Synthesis of novel β-amino ketones containing a p-aminobenzoic acid moiety and evaluation of their antidiabetic activities

Article abstract of DOI:10.1007/s11426-012-4816-2

The synthesis of two series of β-amino ketones containing a p-aminobenzoic acid moiety (TM-1 and TM-2) using a modified protocol of the Mannich reaction is reported. The molecular structures of a total of tweenty three new target compounds were characterized by ¹H NMR, ¹³C NMR, ESI-MS and HR-MS. Subsequently, their antidiabetic activities were screened in vitro. The α-glucodase inhibition (α-GI) activity of compound 1e reached a remarkable level of 66.50%. The peroxisome proliferator-activated receptor (PPAR) relative activation activities of six compounds are above 80%, and in particular 2i displays an unprecedentedly high PPAR of 130.91%. The structure-activity relationships of the compounds were established. 2i is also subject to further in-depth investigation.

Full text of DOI:10.1007/s11426-012-4816-2

SCIENCE CHINA
Chemistry
April 2013 Vol.56 No.4: 490–504
doi: 10.1007/s11426-012-4816-2
• ARTICLES •
Synthesis of novel β-amino ketones containing a p-aminobenzoic
acid moiety and evaluation of their antidiabetic activities
TANG GuangXia1, YAN JuFang2, FAN Li1, XU Jin1, SONG XiaoLi1, JIANG Li3*,
LUO LingFei4 & YANG DaCheng1*
1School of Chemistry and Chemical Engineering, Southwest University, Chongqing 400715, China
2Drug Screening Center, Chengdu Di Ao Pharmaceutical Group Co., Ltd., Chengdu 610041, China
3Schlumberger Technology Corporation, 110 Schlumberger Drive, Sugar Land, TX 77478, USA
4School of Life Science, Southwest University, Chongqing 400715, China
Received June 25, 2012; accepted November 2, 2012; published online January 13, 2013
The synthesis of two series of β-amino ketones containing a p-aminobenzoic acid moiety (TM-1 and TM-2) using a modified
protocol of the Mannich reaction is reported. The molecular structures of a total of tweenty three new target compounds were
1
characterized by H NMR, 13C NMR, ESI-MS and HR-MS. Subsequently, their antidiabetic activities were screened in vitro.
The -glucodase inhibition (-GI) activity of compound 1e reached a remarkable level of 66.50%. The peroxisome proliferator-
activated receptor (PPAR) relative activation activities of six compounds are above 80%, and in particular 2i displays an un-
precedentedly high PPAR of 130.91%. The structure-activity relationships of the compounds were established. 2i is also sub-
ject to further in-depth investigation.
diabetes mellitus, -glucosidase, peroxisome proliferator-activated receptor, β-amino ketone, Mannich reaction
(DPP IV) inhibitors, for example alogliptin (approved in
1 Introduction
Japan in 2010) [2] and linagliptin (approved by the US FDA
in 2011) [3]. Although somewhat useful in the treatment of
T2DM, the long-term effects of these antidiabetic agents
have not been fully evaluated and may lead to a variety of
adverse consequences. For instance, insulin may cause
weight gain and hypoglycaemia, and TZD insulin sensitiz-
ers may lead to multiple adverse effects including weight
gain, edema, hepatotoxicity and elevated level of low den-
sity lipoprotein cholesterol (LDL-C) [4]. Thus the demand
remains urgent for the development of new antidiabetic
drugs with higher efficacy and lower toxicity for long term
treatment.
Many of the commercially available T2DM drugs, de-
spite being structurally diverse, share a few key features that
are deemed critical to their functionalities. For example,
3-phenylpropanoic acid based nateglinide [5] (Novartis,
1999), an oral formulation of insulin secretion promoter
Diabetes mellitus (DM) is a common metabolic disorder
ever growing into epidemic proportions following the im-
provement in quality of life for an increasingly greater pop-
ulation in recent years [1], and the condition is now ranked
the 3rd most fatal non-infectious chronic disease after cardi-
ovascular disease and cancer. For the most widespread type
2 diabetes (T2DM), clinical drugs may be classified into
insulin secretagogues (glinides and sulfonylureas), insulin
sensitizers (biguanides, thiazolidinediones (TZDs)), aldose
reductase inhibitors (e.g. epalrestat and tolrestat), glucose
absorption effectors (e.g. a-glucosidase inhibitors such as
acarbose, miglitol and voglibose). Also falling in the same
category are the newly introduced dipeptidyl peptidase IV
*Corresponding authors (email: ljiang2008@live.com; hxydc@swu.edu.cn)
© Science China Press and Springer-Verlag Berlin Heidelberg 2013
chem.scichina.com www.springerlink.com

Tang GX, et al. Sci China Chem April (2013) Vol.56 No.4
491
effective in reducing blood glucose, and benzoic acid based
repaglinide [6] (Novo Nordisk, 1998) and L-tyrosine deriv-
atives [7] both contain carboxyl and amino functional
groups. In addition, drugs like rosiglitazone (GSK, 1999),
pioglitazone (Takeda, 1999) and epalrestat (Pharmaceutical,
1992), plus drug candidates muraglitazar (BMS 298585),
DRF2189 [8], MCC555 [9] and GW1929, PTP1B inhibitors
B [10] and C [11] (Chart 1), contain some sort of acidic
part(s) or other functional group(s) with or without amino
groups. Additionally, the majority of these molecules con-
sist of two aromatic ring modules separated by a linker.
These structural factors have to be taken into consideration
in the development of new T2DM drugs.
ganic synthesis [12, 13, 14], is commonly used for the
preparation of -amino ketones, including those potent for
analgesic [15], tubercle [16, 17], inflammation [18], cancer
[19] and bacteria [20]. Our group has recently identified a
large group of -amino ketones exhibiting androgen recep-
tor modulating activity [21] and antidiabetic activities [22],
It is apparent that many potent -amino ketones contain a
p-aminobenzoic acid moiety. These compounds often have
the modular composition as outlined in Figure 1, including a
hydrophilic portion A and a hydrophobic portion C con-
nected by a linker portion B, and closely resemble the
structural characteristics of TZDs and some L-tyrosine de-
rivatives [7].
The search for effective antidiabetic target molecules
needs to start from examining the structural features of ex-
isting drugs, followed by designing a novel class of specific
compounds and adopting facile synthetic methods. An im-
portant way is to obtain the desired bioactive molecules
from domino or multicomponent reactions (MCRs). The
Mannich reaction, one of the most important MCRs in or-
We are interested in the development of non-thiazoli-
dinedione peroxisome proliferator-activated receptor (PPARγ)
agonists in an attempt to surmount the pitfalls associated
with the known TZDs [7]. Based on our previous research
results, we decided to modify discrete parts of -amino ke-
tone lead compounds (Figure 1). As a well accepted scaffold
in medicinal chemistry, p-aminobenzoic acid has been in-
Chart 1 Representative candidates in clinical trials or active compounds reported.
Figure 1 Engineering strategies for target molecules.

492
Tang GX, et al. Sci China Chem April (2013) Vol.56 No.4
corporated in compounds displaying a variety of pharmaco-
logical effects, including anti-tumor [23], anticancer [24],
anti-hepatitis B virus [25], anti-HIV integrase [26], antidia-
betic [22, 27] and antibacterial activities [28, 29]. The
p-aminobenzoic acid moiety is conserved as an essential
pharmacophore which, similar to glinide, also acts as the
indispensable amino component in the Mannich reaction.
Alternatively, ethyl p-aminobenzoate, commonly present in
toponarcosis, analgesic and antipruritic drugs, may be used
as the hydrophilic portion A since it can be converted into
p-aminobenzoic acid via in vivo metabolism.
It was reported [30] that the presence of a naphthalene
ring led to more pronounced PPAR activities than analogues
with a simple benzene ring. Presumably, the introduction of
hydrocarbon chain between the carbonyl group and the ar-
omatic ring in an aryl alkyl ketone alters the flexibility of
the backbone which, in turn, changes its solubility and bio-
activity. As such, inserting a nabumetone building block
into -amino ketones in place of portion C and part of por-
tion B is desirable by virtue of its two methylene groups
between the carbonyl group and the naphthalene ring and
the two reactive sites for the Mannich reaction, which
should lead to more backbone flexibility and, in turn, en-
hanced bioactivity. Currently, nabumetone is present in
common non-steroidal anti-inflammatory drugs (NSAID)
for rheumatic and inflammatory conditions [31]. The two
reactive sites allow ready construction but may result in two
kinds of products, which result in difficulties in obtaining
high purity target molecules. Finally, an aromatic aldehyde
with different substituent groups was selected as part of
linker B. Based on our previous work and fresh considera-
tions as highlighted above, we designed two series of
-amino ketones containing target molecules TM-1 and
TM-2 as outlined in Figure 1. By synthesizing the two se-
ries of -amino ketones through the Mannich reaction with
the modified “one-pot, two-step and three-component” pro-
tocol and screening their PPAR activation and -gluco-
sidase inhibition, we were able to identify new lead mole-
cules for antidiabetic drug and establish the all important
structure-function relationship.
Figure 2 Current synthetic methodologies for Mannich bases. (a)
Three-component one-pot reaction; (b) condensation of Schiff base with
ketone; (c) exchange of ketone; (d) Michael addition; (e) exchange of
amine.
and CN bond forming reactions in organic synthesis. Prior
to 1980, the conventional starting materials were formalde-
hyde, fatty amines and ketones, and to a lesser extent aro-
matic aldehydes and fatty amines [39]. It was not until 1991
when aromatic aldehydes, arylamines and ketones were first
used as the starting materials [40]. Subsequently, the use of
aromatic aldehydes and arylamines as starting materials was
further explored [41, 42, 43]. We adopted a “one-pot,
two-step and three-component” protocol to yield -amino
ketones (Scheme 1). Many target molecules showing vari-
ous levels of potent antidiabetic activities have since been
obtained [22, 44, 45, 46].
A
model reaction system involving nabumetone,
p-chlorobenzaldehyde and p-aminobenzoic acid as starting
materials was tested at different temperatures, namely 25,
30 and 40 °C; in different solvents including ethanol, meth-
anol, benzene, chloroform or an ethanol-chloroform binary
mixture; using different catalysts including concentrated
hydrochloric acid, glacial acetic acid, nitric acid, phosphoric
acid, iron(III) chloride, aluminium trichloride, or a solution
of hydrogen chloride in ethyl acetate. The optimal reaction
conditions were found to be an ethanol-chloroform mix-
ture solution (V:V = 3:1), catalyzed by concentrated hydro-
chloric acid (0.10.2 mL) in the temperature range of
2530 °C. In general, these parameters are not too dissimi-
lar to previous findings for other β-amino ketone systems
[22, 45].
As shown in Table 1, there appears to be a general trend
for both TM-1 and TM-2 in that the yield of the final product
-amino ketone is closely dependent on the substituent
group(s) on the benzene ring of the aromatic aldehyde. More
specifically, aromatic aldehydes with electron-donating
groups react more readily than those with electron-with-
drawing groups, following the order 2l(4-O-Bun) >2j(3-
2 Results and discussion
2.1 Chemistry
The Mannich reaction is a classical tool for the synthesis of
-amino ketones, which may be achieved via a number of
protocols. These include either a direct approach involving
all-three starting materials aldehydes, ketones and amines
[32, 33], exchange of amines [34], Michael addition of ar-
omatic amines with , -unsaturated ketones [35], forming
amine hydrochlorides [36], exchange of ketones [37], or
condensation of Schiff bases with aromatic ketones [38]
(Figure 2).
The Mannich reaction is one of the most important CC

Tang GX, et al. Sci China Chem April (2013) Vol.56 No.4
493
also generally valid that aromatic aldehydes with para-
substituents react faster than those with ortho- or me-
ta-substituents. For instance, 1k(4-CH3O) > 1j(3-CH3O),
1e(4-Cl) > 1q(2-Cl) > 1d(3-Cl), 1b(3,4-diCl) > 1o(2,4-
diCl). All of these are correlated to the nature, and location,
of the substitutent.
It is worth noting that aromatic aldehydes substituted by
either 2,4-diCl, 3-OH or 2-Cl groups reacted readily with
p-aminobenzoic acid to yield 1o, 1p and 1q, respectively,
but the expected analogous reaction with ethyl
p-aminobenzoate did not proceed. The root cause remains
unclear and is under continuing investigation. Fortunately,
aromatic aldehydes substituted by a 4-OH group produced
the corresponding -amino ketones in moderate yields.
Again, to our surprise, aromatic aldehydes substituted by
4-NO2, 4-N(CH3)2, or furfuraldehyde, did not produce any
of the corresponding -amino ketones under the established
conditions.
2.2 Biological activity
Positive controls for the assay of -glucosidase inhibition
and PPAR activation are Acarbose (100 µg mL−1, 74%) and
Rosiglitazone(10 µg mL−1, 100%), respectively. The bioas-
say results of all target compounds are shown in Table 1.
2.2.1 α-Glucosidase inhibition
Scheme 1 Synthesis of target -amino ketones.
-Glucosidase plays an important role in the digestion of
carbohydrates and biosynthesis of viral envelope glycopro-
teins. -Glucosidase inhibitors are promising candidates for
antidiabetics [47], anti-AIDS [48], anti-viral hepatitis [49]
and anticancer drugs [50]. In order to evaluate the antidia-
betic activity of the target -amino ketones, we adopted the
assay of -GI activity. From Table 1 and Figure 3, it can be
found that the -GI activities were weak for the majority of
the -amino ketones containing p-aminobenzoic acid (TM-
1) or ethyl p-aminobenzoate moiety (TM-2) at a concentra-
tion of 10.0 µg mL−1. A few compounds previously reported
by our group are also listed in Table 1 for the purpose of
investigating the structure-activity relationship (SAR) [22,
44]. As shown in Figure 3, the two series of compounds
exhibit fairly consistent performances in their biological
activities, except for the case of 3-CH3. The products with
the strong electron-withdrawing group 3-NO2 and the strong
electron-donating groups 3-OCH3, 4-CH3O or 3, 4-OCH2O
show very weak -GI activities. However, those with the
weak electron-withdrawing groups 4-Cl or 3,4-diCl are sig-
nificantly better. For instance, the -GI activity of 1e (4-Cl)
reached 66.50% at 10 µg mL−1, very close to the positive
control of acrobose at 10-fold higher concentration (74.12%,
100 µg mL−1). In addition, the -GI activity of 1b (3,4-diCl)
reached 48.50%. Furthermore, 1n and 2b gave values of
41.10% and 40.72%, respectively. These represent approx-
imately 55% of the positive control activity of acrobose at a
10-fold higher concentration level.
CH3O) > 2i(4-CH3) > 2c(3-F) > 2e(4-Cl). In particular, the
substituent 4-O-Bun leads to the highest yield of over 90%.
This can be attributed to its fairly low solubility in water,
which is favorable for its separation from the reaction mix-
ture, hence promoting the reaction toward completion and
the subsequent recrystallization. However, aromatic alde-
hyde bearing stronger electron-withdrawing groups led to
lower yield, following the order 2a(3-NO2) < 2d(3-Cl) <
2e(4-Cl) < 2c(3-F) < 2i(4-CH3) < 2j(3-CH3O). This is
probably correlated to the steric hindrance effect of the sub-
stituent and the resultant electronic density distribution of
the molecule. Based on the results shown in Table 1, it be-
comes apparent that the nature, and location, of the substit-
uent on the aromatic aldehyde have significant impact on
the yield of the target molecules. Thus aromatic aldehydes
with the strongest electron-withdrawing group (4-NO2) or
electron-donating group (4-OH) lead to the lowest yield, or
even worse, do not undergo the reaction at all. In addition,
aromatic aldehydes with moderately strong electron-
withdrawing groups (e.g. 4-Br), or electron-donating groups
(e.g. 4-Me, 4-MeO) produce reasonable yields in the range
of 7080%. The yields follow the order 1m(3,4- OCH2O) >
1l(4-O-Bun) > 1k(4-CH3O) > 1j(3-CH3O) > 1h(3-CH3) >
1i(4-CH3); 1f(4-Br) > 1b(3,4-diCl) > 1o(2,4- diCl) >
1d(3-Cl) in TM-1, and 2l(4-O-Bun) > 2j(3-CH3O) >
2i(4-CH3); 2f(4-Br) > 2e(4-Cl) > 2d(3-Cl) in TM-2. It is

494
Tang GX, et al. Sci China Chem April (2013) Vol.56 No.4
Table 1 Yield, -Glucosidase inhibition and PPAR relative activation of β-amino ketones in vitro
PPAR relative activation (%) b)
1 a) (nmol mL1)
-glucosidase inhibition (%)
TM
R1
Yield (%)
C
PPAR (at conc. C1)
1a
1b
1c
1d
1e
1f
3-NO2
3,4-diCl
3-F
60.1
62.1
75.6
45.0
84.4
75.1
72.3
81.2
80.3
75.3
81.5
82.5
89.0
57.8
51.2
78.0
50.2
30.7
76.8
73.7
64.2
71.1
83.2
82.5
76.4
82.9
86.5
84.8
93.6
73.0
56.4
20.06
19.14
21.22
20.49
20.49
18.78
22.05
21.38
21.38
20.68
20.68
19.02
20.10
21.31
19.14
21.31
20.49
18.99
18.17
20.02
19.38
19.38
17.84
20.76
20.18
20.18
19.55
19.55
18.06
19.03
20.10
6.40
48.50
10.00
21.80
66.50
33.19
10.21
6.60
8.75
73.66
28.65
8.49
3-Cl
4-Cl
3.34
4-Br
17.88
41.55
11.76
24.15
12.59
22.87
12.66
13.33
39.79
33.89
24.10
29.40
4.58
1g
1h
1i
H
3-CH3
4-CH3
3-CH3O
4-CH3O
4-O-Bu-n
3,4-OCH2O
4-OH
11.17
9.35
1j
1k
1l
13.29
12.21
7.93
1m
1n
1o
1p
1q
2a
2b
2c
2d
2e
2f
41.10
8.26
2,4-diCl
3-HO
1.56
2-Cl
6.39
3-NO2
3,4-diCl
3-F
10.69
40.72
13.84
8.71
1.71
25.68
1.26
3-Cl
4-Cl
24.18
29.24
13.32
25.45
13.89
10.34
29.26
10.37
1.75
3.63
4-Br
20.31
1.15
2g
2h
2i
H
3-CH3
4-CH3
3-CH3O
4-CH3O
4-O-Bu-n
3,4-OCH2O
4-OH
5.56
15.66
29.32
4.49
2j
2k
2l
14.26
11.56
3.23
2m
2n
34.38
74
Positive control
100
a) The concentration of each test compound is 10 µg mL−1 (the corresponding concentration in nmol mL−1 is shown in the Table); b) the value of PPAR is
an average value of four measurements.
Table 2 PPAR relative activation of some -amino ketones in vitro
Compound
C2 a)
PPARb)
1c
1f
1i
1j
1k
1l
1o
1p
1q
2c
2f
2i
2j
2l
10.61
35.91
9.39
85.93
10.69
93.47
10.34
72.31
10.34
103.32
9.51
68.53
9.57
53.20
10.66
23.11
10.24
64.92
10.01
63.70
8.92
88.48
10.09
130.91
9.78
80.66
9.03
40.41
a) The concentration of each test compound is 5.0 µg mL−1 (the corresponding concentration in nmol mL−1 is shown in the Table); b) the value of PPAR
is an average value of four measurements. The positive control is rosiglitazone, with an activation of 100% (10 μg mL−1).
The biological activity of the target molecules in the
TM-1 series is closely correlated to the intrinsic property of
the substitutent group R1 on the aromatic aldehyde. The
-GI activities of -amino ketones with the electron-
donating groups 3-CH3, 4-CH3, 3-CH3O, 4-CH3O, 4-C4H9O
or 3,4-OCH2O on the benzene ring of the aromatic aldehyde
are all at the 10% level with no significant difference be-
tween them, with the exception of 1n(4-OH) whose activity
is about 41%. It is probable that the compound with com-
paratively larger electronic density variation across the ar-
omatic aldehyde moiety has lower solubility which, in turn,
weakens the competitive binding of -glucosidases at the
hypso-concentration. This inevitably leads to limited -GI
activity. In contrast, 1n(4-OH) showed good solubility due
to smaller 4-OH substituent. Therefore it is more likely to
bind effectively to the protein substrate, an effect supple-
mented by the extended hydrogen bonding network and 3-D
matching. This naturally leads to considerably higher bioac-
tivities. These results indicate that the electronic density
distribution is an important, but not the only, factor that

Tang GX, et al. Sci China Chem April (2013) Vol.56 No.4
495
be inversely proportional to the electron-withdrawing
strength of the substituents.
2.2.2 PPAR activation
The target molecules were evaluated in vitro for PPAR ago-
nist activity at a concentration level of 10 µg mL−1, showing
mainly moderate activities. However, 1b (R1 = 3,4-diCl)
exhibits an outstanding PPAR relative activation activity of
73.66%, outperforming the rest of the group by a significant
margin. Examination of the SAR for PPAR agonist activity
demonstrates that TM-1 with a p-aminobenzoic acid moiety
is more active than its TM-2 counterpart (with an ethyl
p-aminobenzoate moiety). This is primarily a reflection of
the SAR: the built-in molecular pattern requires the pres-
ence of an acidic terminal. This is again probably related to
the disparity of their solubilities. Moreover, enhancement in
solubility not only increases the availability of molecules in
the solution phase, it also improves the binding strength to
PPAR receptor. It also probably affects the kinetic rate of
hydrolysis of carboxylic acid esters which, in turn, modu-
lates the biological activity. As shown in Figure 4, the
changes in PPAR of 1 and 2 follow an almost identical
trend with the notable exceptions of b(R1 = 3, 4-diCl), g(R1
= H) and n(R1 = 4-OH) where some divergence is observed.
In addition, j(R1 = 3-CH3O) and k(R1 = 4-CH3O) exhibit
divergence on a much smaller scale. As for chloro-
substituted target molecules, the difference in activity for
the dichloro-substituted compound (1b) is greater than that
for monochloro-substituted compound, especially in the
TM-1 series.
Figure 3 Comparison of the α-glucosidase inhibition of TM-1 and TM-2.
dictates the bioactivities. Overall, the -GI bioactivities of
-amino ketones with electron-withdrawing groups have
considerable disparity. For chloro-substituted aromatic alde-
hydes, such as with 4-Cl, 3-Cl or 2-Cl substitutuents, the
-GI activities of the corresponding TM-1 are 66.50%,
21.80% and 6.39%, respectively, showing additional trend
beyond the electronic density distribution. This is most
probably related to the 3-D conformation, steric hindrance
and molecular rigidity—in short, the compatibility between
the target molecule and the enzyme. The higher -GI activ-
ity (48.50%) of 3,4-diCl (1b), as compared to 2,4-diCl (1o,
8.26%) and 2-Cl (1q, 6.39%) analogues, probably originates
from the steric hindrance.
Further investigation into the SAR of the -GI activity
reveals that the compounds of both series 1 and 2 with
electron-withdrawing groups on the aromatic aldehyde
show a similar trend. However, in the presence of elecron-
donating groups, TM-2 outperforms TM-1 in biological
assay, showing the overall molecular structure dictates its
bioactivities. Generally, for TM-2 originating from pa-
ra-substituted aromatic aldehydes, the -GI activities are
stronger than those with a meta-substituent. The -GI activ-
ities vary as follows: 2n(4-OH) > 2l (4-C4H9O) > 2m(3,
4-OCH2O). This is intimately associated with molecular
electronegativity as an exposed OH group serves to en-
hance the hydrogen bonding network and solubility. This in
turn improves the probability of colliding with the target
species, and ultimately leads to higher bioactivities, exactly
as in the case of 2n(>40%). Based on these results, we pro-
pose that the OH substituent in the product improves
water-solubility and results in stronger binding to
-glucosidase. The -GI activities of certain products with
electron-withdrawing groups (3-NO2, 3-F, 4-Br) appear to
In TM-1, the -amino ketones with electron-donating
alkoxy or hydroxy groups, including 3-CH3O, 4-CH3O,
4-C4H9O, 3,4-OCH2O, 3-OH and 4-OH, show similar PPAR
relative activation activities under the 40% mark, except for
4-OH whose PPAR relative activation activity is over 70%.
Figure 4 Comparison of PPAR relative activation of 1 and 2.

496
Tang GX, et al. Sci China Chem April (2013) Vol.56 No.4
This is likely attributed to its hydrophilic feature and more
effective binding to its protein receptor. However, the
product with a 3-OH substituent leads to a weaker PPAR
than that with a 4-OH. We propose that a 3-OH substituent
has a larger steric hindrance than a 4-OH substituent due to
different localization and space volume. This is also the
origin of the trend 1g(H) > 1i(4-CH3) > 1h(3-CH3). As for
chloro-substituted TM-1, the activity of the 3,4-diCl substi-
tuted compound is the strongest and the ortho-chlorosub-
stituted compound had better activity than the meta-chloro-
or para-chloro-substituted compound. The root cause of this
phenomenon remains unclear and is under continuing inves-
tigation. The trend in PPAR for TM-2 is analogous to that
for TM-1 but with a somewhat smaller disparity.
TM-1 exhibits similar trend in the change of -GI and
PPAR relative activation (Figure 6(a)) with only one ap-
parent departures (1e(4-Cl)), indicating that most of the
TM-1 series possess a dual-target property, while TM-1e is
a single-target molecule. However, the correlation of the
two bioactivities in the TM-2 series is significantly weaker
(Figure 6(b)), and indeed TM-2b, TM-2j, TM-2k, and
TM-2n show completely the opposite trend. Such differ-
ences for TM-2b and TM-2n are particularly great, and
characteristic of single-target binding.
It is interesting to observe, as shown in Figure 7, that the
lipo-hydro distribution coefficients for TM-1 and TM-2 as
a function of the individual substituent show a very similar
trend, which once again conforms the critical role played by
the substituent. For a given substituent, the magnitude of the
It is surprising to note that both TM-1 and TM-2 exhibit
greater PPAR relative activation at a 5.0 µg mL−1 concen-
tration level than at a higher concentration of 10 µg mL−1
(Table 2 and Figure 5). 2i and 1k show the highest activities
of 130.91% and 103.32%, respectively, indicating more
potent performance than the deployed positive control of
rosiglitazone (10 µg mL−1, 100%). 1i, 2f, 1f and 2j also
show remarkably high PPAR relative activation activities of
93.47%, 88.48%, 85.93% and 80.66%, respectively. The
change in 2i is most dramatic from a moderate 15.66% at 10
µg mL−1 to 130.91% at 5.0 µg mL−1. The fact that higher
concentrations of these compounds in Table 2 lead to re-
duced activation (Figure 5) is an important observation.
Although its root cause remains unclear up to this point,
further research is underway. We suppose that such dispari-
ties in PPAR may be related to the solubility, stability and
conformation of the compounds, and may also reflect the
activation effect at lower concentrations and the inhibition
effect at higher concentrations.
Figure 5 Comparison of PPAR relative activation at different concentra-
tions for selected compounds.
Figure 6 Comparison of -GI and PPAR relative activation.

Tang GX, et al. Sci China Chem April (2013) Vol.56 No.4
497
rosiglitazone in the 3-D conformations. For those species
with lower bioactivity such as 2b, 2d and 2g, the naphtha-
lene and ethyl 4-aminobenzoate rings are nearly coplanar, in
contrast to the case for their high activity counterparts and
Rosiglitazone. It has been reported [30] that the target mol-
ecules are bound in the cavity of a PPAR receptor in a
U-shaped conformation. However, the high profile drug
rosiglitazone does not conform to this general trend since it
does not show any U-shaped conformation at low internal
energy levels. It is hence plausible that fairly flexible mole-
cules including rosiglitazone are able to accommodate the
target molecules in their rectangular conformation before
entering the receptor cavity for binding. Those molecules
with sufficient flexibility show higher bioactivity. It tran-
spires that, in addition to its own structural features, the
target molecule should possess a high level of temporary
spatial compatibility with the receptor at the very moment
of binding [30, 47]. The target molecules 2i, 1k, 1j, 2f, 1f
and 2j probably owe their high levels of bioactivity to such
temporary structural compatibility with the receptor. It can
be observed from Figure 9 that the PPAR activation and
other parameters exhibit an irregular trend. Hence, taking
into consideration the multiple factors governing the bioac-
tivities—such as electronic density distribution, hydrogen
bonding network, space volume and molecular flexibil-
ity—the trend in bioactivity as shown by the large group of
high performance target molecules presented in this article
clearly merits further investigation.
Figure 7 Relationship of CLogP-substituent for selected compounds.
lipo-hydro distribution coefficient for TM-1 is smaller than
that for TM-2. Furthermore, the coincidence of the trends of
lipo-hydro distribution coefficient and biological activities
for different substituents shows the close correlation be-
tween the two parameters. Therefore the impact of the lipo-
hydro distribution coefficient should be taken into consider-
ation in further optimization of the target molecules.
Examination of the 3-D structures (Figure 8) of a group
of highly active species including 2i, 1k, 1j, 2f, 1f and 2j
shows the absence of a U-shaped geometry, which has been
suggested as an indispensible prerequisite for effective
binding [30]. Instead, these potent molecules have a rec-
tangular geometry, which is apparently consolidated by an
R1-bearing aromatic ring. Although the naphthalene ring
and the other two benzene rings are non-coplanar, the
structural module composed of naphthalene and 4-amino-
benzoic acid (or ethyl 4-aminobenzoate) closely resembles
3 Experimental
3.1 General methods
Melting points were determined using an Electrothermal
1
X-6 apparatus. H nuclear magnetic resonance (NMR) and
13C NMR spectra were recorded on a Bruker AV 300 spec-
trometer using tetramethylsilane (TMS) as an internal
Figure 8 Three-dimensional structures of selected target molecules (PPAR).

498
Tang GX, et al. Sci China Chem April (2013) Vol.56 No.4
Figure 9 Diagrams of a set of additional characteristics for selected compounds and the corresponding PPAR. (a) PPAR-Stretch diagram; (b) PPAR-bend
diagram; (c) PPAR-Stretch/bend diagram; (d) PPAR-Torsion diagram; (e) PPAR-Dipole/dipole diagram; (f) PPAR-Total energy diagram.
standard. The chemical shifts are reported in parts per mil-
lion (ppm), the coupling constants (J) are expressed in hertz
(Hz) and signals are described as singlet (s), doublet (d),
triplet (t), quartet (q), as well as multiplet (m). Electrospray
ionization mass spectra (ESI-MS) were obtained on an Ag-
ilent 1946B instrument. High resolution mass spectra (HR

Tang GX, et al. Sci China Chem April (2013) Vol.56 No.4
499
MS) analyses are conducted on a Daltonics Data Analysis
3.2 (Bruker) using the ESI technique. The progress of reac-
tions and the purity of the compounds were monitored by
thin layer chromatography (TLC) using solvent systems of
different polarities. All chemical reagents and solvents were
commercially available, and were used without further puri-
fication.
49.9, 43.5, 28.8; ESI-MS (m/z): 523 (M+1)+; HRMS: calcd
for C29H26BrNNaO4, [M+Na]+ 554.0937; found, 554.0915.
4-(5-(6-Methoxynaphthalen-2-yl)-3-oxo-1-p-tolylpentyla
1
mino)benzoic acid (1i). Yield 80.3 %, Mp 190192 ℃. H
NMR (DMSO-d6, 300 MHz): δ 2.23 (3H, s), 2.78 (1H, dd,
J1 = 4.6, J2 = 6.5 Hz), 2.80-2.86 (4H, m), 3.00 (1H, dd, J1 =
8.5, J2 = 16.3 Hz), 3.85 (3H, s), 4.90 (1H, dd, J1 = 7.8, J2 =
12.5Hz), 6.53 (2H, d, J = 8.5 Hz), 6.92 (1H, d, J = 7.7 Hz),
7.08 (2H, d, J = 7.6 Hz), 7.12 (1H, d, J = 2.2 Hz), 7.25 (2H,
d, J = 7.6 Hz), 7.257.29 (2H, m), 7.55 (1H, s), 7.58 (2H, d,
J = 8.5 Hz), 7.67 (1H, d, J = 7.6 Hz), 7.69 (1H, d, J = 7.7
Hz), 11.98 (1H, s); 13C NMR (DMSO-d6, 75MHz): δ 207.1,
167.4, 156.8, 151.6, 140.1, 136.2, 136.0, 132.7, 130.9(2C),
129.0, 128.8, 128.5, 127.7(2C), 126.7, 126.3, 125.9(2C),
118.5, 117.4, 111.8(2C), 105.7, 55.1, 52.1, 50.2, 43.6, 28.8,
20.6; ESI-MS (m/z) 468 (M+1)+; HRMS: calcd for
C30H29NNaO4, [M+Na]+ 490.1989; found, 490.1969.
3.1.1 General procedure for the synthesis of β-amino ke-
tones
To an alcoholic solution of aromatic aldehyde (2 mmol, 35
mL) in a round-bottom flask was added arylamine (2 mmol).
The solution was stirred continuously at ambient tempera-
ture. When a great quantity of precipitate appeared,
nabumetone (2 mmol) in chloroform (13 mL) and concen-
trated hydrochloric acid (0.10.2 mL) were added sequen-
tially. The mixture was stirred vigorously at controlled
temperature. The progress of the reaction was monitored by
TLC. On completion of the reaction, the mixture was frozen
in a refrigerator. A few hours later, the solid was subse-
quently collected by filtration. The filter cake was washed
with water (2 × 3 mL) and 95% ethanol (2 × 3 mL), and
then recrystallized from 95% ethanol. The yields vary from
3097%. Melting points were determined, and the struc-
4-(5-(6-Methoxynaphthalen-2-yl)-1-(3-methoxyphenyl)-
3-oxopentylamino)benzoic acid (1j). Yield 75.3%, Mp
1
145147 °C. H NMR (DMSO-d6, 300 MHz): δ 2.79 (1H,
dd, J1 = 4.4, J2 = 16.2 Hz), 2.842.91 (4H, m), 3.01 (1H, dd,
J1 = 9.2, J2 = 16.3 Hz), 3.70 (3H, s), 3.85 (3H, s), 4.93 (1H,
dd, J1 = 7.6, J2 = 12.3 Hz), 6.55 (2H, d, J = 8.6 Hz), 6.76
(1H, d, J = 7.8 Hz), 6.93 (1H, d, J = 6.1 Hz), 6.916.98 (2H,
m), 7.10 (1H, dd, J1 = 2.3, J2 = 8.9 Hz), 7.197.23 (1H, m),
7.25 (1H, d, J = 2.3 Hz), 7.29 (1H, d, J = 8.4 Hz), 7.56 (1H,
s), 7.59 (2H, d, J = 8.5 Hz), 7.67 (1H, d, J = 7.0 Hz), 7.70
(1H, d, J = 7.1 Hz), 12.01 (1H, s); 13C NMR (DMSO-d6, 75
MHz,): δ 207.0, 167.4, 159.4, 156.8, 151.4, 145.0, 136.2,
132.7, 131.0(2C), 129.5, 128.8, 128.5, 127.7, 126.7, 125.8,
118.6, 118.5, 117.5, 112.3, 112.0(2C), 111.8, 105.7, 55.1,
54.9, 52.3, 50.2, 43.6, 28.8. HRMS: calcd for C30H29NNaO5,
[M+Na]+ 506.1938; found, 506.1932.
1
tures were confirmed by H NMR, 13C NMR, ESIMS and
HRMS techniques.
3.1.2 Characterization of new target compounds
4-(1-(3-Fluorophenyl)-5-(6-methoxynaphthalen-2-yl)-3-oxo
pentylamino)benzoic acid (1c). Yield 75.6%, Mp
1
188189 °C. H NMR(DMSO-d6, 300 MHz): δ 2.58–3.05
(6H, m), 3.90 (3H, s), 4.894.93 (1H, m), 5.87 (1H, s), 6.48
(2H, d, J = 8.6 Hz), 6.90 (1H, d, J = 6.9 Hz), 7.037.11 (3H,
m), 7.207.25 (2H, m), 7.47 (1H, s), 7.52 (1H, s), 7.607.63
(2H, m), 7.73 (2H, d, J = 8.6 Hz), 11.98 (1H, s); 13C NMR
(DMSO-d6, 75 MHz): δ 206.4, 167.4, 156.7, 151.3, 139.9,
136.0, 132.8, 131.8, 131.0(2C), 129.5, 128.9, 129.0, 128.6,
127.9, 127.6, 127.5, 126.7, 125.9, 118.5, 117.9, 111.6(2C),
105.6, 55.2, 49.1, 47.9, 43.5, 28.8; ESI-MS (m/z): 472
(M+1)+; HRMS: calcd for C29H26FNNaO4 [M+Na]+
494.1738; found, 494.1728.
4-(5-(6-Methoxynaphthalen-2-yl)-1-(4-methoxyphenyl)-
3-oxopentylamino)benzoic acid (1k). Yield 81.5%, Mp
1
181183 °C. H NMR (DMSO-d6, 300 MHz): δ 2.77 (1H,
dd, J1 = 3.8, J2 = 16.2 Hz), 2.82–2.89 (4H, m), 3.01 (1H, dd,
J1 = 9.1, J2 = 16.1 Hz), 3.69 (3H, s), 3.85 (3H, s), 4.90 (1H,
dd, J1 = 7.7, J2 = 12.6 Hz), 6.54 (2H, d, J = 8.0 Hz), 6.84
(2H, d, J = 7.7 Hz), 6.92 (1H, d, J = 6.9 Hz), 7.11 (1H, d, J
= 8.6 Hz), 7.29 (4H, m), 7.54 (1H, s), 7.59 (2H, d, J = 7.9
Hz), 7.67 (1H, d, J = 7.9 Hz), 7.69 (1H, d, J = 6.9 Hz),
12.02 (1H, s); 13C NMR (DMSO-d6, 75 MHz): δ 207.3,
167.5, 158.2, 156.8, 151.6, 136.2, 135.0, 132.8, 131.0(2C),
128.8(2C), 128.6, 127.7, 127.6, 126.7(2C), 125.9, 118.5,
117.3, 113.8, 111.8(2C), 105.7, 55.1, 55.0, 51.7, 50.3, 43.6,
28.8; HRMS: calcd for C30H29NNaO5, [M+Na]+ 506.1938;
found, 506.1933.
4-(1-(4-Butoxyphenyl)-5-(6-methoxynaphthalen-2-yl)-3-
oxopentylamino)benzoic acid (1l). Yield 82.5%, Mp
196198 °C. 1H NMR (DMSO-d6, 300 MHz): δ 0.91 (3H, t,
J = 7.3 Hz), 1.391.44 (2H, m), 1.631.67 (2H, m), 2.77
(1H, dd, J1 = 3.8, J2 = 16.2 Hz), 2.822.89 (4H, m), 3.01
(1H, dd, J1 = 8.4, J2 = 16.3 Hz), 3.85 (3H, s), 3.89 (2H, t, J
4-(1-(4-Bromophenyl)-5-(6-methoxynaphthalen-2-yl)-3-
oxopentylamino)benzoic acid (1f). Yield 75.1%, Mp
1
205207 °C. H NMR (DMSO-d6, 300 MHz): δ 2.81 (1H,
dd, J1 = 4.6, J2 = 16.5 Hz), 2.85-2.90 (4H, m), 3.03 (1H, dd,
J1 = 8.8, J2 = 16.2 Hz), 3.85 (3H, s), 4.95 (1H, dd, J1 = 7.7,
J2 = 12.4 Hz), 6.54 (2H, d, J = 8.5 Hz), 6.97 (1H, d, J = 8.6
Hz), 7.11 (1H, d, J = 8.9 Hz), 7.257.30 (2H, m), 7.34 (2H,
d, J = 8.3 Hz), 7.47 (2H, d, J = 8.3 Hz), 7.55 (1H, s), 7.60
(2H, d, J = 8.4 Hz), 7.67 (1H, d, J = 6.8 Hz), 7.70 (1H, d, J
= 6.3 Hz), 12.04 (1H, s); 13C NMR (DMSO-d6, 75 MHz): δ
206.9, 167.4, 156.8, 151.3, 142.7, 136.2, 132.8, 131.3,
131.0(2C), 129.6, 128.8(2C), 128.6 (2C), 127.7, 126.7,
125.9, 120.0, 118.5, 117.7, 111.8(2C), 105.7, 55.1, 51.7,

500
Tang GX, et al. Sci China Chem April (2013) Vol.56 No.4
= 6.3 Hz), 4.91 (1H, dd, J1 = 7.8, J2 = 12.5 Hz), 6.54 (2H,
d, J = 8.5 Hz), 6.82 (2H, d, J = 8.3 Hz), 6.90 (1H, d, J = 7.7
Hz), 7.10 (1H, d, J = 6.9 Hz), 7.26 (2H, d, J = 8.2 Hz), 7.28
(2H, d, J = 6.5 Hz), 7.54 (1H, s), 7.59 (2H, d, J = 8.5 Hz),
7.67 (1H, d, J = 8.2 Hz), 7.69 (1H, d, J = 6.9 Hz), 12.00 (1H,
s); 13C NMR (DMSO-d6, 75 MHz): δ 207.2, 167.4, 157.6,
156.8, 151.6, 136.2, 134.8, 132.7, 130.9(2C), 128.8(2C),
128.5, 127.7, 127.5, 126.7, 125.8(2C), 118.5, 117.3, 114.3,
111.8(2C), 105.7, 67.0, 55.1, 51.8, 50.3, 43.6, 30.8, 28.8,
18.8, 13.7; ESI-MS (m/z): 564 (M+K)+; HRMS: calcd for
C33H35NNaO5, [M+Na]+ 548.2407; found, 548.2413.
calcd for C29H26ClNNaO4, [M+Na]+ 510.1443; found,
510.1425.
Ethyl 4-(5-(6-methoxynaphthalen-2-yl)-1-(3-nitrophenyl)-
3-oxopentylamino)benzoate (2a). Yield 30.7%, Mp
106108 °C. 1H NMR (CDCl3, 300 MHz): δ 1.35 (3H, t, J =
7.1 Hz), 2.773.00 (6H, m), 3.91 (3H, s), 4.28 (2H, q, J =
7.1 Hz), 4.93 (1H, t, J = 5.8 Hz), 4.95 (1H, s), 6.40 (2H, d, J
= 8.5 Hz), 7.087.14 (2H, m), 7.18 (1H, d, J = 8.2 Hz),
7.377.44 (2H, m), 7.587.61 (3H, m), 7.77 (2H, d, J = 8.5
Hz), 8.02 (1H, d, J = 8.5 Hz), 8.14 (1H, s); 13C NMR
(CDCl3, 75 MHz): δ 207.5, 166.7, 157.5, 150.0, 148.7,
144.3, 135.4, 133.2, 132.7, 131.8, 129.8(2C), 129.0, 128.1,
127.4, 127.2, 126.4, 122.6, 121.2, 120.0, 119.0, 112.6(2C),
105.7, 60.4, 55.4, 53.0, 49.8, 45.0, 29.5, 14.5; ESI-MS (m/z):
549 (M+Na)+; HRMS: calcd for C31H30N2NaO6, [M+Na]+
549.1996; found 549.2003; calcd for C31H30N2KO6, [M+K]+
565.1741; found, 565.1748.
Ethyl 4-(1-(3,4-dichlorophenyl)-5-(6-methoxynaphthalen-
2-yl)-3-oxopentylamino)benzoate (2b). Yield 76.8%, Mp
142145 °C. 1H NMR (CDCl3, 300 MHz): δ 1.30 (3H, t, J =
7.1 Hz), 2.73 (2H, t, J = 7.2 Hz), 2.83 (2H, d, J = 5.9 Hz),
2.94 (2H, t, J = 7.2 Hz), 3.91 (3H, s), 4.25 (2H, q, J = 7.1
Hz), 4.75 (1H, t, J = 5.9 Hz), 4.91 (1H, s), 6.36 (2H, d, J =
8.5 Hz), 7.05–7.16 (4H, m), 7.24 (1H, d, J = 8.4 Hz), 7.33
(1H, s), 7.42 (1H, s), 7.58 (2H, d, J = 8.4 Hz), 7.74 (2H, d, J
= 8.5 Hz); 13C NMR (CDCl3, 75 MHz): δ 207.8, 166.7,
157.5, 150.0, 142.3, 135.5, 133.3, 133.1, 131.7, 131.6,
131.5, 130.9(2C), 129.0, 128.3, 127.4, 127.2, 126.4, 125.7,
119.9, 119.1, 112.6(2C), 105.8, 60.4, 55.4, 52.9, 50.0, 45.2,
29.6, 14.5; ESI-MS (m/z): 572 (M+Na)+; HRMS: calcd for
C31H29Cl2NNaO4, [M+Na]+ 572.1366; found, 572.1313.
Ethyl 4-(1-(3-fluorophenyl)-5-(6-methoxynaphthalen-2-
yl)-3-oxopentylamino)benzoate (2c). Yield 73.7%, Mp
129132 ℃. 1H NMR (DMSO-d6, 300 MHz): δ 1.24 (3H, t,
J = 6.9 Hz), 2.73–3.09 (6H, m), 3.85 (3H, s), 4.17 (2H, q, J
= 6.9 Hz), 5.0 (1H, s), 6.57 (2H, d, J = 7.9 Hz), 6.98–7.02
(2H, m), 7.10 (1H, d, J = 7.6 Hz), 7.24–7.34 (5H, m), 7.56
(1H, s), 7.61–7.71 (4H, m); 13C NMR (DMSO-d6, 75 MHz):
δ 206.8, 165.8, 163.9, 160.7, 156.8, 151.6, 146.4, 136.2,
132.7, 130.8, 130.4, 130.3(2C), 128.8, 128.5, 127.6, 126.7,
125.8, 122.6, 118.5, 116.9, 113.0(2C), 105.7, 59.6, 55.1,
51.8, 49.8, 43.5, 28.8, 14.3; ESI-MS (m/z): 522 (M+Na)+;
HRMS: calcd for C31H30FNNaO4, [M+Na]+ 522.2138;
found 522.2049, calcd for C31H30FNKO4 [M+K]+ 538.1883;
found, 538.1796.
4-(1-(2,4-Dichlorophenyl)-5-(6-methoxynaphthalen-2-yl)
-3-oxopentylamino)benzoic acid (1o). Yield 51.2%, Mp
1
159162 °C. H NMR (DMSO-d6, 300 MHz): δ 2.82 (1H,
dd, J1 = 4.4, J2 = 16.2 Hz), 2.892.94 (4H, m), 3.02 (1H, dd,
J1 = 10.0, J2 = 17.0 Hz), 3.85 (3H, s), 5.22 (1H, dd, J1 = 7.6,
J2 = 12.3 Hz), 6.44 (2H, d, J = 8.4 Hz), 7.11 (2H, d, J = 8.6
Hz), 7.26 (1H, s), 7.32 (1H, d, J = 7.9 Hz), 7.40 (1H, s),
7.43 (1H, s), 7.59 (1H, s), 7.63 (2H, d, J = 8.6 Hz), 7.69 (1H,
d, J = 6.4 Hz), 7.71 (1H, d, J = 5.8 Hz); 13C NMR
(DMSO-d6, 75 MHz): δ 206.3, 167.3, 156.8, 150.9, 139.2,
136.1, 132.8, 132.7, 132.4, 131.1(2C), 129.0, 128.9, 128.8,
128.6, 127.9, 127.5, 126.7, 125.9, 118.5, 118.1, 111.6(2C),
105.8, 55.1, 49.1, 47.7, 43.7, 28.8; ESI-MS (m/z): 522
(M+1)+; HRMS: calcd for C29H25Cl2NNaO4, [M+Na]+
544.1053; found, 546.1043.
4-(1-(3-Hydroxyphenyl)-5-(6-methoxynaphthalen-2-yl)-
3-oxopentylamino)benzoic acid (1p). Yield 78.0%, Mp
1
187189 °C. H NMR (DMSO-d6, 300 MHz): δ 2.752.80
(1H, dd, J1 = 4.4, J2 = 16.4 Hz), 2.832.94 (4H, m),
2.963.04 (1H, dd, J1 = 9.1, J2 = 16.3 Hz), 3.85 (3H, s),
4.844.88 (1H, dd, J1 = 8.0, J2 = 11.6 Hz), 6.53 (2H, d, J =
8.6 Hz), 6.60 (1H, d, J = 8.1 Hz), 6.79 (2H, d, J = 8.4 Hz),
6.91 (1H, d, J = 7.4 Hz), 7.057.13 (2H, m), 7.25 (1H, s),
7.29 (1H, d, J = 8.4 Hz), 7.56 (1H, s), 7.60 (2H, d, J = 8.6
Hz), 7.67 (1H, d, J = 6.0 Hz), 7.70 (1H, d, J = 5.4 Hz), 9.35
(1H, s), 12.00 (1H, s); 13C NMR (DMSO-d6, 75 MHz): δ
207.1, 167.5, 157.5, 156.8, 151.6, 144.8, 136.2, 132.8,
131.0(2C), 129.5, 128.8, 128.6, 127.7, 126.7, 125.9, 118.5,
117.4, 117.1, 113.9, 113.1, 111.7(2C), 105.7, 55.1, 52.3,
50.2, 43.6, 28.8; ESI-MS (m/z): 470 (M+1)+; HRMS: calcd
for C29H27NNaO5, [M+Na]+ 492.1781; found, 492.1767.
4-(1-(2-Chlorophenyl)-5-(6-methoxynaphthalen-2-yl)-3-
oxopentylamino)benzoic acid (1q). Yield 50.2%, Mp
1
162164 °C. H NMR (DMSO-d6, 300 MHz): δ 2.74-3.06
Ethyl 4-(1-(3-chlorophenyl)-5-(6-methoxynaphthalen-2-
yl)-3-oxopentylamino)benzoate (2d). Yield 64.2%, Mp
132134 °C. 1H NMR (CDCl3, 300 MHz): δ 1.33 (3H, t, J =
7.1 Hz), 2.692.83 (2H, m), 2.88 (2H, t, J = 5.9 Hz), 2.97
(2H, t, J = 7.2 Hz), 3.91 (3H, s), 4.28 (2H, q, J = 7.1 Hz),
4.81 (1H, t, J = 5.9 Hz), 4.93 (1H, s), 6.41 (2H, d, J = 8.6
Hz), 7.097.20 (7H, m), 7.45 (1H, s), 7.62 (2H, d, J = 8.6
Hz), 7.77 (2H, d, J = 8.6 Hz); 13C NMR (CDCl3, 75 MHz):
δ 207.9, 166.7, 157.5, 150.2, 144.0, 135.6, 134.9, 133.2,
(6H, m), 3.85 (3H, s), 5.225.28 (1H, m), 6.44 (2H, d, J =
8.5 Hz), 7.05 (1H, d, J = 7.3 Hz), 7.11 (1H, d, J = 8.9 Hz),
7.26-7.33 (4H, m), 7.44 (2H, d, J = 6.7 Hz), 7.59 (1H, s),
7.62 (2H, d, J = 8.6 Hz), 7.68 (1H, d, J = 6.7 Hz), 7.71 (1H,
d, J = 6.4 Hz), 12.06 (1H, s); 13C NMR (DMSO-d6, 75
MHz): δ 206.5, 167.4, 156.8, 151.1, 139.9, 136.2, 132.8,
131.8, 131.1(2C), 129.5, 128.9, 128.8, 128.6, 127.8, 127.7,
127.5, 126.7, 125.9, 118.5, 117.9, 111.5(2C), 105.7, 55.1,
49.3, 48.0, 43.7, 28.8; ESI-MS (m/z): 488 (M+1)+; HRMS:

Tang GX, et al. Sci China Chem April (2013) Vol.56 No.4
501
131.4, 130.3(2C), 129.0, 128.6, 127.9, 127.7, 127.2, 126.4,
125.9, 124.4, 119.7, 119.0, 112.6(2C), 105.7, 60.3, 55.4,
53.5, 50.0, 45.3, 29.5, 14.5; ESI-MS (m/z): 538 (M+Na)+;
HRMS: calcd for C31H30ClNNaO4, [M+Na]+ 538.1028;
found 538.1021, calcd for C31H30ClNKO4, [M+K]+
554.0773; found, 554.0732.
Hz), 2.95 (2H, t, J = 7.3 Hz), 3.91 (3H, s), 4.29 (2H, q, J =
7.1 Hz), 4.83 (1H, t, J = 6.2 Hz), 4.95 (1H, s), 6.45 (2H, d, J
= 8.7 Hz), 7.037.13 (4H, m), 7.14-7.19 (3H, m), 7.45 (1H,
s), 7.62 (2H, d, J = 8.6 Hz), 7.77 (2H, d, J = 8.7 Hz); 13C
NMR (CDCl3, 75 MHz): δ 208.5, 166.9, 157.4, 150.6, 141.6,
138.6, 135.8, 133.2, 131.4(2C), 129.0, 128.9, 128.7, 128.5,
127.4, 127.1, 126.9, 126.3, 123.2, 119.2, 119.0, 112.5(2C),
105.7, 60.2, 55.4, 53.9, 50.2, 45.3, 29.4, 21.6, 14.5; ESI-MS
(m/z): 518 (M+Na)+; HRMS: calcd for C32H33NNaO4,
[M+Na]+ 518.2302; found 518.2292, calcd for C32H33NKO4,
[M+K]+ 534.2047; found, 534.2033.
Ethyl 4-(5-(6-methoxynaphthalen-2-yl)-3-oxo-1-p-toly-
lpentylamino)benzoate (2i). Yield 82.9%, Mp 135 137 °C.
1H NMR (DMSO-d6, 300 MHz): δ 1.23 (3H, t, J = 6.9 Hz),
2.23 (3H, s), 2.753.06 (6H, m), 3.85 (3H, s), 4.18 (2H, q, J
= 6.9 Hz), 4.894.94 (1H, m), 6.55 (2H, d, J = 8.4 Hz), 7.00
(1H, d, J = 7.6 Hz), 7.077.13 (3H, m), 7.247.30 (4H, m),
7.54 (1H, s), 7.60 (2H, d, J = 8.4 Hz), 7.667.71 (2H, m);
13C NMR (DMSO-d6, 75 MHz): δ 207.1, 165.8, 156.8,
151.8, 140.0, 136.2, 136.0, 132.8, 130.7(2C), 129.0, 128.8,
128.6, 127.6(2C), 126.7, 126.3, 125.8(2C), 118.4, 116.6,
111.9(2C), 105.8, 59.5, 55.1, 52.1, 50.2, 43.6, 28.8, 20.6,
14.3; ESI-MS (m/z): 518 (M+Na)+; HRMS: calcd for
C32H33NNaO4, [M+Na]+ 518.2302; found 518.2302, calcd
for C32H33NKO4, [M+K]+ 534.2047; found, 534.2051.
Ethyl 4-(5-(6-methoxynaphthalen-2-yl)-1-(3-methoxyph-
enyl)-3-oxopentylamino)benzoate (2j). Yield 86.5%, Mp
136140 °C. 1H NMR (DMSO-d6, 300 MHz): δ 1.24 (3H, t,
J = 7.0 Hz), 2.683.04 (6H, m), 3.70 (3H, s), 3.91 (3H, s),
4.18 (2H, q, J = 7.0 Hz), 4.874.94 (1H, m), 6.57 (2H, d, J
= 8.5 Hz), 6.78 (1H, d, J = 7.6 Hz), 6.937.02 (3H, m), 7.13
(1H, m), 7.207.30 (3H, m), 7.56 (1H, s), 7.62 (2H, d, J =
8.5 Hz), 7.667.71 (2H, m); 13C NMR (DMSO-d6, 7 5
MHz): δ 207.0, 165.8, 159.4, 156.8, 151.8, 144.9, 136.2,
132.8, 130.8(2C), 129.5, 128.8, 128.6, 127.6, 126.7, 125.8,
118.6, 118.5, 116.6, 112.3, 112.1, 111.9(2C), 105.7, 59.6,
55.1, 54.9, 52.2, 50.1, 43.6, 28.8, 14.3; ESI-MS (m/z): 534
(M+Na)+, 512 (M+1)+; HRMS: calcd for C32H33NNaO5,
[M+Na]+ 534.2251; found, 534.2249, calcd for C32H33NKO5,
[M+K]+ 550.1996; found, 550.1995.
Ethyl 4-(5-(6-methoxynaphthalen-2-yl)-1-(4-methoxyph-
enyl)-3-oxopentylamino)benzoate (2k). Yield 84.8%, Mp
158160 °C. 1H NMR (CDCl3, 300 MHz): δ 1.32 (3H, t, J =
7.1 Hz), 2.70 (2H, t, J = 7.3 Hz), 2.87 (2H, d, J = 6.0 Hz),
2.94 (2H, t, J = 7.3 Hz), 3.74 (3H, s), 3.90 (3H, s), 4.27 (2H,
q, J = 7.1 Hz), 4.81 (1H, t, J = 6.0 Hz), 4.95 (1H, s), 6.44
(2H, d, J = 8.7 Hz), 6.79 (2H, d, J = 8.6 Hz), 7.11 (1H, dd,
J1 = 2.0, J2 = 8.9 Hz), 7.16 (2H, d, J = 8.5 Hz), 7.16 (2H, d,
J = 8.6 Hz), 7.43 (1H, s), 7.61 (2H, d, J = 8.4 Hz), 7.77 (2H,
d, J = 8.7 Hz); 13C NMR (CDCl3, 75 MHz): δ 208.6, 166.9,
159.0, 157.4, 150.6, 135.8, 133.5, 133.2, 131.4(2C),
129.0(2C), 127.6, 127.5, 127.3, 127.1, 126.4(2C), 119.2,
119.0, 114.4, 112.5(2C), 105.7, 60.3, 55.4, 55.3, 53.4, 50.1,
45.5, 29.4, 14.5; ESI- MS (m/z): 534 (M+Na)+; HRMS:
Ethyl 4-(1-(4-chlorophenyl)-5-(6-methoxynaphthalen-2-
yl)-3-oxopentylamino)benzoate (2e). Yield 71.1%, Mp
146147 °C. 1H NMR (CDCl3, 300 MHz): δ 1.33 (3H, t, J =
7.1 Hz), 2.72 (2H, t, J = 7.2 Hz), 2.89 (2H, t, J = 5.3 Hz),
2.96 (2H, t, J = 7.2 Hz), 3.91 (3H, s), 4.28 (2H, q, J = 7.1
Hz), 4.82 (1H, t, J = 5.3 Hz), 6.42 (2H, d, J = 8.5 Hz),
7.047.09 (3H, m), 7.167.23 (4H, m), 7.45 (1H, s), 7.62
(2H, d, J = 8.5 Hz), 7.77 (2H, d, J = 8.5 Hz); 13C NMR
(CDCl3, 75 MHz): δ 208.5, 166.1, 156.2, 151.7, 139.8,
133.6, 132.9, 132.3, 131.0, 130.1(2C), 129.4, 128.8(2C),
128.4(2C), 128.0, 127.2, 126.4, 119.2, 118.8, 114.0(2C),
105.9, 59.9, 53.3, 51.8, 50.6, 46.0, 29.1, 14.6; ESI-MS (m/z):
538 (M+Na)+; HRMS: calcd for C31H30ClNNaO4, [M+Na]+
538.1028; found 538.1017, calcd for C31H30ClNKO4,
[M+K]+ 554.0773; found, 554.0737.
Ethyl 4-(1-(4-bromophenyl)-5-(6-methoxynaphthalen-2-
yl)-3-oxopentylamino)benzoate (2f). Yield 83.2%, Mp
137140 °C. 1H NMR (DMSO-d6, 300 MHz): δ 1.24 (3H, t,
J = 6.9 Hz), 2.783.05 (6H, m), 3.85 (3H, s), 4.18 (2H, q, J
= 6.9 Hz), 4.884.95 (1H, m), 6.56 (2H, d, J = 8.6 Hz), 7.04
(1H, d, J = 7.6 Hz), 7.04 (1H, d, J = 8.3 Hz), 7.257.35 (3H,
m), 7.47 (2H, d, J = 7.5 Hz), 7.55 (1H, s), 7.62 (2H, d, J =
8.4 Hz), 7.667.71 (2H, m); 13C NMR (DMSO-d6, 75 MHz):
δ 206.8, 165.8, 156.8, 151.6, 142.6, 136.1, 132.7, 131.3,
130.8(2C), 128.9, 128.8, 128.5, 127.6, 126.7, 125.9, 120.0,
118.5, 116.8, 111.9(2C), 105.7, 59.6, 55.1, 51.7, 49.8, 43.5,
28.8, 14.3; ESI-MS (m/z): 582 (M+Na)+, 560 (M+1)+;
HRMS: calcd for C31H30BrNNaO4, [M+Na]+ 582.1250;
found 584.1259, calcd for C31H30BrNKO4, [M+K]+
598.0995; found, 600.0987.
Ethyl 4-(5-(6-methoxynaphthalen-2-yl)-3-oxo-1-pheny-
lpentylamino)benzoate (2g). Yield 82.5%, Mp 150152 ℃.
1H NMR (CDCl3, 300 MHz): δ 1.32 (3H, t, J = 7.1 Hz),
2.62-2.96 (6H, m), 3.90 (3H, s), 4.27 (2H, q, J = 7.1 Hz),
4.86 (1H, t, J = 6.1 Hz), 5.05 (1H, s), 6.44 (2H, d, J = 8.6
Hz), 7.047.23 (7H, m), 7.44 (1H, s), 7.61 (2H, d, J = 8.6
Hz), 7.76 (2H, d, J = 8.6 Hz); 13C NMR (CDCl3, 75 MHz):
δ 208.5, 166.8, 157.4, 150.5, 141.6, 135.8, 133.2, 131.5,
131.4(2C), 129.0(2C), 127.8, 127.7, 127.5, 127.1, 126.4,
126.2(2C), 119.3, 119.0, 112.5(2C), 105.7, 60.3, 55.3, 53.9,
50.1, 45.4, 29.4, 14.5; ESI-MS (m/z): 504 (M+Na)+; HRMS:
calcd for C31H31NNaO4, [M+Na]+ 504.2145; found
504.2157; calcd for C31H31NKO4, [M+K]+ 520.1890; found,
520.1894.
Ethyl 4-(5-(6-methoxynaphthalen-2-yl)-3-oxo-1-m-tolyl-
pentylamino)benzoate (2h). Yield 76.4 %, Mp 136138 °C.
1H NMR (CDCl3, 300 MHz): δ 1.33 (3H, t, J = 7.1 Hz),
2.30 (3H, s), 2.73 (2H, t, J = 7.3 Hz), 2.89 (2H, d, J = 6.2

502
Tang GX, et al. Sci China Chem April (2013) Vol.56 No.4
calcd for C32H33NNaO5, [M+Na]+ 534.2251; found,
534.2249, calcd for C32H33NKO5, [M+K]+ 550.1996; found,
550.1995.
3.2 Detection of α-glucosidase inhibition
-Glucosidase inhibitory activity was determined in a 100
μL reaction mixer containing optimal rat--glucosidase
(extracted from the rat small intestine of rat), 67 mmol L1
pH 6.8 sodium phosphate buffer and different samples.
Blank control (without enzyme and samples) or negative
control (without sample) was set as above. After incubation
at 37 °C for 10 min, 0.1 mol L1 maltose was added and
incubated for another 10 min at room temperature. The re-
action was stopped with 200 μL of glucose and the optical
density (OD) values at 490 nm recorded. The inhibition
ratio was calculated according to the following equation: I%
= [1(ODSampleODBlank) / (ODNegativeODBlank)] × 100%.
Based on the inhibition value, IC50 was calculated using the
4 Parameter Logistic Model in Xlfit.
Ethyl 4-(1-(4-butoxyphenyl)-5-(6-methoxynaphthalen-2-
yl)-3-oxopentylamino)benzoate (2l). Yield 93.6%, Mp
142144 °C. 1H NMR (DMSO-d6, 300 MHz): δ 0.90 (3H, t,
J = 7.3 Hz), 1.24 (3H, t, J = 7.0 Hz), 1.361.43 (2H, m),
1.621.67 (2H, m), 2.62–3.02 (6H, m), 3.85 (3H, s),
3.863.90 (2H, m), 4.18 (2H, q, J = 7.0 Hz), 4.854.91 (1H,
m), 6.56 (2H, d, J = 8.5 Hz), 6.82 (2H, d, J = 8.3 Hz), 6.97
(1H, d, J = 7.7 Hz), 7.11 (1H, d), 7.257.29 (4H, m), 7.54
(1H, s), 7.61 (2H, d, J = 8.5 Hz), 7.667.71 (2H, m); 13C
NMR (DMSO-d6, 75 MHz): δ 207.2, 165.8, 157.7, 156.8,
151.8, 136.2, 134.7, 132.8, 130.7(2C), 128.8, 128.6(2C),
127.6, 127.5, 126.7, 125.8(2C), 118.5, 116.5, 114.3,
111.9(2C), 105.7, 67.0, 59.5, 55.1, 51.8, 50.2, 43.6, 30.8,
28.8, 18.8, 14.3, 13.7; ESI-MS (m/z): 576 (M+Na)+, 554
(M+1)+; HRMS: calcd for C35H39NNaO5, [M+Na]+
3.3 Determination of PPAR activation
576.2720; found,
576.2739, calcd for C35H39NKO5,
[M+K]+ 592.2465; found, 592.2468.
HepG2 cells were cultured in low glucose Dulbecco’s Mod-
ified Eagle’s medium (DMEM) supplemented with 100
U mL1 streptomycin and penicillin. One day prior to trans-
fection, the cells were plated in 96-well plates with 1.5 ×
104 cells per well. When the cells grew at a confluence of
70%, plasmid pPPRE-Luc with firefly luciferase reporter
gene and the control plasmid phRL-TK with Renilla lucif-
erase reporter gene were transfected into the cells. 24 hours
after the transfection, the medium was replaced with fresh
medium containing either different samples, rosiglitazone
(positive control) or without sample (negative control). The
cells without transfection were used as a blank control. Af-
ter incubation for a further 24 h, the expression of lucifer-
ases was measured with a Dual-Luciferase Reporter Gene
Assay Kit (Promega). The times the samples became acti-
vated (T%) were calculated as the following equation:
Ethyl 4-(1-(benzo[d][1,3]dioxol-5-yl)-5-(6-methoxyna-
phthalen-2-yl)-3-oxopentylamino)benzoate (2m). Yield
73.0%, Mp 131132 °C. H NMR (CDCl3, 300 MHz): δ
1
1.33 (3H, t, J = 7.1 Hz), 2.732.75 (2H, m), 2.932.98 (4H,
m), 3.91 (3H, s), 4.28 (2H, q, J = 7.1 Hz), 4.75 (1H, t, J =
5.8 Hz), 5.91 (2H, s), 6.55 (2H, d, J = 8.4 Hz), 6.70 (2H, dd,
J1 = 7.9, J2 = 16.9 Hz), 6.78 (1H, s), 7.087.20 (3H, m),
7.46 (1H, s), 7.62 (2H, d, J = 8.6 Hz), 7.78 (2H, d, J = 8.4
Hz); 13C NMR (CDCl3, 75 MHz): δ 208.4, 166.9, 155.7,
151.3, 148.3, 145.5, 137.7, 135.6, 133.2, 129.9(2C), 129.1,
127.5, 127.2, 126.4, 126.0, 125.4, 118.8, 117.4, 112.5(2C),
107.1, 107.0, 105.7, 60.3, 55.4, 55.1,53.2, 49.8, 45.3, 29.5,
14.6; ESI-MS (m/z): 548 (M+Na)+, 526 (M+1)+; HRMS:
calcd for C32H31NNaO6, [M+Na]+ 548.2044; found,
548.2045, calcd for C32H31NKO6, [M+K]+ 564.1789; found,
564.1800.
Ethyl 4-(1-(4-hydroxyphenyl)-5-(6-methoxynaphthalen-
2-yl)-3-oxopentylamino)benzoate (2n). Yield 56.4%, Mp
159161 °C. 1H NMR (CDCl3, 300 MHz): δ 1.24 (3H, t, J =
7.0 Hz), 2.723.04 (6H, m), 3.85 (3H, s), 4.18 (2H, q, J =
7.0 Hz), 4.84 (1H, dd, J1 = 5.0, J2 = 8.3 Hz), 6.55 (2H, d, J
= 8.6 Hz), 6.68 (2H, d, J = 8.3 Hz), 7.10 (1H, dd, J = 2.0,
8.9 Hz), 7.17 (2H, d, J = 8.3 Hz), 7.28 (2H, d, J = 12.7 Hz),
7.54 (1H, s), 7.61 (2H, d, J = 8.6 Hz), 7.69 (2H, d, J = 8.6
Hz); 13C NMR (CDCl3, 75 MHz): δ 207.6, 166.0, 156.9,
156.3, 151.9, 136.3, 133.3, 132.9, 130.9(2C), 129.0(2C),
128.7, 127.8, 127.7, 126.9, 126.0(2C), 118.6, 116.6, 115.2,
112.0(2C), 105.9, 59.8, 55.3, 51.9, 50.4, 43.8, 28.9, 14.5;
ESI-MS (m/z): 520 (M+Na)+, HRMS: calcd for
C31H31NNaO5, [M+Na]+ 520.2094; found, 520.2070, calcd
for C31H31NKO5, [M+K]+ 536.1839; found, 536.1815.
The characterization of other target compounds, includ-
ing 1a, 1b, 1d, 1e, 1g, 1h, 1m, 1n has been reported by the
authors’ group previously [22].
T = [(L1 Sample – L1 Blank) / (L1 Negative – L1 Blank )] / [(L2 Sample
L2 Blank) / (L2 Negative – L2 Blank)] × 100%
–
Here, L1 represents the values for firefly luciferase and L2
represents the values for Renilla luciferase.
4 Conclusion
A group of 31 -amino ketones containing nabumetone,
aromatic aldehyde and p-aminobenzoic acid moieties have
been synthesized through direct Mannich reaction under
optimized synthetic conditions. At a lower concentration
level of 5.0 µg mL−1, compounds 1k and 2i exert the highest
PPAR relative activation activities 130.91% and 103.32%,
respectively, exceeding the positive control value for rosig-
litazone (10 µg mL−1, 100%). Both are proven highly active
target molecules with exceptional potential to become

Tang GX, et al. Sci China Chem April (2013) Vol.56 No.4
503
vinyloxiranes. Org Lett, 2004, 6: 345–347; b) He M, Pan ZX, Song B,
Zhang YP, Jin LH, Hu DY, Yang S, Song BA. Enantioselective syn-
thesis of -amino esters bearing a quinazoline moiety via a Man-
nich-type reaction catalyzed by a cinchona alkaloid derivative. Sci
China Chem, 2012, 10.1007/s11426-012-4822-4
commercially viable. They serve as concrete examples that
-amino ketones containing the 4-aminobenzoic acid unit
represent a group of highly valuable lead molecules in the
search for novel antidiabetic drugs.
14 a) Qiao CH, Xu XJ. The Beckmann rearrangement of 2-
arylaminobenzyl (or methyl) cyclohexanone oxime. J Beijing Normal
Univ (Nat Sci), 1998, 34: 89–92; b) He XW, Shang YJ, Hu JS, Ju K,
Jiang W, Wang SF. Syntheses of N-sulfonyl-N,N-disubstituted ami-
dines via a three-component free-radical coupling reaction of tertiary
amines and arenesulfonyl azides with terminal alkynes. Sci China
Chem, 2012, 55(2): 214-222
15 Karczmarzyk Z, Malinka W. Structural characterization of analgesic
isothiazolopyridines of Mannich base type; X-ray analysis of 2-[(4-
phenylpiperazin-1-yl)ethyl]-and 2-[(4-methylpiperazin-1-yl) methyl]-
4,6-dimethylisothiazolo[5,4-b]pyridin-3(2H)-ones. J Mol Struct, 2008,
888: 160–167
16 Das U, Das S, Bandy B, Stables JP, Dimmock JP. N-Aroyl-3,5-
bis(benzylidene)-4-piperidones: A novel class of antimycobacterial
agents. Bioorg Med Chem, 2008, 16: 3602–3607
17 Ji L, Long QX, Yang DC, Xie JP. Identification of Mannich base as a
novel inhibitor of mycoba-cterium tuberculosis isocitrate by high-
throughput screening. Int J Biol Sci, 2011, 7: 376–382
18 Xu LY, Tang H, Dong JH, Wang WM, She ZZ. Synthesis and
anti-inflammatory activity of the Mannich bases of 2-methyl-
5-(E)-(2-methoxybenzylidene) cyclopentanone. Chin J Med Chem,
2002, 12: 1–4
19 Yang DC, Fan L, Liu HP, Song XL, Xu J, Zhao J, Deng Y, Li CB.
Preparation of compounds with Mannich base structure and medical
application for preventing and/or treating leukemia. Chinese Patent
ZL200710092667.9, 2012-08-01
20 Joshi S, Khosla N, Khare D, Sharda R. Synthesis and in vitro study of
novel Mannich bases as antibacterial agents. Bioorg Med Chem Lett,
2005, 15: 221–226
21 Zhou C, Wu G, Feng Y, Li Q, Su H, Mais D, Zhu Y, Li N, Deng Y,
Yang DC, Wang MW. Discovery and biological characterization of a
novel series of androgen receptor modulators. Br J Pharmacol, 2008,
154: 440–450
22 (a) Wang H, Yan JF, Song XL, Fan L, Xu J, Zhou GM, Jiang L,
Yang DC. Synthesis and antidiabetic performance of -amino ketone
containing nabumetone moiety. Bioorg Med Chem, 2012, 20:
2119–2130; (b) Yang DC, Yan JF, Xu J, Ye F, Zhou ZW, Zhang WY,
Fan L, Chen X. Synthesis and investigation on antidiabetic activity of
4-(1-aryl-3-oxo-5-phenylpentyl-amino) benzenesulfon-amide. Acta
Pharm Sin, 2010, 45: 66–71; (c) Zhang YX, Yan JF, Fan L, Zhang
WY, Zhou ZW, Chen X, Su XY, Tang XM, Yang DC. Synthesis and
preliminary evaluation of antidiabetic activity of 4-(3-(4-
bromophenyl)-3-oxo-1-aryl–propylamino)-N-(5-methylisoxazol-3-yl)
benzenesulfonamide. Acta Pharm Sin 2009, 44: 1244–1251; (d) Xu J,
Yan J. F, Fan L, Song XL, Tang XM, Yang DC. Synthesis and
-glucosidase inhibitory activity of N-(1,5-diaryl-3-pentone-1-yl)
-4-aminobenzoic acid. Acta Pharm Sin, 2009, 44: 48–55; (e) Zhou
ZW, Yan JF, Tang XM, Zhang WY, Zhang YX, Chen X, Su XY,
Yang DC. Synthesis and preliminary evaluation of antidiabetic activ-
ity for -amino ketone containing isoxazole moiety. Chin J Org
Chem, 2010, 30: 582–589
23 Sun J, Yang YS, Li W, Zhang YB, Wang XL, Tang JF, Zhu HL.
Synthesis, biological evaluation and molecular docking studies of
1,3,4-thiadiazole derivatives containing 1,4-benzodioxan as potential
antitumor agents. Bioorg Med Chem Lett, 2011, 21: 6116–6121
24 Prabhu PP, Shastry CS, Pande S, Pai A. Synthesis, characterization
and biological evaluation of novel aryl benzothiazole derivatives. J
Pharm Res, 2011, 4: 2209–2211
25 Qiu JY, Xu BX, Huang ZM, Pan WD, Cao PX, Liu CX, Hao XJ,
Song BA, Liang GY. Synthesis and biological evaluation of Mati-
jing-Su derivatives as potent anti-HBV agents. Bioorg Med Chem,
2011, 19: 5352–5360
The authors are grateful to Dr. Qunli Luo and Mr. Ning Wang for their
assistance in measuring 1H and 13C NMR spectra. We thank the Natural
Science Foundation of Chongqing (2005BB5095), the Ph.D Research
Foundation of Southwest University (SWNU, B2005010), and the Scientific
and Technological Project in Chongqing (CSTC, 2011AB5001) for finan-
cial support.
1
Ottanà R, Maccari R, Giglio M, Corso AD, Cappiello M, Mura U,
Cosconati S, Marinelli L, Novellino E, Sartini S, Motta CL, Settimo
FD. Identification of 5-arylidene-4-thiazolidinone derivatives
endowed with dual activity as aldose reductase inhibitors and
antioxidant agents for the treatment of diabetic complications. Eur J
Med Chem, 2011, 46: 2797–2806
2
Liu KC, Sakya SM, O’Donnell CJ, Flick AC, Ding HX. Synthetic
approaches to the 2010 new drugs. Bioorg Med Chem, 2012, 20:
1155–1174
3
4
http://en.wikipedia.org/wiki/Linagliptin/GEN, News Highlights, 2011-
05-03
Jorgensen MR, Bhurruth-Alcor Y, Røst T, Bohov P, Müller M,
Guisado C, Kostarelos K, Dyrøy E, Berge RK, Miller AD, Skorve J.
Synthesis and analysis of novel glycerolipids for the treatment of
metabolic syndrome. J Med Chem, 2009, 52: 1172–1179
Shinkai H, Nishikawa M, Sato Y, Toi K, Kumashiro I, Seto Y,
Fukuma M, Dan K, Toyoshima S. N-(Cyclohexyl-carbonyl)-
D-phenylalanines and related compounds: A new class of oral
hypoglycemic agents. J Med Chem, 1989, 32: 1436–1441
Grell W, Hurnaus R, Griss G, Sauter R, Rupprecht E, Mark M, Luger
P, Nar H, Wittneben H, Müller P. Repaglinide and related
hypoglycemic benzoic acid derivatives. J Med Chem, 1998, 41:
5219–5246
Henke BR, Blanchard SG, Brackeen MF, Brown KK, Cobb JE,
Collins JL, Harrington WW, Hashim MA, Hull-Ryde EA, Kaldor I,
Kliewer SA, Lake DH, Leesnitzer LM, Lehmann JM, Lenhard JM,
Orband-Miller LA, Miller JF, Mook RA, Noble SA, Oliver W, Parks
DJ, Plunket KD, Szewczyk JR, Willson TM. N-(2-benzoylphenyl)-
L-tyrosine PPARγ agonists. 1. Discovery of a novel series of potent an-
tihyperglycemic and anti-hyperlipidemic agents. J Med Chem, 1998,
41: 5020–5036
5
6
7
8
9
Lohray BB, Bhushan V, Rao BP, Madhavan GR, Murali N, Rao KN,
Reddy AK, Rajesh BM, Reddy PG, Chakrabarti R, Vikramadithyan
RK, Rajagopalan R, Mamidi RN, Jajoo HK, Subramaniam S. Novel
euglycemic and hypolipidemic agents. 1. J Med Chem, 1998, 41:
1619–1630
Reginato MJ, Bailey ST, Krakow SL, Minami KC, Ishii S, Tanaka H,
Lazar MA. A potent antidiabetic thiazolidinedione with unique pe-
roxisome proliferator-activated receptor γ-activating prop-erties. J
Biol Chem, 1998, 273: 32679–32684
10 Klopfenstein SR, Evdokimov AG, Colson A, Fairweather NT, Neu-
man JJ, Maier MB, Gray JL, Gerwe GS, Stake GE, Howard BW,
Farmer JA, Pokross ME, Downs TR, Kasibhatla B, Peters K. 1, 2, 3,
4-Tetrahydro-isoquinolinyl sulfamic acids as phosphatase PTP1B in-
hibitors. Bioorg Med Chem Lett, 2006, 16: 1574–1578
11 Wan ZK, Lee J, Xu WX, Erbe DV, Joseph-McCarthy D, Follows BC,
Zhang YL. Monocyclic thiophenes as protein tyrosine phosphatase
1B inhibitors: Capturing interactions with Asp48. Bioorg Med Chem
Lett, 2006, 16: 4941–4945
12 Li BN, Cai SZ, Du DM, Xu JX. Synthesis of phosphinopeptides via
the Mannich ligation. Org Lett, 2007, 9: 2257–2260
13 a) Lautens M, Tayama E, Nguyen D. Direct vinylogous
Mannich-type reactions via ring opening and rearrangement of
26 Chen XF, Wu YB, Jin J, Wang RZ, Wang Y, Liu J. Design, synthesis
of quinolinone acid-containing compounds with anti-HIV integrase

504
Tang GX, et al. Sci China Chem April (2013) Vol.56 No.4
activity. Acta Pharm Sin, 2010, 45: 263–267
Synthesis of novel -aminoalcohols containing nabumetone moiety
with potential antidiabetic activity. Acta Pharm Sin, 2011, 46:
412–421
27 Tang XM, Fan L, Yu HX, Liao YH, Yang DC. Facile synthesis of
dipeptidomimetics of p-aminobenzoic acid and their antidiabetic ac-
tivity. Chin J Org Chem, 2009, 29: 595–600
28 Mulongo G, Mbabazi J, Odongkara B, Twinomuhwezi H, Mpango
GB. New biologically active compounds from 1, 3-diketones. Res J
Chem Sci, 2011, 1: 102–108
45 Yang DC, Yan JF, Song XL, Zhou ZW, Tang XM, Chen X, Fan L.
Synthesis and preliminary evaluation of antidiabetic activity of
4-[3-(4-bromophenyl)-3-oxo-1-arylpropylamino] benzenesulfonamide.
Acta Chim Sin, 2010, 68: 515–522
46 (a) Zhang YX, Yan JF, Fan L, Zhang WY, Su XY, Chen X, Tang
XM, Zhou ZW, Yang DC. Synthesis and preliminary evaluation of
antidiabetic activity of 4-(3-(4-hydroxyphenyl)-3-oxo-1-aryl-propyl-
amino)-N-(5-methylisoxazol-3-yl)benzenesulfonamide. Chin J Ap-
plied Chem, 2010, 27: 1026–1031; (b) Song XL, Yan JF, Fan L, Xu J,
Zhou ZW, Yang DC. Synthesis and preliminary evaluation of antidi-
abetic activity of 4-(1-aryl-3-aryl/arylalkyl-3-oxopropylamino)-
N-(5-methyl-3-isoxazolyl) benzenesulfonamide. Chin J Org Chem,
2009, 29: 606–613; (c) Tang XM, Yan JF, Zhang YX, Zhang WY, Su
XY, Chen X, Zhou ZW, Yang DC. Synthesis and preliminary study
on α-glucosidase inhibitory activity of 4-[3-(4-bromophenyl)-
3-oxo-1-arylpropylamino]-N-(pyrimidin-2-yl)benzenesulfonamide.
Chin J Org Chem, 2009, 29: 1790–1798; (d) Li QL, Yang DC, Yan
JF, Fan L, Chen X, Liu HP, Zhang WY, Song XL, Yan YH, Xu J,
Zhang J. Preparation of β-aminoketone (alcohol) derivatives as PPAR
agonists or insulin sensitizers. Chinese Patent Application:
200910258861.9; 2009; (e) Yang DC, Yan JF, Fan L, Chen X, Xu J,
Zhang WY, Song XL, Ye F, Liu HP, Jiang HW, Zhou ZW, Tang XM,
Zhang YX, Li TJ, Su XY. Preparation of β-aminoketone derivatives
for treating diabetic mellitus. Chinese Patent Application
200810237001.2; 2008; (f) Wang MW, Deng Y, Zhou CH, Yang DC,
Hui X, Su YR, Gao J. 3-Phenyl-3-(4-nitrobenzeneamino)-1-(4-
chlorophenyl)-1-acetone as non-steridesandrogen receptor moderator.
Chinese Patent ZL200910259690.1, 2011-04-27; (f) Yang DC, Xie
JP, Fan L, Yi L, Tang XM, Zhou ZW, Zhang YX, Su XY. Applica-
tion of β-amino ketone compounds to prepare medical agents for in-
29 Ammar Y, Mohamed Y, EI-Sharief A, EI-Gaby M, Abbas S. Reac-
tivity of 2, 3-pyridine dicarboxylic anhydride towards some nitrogen
nucleophilic reagents: Synthesis and antimicrobial evaluation of
some pyridine carboxamide and pyrrolo[3,4-B]pyridine-5,7-dione
derivatives. Chem Sci J, 2011, 16: 1–11
30 Feng J, Guo YS, Lu Y, Guo ZR. Design, virtual screening and syn-
thesis of PPAR agonists. Acta Chim Sin, 2004, 62: 1544–1550
31 Hedner T, Samulesson O, Wahrborg P, Wadenvik H, Ung K, Ekbom
A. Nabumetone therapeutic use and safety profile in the management
of osteoarthritis and rheumatoid arthritis. Drug, 2004, 64: 2315–2343
32 Zou JH, Yi L. The Mannich reaction among acetophenone,
p-methoxybenzaldehyde and arylamines. J Southwest Chin Norm
Univ (Nat Sci), 1991, 16: 66–70
33 Azizi N, Torkiyan L, Saidi MR. Highly efficient one-pot three-
component Mannich reaction in water catalyzed by heteropoly acids.
Org Lett, 2006, 8: 2079–2082
34 Ma XZ, Yi LN, Liu YX, Mei WJ, She ZZ. Synthesis and antitumor
activities of 2-(E)-(4-cyclopentyloxy-3-methylbenzyl idene) cyclop-
entanone arylamine Mannich bases. Chin J Med Chem, 2006, 16:
144–149
35 Li G, Li ZY, Wu CT. Michael addition of monoazabenzo-15-C-5
with acrylic esters and amides. Chin J Org Chem, 1998, 18: 462–464
36 Chen GX, Xu XJ, Liu LJ. Mannich reaction with arylamine as the
aminecomponent (II). Chem J Chin Univ, 1982, 3: 83–90
37 Xiao J, Wu YR. The preparation of N-Mannich base and its exchange
reaction with arylketones. J Beijing Norm Univ, 1996, 32: 367–369
38 Takaya J, Kagoshima H, Akiyama T. Mannich-type reaction with
trifluoromethylated N, O-hemiacetal: Facile preparation of
-amino--trifluoromethyl carbonyl compounds. Org Lett, 2000, 2:
1577–1579
hibiting
isocitrate
lyase.
Chinese
Patent
Application:
200910104544.1; 2009
47 Rao RR, Tiwari AK, Reddy P, Babu KS, Suresh G, Ali AZ,
Madhusudana K, Agawane SB, Badrinarayan P, Sastry GN, Rao JM.
Synthesis of antihyperglycemic, -glucosidase inhibitory, and DPPH
free radical scavenging furanochalcones. Med Chem Res, 2012, 6:
760–774
39 Xu XJ, Chen GX. The Mannich reaction with arylamines. Acta
Chimica Sin, 1982, 40: 463–467
40 Yi L, Zhou JH, Xu XJ. The Mannich reaction among aromatic ketone,
aromatic aldehyde and arylamines. Chin Chem, 1991, 5: 20–21
41 Yi L, Zhou JH, Lei HS, Lin XM, Zhang MX. The Mannich reaction
of cyclic ketones, aromatic aldehydes and aromatic amines. Org Prep
Proced Int, 1991, 23: 673–676
42 Yang DC. The Mannich reaction of 4-phenyl-2-butanone with aro-
matic aldehydes and aromatic amines. J Southwest Chin Norm Univ
(Nat Sci), 1996, 21: 354–359
48 Tomich CH, Silva PD, Carvalho I, Taft CA. Homology modeling and
molecular interaction field studies of -glucosidases as a guide to
structure-based design of novel proposed anti-HIV inhibitors. J
Comput Aided Mol Des, 2005, 19: 83–92
49 Lazar C, Durantel D, Macovei A, Zitzmann N, Zoulim F, Dwek RA,
Branza-Nichita N. Treatment of hepatitis B virus-infected cells with
a-glucosidase inhibitors results in production of virions with altered
molecular composition and infectivity. Antiviral Res, 2007, 76:
30–37
43 Yang DC, Zhang GL, Yang Y, Zhong YG. The Mannich reaction of
4-methylacetophenone with aromatic aldehydes and aromatic amines.
Chem J Chin Univ, 2000, 21: 1694–1696
50 Hakamata W, Kurihara M, Okuda H, Nishio T, Oku T. Design and
screening strategies for -glucosidase inhibitors based on enzymo-
logical information. Curr Top Med Chem, 2009, 9: 3–12
44 Zhang K, Yan JF, Tang XM, Liu HP, Fan L, Zhou GM, Yang DC.