Solution-phase parallel synthesis and screening of anti-tumor activities from fenbufen and ethacrynic acid libraries

Article abstract of DOI:10.1016/j.bmcl.2011.01.068

The derivatives with fenbufen and ethacrynic acid core compounds was synthesized through a facial preparation of 1-amino-4-azidobutane. The subsequent coupling with 102 members of carboxylic acids afforded amide products. The in situ screening using colorimetric assay with 3-(4.5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide showed that fenbufen but not ethacrynic acid butyl amide members displayed the cytotoxicities to tumor cells substantially, including two human cell lines (MCF7 and A549) and two murine cell lines (C26 and TRAMP-C1). Three fenbufen analogs were found to have a good anti-tumor activity comparable to cisplatin.

Full text of DOI:10.1016/j.bmcl.2011.01.068

Bioorganic & Medicinal Chemistry Letters 21 (2011) 1320–1324
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
Solution-phase parallel synthesis and screening of anti-tumor activities
from fenbufen and ethacrynic acid libraries
Yuan-Hsiao Su a, Li-Wu Chiang a, Kee-Ching Jeng b, Ho-Lien Huang a, Jenn-Tzong Chen c, Wuu-Jyh Lin c,
Chia-Wen Huang a, Chung-Shan Yu a,d,
⇑
a Department of Biomedical Engineering and Environmental Sciences, National Tsing-Hua University, No. 101 Sec.2, Guang-Fu Rd., Hsinchu 30043, Taiwan
b Taichung Veterans General Hospital, Taichung 40705, Taiwan
c Institute of Nuclear Energy Research, Taoyuan 32546, Taiwan
d Institute of Nuclear Engineering and Science, National Tsing-Hua University, Hsinchu 30043, Taiwan
a r t i c l e i n f o
a b s t r a c t
Article history:
The derivatives with fenbufen and ethacrynic acid core compounds was synthesized through a facial
preparation of 1-amino-4-azidobutane. The subsequent coupling with 102 members of carboxylic acids
afforded amide products. The in situ screening using colorimetric assay with 3-(4.5-dimethylthiazol-2-
yl)-2,5-diphenyltetrazolium bromide showed that fenbufen but not ethacrynic acid butyl amide mem-
bers displayed the cytotoxicities to tumor cells substantially, including two human cell lines (MCF7
and A549) and two murine cell lines (C26 and TRAMP-C1). Three fenbufen analogs were found to have
a good anti-tumor activity comparable to cisplatin.
Received 3 November 2010
Revised 31 December 2010
Accepted 17 January 2011
Available online 22 January 2011
Keywords:
1-Amino-4-azidobutane
Library
Ó 2011 Elsevier Ltd. All rights reserved.
1-{4-[(4-Aminobutylaminooxy)methyl]-
2,3-dichlorophenyl}-2-methylenebutan-1-
one
N-(4-Aminobutyl)-4-(biphenyl-4-yl)-4-
oxobutanamide
Cell-based screening
Parallel solution-phase synthesis (psps) has been used for lead
generation and lead optimization of synthetic compounds.1 The ra-
pid purification or isolation of desired compounds from a reaction
mixture represents a bottleneck in the synthetic procedure.2 Thus,
numerous approaches using solid-supported reagents and scav-
engers have been developed for isolation of library members.3,4
On the other hand, in situ enzyme-based screening of the library
mixture without prior purification has been developed by Wong
and co-workers.5 A number of potential compounds could be dis-
covered by this approach.
By adopting a similar strategy, an in situ cell-based assay might
provide another methodology for drug discovery. As an ongoing
program of discovery for potent antitumor compounds, we were
interested in developing a cell-based in situ screening protocol
before studying their in vivo pharmacokinetics through an imaging
modality in conjunction with radiofluoro-substituted (18F) probe
H
N
O
O
H
N
Cl
Cl
O
IC50 = 18 uM vs. A549
IC50 = 8 uM vs. MCF7
O
butyl fenbufen
O
IC50 = 100 uM vs. A549
IC50 = 87 uM vs. MCF7
butyl ethacrynic acid
Abbreviations: A549 tumor, human type II epithelial cell line; MCF-7 tumor, solid human estrogen receptor positive breast carcinoma; TRAMP-C1 prostate cancer cell line,
transgenic adenocarcinoma of the mouse prostate; C26 tumor, solid murine colon carcinoma; MTT assay, cell viability was measured with blue formazan that was
metabolized from colorless 3-(4,5-dimethyl-thiazol-2-yl)-2,5-diphenyl tetrazolium bromide (MTT) by mitochondrial dehydrogenases which are active only in live cells;
NSAIDs, non-steroid anti-inflammatory drugs.
⇑
Corresponding author. Tel.: +886 3 5715131x35582; fax: +886 3 5718649.
E-mail address: csyu@mx.nthu.edu.tw (C.-S. Yu).
0960-894X/$ - see front matter Ó 2011 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2011.01.068

Y.-H. Su et al. / Bioorg. Med. Chem. Lett. 21 (2011) 1320–1324
1321
and positron emission tomography (PET). The initial concept has
been performed through a cell-based screening of a minilibrary
by psps via amide bond formation.6 Two amide compounds,
butylfenbufen and butyl ethacrynic acid, derived from fenbufen 3
and ethacrynic acid (EA) 8 coupled with butanamine showed cyto-
and the in situ cell-based screening. Thus, a desired member com-
pound exhibiting inhibition concentration ranging in submicromo-
larity should be our ultimate goal. Therefore, fenbufen butanamine
5 and ethacrynic acid butanamine 10 emerged as the rational
building templates for synthesis of libraries and in situ screening.
Since the structural features of the biphenyl ring moieties and
the 1,2-dichloro-4-carbonyl benzene ring moieties are each re-
quired for the bioactivities of individual compound class, the
appropriate structural diversity would be extended from the ter-
minal end of the butyl chain. Thus, a bifunctional linker consisting
of a protected amino group and a free amino group is prerequisite
toxicities of IC50 of 87 and 18 lM to MCF7 and A549 tumor cells,
respectively.6 Furthermore, Lu and co-workers recently report that
EA and its derivatives exhibit a selective toxicity to chronic lym-
phocytic leukemia by inhibition of the Wnt/bcatenin pathway.7,8
Based on the above findings, it would be interesting to find out
whether more potent analogs could be discovered through psps
Aromatic ring-based components
Heterocyclic or structurally complicated components
O
O
HO
O
HO
H
N
O
NH2
HO
O
O
O
OH
O
HO
HO
HO
N
O
HN
B3
OH
OH
OH
HO
A1
A5
B4
B2
HO
A2
A3
A4
OH
B1
O
O
NH2
NH
O
OH
O
O
O
-O
O
NH
O
O
HO
O
N+
OH
NH2
HO
HO
HO
NH
HO
N
O
HO
N
B6
A6
A7
O
A8
N
B7
HO
B8
B5
HO
O
OH
H
H
H
HO
N
O
N
N
O
HN
OH
HO
O
OH
O
N
O
O
HO
O
A9
O
A10
A11
O
A12
OH
O
B11
HO
HO
B12
B10
B9
HO
HO
-O
OH
OH
HO
N+
O-
OH
N
HO
N+
O
H
O
H
HO
O
O
HO
O
NH2
O
O
O
N
N
O
A13
N
O
O
A14
A15
A19
A16
O
N
H2N
N
H
B15
HO
HO
H
O
O
OH
O
OH
O
B14
H
N
OH
O
B13
O
B16
O
OH
O
O
OH
O
O
A17
A18
A20
O
O
O
HO
HO
HO
O
O
OH
HO
N
OH
O
O
-O
O
HN
HO
N
N+
O
N
NH
O
O
O
OH
O
O
N
O
O
OH
A22
B18
B19
OHO
B20
A21
B17
A23
Halosubstituted aromatic components
Aliphatic components
Br
HO
Cl
O
HO
HO
O
NH2
Cl
O
Cl
HO
OH
D1
OH
D2
O
O
N
O
O
S
O
HO
NH2
O
HO
O
O
O
OH
H2N
HO
N
OH
HO
N
HO
O
Cl
OH
OH
D6
O
Cl
D4
C9
D3
C1
C4
C2
HO
C3
D5
OH
D8
C5
D7
F
O
O
OH
NH2
O
O
O
HO
O
O
Cl
Cl
O
O
S
HO
H2N
F
F
OH
HO
OH
HO
HO
HO
OH
F
O
Cl F
OH
7
D14
O
5
O
NH2
D10
O
O
O
HO
Cl
D13
O
C8
D12
C10
C11
D9
NH
D11
O
C7
C6
Cl
NH2
Cl
O
O
O
F
F
H2N
HO
F
F
HO
O
OH
NH2
HO
NH2
O
O
O
OH
HO
NH
F
OH
HO
H2N
N
O
OHO
D16
2
N
OH
O
F
O
O
HO
HO
Cl
Cl
C15
D17
F
F
D18
C16
D15
C14
OH
C13
C12
F
HO
O
O
F
O
H
F
N
Cl
OH
OH
F
HO
Cl
Heteroatom contained components
HO
F
O
O
F
F
O
O
O
HO
F
O
F
HO
O
O
O
F
O
P
F
OH
OH
O
S
C20
I
C21
I
HN
HN
O
OH
C17
C18
Cl
C19
Cl
S
HO
Br
OH
O
F
E1
O
OH
Cl
Cl
Cl
E3
E2
E4
ONa
HO
HN
E5
O
E6
N
OH
NH
OH
O
O
O
O
I
HO
F
O
I
I
O
O
P
O
H2N
HO
O
O
O
OH
OH
O
O
S
NH
I
C24
Br
O
C23
C22
OH
P
Br
N
C26
OH
HO
HO
C25
Br
E10
HO
O
O
O
Br
E9
HO
O
E8
E7
E11
O
HO
O
NH2
HO
Br
O
S
OH
HO
O
S
H2N
O
S
N
S
S
6
E15
O
OH
HO
O
E12
E14
E13
Figure 1. Carboxylic acid moieties for preparing the amide product.

1322
Y.-H. Su et al. / Bioorg. Med. Chem. Lett. 21 (2011) 1320–1324
TfN3, CuSO4
40%
N3
NH2
O
H2N
H2N
2
1
O
H
N
OR1
2, HBTU
N3
DIEA, DMSO,
1 h, 77%
O
O
4
3 (Fenbufen) R1 = H
CH2N2
85%
H2/Pd/C
86%
2, NEt3, 16-23%
7 R1 = CH3
O
H
O
H
N
O
N
R2
NH2
N
H
O
O
5
6
(102 members)
R2COOH, HBTU
DIEA, DMSO
Scheme 1. Reagents and conditions for preparation of libraries of butylfenbufen.
H
N
H
N
O
OH
2, HBTU,
O
O
R1
O
R2
O
N
H
Cl
Cl
R2COOH, HBTU,
Cl
Cl
O
Cl
Cl
O
DIEA, DMF
86%
DIEA, DMSO
O
O
O
R1 = N3
10 R1 = NH2
8
9
11
(102 members)
H2, Pd/C
80%
Ethacrynic acid
Scheme 2. Reagents and conditions for preparation of libraries of EA butylamide.
fenbufen butyl amide
ethacrynic acid butyl amide
120
100
80
60
40
20
0
120
100
80
60
40
20
0
120
100
80
60
40
20
0
120
100
80
60
40
20
0
survival ratio
the number of compds
survival ratio
the number of compds
molecular weight
molecular weight
Figure 2. Molecular weight distribution of the amide members and the relevant average survival ratio from the four cell lines exerted whereby. (a) Data from fenbufen butyl
amide. (b) Data for EA butylamide.
O
O
H
N
O
O
R
O
N
H
O
O
Cl
12, R = A23
13, R = D5
14, R = C23
O
Cl
C23
A23
D5
Figure 3. Structures of the potential compounds proven by validating experiment.

Y.-H. Su et al. / Bioorg. Med. Chem. Lett. 21 (2011) 1320–1324
1323
for synthesis of libraries. To this end, various synthetic approaches
using either Boc or azido as the protecting groups have been re-
ported.9 Among them, azide as a protecting group particularly at-
tracts our attention because of its easy preparation.10 A number
of synthetic routes toward 1-amino-4-azidobutane 2 have been re-
ported, for example, reductive alkylation of an azide9 or selective
reduction of a diazidoalkane.10 For shortening the synthetic route
toward 1-amino-4-azidobutane 2, here, we described an
alternative synthesis of compound 2 from 1,4-diaminobutane 1
via azide transfer reaction under the catalysis of copper(II) ions
(Scheme 1).11 Its application toward the synthesis of the two core
components of fenbufen 5 and EA 10 and subsequent coupling
reaction were hereby described.
To validate the apparent cytotoxicities induced by the crude
members, an additional preparation and purification for the poten-
tial amides was critical. Hence, fenbufen butyl amide analogs with
carboxylic acid moieties such as A23, D5, and D11 and EA butyl
amide analogs with acid moieties for example, B15, B17 and E15
deserved a further examination. The self-coupling of acids B15
and D11, as encountered during library preparation, however, dis-
couraged further attempt. Besides, since both EA butyl amides B17
and E15 displayed comparable bioactivity profiles, the B17 amide
was chosen for further analysis. Thus, three potential compounds
12–14 were prepared separately and tested for MTT assays (Fig. 3).
Instead of using column chromatography, compound 2 was ob-
tained in an acceptable yield (40%) through partition. Though a sig-
nificant amount of the byproduct, 1,4-diazido butane, was always
accompanied in the mixture, the desired amide product 4 could
be isolated by column chromatography in the subsequent reaction.
Activation of the compound 3 by means of formation of an ester 7
to couple with compound 2 failed, however, to provide satisfactory
yield of compound 4 at either room temperature or under reflux
condition. Thus the usual activation condition using O-(ben-
zotriazol-1-yl)-N,N,N0,N0-tetramethyluronium (HBTU) emerged as
the proper choice.12 After amide formation and subsequent
reductive hydrogenation, the amine compound 5 obtained was
readily served as a core compound for preparing the libraries of
compound 6.
140
cisplatin
A549
compd12
compd13
compd14
fenbufen
120
100
80
60
40
20
0
However, purification of core compound 5 under the usual
chromatographic condition was not an appropriate choice since
very polar eluents such as ammonia (5%) in combination with
MeOH and CHCl3 would dissolve silica gel and make mess. By
adopting a simple partition between CH2Cl2 and aqueous layer
with pH value ranging from 10 to 11 (adjusted by a combination
of NaHCO3 (0.1 M) and NaOH (0.1 M)), the resultant organic salts
and residual bases could be successfully removed. Likewise, a sim-
ilar synthetic approach could be used for the preparation of EA
amine 10 as the other core compound (Scheme 2). The two amino
core compounds 5 and 10 were each obtained in 68–70% yields via
two steps. The advantage of the current coupling method was
attributed to a straightforward purification and thereby simplifies
the overall synthesis. Coupling of both core compounds with 102
carboxylic acids provided the corresponding amide products
(Fig. 1).
As inspired by a recent psps that showed a GAUSSIAN distribution
for library member population versus MW sampling,4a we inte-
grated the diagram of bioactivity profile against MW sampling
(Fig. 2). As compared to EA butyl amide members, fenbufen butyl
amide members, in general, exhibited cytotoxic activities against
the four cell lines substantially. Compared to 100% average survival
ratio by treatment with EA butyl amide members, the average sur-
vival ratio by fenbufen butyl amide members was down to 60%.
Furthermore, the survival-ratio pattern showed a valley in the
range of 540–620 D (Fig. 2a). Since the bioactivity trend could be
biased with less sampling numbers, the present data would be sta-
tistically significant with sufficient samples with MW ranging from
420 to 580 D. A similar bioactivity pattern was also observed irre-
spective of the molecular weight sampling interval that was 50 or
30 D (Supplementary Figs. 3 and 4). Therefore, the pattern of bioac-
tivity might be molecular weight-dependent. In contrast, EA butyl
amide members showed two valleys at 430–470 and 670–710 D
for bioactive regions (Fig. 2b). Whereas other MW sampling inter-
val (MWSI) such as 30 D did not alter the biologic pattern, a signif-
icant change was found at the MWSI of 50 D that formed a big
valley at 680–730 D. The dependency of bioactivity on MW is
biased by MWSI. Thus, the current empirical manipulation needs
further investigation.
3.125 6.25 12.5 25
Concentration (µM)
50
50
50
100
0.78 1.56
140
120
100
80
cisplatin
Tramp-C1
compd12
compd13
compd14
fenbufen
60
40
20
0
3.125 6.25 12.5 25
Concentration (µM)
100
0.78 1.56
140
120
100
80
cisplatin
compd12
compd13
compd14
fenbufen
C26
60
40
20
0
3.125 6.25 12.5 25
Concentration (µM)
100
0.78 1.56
Figure 4. Effects of three potential compounds on a survival ratio of A549, TRAMP-
C1, and C26 cells. Cisplatin and fenbufen were used as positive control and negative
control, respectively. Error bar was calculated from the measurements in triplicate.

1324
Y.-H. Su et al. / Bioorg. Med. Chem. Lett. 21 (2011) 1320–1324
Table 1
Supplementary data
Concentration determined for half surviving cells by MTT assay
Tumor cells
IC50 value (
13
lM)
Supplementary data associated with this article can be found, in
the online version, at doi:10.1016/j.bmcl.2011.01.068.
Cisplatin
12
14
Fenbufen
A549
MCF7
TRAMP-C1
C26
6a
9
2
40
10
17
1
18
13
19
5
80
35
>100
>50
>100
>100
25b
22
17
References and notes
1. Boger, D. L.; Desharnais, J.; Capps, K. Angew. Chem., Int. Ed. Engl. 2003, 42, 4138.
2. Kaldor, S. W.; Siegel, M. G. Curr. Opin. Chem. Biol. 1997, 1, 101.
3. An, H.; Cook, D. Chem. Rev. 2000, 100, 3311.
a
IC50 = 1.3–11
IC50 = 14–31
l
l
M.6,14
M.6,15
b
4. (a) Zhang, L.; Lushingston, G. H.; Neuenswander, B.; Hershberger, J. C.;
Malinakova, H. C. J. Comb. Chem. 2008, 10, 285; (b) Manyem, S.; Sibi, M. P.;
Lushington, G. H.; Neuenswander, B.; Schoenen, F.; Aube, J. J. Comb. Chem. 2007,
9, 20; (c) Nieto, M. J.; Philip, A. E.; Poupaert, J. H.; McCurdy, C. R. J. Comb. Chem.
2005, 7, 258; (d) Radi, M.; Petricci, E.; Maga, G.; Corelli, F.; Botta, M. J. Comb.
Chem. 2005, 7, 117; (e) Ranise, A.; Spallarossa, A.; Cesarini, S.; Bondavalli, E.;
Schenone, S.; Bruno, O.; Menozzi, G.; Fossa, P.; Mosti, L.; La Colla, M.; Sanna, G.;
Murreddu, M.; Collu, G.; Busonera, B.; Marongiu, M. E.; Pani, A.; La Colla, P.;
Loddo, R. J. Med. Chem. 2005, 48, 3858.
5. (a) Brik, A.; Wu, C.-Y.; Wong, C.-H. Org. Biomol. Chem. 2006, 4, 1446; (b) Best,
M.; Brik, A.; Chapman, E.; Lee, L.; Cheng, W.-C.; Wong, C.-H. Chem. Biol. Chem.
2004, 5, 811; (c) Wu, C.-Y.; Chang, C.-F.; Chen, J. S. Y.; Lee, S.-T.; Wong, C.-H.;
Lin, C.-H. Angew. Chem., Int. Ed. Engl. 2003, 42, 4661; (d) Brik, A.; Lin, Y.-C.;
Elder, J.; Wong, C.-H. Chem. Biol. 2002, 9, 891.
6. Chiang, L.-W.; Pei, K.; Chen, S.-W.; Huang, H.-L.; Lin, K.-J.; Yen, T.-C.; Yu, C.-S.
Chem. Pharm. Bull. 2009, 57, 714.
7. Jin, G.; Lu, D.; Yao, S.; Wu, C. C. N.; Liu, J. X.; Carson, D. A.; Cottam, H. B. Bioorg.
Med. Chem. Lett. 2009, 19, 606.
8. Lu, D.; Liu, J. X.; Endo, T.; Zhou, H.; Yao, S.; Willert, K.; Schmidt-Wolf, I. G. H.;
Kipps, T. J.; Dennis, A.; Carson, D. A. Open Access Plos One 2009, 4, e8294, 1.
9. Carboni, B.; Benalil, A.; Vaultier, M. J. Org. Chem. 1993, 58, 3736.
10. Lee, J. W.; Jun, S. I.; Kim, K. Tetrahedron Lett. 2001, 42, 2709.
11. Alper, P. B.; Hung, S.-C.; Wong, C.-H. Tetrahedron Lett. 1996, 37, 6029.
12. Valeur, E.; Bradley, M. Chem. Soc. Rev. 2009, 38, 606.
The results confirmed that compound 12 had an equivalent
cytotoxicity of cisplatin against A549 cells (Fig. 4, Table 1). Inter-
estingly, compound 14 was not potent although it retained both
structural characteristics of fenbufen and ethacrynic acid.
The probable cause of lacking improvement on the bioactivity
profile by the ethacrynic acid members might be due to the inap-
propriateness of butyl linkage. On the other hand, Lipinski’s rule
of 5 might provide the other reasonable rationale.13 The rule pre-
dicts that the drug-like compounds have less than 5H-bond donors
and 10 H-bond acceptors and the molecular weight is less than
500 D. Since the core of EA members contained 7H-bond acceptors,
its coupling products would contain more than 10H-bond accep-
tors, such as the coupling products from B17, E10 and E15.
In brief, the present parallel synthesis and in situ screening of
the amide library through 4-azido butanamine as a linker allowed
a discovery of the fenbufen analogs 12–14 with various anti-tumor
activities, as compared to cisplatin. In addition, in view of the anti-
tumor potencies accompanied with the initial role as NSAIDs, the
biologic roles played by the fenbufen analogs warrant further
study.
13. Lipinski, C. A.; Lombardo, F.; Dominy, B. W.; Feeney, P. J. Adv. Drug Deliv. Rev.
1997, 23, 3.
14. (a) Dolfen, D.; Schottler, K.; Valiahdi, S.-M.; Jakupe, M. A.; Keppler, B. K.;
Tiekink, E. R. T.; Mohr, F. J. Inorg. Biochem. 2008, 102, 2067; (b) Ulukaya, E.;
Ozdikicioglu, F.; Oral, A. Y.; Demirci, M. Toxicol. In Vitro 2008, 22, 232; (c) Song,
X.; Liu, X.; Chi, W.; Liu, Y.; Wei, L.; Wang, X.; Yu, J. Cancer Chemother.
Pharmacol. 2006, 58, 776; (d) Causey, P. W.; Baird, M. C. Organometallics 2004,
23, 4486.
Acknowledgments
15. (a) Zhang, J.; Liu, L.; Gong, Y.; Zheng, X.; Yang, M.; Cui, J.; Shen, S. Eur. J. Med.
Chem. 2009, 44, 2322; (b) Johansson, K.; Åhlen, K.; Rinaldi, R.; Sahlander, K.;
Siritantikorn, A.; Morgenstern, R. Carcinogenesis 2007, 28, 465.
We are grateful to National Science Council of Taiwan for pro-
viding financial support (NSC-98-2113-M-007-012). We acknow-
ledge Dr. Chi-Shun Chiang for providing TRAMP-C1 tumor cells.