
cule, and it had to be performed earlier in the synthetic
sequence. In addition, and to facilitate the scale-up of
enantiomerically pure compounds, we decided to avoid
the tedious chromatographic resolution of the racemic
b-lactam. Both requirements were fulfilled by following
the synthetic protocol outlined in Scheme 2. In particu-
lar, the resolution of compound 3a was straightforward-
ly accomplished by crystallization with (+)-mandelic
acid.14 Compounds 9–17 (Table 1) were prepared from
compound 4a by using the conditions described in
Scheme 1.

360
P. Imbach et al. / Bioorg. Med. Chem. Lett. 17 (2007) 358–362
Table 1. Inhibition of the chymotrypsin-like activity of the 20S proteasome by b-lactam derivatives
R2
R2
R2
O
R3
O
H
3
N
R1
N
N
H
H
NH
O
O
Compound
R1
R2
R3
R/Sa
IC50 (lM)
8
9
CH2Ph-(3)–OPh
CH2Ph-(3)–OPh
CH2Ph-(3)–OPh
CH2Ph-(3)-OPh
O–CH2Ph
H
H
Isobutyl
Isobutyl
Isobutyl
Isobutyl
Benzyl
Benzyl
Benzyl
Benzyl
Benzyl
Benzyl
R
S
R
S
R
S
R
S
R
S
0.020
38
10
11
12
13
14
15
16b
17b
OCH3
OCH3
OCH3
OCH3
OCH3
OCH3
OCH3
OCH3
0.0025
1.3
0.0050
1.4
O–CH2Ph
CH2Ph-(3)–OPh
CH2Ph-(3)–OPh
CH2Ph-(3)–OPh
CH2Ph-(3)–OPh
0.0014
0.17
0.0040
0.28
The IC50 value is the concentration of inhibitor at which the rate of the 20S proteasome catalyzed hydrolysis of the substrate Suc-Leu-Leu-Val-Tyr-
AMC (chymotrypsin-like activity) is reduced by 50%.
a Absolute configuration at C(3).
b Compounds 16 and 17 contain a tert-leucine instead of a valine residue N-terminal to the b-lactam group.
Cbz
O
HN
H2N
O
H2N
O
a
b,c
N
H
O
O
N
N
NH
O
O
O
3a
4a
4
Scheme 2. Reagents and conditions: (a) crystallization with (+)-mandelic acid; (b) Na-Cbz-L-Val-OH, TPTU, DIEA, DMF, rt; (c) K2S2O8/Na2HPO4
in CH3CN/H2O.
no-azetidine-2-one moiety is shown in Table 1. Accord-
ing to our modeling studies (Fig. 3), it was predicted that
Table 2. 20S proteasome chymotrypsin-like (ChyL), post-glutamyl-
peptide hydrolytic (PGPH), and trypsin-like (TryL) inhibitory activ-
the R configuration at position 3 of the b-lactam ring
was necessary to properly orient the bulky substituent
in the S1 pocket. This key structural requirement is
ities as well as antiproliferative activity of some b-lactam derivatives
Compound ChyL IC50 PGPH IC50 TryL IC50 MDA-MB435
(lM)
(lM)
(lM)
IC50 (lM)
exemplified by the 70- to 2000-fold higher inhibitory
activity observed for the R-epimer (Table 1). As ob-
served with our noncovalent inhibitors introduction of
methoxy groups in the central phenyl ring to form
hydrogen bonds with residues of the S3 pocket improves
potency (compound 8 vs 10).21 However, less sensitivity
of the inhibitory activity to the optimal filling of the S1,
AS1, and AS2 pockets is observed with the covalent
inhibitors. The isobutyl and benzyl groups in the S1
pocket give compounds of equal potency (compound
10 vs 14) while compound 12 whose N-terminal group
can occupy only one of the two accessory pockets AS1
and AS2 is not dramatically less active than 14 that
occupies both. This comes in support of the covalent
nature of the inhibitory activity of these compounds.
Covalent inhibitors are less dependent on the creation
of optimal intermolecular interactions to achieve
potency.
8
10
14
0.020
0.0025
0.0014
>20
>100
>20
n.d.
1.1
19
1.2
0.034
0.032
of potent noncovalent inhibitors of the chymotrypsin-
like proteolytic activity of the proteasome previously de-
scribed by us.8 Thus, we undertook molecular modeling
studies20 aimed at designing the most promising b-lac-
tam analogues of these new 20S proteasome inhibitors.
On the basis of our binding model hypotheses,8 we
determined that replacing the C-terminal benzylic group
of our noncovalent inhibitors by a chiral b-lactam moi-
ety (Fig. 2) would provide potent and covalent 20S pro-
teasome inhibitors. The biological activity obtained with
representative examples of compounds containing a C-
terminal 3-substituted (isobutyl or benzyl) (R)-3-ami-

P. Imbach et al. / Bioorg. Med. Chem. Lett. 17 (2007) 358–362
361
O
R
R
O
OH
O
O
NH
O
R'
H
O
H
R'
N
N
N
N
N
N
H
H
H
H
O
O
Figure 2. Schematic representation of the transformation of a noncovalent inhibitor into a covalent 20S proteasome inhibitor.
References and notes
1. Tanaka, K. J. Biochem. 1998, 123, 195.
2. Orlowski, M.; Cardozo, C.; Michaud, C. Biochemistry
1993, 32, 1563.
3. For recent reviews on this target and 20S proteasome
´
´
inhibitors, see: (a) Garcıa-Echeverrıa, C. Int. J. Pept. Res.
Ther. 2006, 12, 49; (b) Delcros, J.; Baudy Floc’h, M.;
Prigent, C.; Arlot-Bonnemains, Y. Curr. Med. Chem.
´
´
2003, 10, 479; (c) Garcıa-Echeverrıa, C. Mini-Rev. Med.
Chem. 2002, 2, 247; (d) Elliott, P.; Ross, J. Am. J. Clin.
Pathol. 2001, 116, 637; (e) Shah, S.; Potter, M.; Callery, A.
Surg. Oncol. 2001, 10, 43, and references therein.
Figure 3. Model of compound 14 bound to proteasome X/HC5
subunits after acylation of the catalytic threonine (Thr1). In particular,
the designed b-lactam analogues can form the same key hydrogen
bond interactions (with residues Thr21, Gly 47, Ala 19, and Asp 153)
as their parent noncovalent inhibitors.
4. A dipeptide boronic acid proteasome inhibitor (Bortezo-
mib; PS-341) was approved in May 2003 by the US FDA
for the treatment of patients with relapsed or refractory
multiple myeloma. For additional data on this compound,
see: Nawrocki, S.; Bruns, C.; Harbison, M.; Bold, R.;
Gotsch, B.; Abbruzzese, J.; Elliott, P.; Adams, J.;
McConkey, D. Mol. Cancer Therap. 2002, 1, 1243;
Adams, J.; Palombella, V. J.; Elliott, P. J. Invest. New
Drugs 2000, 18, 109.
Confirmation of a covalent interaction with the enzyme
was obtained by mass spectrometry. The b5 subunit,
which is responsible for the chymotrypsin-like activity
of the 20S proteasome, was shown by LC–MS analy-
ses to be covalently bound by one molecule of
compound 8.22
´
5. Garcıa-Echeverrıa, C.; Imbach, P.; France, D.; Furst, P.;
¨
Lang, M.; Noorani, M.; Scholz, D.; Zimmermann, J.;
´
Furet, P. Bioorg. Med. Chem. Lett. 2001, 11, 1317.
6. Furet, P.; Imbach, P.; Furst, P.; Lang, M.; Noorani, M.;
¨
´
´
Zimmermann, J.; Garcıa-Echeverrıa, C. Bioorg. Med.
Chem. Lett. 2001, 11, 1321.
Compounds containing the C-terminal chiral b-lactam
group are not only potent inhibitors of the chymotryp-
sin-like activity of the 20S proteasome but are also very
selective for this catalytic site. As shown in Table 2,
compounds 8, 10, and 14 showed at least 350-fold selec-
tivity over the trypsin-like and post-glutamyl-peptide
hydrolytic activities of the proteasome.23 In addition,
these compounds displayed a pronounced antiprolifera-
tive effect as illustrated with the inhibitory activity
values obtained against the human breast carcinoma
cell line MDA-MB-435 (e.g., IC50 = 32 nM for
compound 14).
7. Furet, P.; Imbach, P.; Furst, P.; Lang, M.; Noorani, M.;
´
¨
´
Zimmermann, J.; Garcıa-Echeverrıa, C. Bioorg. Med.
Chem. Lett. 2002, 12, 1331.
8. Furet, P.; Imbach, P.; Noorani, M.; Koeppler, J.; Laumen,
K.; Lang, M.; Guagnano, V.; Fuerst, P.; Roesel, J.;
Zimmermann, J.; Garcia-Echeverria, C. J. Med. Chem.
2004, 47, 4810.
9. Overman, L. E.; Osawa, T. J. Am. Chem. Soc. 1985, 107,
1698.
10. Okawara, T.; Harada, K. J. Org. Chem. 1972, 37, 3286.
11. To obtain optically pure material, the racemic product 3
was separated by HPLC using a chiral column. Single
peak at tR = 7.41 min (CHIRALCEL OD (1118);
250 · 4.6 mm; eluting with hexane/ethanol, 90/10; flow
In summary, the data reported in this letter demonstrate
that a C-terminal chiral b-lactam group can effectively
serve as the basis for designing potent and selective
covalent inhibitors of the chymotrypsin-like activity of
the 20S proteasome. These novel inhibitors open a
new avenue for further investigation of the proteasome
as a therapeutic target in oncology drug discovery.
1 mL/min;
DMSO).
12. Podlech, J.; Steurer, S. Synthesis 1999, 4, 650.
k = 210 nm);
[a]D = +20.4 ꢁ
(c = 0.525;
13. Suitable crystals were obtained from an acetone solution
by slow evaporation of the solvent. An Enraf-Nonius
CAD4 automatic diffractometer was used for data collec-
tion with CuKa radiation and a graphite monochromator.
The structure was solved by direct methods (SHELXS).
The parameters were refined by full-matrix least-squares
calculations (SHELXL) with anisotropic displacement
parameters for all non-H atoms. A subsequent difference
Fourier map showed 38 of 58 hydrogen atoms. The
positions of the remaining ones were calculated assuming
normal geometry. Hydrogen atom parameters were ideal-
ized and not refined. Crystallographic data (excluding
Acknowledgment
We thank W. Beck, E. Boss, J.-M. Groell, J. Koeppler,
V. Huy Luu, E. Masso, and H. Walter for their technical
assistance.

362
P. Imbach et al. / Bioorg. Med. Chem. Lett. 17 (2007) 358–362
structure factors) for the structure have been deposited P. A.; Pritchard, G. J.; Schofield, C. J. Biochemistry 1999,
with the Cambridge Crystallographic Data Centre as
supplementary publication numbers CCDC 620075. Cop-
ies of the data can be obtained, free of charge, on
application to CCDC, 12 Union Road, Cambridge CB2
1EZ, UK [Fax: +44 (0)1223-336033 or e-mail:
deposit@ccdc.cam.ac.uk].
38, 7989.
19. An article confirming the interest of b-lactams as
proteasome inhibitors has been published after the com-
pletion of the work reported here. The authors report the
synthesis and proteasome inhibitory activity of a b-lactam
analogue of the natural compound salinosporamide A, a
b-lactone proteasome inhibitor Hogan, P. C.; Corey, E. J.
J. Am. Chem. Soc. 2005, 127, 15386.
20. These modeling experiments were performed in Macro-
model using the AMBER/H2O/GBSA force field for
energy-minimization. Macromodel: Mohamadi, F.; Rich-
ards, N. G. J.; Guida, W. C.; Liskamp, R.; Lipton, M.;
Caufield, C.; Chang, G.; Hendrickson, T.; Still, W. C. J.
Comput. Chem. 1990, 11, 440.
21. The most potent noncovalent inhibitors disclosed in Ref. 8
contain a 3,4,5-trimethoxy-phenyl moiety to fill the S3
pocket, whereas the corresponding group in b-lactam
analogues described here is a 2,3,4-trimethoxy-phenyl
moiety. The latter gives inhibitors of similar potency as
the former in the noncovalent series (manuscript in
preparation).
22. The 20S proteasome preparation was obtained from Dr. J.
Zimmermann (Novartis Pharma Research, Oncology).
The material was dissolved in a Tris buffer, pH 8.0, at a
concentration of approximately 1 mg/mL for each sub-
unit, 10 lL of 10 mM Tris–HCl, pH 8.0, buffer containing
10 mM CaCl2 (Buffer A) was added to 10 lL of the
proteasome solution in Tris–HCl buffer solution. Two
microliters of 100 lM solution of compound 8 in Tris–
HCl buffer was then added and the reaction mixture was
shaken at 37 ꢁC for 1 h using an Eppendorf thermomixer
(700 minꢀ1). Final concentration of compound 8 was
approximately 10 lM. A blank reaction (10 lL protea-
some + 10 lL Buffer A + 2 lL DMSO) was run under the
same conditions. Mass spectrometry was carried out using
a Q-Tof (Micromass, Manchester, UK) quadrupole time-
of-flight hybrid tandem mass spectrometer equipped with
a Micromass Z-type electrospray ionization source (ESI).
23. These compounds also show high selectivity against
important serine and cysteine proteases such as
chymotrypsin, thrombin, and calpain. No inhibition of
these enzymes was detected at a concentration as high as
25 lM.
14. Compound 3a was crystallized with L-(+) mandelic acid
(388 mg; 2.55 mM) from methanol (4 mL). After addition
of diethyl ether, the crystalline product is filtered off. The
crude product is taken up into ethyl acetate/saturated.
NaHCO3 soln and the organic layer is washed with water,
brine, and dried over Na2SO4. After filtration, the solvent
is removed by evaporation under reduced pressure to
obtain 515 mg of enantiomerically pure (S)-2-[(S)-2-ami-
no-3-(2,3,4-trimethoxy-phenyl)-propio-nylamino]-N-((R)-
3-isobutyl-2-oxo-azetidin-3-yl)-3-methyl-butyramide
ES-MS: 293.1 [M+H]+; single peak at tR = 2.99 min
(System 1); Rf = 0.56 (CH2Cl2/CH3OH 9/1);
[a]D = ꢀ22.8 ꢁ (c = 1.015; methanol).
4.
15. The 20S proteasome was obtained in-house from human
blood. Fluorogenic peptides used: Suc-Leu-Leu-Val-Tyr-
AMC (substrate for chymotrypsin-like assay), Boc-Leu-
Arg-Arg-AMC (substrate for trypsin-like assay), and
Z-Leu-Leu-Glu-AMC (substrate for post-glutamyl-pep-
tide hydrolytic-like assay). Fluorescence excitation/emis-
sion wavelengths were 355 nm/460 nm for 7-amido-4-
methyl-coumarin (AMC).
16. A Q-Tof (Micromass, Manchester, UK) quadrupole time-
of-flight hybrid tandem mass spectrometer equipped with
a Micromass Z-type electrospray ionization source was
used.
17. Inhibition of cell growth by test compounds was assessed
by using a MTT-based proliferation assay. The human
breast carcinoma cell line MDA-MB-435 was cultivated in
MEM, supplemented with 10% FCS, 100 U/mL penicillin/
streptavidin, 4 mM L-glutamine, 1 mM sodium pyruvate,
nonessential amino acids (1·), and 20 mM Hepes. After
24, 48 or 72 h of compound incubation, the rate of cell
TM
proliferation was assessed with the CellTiter96 essay. The
plates were read in a microplate reader (Dynatech
MR5000) at 550 versus 630 nm.
18. Wilmouth, R. C.; Kassamally, S.; Westwood, N. J.;
Sheppard, R. J.; Claridge, T. D. W.; Aplin, R. T.; Wright,