Design and synthesis of a novel class of furan-based molecules as potential 20S proteasome inhibitors

Article abstract of DOI:10.1016/j.bmcl.2006.11.020

A novel class of furan-based compounds as potential 20S proteasome inhibitors have been designed and synthesized, among which nine compounds are peptide derivatives and six molecules are statine peptidomimetics. The C-terminal furanyl moiety was introduced to target molecules as furan-based amino acids. All the compounds were obtained steadily with moderate to high yield. Compound 12 was a selective moderate potent proteasome peptidomimetic inhibitor. It inhibited HepG2 and HL-60 proliferation effectively.

Full text of DOI:10.1016/j.bmcl.2006.11.020

Bioorganic & Medicinal Chemistry Letters 17 (2007) 1102–1106
Design and synthesis of a novel class of furan-based molecules
as potential 20S proteasome inhibitors
Yiqiu Fu, Bo Xu, Xiaomin Zou, Chao Ma, Xiaoming Yang, Ke Mou, Gang Fu,
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 23 June 2006; revised 11 October 2006; accepted 7 November 2006
Available online 10 November 2006
Abstract—A novel class of furan-based compounds as potential 20S proteasome inhibitors have been designed and synthesized,
among which nine compounds are peptide derivatives and six molecules are statine peptidomimetics. The C-terminal furanyl moiety
was introduced to target molecules as furan-based amino acids. All the compounds were obtained steadily with moderate to high
yield. Compound 12 was a selective moderate potent proteasome peptidomimetic inhibitor. It inhibited HepG2 and HL-60 prolif-
eration effectively.
Ó 2006 Elsevier Ltd. All rights reserved.
The 26S proteasome identified more than 20 years ago1,2
is an unusually high molecular weight (about 700 kDa)
multicatalytic protein complex.3 The proteasome is com-
posed of the 20S catalytic core that has a cylindrical
shape with a and b subunits forming four stacking
rings,4–6 and two 19S regulatory caps which recognize
protein substrates labeled with multiple ubiquitin
chains7 and facilitate the removal of the ubiquitin mole-
cules from the substrate, and promote the unfolding of
the substrate protein as it enters the central catalytic
core.8,9 Three major proteolytic activities of proteasome
can be distinguished as trypsin-like (T-L), chymotryp-
sin-like (CT-L), and peptidyl-glutamyl peptide hydro-
lase (PGPH) activities which cleave peptide bonds on
the carboxyl side of basic, hydro-phobic, and acidic ami-
no acid residues, respectively.10 The hydroxyl group on
the side chain of Thr1 is the catalytic core of 20S protea-
some and responsible for the catalytic cleavage of
substrates through nucleophilic attack (addition-elimi-
nation mechanism).11 Besides some essential physiolog-
ical roles,12 the proteasome is also responsible for the
inappropriate or accelerated protein degradation that
occurs as a result or cause of pathological conditions.
One notable example is cancer, in which the unregulated
proteasome-mediated degradation of cell cycle regulato-
ry proteins13–15 results in accelerated and uncontrolled
mitosis, thereby promoting cancer growth and spread.16
Inhibition of this enzymatic activity with b-subunit-spe-
cific proteasome inhibitors may provide an anti-tumor
effect by inhibiting cell proliferation and angiogenesis,
and selectively inducing apoptosis of tumor cells.17,18
To date, several classes of proteasome inhibitors have
been described.11,19–27 PS-341, a peptide boronic acid,
was approved for cancer treatment for multiple myelo-
ma patients by FDA of The United States In 2003.28
However, many of these molecules cannot act as drug-
like compounds due to their high toxicity, poor selectiv-
ity and difficulties of synthesis, as well as the lack of
diversity of C-terminal groups. With the inspiration
from the inhibition mechanism of wortmannine
(Fig. 1a), a phosphatidylinositol-3 inhibitor, in which
the acylfuran reacts with a nitrogen nucleophile by an
addition-elimination process, producing a ring-opened
Michael-adduct,29 we choose the furanyl group as the
C-terminal of our new proteasome inhibitors and dipep-
tide core structure I (Fig. 1b) as parent skeleton based
on the amino acid residuals in the active site of 20S pro-
teasome. In addition, the statine structure II (Fig. 1b)
can exhibit a good recognition between the inhibitor
and proteasome.30 Furthermore, peptidomimetics are
also more stable than peptide-based molecules in vivo
and have improved transmembrane characteristics and
Keywords: Proteasome inhibitor; Peptidomimetic; Furan; Synthesis;
Antitumor.
*
Corresponding author. Tel./fax: +86 10 82801505; e-mail: pingxu@
bjmu.edu.cn
0960-894X/$ - see front matter Ó 2006 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2006.11.020

Y. Fu et al. / Bioorg. Med. Chem. Lett. 17 (2007) 1102–1106
1103
a
O
OAc
O
O
O
OAc
H3C
O
O
O
H3C
Enz-NH2
H
H
O
O
O
OH
HN
Enz
O
b
R2
HN
O
O
R1
O
H
H
N
O
N
BocHN
BocHN
N
H
OH
II
O
R1
O
R2
O
I
Figure 1. (a) The inhibitory mechanism of wortmannine. (b) The structures of two series of designed proteasome inhibitors.
bioavailability.19 Herein, we report the design and syn-
thesis of two series of furan-based compounds I and II
(Fig. 1b) as well as the activity evaluation of these novel
classes of proteasome inhibitors.
R2
R2
O
i
ii
H
OH
OCH3
N
H2N
H2N
BocHN
OCH3
O
O
R2
O
4a-c
5a-c
The synthesis of the furan-based amino acids 3 began
with the Boc-protected a-amino acid (Scheme 1). Wein-
reb amide 1 was obtained with published approach,31
and then treated with n-BuLi and furan to give the
Boc-protected furan-based amino acid 2 with high yield
(92% for 2a, 80% for 2b, and 88% for 2c). Based on the
yields and TLC data, the favorable reaction temperature
was ꢀ78 °C. Then, 2a–c were deprotected with 50%
TFA/CH2Cl2 to yield the intermediates 3a–c which were
neutralized in situ by N-methylmorpholine (NMM) and
used for next step without further purification.
O
R1
Ο
O
iv
iii
H
H
N
N
BocHN
N
BocHN
OH
H
R2
O
R2
I (7-15)
a R2=PhCH2 b R2=(CH3)2CHCH2 c R2=(CH3)2CH
O
O
6a-c
Scheme 2. The synthesis of peptide-based inhibitors. Reagents and
conditions: (i) SOCl2, MeOH, rt, 5 h; (ii) Boc-Phe-OH, N-ethyl-N0-(3-
dimethylaminopropyl)-carbodiimide hydrochloride (EDCÆHCl), 1-
hydroxybenzotriazole (HOBt), NMM, DMF, rt, 12 h; (iii) 1 N NaOH,
THF, rt, 10 h; (iv) 3, EDCÆHCl, HOBt, NMM, DMF, rt, 12 h.
The synthetic route of the target series I is described in
Scheme 2. The amino acids with hydrophobic side chain
conducted methylation to give the methyl esters 4 which
were further coupled with N-protected (S)-phenylala-
nine under typical peptide coupling conditions with
the yields of 70–80% after recrystallization (ethyl ace-
tate/petroleum ether). The obtained dipeptide esters 5
were hydrolyzed to give the acids 6, followed by the cou-
pling with 3 to form 7–15 (30–57% yields), which were
characterized by 1H NMR and MS. The results are sum-
marized in Table 1.
The series II were synthesized with the approaches de-
scribed in Scheme 3. Compound 16 was obtained from
N-protected (S)-phenylalanine,32 followed by the
oxidation with PyÆSO3, the obtained aldehyde was trans-
formed in Wittig reaction to the corresponding c-amino-
a,b-unsaturated ester 17. The trans conformation of the
double bond in 17 was proved by the coupling constant
(J = 15.6 Hz) between the two hydrogens of alkene.
Upon the treatment of c-amino olefin 17 with m-chlor-
operbenzoic acid (m-CPBA), the glycidic ester 18 was
obtained. During silica gel chromatography, about half
of inputing 17 was recovered leading to the low yield of
18 (40%), but the conversion rate of 17 to 18 was rather
high (91%). After the attack of phenylamine or benzyl-
amine, a smooth ring-opening reaction took place to
give the desired 2-phenylamino- or 2-benzylamino
substituted statine derivatives 19 and 20 (87.5% and
60% yields, respectively) which were subjected to hydro-
lysis in the presence of THF and 1 N solution of sodium
hydroxide followed by peptidyl bond formation under
R1 CH3
R1
N
O
CH3
O
i
ii
OH
O
BocHN
BocHN
BocHN
O
O
1 a-c
R1
R1
iii
H2N
O
O
2 a-c
3 a-c
a R1=(CH3)2CH b R1=(CH3)2CHCH2 c R1=CH3CH2CH(CH3)
Scheme 1. The synthesis of furan-based amino acids. Reagents and
conditions: (i) isobutyl chloroformate (IBCF), N-methylmorpholine
(NMM), HClÆHN(OCH3)CH3, CH2Cl2, 0 °C, then rt, 5 h; (ii) n-BuLi,
furan, THF, 0 °C, 30 min, then ꢀ78 °C, 2.5 h; (iii) 50% TFA
(trifluoroacetic acid)/CH2Cl2, rt, 2 h.

1104
Y. Fu et al. / Bioorg. Med. Chem. Lett. 17 (2007) 1102–1106
Table 1. Yields and melting points of peptide-based compounds 7–15
Compound
R1
R2
Yield (%)
Mp (°C)
7
8
(CH3)2CH
(CH3)2CH
(CH3)2CH
(CH3)2CH
(CH3)2CHCH2
PhCH2
39
43
30
57
45
35
49
37
45
196–197
165–166
170–172
149–151
151–153
145–147
181–182
164–165
190–192
9
10
11
12
13
14
15
(CH3)2CHCH2
(CH3)2CHCH2
(CH3)2CHCH2
CH3CH2CH(CH3)
CH3CH2CH(CH3)
CH3CH2CH(CH3)
(CH3)2CH
(CH3)2CHCH2
PhCH2
(CH3)2CH
(CH3)2CHCH2
PhCH2
pounds, 0.1% DMSO was used in the medium. The
different peptidase activities of the proteasome were as-
sayed using the following fluorogenic peptides: Suc-Leu-
leu-Val-Tyr-AMC (Suc represents succinyl and AMC
represents 7-amido-4-methylcoumarin) for chymotryp-
tic-like (ChT-L) activity; Z-Ala-Arg-Arg-AMC (Z
represents benzyloxycarbonyl) for trypsin-like activity;
Z-Leu-Leu-Glu-bNA (bNA represents b-naphthyl-
amide) for PGPH activity. MG132 (1 lM), an known
inhibitor of proteasome ChT-L activity, and 0.1%
DMSO were used as positive and solvent control. The
results in Table 3 shows that only compound 12 had
moderate potent inhibition activity (IC50 = 7.85 lM).
All the other compounds were not conspicuously
bioactive for proteasome inhibition (IC50 > 100 lM).
Furthermore, compound 12 shows no inhibitory effect
for T-L and PGPH subunit activity (IC50 > 100 lM).
Basing on these primary bioactive data, the cell growth
inhibition effects of compound 12 on human hepatoma
cell line (HepG2) and acute human myeloid leukemic
cell line (HL-60) were further tested using a standard
MTT or SRB-based colorimetric assay. IC50 values were
34.2 and 37.07 lM for HepG2 and HL-60, respectively.
i, ii
iii, iv
OH
OH
OEt
BocHN
BocHN
BocHN
O
O
16
17
NHR2
OEt
v
vi
OEt
BocHN
BocHN
O
OH
O
O
19 R2=PhCH2NH
20 R2=PhNH
18
R2
NHR2
OH
HN
O
vii
H
viii
N
O
BocHN
BocHN
OH
O
R1
OH
O
21 R2=PhCH2NH
22 R2=PhNH
II (23-28)
Scheme 3. The synthesis of statine-based inhibitors. Reagents and
conditions: (i) CH3I, NaHCO3, DMF, rt, 5 h; (ii) CaCl2, NaBH4,
EtOH/THF, 0 °C then rt, 5 h; (iii) PyÆSO3, Et3N, CH2Cl2/DMSO,
ꢀ10 °C then rt, 30 min; (iv) Ph3PCHCOOEt, toluene, 80 °C, 1 h; (v) m-
CPBA, CH2Cl2, 5 days; (vi) benzylamine or phenylamine, 60 °C, 3
days; (vii) 1 N NaOH, THF, 10 h; (viii) 3, EDCÆHCl, HOBt, NMM,
DMF, rt, 12 h.
The molecular modeling of inhibitor binding was done
to compare the binding modes of MG132 and com-
pound 12 to proteasome (Fig. 2). The distance between
the furanyl group of 12 and the Thr1 amino group was
˚
7.37 A (Fig. 2b), which was very close to the distance
from the aldehyde carbonyl group of MG132 to the
common conditions. The yields of series II compounds
23–28 were 24–60%. The structures of 23–28 were con-
firmed by the H NMR and MS data. Table 2 shows
˚
Thr1 Oc (6.03 A, Fig. 2d). This demonstrated that the
1
active centers of the two inhibitors entered the same
pocket when they bound to proteasome. As shown in
Figure 2a, the side-chain substitute groups of 12 fitted
into the individual pocket of the 20S proteasome,
among which the benzyl groups showed good recogni-
tion, and this gave a possible explanation to the fact that
compound 12 showed inhibitory activity, and also dem-
onstrated that the P1 and P2 substitutions of molecules
the results of series II.
The proteasome chymotryptic-like (ChT-L) activity
inhibitory effects of compounds 7–15 and 23–28, and
also the trypsin-like (T-L), peptidyl-glutamyl peptide
hydrolase (PGPH) activities of compound 12 were as-
sayed in vitro. Due to the poor solubility of these com-
Table 2. Yields and melting points of statine peptidomimetics 23–28
Compound
R1
R2
Yield (%)
Mp (°C)
23
24
25
26
27
28
(CH3)2CH
PhCH2
PhCH2
PhCH2
Ph
37
60
30
25
26
24
147–148
77–78
67–69
(CH3)2CHCH2
CH3CH2CH(CH3)
(CH3)2CH
181–182
206–208
189–190
(CH3)2CHCH2
CH3CH2CH(CH3)
Ph
Ph

Y. Fu et al. / Bioorg. Med. Chem. Lett. 17 (2007) 1102–1106
1105
Table 3. Proteasome inhibition activities
requirement at P2 cannot be disregarded in inhibitor de-
sign. Furthermore, the lack of acidic or basic residues in
12 led to the selectivity of CT-L activity of proteasome
over the T-L or PGPH activity. According to the pub-
lished literatures, statine isostere derivatives can act as
good proteasome inhibitors, but the diminution of activ-
ity of 23–28 showed that the skeleton variation cannot
be ignored when the inhibition mechanism changes
(covalent vs noncovalent). Besides, the N-terminal tert-
butoxycarbonyl group may be an unfavorable N-pro-
tecting group for the proteasome binding.
Compound
IC50 (ChT-L)/lM
7
8
>100
>100
>100
>100
>100
7.85
9
10
11
12a
13
14
15
23
24
25
26
27
28
>100
>100
>100
>100
>100
>100
>100
>100
>100
In conclusion, a new class of potential proteasome
inhibitors with two different parent skeletons was de-
signed and synthesized. The C-terminal furanyl substi-
tution may contribute to the inhibition of the activity
of 20S proteasome with unique mechanism. Except for
the typical peptide coupling procedures which yielded
the series of peptide-based target molecules, the syn-
thetic routes of statine-based peptidomimetics and fur-
an-based amino acids were also described in detail. All
the reaction steps were demonstrated to be easily per-
formed with moderate or high yield. The primary
evaluation of proteasome inhibitory activity and tu-
mor cell proliferation inhibition effect of these com-
pounds showed that compound 12 was a selective
and moderate potent proteasome ChT-L activity
inhibitor, which inhibited HepG2 and HL-60 prolifer-
ation effectively.
a No inhibitory activity for T-L and PGPH subunit activity
(IC50 > 100 lM).
were important for the potency of inhibitors. Neither
valine nor isoleucine residue was suitable for the S1
pocket of b5 subunit, whereas the isobutyl group can
interact with the Met45 well,33 which partly contributed
to the activity of 12. The P2 substitution was once
thought not important for the recognition between
inhibitor and proteasome, but the activity loss of 10
and 11 compared to 12 showed that specific binding
Figure 2. (a) Binding mode of compound 12. (b) The distance between the 50-C of furanyl group and amino group of proteasome Thr1. (c) Binding
mode of MG-132. (d) The distance between the aldehyde carbonyl and proteasome Thr1 Oc. The two inhibitors were 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).

1106
Y. Fu et al. / Bioorg. Med. Chem. Lett. 17 (2007) 1102–1106
Brand, S.; Elliott, P. J.; Lazarus, D.; McCormack, T.;
Acknowledgments
Parent, L.; Stein, R.; Adams, J. Proc. Natl. Acad. Sci.
U.S.A. 1998, 95, 15671.
16. Pagano, M.; Tam, S. W.; Tehodoras, A. M.; Beer-
Romero, P.; Del Sal, G.; Chau, V.; Yew, P. R.; Draetta,
G. F.; Rolfe, M. Science 1995, 269, 682.
This work was supported by the National Natural Sci-
ence Foundation of China (20272004 and 20572006),
and 985 program, Ministry of Education of China.
17. Oikawa, T.; Sasaki, T.; Nakamura, M.; Shimamura, M.;
Tanahashi, N.; Mura, S.; Tanaka, K. Biochem. Biophys.
Res. Commun. 1998, 246, 243.
Supplementary data
18. Drexler, H. C.; Risau, W.; Konerding, M. A. FASEB J.
2000, 14, 65.
Supplementary data associated with this article can be
found, in the online version, at doi:10.1016/j.bmcl.
2006.11.020.
19. Delcros, J. G.; Baudy Floc’h, M.; Prigent, D.; Arlot-
Bonnemains, Y. Curr. Med. Chem. 2003, 10, 479.
20. Fenteany, G.; Standaert, R. F.; Lane, W. S.; Choi, S.;
Corey, E. J.; Schreiber, S. L. Science 1995, 268, 726.
21. Kohno, J.; Koguchi, Y.; Nishio, M.; Nakao, K.; Kuroda,
M.; Schimizu, R.; Ohnuki, T.; Komatsubara, S. J. Org.
Chem. 2000, 65, 990.
References and notes
1. Hershko, A. Trends Biochem. Sci. 1996, 21, 445.
2. Etlinger, J. D.; Goldberg, A. L. Proc. Natl. Acad. Sci.
U.S.A. 1977, 74, 54.
22. Vitinsky, A.; Michaud, C.; Powers, J. C.; Orlowski, M.
Biochemistry 1992, 31, 9421.
3. Coux, O.; Tanaka, K.; Goldberg, A. L. Annu. Rev.
Biochem. 1996, 65, 801.
23. Meng, L. H.; Kwok, B. H. B.; Sin, N.; Crews, C. M.
Cancer Res. 1999, 59, 2798.
4. Lo¨we, J.; Stock, D.; Jap, B.; Zwickl, P.; Baumerster, W.;
Hubert, R. Science 1995, 268, 533.
24. Lynas, J. F.; Harriott, P.; Healy, A.; McKervey, M. A.;
Walker, B. Bioorg. Med. Chem. Lett. 1998, 8, 373.
25. Bogyo, M.; Shin, S.; McMaster, J. S.; Ploegh, H. L. Chem.
Biol. 1998, 5, 307.
26. Adams, J. Curr. Opin. Chem. Biol. 2002, 6, 493.
27. Nam, S.; Smith, D. M.; Dou, Q. P. J. Biol. Chem. 2001,
276, 13322.
5. Groll, M.; Ditzel, L.; Lo¨we, J.; Stock, D.; Bocheler, M.;
Bartunik, H. D.; Huber, R. Nature 1997, 286, 463.
6. Unno, M.; Mizushima, T.; Motimoto, Y.; Tomisugi, Y.;
Tanaka, K.; Yasuoka, N.; Tsukiharal, T. Structure 2002,
10, 609.
7. Voges, D.; Zwickl, P.; Baumeister, W. Annu. Rev.
Biochem. 1999, 68, 1015.
8. Larsen, C. N.; Finley, D. Cell 1997, 91, 431.
28. 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.
29. 17Powis, G.; Bonjouklian, R.; Berggren, M. M.; Galegos,
A.; Abraham, R.; Ashendel, C.; Zalkow, L.; Matter, W.
F.; Dodge, J.; Grindey, G. Cancer Res. 1994, 54, 2419.
9. Baumeister, W.; Walz, J.; Zuhl, F.; Seemuller, E. Cell
1998, 92, 367.
¨
¨
10. Orlowski, M.; Wilk, S. Biochem. Biophys. Res. Commun.
1981, 101, 84.
11. Kisselev, A. F.; Goldberg, A. L. Chem. Biol. 2001, 8, 739.
12. Goldberg, A. L.; Stein, R.; Adams, J. Chem. Biol. 1995, 2,
503.
30. Furet, P.; Imbach, P.; Furst, P.; Lang, M.; Noorani, M.;
¨
Zimmermann, J.; Garc´ıa-Echeverr´ıa, C. Bioorg. Med.
Chem. Lett. 2001, 11, 1321.
13. Koepp, D. M.; Harper, J. W.; Elledge, S. J. Cell 1999, 97,
431.
31. Zhang, X.; Zou, X. M.; Xu, P. Synth. Commun. 2005, 35,
1881.
14. Hershko, A.; Ciechanover, A. Annu. Rev. Biochem. 1998,
67, 425.
15. Palombella, V. J.; Conner, E. M.; Fuseler, J. W.; Destree,
A.; Davis, J. M.; Laroux, F. S.; Wolf, R. E.; Huang, J.;
32. Luly, J. R.; Dellaria, J. F.; Plattner, J. J.; Soderquist, J.;
Yi, L. N. J. Org. Chem. 1987, 52, 1487.
33. Groll, M.; Berkers, C.; Ploegh, H. L.; Ovaa, H. Structure
2006, 14, 451.