Rational design, synthesis and in vitro evaluation of novel exo-methylene butyrolactone salicyloylamide as NF-κB inhibitor

Article abstract of DOI:10.1016/j.bmcl.2016.12.017

(?)-Dehydroxymethylepoxyquinomicin ((?)-DHMEQ, 1) is a specific inhibitor of NF-κB. It binds to SH group in the specific cysteine residue of NF-κB components with its epoxide moiety to inhibit DNA binding. In the present research, we have designed and synthesized an epoxide-free analog called (S)-β-salicyloylamino-α-exo-methylene-?-butyrolactone (SEMBL, 3). SEMBL inhibited DNA binding of NF-κB component p65 in vitro. It inhibited LPS-induced NF-κB activation, iNOS expression, and inflammatory cytokine secretions. It also inhibited NF-κB and cellular invasion in ovarian carcinoma ES-2 cells. Moreover, its stability in aqueous solution was greatly enhanced compared with (?)-DHMEQ. Thus, SEMBL has a potential to be a candidate for a new anti-inflammatory and anticancer agent.

Full text of DOI:10.1016/j.bmcl.2016.12.017

Bioorganic & Medicinal Chemistry Letters xxx (2016) xxx–xxx
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
Rational design, synthesis and in vitro evaluation of novel exo-methylene
butyrolactone salicyloylamide as NF-jB inhibitor
Kulrawee Sidthipong a, Jun Ma b, Wei Lin Yu b, Yan Feng Wang b, Susumu Kobayashi c, Satoshi Kishino d,
Naoki Koide e, Takashi Yokochi e, Kuniki Kato a, Shoshiro Okada f, Kazuo Umezawa a,
⇑
a Department of Molecular Target Medicine, Aichi Medical University School of Medicine, 1-1 Yazako-Karimata, Nagakute 480-1195, Japan
b Shenzhen Wanhe Pharmaceutical Co., Ltd., Hi-tech Industrial Park, Nanshan District, Shenzhen, China
c Faculty of Pharmaceutical Sciences, Tokyo University of Science, 2641, Chiba 278-8510, Japan
d Department of Medication Use Analysis and Clinical Research, Meiji Pharmaceutical University, 2-522-1, Kiyose, Tokyo 204-0004, Japan
e Department of Microbiology and Immunology, Aichi Medical University School of Medicine, 1-1 Yazako-Karimata, Nagakute 480-1195, Japan
f Department of Pharmacology, Aichi Medical University School of Medicine, 1-1 Yazako-Karimata, Nagakute 480-1195, Japan
a r t i c l e i n f o
a b s t r a c t
Article history:
(À)-Dehydroxymethylepoxyquinomicin ((À)-DHMEQ, 1) is a specific inhibitor of NF-
jB. It binds to SH
Received 17 September 2016
Revised 2 December 2016
Accepted 6 December 2016
Available online xxxx
group in the specific cysteine residue of NF-jB components with its epoxide moiety to inhibit DNA bind-
ing. In the present research, we have designed and synthesized an epoxide-free analog called (S)-b-sali-
cyloylamino- -exo-methylene-ƴ-butyrolactone (SEMBL, 3). SEMBL inhibited DNA binding of NF-
component p65 in vitro. It inhibited LPS-induced NF- B activation, iNOS expression, and inflammatory
cytokine secretions. It also inhibited NF- B and cellular invasion in ovarian carcinoma ES-2 cells.
a
jB
j
j
Keywords:
(À)-DHMEQ
Cysteine
Moreover, its stability in aqueous solution was greatly enhanced compared with (À)-DHMEQ. Thus,
SEMBL has a potential to be a candidate for a new anti-inflammatory and anticancer agent.
Ó 2016 Elsevier Ltd. All rights reserved.
NF-
-Exomethylene-
Bioconjugate
jB
a
c-lactone
NF-jB is a transcription factor that promotes the expression of
NF-jB cysteine sulfhydryl groups of inhibitors, and their bioactiv-
many inflammatory cytokines, adhesion molecules, and anti-apop-
tosis proteins. However, its excess activation often induces inflam-
ity can be explained by a Michael-type conjugate addition (e.g.
compound 5) of the nucleophilic Cys38 and Cys120 in the p65 mono-
mation and contributes to cancer progression. Thus, NF-
considered to be an attractive target for drug discovery. We previ-
j
B is
mer of NF-
jB to one or more ab-unsaturated carbonyl moieties of
the inhibitor.6
ously designed and synthesized an NF-
j
B inhibitor (À)-DHMEQ
Based on the experimental results above mentioned, we
(1).1,2 It showed potent anti-inflammatory and anticancer activities
designed an epoxide-free analog of (À)-DHMEQ, called (S)-b-sali-
in animal experiments.3 In one hand, we clarified compound 1
cyloylamino-
a-exo-methylene-c-butyrolactone (SEMBL, 3), in
directly binds to the specific cysteine residue of NF-
jB components
which the a-methylene-c-butyrolactone moiety is likely able to
to inhibit their DNA binding activity, which consequently induces
an instability and loss of importin affinity in the case of RelB.3,4
Furthermore, our previous studies revealed that the sulfhydryl
group of cysteine might make a nucleophilic attack on the epoxide
of (À)-DHMEQ and form the adduct molecule with an epoxide
ring-opened structure.5 In general, however, epoxide moiety is
known to be highly reactive and the molecule with an epoxide
may react with various biological molecules in the body. So we
searched for an epoxide-free analog considered to be more stable
than (À)-DHMEQ. Sesquiterpene lactones (e.g. Helenalin 2) are
the most widely published class of natural products cited as
react with the SH group in cysteine to afford compound 6 as
depicted in Fig. 1.
In this study, we report an efficient and practical route for the
synthesis of 3. Additionally, the in vitro evaluation of 3 as NF-
inhibitor was investigated. The target compound 3 was synthe-
sized by condensation of b-amino- -methylene- -butyrolactone
jB
a
c
13 with salicylic acid as depicted in Scheme 1. Although compound
13 was prepared from (S)-tryptophan via ozonolysis by Maxlean
et al. in 1987, their synthetic route is evidently unfit for the quan-
tity production.7 Thus, we developed a new synthetic approach to
get compound 13 from inexpensive, commercially available
N-t-Boc-l-aspartic acid b-benzyl ester 7 as depicted in Scheme 1.
Reduction of compound 7 with NaBH4 in the presence of isobutyl
chloroformate afforded the alcohol 8, which was subsequently
⇑
Corresponding author.
E-mail address: umezawa@aichi-med-u.ac.jp (K. Umezawa).
http://dx.doi.org/10.1016/j.bmcl.2016.12.017
0960-894X/Ó 2016 Elsevier Ltd. All rights reserved.
Please cite this article in press as: Sidthipong K., et al. Bioorg. Med. Chem. Lett. (2016), http://dx.doi.org/10.1016/j.bmcl.2016.12.017

2
K. Sidthipong et al. / Bioorganic & Medicinal Chemistry Letters xxx (2016) xxx–xxx
to the cells 30 min before LPS. SEMBL inhibited the activation of
NF-
B, and the activity was stronger than (À)-DHMEQ (Fig. 2C).
LPS induces NF- B-dependent inducible NO synthase (iNOS) in
j
j
macrophages to produce NO. As shown in Fig. 2D, LPS induced
NO production in RAW264.7 cells, which was inhibited by SEMBL
dose-dependently at 0.03–1 lg/ml without any toxicity. The inhi-
bitory activity was again stronger than that of (À)-DHMEQ. Consis-
tent with this result, SEMBL inhibited the LPS-induced expression
of iNOS mRNA in PCR analysis at 0.1–1
NF- B-dependent expressions and secretions of many inflamma-
tory cytokines including interleukin-6 (IL-6), IL-1b, and TNF- in
macrophages. As shown in Fig. 2F, SEMBL inhibited LPS-induced
IL-6, IL-1b, and TNF- secretions. Thus, SEMBL showed in vitro
anti-inflammatory activity in macrophage-like RAW264.7 cells.
Ovarian carcinoma often metastasizes in the lung, liver and
peritoneal cavity.10 We then evaluated the cellular anti-metastasis
activity of SEMBL using ovarian clear cell carcinoma ES-2 cells.
Using the nuclear fraction from ES-2 cells, SEMBL inhibited the
lg/ml (Fig. 2E). LPS induces
j
a
a
Fig. 1. Molecular design of epoxide-free analog of (À)-DHMEQ (1). Epoxide moiety
has been replaced by the exomethylene carbonyl moiety in SEMBL (3). Compound 3
may react with cysteine SH to give the adduct molecule 6
NF-
SEMBL on cell viability. SEMBL inhibited the constitutively
activated NF- B activity when added to the cells without any
jB-DNA binding in vitro (Fig. 3A). Fig. 3B shows the effect of
j
toxicity. The inhibitory activity was significantly stronger than that
of DHMEQ as shown in Fig. 3C. We then studied the effect of
SEMBL on cellular migration. It inhibited the cellular migration
when monitored with a Radius 96-well cell migration assay (Cell
Biolabs, Inc., CA), as shown in Fig. 3D. It also inhibited the cellular
invasion monitored by a Matrigel chamber assay (Discovery
Labware, Inc., MA) at lower concentrations and the inhibitory
activity is stronger than DHMEQ (Fig. 3E). As the mechanism of
inhibition, Matrix metalloproteinase-2 (MMP-2) expression was
inhibited by SEMBL (Fig. 3F and G). Thus, SEMBL inhibited cellular
migration and invasion without toxicity in ovarian clear cell
carcinoma ES-2 cells.
Scheme 1. Reagents and conditions: (i) N-methylmorpholine, i-BuCO2Cl, NaBH4,
MeOH/THF À10 °C to 0 °C; (ii) 2N HCl/Et2O; (iii) Ph2C@NH, CH2Cl2; (iv) NaH,
HCO2Et, Et2O: (v) (HCHO)n, THF, reflux, 3hr; (vi) 1 N HCl, Et2O; (vii) HOBt(leq.), EDCl
(2 eq.), salicylic acid(1 eq.), i-Pr2NEt(3 eq.), DMF
We then compared the stability of SEMBL and (À)-DHMEQ in
phosphate-buffered saline (PBS). Each chemical was dissolved
and incubated in PBS at 100 lg/ml for 0, 30 min, 1 h, 2 h, 6 h,
12 h at 37 °C. We carried out HPLC analysis of SEMBL and DHMEQ
after incubation in PBS. The results showed that more SEMBL than
DHMEQ remained intact at 30 min-12 h of incubation (Fig. 4A). We
also evaluated the inhibitory activity on NO production. Each com-
treated with 2N-HCl in Et2O to give the amino-lactone 9. After the
protection of amino group of 9 with benzophenone imine, the
imine-lactone 10 was reacted with ethyl formate followed by
pound was incubated in PBS again at 100 lg/ml. Then, the inhibi-
paraformaldehyde in order to obtain
a
-methylene-
c
-butyrolac-
tory activity of the solution was tested for LPS-induced NO
production in RAW264.7 cells. Fig. 4B shows the amount of NO
production. (À)-DHMEQ was quickly inactivated in PBS at
30 min-12 h of incubation. In comparison to (À)-DHMEQ, SEMBL
in PBS showed stronger inhibitory activity at 30 min-12 h of
incubation.
tone 12 according to the protocol of Reid et al., in a 27% yield in
2 steps.8 Next, deprotection of the amino group of 12 with
1N-HCl afforded b-amino-a-methylene-c-butyrolactone 13 in a
quantitative yield. Finally, compound 13 was condensed with sal-
icylic acid in the presence of 1-ethyl-3-(3-dimethylaminopropyl)
carbodiimide (EDCI) and N,N-diisopropylethylamine in DMF to
give the target compound SEMBL 3 in 35% yield.
The presence of a conjugated exomethylene group was shown
to be significant in eliciting NF-jB inhibition in sesquiterpene lac-
Next, we evaluated the biological activity of SEMBL. Firstly, we
tones.12–14 In this study, we employed an exomethylene carbonyl
moiety for the substitution of epoxide but also tried to maintain
the structure of (À)-DHMEQ as much as possible, since the hydro-
xyl group at the 2-position of the benzamide ring system was
found to be essential for the inhibitory activity.15 We could suc-
cessfully synthesize an epoxide free DHMEQ analog, SEMBL, with
a 35% yield starting from 13. This synthesis route is more attain-
able than the previous one reported by Maclean et al. because
the starting compound, N-t-Boc-l-aspartic acid b-benzyl ester,
has a simpler structure and is more commercially available. Biolog-
ical evaluation demonstrated that this new analog, SEMBL, is able
examined the inhibitory activity of SEMBL on in vitro NF-
ing to B DNA. Mouse monocytic leukemia RAW264.7 cells were
incubated with LPS for 2 h to accumulate NF- B in the nucleus.
Then, the nuclear extract was prepared and used for NF- B. As
shown in Fig. 2A, the addition of LPS led to the increase of nuclear
NF- B. SEMBL was added directly to the nuclear extract, and it
inhibited the p65 binding to B DNA at 3–10 g/ml. The inhibitory
activity was similar to that of (À)-DHMEQ (Fig. 2A). The p65-
DNA binding was assessed by the TransAM NF- B p65 transcrip-
jB bind-
j
j
j
j
j
l
jB
j
tion factor assay kit (Active Motif, Tokyo, Japan). Previously,
DHMEQ was reported to have anti-inflammatory effects in macro-
phage cells.9 Hence, we studied the cellular anti-inflammatory
effect of SEMBL in RAW264.7 cells. SEMBL did not show any cellu-
to inhibit NF-
j
B-DNA binding as (À)-DHMEQ, and it also inhibits
NF- B-dependent inflammatory cytokine secretions. It also inhib-
j
ited cellular metastatic activity in ovarian carcinoma cells. SEMBL
is likely to bind to Cys38 as (À)-DHMEQ, since it directly inhibits
p65-DNA binding (Figs. 2A, 3A). Reaction of SEMBL with protected
lar toxicity at 1
lg/ml (Fig. 2B). LPS induced accumulation of p65 in
2 h in cultured RAW264.7 cells. SEMBL or (À)-DHMEQ was added
Please cite this article in press as: Sidthipong K., et al. Bioorg. Med. Chem. Lett. (2016), http://dx.doi.org/10.1016/j.bmcl.2016.12.017

K. Sidthipong et al. / Bioorganic & Medicinal Chemistry Letters xxx (2016) xxx–xxx
3
Fig. 2. Inhibition of NF-
stimulated by LPS for 2 h and nuclear extract was used for the source of p65. SEMBL or (À)-DHMEQ was added to the nuclear extract, and the mixture was assayed in
coated dish. DNA binding was detected by labeled p65 antibody. Data represent the mean SD of 3 independent experiments. *P < 0.05, **P < 0.005. (B) Effect on cell viability.
The cells were treated with each chemical for 24 h, and viability was assessed by MTT. Each value is the mean SD of 3 determinations. (C) Inhibition of NF- B in cultured
j
B and downstream signaling by SEMBL in mouse monocytic leukemia RAW264.7 cells (A) Inhibition of p65-DNA binding in vitro. Cells were
j
B DNA-
j
RAW264.7 cells. SEMBL or (À)-DHMEQ was added to the cultured cells, and then stimulated for 2 h with LPS. The nuclear extract was prepared and the DNA binding was
detected as in Fig. 2(A). *P < 0.05, **P < 0.005 (D) Inhibition of LPS-induced NO-production. SEMBL or (À)-DHMEQ was added to the cells for 24 h. Amount of NO in the medium
was measured by the Griess assay. Each value is the mean SD of 3 determinations. *P < 0.05, **P < 0.005. (E) Inhibition of iNOS mRNA expression. The cells were incubated
with SEMBL for 24 h, and iNOS was detected by semi-quantitative PCR. (F) Inhibition of LPS-induced inflammatory cytokine secretions. The cells were incubated with SEMBL
for 24 h. Each protein amount in the medium was measured by ELISA. Each value is the mean SD of 3 determinations. *P < 0.05, **P < 0.005, ***P < 0.0005.
cysteine to form the conjugate (6) remains to be studied. As a char-
acteristic of this compound, SEMBL in aqueous solution, such as
PBS, was shown to be more stable than (À)-DHMEQ (Fig. 4). The
improved stability may contribute to the stronger cellular activity.
SEMBL is more lipophilic than DHMEQ as the partition coefficient
of SEMBL is higher than DHMEQ (SEMBL CLogP = 1.659, DHMEQ
CLogP = 0.871699). Hence SEMBL is likely able to penetrate into
the cells more easily than (À)-DHMEQ. Therefore, the stronger
cellular activity of SEMBL could also be resulting from its superior
cellular penetration.
In summary, a novel exo-methylene butyrolactone salicyloy-
lamide analog, SEMBL, was synthesized and evaluated for biologi-
cal activity as an NF-jB inhibitor. Our studies indicated that SEMBL
has the potential to be developed as a new anti-inflammatory and
anticancer agent for the second generation of (À)-DHMEQ, since
improved stability is preferable for drug formulation and therapeu-
tic application. The synthetic strategy outlined in this report is
likely to be efficient and applicable to the synthesis of a number
of potential anti-inflammatory and anticancer agents of this new
class.
Please cite this article in press as: Sidthipong K., et al. Bioorg. Med. Chem. Lett. (2016), http://dx.doi.org/10.1016/j.bmcl.2016.12.017

4
K. Sidthipong et al. / Bioorganic & Medicinal Chemistry Letters xxx (2016) xxx–xxx
Fig. 3. Inhibition of NF-jB and cellular invasion by SEMBL in ovarian clear cell carcinoma ES-2 cells (A) In vitro inhibition of p65-DNA binding by SEMBL. Nuclear extract from
ES-2 cells was used for the source of p65. SEMBL was added to the nuclear extract, and the mixture was assayed in
p65 antibody. Data represent the mean SD of 3 independent experiments. *P < 0.05, **P < 0.005. (B) Effect on cell viability. The cells were treated with SEMBL for 24 h, and
viability was assessed by MTT. Each value is the mean SD of 3 determinations. (C) Inhibition of NF- B in cultured ES-2 cells. SEMBL or DHMEQ was added to the cultured
jB DNA-coated dish. DNA binding was detected by labeled
j
cells, and the cells were incubated for 2 h. Then, the nuclear extract was prepared and the DNA binding was detected as in Fig. 3(A) *P < 0.05, **P < 0.005. (D) Inhibition of
cellular migration. The cell were incubated with SEMBL for 24 h and the migration was measured by Radius 96-well migration assay. Each value is the mean SD of 3
determinations. *P < 0.005 (E) Inhibition of cellular invasion. The cells were incubated with SEMBL or DHMEQ for 24 h. and the invasion was measured by Matrigel chamber
assay. Data are expressed as mean SD of 4 fields *P < 0.05, **P < 0.005. (F) Inhibition of MMP-2 secretion. The cells were incubated with SEMBL for 24 h. MMP-2 in the
medium was measured by ELISA. Each value is the mean SD of 3 determinations. *P < 0.05. (G) Inhibition of MMP-2 mRNA expression. The cells were incubated with SEMBL
for 24 h, and MMP-2 was detected by semi-quantitative PCR.
Aichi Medical University 2011–2015 (S1101027). This work was
also supported by grant (A20) from Chubu Advanced Medicine
Consortium of Nagoya University. This work was also supported
by National Key Programs in Science and Technology-Discovery
and Development of New Drugs with High Clinical Impact (NO.
Acknowledgments
This work was supported in part by JSPS Kakenhi Grant Number
23310163 and 26350975 and the MEXT-Supported Program for the
Strategic Research Foundation at Private Universities, which is for
Please cite this article in press as: Sidthipong K., et al. Bioorg. Med. Chem. Lett. (2016), http://dx.doi.org/10.1016/j.bmcl.2016.12.017

K. Sidthipong et al. / Bioorganic & Medicinal Chemistry Letters xxx (2016) xxx–xxx
5
2012ZX09103101-004), Guangdong Provincial Programs for
Achieving Core Technological Breakthroughs in Strategic Emerging
Industries (NO. 2012A080800010) and Shenzhen Municipal
Programs for the Research and Development of Technologies
(Three Key Industries) (NO. JSA201105090126A).
A. Supplementary data
Supplementary data associated with this article can be found, in
the online version, at http://dx.doi.org/10.1016/j.bmcl.2016.12.
017.
References
1. Matsumoto N, Ariga A, To-e S, et al. Bioorg Med Chem Lett. 2000;10:865.
2. Ariga A, Namekawa J, Matsumoto N, Inoue J, Umezawa K.
2002;277:27625.
J Biol Chem.
3. Umezawa K. (Review) Forum of Immunopathological Diseases and Therapeutics.
2013;4:63.
4. Takeiri M, Horie K, Takahashi D, et al. Org Biomol Chem. 2012;10:3053.
5. Kozawa I, Kato K, Teruya T, Suenaga K, Umezawa K. Bioorg Med Chem Lett.
2009;19:5380.
6. Garcia-Pineres AJ, Castro V, Mora G, et al. J Biol Chem. 2001;276:39713.
7. Torsten H, Walter SA, David MB. Can J Chem. 1988;66:779.
8. Murray AW, Reid RG. J Chem Soc Chem Commun. 1984;132.
9. Suzuki E, Umezawa K. Biomed Pharmaco. 2006;60:578.
10. Kristensen GB, Trope C. Lancet. 1997;349:113.
11. HPLC conditions are as follows. SEMBL and (À)-DHMEQ concentrations were
measured using a high performance liquid chromatography (HPLC) system
(LC-20AD pump, Shimadzu, Kyoto, Japan) equipped with an auto-sampler
(SIL-20A),
controller (CBM-20A). The analytical column consisted of a Shim-Pack VP-
ODS (150 mm  2.0 mm i.d., 5 m particle size) and a Shim-Pack (GVP-ODS,
m particle size) as guard column. The column was
heated to 40 °C. The mobile phase consisted of acetonitrile and 0.1% (w/v)
ammonium acetate (22:78, v/v). The flow rate was 0.3 mL/min. The wavelength
of the UV detector was set at 254 nm.
a column oven (CTO-20A), degasser (DGU-20A3) and system
l
5 mm  2.0 mm i.d., 5
l
Fig. 4. Stability of SEMBL and (À)-DHMEQ in aqueous solution. (A) HPLC analysis.
Each chemical was incubated in PBS for 30 min, 1 h, 2 h, 6 h, 12 h at 100 g/ml.
l
After the incubation, ethyl acetate (100 ml) was added and vortex immediately. The
ethyl acetate portion was transferred to the new tube and allowed to dry. The
concentration of SEMBL and DHMEQ were measured by HPLC analysis.11 (B)
Remained biological activity. Each chemical was incubated in PBS as in Fig 4A. After
the incubation, the solution was added to the cultured RAW 264.7 cells to provide
12. Bremner P, Heinrich M. J Pharm Pharmcol. 2002;54:453.
13. Dirsch VM, Stuppner H, Ellmerer-Müller EP, Vollmar AM. Bioorg Med Chem.
2000;8:2747.
14. Rüngeler P, Castro V, Mora G, et al. Bioorg Med Chem. 1999;7:2343.
15. Chaicharoenpong P, Kato K, Umezawa K. Bioorg Med Chem Lett. 2002;10:3933.
final concentration of
chemical was added from the 10 mg/ml DMSO solution to give a final concentration
of 1 g/ml. After 30 min of chemical addition, the cells were stimulated with
1 lg/ml for each chemical. For the 0-time point, each
l
10 ng/ml LPS for 24 h. NO production was detected by Griess assay. Each points in A
and B is the mean SD of 3 determinations. *P < 0.05, **P < 0.005.
Please cite this article in press as: Sidthipong K., et al. Bioorg. Med. Chem. Lett. (2016), http://dx.doi.org/10.1016/j.bmcl.2016.12.017