Design, synthesis, and activity of caffeoyl pyrrolidine derivatives as potential gelatinase inhibitors

Article abstract of DOI:10.1016/j.bmc.2004.07.025

A group of caffeoyl pyrrolidine derivatives has been developed to act as potential gelatinase inhibitors. The synthesis and biological evaluation of caffonyl pyrrolidine derivatives as MMPs inhibitors are reported in this paper. Inhibiting activities of synthesized compounds on gelatinase (MMP-2 and -9) were tested by using succinylated gelatin as substrate. Structure-activity relationship results from these tested compounds demonstrated that longer and more flexible side chain linked to the pyrrolidine ring at C₄ produced higher activity at gelatinase. Furthermore, aromatic heterocycle and sulfamide in the same position could enhance the activities. Compounds with free phenol hydroxyl group showed higher activity compared to methylated derivatives (or counterparts), which confirms the importance of phenol hydroxyl functionality in the interaction with gelatinase. The anti-metastasis model of mice bearing H₂₂ tumor cell was used to evaluate their in vivo inhibiting activities. All tested compounds were orally administered at a dose of 50 or 100 mg/kg, 6 days/week for two weeks. The test results demonstrated that most of these inhibitors showed significant anti-cancer activities (inhibitory rate > 35%) and were devoid of toxic effects. Compound 29 showed the highest inhibitory rate at 69.25%, indicating that it might be a promising lead compound.

Full text of DOI:10.1016/j.bmc.2004.07.025

Bioorganic & Medicinal Chemistry 12 (2004) 5171–5180
Design, synthesis, and activity of caffeoyl pyrrolidine derivatives
as potential gelatinase inhibitors
Ya-Lin Li and Wen-Fang Xu*
Department of Medicinal Chemistry, School of Pharmacy, Shandong University, Jinan, Shandong 250012, PR China
Received 8 April 2004; revised 11 July 2004; accepted 11 July 2004
Available online 7 August 2004
Abstract—The synthesis and biological evaluation of caffonyl pyrrolidine derivatives as MMPs inhibitors are reported in this paper.
Inhibiting activities of synthesized compounds on gelatinase (MMP-2 and -9) were tested by using succinylated gelatin as substrate.
Structure–activity relationship results from these tested compounds demonstrated that longer and more flexible side chain linked to
the pyrrolidine ring at C4 produced higher activity at gelatinase. Furthermore, aromatic heterocycle and sulfamide in the same posi-
tion could enhance the activities. Compounds with free phenol hydroxyl group showed higher activity compared to methylated
derivatives (or counterparts), which confirms the importance of phenol hydroxyl functionality in the interaction with gelatinase.
The anti-metastasis model of mice bearing H22 tumor cell was used to evaluate their in vivo inhibiting activities. All tested com-
pounds were orally administered at a dose of 50 or 100mg/kg, 6days/week for two weeks. The test results demonstrated that most
of these inhibitors showed significant anti-cancer activities (inhibitory rate > 35%) and were devoid of toxic effects. Compound 29
showed the highest inhibitory rate at 69.25%, indicating that it might be a promising lead compound.
Ó 2004 Elsevier Ltd. All rights reserved.
1. Introduction
The matrix metalloproteinases (MMPs) are a family of
zinc-dependent protienases involved in the degradation
of the extracellular matrix. The MMPs have been impli-
cated in the processes of tumor growth, invasion, and
metastasis, frequently overexpressed by malignant tum-
ors and have been associated with an aggressive malig-
nant phenotype and poor prognosis in patients with
cancer.1–3 Up to now, the mammal MMP gene family
consists of at least 24 structurally related members,
among which, gelatinase (MMP-2 and -9) are proved
to be high correlation with cancer.
The interaction of MMP inhibitors with MMPs is illus-
trated in Figure 1 by the complex of CGS27023A with
MMP-3.4 The figure shows the p-(methoxyphenyl)sulfo-
nyl extend into S01 pocket of the enzyme, pyridine is in S20
pocket, and isopropyl is in S1 domain. There are two
hydrogen bond between the inhibitor with the backbone
Figure 1. Mode of binding of CGS 27023A with MMP-3.
nyl O of Ala165, respectively. The S01 pocket of gelatinase
of the enzyme, which are the O of sulfonylamide with
is deeper than that of MMP-3 and can accommodate
NH of Leu164 and NH of hydroxamic acid with carbo-
larger group. In our study, we designed compounds with
caffonyl group to extend into S01 pocket.
Keywords: Caffonyl pyrrolidine derivatives; MMPs inhibitors; IC50
*
.
L-Hydroxyproline is known as one of specific amine acid
of collagens, which are the substrate of MMPs. So the
Corresponding author. Tel./fax: +86-531-8382264; e-mail: xuwenfang@
sdu.edu.cn
0968-0896/$ - see front matter Ó 2004 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmc.2004.07.025

5172
Y.-L. Li, W.-F. Xu / Bioorg. Med. Chem. 12 (2004) 5171–5180
Figure 2. Comparison between designed compound with existing inhibitors. C, H, N, O, S are shown as black, green, blue, red, and yellow,
respectively.
derivatives of hydroxyproline may specifically recognize
MMPs and then combine with the active site. The caffic
acid is proved to inhibit MMP-2 and -9,5 so we linked
caffic acid with hydroxyproline derivatives as to find
compounds with activity inhibiting MMPs.
azide, hydrogenation over 5% Pd–C/CaCO3, acylation
and ester exchange upon treatment with hydroxylamine
as illustrated in Scheme 1.
Considered that the caffic acid, which contained free
phenol hydroxyl groups could inhibit the activity of
MMP-2 and -9, we prepared several compounds with
polyphenols as Scheme 2.
The 3-D structure of one of aimed compounds, com-
pound 30, was simulated by Chem3D Ultra 7.0 program
and showed in Figure 2. From Figure 2, we can deduce
that either the caffonyl group or the p-(methylphenyl)-
sulfonyl group can extend into the S01 group, which
may enhance the flexibility of the compound to combine
with the active site.
3. Results and discussion
3.1. Structure–activity relationship in vitro
Compounds were evaluated as inhibitors of gelatinases.
The results of their inhibitory activities (IC50) were re-
ported in Table 1.
2. Chemistry
The target compounds were synthesized efficiently fol-
lowing the procedures shown in Schemes 1 and 2.
3.1.1. Effect of the size of the side chains linked to the
pyrrolidine ring at C4. It is reported that MMPs have
two hydrophobic domains, which are called S01 pocket
and S02 pocket besides their catalytic activity center as
in Figure 3.5 The side chain R1 is designed to be located
in either of these two pockets. As seen in Table 1, to
Briefly, all compounds were derived from 4-L-hydroxy-
proline, and obtained through a reaction sequence
including methylation, esterification, condensation,
mesylation, SN2 reaction upon treatment with sodium

Y.-L. Li, W.-F. Xu / Bioorg. Med. Chem. 12 (2004) 5171–5180
5173
O
O
O
OMe
O
HO
HO
MeO
OH
Cl
d, e
MeO
MeO
b, c
N
MeO
OMs
O
H
N
O
H
N
HCl
OH
OH
a
OMe
f, g
HO
HO
O
OMe
O
H
N
O
MeO
MeO
O
N
h
i
MeO
MeO
N
N3
NH
R
Scheme 1. Reagents: (a) MeOH, HCl; (b) Me2SO4, NaOH; (c) SOCl2, C6H6; (d) Py, Et3N; (e) MsCl, Et3N, CH2Cl2; (f) NaN3, DMF; (g) 5% Pd–C,
H2, EtOH; (h) Et3N, CH2Cl2, R–Cl (R = carbonyl or sulfonyl group); (i) NH2OK, MeOH.
O
O
O
OMe
O
O
HO
HO
AcO
AcO
OH
Cl
d, e
AcO
AcO
b, c
N
OR
O
H
N
O
f
H
N
HCl
O
OMe
OH
a
OMe
HO
HO
N
HO
HO
OR
Scheme 2. Reagents: (a) MeOH, HCl; (b) Ac2O, H2SO4; (c) SOCl2, C6H6; (d) Py, Et3N; (e) Et3N, CH2Cl2, R–Cl (R = carbonyl or sulfonyl group);
(f) MeOH, THF, HCl.
some extent, the longer the side chain of R1 was, the
more strongly the compounds inhibited the enzyme.
figuration to match the 3-D structure (bioactive
conformation maybe better) of the enzyme.
Comparing 23, 24, 25, or 27, 29, 31, 32, we could con-
firm that the length of side chains of R1 was positively
relative with the inhibitory activities with an exception
of compound 12. This could be due to its bulky side
chain in the interaction with the enzymesÕ hydrophobic
domains.
3.1.3. Effect of the heteroaryl group of the side chain
linked to the pyrrolidine ring at C4. High activities of 19
and 33 indicated that the heteroaromyl group might be
better than aryl group.
3.1.4. Effect of the free phenol hydroxyl groups. Compar-
ing 9, 14, or 10, 13, we could draw a conclusion that free
phenol hydroxyl group do favor to the inhibitory activity.
Effect of the sulfonyl or acryl groups in the side chain
linked to the pyrrolidine ring at C4. The activity of 20
was better than 17, which confirmed that sulfonyl amide
is better than acryl, because sulfonyl groups was easier
to form a hydrogen bond with the active domain of en-
zyme.
3.1.5. Effect of symmetrical structure. The side chain of
32 is p-methoxyphenyl-2-propenoyl amide, which was
similar to caffonyl group, and so formed a butterfly-
like structure, therefore 32 had potent inhibitory abili-
ties.
3.1.2. Effect of the flexibility of the side chain linked to the
pyrrolidine ring at C4. Compounds 18 and 28 showed
higher activities, which might be favored from the flexi-
bility of the hexanyl group, which could modify the con-
3.2. Anti-metastasis activity against H22 tumor cell in vivo
Compounds were evaluated in the anti-metastasis model
of mice bearing H22 tumor cell. The results of their

5174
Y.-L. Li, W.-F. Xu / Bioorg. Med. Chem. 12 (2004) 5171–5180
Table 1. SAR of the compounds 6–33
O
R2
O
R3O
R3O
N
*
4
R1
No.
R1
R2
R3
4*
IC50 (nmol)
MR
8.9441
6
OH
OH
MeO
MeO
Me
Ac
S
657.9 34.6
ND*
7
S
9.9431
10.280
11.279
13.102
14.101
15.650
11.864
9.3528
9.7197
9.2345
10.586
11.978
11.959
10.496
13.007
11.239
12.634
13.098
13.993
8.849
8
OSO2CH3
OSO2CH3
MeO
MeO
Me
Ac
S
198.9 11.2
234.5 21.3
182.3 26.2
160.0 10.7
731.4 64.0
104.9 15.3
168.1 14.9
224.9 46.6
ND*
9
S
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
28
27
29
30
31
32
33
OSO2C6H4CH3-p
OSO2C6H4CH3-p
OCOC6H2(OCOCH3)3-30,40,50
OSO2C6H4CH3-p
OSO2CH3
MeO
MeO
Me
Ac
S
S
MeO
MeO
Ac
H
S
S
MeO
MeO
H
S
N3
NH2
Me
Me
Me
Me
Me
Me
Me
Me
Me
Me
Me
Me
Me
Me
Me
Me
Me
Me
Me
R
R
R
R
R
R
R
R
R
R
R
R
R
R
R
R
R
R
R
MeO
MeO
NHCOCH2CH3
NHCO(CH2)4CH3
NHCOC5H4N-30
NHSO2CH3
479.4 30.4
130.7 15.9
91.2 8.1
155.5 12.7
178.0 10.0
103.1 27.4
110.6 9.1
93.4 13.3
65.7 7.7
9.7 1.6
MeO
MeO
MeO
MeO
NHSO2C6H4CH3-p
NHCO(CH2)2COOH
NHCOCH2C6H5
NHCO(CH2)2C6H5
NHCOCH@CHC6H3(OMe)-40(E)
OH
MeO
MeO
MeO
MeO
NHOH
NHOH
NHOH
NHOH
NHOH
NHOH
NHOH
NHOH
NHCO(CH2)4CH3
NHCOCH2CH3
NHCOCH2C6H5
NHSO2C6H4CH3-p
NHCO(CH2)2C6H5
NHCOCH@CHC6H3(OMe)-40(E)
NHCOC5H4N-30
7.8 0.9
11.883
10.491
12.539
12.912
13.002
13.898
11.864
15.7 1.5
11.5 0.5
9.7 0.6
9.4 0.4
6.7 0.17
9.0 0.6
ND*: Not determined; MR was determined by Chem3D Ultra 7.0 program.
S2
S1
S2'
158
Ala
H
N
O
Leu160
Ala161
O
H
N
163
NH
N
H
N
H
N
H
159
His162
O
O
O
O
157
His162
151
H
O
H
O
Me
H
N
O
S3'
P1
P3'
P2
N
N
P1'
S3
P3
188
189
N
N
H
N
H
NH2
O
P2'
H
O
O
197
O
Phe/Tyr164
H
iPr
216
Zn2+
N
218
194
Tyr219
193 Ph
222
Leu/Val214
S2
O
Tyr219
catalytic activity
center
NH
220
O
S2' solvent exposed
S1' MMP special
pocket long and narrow
Figure 3. Schematic depiction of the bond between MMPs and substrate.
activities (inhibitory rate) following oral administration
in mice were reported in Table 2.
The tested compounds can significantly inhibit the lung
metastasis following administration in mice (P < 0.01).

Y.-L. Li, W.-F. Xu / Bioorg. Med. Chem. 12 (2004) 5171–5180
5175
Table 2. Anti-tumor activities in vivo
Compound no.
Mice of
survived (n)
Body weight (g)
Lung weight (g)
Metastasized nodes
on lung surface (n)
Inhibitory
rate (%)
lgP
Control
10
9(10)*
10
9(10)*
10
10
23.50 3.42
16.82 3.16
17.88 3.27
20.63 2.42
23.20 1.16
21.65 2.31
22.22 3.26
22.48 2.50
23.17 4.25
22.68 3.41
21.95 3.42
0.168 0.021
0.148 0.062
0.143 0.041
0.153 0.068
0.163 0.014
0.143 0.012
0.166 0.029
0.161 0.023
0.164 0.022
0.178 0.073
0.163 0.019
46.5 2.12
22.0 3.78
19.5 1.96
17.5 0.93
28.8 9.34
24.5 8.32
14.3 0.57
21.7 4.72
20.0 1.17
30.2 12.19
20.6 3.25
12 (100mg/kg/d)
14 (100mg/kg/d)
19 (100mg/kg/d)
20 (100mg/kg/d)
22 (100mg/kg/d)
29 (50mg/kg/d)
30 (50mg/kg/d)
31 (50mg/kg/d)
32 (50mg/kg/d)
33 (50mg/kg/d)
52.69
58.06
62.37
38.07
47.31
69.25
53.33
56.99
35.05
55.70
1.2687
ꢀ0.474
0.6211
2.939
ꢀ0.2714
1.1552
1.236
10
10
10
10
9(10)*
1.5725
1.4258
ꢀ0.1262
*The animal number in brackets is the original number; lgP was determined by Chem3D Ultra 7.0 program.
All animals administrated the tested compounds except
12 did not significantly lose weights (P < 0.05), which
meant low toxicity of the compounds.
4. Conclusion
The pyrrolidine derivatives we designed could conform-
ationally match the active site of MMP-2 and -9, so they
showed high anti-MMP-2 and -9 activity in vitro. Fur-
thermore they could be metabolized into active seg-
ments, such as derived hydroxyl prolidines and caffic
acid, which might also be expected to have anti-neoplas-
tic activities, so they also showed high anti-tumor activ-
ity in vivo. Among the inhibitors the best one was
compound 29, which had inhibitory rate as high as
69.25%, showing that it might be a promising lead com-
pound.
3.2.1. Effect of zinc-binding group. Hydroxamic acid and
carbomethoxy are both zinc-binding group and can che-
late with zinc at catalytic activity center of the enzyme.
Although hydroxamates showed more than 10 times
stronger inhibitory activities against gelatinases than
carboxylates in vitro, they did not do the same in vivo.6
Maybe because hydroxamates are not stable under the
metabolic system, but carbomethoxy can be metabo-
lized to carboxylic acid, which is also a zinc-binding
group.
3.2.2. Effect of the lgP of inhibitors. The side chain of 33,
29, 31, and 32 was nicotimide, phenylethanoylamide,
phenylpropanoylamide, and p-methoxyphenyl-2-prop-
enoylamide, respectively, and we found that phenyletha-
noylamide was the best side chain in activity.
Compound 29 was not too active in vitro, but it showed
impressive in vivo activity, which might benefit from its
fitting lgP. Moreover no toxic effects were observed in
compound 29 and this might be a promising lead com-
pound. Compound 29 also suggested that we could
modify 31 and enhance its oral bioavailability to gain
better inhibitor.
5. Experimental section
All reactions except those in aqueous media were carried
out by standard techniques for the exclusion of mois-
ture. All reactions were monitored by thin-layer chro-
matography on 0.25-mm silica gel plates (60GF-254)
and visualized with UV light, or iodine vapor. 1H
NMR spectra were determined on a Brucker AM300
spectrometer using TMS as an internal standard. ESI
MS were determined on an API 4000 spectrometer.
5.1. (2S,4R) Methyl 4-hydroxy-2-pyrrolidinecarboxylate
hydroxychloride (1)
Compound 19 showed higher activity and lower toxic
effect, attributing to its well lgP and heteroaromyl.
The title compound was prepared as described by Jordis
in (1S,4S)-2-thia-5-azabicyclo[2.2.1]heptane.7
3.2.3. Effect of the flexibility of the side chain linked to
the pyrrolidine ring at C4. Compound 32 was very
active in vitro, but was not active in vivo. This
result might be due to its high lgP value, which
makes it difficult to be absorbed. Its molecular rigidity
and high polarity also make this compound difficult
to transfer across the biomembrane to reach its active
site.
5.2. (E)-3-(3,4-Dimethoxyphenyl)-2-propenoic acid (2)
Caffic acid (18g, 0.1mol) was dissolved in cool 90mL
NaOH solution (4mol/L), in which process the inner
temperature was required under 20°C. After Me2SO4
(20mL) was added dropwise and stirred for 20min,
50mL NaOH solution (4mol/L) and 20mL Me2SO4
were dropped into the reaction system constantane-
ously. The resulting mixture was slowly heated to
90°C in 1h and maintained at this temperature for 1h,
then the solution was refluxed for 2h, after 50mL
NaOH solution (4mol/L) was added, another 2h of re-
fluxing was applied. The pH of the resulting solution
was regulated to 2 by concentrated HCl. Two to three
3.2.4. Effect of the sulfonyl in the side chains linked to the
pyrrolidine ring at C4. The high activity of 14 might be
benefited from sulfonyl group, but it also had intolera-
ble toxic effect, resulting in the rush decrease of body
weight of the mice.

5176
Y.-L. Li, W.-F. Xu / Bioorg. Med. Chem. 12 (2004) 5171–5180
hours later, mass brown solid were attained by filtrating,
and washed to neutrality with water. The gross was re-
crystallized in EtOH and H2O to get 16.4g (yield
78.8%) light yellow crystal, which did not appear blue
in Fe(SCN)3/FeCl3.
(1H), 7.606 (d, J = 15.6Hz, 1H). ESI MS: m/z (rel inten-
sity) 389.89.
5.8. (2S,4R) Methyl 1-[(E)-3-(3,4-dimethoxyphenyl)-2-
propenoyl]-4-[(methylsulfonyl)oxy]-2-pyrrolidinecarboxyl-
ate (8)
5.3. (E)-3-(3,4-Diacetyloxyphenyl)-2-propenoic acid (3)
In nitrogen atmosphere, compound 6 (3.35g, 10mmol)
was taken in CH2Cl2 (5mL) and Et3N (4.5mL) was
added at 0°C. Methanesulfonyl chloride was added
dropwise and the resulting mixture was stirred for 4h
at room temperature. The following mixture was parti-
tioned between water and EtOAc. The organic layer
was washed with saturated NaHCO3 solution, H2O,
and brine in turn, dried over MgSO4, filtered, and evap-
orated to give pale yellow oil, which was chromato-
graphed over flash silica with petro ether–acetone (4:1
to 1:2) to provide 3.4g white crystal, yield 82.3%, mp
Caffic acid (18g, 0.1mol) was dissolved in 30mL anhy-
dride acetate and 10 d H2SO4, and stirred for 10min
at 60°C. The cold solution was poured into icy water,
stirred vigorously, and filtrated to get white solid,
washed with water until neutral. The gross was recrys-
tallized in AcOEt and hexane to get 16.4g (yield
83.0%) white crystal, which did not appear blue in
Fe(SCN)3/FeCl3.
5.4. (E)-3-(3,4-Dimethoxyphenyl)-2-propenoic chloride
(4)
1
139.1–140.0°C. H NMR (CDCl3, 300MHz): d 2.258–
2.348 (m, 2H), 2.562–2.645 (m, 1H), 3.067 (s, 3H),
3.770 (s, 3H), 3.899–3.916 (6H), 4.046–4.152 (m, 2H),
4.714 (t, J = 8.1Hz, 1H), 6.490 (d, J = 15.3Hz, 1H),
6.850 (d, J = 8.1Hz, 1H), 7.090 (s, 1H), 7.103 (d,
J = 8.1Hz, 1H), 7.674 (d, J = 15.3Hz, 1H). ESI MS:
m/z (rel intensity) 412.3.
Compound 2 (20.8g, 10mmol) was dissolved in 40mL
SOCl2 and 320mL benzene, and refluxed for 3h. The
resulting solution was rotary evaporated to get pale yel-
low crystal.
5.5. (E)-3-(3,4-Diacetyloxyphenyl)-2-propenoic chloride
(5)
5.9. (2S,4R) Methyl 1-(E)-3-(3,4-diacetyloxyphenyl)-2-
propenoyl-4-[(methylsulfonyl)oxy]-2-pyrrolidinecarboxyl-
ate (9)
Compound 3 was converted to the title compound as
described for compound 4 to get white solid.
Compound 7 was converted to the title compound as de-
scribed for compound 8 to get white solid, yield 81.0%,
mp 77.8–80.0°C. 1H NMR (CDCl3, 300MHz): d 2.296–
2.370 (m, 8H), 3.072 (s, 3H), 3.781 (s, 3H), 4.024–4.033
(m, 3H), 4.717 (t, J = 8.1Hz, 1H), 6.496 (d, J = 15.6Hz,
1H), 7.215 (d, J = 8.4Hz, 1H), 7.380 (d, J = 1.8Hz, 1H),
7.397 (d, J = 8.4Hz, 1H), 7.687 (d, J = 15.6Hz, 1H),
J = 15.6Hz, 1H). ESI MS: m/z (rel intensity) 426.85
[MꢀCH3CO], 384.75 [Mꢀ(CH3CO)2].
5.6. (2S,4R) Methyl 1-[(E)-3-(3,4-dimethoxyphenyl)-2-
propenoyl]-4-hydroxy-2-pyrrolidinecarboxylate (6)
Compound 1 (1.85g, 11mmol) was suspended in 20mL
Py and 3mL Et3N. After stirred for 20min at room tem-
perature, the resulting mixture was filtrated. The filtrate
was cooled to ꢀ5°C, and at this temperature 10mL
CH2Cl2 with 2.82g 4 (10mmol) was added. After stirred
for 3.5h, the resulting mixture was filtrated to remove
white precipitation, and the filtrate was rotary evapo-
rated. The resulting crude oil was chromatographed
over flash silica with petro ether–EtOAc (4:1 to 1:4) to
provide 2.4g of pale yellow solid, yield 71.6%, mp
62.5–63.5°C. 1H NMR (CDCl3, 300MHz): d 2.027–
2.134 (m, 1H), 2.305–2.335 (m, 1H), 3.050 (s, 1H),
3.706 (s, 1H), 3.732 (s, 3H), 3.849 (s, 3H), 3.873 (s,
3H), 3.896–3.932 (m, 1H), 4.589 (s, 1H), 4.696 (t,
J = 7.8Hz, 1H), 6.487 (d, J = 15.6Hz, 1H), 6.783 (d,
J = 8.4Hz, 1H), 7.018 (s, 1H), 7.032 (d, J = 8.4Hz,
1H), 7.582 (d, J = 15.6Hz, 1H).
5.10. (2S,4R) Methyl 1-[(E)-3-(3,4-dimethoxyphenyl)-2-
propenoyl]-4-[(4-methylphenyl)sulfonyl]oxy-2-pyrrolidine-
carboxylate (10)
Compound 6 (3.35g, 10mmol) was taken in CH2Cl2
(5mL) and Et3N (4.5mL) was added at 0°C. Tolylsul-
fonyl chloride (3.8g, 20mmol) was added dropwise
and the resulting mixture was stirred overnight at room
temperature. The following mixture was diluted with
CH2Cl2 and washed with 10% HCl, saturated NaHCO3
solution, and H2O, dried over MgSO4, filtered, and
evaporated to give pale yellow oil, which was chromato-
graphed over flash silica with petro ether–acetone (8:1 to
1:1) to provide 3.9g white crystal, yield 79.8%, mp 60.2–
62.1°C. 1H NMR (CDCl3, 300MHz): d 2.039–2.208 (m,
1H), 2.412 (s, 3H), 2.444–2.498 (m, 1H), 3.873 (s, 3H),
3.905–3.925 (6H), 3.997–4.014 (m, 3H), 4.651 (t,
J = 7.8Hz, 1H), 6.380 (d, J = 15.3Hz, 1H), 6.858 (d,
J = 8.4Hz, 1H), 7.004 (d, J = 1.5Hz, 1H), 7.086 (d,
J = 8.4Hz, 1H), 7.639 (d, J = 15.3Hz, 1H), 7.348 (d,
J = 7.8Hz, 2H), 7.787 (d, J = 7.8Hz, 2H). ESI MS:
m/z (rel intensity) 488.2.
5.7. (2S,4R) Methyl 1-(E)-3-(3,4-diacetyloxyphenyl)-2-
propenoyl-4-hydroxy-2-pyrrolidinecarboxylate (7)
The title compound was prepared as described for com-
1
pound 6 to get 3.0g white solid, mp 138.7–140.2°C. H
NMR (CDCl3, 300MHz): d 2.071–2.080 (m, 1H), 2.129
(dd, J = 8.1, 4.8Hz, 1H), 2.293–2.299 (6H), 3.689 (s,
1H), 3.904 (dd, J = 4.5, 10.8Hz, 2H), 4.604 (1H), 4.684
(t, J = 8.1Hz, 1H), 6.604 (d, J = 15.6Hz, 1H), 7.181
(d, J = 7.8Hz, 1H), 7.326 (d, J = 1.5Hz, 1H), 7.358

Y.-L. Li, W.-F. Xu / Bioorg. Med. Chem. 12 (2004) 5171–5180
5177
5.11. (2S,4R) Methyl 1-(E)-3-(3,4-diacetyloxyphenyl)-2-
propenoyl-4-[(4-methylphenyl)sulfonyl]oxy-2-pyrrolidine-
carboxylate (11)
NMR (CDCl3, 300MHz): d 2.234—2.279 (m, 2H),
2.497 (s, 3H) ESI MS: m/z (rel intensity) 384.52.
5.15. (2S,4S) 1-[(E)-3-(3,4-Dimethoxyphenyl)-2-prop-
enoyl]-2-carbomethoxy-4-azidopyrrolidine (15)
Compound 7 was converted to the title compound as de-
scribed for compound 10 to get white solid, yield 78.9%,
mp 117.1–120.5°C (decomposed). 1H NMR (CDCl3,
300MHz): d 2.181 (m, 2H), 2.229 (s, 6H), 2.317 (s,
3H), 3.777 (s, 3H), 3.835–3.999 (m, 3H), 4.647 (t,
J = 8.1Hz, 1H), 6.438 (d, J = 15.6Hz, 1H), 7.207 (d,
J = 4.5Hz, 1H), 7.333 (d, J = 4.5Hz, 1H), 7.353
(d, J = 7.8Hz, 2H), 7.622 (d, J = 15.6Hz, 1H), 7.784
(d, J = 7.8Hz, 2H). ESI MS: m/z (rel intensity) 502.82
[MꢀCH3CO], 460.505 [Mꢀ(CH3CO)2].
In nitrogen atmosphere, compound 7 (4.13g, 10mmol)
was taken in 15mL dry DMF in the presence of ground
NaN3 (695mg, 10.7mmol). The resulting mixture was
heated to 55°C for 10h and then partitioned between
water and EtOAc. The organic layer was then washed
with brine, dried over MgSO4, filtered, and evaporated.
The resulting crude oil was chromatographed over flash
silica with hexane–EtOAc (5:1 to 3:1) to provide 2.87g
of pale yellow oil, which solidified upon standing, yield
79.0%, mp 179.0–180.5°C. IR: mCH 2953, mN 2104, mC@O
5.12. (2S,4R) Methyl 1-[(E)-3-(3,4-dimethoxyphenyl)-2-
propenoyl]-4-(3,4,5-triacetyloxybenzoyl)oxy-2-pyrrolidine-
carboxylate (12)
3
3
1749, 1650, mC@C 1597, 1513.
5.16. (2S,4S) 4-Amine-1-[(E)-3-(3,4-dimethoxyphenyl)-2-
propenoyl]-2-pyrrolidinecarboxylate (16)
3,4,5-Trimethoxybenzoic acid was converted to 3,4,5-
trimethoxybenzoyl chloride as described for compound
4, then the title compound was prepared as described
for compound 10 to get white solid, yield 80.7%, mp
In hydrogen atmosphere, Compound 15 (11g, 30mmol)
was dissolved in alcohol, and 5% Pd/CaCO3 (3g) was
added. The resulting solution was stirred for 10h under
760mmHg pressure, during which the hydrogen was in-
put intermittently. The resulting mixture was filtered
and the filtrate was evaporated to give brown oil, which
was chromatographed over flash silica with CHCl2–
MeOH (4:1 to 1:4) to provide aqua oil, which solidified
upon standing, yield 60.8%. ESI MS: m/z (rel intensity)
333.4. IR: mNH 3345.94, 3257.87, mCH 2936, mC@O 1726,
1
81.7–84.0°C. H NMR (CDCl3, 300MHz): d 2.070 (s,
3H), 2.246–2.310 (12H), 2.404–2.474 (m, 2H), 3.777 (s,
3H), 3.756–3.925 (m, 1H), 3.963–4.200 (t, J = 7.8Hz,
1H), 6.590 (d, J = 15.3Hz, 1H), 7.208 (d, J = 8.1Hz,
1H), 7.315 (s, 1H), 7.370 (s, 2H), 7.390 (d, J = 8.1Hz,
1H), 7.665 (d, J = 15.3Hz, 1H). ESI MS: m/z (rel inten-
sity) 626.91 [MꢀCH3CO].
2
3
1641, mC@C 1590, 1514.
5.13. (2S,4R) Methyl 1-[(E)-3-(3,4-dihydroxyphenyl)-2-
propenoyl]-4-[(4-methylphenyl)sulfonyl]oxy-2-pyrrolidine-
carboxylate (13)
5.17. (2S,4S) Methyl 1-[(E)-3-(3,4-dimethoxyphenyl)-2-
propenoyl]-4-propanoylamid-2-pyrrolidinecarboxylate
(17)
Compound 11 (0.9g, 1.66mmol) was dissolved in 8mL
MeOH and 8mL THF, and then 3mL concentrated
HCl was added. The resulting solution was stirred for
20min at 60°C and was diluted with H2O after cooled
to room temperature. The following mixture was parti-
tioned between water and EtOAc and the organic layer
was washed with brine, dried over MgSO4, filtered, and
evaporated to give white puff, which was chromato-
graphed over flash silica with petro ether–EtOAc–acetic
acid (30:10:1 to 10:30:1) to provide 0.46g white crystal,
which appeared blue in FeCl3 and Fe(SCN)3, yield
60.2%, mp 142.0–144.5°C. 1H NMR: 2.070 (s, 3H),
2.246–2.310 (12H), 2.404–2.474 (m, 2H), 3.777 (s, 3H),
3.756–3.925 (m, 1H), 3.963–4.200 (t, J = 7.8Hz, 1H),
6.590 (d, J = 15.3Hz, 1H), 7.208 (d, J = 8.1Hz, 1H),
7.315 (s, 1H), 7.370 (s, 2H), 7.390 (d, J = 8.1Hz, 1H),
7.665 (d, J = 15.3Hz, 1H). ESI MS: m/z (rel intensity)
462.29.
Compound 16 (334mg, 1mmol) was dissolved in 2mL
dry CHCl2 and 0.5mL Et3N, and 2mL CHCl2 with
0.16mL propanoyl chloride was added dropwise. The
resulting solution was stirred for 3h. The following mix-
ture was partitioned between water and CHCl2 and the
organic layer was washed with 1% HCl, 5% NaCO3, and
water, in turn dried over MgSO4, filtered, and evapo-
rated to give yellow puff, which was chromatographed
over flash silica with petro ether–EtOAc (4:1 to 1:4) to
provide 291mg white crystal, yield 74.6%, mp 132.0–
1
133.7°C. H NMR: 1.144 (t, J = 7.5Hz, 3H), 1.942 (s,
1H), 1.989 (s, 1H), 2.216 (q, J = 7.5Hz, 2H), 2.487 (m,
1H), 3.817 (s, 3H), 3.905 (s, 3H), 3.913 (s, 3H), 4.605
(d, J = 9.9Hz, 1H), 4.806 (s, 1H), 6.518 (d,
J = 15.3Hz, 1H), 6.853 (d, J = 8.1Hz, 1H), 6.984 (s,
1H), 7.028 (s, 1H), 7.097 (d, J = 8.1Hz, 1H), 7.668 (d,
J = 15.3Hz, 1H)ESI MS: m/z (rel intensity) 389.8.
5.14. (2S,4R) Methyl 1-[(E)-3-(3,4-dihydroxyphenyl)-2-
propenoyl]-4-[(methylsulfonyl)oxy]-2-pyrrolidinecarboxyl-
ate (14)
5.18. (2S,4S) Methyl 1-[(E)-3-(3,4-dimethoxyphenyl)-2-
propenoyl]-4-hexanoylamid-2-pyrrolidinecarboxylate (18)
Compound 9 was converted to the title compound as de-
scribed for compound 11 to get white solid. The mobile
phase for VLC employed petro ether–EtOAc–acetic acid
(30:10:1–10:30:1), yield 54.8%, mp 179.0–180.5°C. H
Compound 16 was converted to the title compound as
described for compound 17, yield 62.7%. 1H NMR:
0.870 (t, J = 6.6Hz, 3H), 1.258–1.334 (m, 4H), 1.614
(t, J = 7.1Hz, 2H), 2.169 (t, J = 7.1Hz, 2H), 2.487 (m,
1

5178
Y.-L. Li, W.-F. Xu / Bioorg. Med. Chem. 12 (2004) 5171–5180
2H), 3.818 (s, 3H), 3.905–3.920 (6H), 4.591–4.812 (m,
3H), 6.542 (d, J = 15.3Hz, 1H), 6.840–7.117 (m, 3H),
7.680 (d, J = 15.3Hz, 1H). ESI MS: m/z (rel intensity)
341.75.
over flash silica with petro ether–EtOAc–AcOH
(40:10:1 to 10:40:1) to provide 308mg pale yellow crys-
tal, yield 74.6%, mp 155.3–157.0°C. H NMR: 1.952–
2.051 (m, 2H), 2.455 (t, J = 9.6Hz, 2H), 2.677 (t,
J = 9.6Hz, 2H), 3.807 (s, 3H), 3.896–3.991 (6H), 4.592
(dd, J = 1.8, 9.6Hz, 1H), 4.762 (d, J = 3.3Hz, 1H),
6.524 (d, J = 15.6Hz, 1H), 6.846 (d, J = 8.1Hz, 1H),
7.039 (d, J = 1.5Hz, 1H), 7.103 (dd, J = 1.5, 8.1Hz),
7.234 (d, J = 8.7Hz, 1H), 7.662 (d, J = 15.6Hz, 1H).
ESI MS: m/z (rel intensity) 433.1.
1
5.19. (2S,4S) Methyl 1-[(E)-3-(3,4-dimethoxyphenyl)-2-
propenoyl]-4-[(3-pyridinylcarbonyl)amid]-2-pyrrolidine-
carboxylate (19)
Compound 16 was converted to the title compound as
described for compound 17 to get pale yellow solid, yield
68.9%. H NMR: 2.041–2.155 (m, 2H), 2.535–2.636 (m,
5.23. (2S,4S) 1-[(E)-3-(3,4-Dimethoxyphenyl)-2-prop-
enoyl]-4-[(2-phenylethanoyl)amide]-2-pyrrolidinecarboxyl-
ate (23)
1
2H), 3.856 (s, 3H), 3.901 (s, 3H), 3.909 (s, 3H), 4.035 (m,
1H), 4.680 (d, J = 9.6Hz, 1H), 5.061 (t, J = 3.9Hz, 1H),
6.526 (d, J = 15.3Hz, 1H), 6.831–7.261 (m, 3H), 7.415
(dd, J = 4.8, 7.8Hz, 1H), 7.679 (d, J = 15.3Hz, 1H),
8.202 (d, J = 4.8 Hz, 1H), 8.302 (d, J = 7.8Hz, 1H),
8.737 (d, J = 4.2Hz, 1H), 9.122 (s, 1H). ESI MS: m/z
(rel intensity) 438.5.
Compound 16 was converted to the title compound as
described for compound 17 to get white crystal, yield
1
68.9%, mp 103.0–104.0°C. H NMR: 2.415–2.465 (m,
2H), 3.481 (m, 2H), 3.700 (s, 3H), 3.900 (s, 3H), 3.914
(s, 3H), 3.853–3.931 (m, 2H), 4.549 (dd, J = 2.4,
7.5Hz, 1H), 4.720 (d, J = 3Hz, 1H), 6.461 (d,
J = 15.6Hz, 1H), 6.950 (d, J = 7.5Hz, 1H), 6.811–
7.091 (3H), 7.197–7.313 (5H), 7.639 (d, J = 15.6Hz,
1H). ESI MS: m/z (rel intensity) 452.5.
5.20. (2S,4S) Methyl 1-[(E)-3-(3,4-dimethoxyphenyl)-2-
propenoyl]-4-[(methylsulfonyl)amide]-2-pyrrolidinecarb-
oxylate (20)
Compound 16 was converted to the title compound as
described for compound 17 to get white crystal, yield
68.9%, mp 80.3–84.7°C. 1H NMR: 2.028–2.095 (m,
1H), 2.477–2.575 (m, 1H), 2.979 (s, 3H), 3.776 (s, 3H),
3.884–3.926 (6H), 3.988 (t, J = 5.4Hz, 1H), 4.024–
4.238 (m, 2H), 4.584 (dd, J = 3.3, 15.6Hz, 1H), 6.036
(d, J = 9.6Hz), 6.510 (d, J = 15.6Hz, 1H), 6.837 (d,
J = 7.8Hz, 1H), 7.081 (d, J = 1.5Hz, 1H), 7.097 (d,
J = 7.8Hz, 1H), 7.650 (d, J = 15.6Hz, 1H). ESI MS:
m/z (rel intensity) 411.90.
5.24. (2S,4S) Methyl 1-[(E)-3-(3,4-dimethoxyphenyl)-2-
propenoyl]-4-[(2-phenylpropanoyl)amide]-2-pyrrolidine-
carboxylate (24)
The starting amine 16 was converted to the title com-
pound as described for compound 17 to get white
crystal, yield 59.0%, mp 80.3–81.5°C. 1H NMR:
1.7852–1.8726 (m, 2H), 2.5115–2.5564 (m, 2H), 2.9756
(t, J = 7.56Hz, 2H), 3.6467 (d, J = 10.44Hz, 1H),
3.7808 (s, 3H), 3.7928–3.8414 (m, 2H), 3.9497 (s, 3H),
3.9816 (s, 3H), 4.5796 (d, J = 9.84Hz, 1H), 4.9794 (m,
1H), 6.4656 (d, J = 15.36Hz, 1H), 6.8892 (d,
J = 8.50Hz, 1H), 7.0021 (d, J = 8.50Hz, 1H), 7.0476
(s, 1H), 7.1023–7.1606 (m, 2H), 7.1828–7.1252 (m,
2H), 7.2519–7.2767 (m, 1H), 7.6829 (d, J = 15.36Hz,
1H), 8.2493 (d, J = 7.86Hz, 1H), 8.9080 (1H), 10.7505
(1H). ESI MS: m/z (rel intensity) 467.05.
5.21. (2S,4S) Methyl 1-[(E)-3-(3,4-dimethoxyphenyl)-2-
propenoyl]-4-[(4-methylphenyl)sulfonyl]amide-2-pyrrol-
idinecarboxylate (21)
Compound 16 was converted to the title compound as
described for compound 17 to get white crystal, yield
1
68.9%, mp 103.0–104.0°C. H NMR: 1.760–1.826 (m,
5.25. (2S,4S) Methyl 1-[(E)-3-(3,4-dimethoxyphenyl)-2-
propenoyl]-4-[(E)-3-(4-methoxy)-2-propenoyl]amide-2-
pyrrolidinecarboxylate (25)
2H), 2.235–2.334 (m, 2H), 2.363 (s, 3H), 3.779 (s, 3H),
3.899–3.907 (6H), 4.098–4.145 (m, 1H), 4.472 (dd,
J = 3.0, 9.6Hz, 1H), 6.215 (d, J = 9.6Hz, 1H), 6.348
(d, J = 15.3Hz, 1H), 6.828–7.082 (m, 3H), 7.284 (d,
J = 8.1Hz, 1H), 7.616 (d, J = 15.3Hz, 1H), 7.750 (d,
J = 8.1Hz, 1H). ESI MS: m/z (rel intensity) 487.55.
Compound 16 was converted to the title compound as
described for compound 17 to get pale yellow crystal,
1
yield 67.9%, mp 101.9–103.4°C. H NMR: 2.534–2.584
(m, 2H), 3.800 (s, 3H), 3.843 (s, 3H), 3.937 (s, 3H),
3.962 (s, 3H), 3.970–3.988 (2H), 4.656 (d, J = 9.6Hz,
1H), 4.917 (1H), 5.942 (d, J = 12.6Hz, 1H), 6.153 (d,
J = 15.6Hz, 1H), 6.832–7.504 (7H), 7.590 (d,
J = 12.6Hz, 1H), 7.700 (d, J = 15.6Hz, 1H). ESI MS:
m/z (rel intensity) 495.45.
5.22. (2S,4S) 4-[1-[(E)-3(3,4-Dimethoxyphenyl)-2-prop-
enoyl]-5-(carbomethoxy)tetrahydro-1H-3-pyrroli-
dine]amide-4-oxobutanoic acid (22)
Compound 16 (334mg, 1mmol) was dissolved in 2mL
dry CH2Cl2 and 0.5mL Et3N, and 2mL CHCl2 with
110mg succinic anhydride was added dropwise. The
resulting solution was stirred for 3h. The following mix-
ture was partitioned between water and CH2Cl2 and the
organic layer was washed with 1% HCl, 5% NaCO3, and
water, in turn dried over MgSO4, filtered, and evapo-
rated to give yellow puff, which was chromatographed
5.26. (2S,4S) 1-[(E)-3-(3,4-Dimethoxyphenyl)-2-prop-
enoyl]-4-hydroxy-2-N-hydroxyamid-pyrrolidine (26)
Compound 6 (670mg, 2mmol) was dissolved in 7mL
dry methanol and 1.5mL methanol with NH2OK (pre-
pared by Fieser and Fieser. Vol. 1, p 478). The resulting

Y.-L. Li, W.-F. Xu / Bioorg. Med. Chem. 12 (2004) 5171–5180
5179
solution was stirred at rt for 24h, then 1.5g silica gel was
added and evaporated to give pale yellow powder, which
was chromatographed over flash silica with CH2Cl2–
MeOH (50:1 to 1:50) to provide 392mg (58.3%) pale
yellow crystal, which appeared red in FeCl3, mp
117.3–118.8°C. IR: mOH,NH 3350–2800 (w), mC@O 1634,
mC@C 1588, 1513, bOH 1141. ESI MS: m/z (rel intensity)
335.1.
J = 15.3Hz, 1H), 7.497 (d, J = 8.1Hz, 2H), 7.796 (d,
J = 8.1Hz, 2H) ESI MS: m/z (rel intensity) 488.35.
5.31. (2S,4S) 1-[(E)-3-(3,4-Dimethoxyphenyl)-2-prop-
enoyl]-2-N-hydroxyamid-4-(3-phenylpropanoyl)amidepyr-
rolidine (31)
Compound 22 was converted to the title compound as
described for compound 26 to get white crystal, yield
52.7%, mp 137.9–140.5°C. 1H NMR: 2.8188 (t,
J = 7.56Hz, 2H), 2.2935–2.3745 (m, 4H), 3.7205–
3.7601 (m, 2H), 3.7871 (s, 3H), 3.8221 (s, 3H) 4.1767
(t, J = 7.62Hz, 1H), 4.3482 (dd, J = 7.14, 19.68Hz,
1H), 6.8537 (d, J = 15.36Hz, 1H), 6.9660 (d,
J = 8.16Hz, 1H), 7.1672 (s, 1H), 7.1982–7.2754 (m,
3H), 7.3577–7.4011 (m, 2H), 7.1478 (d, J = 8.16Hz,
1H), 7.3577 (m, 1H), 7.3887 (d, J = 15.36Hz, 1H). ESI
MS: m/z (rel intensity) 466.99.
5.27. (2S,4S) 1-[(E)-3-(3,4-Dimethoxyphenyl)-2-prop-
enoyl]-2-N-hydroxyamid-4-propanoylamidepyrrolidine
(27)
Compound 17 was converted to the title compound as
described for compound 26 to get white crystal, yield
50.7%, mp 123.0–125.4°C. 1H NMR: 0.996 (t,
J = 7.2Hz, 3H), 1.742–1.763 (m, 1H), 2.073 (q,
J = 7.2Hz, 2H), 2.357–2.383 (m, 1H), 3.776 (s, 3H),
3.807 (s, 3H), 4.240–4.289 (m, 1H), 4.354 (t,
J = 6.6Hz, 1H), 6.850 (d, J = 15.3Hz, 1H), 6.946 (d,
J = 8.4Hz, 1H), 7.184 (dd, J = 1.5, 8.4Hz, 1H), 7.333
(d, J = 1.5Hz, 1H), 7.384 (d, J = 15.3Hz, 1H). ESI
MS: m/z (rel intensity) 390.1.
5.32. (2S,4S) 1-[(E)-3-(3,4-Dimethoxyphenyl)-2-prop-
enoyl]-2-N-hydroxyamid-4-[(E)-3-(4-methoxy)-2-prop-
enoyl]amide]pyrrolidine (32)
Compound 25 was converted to the title compound as
described for compound 26 to get pale yellow crystal,
yield 56.6%, mp 154.4–155.7°C. IR: mOH,NH 3203–
2836, mC@O 1647, mC@C 16.2, 1512ESI MS: m/z (rel inten-
sity) 496.44.
5.28. (2S,4S) 1-[(E)-3-(3,4-Dimethoxyphenyl)-2-prop-
enoyl]-4-hexanoylamid-2-N-hydroxyamidepyrrolidine (28)
Compound 18 was converted to the title compound as
described for compound 26 to get white crystal, yield
50.8%, mp 105.0–107.5°C. 1H NMR: 0.841 (t,
J = 6.9Hz, 3H), 1.212–1.298 (m, 4H), 1.444–1.540 (m,
2H), 2.046 (t, J = 7.8Hz, 2H), 2.484–2.496 (m, 2H),
3.287 (2H), 3.776–3.805 (6H), 4.027–4.061 (m, 1H),
4.233–4.282 (t, J = 6.9Hz, 1H), 6.847 (d, J = 15.3Hz,
1H), 6.958 (d, J = 8.4Hz, 1H), 7.186 (dd, J = 1.2,
8.4Hz, 1H), 7.329 (d, J = 1.2Hz, 1H), 7.409 (d,
J = 15.3Hz, 1H). ESI MS: m/z (rel intensity) 432.3.
5.33. (2S,4S) 1-[(E)-3-(3,4-Dimethoxyphenyl)-2-prop-
enoyl]-2-N-hydroxyamid-4-(3-pyridinylcarbonyl)amide-
pyrrolidine (33)
Compound 19 was converted to the title compound
as described for compound 26 to get pale yellow crystal,
yield 56.6%, mp 154.4–155.7°C. 1H NMR: 1.9867–
2.0840 (m, 2H), 3.8539 (s, 3H), 3.9519 (s, 3H), 6.251
(d, J = 15.36Hz, 1H), 6.9551 (d, J = 8.28Hz, 1H),
7.1174 (d, J = 8.28Hz, 1H), 7.1763 (s, 1H), 7.4110 (d,
J = 15.36Hz, 1H), 7.5421 (1H), 8.7012–8.7295 (m, 1H).
ESI MS: m/z (rel intensity) 439.94.
5.29. (2S,4S) 1-[(E)-3-(3,4-Dimethoxyphenyl)-2-prop-
enoyl]-2-N-hydroxyamid-4-(2-phenylethanoyl)amidepyr-
rolidine (29)
5.34. Assays of MMP inhibition and anti-tumor activity
Compound 23 was converted to the title compound as
described for compound 26 to get white crystal, yield
50.7%, mp 124.9–125.7°C. H NMR: 1.187–1.778 (m,
1H), 2.381–2.409 (m, 1H), 3.401 (s, 2H), 3.704–3.742
(m, 1H), 3.775 (s, 3H), 3.801 (s, 3H), 4.009–4.342 (m,
3H), 6.845 (d, J = 15.3Hz, 1H), 6.934 (d, J = 8.4Hz,
1H), 7.172–7.328 (m, 7H), 7.382 (d, J = 15.3Hz, 1H).
ESI MS: m/z (rel intensity) 528.3.
5.34.1. MMP inhibition. Gelatinase (MMP-2, -9) and
TNBS were purchased from Sigma, and the substance
was synthesized as described by Vijaykumar et al.8 The
gelatinase, substance, and inhibitor were dissolved in so-
dium borate (pH8.5, 50mmol/L) and incubated for
30min at 37°C, then 0.03%TNBS was added and incu-
bated for another 20min, the resulting solution was de-
tected under 450nm wavelength to gain absorption.
1
5.30. (2S,4S) 1-[(E)-3-(3,4-Dimethoxyphenyl)-2-prop-
enoyl]-2-N-hydroxyamid-4-[(4-methylphenyl)sulfonyl]-
amidepyrrolidine (30)
5.35. Anti-tumor activities
Male Kunming mice were obtained from Lab Animal
Center, Shandong University.
Compound 21 was converted to the title compound as
described for compound 26 to get white crystal, yield
50.7%, mp 124.9–125.7°C. H NMR: 2.052–2.266 (m,
2H), 3.6122 (m, 1H), 3.784 (s, 3H), 3.813 (s, 6H),
3.875–3.927 (m, 2H), 4.312 (t, J = 5.8Hz, 1H), 6.553
(d, J = 15.3Hz, 1H), 6.965 (d, J = 7.8Hz, 1H), 7.189
(d, J = 7.8Hz, 1H), 7.301 (s, 1H), 7.351 (d,
1
The mice bearing H22 tumor cells ascites were obtained
from Shandong Medicine Academy of Sciences.
Materials: Aqueous suspensions of inhibitors were
prepared by homogenizing the compounds in 0.5%

5180
Y.-L. Li, W.-F. Xu / Bioorg. Med. Chem. 12 (2004) 5171–5180
carboxymethyl cellulose Na, 0.9% benzyl alcohol and
0.4% Tween 80 in saline. A control solution of the vehi-
cle without inhibitors was also prepared.
References and notes
1. Tumor cell Interactions with the Extracellular Matrix
During Invasion and Metastasis. Stetler-Stevenson, W.
G.; Aznavoorian, S.; Liotta, L. A. Annu. Rev. Cell Biol.
1993, 9, 541–573.
2. A matrix Metallo-proteinase Expressed on the Surface of
Invasive Tumor Cells. Sato, H.; Takino, T.; Okada, Y.
Nature 1994, 370, 61–65.
3. Changing Views of the Role of Matrix Metalloproteinases
in Metastasis. Chambers, A. F.; Matrisian, L. J. Natl.
Cancer Inst. 1997, 89, 1260–1270.
4. Bioactive conformation of a potent stromelysin inhibitor
determined by X-nucleus filtered and multidimensional
NMR spectroscopy. Gonnella, N. C.; Li, Y. C.; Zhang, X.;
Paris, C. G. Bioorg. Med. Chem. 1997, 5, 2193–2201.
5. Feng, X.; Danqing, S.; Yongsu, Z. In Discussion of the
Inhibition of Angiogenesis and its Mechanism In Progress
of Anti-neoplasm-drugs and Chemotherapy; Science Publish-
ing Company: Beijing, 2001; Vol. 10, p 243.
Bouin solution: Saturated trinitrophenol:saturated for-
maldehyde:glacial acetic acid = 15:5:1.
Experimental design: Mice bearing H22 tumor ascites
were injected via the caudal vein and randomly divided
into 11 groups. The animals of the control group were
treated with the same volume of vehicle, the other
groups were given the inhibitors by oral administration,
respectively, at a dose of 100mg/kg/d (carboxylates) or
50mg/kg/d (hydroxamates) 6days/week for two weeks.
The mice were weighed and sacrificed for autopsy imme-
diately. The lungs with tumor nodes were removed and
were weighed, then placed in bouin solution. One day
later, the metastasized nodes on the surface of lungs
were counted.
6. Grams, F.; Reinemer, P.; Powers, J. C., et al. Eur. J.
Biochem. 1995, 288, 830–841.
7. Jordis, U.; Sauter, F.; Siddiqi, S. M. (1S,4S)-2-Thia-5-
azabicyclo[2.2.1]heptane. Indian J. Chem. Sec. B 28, April
1989; 294–296.
Acknowledgement
8. Vijaykumar, M. B.; Bonnie, J. S.; Richard, R. R.;
Paul, J. K.; Howard, G. W.; Mahesh, M.; Jon, R. C.;
Srinivasa, K. R. Matrix Biol. 2000, 19, 26.
This work was supported by the National Nature Sci-
ence Foundation of China (Grant No. 20272033).