Structure based optimization of chromen-based TNF-α converting enzyme (TACE) inhibitors on S1′ pocket and their quantitative structure-activity relationship (QSAR) study

Article abstract of DOI:10.1016/j.bmc.2010.10.006

A series of coumarin based TACE inhibitors were designed to bind in S1′ pocket of TACE enzyme based on their docking study. Twelve analogues were synthesized and most of compounds were active in vitro TACE enzyme inhibition as well as cellular TNF-α inhibition. Among these, 15l effectively inhibited the production of serum TNF-α by oral administration at a dose of 30 mg/kg. Compound 15l also showed a good oral bioavailability at 42% and effectively inhibited paw edema in rat carrageenan model. Quantitative structure-activity relationship (QSAR) study using genetic function approximation technique (GFA) and docking study were performed to confirm the series of coumarin core TACE inhibitors. QSAR model have been evaluated internally and externally using test set prediction. Through docking study of each molecule, it is validated that the electrostatic descriptors from the QSAR equation could explain the importance of S1′ pocket and the TACE inhibitory activity well.

Full text of DOI:10.1016/j.bmc.2010.10.006

Bioorganic & Medicinal Chemistry 18 (2010) 8618–8629
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry
journal homepage: www.elsevier.com/locate/bmc
Structure based optimization of chromen-based TNF-a converting enzyme
(TACE) inhibitors on S10 pocket and their quantitative structure–activity
relationship (QSAR) study
Jee Sun Yang a, Kwangwoo Chun a,b, Jung Eun Park a, Misun Cho a, Jeongjea Seo a, Doona Song a,
a,c,
Hongchul Yoon a, Chun-Ho Park b, Bo-Young Joe b, Jong-Hee Choi b, Myung-Hwa Kim b, , Gyoonhee Han
⇑
⇑
a Translational Research Center for Protein Function Control, Department of Biotechnology, Yonsei University, Seodaemun-gu, Seoul 120-749, Republic of Korea
b Drug Discovery Laboratory, R&D Center, Jeil Pharmaceutical Co., Ltd, 117-1, Keungok-Ri, Baekam-Myun, Cheoin-Gu, Yongin-City, Kyunggi-Do 449-861, Republic of Korea
c Department of Integrated OMICS for Biomedical Sciences (WCU Program), Yonsei University, Seodaemun-gu, Seoul 120-749, Republic of Korea
a r t i c l e i n f o
a b s t r a c t
A series of coumarin based TACE inhibitors were designed to bind in S10 pocket of TACE enzyme based on
their docking study. Twelve analogues were synthesized and most of compounds were active in vitro
Article history:
Received 4 August 2010
Revised 3 October 2010
Accepted 5 October 2010
TACE enzyme inhibition as well as cellular TNF-
a inhibition. Among these, 15l effectively inhibited the
production of serum TNF- by oral administration at a dose of 30 mg/kg. Compound 15l also showed a
a
good oral bioavailability at 42% and effectively inhibited paw edema in rat carrageenan model. Quantita-
tive structure–activity relationship (QSAR) study using genetic function approximation technique (GFA)
and docking study were performed to confirm the series of coumarin core TACE inhibitors. QSAR model
have been evaluated internally and externally using test set prediction. Through docking study of each
molecule, it is validated that the electrostatic descriptors from the QSAR equation could explain the
importance of S10 pocket and the TACE inhibitory activity well.
Keywords:
Chromen-based TACE inhibitors
Optimization
Docking
Quantitative structure–activity relationship
Genetic function approximation
Ó 2010 Elsevier Ltd. All rights reserved.
1. Introduction
The orally bioavailable small molecule TACE inhibitors could be
considered as the effective treatment for rheumatoid arthritis and
Tumor necrosis factor-
inflammatory cytokines produced by activated macrophages and
monocytes.1,2 The overproduction of TNF-
is responsible for many
a
(TNF-
a
) is one of the major pro-
other inflammatory diseases by reducing the levels of soluble-
TNF-a 4,13
.
a
As the catalytic site of TACE and matrix metalloproteinase (MMP)
has high amino acid sequence similarity, the early MMP inhibitors,
such as marimastat, prinomastat, and CGS27023A showed TACE
inhibitory activity.14 But the MMP inhibitors showed musculoskele-
tal side effects and failed in clinical trials for unknown reason.15
Though the reason for the side effect is obscure, some researchers
reported that the inhibition of MMP-1 and/or MMP-14 causes
toxicity.16 Hence, developing selective TACE inhibitors or dual TACE
and MMP inhibitors without MMP-1 inhibition are desirable.
On our ongoing discovery program of potent TACE inhibitors,17
a series of chromen-base analogues were prepared and evaluated
for their in vitro potencies.18 Among these inhibitors, the most
active compound, 1a (Fig. 1) has IC50 value of 3 nM for TACE inhi-
bition and found 347 time more selective to TACE over MMP-2
autoimmune disorders such as rheumatoid arthritis, psoriasis,
Crohn’s disease,3 ulcerative colitis,4 diabetes,5 multiple sclerosis,6
atheroscelorosis,7 and stroke.8 The clinical success of anti-TNF-
a
biol-
ogics for treating inflammatory diseases, such as TNF-
a antibody
RemicadeÒ (infliximab) and HumiraÒ (adalimumab), and the soluble
TNF receptor EnbrelÒ (etanercept), have suggested that inhibition of
TNF-
a would be a valid approach for an effective treatment of
rheumatoid arthritis, Crohn’s disease, and psoriasis.9–11 Despite the
success of the biological agents and discovery of a cost-effective, pre-
cisely controllable, orally active, and selective small molecule of anti-
TNF-
getedattwodifferentlevels;oneisinhibitionofpro-TNF-
by inhibiting TNF- converting enzyme (TACE) to reduce the amount
of active and soluble TNF- , and the other is inhibition of pro-TNF-
a
would be desirable.4 Inhibition of TNF-
a
function can be tar-
a
processing
a
a
a
with no inhibition of MMP-9 at 10 lM. a-Methylated analogue
synthesis by blocking nuclear factor (NF)-
j
B activation cascade.12
(1a) and benzylated one (1b) show 20 and 12 times increase in
the inhibitory potency to TACE compared to the non-alkylated
analogue, respectively. Attaining further understanding on the
relationship between the structure and biological activity based
on quantitative structure–activity relationship (QSAR) is useful
tool for design of more potent inhibitors.
⇑
Corresponding authors. Tel.: +82 31 332 4457; fax: +82 31 333 0337 (M.-H.K.);
tel.: +82 2 21232882; fax: +82 2 3627265 (G.H.).
E-mail addresses: mhkim@jeilpharm.co.kr (M.-H. Kim), gyoonhee@yonsei.ac.kr
(G. Han).
0968-0896/$ - see front matter Ó 2010 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmc.2010.10.006

J. S. Yang et al. / Bioorg. Med. Chem. 18 (2010) 8618–8629
8619
O
O
O
O
O
R
O
R
R
a
b
O
OH
O
O
H
n
N
H
N
O
n
O
O
HOHN
N
O
O
9 n= 0, 1, 3, 4
NH2
7 n= 0, 1, 3, 4
8
R= H, OBn
R= H, OBn
O
1a (R=H)
1b (R=Ph)
DPC-333 or BMS-561392
O
P
O
EtO
EtO
O
OEt
n
Figure 1. (a) Structure of chromen-based TACE Inhibitor (1a, b) and DPC-333.
c
R
EtO
EtO
O
n
11 n= 0, 1, 3, 4
10 n= 0, 1, 3, 4
R= H, OBn
In this paper, we will report design and synthesis of the chromen-
core TACE inhibitors for the optimization of S10 pocket modification,
and QSAR study using genetic function approximation (GFA) tech-
R= H, OBn
R
Scheme 2. Reagents and conditions: (a) BH3Me2NH, MeOH; (b) Eschenmoser’s salt,
EtOH, heating; (c) triethyl phosphonoacetate, NaH, THF.
nique. Furthermore we have docked a-substituted chromen-based
analogues into the TACE enzyme to attain a clear percept of the
structure–activity relationship of this series of compounds.
lether (17a–l). The esters were converted to the racemic
tuted hydroxamate analogues (15b–l).
a-substi-
2. Methods
Next, all
a-alkoxy aromatic alkyl substituted chromen-based
analogues (15a–l) designed occupy S10 pocket of TACE enzyme
2.1. Synthesis
was evaluated for in vitro and in vivo activities.
The series of
a-substituted chromen-based analogues were
2.2. QSAR and docking study
prepared using the Wittig coupling reaction of the substituted
2-hydroxybenzaldehydes and phosphonates.18 The starting alde-
hydes were prepared from 2-hydroxybenzoic acids 2 (Scheme 1).
Esterification of 4-methyl-2-hydroxybenzoic acids (2a,b) yielded
methyl esters and they were protected with a p-methoxybenzyl
(PMB) group. The esters (3a,b) were converted to aldehyde (4a,b)
by reduction with lithium aluminum hydride (LAH) and the
following oxidation with pyridinium chlorochromate (PCC). The
4-fluorinated aldehyde 4c was prepared from the alcohol 5 by
reacting with paraformaldehyde and the following protection with
p-methoxybenzyl chloride (PMBCl).
Preparation of phosphonates for Wittig reaction is shown in
Scheme 2. The aldehydes (7) and 2,2-dimethyl-1,3-dioxane-4,6-
dione (meldrum’s acid, 8) was reacted with borane dimethylamine
in dry methanol to synthesize the 4,6-dione (9). Mannich type
reaction of the 4,6-dione (9) with N,N-dimethylmethylene ammo-
nium iodide (Eschenmoser’s salt)19 and the following reaction with
triethyl phosphonoacetate and sodium hydride (NaH) gave the
phosphonate (11).
2.2.1. Chemical and biological data
A series of 26
were used for the present study (see Supplementary data,
Fig. S1).18,20
-Alkoxy aromatic alkyl group substituted chromen-
a-substituted chromen-based TACE inhibitors
a
based analogues (S1a–k, S2a–o) were comprised of 18 molecules
in training set and eight molecules in test set. Another set of
molecules,
c-lactam hydroxamic acid derivatives (S3a–d), repre-
sented by DPC-333 (S3b)18,20 were taken from the literature and
used only as a test set for the QSAR model.
The biological activities of 26 molecules measured as IC50 which
ranging from 1 nV to 40 lM against human TACE enzyme. Then,
they were converted to pIC50 (pIC50 = log 1/IC50) so that they can
be used in QSAR as dependent variable where IC50 is the concentra-
tion of the inhibitor producing 50% inhibition of the enzyme. The
structure observed and predicted biological activities of the each
molecule are presented in Supplementary data (Tables S1–S3). Also
all of 26 molecules were prepared and minimized in Discovery
Studio 2.5. The energy minimization of each molecule was carried
out by the CHARMm force field of Discover program.
Wittig reaction with aldehyde (4a–c) gave Z-isomer 12a–i
(Scheme 3). p-Methoxybenzyl (PMB) group was deprotected with
hydrochloride and the subsequent cyclization gave the coumarin
ring (14a–i). The treatment of the ester 14a with potassium hydrox-
2.2.2. Calculation of descriptors and generation of QSAR models
QSAR analysis is an area of computational research which builds
models of biological activity using physicochemical properties of a
series of compounds. Different types of descriptors were calculated
for each molecule in the study table using Discover Studio 2.5
and preADMET. These descriptors include constitutional, electro-
static, geometrical, physicochemical, and topological descriptors
(Table S4).
ylamide gave the racemic
4). The benzyloxy group on aromatic ring at
a
-substituted hydroxamate 15a (Scheme
-position of coumarin
a
ring (14b–i) was deprotected with Pd/C under H2 atmosphereto give
the alcohol 16a–h. The various alkylated analogues of phenol 16a–h
were prepared by alkylation with 1-(2-chloroethyl)piperidine
hydrochloride or N,N-dialkylaminoethyl chloride to give phenethy-
QSAR model generation was performed by genetic function
approximation (GFA) technique with Discovery Studio 2.5. The
application of GFA algorithm allows the construction of high-qual-
ity predictive models and makes additional information not pro-
vided by standard regression techniques. GFA was performed
using 50,000 crossovers, 1500 population size, and smoothness
value of 2.0. The number of terms in the equation was set as initial
equation length 4 to maximum equation length 6. The set of equa-
tions generated were evaluated with the following notation: r2 is
the coefficient of determination; r2 (adj) is r2 adjusted for the num-
ber of terms in the model; r2 (pred) is the prediction (PRESS) r2,
equivalent to q2 from a leave-1-out cross-validation; Friedman
L.O.F. is the Friedman lack-of-fit score; S.O.R. P-value is the P-value
for significance of regression.
O
O
a, b
OH
OMe
OPMB
R
c, d
O
OH
R
H
3a R=5-Me
2a R=5-Me
3b R=4-Me
2b R=4-Me
OPMB
O
f
4aR=5-Me
4bR=4-Me
4cR=4-F
e
H
F
OH
F
OH
5
6
Scheme 1. Reagents and conditions: (a) H2SO4, MeOH, reflux; (b) PMBCl, K2CO3, KI,
acetone, 65 °C; (c) LiAlH4, THF; (d) PCC, SiO2, CH2Cl2; (e) paraformaldehyde, 10%
H2SO4, MeOH/toluene, reflux; (f) PMBCl, K2CO3, DMF.

8620
J. S. Yang et al. / Bioorg. Med. Chem. 18 (2010) 8618–8629
OPMB
EtO
O
O
O
P
O
EtO
EtO
a
R1
OEt
H
OEt
+
R1
O
n
OPMB
EtO
O
n
R2
R2
12a
12b
4a
n=3, R1=5-Me, R2=H
n=0, R1=5-Me, R2=4-OBn
R1= 5-Me
11a
11b
n=3, R2=H
n=0, R2=4-OBn
4b R1= 4-Me
4c
12c n=0, R1=5-Me, R2=3-OBn
R1= 4-F
11c n=0, R2=3-OBn
12d
12e
n=1, R1=5-Me, R2=4-OBn
n=1, R1=5-Me, R2=3-OBn
11d
11e
n=1, R2=4-OBn
n=1, R2=3-OBn
12f n=2, R1=5-Me, R2=4-OBn
11f n=2, R2=4-OBn
12g
12h
n=2, R1=5-Me, R2=3-OBn
n=1, R1=4-F, R2=4-OBn
11g
n=2, R2=3-OBn
12i n=1, R1=4-Me, R2=4-OBn
O
OH
O
OEt
R1
c
b
OEt
R1
O
O
n
O
n
EtO
R2
R2
13a
14a
n=3, R1=6-Me, R2=H
n=3, R1=5-Me, R2=H
13b n=0, R1=5-Me, R2=4-OBn
13c n=0, R1=5-Me, R2=3-OBn
14b n=0, R1=6-Me, R2=4-OBn
14c n=0, R1=6-Me, R2=3-OBn
13d
14d
n=1, R1=6-Me, R2=4-OBn
n=1, R1=5-Me, R2=4-OBn
13e n=1, R1=5-Me, R2=3-OBn
13f n=2, R1=5-Me, R2=4-OBn
14e n=1, R1=6-Me, R2=3-OBn
14f n=2, R1=6-Me, R2=4-OBn
13g
14g
n=2, R1=6-Me, R2=3-OBn
n=2, R1=5-Me, R2=3-OBn
13h n=1, R1=4-F, R2=4-OBn
13i n=1, R1=4-Me, R2=4-OBn
14h n=1, R1=7-F, R2=4-OBn
14i n=1, R1=7-Me, R2=4-OBn
Scheme 3. Reagents and conditions: (a) t-BuOLi, THF; (b) 4 N HCl, 1,4-dioxane; (c) xylene, 155 °C.
2.2.3. Docking study
has TACE inhibition at 0.3 nM of IC50 as well as cellular TNF-
inhibition, but the phenylbutyl analogue (15a) showed poor
cellular TNF- inhibitory activity with relatively weak TACE inhibi-
tion (IC50 = 33 nM). The length between the chromen ring and the
aromatic group of 15a is too long to fit in the S10 pocket of TACE.
Thus, the tertiary aminoethyl groups were introduced to 3- or
a
All the molecular docking studies were performed by the dock-
ing software LigandFit21 module of Discovery Studio 2.5 (Accelrys,
San Diego, CA). In this paper, the docking of chromen-based TACE
inhibitors into the active site of human TACE was performed. The
crystal structure of human TACE (pdb code: 3EDZ)22 obtained from
the Protein Data Bank was refined to remove water molecules and
to add hydrogen atoms to the whole enzyme under the condition
of pH 7.40. A binding pocket of the native ligand (methyl (1R,2S)-
2-(hydroxycarbamoyl)-1-{4-[(2-methylquinolin-4-yl)methoxy]ben-
zyl}cyclopropanecarboxylate) was selected as the binding site for
the study. The CFF (Consistent Force Field) function was selected
as the energy grid and Monte Carlo trial method was used to
roughly search the conformations when compounds are docked
into the TACE enzyme. Then the ligands were optimized using a
forcefield function, CHARMm. Thirty poses were docked for each
ligand. Among the docked conformations, the over-fitted confor-
mation was eliminated and the poses with higher consensus score
among LigScore1, LigScore2, PLP1, PLP2, Jain, PMF, and DockScore
were selected for further analysis.
a
4-hydroxyl group on the phenyl rings of the
(n = 0), phenethyl analogues (n = 1) and phenpropyl analogues
(n = 2) to explore the -substituent effect. The tertiary amino
a-benzyl analogues
a
group on the alkylated group was intended to increase the water
solubility. The prepared piperidinylethyl substituted analogues
(15c–h) showed almost same potencies at around 0.1 nM and
0.2 lM of IC50 in inhibition of TACE and cellular TNF-a, respec-
tively, except 15c. For the benzyl analogues (15c and 15d), 3-posi-
tion is better than 4-position for in vitro activities. Among these,
4-alkylated phenethyl analogue (15e) showed the best IC50 as
0.00104 and 0.20 lM for TACE and TNF-a inhibition, respectively.
And other close amino analogues of 15e were prepared. The
corresponding dimethylamino analogue (15i) and diethylamino
analogue (15j) were 1.8- and 1.4-fold more potent than 15e in
cellular TNF-a inhibition, respectively. Finally, 7-substituted
coumarin analogues (15k and 15l) were prepared to compare
in vitro activities to 6-substituent analogues and showed twofold
3. Results and discussion
more potent in cellular TNF-
pared 12 -substituted analogues, 15l showed the most potent
in vitro inhibitory activities, 0.28 nM and 60 nM of IC50 values for
TACE inhibition and the cellular TNF- inhibition, respectively.
a inhibition than 15i. Among the pre-
3.1. Pharmacological optimization
a
Series of the racemic
a-substituted analogues (15a–l) were
a
tested in vitro for inhibition of TACE enzyme and cellular TNF-
a
The most potent TACE inhibitor 15l shows very good selectivity
to MMP-1 by 1889 fold but some selectivity to MMP-3 and MMP-9
by only 29-fold and 73-fold, respectively (Table 2). The excellent
selectivity of 15l to MMP-1 gave a good potential to reduce the
possible toxicity of TACE inhibition. Though the structural
difference of S10 pocket between TACE and MMPs (MMP-2 and
and NO production (Table 1). All of the prepared analogues did
not inhibit the production of NO on RAW264.7 cells. Most of
compounds showed very strong TACE inhibitory activities ranging
0.3–30 nM of IC50 values, well correlated with their cellular TNF-
a
inhibitory activities. The 4-hydroxyphenylethyl analogue (15b)
showed good activity comparable to the benzyl analogue (1b). It

J. S. Yang et al. / Bioorg. Med. Chem. 18 (2010) 8618–8629
8621
O
O
OEt
a
NHOH
R1
R1
O
O
n
O
O
n
R2
R2
14a n=3, R1= 6-Me, R2=H
15a n=3, R1= 6-Me, R2=H
O
O
OEt
NHOH
R1
R1
O
O
O
O
n
n
a
R2
R2
b
O
14b-i
n=0-2,
R1= 6-Me, 7-Me, or 7-F,
R2=OBn
15b n=1, R1= 6-Me, R2=4-OH
OEt
R1
O
O
n
16a-h
n=0-2,
R1= 6-Me, 7-Me, or 7-F,
R2=4-OH
c
R2
O
O
a
NHOH
OEt
R1
R1
O
O
O
O
n
n
OR2
OR2
17a-l
15c
R1= 6-Me, n= 0; R2= 4-piperidinylethyl
n=0-2,
15d R1= 6-Me, n= 0; R2= 3-piperidinylethyl
R1= 6-Me, 7-Me, or 7-F,
R2=3-piperidinylethyl,
4-piperidinylethyl,
15e
15f
R1= 6-Me, n= 1; R2= 4-piperidinylethyl
R1= 6-Me, n= 1; R2= 3-piperidinylethyl
15g R1= 6-Me, n= 2; R2= 4-piperidinylethyl
4-N,N,-dimethylaminoethyl,
4-N,N,-diethylaminoethyl
15h
15i
R1= 6-Me, n= 2; R2= 3-piperidinylethyl
R1= 6-Me, n= 1; R2= 4-N,N,-dimethylaminoethyl
15j R1= 6-Me, n= 1; R2= 4-N,N,-diethylaminoethyl
15k
15l
R1= 7-F, n= 1; R2= 4-N,N,-diethylaminoethyl
R1= 7-Me, n= 1; R2= 4-N,N,-diethylaminoethyl
Scheme 4. Reagents and conditions: (a) NH2OK, MeOH; (i) LiOH, MeOH/THF; (ii) pentafluorophenol, DMAP, EDCI, CH2Cl2; (iii) NH2OK, MeOH; (b) Pd/C, H2, EtOH; (c) 1-(2-
chloroethyl)piperidine hydrochloride, Cs2CO3, NaI, DMF, 80–85 °C or 2-chloro-N,N-dialkylethanamine hydrochloride, NaH, DMF.
MMP-9)18 is the rationale for design of the selective TACE inhibitor,
long -substituents of chromen-based inhibitors increase the
inhibitory activity to TACE as well as MMPs. Thus, 15l is the dual
was relatively late, with Tmax occurring at 2.00 h. The apparent ter-
minal elimination phase in plasma appeared to begin at 5 h, with
an estimated half-life (t1/2) of 3.19 h. The oral bioavailability was
41.6% in rats.
a
functional inhibitor to TACE and MMPs.
Compound 15l showed higher efficacy than DPC-333 in in vivo
Finally, the pharmacological efficacy of 15l was evaluated using
the acute carrageenan induced arthritis model in SD rat (Fig. 2).
The edema reached to maximum 6 h after the induction and the
inhibition activity of each test group was determined at 6 h and
12 h after peripheral (ip) injection. The positive control using indo-
methacin and the two test groups (30 mg/kg and 100 mg/kg of 15l)
inhibited rat paw edema by 15%, 8%, and 20% at 6 h, and 4%, 14%,
and 35% at 24 h, respectively, compared to the vehicle. Accord-
ingly, it was proved that 15l effectively inhibited rat paw edema.
TNF-
than DPC-333 in cellular inhibition of TNF-
ally administered 30 mg/kg of 15l inhibited 74% of serum TNF-
a
inhibition experiment although it was five times less active
a
in RAW264.7 cells. Or-
a
production in LPS induced endotoxemia animal model whereas
DPC-333 showed 61% in the same condition (Table 3). The good
in vivo efficacy of 15l in po administration may be explained by
exploring the pharmacokinetic profiles. First, liver microsomal
stability, one of several major determining factors for the oral
bioavailability and systemic clearance was examined using human
liver microsome. As a result, 15l showed a long half-life (t1/2) at
73 min and good microsomal clearance (Table 4). Encouraged by
this in vitro result, in vivo pharmacokinetic profiles of 15l in
Sprague–Dawley (SD) rats were evaluated (Table 5). The systemic
plasma clearance (CL) was moderate with an average of 2.66 l/h/
kg. The apparent volume of distribution at steady state (Vss) was
estimated to be 3.04 l/kg indicating that 15l is likely to be readily
distributed inside of the vasculature. The apparent terminal
elimination phase in plasma began approximately 5–8 h after
dosing, with an average half-life (t1/2) of 1.14 h following the
intravenous administration. The oral absorption of 15l in rats
3.2. Docking and QSAR study
3.2.1. GFA models
In the present study, QSAR model was generated using the
training set of a-alkoxyaryl alkyl group substituted chromen-based
analogues. Test set with regularly distributed biological activities
was used to evaluate the predictive ability of the generated QSAR
models.
Six descriptors namely Balaban_index_JX, FNSA3, Log D,
Num_Atoms, Num_Chains, and WNSA2, significantly explain the var-
iance in the biological activity for
a-alkoxyaryl alkyl substituted

8622
J. S. Yang et al. / Bioorg. Med. Chem. 18 (2010) 8618–8629
Table 1
In vitro inhibitory activities of compounds 15a–l
O
NHOH
R1
O
O
n
R2
a
Compound
R1
n
R2
IC50 (lM)
TACE
TNF-ab
NOb
DPC-333
1b
—
—
0
3
1
0
0
1
1
2
2
1
1
1
1
—
H
H
OH
0.00010
0.00050
0.03282
0.00030
0.01094
0.00107
0.00104
0.00090
0.00107
0.00027
0.00066
0.00053
0.00046
0.00028
0.012
0.52
>3.0
0.92
1.5
0.26
0.20
0.24
0.32
0.23
0.11
0.14
0.07
0.06
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
6-Me
6-Me
6-Me
6-Me
6-Me
6-Me
6-Me
6-Me
6-Me
6-Me
6-Me
7-F
15a
15b
15c
15d
15e
15f
15g
15h
15i
4-(2-Piperidin-1-yl-ethoxy)
3-(2-Piperidin-1-yl-ethoxy)
4-(2-Piperidin-1-yl-ethoxy)
3-(2-Piperidin-1-yl-ethoxy)
4-(2-Piperidin-1-yl-ethoxy)
3-(2-Piperidin-1-yl-ethoxy)
4-(2-Dimethylamino-ethoxy)
4-(2-Diethylamino-ethoxy)
4-(2-Dimethylamino-ethoxy)
4-(2-Dimethylamino-ethoxy)
15j
15k
15l
7-Me
a
Values are means of three experiments.
RAW264.7 cells were used, NA = no inhibition at 10 lM.
b
Table 2
Table 5
In vitro inhibitory activities of compound 15l and DPC-333
In vivo pharmacokinetic profiles of 15l
a
Compound
IC50
MMP-1
(l
M)
Parameter
Unit
iv (N = 3)
po (N = 3)
TACE
MMP-3
MMP-9
Dose
Tmax
Cmax
AUC0–inf
CL
Vss
t1/2
MRTinf
F
(mg/kg)
(h)
5.00
—
—
1.88
2.66
3.04
2.80
1.14
—
30.00
2.00
1.14
4.69
—
15l
DPC-333
0.00031
0.00010
557
4926
8.7
3.9
23
460
(
lg/ml)
(h
lg/ml)
a
(l/h/kg)
(l/kg)
(h)
(h)
(%)
Values are means of three experiments.
—
3.19
2.90
41.61
Table 3
In vivo inhibition of LPS induced TNF-
a by 15l and DPC-333
Sprague–Dawley (SD) rats were used.
Compound
Dose (mg/kg)
% Inhibitiona
**
**
15l
DPC-333
30 (po)
30 (po)
74%
61%
80% of the variance. The correlation coefficient (r2) between ob-
served and predicted activity of test set was found to be 0.7719
(Fig. 3). The constitutional descriptors, Num_Atoms and Num_-
Chains, positively correlate with biological activity. It underlines
the importance of size of molecules. FNSA3 and WNSA2 which are
electrostatic descriptors, denote the fractional negative surface
area computed with quantum charges and weighted negatively
charged partial surface area, respectively. Though these descriptors
have similar components, the equation (1) is weighted more on
FNSA3 rather than WNSA2 as shown on coefficients.
Balaban_index_JX is a topological descriptor weighted on relative
electronegativity, and Log D is the octanol–water partition coeffi-
cient calculated by taking the ionization states of the molecule
which represents lipophilicity of a molecule into account.
LPS (1.5 mg/kg, ip) was pre-treated twice (ip) at 24 h and 0.5 h before 15l and DPC-
333 were orally administered. Serum TNF-a
was measured after 60 min. *P <0.05;
**P <0.01.
a
BALB/c mice (four per group) were treated.
Table 4
Human liver microsomal stability of 15l
Compound
t1/2 (min)
Microsomal clearance (lL/mim/mg)
15l
DPC-333
73.0
94.5
9.5
7.3
15l and DPC-333 (40 lM) was incubated with human liver microsomes (1 mg/mL)
in the presence of NADPH (1 mM), followed by liquid chromatography MS/MS mass
spectrometry analysis. The analytical data were processed by Metabolite ID™
program.
3.2.2. Docking study and binding mode analysis
To understand the interactions between the TACE and its inhib-
itors and to explore their binding mode, docking study was
performed using LigandFit21 of Discovery Studio 2.5. The active site
cleft of TACE is deep and contains zinc, and the right-hand side of
the zinc corresponds to the S10, S20, and S30 pocket. The ligand–
enzyme interaction analysis shows that Thr347, Leu348, Gly349,
His405, Glu406, and His409 are the important residues at the
chromen-based analogues in equation (1) (Table 6). The variance of
this model was found to be 88%. The equation (1) also shows good
correlation coefficient for leave-one-out validation (r2 (pred)) with

J. S. Yang et al. / Bioorg. Med. Chem. 18 (2010) 8618–8629
8623
350
300
250
200
150
100
FNSA3 stands for the sum of each surface area for a negatively
charged atom multiplied by negative charge divided by the total
surface area. Because it is multiplied by a negative charge, FNSA3
has negative value. Thus surface area for a negatively charged atom
should be increased for better activity. The receptor–ligand inter-
action is stabilized by van der Waals interaction between alkoxy
* p < 0.05
A
moiety of
a-alkoxy aromatic alkyl group substituent and positive
charged S30. This is also proved by negative correlation of Log D,
signifying lipophilicity of a molecule, to the biological activity.
The docking study of 15l, which showed the best in vitro and
in vivo pharmacokinetic profiles among 12 synthesized analogues,
also fitted well into the QSAR equation. (Fig. 5) Moreover, the (S)-
isomer of 15l was docked into the catalytic domains of the human
TACE and MMP-3 (1G49.pdb)23 enzyme. The docking data suggest
that the hydroxamic acid moiety of 15l binds the zinc metal in the
active sites of TACE and MMP-3 enzyme. The carbonyl group in
chromen ring interacts with the hydroxyl group of Thr347 of TACE
through hydrogen bonding whereas the carbonyl group of chro-
men ring forms a hydrogen bonding with the Leu664 of MMP-3.
This difference on the docking poses between TACE and MMP-3 ex-
plains the TACE selectivity of 15l over MMP-3. It should also be
*
Control
Indomethacin
15l 100 mpk
15l 30 mpk
0 hr
2 hr
4 hr
6 hr
8 hr
24 hr
Time after carrageenan injection (h)
120
100
80
60
40
20
0
* p < 0.01
** p < 0.05
*** p < 0.001
B
6 hr
24 hr
*
pointed out that the only (S)-configuration of
a-position of
**
hydroxamic acid would be allowed to fit into the empty S10 pocket
of TACE and could afford an extra stabilization through the hydro-
***
gen bonding between the oxygen at the
a-position of hydroxamic
acid and the hydrogen of imidazole ring of His405, and the pi–pi
interaction between
His405 of TACE.
a-aromatic ring of 15l and imidazole ring of
Finally, the docking study suggests that the
a
-alkoxy aromatic
alkyl group substituted chromen-based analogues are stabilized
by binding inside the active cleft (S10 and S30) of TACE, due to the
presence of an extra hydrogen bond and van der Waals interaction.
Thus, the QSAR equation may be useful for the design and develop-
Control
Indomethacin 15l 100 mpk
15l 30 mpk
ment of novel
a-substituted chrome-based TACE inhibitor having
Figure 2. (A) Effects of 15l on carrageenin induced paw edema model in SD rat. (B)
Effects of inhibition of indomethacin, 15l (30 mg/kg) and 15l (100 mg/kg) on paw
edema at 6 h and 24 h after carrageenan induction. The edema reached maximum
6 h after the induction and the inhibition activity compared to vehicle was
determined at 6 h and 12 h after peripheral (ip) injection. Indomethacin (10 mg/kg)
was used for the positive control. (*P <0.05; **P <0.01).
better inhibitory activity.
3.2.3. Validation
For the external validation, four c-lactam hydroxamate deriva-
tives18,20 were used as an additional test set. It has been previously
shown that van der Waals interaction between the fractional nega-
tive surface area of
acid residues inside the S10 pocket increases biological activity of
TACE. Thus, four -lactam hydroxamate derivatives were selected
because of the similarity in -alkoxy aromatic ring. (Table S3) The
test set was applied in equation (1) and the squared regression coef-
ficient (r2) resulted as 0.8849. (Fig. 3) The model is illustrated on a
specific case, comparing S3d with S3a and S3c. Compound S3d has
a-alkoxyaryl alkyl substituent and the amino
active site and are the main contributors to the receptor–ligand
interaction.
c
For
a
-alkoxyaryl alkyl group substituted chromen-based ana-
-substituent comes into
a
logues which are bulky and long, the
a
the deep S10 pocket merged with S30 pocket. The chromen ring
extends away from the enzyme, but still the carbonyl group in
chromen ring forms hydrogen bond with the amide nitrogen
proton of Leu348 and/or hydroxyl group of Thr347. It also has
pi–pi interaction between imidazole ring of His405 of TACE and
benzyloxy group for alkoxy aromatic moiety of
a-position when
S3a and S3c has electron density rich atom. The negative correlation
of FNSA3 and Log D could be elucidated and demonstrated by this
example. Another set of compound, S3b and S3c, could also validate
the equation with size-related descriptors, Num_Atoms and
Num_Chains. Compound S3b, which has bulky group (iso-butyl) that
may place on S1 pocket of TACE, showed better inhibitory activity
than compound S3c which has simple methyl group (Table S3).
a
-aromatic ring of ligand. Another hydrogen bond has been formed
between oxygen on -alkoxyaromatic ring and amine nitrogen
proton on imidazole ring of His405. Thus -alkoxy aromatic alkyl
a
a
group substituents fit well in the pocket of the enzyme through
S10 to S30. (Fig. 4) While S10 pocket has hydrophobic surface inside,
partial positive surface was observed in adjacent to S30 pocket.
Table 6
QSAR equations generated using genetic function approximation for the training set of the molecules in each group
No. Equation
r2
r2
r2
RMS
Friedman S.O.R. P-
(adj)
(pred) residual
error
L.O.F.
value
1
pIC50 = ꢀ120.91 + 12.369 (Balaban_index_JX) ꢀ 736.95 (FNSA3) ꢀ 0.85767 (Log D) + 1.5867
(Num_Atoms) + 1.0541 (Num_Chains) + 0.070294 (WNSA2)
0.9214 0.8785 0.7978 0.2687
0.1045
1.79 Eꢀ05

8624
J. S. Yang et al. / Bioorg. Med. Chem. 18 (2010) 8618–8629
12
11.5
11
10.5
10
R² = 0.8849
R² = 0.9214
9.5
9
8.5
8
R² = 0.7719
7.5
7
7
7.5
8
8.5
Activity
9
9.5
10
10.5
Figure 3. A graph of actual versus predicted activities of training set (ꢀ) and test sets (j and N (S3)) of molecules using equation (1). Balaban_index_JX: relative
electronegativity; FNSA3: fractional charged partial negative surface area third type; Log D: the octanol–water partition coefficient calculated taking into account the
ionization states of the molecule; Num_Atoms: heavy atoms; Num_Chains: unbranched chains needed to cover all the non-ring bonds in the molecule; WNSA2: surface
weighted charged partial negative surface area second type.
Figure 4. Docked conformations of
a-alkoxyaryl alkyl group substituted chromen-
based analogues (S1a, S1i, S2k, S2m, and S2n): The
a
-substituents of TACE
inhibitors binds through the S10 to S30 pocket, which is the active site of TACE
enzyme. And the hydrogen bond between the carboxyl group of the chromen ring
and Thr347 or Leu348 of TACE enzyme fixes the chromen ring on the surface of S1
Figure 6. Docked conformations of
and S3d).
c-lactam hydroxamic acid derivatives (S3a, S3c,
pocket. Depending on the
a-substituents, the chromen ring could be rotated and
placed on different directions on the surface of S1 pocket.
moiety at
a-position of c-lactam hydroxamates give extra stabil-
ization for the receptor–ligand interaction.
4. Conclusion
A series of
a-substituted chromen-core TACE inhibitors were
rationally designed for filling in S10 pocket of the enzyme. Twelve
analogues were synthesized by varying the chain length and
alkylation pattern on aromatic ring of side chain. Most of
compounds were active in vitro TACE enzyme inhibition as well
as cellular TNF-
a inhibition. Among these, 15l showed most
promising in vitro and in vivo activity profiles. Based on these re-
sults, 15l was evaluated on in vitro and in vivo pharmacokinetic
studies and found moderate systemic clearance and good oral
bioavailability at 42%. It was also evaluated in carrageenan induced
RA animal model and found that it effectively inhibited paw edema
in rat. QSAR study and docking were also performed to confirm the
Figure 5. The docking model for the (S)-a-substituted chromen analogue (15l)
binding to the binding pocket of TACE.
series of a-alkoxyaryl alkyl substituted chromen-based analogues.
a
-Substituent with long and bulky groups of coumarin core TACE
inhibitors, gets inside the S10 and S30 pocket and formed van der
Waals interaction so that increases the inhibitory activity. Hence,
According to the result, pyrrolidinone in
mates would replace alkyl group in -alkoxy aromatic alkyl substi-
tuted chromen-based analogues. The binding modes of -lactam
hydroxamates also support the result. (Fig. 6) The carbonyl group
on pyrrolidinone ring also plays a role as a hydrogen bond acceptor
like the carbonyl group on chromen ring. And the alkoxy aromatic
c-lactam hydroxa-
a
the docking studies suggest that a-alkoxyaryl alkyl group shows
c
great activity by stabilizing receptor–ligand interaction. Also the
docking study of 15l showed the dual function inhibitory activity
to TACE and MMP-3. On the basis of the descriptors thus obtained
from the QSAR equation, novel a-substituted chromen-based TACE

J. S. Yang et al. / Bioorg. Med. Chem. 18 (2010) 8618–8629
8625
inhibitors can be designed and developed that are predicted to at-
tain improved TACE inhibitory activity.
400 MHz) d 7.44 (d, J = 8.4 Hz, 2H), 7.11 (d, J = 2.0 Hz, 1H), 7.06
(d, J = 8.4 Hz, 2.0, 1H), 6.92 (d, J = 8.4 Hz, 2H), 6.86 (d, J = 8.4 Hz,
1H), 5.02 (s, 2H), 4.67 (s, 2H), 3.82 (s, 3H), 2.46 (br, 1H), 2.30 (s, 3H).
PCC (42.5 g, 197.4 mmol, 3.0 equiv) and SiO2 (42.5 g) were
added to a solution of the above alcohol (17.0 g, 65.8 mmol,
1.0 equiv) in methylene chloride (200 mL). The mixture was stirred
at room temperature for 1 h. The reaction mixture was filtered
through pad of Celite. The filtrate was concentrated and purified
by column chromatography on silica gel (hexane/ethyl ace-
tate = 3:1) to give compound 4a (16.4 g, 97%); 1H NMR (CDCl3,
400 MHz) d 10.49 (s, 1H), 7.64 (d, J = 2.0 Hz, 1H), 7.36–7.34 (m,
3H), 6.96 (d, J = 8.8 Hz, 2H), 6.92 (d, J = 8.8 Hz, 1H), 5.09 (s, 2H),
3.82 (s, 3H), 2.31 (s, 3H); MS (ESI) m/z 279 (MNa+).
5. Experimental
All chemicals were obtained from commercial suppliers and
used without further purification. Flash column chromatography
was performed with silica (Merck EM9385, 230–400 mesh).
1H and 13C NMR spectra were recorded at 300 and at 75 MHz,
respectively. Proton and carbon chemical shifts are expressed in
ppm relative to internal tetramethylsilane, and coupling constants
(J) are expressed in Hertz. LC/MS (liquid chromatography–mass
spectrometry) spectra were recorded by electrospray ionization
(ESI) on Shimazu LC/MS instruments (20% 0.1% TFA in H2O/0.1%
TFA in acetonitrile) scan (from 0 to 600 amu/z) mode and the
detected ion peaks are (M+z)/z and (Mꢀz)/z in positive and
negative ion mode, respectively, where M represents the molecular
weight of the compound and z the charge (number of protons). LC/
MS/MS were recorded by electrospray ionization (ESI) on Agilent
1100/API2000 system. HRMS (High resolution mass spectrometry)
were obtained from fast atom bombardment (FAB) on JMS-700
(Jeol, Japan) High Resolution Tandem Mass Spectrometry in Korea
Basic Science Institute in Seoul, Korea. Final products were purified
to a minimum purity of 96%, as determined by reversed phase
HPLC (see Supplementary data: Table S5).
5.1.3. 2-(4-Methoxybenzyloxy)-4-methylbenzaldehyde (4b)
1H NMR (CDCl3, 400 MHz) d 10.45 (s, 1H), 7.75 (d, J = 8.0 Hz,
1H), 7.37 (d, J = 8.4 Hz, 2H), 6.94 (d, J = 8.4 Hz, 2H), 6.87 (s, 1H),
6.85 (d, J = 8.0 Hz, 1H), 5.10 (s, 2H), 3.83 (s, 3H), 2.40 (s, 3H); MS
(ESI) m/z 279 (MNa+).
5.1.4. 4-Fluoro-2-hydroxybenzaldehyde (6)
3-Fluorophenol (5, 5.0 ml, 56.0 mmol, 1.0 equiv) was added to
8 wt % magnesium methoxide solution in MeOH (44.5 mL) and
the mixture was distilled. Approximately, half of the methanol
was distilled off and toluene (42 mL) was added and the azeotropic
mixture was removed by fractional distillation until the tempera-
ture of the reaction mixture rose to 95 °C. Slurry of paraformalde-
hyde powder (5.18 g, 172.5 mmol, 3.1 equiv) in toluene (10 mL)
was added in small portion over 1 h with concurrent removal of
volatile materials by distillation. The mixture was stirred at 95 °C
for 1 h and cooled to room temperature. The reaction mixture
was added slowly to 10% sulfuric acid solution (63 mL) and stirred
at 30–40 °C for 2 h. The reaction mixture was extracted with
toluene, washed with 10% sulfuric acid solution and water, dried
over anhydrous MgSO4, filtered, and concentrated in vacuo. The
residue was purified by column chromatography on silica gel
(hexane/ethyl acetate = 8:1) to give compound 6 (4.05 g, 52%);
1H NMR (CDCl3, 400 MHz) d 11.37 (s, 1H), 9.84 (s, 1H), 7.57 (dd,
J = 8.6, 6.4 Hz, 1H), 6.76–6.66 (m, 2H).
5.1. Procedures for 4a–c
5.1.1. Methyl 2-(4-methoxybenzyloxy)-5-methylbenzoate (3a)
2-Hydroxy-5-methylbenzoic acid (2a, 12.0 g, 78.9 mmol,
1.0 equiv) was dissolved in MeOH (100 mL) and H2SO4 (1.2 mL)
was added at room temperature. The mixture was heated at reflux
for 2 days (TLC). The mixture was cooled to room temperature.
NaHCO3 Solution was added to reaction mixture and then the
mixture was extracted with ethyl acetate. The organic layer was
washed with brine, dried over anhydrous MgSO4, filtered, and
concentrated. The residue was purified by column chromatography
on silica gel (hexane/ethyl acetate = 3:1) to give the ester (12.0 g,
91%); 1H NMR (CDCl3, 400 MHz)
d 10.57 (s, 1H), 7.61 (d,
J = 2.0 Hz, 1H), 7.25 (d, J = 8.4 Hz, 2.0, 1H), 6.87 (d, J = 8.4 Hz, 1H),
3.92 (s, 3H), 2.27 (s, 3H).
5.1.5. 4-Fluoro-2-(4-methoxybenzyloxy)benzaldehyde (4c)
4-Methoxybenzyl chloride (0.85 mL, 6.6 mmol, 1.1 equiv) was
added to a solution of 59 (0.85 g, 6.0 mmol, 1.0 equiv), K2CO3
(2.50 g, 18.1 mmol, 3.0 equiv) in DMF (6 mL) at room temperature.
The mixture was stirred at room temperature for 12 h. Water was
added to the reaction mixture and then the mixture was extracted
with ethyl acetate. The organic layer was washed with brine, dried
over anhydrous MgSO4, filtered, and concentrated. The residue was
purified by column chromatography on silica gel (hexane/ethyl
acetate = 6:1) to give compound 4c (1.45 g, 92%); 1H NMR (CDCl3,
400 MHz) d 10.40 (s, 1H), 7.87 (dd, J = 8.6, 6.4 Hz, 1H), 7.36
(d, J = 8.4 Hz, 2H), 6.94 (d, J = 8.4 Hz, 2H), 6.78–6.71 (m, 2H), 5.09
(s, 2H), 3.83 (s, 3H); MS (ESI) m/z 261 (MH+).
4-Methoxybenzyl chloride (10.7 mL, 78.8 mmol, 1.1 equiv) was
added to
a solution of the above ester (11.9 g, 71.6 mmol,
1.0 equiv), K2CO3 (29.7 g, 214.8 mmol, 3.0 equiv) and KI (1.2 g,
7.16 mmol, 0.1 equiv) in acetone (100 mL) at room temperature.
The mixture was heated at 65 °C for 12 h (TLC). The mixture was
cooled at room temperature. Water was added to the reaction
mixture and then the mixture was extracted with ethyl acetate.
The organic layer was washed with brine, dried over anhydrous
MgSO4, filtered, and concentrated. The residue was purified by
column chromatography on silica gel (hexane/ethyl acetate = 4:1)
to give compound 3a (19.0 g, 93%); 1H NMR (CDCl3, 400 MHz) d
7.62 (d, J = 2.0 Hz, 1H), 7.41 (d, J = 8.8 Hz, 2H), 7.22 (d, J = 8.4 Hz,
2.0, 1H), 6.92–6.90 (m, 3H), 5.08 (s, 2H), 3.89 (s, 3H), 3.81 (s, 3H),
2.30 (s, 3H); MS (ESI) m/z 309 (MNa+).
5.2. General procedures for 9
Meldrum’s acid (8, 1.0 equiv) was added to a solution of borane
dimethylamine complex (1.0 equiv) in MeOH (0.5–1 M solution).
Benzaldehyde 7 (3.0 equiv) in MeOH (1 M solution) was added
dropwise to the mixture and then stirred at room temperature
for 20 min and the reaction mixture was stirred at room tempera-
ture for 30 min. Ice water was added to the reaction mixture and
the solution was acidified with 2 N HCl to pH 1. The mixture was
stirred at 0 °C for 1 h and the solid was filtered and washed with
water to give compound 9 in 60–80% yield.
5.1.2. 2-(4-Methoxybenzyloxy)-5-methylbenzaldehyde (4a)
LAH (2.52 g, 66.4 mmol, 1.0 equiv) was added to a solution of 3a
(19.0 g, 66.4 mmol, 1.0 equiv) in THF (200 mL) at 0 °C. The reaction
mixture was warmed to room temperature and then stirred for
30 min. H2O (2.5 mL), 15% NaOH aqueous solution (2.5 mL) and
H2O (2.5 mL) were added sequently to the reaction mixture. The
white solid was filtered through pad of Celite and the solvent
was evaporated to give the alcohol (17.1 g, 99%); 1H NMR (CDCl3,

8626
J. S. Yang et al. / Bioorg. Med. Chem. 18 (2010) 8618–8629
5.2.1. 2,2-Dimethyl-5-(4-phenylbutyl)-1,3-dioxane-4,6-dione
1H NMR (CDCl3, 400 MHz)
7.29–7.16 (m, 5H), 3.49 (t,
5.4.2. Diethyl 2-(diethoxyphosphoryl)-4-(4-phenylbutyl)-
pentanedioate
d
J = 5.2 Hz, 1H), 2.63 (t, J = 7.6 Hz, 2H), 2.16–2.11 (m, 2H), 1.77 (s,
1H NMR (CDCl3, 400 MHz) d 7.29–7.15 (m, 5H), 4.21–4.06 (m,
8H), 3.02–2.92 (m, 1H), 2.61–2.03 (m, 5H), 1.73–1.47 (m, 6H),
1.39–1.20 (m, 12H); MS (ESI) m/z 479 (MNa+).
3H), 1.75 (s, 3H), 1.72–1.49 (m, 4H); MS (ESI) m/z 277 (MH+).
5.2.2. 5-(4-(Benzyloxy)benzyl)-2,2-dimethyl-1,3-dioxane-
4,6-dione
5.5. General procedures for 12
1H NMR (CDCl3, 400 MHz) d 7.43–7.32 (m, 5H), 7.23 (d,
J = 8.8 Hz, 2H), 6.89 (d, J = 8.8 Hz, 2H), 5.04 (s, 2H), 3.72 (t,
J = 4.8 Hz, 1H), 3.44 (d, J = 4.8 Hz, 2H), 1.72 (s, 3H), 1.46 (s, 3H);
MS (ESI) m/z 341 (MH+).
One molar t-BuOLi (1.5 equiv, THF solution) was added to a
solution of 11 (1.5 equiv) in THF (0.2 M solution) and the mixture
was stirred at room temperature for 30 min. Compound
(1.0 equiv) was added and then the reaction mixture was stirred
at room temperature for 12 h. The reaction mixture was extracted
with ethyl acetate. The organic layer was washed with brine, dried
over anhydrous MgSO4, filtered, and concentrated. The residue was
purified by column chromatography on silica gel (hexane/ethyl
acetate = 3:1) to give 12 in 80–90% yield.
4
5.2.3. 5-(3-(Benzyloxy)benzyl)-2,2-dimethyl-1,3-dioxane-
4,6-dione
1H NMR (CDCl3, 400 MHz) d 7.44–7.18 (m, 6H), 6.97–6.84 (m,
3H), 5.05 (s, 2H), 3.74 (t, J = 4.8 Hz, 1H), 3.46 (d, J = 4.8 Hz, 2H),
1.73 (s, 3H), 1.50 (s, 3H); MS (ESI) m/z 341 (MH+).
5.5.1. Diethyl 2-(2-(4-methoxybenzyloxy)-5-methylbenzylidene)-
4-(4-phenylbutyl)pentane-dioate (12a)
5.2.4. 5-(3-(Benzyloxy)phenethyl)-2,2-dimethyl-1,3-dioxane-
4,6-dione
1H NMR (CDCl3, 400 MHz) d 7.87 (s, 1H), 7.31 (d, J = 8.4 Hz, 2H),
7.26–7.10 (m, 6H), 7.06 (d, J = 8.4 Hz, 1H), 6.89 (d, J = 8.4 Hz, 2H),
6.83 (d, J = 8.4 Hz, 1H), 4.98 (s, 2H), 4.25 (q, J = 7.2 Hz, 2H), 4.06–
3.94 (m, 2H), 3.80 (s, 3H), 2.86–2.64 (m, 3H), 2.51 (t, J = 7.6 Hz,
2H), 2.29 (s, 3H), 1.57–1.41 (m, 4H), 1.33 (t, J = 7.2 Hz, 3H), 1.27–
1.16 (m, 2H), 1.12 (t, J = 7.2 Hz, 3H); MS (ESI) m/z 581 (MNa+).
1H NMR (CDCl3, 400 MHz) d 7.45–7.20 (m, 6H), 6.88–6.83 (m,
3H), 5.05 (s, 2H), 3.48 (t, J = 5.2 Hz, 1H), 2.84 (t, J = 7.5 Hz, 2H),
2.43–2.38 (m, 2H), 1.76 (s, 3H), 1.73 (s, 3H).
5.3. General procedures for 10
5.5.2. Diethyl 2-(4-(benzyloxy)benzyl)-4-(2-(4-methoxybenzyl-
oxy)-5-methylbenzylidene)-pentanedioate (12b)
A mixture of 9 (1.0 equiv) and N,N-dimethylmethyleneiminium
iodide (2.5 equiv) in EtOH (0.5 M solution) was stirred at 85 °C.
After 4.5 h, reaction mixture was concentrated and extracted with
ethyl acetate. The organic layer was washed with NaHCO3 solution,
dried over anhydrous MgSO4, filtered, and concentrated to give
compound 10 in 70–90% yield.
1H NMR (CDCl3, 400 MHz) d 7.89 (s, 1H), 7.42–7.39 (m, 8H),
7.08–6.97 (m, 4H), 6.87–6.79 (m, 4H), 5.00 (s, 2H), 4.98 (s, 2H),
4.26–4.21 (m, 2H), 3.94–3.85 (m, 2H), 3.79 (s, 3H), 2.97–2.63 (m,
5H), 2.29 (s, 3H), 1.34–1.25 (m, 3H), 1.03–0.99 (m, 3H).
5.5.3. Diethyl 2-(4-(benzyloxy)phenethyl)-4-(2-(4-methoxyben-
zyloxy)-4-methylbenzyl-idene)pentanedioate (12i)
5.3.1. Ethyl 2-methylene-6-phenylhexanoate
1H NMR (CDCl3, 400 MHz) d 7.29–7.16 (m, 5H), 6.12 (d,
J = 1.2 Hz, 1H), 5.49 (d, J = 1.2 Hz, 1H), 4.20 (q, J = 7.2 Hz, 2H),
2.63 (t, J = 7.6 Hz. 2H), 2.33 (t, J = 7.6 Hz, 2H), 1.67–1.50 (m, 4H),
1.29 (t, J = 7.2 Hz, 3H); MS (ESI) m/z 233 (MH+).
1H NMR (CDCl3, 400 MHz) d 7.89 (s, 1H), 7.43–7.30 (m, 7H), 7.18
(d, J = 7.6 Hz, 1H), 6.99 (d, J = 8.8 Hz, 2H), 6.89 (d, J = 8.8 Hz, 2H),
6.83 (d, J = 8.8 Hz, 2H), 6.78 (d, J = 7.6 Hz, 1H), 6.74 (s, 1H), 5.01
(s, 2H), 4.96 (s, 2H), 4.23–3.98 (m, 4H), 3.80 (s, 3H), 2.87–2.69
(m, 3H), 2.44–2.39 (m, 2H), 2.34 (s, 3H), 1.90–1.60 (m, 2H), 1.31–
1.18 (m, 6H); MS (ESI) m/z 659 (MNa+).
5.3.2. Ethyl 2-(4-(benzyloxy)benzyl)acrylate
1H NMR (CDCl3, 400 MHz) d 7.44–7.32 (m, 5H), 7.18 (d,
J = 8.8 Hz, 2H), 6.91 (d, J = 8.8 Hz, 2H), 6.20 (d, J = 1.4 Hz, 1H),
5.44 (d, J = 1.4 Hz, 1H), 5.04 (s, 2H), 4.18 (q, J = 7.2 Hz, 2H), 3.57
(s, 2H), 1.27 (t, J = 7.2 Hz, 3H); MS (ESI) m/z 297 (MH+).
5.6. General procedures for 13
4 N HCl (1 equiv, in 1,4-dioxane solution) was added to a solu-
tion of 12 in 1,4-dioxane (0.1 M solution) and the mixture was stir-
red at room temperature for 3 h. The reaction mixture was
extracted with ethyl acetate. The organic layer was washed with
brine, dried over anhydrous MgSO4, filtered, and concentrated.
The residue was purified by column chromatography on silica gel
(hexane/ethyl acetate = 3:1) to give compound 13 in 70–80% yield.
5.4. General procedures for 11
NaH (2.5 equiv) was added to a solution of triethyl phosphono-
acetate (1.0 equiv) in THF (0.5 M solution) at 0 °C. After 30 min, 10
(1.0 equiv) in THF (1 M solution) was added and the reaction mix-
ture was stirred at room temperature for 12 h. The mixture was
poured into ice water and extracted with ethyl acetate. The organic
layer was washed with brine, dried over anhydrous MgSO4,
filtered, and concentrated. The residue was purified by column
chromatography on silica gel (hexane/ethyl acetate = 1:1) to give
compound 11 in 40–50% yield.
5.6.1. Diethyl 2-(2-hydroxy-5-methylbenzylidene)-4-(4-phenyl-
butyl)pentanedioate (13a)
1H NMR (CDCl3, 400 MHz) d 7.66 (s, 1H), 7.27–7.00 (m, 6H), 6.92
(s, 1H), 6.78 (d, J = 8.4 Hz, 1H), 5.66 (s, 1H), 4.29–3.92 (m, 4H),
2.73–2.59 (m, 3H), 2.53 (t, J = 7.6 Hz, 2H), 2.27 (s, 3H), 1.61–1.20
(m, 6H), 1.34 (t, J = 7.2 Hz, 3H), 1.13 (t, J = 7.2 Hz, 3H); MS (ESI)
m/z 461 (MNa+).
5.4.1. Diethyl 2-(4-(benzyloxy)benzyl)-4-(diethoxyphosphoryl)-
pentanedioate
1H NMR (CDCl3, 400 MHz) d 7.43–7.32 (m, 5H), 7.06 (d,
J = 8.0 Hz, 2H), 6.87 (d, J = 8.0 Hz, 2H), 5.03 (s, 2H), 4.21–4.02 (m,
8H), 3.10–2.05 (m, 6H), 1.35–1.24 (m, 12H); MS (ESI) m/z 543
(MNa+).
5.6.2. Diethyl 2-(4-(benzyloxy)benzyl)-4-(2-hydroxy-5-methyl-
benzylidene)pentanedioate (13b)
1H NMR (CDCl3, 400 MHz) d 7.69 (s, 1H), 7.43–7.30 (m, 5H),
7.03–6.77 (m, 7H), 5.79 (br, 1H), 5.01 (s, 2H), 4.27–4.22 (m, 2H),

J. S. Yang et al. / Bioorg. Med. Chem. 18 (2010) 8618–8629
8627
4.01–3.84 (m, 2H), 2.99–2.60 (m, 5H), 2.27 (s, 3H), 1.32 (t,
J = 7.2 Hz, 3H), 1.26 (t, J = 7.2 Hz, 3H).
(q, J = 7.2 Hz, 2H), 2.93–2.58 (m, 5H), 1.99–1.77 (m, 2H), 1.16 (t,
J = 7.2 Hz, 3H); MS (ESI) m/z 407 (MNa+).
5.6.3. Diethyl 2-(4-(benzyloxy)phenethyl)-4-(2-hydroxy-4-
methylbenzylidene)pentane-dioate (13i)
5.9. General procedures for 17
1H NMR (CDCl3, 400 MHz) d 7.69 (s, 1H), 7.44–7.29 (m, 6H),
7.03–6.98 (m, 2H), 6.89–6.85 (m, 2H), 6.73 (d, J = 8.0 Hz, 1H),
6.69 (s, 1H), 5.65 (br, 1H), 5.03 (s, 2H), 4.26–3.93 (m, 4H), 2.82–
2.42 (m, 5H), 2.30 (s, 3H), 1.92–1.65 (m, 2H), 1.30 (t, 7.0 Hz, 3H),
1.17 (t, J = 7.0 Hz, 3H); MS (ESI) m/z 539 (MNa+).
A mixture of 16 (1.0 equiv), 1-(2-chloroethyl)piperidine hydro-
chloride (1.5 equiv), NaI (1.5 equiv) and Cs2CO3 (3.0 equiv) in DMF
(0.2 M solution for 16) was stirred at 85 °C. After 6 h, the reaction
mixture was concentrated. The residue was purified by column
chromatography on silica gel (chloroform/methanol = 8:1) to give
compound 17a–f in 60–80% yield.
5.7. General procedures for 14
5.9.1. Ethyl 3-(6-methyl-2-oxo-2H-chromen-3-yl)-2-(4-(2-(piperi-
din-1-yl)ethoxy)benzyl)-propanoate (17a)
A solution of 14 in xylene (0.1 M solution) was stirred in sealed
tube at 155 °C for 2 days and then the reaction mixture was con-
centrated. The residue was purified by column chromatography
on silica gel (hexane/ethyl acetate = 6:1) to give compound 14 in
50% yield.
1H NMR (CDCl3, 400 MHz) d 7.45 (s, 1H), 7.29–7.18 (m, 3H), 7.11
(d, J = 8.4 Hz, 2H), 6.78 (d, J = 8.4 Hz, 2H), 4.23 (t, J = 5.6 Hz, 2H),
3.96 (q, J = 7.2 Hz, 2H), 3.20–2.73 (m, 11H), 2.39 (s, 3H), 1.82–
1.55 (m, 6H), 1.02 (t, J = 7.2 Hz, 3H).
5.7.1. Ethyl 2-((6-methyl-2-oxo-2H-chromen-3-yl)methyl)-6-
phenylhexanoate (14a)
5.9.2. Ethyl 4-(4-(2-(diethylamino)ethoxy)phenyl)-2-((7-methyl-
2-oxo-2H-chromen-3-yl)methyl)butanoate (17i)
1H NMR (CDCl3, 400 MHz) d 7.44 (s, 1H), 7.38–7.15 (m, 8H), 4.03
(m, 2H), 2.88–2.74 (m, 3H), 2.60 (t, J = 7.6 Hz, 2H), 2.38 (s, 3H), 1.75–
1.37 (m, 6H), 1.09 (q, J = 7.2 Hz, 3H); MS (ESI) m/z 415 (MNa+).
1H NMR (CDCl3, 400 MHz) d 7.45 (s, 1H), 7.29 (d, J = 7.6, 1H),
7.11–7.05 (m, 4H), 6.82 (d, J = 8.4 Hz, 2H), 4.10–4.01 (m, 4H),
2.96–2.57 (m, 11H), 2.44 (s, 3H), 2.02–1.81 (m, 2H), 1.15 (t,
J = 7.2 Hz, 3H), 1.07 (t, J = 7.2 Hz, 6H).
5.7.2. Ethyl 2-(4-(benzyloxy)benzyl)-3-(6-methyl-2-oxo-2H-
chromen-3-yl)propanoate (14b)
1H NMR (CDCl3, 400 MHz) d 7.44–7.10 (m, 11H), 6.88 (d,
J = 8.4 Hz, 2H), 5.02 (s, 2H), 3.94 (q, J = 7.2 Hz, 2H), 3.22–2.74 (m,
5H), 2.38 (s, 3H), 1.00 (t, J = 7.2 Hz, 3H); MS (ESI) m/z 457 (MH+).
5.9.3. Ethyl 4-(4-(2-(diethylamino)ethoxy)phenyl)-2-((7-fluoro-
2-oxo-2H-chromen-3-yl)methyl)butanoate (17j)
1H NMR (CDCl3, 400 MHz) d 7.46 (s, 1H), 7.39 (dd, J = 8.6, 6.0 Hz,
1H), 7.09 (d, J = 8.6 Hz, 2H), 7.05–6.98 (m, 2H), 6.82 (d, J = 8.6 Hz,
2H), 4.10–4.02 (m, 4H), 2.92–2.58 (m, 11H), 2.06–1.81 (m, 2H),
1.08 (t, J = 7.2 Hz, 6H), 1.16 (t, J = 6.8 Hz, 3H).
5.7.3. Ethyl 4-(4-(benzyloxy)phenyl)-2-((7-methyl-2-oxo-2H-
chromen-3-yl)methyl)-butanoate (14i)
1H NMR (CDCl3, 400 MHz) d 7.44–7.27 (m, 7H), 7.11–7.05 (m,
4H), 6.89 (d, J = 8.4 Hz, 2H), 5.04 (s, 2H), 4.06 (q, J = 7.2 Hz, 2H),
2.95–2.58 (m, 5H), 2.44 (s, 3H), 2.02–1.80 (m, 2H), 1.14 (t,
J = 7.2 Hz, 3H); MS (ESI) m/z 493 (MNa+).
5.9.4. Ethyl 4-(4-(2-(dimethylamino)ethoxy)phenyl)-2-((6-
methyl-2-oxo-2H-chromen-3-yl)methyl)butanoate (17g)
60% NaH (52.6 mg, 1.31 mmol, 2.5 equiv) was added to a solution
of 16 (200.0 mg, 0.53 mmol) in DMF (20 mL) at 0 °C. After 30 min,
5.8. General procedures for 16
2-chloro-N,N-dimethylethanamine
hydrochloride
(151.7 mg,
1.05 mmol, 2.0 equiv) was added and the reaction mixture was
heated at 70 °C for 12 h. The mixture was poured into ice water
and extracted with ethyl acetate. The organic layer was washed with
brine, dried over anhydrous MgSO4, filtered, and concentrated. The
residue was purified by column chromatography on silica gel (chlo-
roform/methanol = 10:1) to give compound 17g (114.0 mg, 48%). 1H
NMR (CDCl3, 400 MHz) d 7.43 (s, 1H), 7.29–7.19 (m, 3H), 7.09 (d,
J = 8.4 Hz, 2H), 6.84 (d, J = 8.4 Hz, 2H), 4.10–4.03 (m, 4H), 2.96–
2.58 (m, 7H), 2.39 (s, 3H), 2.34 (s, 6H), 2.02–1.17 (m, 2H), 1.15 (t,
J = 8.8 Hz, 3H); MS (ESI) m/z 474 (MNa+).
Pd/C (10 wt %, 40.0 mg) was added to
a solution of 15
(1.0 equiv) in EtOH (0.2 M solution). The mixture was stirred under
H2 atmosphere at room temperature for 5 h. The catalyst was re-
moved by through pad of Celite and washed with EtOH. The sol-
vent was evaporated in vacuo to give compound 16 in 80–90%
yield.
5.8.1. Ethyl 2-(4-hydroxybenzyl)-3-(6-methyl-2-oxo-2H-
chromen-3-yl)propanoate (16a)
1H NMR (CDCl3, 400 MHz) d 7.47 (s, 1H), 7.29–7.05 (m, 3H), 7.06
(d, J = 8.4 Hz, 2H), 6.75 (d, J = 8.4 Hz, 2H), 5.65 (br, 1H), 3.96 (q,
J = 7.2 Hz, 2H), 3.20–2.73 (m, 5H), 2.38 (s, 3H), 1.01 (t, J = 7.3 Hz,
3H).
5.10. General procedures for 15
NH2OH in MeOH (1.76 M solution, 20 equiv) was added to a
solution of 16 (1.0 equiv) in MeOH (0.25 M solution). The mixture
was stirred at room temperature for 4 h and concentrated. The res-
idue was purified by column chromatography on silica gel (hex-
ane/ethyl acetate = 2:1) to give compound 15 in 10–30% yield.
5.8.2. Ethyl 4-(4-hydroxyphenyl)-2-((7-methyl-2-oxo-2H-
chromen-3-yl)methyl)butanoate (16g)
1H NMR (CDCl3, 400 MHz) d 7.45 (s, 1H), 7.27 (d, J = 7.6 Hz, 1H),
7.12 (s, 1H), 7.08–7.04 (m, 3H), 6.75 (d, J = 8.4 Hz, 2H), 4.81 (br,
1H), 4.07 (q, J = 7.2 Hz, 2H), 2.94–2.44 (m, 5H), 2.39 (s, 3H), 2.02–
1.80 (m, 2H), 1.15 (t, J = 7.2 Hz, 3H); MS (ESI) m/z 403 (MNa+).
5.10.1. 2-((6-Methyl-2-oxo-2H-chromen-3-yl)methyl)-6-phenyl-
hexanoic acid hydroxyamide (15a)
1H NMR (DMSO-d6, 400 MHz) d 10.39 (s, 1H), 8.67 (s, 1H), 7.63
(s, 1H), 7.41 (s, 1H), 7.37–7.12 (m, 7H), 2.60–2.34 (m, 5H), 2.34 (s,
3H), 1.56–1.22 (m, 6H); MS (FAB) m/z 380 (MH+), HRMS (FAB) calcd
for C23H25NO4(MH+) 380.1862, found 380.1862.
5.8.3. Ethyl 2-((7-fluoro-2-oxo-2H-chromen-3-yl)methyl)-4-(4-
hydroxyphenyl)butanoate (16h)
1H NMR (CDCl3, 400 MHz) d 7.46 (s, 1H), 7.39 (dd, J = 8.4, 5.9 Hz,
1H), 7.05–6.97 (m, 4H), 6.76 (d, J = 8.4 Hz, 2H), 4.96 (br, 1H), 4.07

8628
J. S. Yang et al. / Bioorg. Med. Chem. 18 (2010) 8618–8629
5.10.2. N-Hydroxy-4-(4-hydroxyphenyl)-2-((6-methyl-2-oxo-
2H-chromen-3-yl)methyl)butan-amide (15b)
5.10.9. 4-(4-(2-(Dimethylamino)ethoxy)phenyl)-2-((6-methyl-2-
oxo-2H-chromen-3-yl)-methyl)butanoic acid hydroxyamide
(15i)
1H NMR (DMSO-d6, 400 MHz) d 10.45 (s, 1H), 8.74 (s, 1H), 7.64
(s, 1H), 7.41 (s, 1H), 7.36 (d, J = 8.4 Hz, 1H), 7.28 (d, J = 8.0 Hz, 1H),
6.94 (d, J = 7.7 Hz, 2H), 6.56 (d, J = 8.0 Hz, 2H), 2.59–2.61 (m, 2H),
2.37–2.41 (m, 3H), 2.35 (s, 3H), 1.74–1.78 (m, 1H), 1.57–1.61 (m,
1H); MS (FAB) m/z 368 (MH+), HRMS (FAB) calcd for
1H NMR (DMSO-d6, 400 MHz) d 10.43 (s, 1H), 8.71 (br, 1H), 7.61
(s, 1H), 7.38–7.23 (m, 3H), 7.03 (d, J = 8.4 Hz, 2H), 6.79 (d, J = 8.4 Hz,
2H), 3.97 (t, J = 6.0 Hz, 2H), 2.56–2.39 (m, 7H), 2.31 (s, 3H), 2.17
(s, 6H), 1.73–1.52 (m, 2H); MS (FAB) m/z 439 (MH+), HRMS (FAB)
calcd for C25H30N2O5(MH+) 439.2233, found 439.2236.
C
21H21NO5(MH+) 368.1498, found 368.1500.
5.10.10. 4-(4-(2-(Diethylamino)ethoxy)phenyl)-2-((6-methyl-2-
oxo-2H-chromen-3-yl)-methyl)-butanoic acid hydroxyamide
(15j)
5.10.3. 3-(6-Methyl-2-oxo-2H-chromen-3-yl)-2-(4-(2-(piperi-
din-1-yl)ethoxy)benzyl)propanoic acid hydroxyamide (15c)
1H NMR (DMSO-d6, 400 MHz) d 10.37 (s, 1H), 8.46 (s, 1H),
7.65 (s, 1H), 7.41 (s, 1H), 7.37–7.35 (m, 2H), 7.08 (d, J = 8.4 Hz,
2H), 6.81 (d, J = 8.4 Hz, 2H), 3.99 (d, J = 5.6 Hz, 2H), 2.82–2.49
(m, 11H), 2.35 (s, 3H), 1.53–1.36 (m, 6H); MS (FAB) m/z 465
(MH+), HRMS (FAB) calcd for C27H32N2O5(MH+) 465.2389, found
465.2390.
1H NMR (DMSO-d6, 400 MHz) d 10.46 (s, 1H), 8.74 (s, 1H), 7.65
(s, 1H), 7.42 (s, 1H), 7.38 (d, J = 8.4 Hz, 1H), 7.28 (d, J = 8.4 Hz, 1H),
7.06 (d, J = 8.4 Hz, 2H), 6.82 (d, J = 8.4 Hz, 2H), 3.97 (t, J = 6.0 Hz,
2H), 2.77–2.42 (m, 11H), 2.35 (s, 3H), 1.80–1.60 (m, 2H), 0.97 (t,
J = 7.2 Hz, 6H); MS (FAB) m/z 467 (MH+), HRMS (FAB) calcd for
C
27H34N2O5(MH+) 467.2546, found 467.2547.
5.10.4. 3-(6-Methyl-2-oxo-2H-chromen-3-yl)-2-(3-(2-(piperi-
din-1-yl)ethoxy)benzyl)propanoic acid hydroxyamide (15d)
1H NMR (DMSO-d6, 400 MHz) d 10.37 (s, 1H), 8.65 (s, 1H), 7.66
(s, 1H), 7.42–7.13 (m, 4H), 6.80–6.70 (m, 3H), 4.04 (t, J = 5.6 Hz,
2H), 3.58–1.23 (m, 17H), 2.35 (s, 3H); MS (FAB) m/z 481 (MH+).
5.10.11. 4-(4-(2-(Diethylamino)ethoxy)phenyl)-2-((7-fluoro-2-
oxo-2H-chromen-3-yl)methyl)-butanoic acid hydroxyamide
hydrochloride (15k)
1H NMR (DMSO-d6, 400 MHz) d 10.49 (s, 1H), 9.80 (br, 1H), 7.69
(s, 1H), 7.52 (d, J = 8.0 Hz, 1H), 7.21 (s, 1H), 7.17–7.12 (m, 3H), 6.90
(t, J = 5.2 Hz, 2H), 4.29 (t, J = 5.2 Hz, 2H), 3.60 (br, 1H), 3.50–3.21
(m, 6H), 2.59–2.44 (m, 5H), 2.40 (s, 3H), 1.80–1.59 (m, 2H), 1.23
(t, J = 7.2 Hz, 6H); MS (FAB) m/z 471 (MH+), HRMS (FAB) calcd for
5.10.5. 2-((6-Methyl-2-oxo-2H-chromen-3-yl)methyl)-4-(4-(2-
(piperidin-1-yl)ethoxy)phenyl)-butanoic acid hydroxyamide
(15e)
C
26H31FN2O5(MH+) 471.2295, found 471.2294.
1H NMR (DMSO-d6, 400 MHz) d 10.46 (s, 1H), 8.72 (br, 1H), 7.64
(s, 1H), 7.41 (s, 1H), 7.37 (d, J = 8.4 Hz, 1H), 7.27 (d, J = 8.4 Hz, 1H),
7.06 (d, J = 8.8 Hz, 2H), 6.82 (d, J = 8.8 Hz, 2H), 4.00 (t, J = 6.0 Hz,
2H), 2.64–2.42 (m, 11H), 2.35 (s, 3H), 1.85–1.55 (m, 2H), 1.51–
1.32 (m, 6H); MS (FAB) m/z 479 (MH+), HRMS (FAB) calcd for
5.10.12. 4-(4-(2-(Diethylamino)ethoxy)phenyl)-2-((7-methyl-2-
oxo-2H-chromen-3-yl)methyl)-butanoic acid hydroxyamide
hydrochloride (15l)
1H NMR (DMSO-d6, 400 MHz) d 10.48 (s, 1H), 10.12 (s, 1H), 7.67
(s, 1H), 7.50 (d, J = 8.0 Hz, 1H), 7.20 (s, 1H), 7.14 (d, J = 8.0 Hz, 1H),
7.10 (d, J = 8.4 Hz, 2H), 6.88 (d, J = 8.4 Hz, 2H), 4.4 (br, 1H), 4.29–
2.38 (m, 13H), 1.80–1.57 (m, 2H), 1.22 (t, J = 7.2 Hz, 6H); MS
(FAB) m/z 467 (MH+), HRMS calcd for C27H34N2O5(MH+)
467.2546, found 467.2549.
C
28H34N2O5(MH+) 479.2546, found 479.2546.
5.10.6. 2-((6-Methyl-2-oxo-2H-chromen-3-yl)methyl)-4-(3-(2-
(piperidin-1-yl)ethoxy)phenyl)-butanoic acid hydroxyamide
(15f)
1H NMR (DMSO-d6, 400 MHz) d 10.49 (br, 1H), 8.32 (s, 1H), 7.66
(s, 1H), 7.47–6.90 (m, 7H), 4.02 (t, J = 4.9 Hz, 2H), 2.66–2.40 (m,
11H), 2.35 (s, 3H), 1.83–1.64 (m, 2H), 1.52–1.36 (m, 6H); MS
(FAB) m/z 480 (MH+), HRMS (FAB) calcd for C28H34N2O5(MH+)
479.2546, found 479.2546.
5.11. Biology
5.11.1. TACE, MMP-3, and MMP-9 enzyme inhibition assay
The enzymatic activity of human TACE (R&D Systems,
Minneapolis, MN, USA) was determined at 25 °C with a synthetic
fluorogenic substrate, Mca-Pro-Leu-Ala-Gln-Ala-Val-Dpa-Arg-Ser-
Ser-Ser-Agr-NH2 (Mca: [7-methoxycoumarin-4-yl] acetyl; Dpa:
5.10.7. 2-((6-Methyl-2-oxo-2H-chromen-3-yl)methyl)-5-(4-(2-
(piperidin-1-yl)ethoxy)phenyl)-pentanoic acid hydroxyamide
(15g)
N-3-[2,4-dinitrophenyl]-L-2,3-diaminopropionyl, R&D Systems,
Minneapolis, MN, USA). The enzymatic activities of MMP-3 and
MMP-9were determinedwithrecombinanthumanversioncatalytic
domains (Roche Applied Science, D-68298 Manheim, Germany) and
a fluorogenic peptide substrate, Mca-Pro-Leu-Gly-Leu-Dpa-Ala-
Arg-NH2 (Sigma Chemical Co., PO Box 14508, St. Louis, MO 63178,
USA). Fluorescence measurements were performed in a Perkin–El-
mer Luminescence Spectrometer LS50B. Cleavage of internal
quenched substrate liberates emission-active Mca product, causing
an increase in fluorescence emission at 420 nM (the excitation
wavelength is 324 nM). The rate of emission change is proportional
to enzyme activity.
1H NMR (DMSO-d6, 400 MHz) d 10.43 (s, 1H), 8.69 (s, 1H), 7.65
(s, 1H), 7.43 (s, 1H), 7.38 (d, J = 8.4 Hz, 1H), 7.28 (d, J = 8.4 Hz, 1H),
7.10 (d, J = 8.0 Hz, 2H), 6.86 (d, J = 8.0 Hz, 2H), 4.23–4.15 (m, 2H),
3.42–2.40 (m, 11H), 2.35 (s, 3H), 1.84–1.36 (m, 10H); MS (FAB)
m/z 493 (MH+), HRMS (FAB) calcd for C29H36N2O5(MH+)
493.2702, found 493.2702.
5.10.8. 2-((6-Methyl-2-oxo-2H-chromen-3-yl)methyl)-5-(3-(2-
(piperidin-1-yl)ethoxy)phenyl)-pentanoic acid hydroxyamide
(15h)
1H NMR (CDCl3, 400 MHz) d 7.45 (s, 1H), 7.30–7.11 (m, 4H), 6.72
(d, J = 8.0 Hz, 1H), 6.68–6.65 (m, 2H), 5.40 (br, 1H), 4.09–4.01 (m,
2H), 2.94–2.71 (m, 3H), 2.59 (t, J = 7.2 Hz, 2H), 2.39 (s, 3H),
1.75–1.55 (m, 4H), 1.12 (t, 7.2 Hz, 3H); MS (FAB) m/z 493 (MH+),
5.11.2. Cellular NO assay and TNF-a immunoassay
ꢀ
RAW264.7 cells were cultured with inhibitors for 24 h. NO2
accumulation was used as an indicator of NO production in the
medium. The isolated supernatants were mixed with an equal
volume of Griess reagent (1% sulfanilamide, 0.1% naphthylethylen-
HRMS (FAB) calcd for
493.2702.
C
29H36N2O5(MH+) 493.2702, found

J. S. Yang et al. / Bioorg. Med. Chem. 18 (2010) 8618–8629
8629
ediamine dihydrochloride, and 2% phosphoric acid) and incubated
at room temperature for 10 min. Nitrite production was deter-
mined by measuring absorbance at 540 nm versus a NaNO2
ti-tube vortexer. The 96-well cluster tube plate was then centrifuged
at 12,000 rpm for 10 min and the supernatant was analyzed for
quantitation of 15l. Sample analysis was performed by API2000
LC/MS/MS system in a positive MRM mode (column: Waters Xterra
standard curve. The concentration of TNF-a secreted in the culture
supernatant of RAW264.7 cells was determined by ELISA, accord-
ing to the manufacture’s instruction (R&D Systems, Minneapolis,
MN, USA).
MSC18 (2.1 ꢁ 50 mm, 3.5
lm); mobilephase: acetonitrile/DI water/
0.1% formic acid; ion source: turbo ion spray). Analytical data were
processed using Analyst™ 1.4.1. (Applied Biosystems, Concord,
Canada). Pharmacokinetic parameters were obtained by non-
compartmental analysis of the plasma concentration–time profiles
using PK Solution 2.0 (Summit Research Services, Montrose, CO,
USA).
5.12. Pharmacology and pharmacokinetics
5.12.1. In vivo TNF-
To test the inhibitory activity of TNF-
TNF- in blood of BALB/c mice (7–9 weeks, Japan SLC, Inc.) was
measured. 10 g/mouse of LPS (Sigma, L-2880) was intraperitone-
a
reproduction assay
a
reproduction, the level of
a
Acknowledgments
l
ally administrated into the mice to induce inflammation and mice
had been starved for 24 h before use. Appropriate concentration of
the compound (0.2 mg/mouse) was orally administrated thereto
and lipopolysaccharide was further treated after 30 min. After
90 min, the serum was collected from the heart, and level of
This research was supported by National Research Foundation
(2009-0092966), Korean Research WCU grant (R31-2008-000-
10086-0) and the Brain Korea 21 project, the Republic of Korea.
Supplementary data
TNF-a in the isolated serum was determined by ELISA according
to the manufacture’s instruction (Biosource International, USA)
Supplementary data associated with this article can be found, in
the online version, at doi:10.1016/j.bmc.2010.10.006. These data
include MOL files and InChiKeys of the most important compounds
described in this article.
5.12.2. Carrageenan-induced paw edema test in rat
Sprague–Dawley rats (150 g, 5 weeks, SPF) were divided into
four groups consisting of five rats each group: vehicle, 10 mg/kg
of indomethacin, 30 mg/kg of 15l, and 100 mg/kg of 15l. Indometh-
acin was diluted with mixture solution of DMSO, Tween 80, and
H2O (1:4:95) to use as positive control group and 15l was diluted
with mixture solution of ethanol, PEG (polyethylene glycol) 400,
References and notes
1. Tracey, K. J.; Cerami, A. Annu. Rev. Med. 1994, 45, 491.
2. Goldfeld, A. E.; Strominger, J. L.; Doyle, C. J. Exp. Med. 1991, 174, 73.
3. Le, G. T.; Abbenante, G. Curr. Med. Chem. 2005, 12, 2963.
and H2O (5:20:75) to use as a test group. After 30 min, 100 lL of
2% carrageenan (Sigma, c-1013) solution was injected into inner
rat to induce edema. The thickness of ankle was measured by ber-
nier’s caliper to determine the degree of the edema after 6 and
24 h. The thickness ratio of ankle (%) = [(maximum thickness of
rat paw edema) ꢀ (the thickness of pre-treated rat paw edema)]/
(the thickness of pre-treated rat paw edema) ꢁ 100%.
4. Newton, R. C.; Decicco, C. P. J. Med. Chem. 1999, 42, 2295.
5. Hotamisligil, G. S.; Arner, P.; Caro, J. F.; Atkinson, R. L.; Spiegelman, B. M. J. Clin.
Invest. 1995, 95, 2409.
6. Selmaj, K.; Raine, C. S.; Cannella, B.; Brosnan, C. F. J. Clin. Invest. 1991, 87, 949.
7. Rus, H. G.; Niculescu, F.; Vlaicu, R. Atherosclerosis 1991, 89, 247.
8. Lovering, F.; Zhang, Y. Curr. Drug Targets CNS Neurol. Disord. 2005, 4, 161.
9. Moreland, L. W.; Baumgartner, S. W.; Schiff, M. H.; Tindall, E. A.; Fleischmann,
R. M.; Weaver, A. L.; Ettlinger, R. E.; Cohen, S.; Koopman, W. J.; Mohler, K.;
Widmer, M. B.; Blosch, C. M. N. Eng. J. Med. 1997, 337, 141.
10. Lipsky, P. E.; van der Heijde, D. M.; St Clair, E. W.; Furst, D. E.; Breedveld, F. C.;
Kalden, J. R.; Smolen, J. S.; Weisman, M.; Emery, P.; Feldmann, M.; Harriman, G.
R.; Maini, R. N. N. Eng. J. Med. 2000, 343, 1594.
11. Machold, K. P.; Smolen, J. S. Expert Opin. Biol. Ther. 2003, 3, 351.
12. Moss, M. L.; Sklair-Tavron, L.; Nudelman, R. Nat. Clin. Pract. Rheumatol. 2008, 4,
300.
13. Nelson, F. C.; Zask, A. Expert Opin. Invest. Drugs 1999, 8, 383.
14. Wasserman, Z. R.; Duan, J. J.; Voss, M. E.; Xue, C. B.; Cherney, R. J.; Nelson, D. J.;
Hardman, K. D.; Decicco, C. P. Chem. Biol. 2003, 10, 215.
15. Shaw, T.; Nixon, J. S.; Bottomley, K. M. Expert Opin. Invest. Drugs 2000, 9, 1469.
16. Aranapakam, V.; Davis, J. M.; Grosu, G. T.; Baker, J.; Ellingboe, J.; Zask, A.; Levin,
J. I.; Sandanayaka, V. P.; Du, M.; Skotnicki, J. S.; DiJoseph, J. F.; Sung, A.; Sharr,
M. A.; Killar, L. M.; Walter, T.; Jin, G.; Cowling, R.; Tillett, J.; Zhao, W.; McDevitt,
J.; Xu, Z. B. J. Med. Chem. 2003, 46, 2376.
17. Park, S. K.; Han, S. B.; Lee, K.; Lee, H. J.; Kho, Y. H.; Chun, H.; Choi, Y.; Yang, J. Y.;
Yoon, Y. D.; Lee, C. W.; Kim, H. M.; Choi, H. M.; Tae, H. S.; Lee, H. Y.; Nam, K. Y.;
Han, G. Biochem. Biophys. Res. Commun. 2006, 341, 627.
18. Chun, K.; Park, S. K.; Kim, H. M.; Choi, Y.; Kim, M. H.; Park, C. H.; Joe, B. Y.; Chun,
T. G.; Choi, H. M.; Lee, H. Y.; Hong, S. H.; Kim, M. S.; Nam, K. Y.; Han, G. Bioorg.
Med. Chem. 2008, 16, 530.
19. Hin, B.; Majer, P.; Tsukamoto, T. J. Org. Chem. 2002, 67, 7365.
20. Duan, J. J.; Chen, L.; Wasserman, Z. R.; Lu, Z.; Liu, R. Q.; Covington, M. B.; Qian,
M.; Hardman, K. D.; Magolda, R. L.; Newton, R. C.; Christ, D. D.; Wexler, R. R.;
Decicco, C. P. J. Med. Chem. 2002, 45, 4954.
5.12.3. Human liver microsomal stability test
Compound 15l or DPC-333 (1
some (BD gentest, 20 L, final concn 1 mg/mL) and MgCl2 (40
was added to potassium phosphate buffer. The solution was incu-
bated for 5 min at 37 °C and added b-NADPH (Sigma, 100 L, final
concn 1 mM). Aliquots of 100 L are removed from the incubations
at varying time intervals; 0, 15, 45, and 80 min. The reactions are
terminated by the addition of ice cold acetonitrile 200 L and
internal standard 100 (carbamazepine, 1000 ng/mL). The
lM, 40
lL), Human liver micro-
l
l
L)
l
l
l
lL
samples were centrifuged at 3000 rpm for 10 min and analysis by
LC/MS/MS.
5.12.4. Rat pharmacokinetic (PK) study
Male SD rats were acclimated to the testing facility in a temper-
ature and humidity controlled condition for approximately a week
prior to thestudy. Male SD rats (10-weekold, 285–305 g) were cann-
ulated in the right jugular vein. 15l dissolved saline were dosed iv or
po in SD rats. About 300 lL of blood samples were collected in BD
microtainer plasma separator tubes at selected times through the
jugular vein cannula over 24 h post-dosing. Blood samples were
centrifuged at 3000 rpm for 10 min and stored in a freezer until
analyzed. Plasma samples were prepared for analysis by the follow-
ing protein precipitation method. One hundred microliters of the
plasma samples were transferred to a 96-well cluster tube plate.
Three volumes of acetonitrile containing carbamazepine (IS) were
added and the resulting mixture was vortexed for 5 min on a mul-
21. Venkatachalam, C. M.; Jiang, X.; Oldfield, T.; Waldman, M. J. Mol. Graphics
Modell. 2003, 21, 289.
22. Mazzola, R. D., Jr.; Zhu, Z.; Sinning, L.; McKittrick, B.; Lavey, B.; Spitler, J.;
Kozlowski, J.; Neng-Yang, S.; Zhou, G.; Guo, Z.; Orth, P.; Madison, V.; Sun, J.;
Lundell, D.; Niu, X. Bioorg. Med. Chem. Lett. 2008, 18, 5809.
23. Natchus, M. G.; Bookland, R. G.; De, B.; Almstead, N. G.; Pikul, S.; Janusz, M. J.;
Heitmeyer, S. A.; Hookfin, E. B.; Hsieh, L. C.; Dowty, M. E.; Dietsch, C. R.; Patel,
V. S.; Garver, S. M.; Gu, F.; Pokross, M. E.; Mieling, G. E.; Baker, T. R.; Foltz, D. J.;
Peng, S. X.; Bornes, D. M.; Strojnowski, M. J.; Taiwo, Y. O. J. Med. Chem. 2000, 43,
4948.