Novel CADD-based peptidyl vinyl ester derivatives as potential proteasome inhibitors

Article abstract of DOI:10.1016/j.bmcl.2007.12.077

A series of peptidyl vinyl ester derivatives bearing three different P1 substitutions as potential proteasome inhibitors were studied. The target molecules were designed based on CADD (computer aided drug design) protocol and synthesized. Their activities

Full text of DOI:10.1016/j.bmcl.2007.12.077

Available online at www.sciencedirect.com
Bioorganic & Medicinal Chemistry Letters 18 (2008) 2198–2202
Novel CADD-based peptidyl vinyl ester derivatives
as potential proteasome inhibitors
Ke Mou, Bo Xu, Chao Ma, Xiaoming Yang, Xiaomin Zou, Yang Lu and Ping Xu*
¨
Department of Medicinal Chemistry, State Key Laboratory of Natural and Biomimetic Drugs,
School of Pharmaceutical Sciences, Peking University, Beijing 100083, PR China
Received 22 October 2007; revised 30 November 2007; accepted 14 December 2007
Available online 2 February 2008
Abstract—A series of peptidyl vinyl ester derivatives bearing three different P1 substitutions as potential proteasome inhibitors were
studied. The target molecules were designed based on CADD (computer aided drug design) protocol and synthesized. Their
activities toward proteasome and four human cancer cell lines (including hepatoma cell line (Bel-7402), myeloid leukemic cell line
(HL-60), gastric cancer cell line (BGC-823) and nasopharyngeal cancer cell line (KB)) were tested using fluorescence assay. Two
compounds showed proteasome inhibitory activities, and four compounds showed weak antiproliferative activities toward HL-60
and BGC-823.
Ó 2008 Elsevier Ltd. All rights reserved.
The ubiquitin–proteasome pathway (UPP) is one of the
two ways cells utilize for protein degradation, in which
the 26S proteasome complex makes a functional
machinery where hydrolysis of targeted proteins takes
place.1 It consists of a 20S core catalytic cylindrical com-
plex capped at both ends by 19S regulatory subunits.
The 20S proteasome is formed by four stacked rings,
with each of the two inner rings composed of seven dif-
ferent b subunits.2–4 b1, b2, and b5 subunits contain the
post-acidic-like (PGPH), trypsin-like (T-L), and chymo-
trypsin-like (ChT-L) active sites where peptide bonds are
cleaved on the carboxyl side of acidic, basic and hydro-
phobic amino acid residues, respectively.5–7 Substrates
of 20S proteasome are intracellular proteins, including
cyclins, caspases, BCL2 and nuclear factor of jB (NF-
jB) , representative of mediators during cell-cycle pro-
gression and apoptosis. A large body of experiments
have already proven that the effects of these proteins
are regulated by 20S proteasome. Therefore, it is pro-
posed to be a new target for antineoplastic and anti-
inflammatory therapy.8 To date, two compounds have
entered the clinic as antineoplastic (PS-341, also known
as bortezomib and Velcade, a peptidyl boronic acid ap-
proved for the treatment of multiple myeloma patients
by FDA in 2003) and anti-inflammatory drugs (PS-
519, also known as MLN519).9,10
Currently, the reported proteasome inhibitors included
peptidyl aldehydes, a-ketoamides, peptidyl boronic
acids, a0,b0-epoxyketones, and non-covalent binding
inhibitors, etc.,11–17 all of which targeted the chymotryp-
sin-like (b5) subunit because of its importance on pro-
tein hydrolysis.18–20 Among them peptidyl boronic
acids are the most potent and a0,b0-epoxyketones are
the most selective.21 However, Mauro Marastoni and
coworkers22 argued that the tripeptide vinyl ester deriv-
atives also had the trypsin-like (b2) inhibitory activity.
After modifications, Hmb-Val-Gln-Leu-VE (Hmb:
3-hydroxy-2-methylbenzoyl) was found to be the most
potent compound toward the b2 subunit, and was about
500-fold less selective for the b5 subunit.
Peptides containing a Michael acceptor are known
irreversible cysteine protease inhibitors.23 They have
already been proven to inhibit proteasome by forming
a Michael adduct with the catalytic Thr1 of proteasome
(Fig. 1) in an irreversible manner. We adopted the
CADD methodology (SYBYL 7.1, DOCK 4.0 and
Pymol 0.98) to design new peptidyl vinyl ester derivatives,
hopefully with enhanced b5 subunit inhibitory potency.
All binding modes are represented via Pymol 0.98.
Keywords: Binding mode; CADD; Synthesis; Bioactivity; Proteasome
inhibitor; Peptidomimetic; Peptidyl vinyl ester derivatives.
*
First, we used Z-Leu-Leu-Leu-VE (designated as M1) as
a model compound to investigate the binding mode.
Corresponding author. Tel./fax: +86 10 82801505; e-mail:
pingxu@bjmu.edu.cn
0960-894X/$ - see front matter Ó 2008 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2007.12.077

K. Mou et al. / Bioorg. Med. Chem. Lett. 18 (2008) 2198–2202
2199
nal exchanged their pockets (M1 had two hydrogen
bonds with Thr1 and Asp125, and the distance was
˚
4.58 A, Fig. 2B), and this special binding mode was also
seen in non-covalent inhibitors.24 The binding energy of
mode a was lower than that of mode b, suggesting its pre-
dominance in binding. Then, three larger groups, benzyl
group (C1), 4-hydroxyl benzyl group (C2) and (3S)-2-
oxo-3-pyrrolidinyl-methyl group (C3), were designed
to replace the isobutyl group at the P1 position
(Fig. 3), and another three model compounds were used
in binding, Z-Leu-Leu-Phe-VE (M2 with C1 group), Z-
Leu-Leu-Tyr-VE (M3 with C2 group) and Z-Leu-Leu-
MK-VE (M4 with C3 group) (the order of the size of
C-terminal groups is: isobutyl group < C3 < C1 < C2).
Interestingly, for M4 and M2 (with more bulky C-termi-
nal groups), the conventional binding modes were ob-
served (M4 had two hydrogen bonds with Thr21 and
Figure 1. Vinyl ester inhibitors: potential interactions with proteasome
20S catalytic subunits.
Two major binding modes were observed (mode a and
mode b). Mode a was the conventional binding mode
(M1 had two hydrogen bonds with Thr1 and Ser129,
and the distance between the catalytic Thr1 and its reac-
˚
tion site was 5.24 A, Fig. 2A), while mode b was a
reversed one in which the P1 side chain and the C-termi-
Figure 2. Binding modes of M1 to M4 docked into the b5 and b6 subunits of the bovine proteasome structure 1IRU using the AMBER_Parm99
force field in DOCK (UCSF, San Francisco, CA) as rendered in PyMOL (DeLano Scientific, South San Francisco, CA). (A) Binding mode of M1,
mode a; (B) Binding mode of M1, mode b; (C) Binding mode of M4; (D) Binding mode of M2; (E) Binding mode of M3; (F) S1 pocket is composed by
Met45, Gly47 and Ala49, and the adjacent cavity is composed by Thr1, Ala20 and Thr21.

2200
K. Mou et al. / Bioorg. Med. Chem. Lett. 18 (2008) 2198–2202
3,27 followed by reduction with NaBH4 and CaCl2 in
ethanol and THF (1:0.6) to give 4.25 One-pot reaction
integrating oxidation of 4 by PyÆSO3 to aldehyde 5
and subsequent reaction with the ylide Ph3P@CHCO2Et
afforded the vinyl ester 6 (yield for this one-pot reaction
is about 76%).26 At last, 6 was deprotected with 50%
TFA/CH2Cl2 and then coupled with different amino
acids, dipeptides and tripeptides under typical peptide
coupling conditions using HOBt and EDC to obtain
the final products.
Figure 3. Newly designed C-terminus.
The different peptidase activities of the proteasome were
assayed using the following fluorogenic peptides: Suc-
Leu-leu-Val-Tyr-AMC for chrymotryptic-like (ChT-L)
activity; Z-Ala-Arg-Arg-AMC for trypsin-like (T-L)
activity and Z-Leu-Leu-Glu-bNA for post-acidic-like
(PGPH) activity (Suc for succinyl and AMC for 7-ami-
do-4-methylcoumarin).
˚
Ala49, and the distance was 4.96 A, while M2 had four
hydrogen bonds with Thr21, Ala49, Ser129 and Ser130,
˚
respectively, and the distance was 4.24 A, Fig. 2C and
D); however, in case of M3 (with the largest C-terminal
group), the reversed mode was the only observation (M3
had two hydrogen bonds with Thr21, and the distance
˚
was 3.89 A, Fig. 2E).
Cell growth inhibition effects of final products on four
kinds of human cancer cell lines (including hepatoma
cell line (Bel-7402), myeloid leukemic cell line (HL-60),
gastric cancer cell line (BGC-823) and nasopharyngeal
cancer cell line (KB)) were tested through a standard
MTT or SRB-based colorimetric assay.
It was clear that S1 pocket is smaller than the cavity next
to it (Fig. 2F), and when P1 side chain became larger,
the reversed binding mode would be favored. In order
to study the potency and selectivity of this special bind-
ing mode toward b5 subunit, a series of compounds with
these C-terminus were synthesized and their inhibitory
activities were compared (C1 and C3 represented the
conventional binding mode while C2 represented the
reversed one).
Structures of the final products and their inhibitory
activities are listed in Table 1. The data revealed that
tetrapeptides with C3 terminus had better proteasome
inhibitory activities: two compounds (32 and 35) had
the best activities (IC50 values were 48.31 and
63.41 lm toward the b5 subunit, respectively), but they
showed poor inhibitory activities in the T-L active site.
For cell proliferative activity, smaller inhibitors exhib-
ited better effects: dipeptides had a mean antiprolifera-
tive rate of more than 20% toward HL-60, while
tripeptides and tetrapeptides were about 10%, but one
tripeptide, compound 27, was an exception. By compar-
ing the three terminus (compounds 17–31), C3 had the
best activity (compounds 27 and 28 had antiprolifera-
tive rates of 36.97% and 31.04%), C2 had a moderate
activity (compounds 24 and 26 had antiproliferative
rates of 26.73% and 26.50%), and C1 had the lowest
Herein, 30 peptidyl vinyl ester derivatives with three dif-
ferent P1 side chains (Fig. 3) were synthesized according
to standard solution-phase peptide chemistry. All amino
acids used are of L configuration. Compound 32 was
synthesized as an example shown in Scheme 1. Boc-pro-
tected Glutamic acid 1 was treated with CH3I under
NaHCO3 and DMF to give the dimethyl ester,25 which
was then reacted with BrCH2CN in the presence of lith-
ium hexamethyldisilazide (LiHMDS) at ꢀ78 °C to af-
ford the acetonitrile-substituted product 2 in a very
high yield (crude yield for the first two steps is over
96%).26 Compound 2 was then reduced by NaBH4 and
CoCl2Æ6H2O in methanol to give the cyclic product
Scheme 1. Reagents and conditions: (a) NaHCO3, CH3I, DMF; (b) BrCH2CN, LiHMDS, ꢀ78 °C, 3–5 h; (c) NaBH4, CoCl2Æ6H2O, in methanol; (d)
NaBH4, CaCl2 in ethanol and THF; (e) PyÆSO3, TEA, DCM/DMSO; (f) Ph3P@CHCO2Et; (g) TFA, CH2Cl2; (h) Boc-Leu-Leu-Leu-OH, EDC,
HOBt, NMM , DMF.

K. Mou et al. / Bioorg. Med. Chem. Lett. 18 (2008) 2198–2202
2201
Table 1. Structure, proteasome inhibitory activity and antiproliferative rate of the peptidyl vinyl ester derivatives
No.
Compound
Proteasome inhibitory activity
(IC50, lm)
Antiproliferative rate (% at 10 lm)
ChT-L
T-L
PGPH
HL-60a
BGC-823b
Bel-7402b
KBb
7
8
Boc-Ile-C1
Boc-Leu-C1
>100
>100
>100
>100
>100
>100
>100
>100
>100
>100
>100
>100
>100
>100
>100
>100
>100
>100
>100
>100
>100
>100
>100
>100
>100
48.31
>100
>100
63.41
>100
>100
>100
>100
>100
>100
>100
>100
>100
>100
>100
>100
>100
>100
>100
>100
>100
>100
>100
>100
>100
>100
>100
>100
>100
>100
>100
>100
>100
>100
>100
>100
>100
>100
>100
>100
>100
>100
>100
>100
>100
>100
>100
>100
>100
>100
>100
>100
>100
>100
>100
>100
>100
>100
>100
>100
>100
>100
>100
>100
>100
29.56
33.88
25.73
27.13
19.98
21.66
5.87
24.14
30.21
20.90
21.27
19.11
23.06
17.01
33.88
18.44
22.56
9.95
ꢀ7.74
1.55
ꢀ0.18
ꢀ6.67
ꢀ4.66
ꢀ1.97
ꢀ11.24
5.27
9
Boc-Phe-C1
Boc-Val-C1
ꢀ3.61
ꢀ3.57
ꢀ3.14
3.32
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
Boc-Ser-C1
Boc-Ile-C2
Boc-Leu-C2
Boc-Phe-C2
ꢀ10.27
12.68
4.60
1.72
ꢀ0.70
1.76
23.66
22.10
19.26
6.13
Boc-Val-C2
Boc-Ser-C2
ꢀ5.91
ꢀ12.87
ꢀ8.04
ꢀ11.09
ꢀ4.56
ꢀ1.79
ꢀ6.95
3.39
ꢀ15.49
ꢀ1.67
5.23
Boc-Leu-Leu-C1
Boc-Phe-Ser-C1
Boc-Val-Ser-C1
Boc-Leu-Phe-C1
Boc-Ile-Thr-C1
Boc-Leu-Leu-C2
Boc-Phe-Ser-C2
Boc-Val-Ser-C2
Boc-Leu-Phe-C2
Boc-Ile-Thr-C2
Boc-Leu-Leu-C3
Boc-Phe-Ser-C3
Boc-Val-Ser-C3
Boc-Leu-Phe-C3
Boc-Ile-Thr-C3
Boc-Leu-Leu-Leu-C3
Boc-Phe-Ser-Leu-C3
Boc-Val-Ser-Leu-C3
Boc-Leu-Leu-Tyr-C3
Boc-Ile-Thr-Leu-C3
21.08
0.36
8.65
8.28
0.04
ꢀ6.79
4.33
13.32
13.62
0.02
8.17
10.75
ꢀ5.99
7.44
ꢀ2.64
13.37
26.73
7.87
10.28
22.59
13.74
14.69
21.44
16.90
5.52
3.39
ꢀ4.24
3.10
13.09
ꢀ5.34
3.76
26.50
36.97
31.04
ꢀ1.94
ꢀ7.61
0.71
ꢀ2.40
1.66
ꢀ14.03
ꢀ15.60
ꢀ3.45
ꢀ1.56
ꢀ10.29
15.73
ꢀ4.17
9.15
ꢀ4.98
ꢀ1.63
ꢀ16.67
12.02
ꢀ0.39
ꢀ3.55
7.64
13.89
0.69
17.17
4.94
16.14
14.25
20.55
28.90
19.98
ꢀ47.14
17.73
ꢀ2.32
6.64
4.30
ꢀ0.31
activity (only compound 18 had an antiproliferative
rate of 21.08%, whereas others were less than 9%) to-
ward HL-60. Among the four human cancer cell lines,
these inhibitors had better antiproliferative potencies
toward HL-60 (the highest antiproliferative rate was
36.97%, compound 27) and BGC-823 (the highest anti-
proliferative rate was 33.88%, compound 14), but
weaker effects toward Bel-7402 (the highest antiprolifer-
ative rate was 12.68%, compound 14) and KB (the
highest antiproliferative rate was 15.73%, compound
32). Within these inhibitors, compounds 8, 14, 27 and
28 showed weak antiproliferative activities (antiprolifer-
ative rates were more than 30%) toward human cancer
cell lines (HL-60 and BGC-823).
The fact is that when the length of the peptide in-
creased, more interactions could be expected between
the inhibitor and the enzyme, which would lead to a
better proteasome inhibitory activity, however, at the
same time, membrane permeability decreased dramati-
cally which might account for the relatively poor cell
antiproliferative activity. In addition, low water solu-
bility is another problem. Therefore, modifications of
the N-terminal to increase the water medium solubility
and replacement of the peptides with peptidomimetic
structures to improve the membrane permeability
are now in progress and will be reported in due
course.
In summary, a virtual binding model of peptidyl vinyl
ester derivatives with the b5 catalytic center of protea-
some was proposed for the first time in this work.
Based on this model, a series of potential proteasome
inhibitors bearing vinyl ester as the C-terminal phar-
macophore were designed and synthesized. Covalent
Michael adducts can be formed between inhibitors
and the catalytic Thr1, which might cause the inhibi-
tion of proteasome. The proteasome inhibitory and
antiproliferative activities of these compounds toward
four human cancer cell lines were tested. Tetrapeptides
showed better proteasome inhibitory activities, while
dipeptides showed better antiproliferative activities.
Acknowledgments
This work was supported by the National Natural Sci-
ence Foundation of China (20572006, 20772008 and
30772650), and 985 program, Ministry of Education of
China.
Supplementary data
Supplementary data associated with this article can be
found, in the online version, at doi:10.1016/
j.bmcl.2007.12.077.

2202
K. Mou et al. / Bioorg. Med. Chem. Lett. 18 (2008) 2198–2202
16. Elfsson, M.; Splittgerber, U.; Myung, J.; Mohan, R.;
Crews, C. M. Chem. Biol. 1999, 6, 811.
References and notes
17. Garcia-Echeverria, C.; Imbach, P.; France, D.; Furst, P.;
Lang, M.; Noorani, M.; Scholz, D.; Zimmermanna, J.;
Fureta, P. Bioorg. Med. Chem. Lett. 2001, 11, 1317.
18. Chen, P.; Hochstrasser, M. Cell 1996, 86, 961.
19. Rock, K. L.; Gramm, C.; Rothstein, L.; Clark, K.; Stein,
R.; Dick, L.; Hwang, D.; Goldberg, A. L. Cell 1994, 78,
761.
20. Heinemeyer, W.; Fischer, M.; Krimmer, T.; Stachon, U.;
Wolf, D. H. J. Biol. Chem. 1997, 272, 25200.
21. Groll, M.; Kim, K. B.; Kairies, N.; Huber, R.; Crews, C.
M. J. Am. Chem. Soc. 2000, 122, 1237.
1. Ciechanover, A. Cell 1994, 9, 13.
2. Lo¨we, J.; Stock, D.; Jap, B.; Zwickl, P.; Baumerster, W.;
Hubert, R. Science 1995, 268, 533.
3. Groll, M.; Ditzel, L.; Lo¨we, J.; Stock, D.; Bocheler, M.;
Bartunik, H. D.; Huber, R. Nature 1997, 286, 463.
4. Unno, M.; Mizushima, T.; Motimoto, Y.; Tomisugi, Y.;
Tanaka, K.; Yasuoka, N.; Tsukiharal, T. Structure 2002,
10, 609.
5. Orlowski, M.; Wilk, S. Biochem. Biophys. Res. Commun.
1981, 101, 84.
6. Kisselev, A. F.; Akopian, T. N.; Castillo, V.; Goldberg, A.
L. Mol. Cell. 1999, 4, 395.
22. Marastoni, Mauro; Baldisserotto, Anna; Cellini, Silvia;
Gavioli, Riccardo; Tomatis, Roberto J. Med. Chem. 2005,
48, 5038.
7. Nam, S.; Smith, D. M.; Dou, Q. P. J. Biol. Chem. 2001,
276, 13322.
23. Powers, James C.; Asgian, Juliana L.; Ekici, Ozlem
Dogan; James, Karen Ellis Chem. Rev. 2002, 102, 4639.
24. Furet, P.; Imbach, P.; Furst, P.; Lang, M.; Noorani, M.;
Zimmermann, J.; Garcia-Echeverria, C. Bioorg. Med.
Chem. Lett. 2001, 11, 1321.
25. Luly, Jay R.; Dellaria, Joseph F.; Plattner, Jacob J.;
Soderquist, Jeffrey L.; Yi, Nwe J. Org. Chem. 1987, 52,
1487.
26. Tian, Qingping; Nayyar, Naresh K.; Babu, Srinivasan;
Chen, Lijian; Tao, Junhua; Lee, Steven; Tibbetts,
Anthony; Moran, Terence; Liou, Jason; Guo, Ming;
Kennedy, Timothy P. Tetrahedron Lett. 2001, 42, 6807.
27. Amruta Reddy, P.; Hsiang, Bonnie C. H.; Latifi, Tammy
N.; Hill, Matthew W.; Woodward, Karen E.; Rothman,
Steven M.; Ferrendelli, James A.; Covey, Douglas F. J.
Med. Chem. 1996, 39, 1898.
8. Adams, Julian Nat. Rev. Cancer 2004, 4, 349.
9. Adams, J.; Behnke, M.; Chen, S.; Cruickshank, A. A.; Dick,
L. R.; Grenier, L.; Klunder, J. M.; Ma, Y. T.; Plamonda, L.;
Stein, R. L. Bioorg. Med. Chem. Lett. 1998, 8, 333.
10. Shah, I. M.; Lees, K. R.; Pien, C. P.; Elliott, P. J. Br. J.
Clin. Pharmacol. 2002, 54, 269.
11. Vinitsky, A.; Cardozo, C.; Sepp-Lorenzino, L.; Michaud,
C.; Orlowski, M. J. Biol. Chem. 1994, 269, 29860.
12. Lynas, J. F.; Harriott, P.; Healy, A.; McKervey, M. A.;
Walker, B. Bioorg. Med. Chem. Lett. 1998, 8, 373.
13. Chatterjee, S.; Dunn, D.; Mallya, S.; Ator, M. A. Bioorg.
Med. Chem. Lett. 1999, 9, 2603.
14. Wu, S.; Waugh, W.; Stella, V. J. J. Pharma. Sci. 2000, 89,
758.
15. Wu, Yulian; Han, Bing; Luo, Hongyu; Shi, Guixiu; Wu,
Jiangping Transplantation 2004, 78, 360.