Vinyl ester-based cyclic peptide proteasome inhibitors

Article abstract of DOI:10.1016/j.bmcl.2008.02.016

The 20S proteasome is a multicatalytic protease complex responsible for the degradation of many proteins in mammalian cells. Specific inhibition of proteasome enzymatic subunits represents a topic of great interest for the development of new drug therapies. Following our previous development of a new class of peptide-based inhibitors bearing a C-terminal vinyl ester residue as a pharmacophoric unit that are able to interact with the catalytic threonine, we report here the synthesis and biological properties of a new series of vinyl ester cyclopeptide analogues. Some of these derivatives were shown to inhibit the chymotrypsin-like activity of the proteasome at nanomolar concentration and their potency was found to depend on the size of the tetrapeptidic cyclic portion.

Full text of DOI:10.1016/j.bmcl.2008.02.016

Available online at www.sciencedirect.com
Bioorganic & Medicinal Chemistry Letters 18 (2008) 1849–1854
Vinyl ester-based cyclic peptide proteasome inhibitors
Anna Baldisserotto,a Mauro Marastoni,a,* Stella Fiorini,a Loretta Pretto,b
Valeria Ferretti,b Riccardo Gaviolic and Roberto Tomatisa
aDepartment of Pharmaceutical Sciences and Biotechnology Center, University of Ferrara,
Via Fossato di Mortara 17-19, I-44100 Ferrara, Italy
bChemistry Department and Center for Structural Diffractometry, University of Ferrara, I-44100 Ferrara, Italy
cDepartment of Biochemistry and Molecular Biology, University of Ferrara, I-44100 Ferrara, Italy
Received 7 January 2008; revised 6 February 2008; accepted 7 February 2008
Available online 10 February 2008
Abstract—The 20S proteasome is a multicatalytic protease complex responsible for the degradation of many proteins in mammalian
cells. Specific inhibition of proteasome enzymatic subunits represents a topic of great interest for the development of new drug ther-
apies. Following our previous development of a new class of peptide-based inhibitors bearing a C-terminal vinyl ester residue as a
pharmacophoric unit that are able to interact with the catalytic threonine, we report here the synthesis and biological properties of a
new series of vinyl ester cyclopeptide analogues. Some of these derivatives were shown to inhibit the chymotrypsin-like activity of
the proteasome at nanomolar concentration and their potency was found to depend on the size of the tetrapeptidic cyclic portion.
ꢀ 2008 Elsevier Ltd. All rights reserved.
The proteasome, a multicatalytic protease complex,1 is
the essential component of the nonlysosomal protein
degradation pathway in prokaryotes and eukaryotes. It
is involved in many biological processes, including the re-
moval of abnormal, misfolded or improperly assembled
proteins, the stress response, and cell cycle control and
differentiation, as well as metabolic adaptation and the
generation of peptide antigens for presentation to
CD8+ cytotoxic T cells by major histocompatibility com-
plex (MHC) class I molecules. These cellular functions
are linked to an ubiquitin- and ATP-dependent protein
degradation pathway involving the 26S proteasome
(2.4 MDa), whose core and proteolytic chamber are
formed by the 20S proteasome.2
residue of b subunits as nucleophile, via a catalytic
mechanism similar to those of the serine proteases.3
The regulation of proteasomal activities by inhibition or
modulation of its component molecules has considerable
biological significance. Indeed, the development of spe-
cific and selective multicatalytic complex inhibitors into
novel therapeutic agents represents a stimulating ap-
proach in the treatment of many disease states, including
inflammation and cancer, as well as in the modification
of immune responses. Thus, a variety of natural and syn-
thetic products have been tested as inhibitors of the dif-
ferent proteasomal enzymatic subunits.4
In our earlier studies we synthesized and tested a series
of peptide-based derivatives where vinyl ester moiety
was considered as a potential substrate for catalytic
threonine. The most efficacious compounds, with the
general structure Hmb-Xaa-Xbb-Leu-VE, showed good
inhibition, remarkable selectivity for the trypsin-like
activity of the 20S proteasome and favorable pharmaco-
kinetic properties.5
The 20S proteasome consists of four stacked rings and
each of the two inner rings is composed of seven differ-
ent b subunits. Each b-ring contains three different pro-
teolytically active sites: the b1 subunit contains a post-
acidic (PGPH) active site, the b2 subunit is associated
with trypsin-like (T-L) activity and the b5 subunit per-
forms a chymotrypsin-like (ChT-L) proteolytic function.
All its proteolytic sites utilize an N-terminal threonine
Several of these vinyl ester tripeptides were subjected to
conformational studies in order to define the inhibitor–
enzyme interaction subsite pocket, and through flexible
alignment, our selected vinyl ester compounds demon-
strated conformational similarities with the TMC95A
Keywords: Inhibitors; Proteasome; Chymotrypsin-like activity; Cyclic
peptides.
*
Corresponding author. Tel.: +39 532 291281; fax +39 532 291296;
e-mail: mru@dns.unife.it
0960-894X/$ - see front matter ꢀ 2008 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2008.02.016

1850
A. Baldisserotto et al. / Bioorg. Med. Chem. Lett. 18 (2008) 1849–1854
cyclic inhibitor.6 As the availability of active inhibitors
with a rigid conformation could provide much structural
information, and could also indicate a target investiga-
tion into potency and selectivity,7 we set out to develop
molecules with a constrained conformation in the C-ter-
minal of the vinyl ester pharmacophore.8
(diphenylphosphorylazide) under conditions known to
incur minor racemization and minimal oligomerization.9
After saponification of the cyclic tetrapeptides the exo-
cyclic leucine vinyl ester unit, prepared as described pre-
viously,5a was coupled using WSC/HOBt.
All products were purified by preparative RP-HPLC,
and the homogeneity of the purified products was ac-
cessed by HPLC. Structural characterization was then
achieved by electrospray ionization (ESI) mass spec-
trometry (MICROMASS ZMD 2000) (Table 1) and
1H NMR spectroscopy (Bruker AC 200).10
Here, we describe the synthesis and biological activity of
cyclopeptides derived from our linear prototype inhibi-
tors (Fig. 1). In this series, the endocyclic part of the
molecules included the tetrapeptidic sequence Phe-
AA1-AA2-Glu (or Asp); in derivatives 1–4, the central
dipeptide was Leu-Leu, while the more hydrophilic
Val-Ser sequence was contained in the same position
in analogues 5–8. Cyclization was obtained acylating
the N-terminal by the side chain carboxylic function of
the acidic residue (1, 3, 5, and 7) or head-to-tail in com-
pounds 2, 4, 6, and 8. The exocyclic pharmacophoric
unit Leu-VE was linked in the side chain or at the a-car-
boxy group of the acidic residue.
The activity of the vinyl ester cyclopeptides was tested to
assess inhibition of the b1, b2, and b5 active sites of the
20S proteasome, previously purified from lymphoblas-
toid cell lines.11,12 Suc-LLVY-AMC, Boc-LRR-AMC,
and Ac-YVDA-AMC, specific fluorogenic substrates
for the three main proteolytic activities of the enzymatic
complex, were used to measure chymotrypsin-like, tryp-
sin-like, and caspase-like proteasome activities, respec-
tively. Substrates were incubated at 37 ꢁC for 30 min
with the proteasome, untreated or pre-treated with
incremented concentrations (from 0.001 to 10 lM) of vi-
nyl ester cyclopeptides with the reference inhibitors
epoxomicin and the aldehydic tripeptide derivative
MG132 in activity buffer. Fluorescence was determined
in a fluorimeter (Spectrafluor plus, Tecan, Salzburg,
Austria) using an excitation of 360 nm and an emission
of 465 nm. Activity was evaluated in fluorescence units,
and the inhibitory activity of the compounds is ex-
pressed here as IC50. The data were then plotted as per-
centage control (the ratio of percentage conversion in
the presence and absence of the inhibitor) versus the
inhibitor concentration, and fitted with the equation
Y = 100/1 + (X/IC50)A, where IC50 is the inhibitor con-
centration at 50% inhibition and A is the slope of the
inhibition curve.13
Vinyl ester cyclopeptides 1–8 were synthesized following
the strategy reported in Scheme 1 by the classical solu-
tion method using C-terminal stepwise elongation.
Starting from an acidic residue, monoprotected in the
side chain or at the a-carboxy function, other Na-Boc-
protected amino acids of the sequence were condensed
as succinimidyl esters. After each coupling step, Boc
was removed by TFA. Cyclization of the linear tetrapep-
tides was obtained using the activating reagent DPPA
H
N
HN
O
O
O
O
H
N
O
O
In general, the new cyclic analogues reversed the selec-
tivity against the b2 and b5 proteasome catalytic sub-
sites, as compared to the vinyl ester linear prototypes.
Head-to-tail cyclized derivatives showed improved
inhibitory capacity of the chymotryptic-like activity; in
particular compounds 2 and 4 selectively inhibited the
b5 subunit at a nM concentration. In our experiments,
ChT-L inhibition seemed to depend on the cyclic dimen-
sions; indeed, increasing the ring structure from 12 to 14
atoms, noticeably diminished activity. A more hydro-
phobic endocyclic tetrapeptidic sequence favoured b5
interaction, whereas the endocyclic dipeptidic sequence
Val-Ser in derivatives 5–8 slightly improved inhibition
of the proteasome’s trypsin-like activity. The inhibitory
capacity of the cyclopeptides against the b2 subunit was
less influenced by ring size, and in the corresponding lin-
ear reference, the endocyclic hydrophilic sequence was
shown to be more functional for inhibitory activity. Fi-
nally, all compounds were shown to be inactive against
the 20S proteasome b1 catalytic subsite.
NH
O
HN
Phe-Leu-Leu-Glu-Leu-VE
3
NH
O
O
N
H
HN
O O
O O
O
NH
HN
Phe-Leu-Leu-Glu(Leu-VE)
4
The cell membrane permeation of the most representa-
tive compounds, 2, 4, and 6, was then tested in live cells.
After cell treatment, proteasomes were purified and as-
Figure 1.

A. Baldisserotto et al. / Bioorg. Med. Chem. Lett. 18 (2008) 1849–1854
1851
Boc-Xbb-OSu
Xaa
Boc-Xbb-Xaa
1) TFA
2) Boc-Xcc-OSu
Boc-Xcc-Xbb-Xaa
1) TFA
2) Boc-Phe-OSu
Boc-Phe-Xcc-Xbb-Xaa
TFA
H-Phe-Xcc-Xbb-Xaa
DPPA, NaHCO3
c[Phe-Xcc-Xbb-Xaa]
1) NaOH 1N
2) H-Leu-VE, WSC, HOBt
c[Phe-Leu-Leu-Asp]-Leu-VE 1
c[Phe-Leu-Leu-Asp(Leu-VE)] 2
c[Phe-Val-Ser-Asp]-Leu-VE
5
c[Phe-Val-Ser-Asp(Leu-VE)] 6
c[Phe-Leu-Leu-Glu]-Leu-VE
3
c[Phe-Val-Ser-Glu]-Leu-VE
c[Phe-Val-Ser-Glu(Leu-VE)]
7
8
c[Phe-Leu-Leu-Glu(Leu-VE)] 4
Xaa = H-Asp-OMe, H-Asp(OMe)-OH, H-Glu-OMe, H-Glu(OMe)-OH; Xbb = Leu,Ser;Xcc = Leu,Val
Scheme 1. Synthesis of vinyl ester cyclopeptide derivatives 1–8.
360 min and the incubation was halted by addition of
ethanol. After centrifugation, an aliquot of the clear
supernatant was analyzed in HPLC.13 The half-lives of
the cyclopeptides were obtained by least-squares linear
regression analysis of a plot of logarithmic compound
concentrations versus time, using a minimum of five
points. The data reported in Table 2, as expected on
the basis of molecular features determined by the cycli-
zation, show that all derivatives analyzed have great sta-
bility against human plasma protease with a half-live of
over 6 h.
Table 1. Physicochemical data of vinyl ester cyclopeptides
Compound
HPLCa
P.f. (ꢁC) ½aꢀD
MS
20
KI (a) KI (b)
(C = 1, MeOH) [M+H]+
1
2
3
4
5
6
7
8
11.71
12.72 10.43 234–239 ꢁ43.7
11.85
9.98 198–201 ꢁ39.8
9.90 220–223 ꢁ50.5
655.8
655.7
669.5
669.6
615.2
615.2
629.3
629.3
12.81 10.75 187–191 ꢁ42.1
9.60
10.15
9.74
8.38 108–112 ꢁ29.3
9.06 134–136 ꢁ30.9
8.67 123–126 ꢁ22.4
9.51 143–145 ꢁ25.6
10.68
a Capacity factor (KI) of the cyclic peptides was determined by HPLC
In order to obtain a deeper insight into the molecular
basis of the enzyme–inhibitor recognition process, the
molecule exhibiting the highest affinity, 4, was chosen
to perform a docking simulation to the 20S protea-
some.14 In Figure 2, a schematic view of the ligand-en-
zyme is shown. The binding pocket is delimited,
besides some water molecules, by several residues, most
of which corresponded to those found in the b5 active
site of the proteasome-calpain crystallographic structure
retrieved from PDB.15 Only a small part of the inhibitor
(shaded in blue in the figure) is placed towards the inner
proteasome channel. The most remarkable result is that
the vinyl ester moiety of the molecule was found to be
using two different solvent system gradients.
sayed for proteolytic activity using specific substrate for
T-L, ChT-L, and PGPH activities as previously de-
scribed.13 The results obtained (Table 2) were compara-
ble to those observed in the in vitro assay, thereby
demonstrating that vinyl ester cyclopeptides are cell-per-
meable and able to inhibit the proteasome in vivo.
Resistance to proteolysis of the selected cyclic deriva-
tives was studied in heparinized human plasma. The
degradation kinetics were determined by incubation at
37 ꢁC in human plasma for increments of time up to
˚
located some 3–4 A away from the Thr1 residue, in

1852
A. Baldisserotto et al. / Bioorg. Med. Chem. Lett. 18 (2008) 1849–1854
Table 2. Proteasome inhibition and enzymatic stability of cyclopeptides 1–8 and reference inhibitors epoxomicin and MG132
a
a
No.
Compound
Isolated enzyme IC50 (lM)
In vivo inhibition IC50 (lM)
Half-life (min) plasma
ChT-L
T-L
PGPH
ChT-L
T-L
PGPH
1
2
3
4
5
6
7
8
c[Phe-Leu-Leu-Asp]-Leu-VE
c[Phe-Leu-Leu-Asp(Leu-VE)]
c[Phe-Leu-Leu-Glu]-Leu-VE
c[Phe-Leu-Leu-Glu(Leu-VE)]
c[Phe-Val-Ser-Asp]-Leu-VE
c[Phe-Val-Ser-Asp(Leu-VE)]
c[Phe-Val-Ser-Glu]-Leu-VE
c[Phe-Val-Ser-Glu(Leu-VE)]
Epoxomicin
0.206
0.063
0.310
0.039
0.315
0.097
0.354
0.123
0.005
0.002
0.447
0.389
0.424
0.286
0.134
0.175
0.129
0.155
0.284
1.077
>10
>10
>10
>10
>10
>10
>10
>10
4.560
>10
0.081
0.052
0.110
0.436
0.473
0.197
>10
>10
>10
>360
>360
>360
MG132
a The values reported are the average of two independent determinations.
Figure 2.
Figure 3.

A. Baldisserotto et al. / Bioorg. Med. Chem. Lett. 18 (2008) 1849–1854
1853
Momose, I.; Umezawa, Y.; Hirosawa, S.; Iinuma, H.;
Ikeda, D. Bioorg. Med. Chem. Lett. 2005, 15, 1867; (g)
Vivier, M.; Jarrousse, A. S.; Bouchon, B.; Galmier, M. J.;
Auzeloux, P.; Sauzieres, J.; Madelmont, J. C. J. Med.
Chem. 2005, 48, 6731.
accordance with the hypothesis that this molecular frag-
ment can be considered the potential substrate for cata-
lytic threonine. In Figure 3, the docked ligand is
depicted in the active site.
5. (a) Marastoni, M.; Baldisserotto, A.; Cellini, S.; Gavioli,
R.; Tomatis, R. J. Med. Chem. 2005, 48, 5038; (b)
Marastoni, M.; Baldisserotto, A.; Trapella, C.; Gavioli,
R.; Tomatis, R. Bioorg. Med. Chem. Lett. 2006, 16, 3125;
(c) Marastoni, M.; Baldisserotto, A.; Trapella, C.; Gavioli,
R.; Tomatis, R. Eur. J. Med. Chem. 2006, 41, 978.
6. Baldisserotto, A.; Marastoni, M.; Trapella, C.; Gavioli,
R.; Ferretti, V.; Pretto, L.; Tomatis, R. Eur. J. Med.
Chem. 2007, 42, 586.
7. Borissenko, L.; Groll, M. Chem. Rev. 2007, 107, 687.
8. Baldisserotto, A.; Marastoni, M.; Lazzari, I.; Trapella, C.;
Gavioli, R.; Tomatis, R. Eur. J. Med. Chem. 2007, in
press.
In conclusion, we have presented here a new series of vi-
nyl ester proteasome inhibitors. Cyclization of the linear
prototype, achieved by insertion of an acidic residue into
the sequence, changes the activity profile. In the result-
ing cyclic derivatives, the conformational restriction per-
mits a better interaction with the b5 enzymatic subsite.
The inhibition of proteasome ChT-L activity was also
found to depend on the ring size. In some cases, head-
to-tail cyclization, led to cyclopeptides with a potency
in a nanomolar concentration.
With the aim of designing rigid molecules useful for
close structural studies, we have obtained new inhibitors
of the multicatalytic complex, selective for chymotryp-
sin-like activity and stable to enzymatic degradation,
able to permeate cell membranes. These vinyl ester
cyclopeptides represent an interesting springboard for
the future development of new rigid structures with a
better biological profile.
9. (a) Shiori, T.; Ninomiya, K.; Yamada, S. J. Am. Chem.
Soc. 1972, 94, 6203; (b) Brady, S. F.; Freidinger, R. M.;
Colton, C. D.; Homnick, C. F.; Whitter, W. L.; Curley, P.;
Nutt, R. F.; Veber, D. F. J. J. Org. Chem. 1987, 52, 764.
10. HPLC analysis was performed by a Beckman System
Gold with
a Hypersil BDS C18 column (5 lm;
4.6 · 250 mm). Analytical determination and capacity
factor (K0) of the vinyl ester cyclopeptides were deter-
mined using HPLC conditions in the above solvent system
(solvents A and B) programmed at flow rates of 1 mL/min
using the following linear gradients: (a) from 0% to 100%
B for 25 min and (b) from 30% to 90% B for 25 min. c[Phe-
Leu-Leu-Glu(Leu-VE)] (4). Purified yield 36%; purity
estimated by HPLC >98%; 1H NMR (CDCl3,
200 MHz): d (ppm) 1.01 (d, 6H), 1.09 (m, 12H), 1.31 (t,
3H, J = 7.2), 1.44 (m, 2H), 1.70–1.81 (m, 7H), 2.09 (m,
2H), 2.21 (d, 2H), 2.98 (m, 2H), 4.13 (q, 2H, J = 7.3), 4.27–
4.51 (m, 5H), 5.91 (d, 1H, J = 16.0), 6.98 (dd, 1H,
J = 16.1), 7.19–7.33 (m, 5H), 8.05 (br s, 5H). c[Phe-Val-
Ser-Asp]-Leu-VE (5). Purified yield 31%; purity estimated
Acknowledgments
Financial support of this work by University of Ferrara,
`
by the Ministero dell’Universita e della Ricerca Scientif-
ica e Tecnologica (MURST), the Associazione Italiana
per la Ricerca sul Cancro (AIRC), and the Istituto Supe-
`
riore di Sanita (progetto AIDS). English revision of the
text was carried out by Anna Forster.
1
by HPLC >98%; H NMR (CDCl3, 200 MHz): d (ppm)
0.99 (d, 6H), 1.13 (d, 6H), 1.29 (t, 3H, J = 7.3), 1.52 (m,
2H), 1.80 (m, 1H), 2.11 (s, 1H), 2.58–2.87 (m, 5H), 4.03
(m, 2H), 4.18 (q, 2H, J = 7.1), 4.32–4.60 (m, 5H), 5.93 (d,
1H, J = 16.2), 7.02 (dd, 1H, J = 16.1), 7.12–7.31 (m, 5H),
8.02 (br s, 5H).
References and notes
1. (a) Ciechanover, A. Cell 1994, 79, 13; (b) Coux, O.;
Tanaka, K.; Golberg, A. L. Annu. Rev. Biochem. 1996, 65,
801; (c) Hershko, A.; Ciechanover, A. Annu. Rev.
Biochem. 1998, 67, 425.
11. Gavioli, R.; Vertuani, S.; Masucci, M. G. Int. J. Cancer
2002, 101, 532.
12. Hendil, K. B.; Uerkvitz, W. J. Biochem. Biophys. Methods
1991, 22, 159.
13. Purification of the proteasomes and enzyme assays in vitro
and in live cells are described in ‘‘supporting information’’
of Ref. 5a.
2. (a) Coux, O.; Tanaka, K.; Golberg, A. L. Annu. Rev.
Biochem. 1996, 65, 801; (b) Hershko, A.; Ciechanover, A.
Annu. Rev. Biochem. 1998, 67, 425; (c) Orlowski, R. Z. Cell
Death Differ. 1999, 6, 303, 1695, 19; (d) Kisselev, A. F.;
Songyang, Z.; Goldberg, A. L. J. Biol. Chem. 2000, 275,
14831; (e) Myung, J.; Kim, K. B.; Crews, C. M. Med. Res.
Rev. 2001, 21, 245; (f) Boyo, M.; Wang, E. W. Curr. Top.
Microbiol. Immunol. 2002; (g) Adams, J.; Kauffman, M.
Cancer Invest. 2004, 22, 304.
3. (a) Lo¨we, J.; Stock, D.; Zwickl, P.; Baumeister, W.;
Huber, H. Science 1995, 268, 533; (b) Groll, M.; Ditzel, L.;
Lo¨we, J.; Stock, D.; Bochtler, M.; Bartunik, H. D.; Huber,
R. Nature 1997, 386, 463.
4. (a) Iqbal, M.; Chatterjee, S.; Kauer, J. C.; Das, M.;
Messina, P. A.; Freed, B.; Biazzo, W.; Siman, R. J. Med.
Chem 1995, 38, 2276; (b) Loidl, G.; Groll, M.; Musiol, H.-
J.; Huber, R.; Moroder, L. Proc. Natl. Acad. Sci. U.S.A.
1999, 96, 5418; (c) Kisselev, A. F.; Goldberg, A. L. Chem.
Biol. 2001, 8, 739; (d) Nazif, T.; Bogyo, M. Proc. Natl.
Acad. Sci. U.S.A. 2001, 98, 2967; (e) Furet, P.; Imbach, P.;
Noorani, M.; Koeppler, J.; Lumen, K.; Lang, M.;
Guaniano, V.; Fuerst, P.; Roesel, J.; Zimmermann, J.;
Garcia-Echevevarria, C. J. Med. Chem. 2004, 47, 4810; (f)
14. The equilibrium structure of the molecule 4 was obtained
by molecular mechanics optimization (MMFF94 force
field: Halgren, T. A. The merck force field. J. Comp.
Chem. 1996, 17, 490–512, 520–552, 553–586, 587–615,
616–641). The molecule was docked to the crystal struc-
ture of the yeast 20S proteasome (Groll, M.; Koguchi, Y.;
Huber, R.; Kohno, J. J. Mol. Biol. 2001, 311, 543),
retrieved from the Protein Data Bank (PDB) archive
(www.pdb.org), using the MOE/Dock procedure inte-
grated in the MOE system of programs (Chemical
Computing Group Inc., MOE 2006.08). Before the sim-
ulation, hydrogen atoms were added to the enzyme and
the energy of the structure was minimized using the
Amber99 molecular mechanics forcefield (Ponder, J. W.;
Case, D. A. Adv. Prot. Chem. 2003, 66, 27). During the
first step of the docking application, the ligand was treated
in a flexible manner by rotating routable bonds and a
number of the ligand’s configurations were generated and

1854
A. Baldisserotto et al. / Bioorg. Med. Chem. Lett. 18 (2008) 1849–1854
scored in an effort to determine favorable binding modes
active residue. Out of 34 poses obtained, the one having
the highest score based on the estimate of the free energy
of binding of the ligand from a given pose was retained.
15. Groll, M.; Brandstetter, H.; Bartunik, H.; Bourenkow, G.;
Huber, R. J. Mol. Biol. 2003, 327, 75.
by the application of the Alpha Triangle placement
method. The ligand was docked restricting the search for
binding modes to a specific, small region of the subunit
called the b5 binding site, in which the terminal Thr1 is the