Specific inhibitor of puromycin-sensitive aminopeptidase with a homophthalimide skeleton: Identification of the target molecule and a structure-activity relationship study

Article abstract of DOI:10.1016/S0968-0896(00)00231-5

2-(2,6-Diethylphenyl)-1,2,3,4-tetrahydroisoquinoline-1,3-dione (2: PIQ-22) was found to be a potent and specific inhibitor of puromycin-sensitive aminopeptidase (PSA). Lineweaver-Burk plot analysis showed that PSA is inhibited by PIQ-22 in a non-competitive manner. Structure-activity relationship studies indicated that tautomerism of the imidobenzoylketone group in the cyclic imide moiety of the PIQ-22 skeleton is important for the inhibitory activity.

Full text of DOI:10.1016/S0968-0896(00)00231-5

Bioorganic & Medicinal Chemistry 9 (2001) 121±131
Speci®c Inhibitor of Puromycin-Sensitive Aminopeptidase with a
Homophthalimide Skeleton: Identi®cation of the Target Molecule
and a Structure±Activity Relationship Study
Masato Komoda,a Hiroki Kakuta,a Hiroyasu Takahashi,a Yasuyuki Fujimoto,a
Shizuo Kadoya,a Fuminori Katob and Yuichi Hashimotoa,*
aInstitute of Molecular and Cellular Biosciences, University of Tokyo, 1-1-1 Yayoi, Bunkyo-ku, Tokyo 113-0032, Japan
bR&D Group of Pharmaceuticals, Central Research Institute, Ishihara Sangyo Kaisha Ltd., 2-3-1 Nishishibukawa,
Kusatsu, Shiga 525-0025, Japan
Received 7 July 2000; accepted 11 August 2000
AbstractÐ2-(2,6-Diethylphenyl)-1,2,3,4-tetrahydroisoquinoline-1,3-dione (2: PIQ-22) was found to be a potent and speci®c inhi-
bitor of puromycin-sensitive aminopeptidase (PSA). Lineweaver±Burk plot analysis showed that PSA is inhibited by PIQ-22 in a
non-competitive manner. Structure±activity relationship studies indicated that tautomerism of the imidobenzoylketone group in the
cyclic imide moiety of the PIQ-22 skeleton is important for the inhibitory activity. # 2000 Elsevier Science Ltd. All rights reserved.
Introduction
proteolytic lability, rapid biliary excretion, short duration
of action, and so on, all of which are general features of
peptide-type drugs. Therefore, development of non-
peptide speci®c aminopeptidase inhibitors is desirable.
Aminopeptidases as a class are widely distributed and
play important roles in the turnover of proteins and pep-
tides leading to uptake or re-use of amino acids in various
tissues. Some aminopeptidases possess quite strict sub-
strate-speci®city and limited distribution, and their func-
tion has been shown to be related to physiological and
pathophysiological phenomena such as modulation of the
regulation of physiological events by peptide hormones,
regulation of cell adhesion and mobility by degrading
connective tissue barriers, and so on. Speci®c amino-
peptidase inhibitors would be useful as bioprobes for
investigation of the physiological/pathophysiological
functions of these enzymes, and might also be clinically
useful, e.g., as tumor metastasis inhibitors.
Recently, we have reported novel non-peptide small
molecular inhibitors of aminopeptidase with a phthal-
imide/homophthalimide skeleton.5 9 Some of them were
structurally derived from thalidomide.10 Among them, 2-
(2,6-diethylphenyl)-1,2,3,4-tetrahydroisoquinoline-1,3-
dione (2: PIQ-22) (Fig. 1) is a more potent aminopepti-
dase inhibitor than bestatin in our assay system (vide
infra).6 9 A potent inhibitory e€ect of PIQ-22 on tumor
cell invasion has also been demonstrated.11 Though the
substrate selectivity of the aminopeptidase whose activity
is potently inhibited by PIQ-22 indicated that the enzyme
can be classi®ed as a neutral aminopeptidase,6 8 it has not
been well characterized at the molecular level.9,11
For example, the natural aminopeptidase inhibitor besta-
tin (1) (Fig. 1), which was ®rst isolated from culture ®ltrate
of Streptomyces,1 has attracted considerable attention
since its discovery due to its numerous biological activities,
including as an immune response modi®er, a potential
analgesic, and an inhibitor of tumor metastasis.2 4
Though bestatin is widely employed clinically as an anti-
tumor agent with host-mediated actions, it has drawbacks
for clinical application. These include low bioavailability,
In this article, we ®rst describe the molecular character-
ization of the target aminopeptidase of PIQ-22. Struc-
tural development of PIQ-22 and the results of structure±
activity relationship studies are also described.
Molecular characterization of the target aminopeptidase
of PIQ-22
PIQ-22 (2) showed more potent aminopeptidase-inhi-
biting activity (IC50 value 0.12 mg/mL) than bestatin (1)
*Corresponding author. Tel.: +81-3-5841-7847; fax: +81-3-5841-
8495; e-mail: hashimot@iam.u-tokyo.ac.jp
0968-0896/01/$ - see front matter # 2000 Elsevier Science Ltd. All rights reserved.
PII: S0968-0896(00)00231-5

122
M. Komoda et al. / Bioorg. Med. Chem. 9 (2001) 121±131
Figure 1. Structures of bestatin (1), PIQ-22 (2) and puromycin (3).
(IC50 value of 0.81 mg/mL) when assayed in a system using
human acute lymphoblastic leukemia cells, MOLT-4, as a
source of the enzyme, and l-alanine 7-amino-4-methyl-
coumaryl-7-amide (Ala-AMC) as a substrate.6 9 Amino-
peptidase expressed in another human lymphocytic cell
line, Jurkat, was also sensitive to PIQ-22.12 On the other
hand, the PIQ-22-sensitivity of aminopeptidase(s) expres-
sed in human monocytic cells, such as HL-60 and THP-1,
was much lower (details will be published elsewhere).12
as a single peak with the estimated apparent molecular
weight of 93.3 kDa, which accords reasonably well with
the molecular weight of 102 kDa estimated by SDS±
PAGE of the Mono-Q HPLC fractions (Fig. 2). In the
non-denaturing Superose 6 HPLC, proteins with higher
molecular weight and without aminopeptidase activity
(possibly aggregated and/or inactivated proteins) were
eluted with the retention time of 45±47 min (Fig. 3).
Analysis of each Superose 6 HPLC fraction by SDS±
PAGE again showed that the behavior of only the
102 kDa band (p102) is consistent with that of the amino-
peptidase activity (Fig. 3). Therefore, we concluded that
p102 is the target aminopeptidase of PIQ-22, and its
amino acid sequence was analyzed.
Because monocytic cells exclusively express aminopepti-
dase N (APN), which is identical to the myeloid di€er-
entiation antigen CD13,13 and lymphocytic cells are
reported to be APN/CD13-negative,14 our result sug-
gests that the target molecule of PIQ-22 is di€erent from
APN. In fact, PIQ-22 was inactive toward authentic
APN (porcine, Sigma; data not shown).11,12 Concerning
lymphocytic cells, Wex et al. reported the presence of
aminopeptidase with an isoelectric point of pH 5.4,
which is di€erent from that of APN/CD13 (isoelectric
point: pH 3.5±4.8), though their enzyme has not been
fully characterized.15 Therefore, we conducted a mole-
cular characterization of the PIQ-22-sensitive amino-
peptidase expressed in MOLT-4 cells.
The band of p102 on SDS±PAGE was blotted on a poly-
(vinylidene di¯uoride) (PVDF) membrane ®lter and ana-
lyzed with a gas-phase protein sequencer. The N-terminal
amino acid sequence of PEKRPFERLPADVSPINY was
obtained. A database homology search revealed that
this sequence was identical with the reported sequence of
human puromycin-sensitive aminopeptidase (PSA), and
its calculated molecular weight, 99kDa, is consistent with
that of p102 estimated by gel ®ltration HPLC and SDS±
PAGE, i.e., 93.3 and 102 kDa, respectively.17,18 The result
indicated that the PIQ-22-sensitive aminopeptidase exist-
ing in MOLT-4 cells is PSA. Moreover, it is suggested
that the neutral aminopeptidase predominantly expres-
sed in lymphoid CD13-negative cells is PSA, while that
expressed in monocytic cells is APN/CD13.
Cultured MOLT-4 cells were fractionated and the soluble
fraction was separated successively by ammonium sulfate
fractionation and Mono-Q high-performance liquid col-
umn chromatography (HPLC). Activity of PIQ-22-sensi-
tive aminopeptidase was recovered as a single ammonium
sulfate-precipitated fraction (50±60% ammonium sul-
fate), and eluted as a single peak on Mono-Q HPLC
(Fig. 2). Analysis of each eluted fraction by sodium
Mode of PSA inhibition elicited by PIQ-22
dodecylsulfate±polyacrylamide
gel
electrophoresis
PSA is a neutral aminopeptidase with a similar substrate
speci®city to that of APN, which was ®rst found as a
candidate aminopeptidase for degrading enkephalins,
and is present in brain in much higher amounts than
APN.19 21 Recently, increased anxiety and impaired
pain response in PSA gene-de®cient mice obtained by a
mouse gene-trap method was reported.22 However, the
physiological/pathophysiological role of PSA has not
yet been fully clari®ed.
(SDS±PAGE) showed that the 102 kDa band (p102)
coincided with the aminopeptidase activity (Fig. 2).
Direct activity staining of the electrophoresis gel16
showed a single stained band at the location of p102,
which strongly suggested that MOLT-4 cells express a
unique PIQ-22-sensitive aminopeptidase of 102 kDa.
The active Mono-Q HPLC fractions were collected and
further separated by gel ®ltration HPLC using a Super-
ose 6 column under non-denaturing conditions (Fig. 3).
The PIQ-22-sensitive aminopeptidase activity was eluted
As mentioned before, PIQ-22 is inactive toward APN
(porcine), though bestatin is a potent inhibitor of both

M. Komoda et al. / Bioorg. Med. Chem. 9 (2001) 121±131
123
Figure 2. Mono-Q HPLC chromatogram and SDS±PAGE analysis of the fractions. The chromatogram was monitored by assay of PIQ-22-sensitive
neutral aminopeptidase activity (AP activity: the left vertical scale) measured as described in the Experimental. The horizontal scale is the fraction
number of the HPLC, and the concentration of NaCl in the eluent is shown on the right vertical scale. The SDS±PAGE gel was stained with CBB for
fractions 86±93. The positions of molecular weight markers are also shown.
APN and PSA. PIQ-22 was also inactive towards other
typical neutral aminopeptidases, leucine aminopeptidase
(LAP, porcine, Sigma), dipeptidylpeptidase type IV
(porcine, Sigma), trypsin (bovine, Sigma) and chymo-
trypsin (bovine, Sigma) (data not shown), suggesting
that PIQ-22 is speci®c to PSA. Because PIQ-22 has been
shown to be a potent inhibitor of tumor cell invasion,11
PSA might be one of the target molecules of tumor cell
invasion inhibitors.
competitive manner (Fig. 4(C): the two dose±response
lines cross at the horizontal axis), while puromycin (Fig.
4(A)) and bestatin (Fig. 4(B)) inhibit PSA in a competi-
tive manner (the two dose±response lines cross at the
vertical axis in both cases).9,11 The results suggest that
PIQ-22 does not act as a substrate-mimic, but binds at a
speci®c site of PSA which is not the substrate-binding
site, while puromycin and bestatin act as substrate-
mimics.
PSA is reported to be approximately 100-fold more sen-
sitive to puromycin (3) (Fig. 1) than APN, with an IC50
value of 0.5 mM.19 21 In our assay system using MOLT-4
cells, puromycin inhibited the PIQ-22-sensitive amino-
peptidase with the IC50 value of 0.4 mM, which is in
good agreement with the reported value. Because there
is no obvious structural relationship between PIQ-22 (2)
and puromycin (3), the mode of inhibition elicited by
these two inhibitors, as well as bestatin (1), was analyzed
by Lineweaver±Burk plot analysis (Fig. 4). The analysis
clearly indicated that PIQ-22 inhibits PSA in a non-
Structural development of PIQ-22: e€ects of substituents
introduced into the N-phenyl moiety
The results mentioned above, as well as our previous
studies on neutral aminopeptidase inhibitors with a
phthalimide/homophthalimide skeleton,6 9 suggest that
the structure of PIQ-22 is critically recognized by the
non-competitive site of PSA. We therefore modi®ed the
structure of PIQ-22 in various ways to obtain informa-
tion concerning the non-competitive binding site of PSA
(Table 1).

124
M. Komoda et al. / Bioorg. Med. Chem. 9 (2001) 121±131
Figure 3. Superose 6 HPLC chromatogram and SDS±PAGE analysis of the fractions. The chromatogram was monitored by assay of PIQ-22-sen-
sitive neutral aminopeptidase activity (AP activity: the left vertical scale, the solid line) measured as described in the Experimental and the amount of
protein contained in each fraction (measured by the Lowry method: the right vertical scale, the broken line). The horizontal scale is the retention
time (min). The molecular weight marker proteins were analyzed under the same conditions, and the apparent molecular weight of the amino-
peptidase peak was estimated to be 93.3 kDa by the least-squares method. The SDS±PAGE gel was stained with CBB for fractions with retention
times of 50±57 min.
First, the e€ects of alkyl substituents on the N-phenyl
moiety were examined (2, 4±27). The corresponding
non-alkylated derivative, 4, is almost inactive. Compar-
ison of the PSA-inhibiting activity of monoalkylated
derivatives (5±10) roughly indicated that an ethyl sub-
stituent introduced at the ortho- or meta-position (6 or
7) is superior to a methyl (5) or isopropyl (9) substituent
for potent activity. The substitution position should be
ortho or meta, because the para-monoalkylated deriva-
tives (8 and 10) are inactive.
the most potent inhibitor among these compounds. The
N-(a)-naphthyl derivative (27) was also potent.
Introduction of a methoxy or methylthio group (28±33)
also resulted in the appearance of potent activity. In this
case, introduction at the para-position (30, 33) was also
e€ective.
Dimethoxy derivatives (34±37) were also potent PSA
inhibitors, except the m,p-regioisomer (36). Introduction
of two chlorine atoms at the two ortho positions (38)
resulted in appearance of the activity, but introduction of
two ¯uorine atoms at the corresponding positions (39)
had no e€ect.
The superiority of an ethyl group over other alkyl groups
was also observed in dialkylated derivatives (2, 11±26).
The potency of PSA inhibition elicited by dimethylated
regioisomers decreased in the order of: m,m0-(16)>o,o0-
(11)>o(2),m(5) - (12)>o,p - (13)>m,p - (15)>o(2),m(3) -
(14), while that of diethylated regioisomers decreased in
the order of: o,o0-(2: PIQ-22)>m,m0-(21)>o(2),m(5)-
(17)>m,p-(20)>o,p-(18)>o(2),m(3)-(19). PIQ-22 (2) is
These results suggest that moderately bulky and hydro-
phobic circumstances around the nitrogen atom in the
cyclic imide moiety and perpendicular orientation of the
N-aryl group with respect to the plane of the homo-

M. Komoda et al. / Bioorg. Med. Chem. 9 (2001) 121±131
125
Table 1. PSA-inhibitory activity (IC50 values) of phenyl-homo-
phthalimide analogues
Compound
R1
R2
R3
IC50 (mg/ml)
4
5
6
7
8
9
10
28
29
30
31
32
33
11
12
13
14
15
16
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
20-Me
20-Et
H
H
H
H
H
H
H
H
H
H
H
H
H
60-Me
50-Me
40-Me
30-Me
40-Me
50-Me
60-Et
50-Et
40-Et
30-Et
40-Et
50-Et
60-iPr
60-Et
60-iPr
50-tBu
50-tBu
93.7
41.5
17.8
11.7
>100
54.3
>100
6.2
30-Et
40-Et
20-iPr
40-iPr
20-OMe
30-OMe
40-OMe
20-SMe
30-SMe
40-SMe
20-Me
20-Me
20-Me
20-Me
30-Me
30-Me
20-Et
9.5
7.8
0.9
8.0
6.2
8.7
11.6
16.6
30.0
22.8
1.5
0.12
4.3
19.5
47.0
7.0
2: PIQ-22
17
18
19
20
21
22
23
24
25
26
27
34
35
36
37
20-Et
20-Et
20-Et
30-Et
30-Et
2.5
3.5
0.43
0.78
>100
7.3
1.7
3.2
4.9
>100
5.3
20-iPr
20-Me
20-Me
20tBu
30tBu
20,30-fused-Ph
50-OMe
40-OMe
40OMe
50OMe
60-Cl
50-OMe
40-OMe
40-OMe
50-OMe
60-Cl
H
H
H
H
7-NH2
7-NO2
38
39
51
52
44.1
>100
8.2
60-F
60-F
60-Et
60-Et
60Et
60-Et
28.0
0.81
1: Bestatin
amino (41, 42) or a hydroxyl (43, 44) group into the
phthalimide moiety of inactive 40 resulted in reappear-
ance of the activity. On the other hand, introduction of
an electron-withdrawing nitro group (45, 46) had no
e€ect.
Figure 4. Lineweaver±Burk plot analysis of inhibition of PSA by pur-
omycin (A), bestatin (B) and PIQ-22 (C). &, In the absence of an
inhibitor; ^, in the presence of an inhibitor. V, Rate of enzymatic
hydrolysis of the substrate (Ala-AMC) for PSA (mM/min). Inhibitors
were added at the concentration of 0.5 mM for puromycin (A) and
PIQ-22 (C) and 2.5 mM for bestatin (B). The PSA activity was assessed
as described in the Experimental.
The ring-expanded analogue of PIQ-22, i.e., 47, was also
inactive. Decarbonylation of PIQ-22, i.e., 48, retained
the activity, though the activity was weaker than that of
PIQ-22. On the other hand, carbonylation of PIQ-22,
i.e., 49, resulted in complete disappearance of the activ-
ity. We have found that PIQ-22 is photolabile in a pro-
tic solvent in air, and a methanol solution of PIQ-22
exposed to daylight in air yielded the tricarbonyl analo-
gue (49) quantitatively in a molecular oxygen-dependent
manner (details will be published elsewhere). Brie¯y, a
1H NMR spectroscopic study showed that PIQ-22 was
light-dependently auto-oxidized in methanol to give the
phthalimide moiety are critical factors for potent PSA-
inhibiting activity.
Structural development of PIQ-22: e€ects of derivatization
of the cyclic imide ring
As previously reported,6 9,12 ring reduction of PIQ-22,
i.e., N-2,6-diethylphenylphthalimide (40), resulted in
disappearance of the activity (Fig. 5, Table 2). However,
introduction of an electron-donating group such as an

126
M. Komoda et al. / Bioorg. Med. Chem. 9 (2001) 121±131
Figure 5. Structures of some derivatives of PIQ-22.
Table 2. PSA-inhibitory activity (IC50 values) of some compounds
structurally related to PIQ-22
activity (vide supra). PIQ-22 is the most potent PSA
inhibitor (more potent than bestatin) among the com-
pounds described in this paper.
Compound
IC50 (mg/ml)
2
40
41
42
43
44
45
46
47
48
49
50
0.12
>100
2.5
Comparison of the structures of active compounds with
those of inactive/weakly active compounds, i.e., active
amino/hydroxy-containing arylphthalimides (41±44) ver-
sus inactive arylphthalimides (40 and nitrated analogues
45, 46), active PIQ-22 analogues (PIQ-22: 2 and its dec-
arbonylated analogue 48) versus inactive derivatives
(tricarbonyl analogue 49, di¯uorinated analogue 50 and
ring-expanded 47), led us to consider another possible
interpretation (vide infra). The active compounds listed
above possess a common feature, i.e., they can all be
regarded as existing in a keto/enol tautomeric mixture
(Fig. 6). Keto/enol tautomerism of homophthalic anhy-
dride and trapping of the enol form by Diels±Alder
reaction with a naphthoquinone derivative have been
reported by Tamura et al.24,25
5.6
10.2
9.6
>100
>100
>100
2.4
>100
>100
0.81
Bestatin
corresponding 4-methoxy hemiacetal analogue quanti-
tatively, and this a€orded the tricarbonyl analogue (49)
after work up and recrystalization.12 Therefore, we
designed a non-oxidizable analogue, i.e., the di¯uori-
nated derivative (50). Compound 50 was prepared from
49 by reaction of the carbonyl group at the benzyl
position with (diethylamino)sulfur tri¯uoride (DAST).23
However, 50 was unexpectedly inactive.
In contrast to the putative keto/enol tautomerism of the
active compounds (41±44, PIQ-22, 48), the inactive deri-
vatives listed above (40, 45±47, 49, 50) can be regarded as
compounds which would not readily tautomerize to the
corresponding enol forms. Among the N-arylphthali-
mide analogues, compounds 41±44 are expected to tau-
tomerize to the corresponding enol forms more easily
than 40, 45 and 46 because of the electron-donating
nature of the introduced amino or hydroxyl group. The
tautomerization of the inactive ring-expanded analogue
48 to the corresponding enol form should be dicult
because of ring torsion of the seven-membered ring. The
inactive tricarbonyl analogue 49 and di¯uorinated ana-
logue 50 cannot tautomerize because of the lack of
Possible relevance of tautomerism of the cyclic imide
moiety to the PSA-inhibiting activity
One possible interpretation of the results of our struc-
ture±activity relationship studies mentioned above is
that moderate bulkiness/hydrophobicity around the
cyclic imide nitrogen atom and the orientation of the N-
aryl group are critical factors for potent PSA-inhibiting

M. Komoda et al. / Bioorg. Med. Chem. 9 (2001) 121±131
127
Figure 6. Possible relationship of tautomerism to aminopeptidase-inhibitory activity. (A) Compounds with high activity are considered to exist as
tautomeric mixtures. (B) Compounds with low activity are considered unlikely to tautomerize readily.
active benzyl methylene hydrogens. Therefore, the enol
forms might be the active forms of the compounds
which directly bind to the non-competitive site of PSA.
Experimental
Assay of enzyme activity
Neutral aminopeptidase activity was evaluated in the
usual assay by measuring 7-amino-4-methylcoumarin
(AMC) liberated from l-alanine 4-methylcoumaryl-7-
amide (Ala-AMC).26,27 Brie¯y, Ala-AMC (1 mM) was
dissolved in 50 mM Tris±HCl (pH 7.4) in the presence
or absence of a test inhibitor (various concentrations).
To the solution was added enzymatic activity (MOLT-4
cell extract, commercially available proteases, HPLC
fractions, and so on), and the mixture was incubated at
37 ꢀC for exactly 60 min. Then 6 volumes of 1 M
AcONa±AcOH (pH 4.0) were added to stop the
enzymatic reaction. The amounts of liberated AMC was
measured in terms of ¯uorescence intensity (excitation
at 380 nm, emission at 420 nm). The assay was per-
formed in triplicate, and the mean value was taken.
Though the values deviated from experiment to
experiment, the results (order of potency) were basically
reproducible, and a typical set of data is shown in the
tables.
If this were the case, introduction of a functional group
with an electronic e€ect into the 7-position of the
homophthalimide moiety of PIQ-22 should cause
reduced potency of the PSA-inhibiting activity, because
either an electron-donating or an electron-withdrawing
group at the 7-position would make tautomerization of
the imidobenzoylketone in the cyclic imide moiety
unfavorable (Fig. 6(B)). In fact, introduction of an
amino group (51) or a nitro group (52) at the 7-position
of PIQ-22 decreased the potency of the activity, though
the 7-amino derivative (51) was moderately active
(Table 1).
Conclusion
First, PSA was identi®ed as a PIQ-22-sensitive neutral
aminopeptidase from MOLT-4 cells. This suggests that
the so-far-unidenti®ed aminopeptidase expressed in
lymphocytic cells is PSA. Second, PIQ-22 (2) was
established to be a potent, speci®c, non-peptide, small-
molecular, non-competitive inhibitor of PSA. Because
almost all of the potent protease inhibitors so far known
are substrate mimics/peptide mimics, as in the case of
bestatin (1), and the physiological/pathophysiological
roles of PSA have not yet been well established, PIQ-22
should be a superior bioprobe for investigation of PSA.
The potent inhibitory activity of PIQ-22 on tumor cell
invasion10 12 also suggests the usefulness of the com-
pound as a lead compound for novel drugs, including
antimetastatic agents.
Identi®cation of the neutral aminopeptidase expressed in
MOLT-4
Cells. MOLT-4 cells were maintained in an RPMI1640
medium supplemented with 10% v/v fetal bovine
serum at 37 ꢀC under a 5% CO2 atmosphere. Cells were
collected by centrifugation (2000 rpm, 2 min, 4 ꢀC), and
homogenized in phosphate-bu€ered saline (PBS). The
homogenate was ultracentrifuged at 100,000 g at 4 ꢀC
for 1 h, and the supernatant was separated. The super-
natant was fractionated by successive addition of
ammonium sulfate; the precipitate of each fraction was

128
M. Komoda et al. / Bioorg. Med. Chem. 9 (2001) 121±131
collected by centrifugation (20,000 gÂ3 min at 4 ꢀC).
The aminopeptidase activity was recovered in the frac-
tion of 50±60% saturated (NH4)2SO4. The fraction was
dialyzed and separated by HPLC equipped with a
Mono-Q column by eluting with a linear gradient (Fig.
2) of NaCl (0±1 M) in 20 mM Tris±HCl (pH 8.0)/
250 mM sucrose. The fractions showing aminopeptidase
activity were collected, concentrated by ultra®ltration,
and then further separated by HPLC with a Superose 6
column (eluted with 20 mM Tris±HCl (pH 8.0)/250 mM
sucrose) (Fig. 3). The fractions showing aminopeptidase
activity were collected, concentrated by ultra®ltration,
and then separated by SDS±PAGE according to
Laemmli.28 The bands on the gel were blotted on a
Millipore Immobilon PVDF membrane, and the
membrane was stained with Coomassie Brilliant Blue
(CBB). The band of p102 was cut out and the N-
terminal amino acid sequence was analyzed with a
gas phase amino acid sequencer (Shimadzu PPSQ-10).
Homology search of the obtained sequence was
performed by FASTA search (http://www.fasta.genome.
ad.jp).
2-(4-Ethylphenyl)-1,2,3,4-tetrahydroisoquinoline-1,3-dione
(8). Mp 139±140 ꢀC. 1H NMR (500 MHz, CDCl3) d:
1.28 (3H, t, J=7.5 Hz), 2.72 (2H, q, J=7.5 Hz), 4.22
(2H, s), 7.12 (2H, d, J=8.0 Hz), 7.35 (1H, d, J=7.6 Hz),
7.48 (1H, d, J=8.0 Hz), 7.64 (1H, dt, J=7.6, 1.0 Hz),
8.25 (1H, dd, J=8.0, 1.0 Hz). Anal. calcd for C17H15
NO2: C, 76.96; H, 5.70; N, 5.28. Found: C, 76.94; H,
5.70; N, 5.26.
2-(2-Isopropylphenyl)-1,2,3,4-tetrahydroisoquinoline-1,3-
dione (9). Mp 158±160 ꢀC. H NMR (500 MHz, CDCl3)
1
d: 1.17 (6H, t, J=6.85 Hz), 2.70 (1H, q, J=6.85 Hz),
4.24 (2H, s), 7.05 (1H, d, J=8.32 Hz), 7.26±7.33 (1H,
m), 7.37 (1H, d, J=7.33 Hz), 7.42±7.48 (1H, m), 7.50
(1H, d, J=7.83 Hz), 7.66 (1H, dt, J=7.83, 1.47 Hz),
8.26 (1H, dd, J=7.83, 1.0 Hz). Anal. calcd for C18H17
NO2: C, 77.40; H, 6.13; N, 5.01. Found: C, 77.23; H,
6.37; N, 4.84.
2-(4-Isopropylphenyl)-1,2,3,4-tetrahydroisoquinoline-1,3-
dione (10). Mp 170±172 ꢀC. 1H NMR (500 MHz,
CDCl3) d: 1.29 (6H, t, J=8.85 Hz), 2.97 (1H, q, J=
6.85 Hz), 4.23 (2H, s), 7.10±7.13 (1H, m), 7.12 (2H, d,
J=8.31 Hz), 7.34±7.36 (1H, m), 7.36 (2H, d, J=
8.31 Hz), 7.47 (1H, t, J=7.34 Hz), 7.64 (1H, dt, J=7.83,
1.47 Hz), 8.25 (1H, d, J=7.34 Hz). Anal. calcd for
C18H17NO2: C, 77.40; H, 6.13; N, 5.01. Found: C,
77.33; H, 6.29; N, 4.75.
Chemicals. Cyclic imide derivatives were prepared as
described previously.5,6 Typically, phthalic/homo-
phthalic anhydride was condensed with an appropriate
amine. The syntheses of 4, 11, 15, 17, 22±27, 32, 33, 37,
41, 43, 45 and 47 were already reported.29 32
2-(2,6-Diethylphenyl)-1,2,3,4-tetrahydroisoquinoline-1,3-
dione (PIQ-22: 2). Mp 108±110 ꢀC. 1H NMR (500MHz,
CDCl3) d: 1.14 (6H, t, J=7.33 Hz), 2.40 (4H, q, J=
7.33Hz), 4.25 (2H, s), 7.24 (2H, d, J=7.83 Hz), 7.36±7.40
(2H, m), 7.50 (1H, t, J=7.83 Hz), 7.67 (1H, dt, J=7.83,
1.43 Hz), 8.26 (1H, d, J=6.85 Hz). Anal. calcd for
C19H19NO2: C, 77.79; H, 6.53; N, 4.77. Found: C,
77.69; H, 6.43; N, 4.56.
2-(2,5-Dimethylphenyl)-1,2,3,4-tetrahydroisoquinoline-
1,3-dione (12). Mp 125±126 ꢀC. 1H NMR (500MHz,
CDCl3) d: 2.08 (3H, s), 2.35 (3H, s), 4.23 (2H, s), 6.92 (1H,
s), 7.16 (1H, d, J=7.94 Hz), 7.24 (1H, d, J=7.63 Hz), 7.36
(1H, d, J=7.63 Hz), 7.49 (1H, t, J=7.94 Hz), 7.65 (1H,
dt, J=7.63, 1.22 Hz), 8.25 (1H, d, J=7.94 Hz). Anal.
calcd for C17H15NO2: C, 76.96; H, 5.70; N, 5.28. Found:
C, 77.03; H, 5.71; N, 5.27.
2-(2-Methylphenyl)-1,2,3,4-tetrahydroisoquinoline-1,3-
dione (5). Mp 107±108 ꢀC. H NMR (500 MHz, CDCl3)
2-(2,4-Dimethylphenyl)-1,2,3,4-tetrahydroisoquinoline-
1,3-dione (13). Mp 171 ꢀC. H NMR (500MHz, CDCl3)
1
1
d: 2.14 (3H, s), 4.24 (2H, s), 7.10 (1H, d, J=7.2 Hz),
7.31±7.37 (4H, m), 7.49 (1H, t, J=8.0 Hz), 7.66 (1H, dt,
J=7.5, 1.0 Hz), 8.26 (1H, dd, J=7.5, 1.0 Hz). Anal.
calcd for C16H13NO2: C, 76.48; H, 5.21; N, 5.57. Found:
C, 76.47; H, 5.17; N, 5.28.
d: 2.09 (3H, s), 2.37 (3H, s), 4.23 (2H, s), 6.98 (1H, d,
J=7.93 Hz), 7.16 (1H, d, J=7.93 Hz), 7.17 (1H, s), 7.36
(1H, d, J=7.63 Hz), 7.48 (1H, t, J=7.93 Hz), 7.65 (1H,
dt, J=7.63, 1.53 Hz), 8.25 (1H, dd, J=7.94, 0.92 Hz).
Anal. calcd for C17H15NO2: C, 76.96; H, 5.70; N, 5.28.
Found: C, 76.97; H, 5.75; N, 5.25.
2-(2-Ethylphenyl)-1,2,3,4-tetrahydroisoquinoline-1,3-dione
(6). Mp 123±124 ꢀC. 1H NMR (500 MHz, CDCl3) d:
1.16 (3H, t, J=7.63 Hz), 2.45 (2H, q, J=7.63 Hz), 4.24
(2H, s), 7.31±7.35 (1H, m), 7.37 (1H, d, J=7.63 Hz),
7.41±7.42 (2H, m), 7.49 (1H, t, J=7.94 Hz), 7.66 (1H, dt,
J=7.63, 1.22 Hz), 8.25 (1H, dd, J=7.94, 0.92 Hz). Anal.
calcd for C17H15NO2: C, 76.96; H, 5.70; N, 5.28. Found:
C, 77.04; H, 5.70; N, 5.21.
2-(2,3-Dimethylphenyl)-1,2,3,4-tetrahydroisoquinoline-
1,3-dione (14). Mp 145±146 ꢀC. 1H NMR (500 MHz,
CDCl3) d: 2.02 (3H, s), 2.35 (3H, s), 4.24 (2H, s), 6.95
(1H, d, J=7.02 Hz), 7.20±7.24 (2H, m), 7.36 (1H, d, J=
7.63 Hz), 7.49 (1H, t, J=7.93 Hz), 7.65 (1H, dt, J=7.63,
1.22 Hz), 8.25 (1H, d, J=7.93 Hz). Anal. calcd for
C17H15NO2: C, 76.96; H, 5.70; N, 5.28. Found: C,
76.58; H, 5.69; N, 5.26.
2-(3-Ethylphenyl)-1,2,3,4-tetrahydroisoquinoline-1,3-dione
(7). Mp 76.5±78 ꢀC. 1H NMR (500 MHz, CDCl3) d:
1.17 (3H, t, J=8.0 Hz), 2.72 (2H, q, J=8.0 Hz), 4.23
(2H, s), 7.03 (2H, d, J=8.0 Hz), 7.29 (1H, d, J=8.0 Hz),
7.35 (1H, d, J=8.0 Hz), 7.42 (1H, t, J=8.0 Hz), 7.64
(1H, dt, J=7.5, 1.2 Hz), 8.25 (1H, d, J=8.0 Hz). Anal.
calcd for C17H15NO2: C, 76.96; H, 5.70; N, 5.28. Found:
C, 77.03; H, 5.74; N, 5.30.
2-(3,5-Dimethylphenyl)-1,2,3,4-tetrahydroisoquinoline-
1,3-dione (16). Mp 129±130 ꢀC. 1H NMR (500MHz,
CDCl3) d: 2.36 (6H, s), 4.21 (2H, s), 6.82 (2H, s), 7.07 (1H,
s), 7.34 (1H, d, J=7.63 Hz), 7.47 (1H, t, J=7.63 Hz), 7.63
(1H, dt, J=7.63, 1.22Hz), 8.24 (1H, d, J=7.94 Hz). Anal.
calcd for C17H15NO2: C, 76.96; H, 5.70; N, 5.28. Found:
C, 77.04; H, 5.73; N, 5.23.

M. Komoda et al. / Bioorg. Med. Chem. 9 (2001) 121±131
129
2-(2,4-Diethylphenyl)-1,2,3,4-tetrahydroisoquinoline-1,3-
dione (18). Mp 107±108 ꢀC. 1H NMR (500MHz, CDCl3)
d: 1.15 (3H, t, J=7.5Hz), 1.28 (3H, t, J=7.5 Hz), 2.41
(2H, q, J=7.5 Hz), 2.70 (2H, q, J=7.5 Hz), 4.23 (2H, s),
6.99 (1H, d, J=8.0 Hz), 7.16 (1H, dd, J=8.0, 1.5 Hz),
7.22 (1H, d, J=1.5 Hz), 7.36 (1H, d, J=7.8 Hz), 7.49
(1H, t, J=7.5 Hz), 7.65 (1H, dt, J=7.5, 1.0 Hz), 8.25
(1H, d, J=7.8 Hz). Anal. calcd for C19H19NO2: C, 77.79;
H, 6.53; N, 4.77. Found: C, 77.53; H, 6.53; N, 4.50.
(1H, dt, J=8.0, 1.0 Hz), 7.64 (1H, dt, J=8.0, 1.0 Hz),
8.25 (1H, dd, J=8.0, 1.0 Hz). Anal. calcd for C16H13
NO3: C, 71.90; H, 4.90; N, 5.24. Found: C, 71.83; H,
4.95; N, 4.92.
2-(2-Methylthiophenyl)-1,2,3,4-tetrahydroisoquinoline-
1,3-dione (31). Mp 106±108 ꢀC. 1H NMR (500 MHz,
CDCl3) d: 2.41 (3H, s), 4.20 (1H, d, J=2.5 Hz), 4.30
(1H, d, J=2.5 Hz), 7.17 (1H, d, J=7.83 Hz), 7.31 (1H,
dt, J=7.83, 2.0 Hz), 7.34 (1H, d, J=7.34 Hz), 7.40±7.47
(1H, m), 7.48 (1H, t, J=7.83 Hz), 7.65 (1H, dt, J=7.83,
1.47 Hz), 8.26 (1H, dd, J=7.83, 1.47 Hz). Anal. calcd
for C16H13NO2S: C, 67.82; H, 4.62; N, 4.94. Found: C,
67.86; H, 4.58; N, 4.70.
2-(2,3-Diethylphenyl)-1,2,3,4-tetrahydroisoquinoline-1,3-
dione (19). Mp 155 ꢀC. H NMR (500 MHz, CDCl3) d:
1
1.03 (3H, t, J=7.5 Hz), 1.29 (3H, t, J=7.5 Hz), 2.46
(2H, q, J=7.5 Hz), 2.74 (2H, q, J=7.5 Hz), 4.24 (2H, s),
6.93 (1H, dd, J=7.2, 1.5 Hz), 7.27 (1H, t, J=7.2 Hz),
7.31 (1H, dd, J=7.2, 1.5 Hz), 7.37 (1H, dd, J=7.8,
1.0 Hz), 7.49 (1H, dd, J=7.6, 1.0 Hz), 7.66 (1H, dd,
J=7.6, 1.0 Hz), 8.25 (1H, dd, J=7.8, 1.0 Hz). Anal.
calcd for C19H19NO2: C, 77.79; H, 6.53; N, 4.77. Found:
C, 77.82; H, 6.47; N, 4.70.
2-(2,5-Dimethoxylphenyl)-1,2,3,4-tetrahydroisoquinoline-
1,3-dione (34). Mp 207±209 ꢀC. 1H NMR (500 MHz,
CDCl3) d: 3.73 (3H, s), 3.77 (3H, s), 6.74 (1H, dd,
J=2.44, 0.98 Hz), 6.97 (2H, s), 7.35 (1H, d, J=7.81 Hz),
7.47 (1H, dt, J=7.81, 0.98 Hz), 7.63 (1H, dt, J=7.81,
0.98 Hz), 8.25 (1H, dd, J=7.81, 0.98 Hz). Anal. calcd
for C17H15NO4: C, 68.68; H, 5.09; N, 4.71. Found: C,
68.35; H, 5.08; N, 4.61.
2-(3,4-Diethylphenyl)-1,ꢀ2,3,4-tetrahydroisoquinoline-1,3-
dione (20). Mp 121±122 C. 1H NMR (500MHz, CDCl3)
d: 1.24 (3H, t, J=7.5 Hz), 1.26 (3H, t, J=7.5Hz), 2.69
(2H, q, J=7.5 Hz), 2.70 (2H, q, J=7.5 Hz), 4.22 (2H, s),
6.98 (2H, d, J=8.0 Hz), 7.30 (1H, t, J=8.0 Hz), 7.35
(1H, d, J=7.9 Hz), 7.49 (1H, dt, J=7.6, 1.0 Hz), 7.63
(1H, dt, J=7.6, 1.0 Hz), 8.24 (1H, dd, J=7.9, 1.0 Hz).
Anal. calcd for C19H19NO2: C, 77.79; H, 6.53; N, 4.77.
Found: C, 77.77; H, 6.62; N, 4.76.
2-(2,4-Dimethoxylphenyl)-1,2,3,4-tetrahydroisoquinoline-
1,3-dione (35). Mp 194±195 ꢀC. 1H NMR (500MHz,
CDCl3) d: 3.75 (3H, s), 3.84 (3H, s), 4.21 (2H, d,
J=7.81 Hz), 6.57±6.59 (2H, m), 7.06 (1H, dd, J=9.28,
2.44Hz), 7.34 (1H, d, J=7.81 Hz), 7.46 (1H, t, J=
7.81Hz), 7.62 (1H, dt, J=7.81, 1.47Hz), 8.24 (1H, d, J=
7.81Hz). Anal. calcd for C17H15NO4: C, 68.68; H, 5.09;
N, 4.71. Found: C, 68.22; H, 5.16; N, 4.54.
2-(3,5-Diethylphenyl)-1,ꢀ2,3,4-tetrahydroisoquinoline-1,3-
dione (21). Mp 130±131 C. 1H NMR (500MHz, CDCl3)
d: 1.26 (6H, t, J=7.7Hz), 2.67 (4H, q, J=7.7Hz), 4.22
(2H, s), 6.85 (2H, s), 7.12 (2H, s), 7.34 (1H, d, J=
7.6 Hz), 7.47 (1H, d, J=7.6 Hz), 7.63 (1H, dt, J=7.6,
1.0 Hz), 8.24 (1H, dd, J=7.6, 1.0 Hz). Anal. calcd for
C19H19NO2: C, 77.79; H, 6.53; N, 4.77. Found: C,
77.45; H, 6.72; N, 4.46.
2-(3,4-Dimethoxylphenyl)-1,2,3,4-tetrahydroisoquinoline-
1,3-dione (36). Mp 205±207 ꢀC. 1H NMR (500 MHz,
CDCl3) d: 3.87 (3H, s), 3.92 (3H, s), 4.23 (2H, s), 6.70
(1H, d, J=2.44 Hz), 6.78 (1H, dd, J=8.30, 2.44 Hz),
6.98 (1H, d, J=8.30 Hz), 7.35 (1H, d, J=7.32 Hz), 7.49
(1H, t, J=7.32 Hz), 7.65 (1H, dt, J=7.81, 1.46 Hz), 8.26
(1H, dd, J=7.81, 0.98 Hz). Anal. calcd for C17H15NO4:
C, 68.68; H, 5.09; N, 4.71. Found: C, 68.58; H, 5.06; N,
4.62.
2-(2-Methoxyphenyl)-1,ꢀ2,3,4-tetrahydroisoquinoline-1,3-
dione (28). Mp 129±130 C. 1H NMR (500MHz, CDCl3)
d: 3.78 (3H, s), 4.22 (2H, d, J=11.3 Hz), 7.04±7.09 (2H,
m), 7.17 (1H, dd, J=7.63, 1.52Hz), 7.30 (1H, d,
J=7.63 Hz), 7.43 (1H, dt, J=8.24, 1.83Hz), 7.48 (1H, t,
J=7.63 Hz), 7.65 (1H, dt, J=7.83, 1.47Hz), 8.26 (1H, dd,
J=7.83, 1.47 Hz). Anal. calcd for C16H13NO3: C, 71.90;
H, 4.90; N, 5.24. Found: C, 71.97; H, 4.89; N, 5.19.
2 - (2,6 - Dichlorophenyl) - 1,2,3,4 - tetrahydroisoquinoline -
1,3-dione (38). Bp 250 ꢀC (2.0 mmHg). H NMR (500
1
MHz, CDCl3) d: 4.27 (2H, s), 7.34±7.39 (2H, m), 7.41±
7.52 (3H, m), 7.68 (1H, dt, J=7.32, 1.22 Hz), 8.27 (1H,
d, J=7.32 Hz). Anal. calcd for C15H9Cl2NO2: C, 58.85;
H, 2.96; N, 4.58. Found: C, 58.57; H, 2.91; N, 4.48.
2-(3-Methoxyphenyl)-1,2,3,4-tetrahydroisoquinoline-1,3-
dione (29). Mp 169 ꢀC. p1H NMR (500 MHz, CDCl3) d:
3.82 (3H, s), 4.23 (2H, d, J=11.3 Hz), 6.75 (1H, t,
J=2.0 Hz), 6.80 (1H, dq, J=8.0, 1.0 Hz), 7.00 (1H, ddd,
J=8.0, 2.5, 1.0 Hz), 7.35 (1H, d, J=8.0 Hz), 7.41 (1H, t,
J=8.0 Hz), 7.49 (1H, t, J=7.5 Hz), 7.64 (1H, dt, J=7.5,
1.0 Hz), 8.25 (1H, dd, J=8.0, 1.0 Hz). Anal. calcd for
C16H13NO3: C, 71.90; H, 4.90; N, 5.24. Found: C,
71.81; H, 4.94; N, 4.94.
2 - (2,6 - Di¯uorophenyl) - 1,2,3,4 - tetrahydroisoquinoline -
1,3-dione (39). Mp 209±210 ꢀC. 1H NMR (500 MHz,
CDCl3) d: 4.26 (2H, s), 7.06 (2H, t, J=7.33 Hz), 7.37
(1H, d, J=7.94 Hz), 7.40±7.46 (1H, m), 7.50 (1H, t,
J=7.33 Hz), 7.67 (1H, dt, J=7.33, 0.92 Hz), 8.27 (1H,
d, J=7.94 Hz). Anal. calcd for C15H9F2NO2: C, 65.94;
H, 3.32; N, 5.13. Found: C, 66.01; H, 3.24; N, 5.05.
2-(2,6-Diethylphenyl)-1H-isoindole-1,3-dione (40). Mp
154 ꢀC. 1H NMR (500 MHz, CDCl3) d: 1.16 (6H, t,
J=7.5 Hz), 2.47 (4H, q, J=7.5 Hz), 7.32 (3H, m), 7.91
(4H, m). Anal. calcd for C18H17NO2: C, 77.11; H, 6.05;
N, 4.86. Found: C, 77.39; H, 6.13; N, 5.01.
2-(4-Methoxyphenyl)-1,2,3,4-tetrahydroisoquinoline-1,3-
dione (30). Mp 195 ꢀC. H NMR (500 MHz, CDCl3) d:
3.85 (3H, s), 4.22 (2H, d, J=11.3 Hz), 7.00±7.03 (2H,
m), 7.11±7.14 (2H, m), 7.35 (1H, d, J=8.0 Hz), 7.48
1

130
M. Komoda et al. / Bioorg. Med. Chem. 9 (2001) 121±131
2-(2,6-Diethylphenyl)-5-amino-1H-isoindole-1,3-dione
(500 MHz, CDCl3) d: 1.14 (6H, t, J=7.5 Hz), 2.40 (4H,
q, J=7.5 Hz), 7.26 (2H, d, J=7.5 Hz), 7.42 (1H, t,
J=7.5 Hz), 7.81 (1H, t, J=7.5 Hz), 7.90 (1H, t, J=
7.5 Hz), 8.00 (1H, d, J=7.5 Hz), 8.32 (1H, d, J=7.5 Hz).
Anal. calcd for C19H17NO2F2: C, 69.30; H, 5.17; N,
4.25. Found: C, 69.02; H, 5.11; N, 4.46.
(42). Mp 175 ꢀC. H NMR (500MHz, CDCl3) d: 1.13
1
(6H, t, J=7.5 Hz), 2.46 (4H, q, J=7.5Hz), 4.40 (2H, s),
6.90 (1H, dd, J=8.0, 2.0 Hz), 7.13 (1H, d, J=2.0Hz), 7.22
(1H, d, J=7.5 Hz), 7.36 (1H, t, J=7.50 Hz), 7.72 (1H, d,
J=8.0 Hz). Anal. calcd for C18H18N2O2: C, 73.45; H,
6.16; N, 9.52. Found: C, 73.59; H, 6.02; N, 9.23.
2-(2,6-Diethylphenyl)-7-amino-1,2,3,4-tetrahydroisoqui-
noline-1,3-dione (51). Mp 153±155 ꢀC. HRMS(EI) M+:
308.151 (calcd for C19H20N2O2: 308.153). 1H NMR
(500 MHz, CDCl3) d: 1.13 (6H, t, J=7.3 Hz), 2.39 (4H,
q, J=7.3 Hz), 3.87 (2H, br s), 4.12 (2H, s), 6.97 (1H, dd,
J=8.2, 2.4 Hz), 7.15 (1H, d, J=8.2 Hz), 7.23 (2H, d,
J=7.6 Hz), 7.36 (1H, t, J=7.6 Hz), 7.51 (1H, d, J=
2.4 Hz).
2-(2,6-Diethylphenyl)-5-hydroxy-1H-isoindole-1,3-dione
ꢀ
1
(44). Mp 177±178 C. H NMR (500MHz, CDCl3) d:
1.14 (6H, t, J=7.5 Hz), 2.46 (4H, q, J=7.5 Hz), 7.18 (1H,
dd, J=8.0, 2.8 Hz), 7.23 (1H, d, J=8.0 Hz), 7.36 (1H, d,
J=2.0 Hz), 7.38 (1H, t, J=8.0 Hz), 7.86 (1H, d, J=
8.0 Hz). Anal. calcd for C18H17NO3: C, 73.20; H, 5.80;
N, 4.74. Found: C, 73.20; H, 5.79; N, 4.84.
2-(2,6-Diethylphenyl)-5-nitro-1H-isoindole-1,3-dione (46).
2-(2,6-Diethylphenyl)-7-nitro-1,2,3,4-tetrahydroisoquino-
line-1,3-dione (52). Mp 170±171 ꢀC. HRMS(EI) M+:
338.128 (calcd for C19H18N2O4: 338.127). 1H NMR
(500 MHz, CDCl3) d: 1.15 (6H, t, J=7.5 Hz), 2.39 (4H,
q, J=7.5 Hz), 4.35 (2H, s), 7.26 (2H, d, J=7.5 Hz), 7.41
(1H, t, J=7.5 Hz), 7.60 (1H, d, J=8.5 Hz), 8.50 (1H,
dd, J=8.5, 2.5 Hz), 9.09 (1H, d, J=2.5 Hz).
Mp 134±135 ꢀC. H NMR (500 MHz, CDCl3) d: 1.14
1
(6H, t, J=7.6 Hz), 2.44 (4H, q, J=7.6 Hz), 7.26 (2H, d,
J=7.5 Hz), 7.43 (1H, t, J=7.5 Hz), 8.17 (1H, d, J=
8.1 Hz), 8.70 (1H, dd, J=8.1, 2.0 Hz), 8.80 (1H, d,
J=2.0 Hz). Anal. calcd for C18H16N2O4: C, 66.66; H,
4.97; N, 8.64. Found: C, 66.76; H, 4.86; N, 8.42.
N-(2,6-Diethylphenyl)-3-(2-carboxyphenyl)propionimide
(47). 3-(2-Carboxyphenyl)propionic acid anhydride was
condensed with m-diethylaniline in CH2Cl2. Then, to the
mixture was added pivaloyl chloride and Et3N, and the
whole was stirred for 3 days. Mp 126±158 ꢀC. FAB-Mass:
308 (M+1)+. 1H NMR (500MHz, CDCl3) d: 1.07 (6H, t,
J=7.5Hz), 2.40 (4H, q, J=7.5Hz), 2.76 (2H, t, J=
7.6Hz), 3.50 (2H, t, J=7.6 Hz), 7.04 (2H, d, J=7.5 Hz),
7.16 (1H, t, J=7.5 Hz), 7.39 (1H, t, J=7.5 Hz), 7.49 (1H, d,
J=7.5 Hz), 7.58 (1H, d, J=7.5 Hz), 8.11 (1H, d, J=
7.5 Hz).
Acknowledgements
The work described in this paper was partially sup-
ported by funds for the Promotion of Fundamental
Studies in Health Science from the Organization for
Pharmaceutical Safety Research, and Grants-in-Aid
from the Ministry of Education, Science, Sports, and
Culture of Japan. The authors are grateful to Dr.
Naoko Morisaki and Dr. Hisayoshi Kobayashi (IMCB,
University of Tokyo) for their assistance.
2-(2,6-Diethylphenyl)-3,4-dihydroisoquinolin-1(2H)-one
(48). PIQ-22 was treated with NaBH4 to give 2-(2,6-
diethylphenyl)isoquinolin-1(2H)-one (pale yellow oil,
1H NMR (500 MHz, CDCl3) d: 1.13 (6H, t, J=7.5 Hz),
2.35±2.45 (4H, m), 6.59 (1H, d, J=7.5 Hz), 6.93 (1H, d,
J=7.5 Hz), 7.24 (2H, d, J=7.8 Hz), 7.37 (1H, t, J=
7.8 Hz), 7.53 (1H, dd, J=7.5, 1.0 Hz), 7.60 (1H, d,
J=8.0 Hz), 7.70 (1H, dd, J=7.5, 1.0 Hz), 8.59 (1H, dt,
J=8.0, 0.5 Hz)), which was further hydrogenated by H2
gas catalyzed by 5% Pd/C to give 48. Pale yellow oil. 1H
NMR (500 MHz, CDCl3) d: 1.23 (6H, t, J=7.5 Hz),
2.69 (4H, s), 3.19 (2H, t, J=6.5 Hz), 3.77 (2H, t, J=
6.5 Hz), 7.20 (2H, d, J=7.5 Hz), 7.27 (1H, d, J=7.5 Hz),
7.28 (1H, t, J=7.5 Hz), 7.39 (1H, t, J=7.5 Hz), 7.48
(1H, dt, J=7.5, 1.0 Hz), 8.16 (1H, d, J=7.5 Hz).
References
1. Umezawa, H.; Aoyagi, T.; Suda, H.; Hamada, M.; Takeu-
chi, T. J. Antibiotics 1976, 29, 97.
2. Umezawa, H.; Ishizuka, M.; Aoyagi, T.; Takeuchi, T. J.
Antibiotics 1976, 29, 857.
3. Barclay, R. K.; Phillips, M. A. Biochem. Biophys. Res.
Commun. 1980, 96, 1732.
4. Roques, B. P.; Fournie-Zaluski, M. C.; Soroca, E.;
Lecomte, J. M.; Malfroy, B.; Lloreus, C.; Schwartz, J.-C.
Nature 1980, 288, 286.
5. Shimazawa, R.; Takayama, H.; Kato, F.; Kato, M.;
Hashimoto, Y. Bioorg. Med. Chem. Lett. 1999, 9, 559.
6. Miyachi, H.; Kato, M.; Kato, F.; Hashimoto, Y. J. Med.
Chem. 1998, 41, 263.
7. Shimazawa, R.; Takayama, H.; Hashimoto, Y. J. Syn. Org.
Chem. Jpn 1999, 57, 92.
8. Shimazawa, R.; Takayama, H.; Fujimoto, Y.; Komoda,
M.; Dodo, K.; Yamasaki, R.; Shirai, R.; Koiso, Y.; Miyata,
K.; Kato, F.; Kato, M.; Miyachi, H.; Hashimoto, Y. J.
Enzyme Inhibit. 1999, 14, 257.
9. Hashimoto, Y. Recent Res. Develop. Org. Chem. 2000, 4, 87.
10. Hashimoto, Y. Curr. Med. Chem. 1998, 5, 163.
11. Kagechika, H.; Komoda, M.; Fujimoto, Y.; Koiso, Y.;
Takayama, H.; Kadoya, S.; Miyata, K.; Kato, F.; Kato, M.;
Hashimoto, Y. Biol. Pharm. Bull 1999, 22, 1010.
12. Takahashi, H.; Komoda, M.; Kakuta, H.; Hashimoto, Y.
Yakugaku Zasshi 2000, 120, 0000.
2-(2,6-Diethylphenyl)-1,2,3,4-tetrahydroisoquinoline-1,3,4-
trione (49). Mp 163±165 ꢀC. 1H NMR (500MHz,
CDCl3) d: 1.14 (6H, t, J=7.5Hz), 2.41 (4H, q, J=7.5Hz),
7.26 (2H, d, J=8.5Hz), 7.42 (1H, t, J=7.5 Hz), 7.92 (1H,
td, J=7.5, 1.0 Hz), 7.97 (1H, td, J=7.5, 1.0 Hz), 8.34
(1H, dd, J=7.5, 1.0 Hz), 8.41 (1H, dd, J=7.5, 1.0 Hz).
Anal. calcd for C19H17N3: C, 74.24; H, 5.59; N, 4.56.
Found: C, 74.32; H, 5.84; N, 4.43.
2-(2,6-Diethylphenyl)-4,4-di¯uoro-1,2,3,4-tetrahydroiso-
quinoline-1,3-dione (50). Mp 112±123 ꢀC. 1H NMR

M. Komoda et al. / Bioorg. Med. Chem. 9 (2001) 121±131
131
13. Look, A. T.; Ashmun, R. A.; Shapiro, L. H.; Peiper, S. C.
J. Clin. Invest. 1989, 83, 1299.
23. Middleton, W. J.; Bingham, E. M. J. Org. Chem. 1980, 45,
2883.
14. Ashmun, R.; Look, A. T. Blood 1990, 75, 462.
15. Wex, T.; Lendeckel, U.; Kahne, T.; Ittenson, A.; Frank,
K.; Ansorge, S. Biochim. Biophys. Acta 1997, 1355, 147.
16. Yamamoto, Y.; Li, Y. H.; Huang, K.; Ohkubo, I.; Nishi,
K. Biol. Chem. 1998, 379, 711.
17. Tobler, A. R.; Constam, D. B.; Schmitt-Gra€, A.; Mal-
ipiero, U.; Schlapbach, R.; Fontana, A. J. Neurochem. 1998,
68, 889.
24. Tamura, Y.; Wada, A.; Sasho, M.; Kita, Y. Tetrahedron
Lett. 1981, 22, 4283.
25. Tamura, Y.; Wada, A.; Sasho, M.; Fukunaga, K.; Maeda,
H.; Kita, Y. J. Org. Chem. 1982, 47, 4376.
26. Saiki, I.; Fujii, H.; Yoneda, J.; Abe, F.; Nakajima, M.;
Tsuruo, T.; Azuma, I. Int. J. Cancer 1993, 54, 137.
27. Fujii, H.; Nakajima, M.; Aoyagi, T.; Tsuruo, T. Biol.
Pharm. Bull. 1996, 19, 6.
18. Thompson, M. W.; Tobler, A.; Fontana, A.; Hersh, L. B.
Biochem. Biophys. Res. Commun. 1999, 258, 234.
19. Constam, D. B.; Tobler, A. R.; Rensing-Ehl, A.; Kelmer,
I.; Hersh, L. N.; Fontana, A. J. Biol. Chem. 1995, 270, 26931.
20. Hersh, L. B.; McKelve, J. F. J. Neurochem. 1981, 36, 171.
21. McLellan, S.; Dyer, S. H.; Rodriguez, G.; Hersh, L. B. J.
Neurochem. 1988, 51, 1552.
22. Osada, T.; Ikegami, S.; Takiguchi-Hayashi, K.; Yama-
zaki, Y.; Katoh-Fukui, Y.; Higashinakagawa, T.; Sasaki, Y.;
Takeuchi, T. J. Neurosci. 1999, 19, 6068.
28. Laemmli, U. K. Nature 1970, 227, 680.
29. Miyachi, H.; Azuma, A.; Hioki, E.; Kobayashi, Y.; Iwa-
saki, S.; Hashimoto, Y. Chem. Pharm. Bull. 1996, 44, 1980.
30. Miyachi, H.; Ogasawara, A.; Azuma, A.; Hashimoto, Y.
Bioorg. Med. Chem. 1997, 5, 2095.
31. Miyachi, H.; Azuma, A.; Ogasawara, A.; Uchimura, E.;
Watanabe, N.; Kobayashi, Y.; Hashimoto, Y. J. Med. Chem.
1997, 40, 2858.
32. Shibata, Y.; Sasaki, K.; Hashimoto, Y.; Iwasaki, S. Chem.
Pharm. Bull. 1996, 44, 156.