Structure-activity relationship study of syringolin A as a potential anticancer agent

Article abstract of DOI:10.1016/j.bmcl.2015.06.015

A detailed structure-activity relationship of syringolin A (1), which is a promising antitumor natural product, was described. We previously developed syringolin A analog 2 as a potent proteasome inhibitor by the structure-based drug design of syringolin A. In this Letter, we synthesized a range of analogs of 2, having a different length of the lipophilic chain and substituted aryl group, and their cytotoxicity against human cancer cells was evaluated. It turned out that these modifications greatly affected the cytotoxicity. Further optimization would lead to develop a novel proteasome inhibitor.

Full text of DOI:10.1016/j.bmcl.2015.06.015

Bioorganic & Medicinal Chemistry Letters xxx (2015) xxx–xxx
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
Structure–activity relationship study of syringolin A as a potential
anticancer agent
Takuya Chiba a, Akira Matsuda a,b, Satoshi Ichikawa a,b,
⇑
a Faculty of Pharmaceutical Sciences, Hokkaido University, Kita-12, Nishi-6, Kita-ku, Sapporo 060-0812, Japan
b Center for Research and Education on Drug Discovery, Hokkaido University, Kita-12, Nishi-6, Kita-ku, Sapporo 060-0812, Japan
a r t i c l e i n f o
a b s t r a c t
Article history:
A detailed structure–activity relationship of syringolin A (1), which is a promising antitumor natural
product, was described. We previously developed syringolin A analog 2 as a potent proteasome inhibitor
by the structure-based drug design of syringolin A. In this Letter, we synthesized a range of analogs of 2,
having a different length of the lipophilic chain and substituted aryl group, and their cytotoxicity against
human cancer cells was evaluated. It turned out that these modifications greatly affected the cytotoxicity.
Further optimization would lead to develop a novel proteasome inhibitor.
Received 27 April 2015
Revised 29 May 2015
Accepted 2 June 2015
Available online xxxx
Keywords:
Proteasome inhibitor
Cancer
Ó 2015 Elsevier Ltd. All rights reserved.
Structure–activity relationship
The eukaryotic 20S proteasome is an enzyme, which degrades
ubiquitin-labeled proteins, and responsible for maintenance of an
intracellular protein expression.1 The barrel-shaped proteasome is
HN
HN
O
O
H
N
H
N
composed of seven
a and seven b subunits in a a7b7b7a7
O
O
N
H
N
H
CO2H
N
H
arrangement, and b1, b2 and b5 subunits have a caspase-like, a
trypsin-like and a chymotrypsin-like activities, respectively. After
poly-ubiquitin-labeled proteins are taken up to a hole inside the
proteasome, the proteins were hydrolyzed by proteasomal catalytic
active sites. The inhibition of the proteasome results in accumula-
tion of unnecessary proteins and ultimately causes cell death.2 As
a proteasome inhibitor, bortezomib (VelcadeÒ)3 and carfilzomib
(KyprolisÒ)4 have been approved by U.S. Food and Drug
Administration (FDA) for a treatment of multiple myeloma, and
therefore, the proteasome is considered to be an attractive target
for the development of anticancer agent. However, bortezomib
and carfilzomib have several shortcomings such as side effects and
their drug resistance,5 and there is an urgent need for the next
generation proteasome inhibitor. Syringolin A (1, Fig. 1) is a natural
product isolated from a strain of the plant pathogen Pseudomonas
O
O
O
O
N
H
N
H
1
2
syringolin A ( )
β
β
5 subunit)
Ki' = 0.14 nM (
IC50 = 2.2 nM (RPMI8226 cells)
Ki' = 843 nM ( 5 subunit)
IC50 = 8500 nM (MM1.S cells)
Figure 1. Structure and biological activities of syringolin A and its analog 2.
In 2008, it was identified as a virulence factor by inhibiting
the 20S proteasome.7 Syringolin A reacts irreversibly with the
N-terminal threonine (Thr) of the active site of the b5 subunit
by a 1,4-addition of the hydroxyl group of the Thr to the
a
,b-unsaturated carboxamide moiety of syringolin A. Syringolin A
exhibits a moderate b5 subunit inhibitory activity with the Ki0
value of 0.8 M and a weak b1/b2 subunits inhibitory activity. Its
cytotoxicity is also weak with an IC50 value of 8.5 M for human
myeloma MM1.S cells in vitro. Several studies have been con-
ducted by Kaiser’s group in order to improve its biological activity.8
Our group has accomplished the total synthesis of syringolin A, and
its structure–activity relationship has also been investigated. We
have found the compound 2, where the ureadipeptide moiety
l
syringae pv. Syringae in 1998.6 Its structure consists of
highly strained 12-membered macrolactam, which contains
a
l
b,c-dehydrolysine, and a ureadipeptide side chain.
Abbreviations: Fmoc, 9-fluorenylmethyloxycarbonyl; EDCI, 1-ethyl-3-(3-
dimethylaminopropyl)carboxamide;
HOBt,
1-hydroxybenzotriazole;
DMF,
was replaced with an N-decanoyl-L-phenylalanine, exhibited a
N,N-dimethylformamide.
strong b5 subunit inhibitory activity with a Ki0 value of 0.14 nM
⇑
Corresponding author. Tel.: +81 11 706 3228; fax: +81 11 706 4980.
E-mail address: ichikawa@pharm.hokudai.ac.jp (S. Ichikawa).
as well as strong cytotoxicity with an IC50 value of 2.2 nM against
http://dx.doi.org/10.1016/j.bmcl.2015.06.015
0960-894X/Ó 2015 Elsevier Ltd. All rights reserved.
Please cite this article in press as: Chiba, T.; et al. Bioorg. Med. Chem. Lett. (2015), http://dx.doi.org/10.1016/j.bmcl.2015.06.015

2
T. Chiba et al. / Bioorg. Med. Chem. Lett. xxx (2015) xxx–xxx
human myeloma RPMI8226 cells.9 In this paper, we describe the
structure–activity relationship of 2 aiming at further improvement
of its cytotoxicity.
We planned to synthesize syringolin A analogs by condensation
of the amine unit containing the 12-membered ring 8 with various
carboxylic acid units. The carboxylic acid units 7a–mm were
prepared by a solid phase synthesis in order to synthesize various
analogs efficiently (Scheme 1). Loading amino acids 3a–mm to a
2-chlorotrityl (2-CT) resin gave solid-supported Fmoc amino acids
4a–mm. After deprotection of the Fmoc group, acylation with the
carboxylic acids and cleavage from the 2-CT resin afforded cleanly
the N-acyl carboxylic acid units 7a–mm. Finally, the amine 8
prepared by the previously reported procedure10 was condensed
The X-ray crystal structure of the yeast 20S proteasome bound
to syringolin A (PDB code: 2ZCY)7 revealed several aspects for
molecular interaction in addition to its characteristic covalent
binding. Namely, syringolin A is stabilized by hydrogen bond for-
mation with Asp114 and Gly47 residues (Fig. 2a). The isopropyl
group located on the 12-membered ring is recognized by a small
S1 pocket, and the isopropyl group located on the internal Val side
chain is recognized by a large S3 pocket, which consists of
hydrophobic residues. The S3 pocket could accept a larger sub-
stituent than the isopropyl group, and we have examined the effect
of the size of the substituent and found that 2 possessing a benzyl
group exhibited a stronger proteasome inhibitory activity than syr-
ingolin A. The decanoyl group of 2 was considered to be necessary
both for cell-membrane permeability and proteasome inhibitory
activity (Fig. 2b). In order to optimize the structure in more detail,
we planned to change the length of the lipophilic chain and the
phenylalanine moiety to other amino acids (Fig. 2c).
i
with 7a–mm in the presence of EDCI, HOBt and Pr2NEt in DMF
to yield the corresponding syringolin A analogs (9a–mm).
Cytotoxicity of 9a–mm against several human cancer cells
(human epidermoid carcinoma A431, human colon carcinoma
HCT-116 and RKO, human ovarian carcinoma TOV21G, human
lymphoblastic leukemia CCRF-CEM, human myeloma RPMI8226)
was evaluated.11 Bortezomib and carfilzomib as controls exhibited
cytotoxicity with IC50 values of 11.0–28.2 nM and 6.1–27.1 nM,
respectively (Table 1). Under these assay conditions, compound 2
exhibited a strong cytotoxicity with IC50 values between 4.3 and
(a)
(b)
S3 pocket
S3 pocket
HN
HN
O
O
Gly47
O
Gly47
O
previous study
ref 9
H
N
H
N
O
O
N
H
N
H
CO2H
N
H
O
O
O
O
O
O
N
H
N
H
hydrophobic region
NH2
NH2
Asp114
Asp114
β
of the 6 subunit
Thr1
Thr1
S1 pocket
S1 pocket
compound 2 covalently bound to the proteasome
(c)
this study
R
HN
O
H
N
O
N
n
O
R'
O
N
H
target compounds
Figure 2. (a) Binding mode of syringolin A with proteasome. (b) Expected binding mode of previously synthesized compound 2 with proteasome. (c) The structure of the
target compounds in this structure–activity relationship study.
R1
R1
R1
R1
R1
O
O
a
b
c
d
O
O
O
HO2C
NFmoc
R2
NFmoc
R2
NH
R2
N
R2
R3
HO2C
N
R2
R3
O
O
O
3a-kk
4a-kk
5a-kk
= 2-chlorotrityl chloride resin
6a-kk
7a-kk
H
N
NFmoc
a-d
a-d
HO2C
HO2C
HO2C
N
O
HO2C
3ll
N
Fmoc
O
7ll
3mm
7mm
R
HN
HN
O
·HBr
NH2
H
N
e
O
O
N
n
7a-mm
O
R'
O
O
N
H
N
H
8
9a-mm
Scheme 1. Synthesis of syringolin A analogs (9a–mm). Reagents and conditions: (a) 2-chlorotrityl chloride resin, iPr2NEt, CH2Cl2, rt; (b) 20% piperidine/DMF, rt; (c) acyl
i
chloride, Et3N, CH2Cl2, rt; (d) 5% TFA/CH2Cl2, rt; (e) EDCl, HOBt, Pr2NEt, DMF, rt.
Please cite this article in press as: Chiba, T.; et al. Bioorg. Med. Chem. Lett. (2015), http://dx.doi.org/10.1016/j.bmcl.2015.06.015

T. Chiba et al. / Bioorg. Med. Chem. Lett. xxx (2015) xxx–xxx
3
Table 1
chain could not fit into the hydrophobic pocket at the b6 subunit.
This indicates that the acyl chain having eight to eleven carbon
atoms are suitable to show strong cytotoxicity. The decanoyl
group was revealed to be the best lipophilic chain among the
tested in this study and used for further synthesis of the analogs.
The number of the carbon atoms between the phenyl group and
the peptide backbone was next optimized (Table 2). Decrease and
increase of the number of carbon atoms in 9j–9l resulted in
decrease of the cytotoxicity. Furthermore, the aromatic ring in 2
was important for its potent cytotoxicity because replacement of
the phenyl group by a cyclohexyl group in 9m decreased the activ-
ity about 2 to 5-fold.
With these results in mind, we synthesized and evaluated
cytotoxicity of the analogs possessing a variety of substituents on
the phenyl group (Table 3). We also substituted the phenyl group
with the hetero aryl groups. Introduction of a halogen atom or a
methyl group to the benzene ring in 9n–9y retained the cytotoxicity,
Structure and cytotoxicity of syringolin A analogs 9a–9i
HN
O
H
N
O
N
H
n
O
O
N
H
Compound
IC50 (nM)
A431
HCT-116
RKO
TOV21G
CCRF-
CEM
RPMI8226
Bortezomib
Carfilzomib
2 (n = 6)
28.2
23.8
14.1
413.2
128.3
57.7
20.2
18.2
38.2
44.8
82.2
430.0
27.0
19.3
12.6
402.2
61.4
24.5
14.0
7.1
36.0
37.4
73.9
413.1
20.8
16.7
23.8
12.3
426.6
131.4
60.7
19.0
10.4
19.3
27.5
48.3
424.6
11.0
6.1
4.3
61.1
28.6
11.6
6.6
1.8
7.4
14.8
36.9
409.4
14.4
13.9
9.0
138.4
42.6
18.9
9.9
27.1
11.8
415.4
52.6
33.4
14.8
8.6
19.2
33.6
49.8
469.8
9a (n = 1)
9b (n = 2)
9c (n = 3)
9d (n = 4)
9e (n = 5)
9f (n = 7)
9g (n = 8)
9h (n = 9)
9i (n = 10)
whereas substitution with
a hydroxyl group at the para-
position in 9z caused a substantial decrease of the activity
(IC50 = 28.3–134.3 nM). This could be attributed to the decrease of
hydrophobic interactions with the S3 pocket of the b5 subunit
because of introduction of the polar hydroxyl group. Although the
benzene ring was replaced with the pyridine ring such as 2-pyridyl
(9aa) and 3-pyridyl (9bb) analogs retained the activity, the
4-pyridyl analog (9cc) exhibited a slightly weaker cytotoxicity
(IC50 = 14.5–72.6 nM) than 2. This result was consistent with the
decreased activity observed for compound 9z. Because an aqueous
solubility of 2 is poor, the hydrochloride salts of 9aa and 9bb are
expected to improve its aqueous solubility, which is a preferable
property as a drug candidate. Analogs containing the biphenyl ring
(9dd), naphthyl ring (9ee, 9ff) or heteroaromatic ring (9gg–9jj)
exhibited almost similar cytotoxicity to that of 2.
8.2
14.4
29.6
88.3
734.9
14.1 nM. The lipophilic chain of 2 is expected to be recognized by a
hydrophobic pocket at the b6 subunit, which is located next to the
b5 subunit, upon binding to the b5 subunit.8a First, we evaluated
syringolin A analogs having a different length of the lipophilic
chain (Table 1). Shortening the length of the acyl chain from
decanoyl to pentanoyl group in 9a–9e gradually decreased their
cytotoxicity. This could be attributed to the decrease of the
hydrophobic interaction as well as cell-membrane permeability.
Analogs having over eleven carbons 9f–9i also gradually
decreased the cytotoxicity, suggesting that the longer lipophilic
The analog 9kk, where the nitrogen atom at the amide bond in
the side chain was methylated,12 and conformationally restricted
analog 9ll were also evaluated (Table 4). However, these analogs
lost their cytotoxic activity, suggesting that the hydrogen atom of
Table 2
Structure and cytotoxicity of syringolin A analogs 9j–9m
HN
R
O
H
N
O
N
H
O
O
N
H
Compound
R
IC50 (nM)
TOV21G
A431
27.8
HCT-116
45.1
RKO
20.8
CCRF-CEM
11.0
RPMI8226
26.9
9j
9.5
9.6
9k
16.8
28.4
27.0
11.6
14.9
9l
31.7
52.8
40.6
38.6
39.8
35.1
8.5
9.4
23.5
21.0
9m
48.9
21.4
Please cite this article in press as: Chiba, T.; et al. Bioorg. Med. Chem. Lett. (2015), http://dx.doi.org/10.1016/j.bmcl.2015.06.015

4
T. Chiba et al. / Bioorg. Med. Chem. Lett. xxx (2015) xxx–xxx
Table 3
Structure and cytotoxicity of syringolin A analogs 9n–9jj
HN
R
O
H
N
O
N
H
O
O
N
H
Compound
R
IC50 (nM)
TOV21G
A431
14.9
HCT-116
7.1
RKO
6.8
CCRF-CEM
3.3
RPMI8226
5.6
F
9n
8.2
F
9o
27.7
15.0
11.8
21.1
5.5
8.8
F
9p
9q
33.2
9.9
23.1
8.7
14.8
7.4
19.0
7.1
4.9
8.2
8.6
8.8
Cl
Br
Me
Cl
9r
18.4
15.6
14.9
13.8
5.2
9.4
Cl
9s
9t
23.3
13.4
21.7
13.4
19.6
13.0
12.3
9.6
4.5
3.9
12.0
9.4
Br
9u
15.5
18.0
16.3
12.8
4.6
9.6
Br
9v
25.6
8.9
33.4
12.0
33.7
9.5
24.1
6.2
18.4
3.6
28.3
8.8
9w
Me
9x
9y
6.9
7.0
7.3
6.2
9.1
3.5
1.9
8.4
4.8
Me
18.9
28.1
15.5
HO
9z
134.3
23.5
68.5
19.2
43.7
18.2
151.0
10.7
32.3
8.6
28.3
15.0
N
9aa
Please cite this article in press as: Chiba, T.; et al. Bioorg. Med. Chem. Lett. (2015), http://dx.doi.org/10.1016/j.bmcl.2015.06.015

T. Chiba et al. / Bioorg. Med. Chem. Lett. xxx (2015) xxx–xxx
5
Table 3 (continued)
Compound
R
IC50 (nM)
TOV21G
A431
24.4
HCT-116
29.4
RKO
17.1
CCRF-CEM
27.5
RPMI8226
15.5
N
9bb
9cc
5.1
N
50.3
21.9
56.3
27.1
47.3
30.4
14.5
35.7
12.3
72.6
19.7
9dd
9ee
9ff
16.8
13.2
14.1
29.3
17.8
31.0
18.5
31.3
15.2
1.5
5.0
7.4
10.9
9gg
9hh
9ii
29.7
18.7
15.0
26.7
15.5
9.1
20.1
20.8
11.3
22.3
16.7
10.2
24.9
11.0
4.2
24.1
14.4
5.7
O
S
S
9jj
45.5
44.5
39.4
13.5
16.8
27.5
HN
Table 4
Structure and cytotoxicity of syringolin A analogs 9kk–9mm
Compound
Structure
IC50 (nM)
TOV21G
A431
HCT-116
RKO
CCRF-CEM
RPMI8226
>1000
HN
O
O
H
N
9kk
>1000
>1000
>1000
>1000
196.8
N
O
Me
O
N
H
HN
O
H
N
9ll
>1000
>1000
>1000
>1000
>1000
>1000
N
O
O
O
O
N
H
HN
O
H
N
H
N
9mm
191.1
433.5
243.6
211.6
221.2
254.6
O
O
N
H
Please cite this article in press as: Chiba, T.; et al. Bioorg. Med. Chem. Lett. (2015), http://dx.doi.org/10.1016/j.bmcl.2015.06.015

6
T. Chiba et al. / Bioorg. Med. Chem. Lett. xxx (2015) xxx–xxx
the amide moiety is essential for the activity. These results were
realized by analysis of the binding model of the above mentioned
X-ray crystal structure of the yeast 20S proteasome covalently
bound to syringolin A, and the hydrogen of the amide moiety is
one of the key interactions in inhibition. Although analog 9mm
containing b-amimo acid, which was known to be metabolically
stable, was tested,13 the cytotoxicity as decreased almost 10-fold
(IC50 = 191.1–433.5 nM).
We have conducted the structure–activity relationship of
syringolin A, which indicates the followings. (1) Regarding for a
hydrophobic chain, eight to eleven in number of carbon atoms
are suitable to show strong cytotoxicity. (2) The best number of
carbon atoms linking phenyl group to the peptide backbone is
one. (3) Introduction of the hydrophobic substituent at the ben-
zene ring does not affect the cytotoxicity. (4) The hydrogen atom
of the amide moiety is essential for the cytotoxicity. The knowl-
edge obtained in this study would provide a guide to design more
potent syringolin A analogs and this study is in due course.
References and notes
1. Murata, S.; Yashiroda, H.; Tanaka, K. Nat. Rev. Mol. Cell Biol. 2009, 10, 104.
2. Kisselev, A. F.; van der Linden, W. A.; Overkleeft, H. S. Chem. Biol. 2012, 19, 99.
3. Bross, P. F.; Kane, R.; Farrell, A. T.; Abraham, S.; Benson, K.; Brower, M. E.;
Bradley, S.; Gobburu, J. V.; Goheer, A.; Lee, S.-L.; Leighton, J.; Liang, C. Y.;
Lostritto, R. T.; McGuinn, W. D.; Morse, D. E.; Rahman, A.; Rosario, L. A.; Verbois,
S. L.; Williams, G.; Wang, Y.-C.; Pazdur, R. Clin. Cancer Res. 2004, 10, 3954.
4. Herndon, T. M.; Deisseroth, A.; Kaminskas, E.; Kane, R. C.; Koti, K. M.;
Rothmann, M. D.; Habtemariam, B.; Bullock, J.; Bray, J. D.; Hawes, J.; Palmby,
T. R.; Jee, J.; Adams, W.; Mahayni, H.; Brown, J.; Dorantes, A.; Sridhara, R.;
Farrell, A. T.; Pazdur, R. Clin. Cancer Res. 2013, 19, 4559.
5. (a) Lü, S.; Wang, J. Biomarker Res. 2013, 1, 13; (b) Alsina, M.; Trudel, S.; Furman,
R. R.; Rosen, P. J.; O’Connor, O. A.; Comenzo, R. L.; Wong, A.; Kunkel, L. A.;
Molineaux, C. J.; Goy, A. Clin. Cancer Res. 2012, 18, 4830.
6. Wäspi, U.; Blanc, D.; Winkler, T.; Rüedi, P.; Dudler, R. Mol. Plant-Microbe
Interact. 1998, 11, 727.
7. Groll, M.; Schellenberg, B.; Bachmann, A. S.; Archer, C. R.; Huber, R. Nature
2008, 452, 755.
8. (a) Groll, M.; Schellenberg, B.; Bachmann, A. S.; Archer, C. R.; Huber, R.; Powell,
T. K.; Lindow, S.; Kaiser, M.; Dudler, R. Proc. Natl. Acad. Sci. U.S.A. 2009, 106,
6507; (b) Clerc, J.; Schellenberg, B.; Groll, M.; Bachmann, A. S.; Huber, R.;
Dudler, R.; Kaiser, M. Eur. J. Org. Chem. 2010, 3991; (c) Clerc, J.; Li, N.; Krahn, D.;
Groll, M.; Bachmann, A. S.; Florea, B. I.; Overkleeft, H. S.; Kaiser, M. Chem.
Commun. 2011, 47, 385; (d) Archer, C. R.; Groll, M.; Stein, M. L.; Schellenberg, B.;
Clerc, J.; Kaiser, M.; Kondratyuk, T. P.; Pezzuto, J. M.; Dudler, R. Biochemistry
2012, 51, 6880.
Acknowledgments
9. Chiba, T.; Hosono, H.; Nakagawa, K.; Asaka, M.; Takeda, H.; Matsuda, A.;
Ichikawa, S. Angew. Chem., Int. Ed. 2014, 53, 4836.
This research was supported by JSPS Grant-in-Aid for
Challenging Exploratory Research (SI, Grant number 22659020),
Scientific Research on Innovative Areas ‘Chemical Biology of
Natural Products’ (SI, Grant number 24102502), and Scientific
Research (B) (SI, Grant number 25293026). We thank Taiho
Pharmaceutical Co., Ltd for evaluation of cytotoxicity.
10. Pirrung, M. C.; Biswas, G.; Ibarra-Rivera, T. R. Org. Lett. 2010, 12, 2402.
11. Antiproliferative activities of the compounds against A431, HCT-116, RKO,
TOV21G, CCRF-CEM, RPMI8226 cells were measured by quantifying the
amount of ATP as follows. Briefly, cells (1–2 ꢀ 103 cells/well) in a 96 well
plate were cultured in RPMI medium containing 10% fetal bovine serum in the
presence of test compounds at 37 °C for 72 h under 5% CO2 atmosphere. Then, a
solution of CA-1000 reagents (TOYO INK Co., Ltd) was added, and plates were
shaken and luminescence of oxyluciferin was monitored with the plates held
at 23 °C.
Supplementary data
12. Chatterjee, J.; Gilon, C.; Hoffman, A.; Kessler, H. Acc. Chem. Res. 2008, 41,
1331.
13. Cheng, R. P.; Gellman, S. H.; Degrado, W. F. Chem. Rev. 2001, 101, 3219.
Supplementary data associated with this article can be found, in
the online version, at http://dx.doi.org/10.1016/j.bmcl.2015.06.
015.
Please cite this article in press as: Chiba, T.; et al. Bioorg. Med. Chem. Lett. (2015), http://dx.doi.org/10.1016/j.bmcl.2015.06.015