Synthesis and anticancer activity of 2-alkylaminomethyl-5-diaryl- methylenecyclopentanone hydrochlorides and related compounds

Article abstract of DOI:10.1016/j.bmc.2004.11.009

Eleven new diaryl-methylenecyclopentanone Mannich hydrochlorides and related compounds were synthesized with different modification on Mannich base and α,β-unsaturated bonds. The glutathione-binding ability, glutathione-s-transferase π (GSTπ) inhibition a

Full text of DOI:10.1016/j.bmc.2004.11.009

Bioorganic & Medicinal Chemistry 13 (2005) 1285–1291
Synthesis and anticancer activity of 2-alkylaminomethyl-5-
diaryl-methylenecyclopentanone hydrochlorides and related
compounds
Jingli Wang,a Linxiang Zhao,a Rui Wang,a Min Lu,b Duo Chenb and Yongkui Jingb,*
aSchool of Pharmaceutical Engineering, Shenyang Pharmaceutical University, Shenyang 110015, China
bDepartment of Medicine, Mount Sinai School of Medicine, NewYork, NY 10029, USA
Received 10 August 2004; revised 6 November 2004; accepted 7 November 2004
Available online 25 November 2004
Abstract—Eleven new diaryl-methylenecyclopentanone Mannich hydrochlorides and related compounds were synthesized with dif-
ferent modification on Mannich base and a,b-unsaturated bonds. The glutathione-binding ability, glutathione-S-transferase p
(GSTp) inhibition and antitumor effect of these compounds were compared. Compounds containing both Mannich base and a-
unsaturated bond have GSH binding ability, GSTp inhibitory activity and antitumor effect. Compounds without Mannich base
or having a a-saturated bond lose GSH binding ability and the antitumor effect. Converting of Mannich base from dimethylami-
nomethyl group to morpholino, pyrrolidino, or piperidino-methyl groups do not evidently change the antitumor effect. However
replacement of phenyl group with methylphenyl group on b-chain significantly increases cytotoxic effect in breast cancer cells
but not in immortalized mammary epithelial cells. Our data suggest that diaryl-methylenecyclopentanones represent a new category
of compounds which might inhibit tumor growth through binding to glutathione or thiol proteins.
ꢀ 2004 Elsevier Ltd. All rights reserved.
1. Introduction
strate of antioxidant enzymes to prevent cell death and
as a substrate of glutathione-S-transferases (GSTs)
which involve in the drug resistance.9,10 Among the
GST family it has been found that GSTp was increased
in chemotherapy-resistance cancer cells.11 Recently it
has been found that GSTp functions as an antioxidant
enzyme to block apoptosis-induced by some chemothera-
peutic agents through diminishing free radical oxygen
and directly inhibition of c-Jun N-terminal kinase
(JNK).12,13 Thus GSTp could be used as a potential tar-
get for the cancer treatment.14 The potential GSH bind-
ing ability of these compounds suggests that they may
inhibit GSTp activity directly or indirectly, and would
be potent anticancer agents in tumor cells even with
higher GSTp activity.
Diterpenes and sesquiterpenes containing a,b-unsatu-
rated carbonyl structures have been found to have a
strong antineoplastic activity.1,2 Similarly to these natu-
ral compounds, it has been found that many synthetic
compounds containing a,b-unsaturated carbonyl struc-
tures have remained the antitumor effect.3–5 These stud-
ies suggest that a,b-unsaturated carbonyl structures
would be a critical substitute to mediate the observed
antitumor effect among these compounds. Moreover, it
has been found that compounds containing Mannich
base, which is a masked bioactive group, could produce
the methylene substances upon 1,2-elimination and have
an enhanced antitumor effect.6–8 The mechanism inhib-
iting tumor cell growth of these compounds has been
thought to be mediated by binding of a,b-unsaturated
carbonyl structures to free cysteine sulfhydryl groups
including glutathione (GSH) to form thiol-addition
product.8 GSH plays an important role in cells as a sub-
In our current study, we have introduced two phenyl
groups into b-carbon of previously reported cyclopenta-
none8 to obtain new compounds (7a–h). To determine
the importance of a, or b-unsaturated bonds on the anti-
tumor activity, a selective reduction of the b-unsaturated
bond of 7a and 7e was made into the saturated ketone
analogues 8a and 8b, and further reduction of 8a gener-
ated the a-, b-saturated analogue 9. To test the impor-
tance of a, or b-unsaturated bond or Mannich base,
Keywords: Cyclopentanone; Structure–activity relationship; Glutathi-
one; Glutathione-S-transferase; Growth inhibition.
*
Corresponding author. Tel.: +1 2122416775; fax: +1 2129965787;
e-mail: yongkui.jing@mssm.edu
0968-0896/$ - see front matter ꢀ 2004 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmc.2004.11.009

1286
J. Wang et al. / Bioorg. Med. Chem. 13 (2005) 1285–1291
the GSH binding ability, GSTp inhibition ability and
antitumor activity, were compared among these com-
pounds (5–9 listed in Fig. 1) with modification of a, or
b-unsaturated bond or Mannich base.
The reagents used were as follows, namely: (i) ethylene
glycol; (ii) aryl Grignard reagent; (iii) dilute aqueous–
ethanolic hydrochloric acid; (iv) hydrogen over palla-
dium on carbon powder; (v) concentrated hydrochloric
acid, paraformaldehyde and secondary amine; (vi)
N,N-dimethylmethyleneammonium chloride; (vii) so-
dium borohydride.
2. Results and discussion
2.1. Chemistry
2.2. Interaction with GSH
The compounds 7–9 were prepared according to the
procedure indicated in Figure 1. The starting com-
pound, ethyl 1,4-dioxaspiro[4,4]nonane-6-carboxylate
(4), was synthesized according to previously described
methods.15 2-Diarylmethylenecyclopentanones (5a, b)
were synthesized by the reaction of the appropriate aryl
Grignard reagents with 4 following by hydrolysis and
dehydration using dilute aqueous–ethanolic mineral
acid using a general method reported previously.16
Compounds 7a, b were synthesized from the corre-
sponding 2-diarylmethylenecyclopentanones, secondary
amines and paraformaldehyde and the products were
isolated as hydrochlorides. The ketonic groups in the
compounds 5a, b were reduced into 2-arylmethylcyclo-
pentanones (6a, b), respectively, by catalytic hydrogena-
tion with hydrogen over Pd–C at atmospheric pressure
based on a reported procedure.17 The reactions of 6a,
b with N,N-dimethylmethyleneammonium chloride led
to the formation of 8a, b with Mannich bases, respec-
tively. Compound 8a was reduced by NaBH4 in metha-
nol to yield the corresponding alcohol analogue 9. All
the synthesized Mannich base hydrochlorides and com-
pound 9 were chemically characterized by melting point
(mp), infrared (IR) and nuclear magnetic resonance (1H
NMR) spectra as well as mass spectra (MS).
The interaction of these cyclopentanone Mannich base
hydrochlorides with GSH were measured by reported
method based on the interaction of DTNB [5,50-dithio-
bis(2-nitrobenzoic acid)] with GSH.18
Basically, when DTNB [5,50-dithio-bis(2-nitrobenzoic
acid)] is added to a solution containing GSH, a reaction
listed in Figure 2 will occur and the yellow color product
5-thio-2-nitrobenzoic acid (TNB) of DTNB and GSH
addition will be formed and can quantitatively be used
to measure at 412nm using a spectrometer.19 The
DTNB, GSH, compounds 5–9 and their mixture with
GSH do not have absorption at 412nm at concentra-
tions below 10À3 mol/L.
2-Dimethylaminomethyl-5-(E)-pentylidenecyclopenta-
none hydrochloride (WB852) has been found to directly
bind with GSH measured using MS previously8 and
was used as a positive control in the current method.
We use the decrease of 412nm absorption to evaluate
the competition of WB852 or compounds 5–9 with
DTNB to bind GSH. WB852 reacts completely with an
equimolar amount of GSH in 60min and prevents
TNB formation. Based on the decrease of TNB forma-
tion (the intensity of 412nm absorbance), the addition
O
O
O
O
O
C
i
iii
ii
OC2H5
C
OC2H5
3
4
R1
R1
R1
O
O
O
vi
iv
N
HCl
R1
R1
R1
=
H
R1
R1
R1
5a R1
5b
6a
8a
8b
=
H
=
H
= CH3
6b R1
R1
= CH3
= CH3
vii
v
R1
R1
OH
R1
O
N
HCl
R2
R1
R1
9
=
H
R2
R2
= H
dimethylamino,
dimethylamino,
7a
7e
R1
=
4-morpholinyl , 1-piperidyl 1-pyrrolidinyl
,
~
~
7d
7h
1 = CH3
R
1-piperidyl
1-pyrrolidinyl
,
=
4-morpholinyl ,
Figure 1. The synthetic pathways of compounds 5–9.

J. Wang et al. / Bioorg. Med. Chem. 13 (2005) 1285–1291
NO2
1287
S
S
COOH
pH 7.4
+
S
NO2
GSH
GS
S
NO2
+
H+
+
NO2
COOH
TNB
COOH
COOH
DTNB
Figure 2. The interaction pathway of DTNB with GSH.
Table 1. The influence of WB852 and compounds 5–9 on TNB
formation due to DTNB and GSH reaction
2.3. Inhibition of GSTp activity
To test whether these compounds inhibit GSTp activity,
the cell lysate of HL-60 cells, which has been shown to
contain higher activity of GSTp,20 as a GSTp source
used to measure the inhibitory effect of these compounds
on its activity. The GSTp activity was measured based
on the classic method.21 By using WB852 as a represent
compound of these synthesized cyclopentanones, we cal-
culated the IC50 of its inhibition on GSTp activity and
found that was around 30lM. Thus, for the purpose
to compare the structure–activity relationship of these
compounds on GSTp inhibition, we focused on the
30lM concentration of each compounds. As shown in
Table 2, compounds 7 and 8 had stronger inhibitory ef-
fect on GSTp activity than WB852. However, compound
5a did not have inhibitory effect on GSTp activity and
compound 9 had a decreased inhibitory effect comparing
with WB852. Since the pattern of structure–activity rela-
tionship of these compounds on GSTp activity inhibi-
tion is similar to that of their interaction with GSH
(Table 1), it seems that the inhibitory effect of these com-
pounds on GSTp activity results from their interaction
with GSH, the substrate of GSTp, and inhibit GSTp
activity indirectly. In supporting this conclusion, GSTp
activity in cell lysate isolated from HL-60 cells after
Compound
Concentrations (lmol/L)
40
0
10
2
0
50
160
OD 412nma
WB852
5a
7a
7b
7c
0.0620.060 0.059 0.041 06.02 0
0.063 0.063 0.0620.061 0.0620.063
0.064 0.057 0.045 0.031 0.026 0.007
0.063 0.0520.047 0.038 0.030 0.015
0.0620.055 0.046 0.046 0.036 0.015
0.061 0.054 0.050 0.045 0.040 0.012
0.064 0.064 0.059 0.038 0.03206.02
0.060 0.057 0.053 0.051 0.047 0.025
0.063 0.060 0.0520.048 0.0420.033
7d
7e
7f
7g
7h
8a
8b
9
0.061 0.058 0.055 0.053 0.0520.032
0.060 0.054 0.046 0.035 0.031 0.015
0.063 0.060 0.054 0.045 0.038 0.032
0.063 0.063 0.0620.063 0.061 0.045
a The OD at 412nm was assayed in buffer containing 50 lmol/L GSH
and indicated concentrations of compounds for 30min following
addition of 500lmol/L DTNB.
of compounds 5–9 with GSH is estimated and listed in
Table 1.
All of the compounds with Mannich base hydrochlo-
rides decrease the absorbance at 412nm in a concentra-
tion-dependent manner. Compounds 5a, b without
2-alkylaminomethyl group adjacent to the a-CH2 on the
cyclopentanone ring, do not change the absorbance at
412nm. Compound 7a and 7b decrease the absorbance
at 412nm to 70.3% and 74.6% of that of control solution
at 20lmol/L, respectively, while WB852 only lowers the
GSH level to 95.4% of that of control solution at
20lmol/L (Table 1). Comparing with WB852, compound
7a at 50lmol/L has a stronger activity to decrease the
absorbance at 412nm. For compounds containing the
same alkylidene moiety, the dimethylaminomethyl
analogs are more active than structures containing mor-
pholino, pyrrolidino, or piperidino-methyl groups.
Comparing with compound 8a, compound 9 has a de-
creased ability of blocking TNB formation. The fact
that compounds 5a and 9 do not/weakly decrease the
absorbance at 412nm indicates that both 2-alkylamino-
methyl structure (Mannich base) and cyclopentanone
ring are critical functional groups to mediate GSH-
binding ability. Since compound 8 has similar ability
to decrease the absorbance at 412nm with that of
compounds 7 indicating that b-unsaturated bond is
not necessary for the GSH binding.
Table 2. The inhibitory effect of compounds 7–9 on GSTp activity
in vitro
Compounds
GSTp activity
(nmol/min/mg protein)
Inhibition (%)
Control
WB852
5a
88.4 6.0
35.3 3.0a
88.5 6.4
3.0 2.1a
2.8 1.1a
6.3 2.1a
0
55.8
0
7a
7b
96.2
96.8
92.9
89.6
92.5
97.6
83.4
77.6
98.1
98.7
25.6
7c
7d
a
9.2 2.6
7e
6.0 1.3a
2.1 1.7a
14.7 3.7a
19.8 1.3a
1.5 1.3a
1.1 0.6a
52.5 4.3b
7f
7g
7h
8a
8b
9
GSTp activity was assayed in 2h reaction buffer containing 30 lmol/L
tested compounds, 1mmol/L GSH, 1mmol/L CDNB and 100lg HL-
60 cell lysate. The unit of GSTp activity was defined as the amount of
enzyme which catalyze the formation of 1lmol of product per mg
protein per minute. The data show mean SD of triple samples.
a P < 0.005.
b P < 0.05 in t-test.

1288
J. Wang et al. / Bioorg. Med. Chem. 13 (2005) 1285–1291
treatment with any of these compounds was not inhib-
ited. However the intracellular GSH levels were de-
creased in HL-60 cells after treatment by compounds
showing interaction with GSH in vitro (data not
shown).
out GSTp expression, without severe toxicity to normal
cells at controlled doses.
In summary, both Mannich base hydrochlorides and
cyclopentanone ring are important for GSH binding,
GSTp inhibition and antitumor effect of our synthesized
compounds. The GSTp inhibitory effect of these com-
pounds result from their direct interaction with GSH,
the substrate of GSTp. The significant changes in antitu-
mor activity are apparent in altering phenyl groups to
methylphenyl groups by comparing compounds 7a–d
or 7e–h. Our data support the previous postulation of
that the activity of these compounds may result from di-
rect binding to GSH and other thiol proteins.8
2.4. Antitumor effect
The in vitro antitumor activity of these compounds was
conducted by using MTT assay in two human breast
cancer cell lines MCF-7 and MCF-7/Adr cells and one
immortalized human mammary epithelial cell line
184B5 and by using direct cell number counting in
human leukemia HL-60 cells (Table 3). All of the com-
pounds had antitumor effect except compounds 5a and
5b. Compound 9 had a decreased antitumor effect com-
paring with compound 8a. Since the pattern of cancer
cell growth inhibition (Table 3) is also similar to that
of GSH binding (Table 1), it suggests that the cytotoxic
effect of these compounds is mediated with direct bind-
ing to GSH or other thiol proteins that are critical for
cell growth. The stronger activity of compounds 7e–h
comparing with that of compounds 7a–d, suggest that
methyl group in the phenyl of b-chain will either in-
crease the cell membrane penetration ability or stability
in solution or medium. Correlated with the antitumor
activity, compounds 7e–h have strong GSH decreasing
ability than compounds 7a–c (data not shown). Since
the IC50s of compounds 7e–h in MCF-7 cells (without
expression of GSTp) is much lower than the IC50s in
MCF-7/Adr cells and 184B5 cells (both expressing
GSTp)22 suggesting that the growth inhibitory effect of
these compounds might not result from GSTp inhibi-
tion. Interestingly it seems that GSTp might play a neg-
ative role in cell growth inhibition by these compounds.
It is possible that GSTp catalyzes these compounds to
interact with GSH and the GSH-conjugated compounds
will have a decreased cell growth inhibitory effect. This
assumption needs to be testified in future by synthesiz-
ing GSH combined cyclopentanone. The stronger selec-
tivity of compounds 7h to breast cancer MCF-7 cells
comparing to mammary epithelial 184B5 cells, suggest-
ing it could be developed into a novel drug which will
selectively target tumor cells, at least in tumor cells with-
3. Experimental
GSH, DTNB, 1-chloro-2,4-dinitrobenzene(CDNB) and
3-[4,5-dimethylthiazol-2yl]-2,5-di-phenyltetrazolium
bromide (MTT) were obtained from J&K-ACROS
Chemical Co. Bovine Serum Albumin (BSA) was pur-
chased from Sigma (St. Louis, MO) and prepared in
pure water. IR Spectra were recorded on a Bruker IR-
1
27G spectrometer. The H NMR were measured on a
Bruker ARX-300 spectrometer with an internal stand-
ard of tetramethylsilane. Mass spectra were recorded
on a Shimadzu GCMS QP-1000 mass spectrometer.
4. General procedure for preparation of 2-diarylmethyl-
enecyclopentanones (5a,b)
Grignard reagents were prepared by addition of aryl
bromide (0.26mol) in dry ether (75mL) to magnesium
(0.27mol) in ether (25mL). The mixture was refluxing
for 20min and then a solution of ethyl 2,2-ethylenedi-
oxycyclopentanecarboxylate (0.11mol) in ether (50mL)
was added dropwisely with vigorous stirring. The mix-
ture was boiled and stirred for 60min, cooled and
decomposed with aqueous ammonium chloride
(156mL, 37% w/v). The precipitate was removed by fil-
tration, the ether layer was separated and the aqueous
layer extracted with ether. The combined ethereal solu-
tions were concentrated and methanol (62mL), water
(43mL) and concentrated hydrochloric acid (3.1mL)
were added in. The mixture was refluxed with stirring
for 5h. After cooling down, yellow precipitates were
collected and was recrystallized from ethanol to obtain
yellow needle products.
Table 3. The IC50s of compounds 5–9 on cell growth inhibition in vitro
Compounds
MCF-7a
MCF-7/Adra
184B5a
HL-60b
IC50 (lmol/L)
5a
5b
7a
7b
7c
7d
7e
7f
>50
>50
2.5
8.0
7.7
4.4
1.7
2.5
2.2
2.2
6.4
6.6
>20
>50
>50
5.8
>50
>50
2.0
>50
>50
13.3
13.0
Compound 5a: 2-diphenylmethylenecyclopentanone,
yield 39%, mp 115–116ꢁC.
15.5
14.0
8.0
3.5
4.29.6
1.5
3.0
3.5
3.0
5.5
12.7
3.4
0.4
0.9
1.3
3.2
Compound 5b: 2-[bis-(4-methylphenyl)methylene]cyclo-
pentanone, yield 32%, mp 142–143ꢁC.
3.1
3.8
7g
7h
8a
8b
9
3.5
3.5
4.1. General procedure for synthesis of 2-alkylamino-
methyl-5-diarylmethylenecyclopentanone hydrochlorides
(7a–h)
6.7
7.2N
>20
N
N
4.0
>20
a The cells were treated for 4days.
b The cells were treated for 1 day; N, not tested.
Compounds 7a–h were synthesized according to classi-
cal Mannich reaction.23 A mixture of 2-diarylmethylene-

J. Wang et al. / Bioorg. Med. Chem. 13 (2005) 1285–1291
1289
cyclopentanone (0.005mol), secondary amine hydro-
chloride (0.005mol) and paraformaldehyde (0.013mol)
in the presence of concentrated hydrochloric acid
(0.15mL) was refluxed in absolute ethanol (15mL) for
4h. The resulting solution was concentrated in vacuum.
For compounds 7a and 7e, the residues were treated
with dry acetone yielding crystalline material and the
products were recrystallized from acetone–ethanol.
However, for compounds 7b–d, 7f–h, the following puri-
fication method was employed. The residue was dis-
solved in water, the precipitate was removed and the
filtrate solution was adjust to alkaline with anhydrous
sodium carbonate and was extracted with ethyl ether.
The ethereal solution was dried over anhydrous sodium
sulfate. After filtration the solution was treated drop-
wisely with dry ethereal hydrochloric acid to give a
white precipitate and was then recrystallized from ace-
tone–ethanol.
(ppm): 1.64–1.71 (m, 1H), 2.38 (d, 6H, J = 3.0Hz),
2.79–3.33 (m, 11H), 3.37–3.39 (m, 1H), 6.96–7.16 (m,
8H, H-Ph), 12.57 (s, 1H, HCl); IR (KBr) r/cmÀ1: 2507
(tCH, CH3), 1704 (tC@O), 1641, 1586, 1561, 1508, 1466
(tC@C), 1181 (dCH), 915 (cC@C).
4.1.6. 2-Morpholinomethyl-5-[bis-(4-methylphenyl)meth-
ylene]cyclopentanone hydrochloride (7f). Mp: 168–
170ꢁC. Yield: 40%. EI-MS (m/z): 375 (M+, 13.10), 287
1
(9.03), 273 (18.87), 100 (100.00), 44 (1.32); H NMR
(CDCl3) d (ppm): 1.66–1.77 (m, 1H), 2.38 (d, 6H,
J = 3.0Hz), 2.74–2.81 (m, 2H), 2.93–2.97 (m, 4H),
3.16–3.47 (m, 4H), 3.93–3.96 (m, 2H), 4.21–4.39 (m,
2H), 6.96–7.16 (m, 8H, H-Ph), 13.09 (s, 1H, HCl); IR
(KBr) r/cmÀ1: 2 42 4tC(H, CH3), 1702( tC@O), 1591,
1508, 1446 (tC@C), 1176 (dCH), 819 (cC@C).
4.1.7. 2-Pipyridylmethyl-5-[bis-(4-methylphenyl)methyl-
ene]cyclopentanone hydrochloride (7g). Mp: 161–163ꢁC.
Yield: 36%. EI-MS (m/z): 373 (M+, 15.57), 287 (20.40),
4.1.1. 2-Dimethylaminomethyl-5-diphenylmethylenecyclo-
pentanone hydrochloride (7a). Mp: 155–156ꢁC. Yield:
12%. EI-MS (m/z): 305 (M+, 3.77), 259 (12.28), 58
(100.00), 44 (7.64), 36 (7.30); 1H NMR (CDCl3) d
(ppm): 1.68–1.74 (m, 1H), 2.78–3.33 (m, 11H), 3.35–
3.39 (m, 1H), 7.07–7.35 (m, 10H, H-Ph), 12.56 (s, 1H,
HCl); IR (KBr) r/cmÀ1: 2965 (t@CH), 2455 (tCH, CH3),
1704 (tC@O), 1590, 1469, 1442( tC@C), 1180 (dCH), 948,
758 (cC@C).
1
273 (37.10), 98 (100.00), 44 (1.50); H NMR (CDCl3)
d (ppm): 1.36–1.44 (m, 1H), 1.65–1.71 (m, 1H), 1.79–
1.85 (m, 2H), 2.18–2.33 (m, 2H), 2.38 (d, 6H,
J = 3.0Hz), 2.61–2.97 (m, 7H), 3.12 (m, 1H), 3.28 (m,
1H), 3.47–3.56 (m, 2H), 6.96–7.16 (m, 8H, H-Ph),
12.11 (s, 1H, HCl); IR (KBr) r/cmÀ1: 2502 (tCH
,
CH3), 1707 (tC@O), 1596, 1507, 1449 (tC@C), 1159
(dCH), 917, 820 (cC@C).
4.1.2. 2-Morpholinomethyl-5-diphenylmethylenecyclopen-
tanone hydrochloride (7b). Mp: 172–173ꢁC. Yield: 35%.
EI-MS (m/z): 347 (M+, 9.92), 259 (46.42), 100 (100.00),
44 (40.19), 36 (19.09); 1H NMR (CDCl3) d (ppm):
1.66–1.77 (m, 1H), 2.75–2.83 (m, 2H), 2.92–3.03 (m,
4H), 3.20–3.47 (m, 4H), 3.93–3.97 (m, 2H), 4.20–4.37
(m, 2H), 7.08–7.36 (m, 10H, H-Ph), 13.02(s, 1H,
HCl); IR (KBr) r/cmÀ1: 2 930 t(@CH), 1701 (tC@O),
1586, 1568, 1441, 1410 (tC@C), 1080 (dCH), 872( cC@C).
4.1.8. 2-Pyrrolidylmethyl-5-[bis-(4-methylphenyl)methyl-
ene]cyclopentanone hydrochloride (7h). Mp: 165–167ꢁC.
Yield: 42%. EI–MS (m/z): 359 (M+, 12.56), 287
1
(19.02), 273 (37.93), 84 (100.00), 44 (1.52); H NMR
(CDCl3) d (ppm): 1.66–1.77 (m, 1H), 2.05–2.09 (m,
2H), 2.18–2.25 (m, 2H), 2.38 (d, 6H, J = 3.0Hz), 2.75–
3.10 (m, 7H), 3.40 (m, 1H), 3.71–3.81 (m, 2H), 6.96–
7.16 (m, 8H, H-Ph), 12.47 (s, 1H, HCl); IR (KBr) r/
cmÀ1: 2579 (tCH, CH3), 1706 (tC@O), 1592, 1508, 1448
(tC@C), 1169 (dCH), 820 (cC@C).
4.1.3. 2-Pipyridylmethyl-5-diphenylmethylenecyclopenta-
none hydrochloride (7c). Mp: 168–170ꢁC. Yield: 15%.
EI-MS (m/z): 345 (M+, 9.20), 259 (100.00), 98 (98.23),
4.2. General procedure for synthesis of 2-alkylamino-
methyl-5-diarylmethylcyclopentanone hydrochlorides (8a,
b)
1
36 (30.77); H NMR (CDCl3) d (ppm): 1.36–1.44 (m,
1H), 1.65–1.71 (m, 1H), 1.79–1.85 (m, 2H), 2.20-2.37
(m, 2H), 2.60–2.96 (m, 7H), 3.16–3.28 (m, 2H), 3.47–
3.55 (m, 2H), 7.08–7.36 (m, 10H, H-Ph), 12.10 (s, 1H,
HCl); IR (KBr) r/cmÀ1: 2 942 t@(CH), 1705 (tC@O),
1587, 1569, 1431 (tC@C), 1165 (dCH), 937 (cC@C).
The Mannich bases 8a, b were prepared as follows. N,N-
dimethylmethylene-ammonium chloride (0.01mol), syn-
thesized based on a literature procedure,24 was added to
a solution of 2-diarylmethyl cyclopentanone (0.0025
mol) in acetonitrile (50mL) at 80–90ꢁC and the mixture
was heated under reflux. The reaction was monitored by
TLC using a solvent system of chloroform–methanol
(8:1). After 16h, the mixture was filtered and the solid
product was crystallized from chloroform–acetonitrile
to give 8a or 8b.
4.1.4. 2-Pyrrolidylmethyl-5-diphenylmethylenecyclopen-
tanone hydrochloride (7d). Mp: 164–165ꢁC. Yield: 28%.
EI-MS (m/z): 331 (M+, 11.94), 259 (87.90), 84 (100.00),
1
36 (39.27); H NMR (CDCl3) d (ppm): 1.68–1.73 (m,
1H), 2.05–2.08 (m, 2H), 2.19–2.26 (m, 2H), 2.76–3.12
(m, 7H), 3.37–3.43 (m, 1H), 3.71–3.83 (m, 2H), 7.08–
7.36 (m, 10H, H-Ph), 12.48 (s, 1H, HCl); IR (KBr) r/
cmÀ1: 2 92 1t@(CH), 1706 (tC@O), 1588, 1569, 1491,
1443 (tC@C), 1175 (dCH), 770 (cC@C).
4.2.1. 2-Benzhydryl-5-((dimethylamino)methyl)cyclopen-
tanone hydrochloride (8a). Mp: 175–176ꢁC. Yield: 52%.
EI-MS (m/z): 307 (M+, 0.46), 262 (0.29), 58 (100.00),
44 (1.72), 36 (2.28); 1H NMR (DMSO-d6) d (ppm):
1.41 (m, 1H), 1.60 (m, 1H), 1.99 (m, 1H), 2.27 (m,
1H), 2.64 (m, 1H), 2.72 (s, 6H), 3.05 (m, 1H), 3.33 (m,
2H), 4.25 (d, 1H, J = 8.4Hz), 7.13–7.35 (m, 10H),
4.1.5. 2-Dimethylaminomethyl-5-[bis-(4-methylphenyl)-
methylene]cyclopentanone hydrochloride (7e). Mp: 144–
146ꢁC. Yield: 32%. EI-MS (m/z): 333 (M+, 12.32), 287
1
(16.80), 273 (31.69), 58 (100.00); H NMR (CDCl3) d
10.15 (s, 1H, HCl); IR (KBr) r/cmÀ1: 2481 (tCH
,

1290
J. Wang et al. / Bioorg. Med. Chem. 13 (2005) 1285–1291
CH3), 1739 (tC@O), 1494, 1479, 1450, 1421 (tC@C), 1157
(dCH), 746 (cC@C).
and 10% heat-inactivated fetal bovine serum in a humid-
ified atmosphere of 95% air and 5% CO2 at 37ꢁC. MCF-
7 and MCF-7/Adr cells (obtained from Dr. T.-C.
Chou)25 were cultured in D-MEM medium supple-
mented with 100U/mL pencillin, 100lg/mL streptomy-
cin, 1mmol/L L-glutamine and 5% heat-inactivated
fetal bovine serum in a humidified atmosphere of 95%
air and 5% CO2 at 37ꢁC. 184B5 immortalized mammary
epithelial cells (obtained from ATCC) were cultured in
MEGM (mammary epithelial growth medium) supple-
mented with 1ng/mL cholera toxin in a humidified
atmosphere of 95% air and 5% CO2 at 37ꢁC.
4.2.2. 2-((Dimethylamino)methyl)-5-(di-p-tolylmethyl)-
cyclopentanone hydrochloride (8b). Mp: 172–173ꢁC.
Yield: 49%. EI-MS (m/z): 335 (M+, 1.23), 290 (0.60),
1
58 (100.00), 44 (2.31); H NMR (DMSO-d6) d (ppm):
1.42 (m, 1H), 1.57 (m, 1H), 2.03 (m, 1H), 2.21(s, 6H),
2.24 (m, 1H), 2.58 (m, 1H), 2.70 (s, 6H), 3.05 (m, 1H),
3.23 (m, 2H), 4.20 (d, 1H, J = 8.4Hz), 7.04–7.19 (m,
8H), 10.05 (s, 1H, HCl); IR (KBr) r/cmÀ1: 2922 (t@CH),
2467 (tCH, CH3), 1734 (tC@O), 1511, 1468, 1412( tC@C),
1162( dCH), 767 (cC@C).
4.5. Assay of GSTp activity
4.2.3. Synthesis of 2-benzhydryl-5-((dimethylamino)-
methyl)cyclopentanol (9). 2-Benzhydryl-5-((dimethyl-
HL-60 cells in logarithmic growth (3 · 106) were washed
twice with PBS, resuspended in 300lL of 100mmol/L
potassium phosphate buffer, pH6.8, sonicated for 10s
at 4ꢁC, centrifuged at 14000rpm for 30min at 4ꢁC
and the supernatant lysate was used for enzyme assay.
GSTp activity was determined spectrophotometrically
at 25ꢁC, using CDNB and GSH as substrates according
to the reported method.21 The linear increase in absorp-
tion at 340nm, caused by conjugation of GSH (1mM)
with CDNB (1mM) in HL-60 cell lysate with or without
presence of compounds 5–9 (30lM), was measured. A
complete assay mixture without enzyme was used as a
control. The nonenzymatic reaction was corrected by
blanking the spectrophotometer with the control cuvette
before each reading of the sample cuvette. An extinction
coefficient of 9.6mMÀ1 cmÀ1 was used to calculate
GSTp activity expressed as nanomoles per minute per
milligram of protein.
amino)methyl)cyclopentanone
hydrochloride
(8a,
0.005mol) was dissolved in methanol (30mL) and stir-
red. NaBH4 (1.0g, 0.026mol) was added portionwisely
over a period of 0.5h and stirred at room temperature
for another 6h. Solvent was removed under reduced
pressure, 10ml of water was then added and the precip-
itates were filtered and recrystallized from acetone–ether
to yield 9 as a white powder (76%), mp: 139–141ꢁC.
Yield: 71%. EI-MS (m/z): 309 (M+, 1.10), 281 (0.18),
167 (5.56), 142 (21.50), 58 (100.00), 44 (2.51); 1H
NMR (DMSO-d6) d (ppm): 1.13 (m, 2H), 1.36 (m,
2H), 1.73 (m, 1H), 2.01 (m, 2H), 2.11 (s, 6H), 2.65 (m,
1H), 3.54 (m, 1H), 4.22 (d, 2H, J = 3.8Hz), 4.53 (m,
1H), 7.10 (m, 2H), 7.25 (m, 4H), 7.36 (m, 4H), 10.18
(s, 1H, HCl); IR (KBr) r/cmÀ1: 2956 (t@CH), 2776
(tCH, CH3), 1567, 1493, 1449 (tC@C), 1170 (dCH), 758
(cC@C).
4.3. GSH-binding assay
4.5.1. Assay of cell growth inhibition. HL-60 cells were
seeded at 2 · 105 cells/mL and cultured in the above
noted medium with or without the indicated concentra-
tions of test compounds for 24h. Cell numbers in treated
with or without compounds were determined by hemo-
cytometer and growth inhibition was calculated to un-
treated cells (%).20 The growth inhibition of these
compounds on MCF-7, MCF-7/Adr and 184B5 cells
were measured by MTT assay.26 Briefly, cells (2 · 103)
were plated in each well of a 96-well plate and were al-
lowed to adhere and spread for 24h. The various con-
centrations of compounds 5–9 were then added and
cultured for 4days at 37ꢁC. 50lL of 2mg/mL MTT
solution was added per well and the cultures were con-
tinued for an additional 4h. The medium was removed
by aspiration. The cells were dissolved in 200lL DMSO
and absorbance at 570nm was measured in the 96-well
plate. Growth inhibition was determined as compared
to untreated cells (%).
GSH-binding reaction was carried out in sodium phos-
phate buffer (PBS, 0.1mol/L, pH7.4, containing
0.1mmol/L EDTA) based on a reported method.18
Briefly, 50lmol/L GSH in PBS was incubated with
cyclopentanone derivatives (0, 10, 20, 40, 50 and
160lmol/L) for 30min at 37ꢁC, the unreacted GSH in
the solution was determined by adding an aliquot of
0.3mL to the sample 1.0cm quartz cuvette containing
3.0mL 500lmol/L DTNB solution in the same buffer
and 0.3mL same PBS to the reference 1.0cm quartz cuv-
ette containing the same DTNB solution. The unreacted
GSH reacts with DTNB to form TNB. The absorbance
of TNB was measured with a UV-9100 spectrophoto-
meter at 412nm. The concentrations of the unreacted
GSH can be calculated according to the absorbance of
TNB (e = 13700MÀ1 cmÀ1 at 412nm). The equation of
sample concentration calculation was done as follows:
C1 ¼ ½ðTotal volume=sample volumeÞ
ꢀ ðAbs412Þꢁ=13700
Acknowledgements
4.4. Cell culture
This work was supported by The National Natural Sci-
ence Foundation of China and The National Natural
Science Foundation of Liaoning Province of China.
Technique help of Yi Sha, Wen Li determining the
HL-60 cells (obtained from ATCC) were cultured in
RPMI-1640 medium supplemented with 100U/mL pen-
cillin, 100lg/mL streptomycin, 1mmol/L L-glutamine
1
300MHz H NMR spectra is appreciated.

J. Wang et al. / Bioorg. Med. Chem. 13 (2005) 1285–1291
1291
13. Zhou, L.; Jing, Y.; Styblo, M.; Chen, Z.; Waxman, S.
Blood 2004.
References and notes
14. Tew, K. D.; Dutta, S.; Schultz, M. Adv. Drug Delivery
Rev. 1997, 26, 91–104.
15. Lindtead, R. P.; Meade, E. M. J. Chem. Soc. 1934, 935–
946.
16. Sharp, J. T.; Findlay, R. H.; Thorogood, P. B. J. Chem.
Soc., Perkin Trans. 1 1975, 102–113.
1. Kupchan, S. M.; Eakin, M. A.; Thomas, A. M. Tumor
Inhibitors. J. Med. Chem. 1971, 14, 1147–1152.
2. Lee, K. H.; Huang, E. S.; Piantadosi, C.; Pagano, J. S.;
Geissman, T. A. Cancer Res. 1971, 31, 1649–1654.
3. Lee, K. H.; Kim, S. H.; Furukawa, H.; Piantadosi, C. J.
Med. Chem. 1975, 18, 59–63.
17. Ghera, E.; Shoua, S. J. Org. Chem. 1972, 37, 1292–1297.
18. Mutus, B.; Wagner, J. D.; Talpas, C. J.; Dimmock, J. R.;
Phillips, O. A., et al. Anal. Biochem. 1989, 177, 237–243.
19. Ellman, G. L. Arch. Biochem. Biophys. 1958, 74, 443–
450.
20. Jing, Y.; Dai, J.; Chalmers-Redman, R. M.; Tatton, W.
G.; Waxman, S. Blood 1999, 94, 2102–2111.
21. Habig, W. H.; Pabst, M. J.; Jakoby, W. B. J. Biol. Chem.
1974, 249, 7130–7139.
4. Lee, K. H.; Meck, R.; Piantadosi, C.; Huang, E. S. J. Med.
Chem. 1973, 16, 299–301.
5. Chen, H. T.; Jing, Y. K.; Ji, Z. Z.; Zhang, B. F. Yao. Xue.
Xue. Bao. 1991, 26, 183–192.
6. Chen, H.; Ji, Z.; Wong, L. K.; Siuda, J. F.; Narayanan, V.
L. Pharm. Res. 1996, 13, 1482–1487.
7. Chen, H.; Ji, Z.; Wong, L. K.; Siuda, J. F.; Narayanan, V.
L. Drug Des. Discovery 1996, 14, 43–52.
8. Chen, H.; Ji, Z.; Wong, L. K.; Siuda, J. F.; Narayanan, V.
L. Bioorg. Med. Chem. 1994, 2, 1091–1097.
9. Wu, G.; Fang, Y. Z.; Yang, S.; Lupton, J. R.; Turner, N.
D. J. Nutr. 2004, 134, 489–492.
22. Lin, X.; Nelson, W. G. Cancer Res. 2003, 63, 498–504.
23. Dimmock, J. R.; Nyathi, C. B.; Smith, P. J. J. Pharm. Sci.
1978, 67, 1543–1546.
24. Bohme, H.; Hartke, K. Chem. Ber. 1960, 93, 1305–1309.
25. Chou, T. C.; Zhang, X. G.; Balog, A.; Su, D. S.; Meng, D.,
et al. Proc. Natl. Acad. Sci. U.S.A. 1998, 95, 9642–
9647.
10. Hamilton, D.; Batist, G. Curr. Oncol. Rep. 2004, 6, 116–
122.
11. Shen, H.; Kauvar, L.; Tew, K. D. Oncol. Res. 1997, 9,
295–302.
26. Jing, Y.; Waxman, S.; Mira-y-Lopez, R. Cancer Res. 1997,
57, 1668–1672.
12. Townsend, D. M.; Tew, K. D. Oncogene 2003, 22, 7369–
7375.