Synthesis of 7-benzylamino-6-chloro-2-piperazino-4-pyrrolidinopteridine and novel derivatives free of positional isomers. Potent inhibitors of cAMP- specific phosphodiesterase and of malignant tumor cell growth

Article abstract of DOI:10.1021/jm981021v

7-Benzylamino-6-chloro-2-piperazino-4-pyrrolidinopteridine (7a) is a potent inhibitor of the cAMP-specific phosphodiesterase isoenzyme family PDE4 and induces growth inhibition in a panel of tumor cell lines. In this study, we describe a synthesis that yi

Full text of DOI:10.1021/jm981021v

J . Med. Chem. 1998, 41, 4733-4743
4733
Syn th esis of 7-Ben zyla m in o-6-ch lor o-2-p ip er a zin o-4-p yr r olid in op ter id in e a n d
Novel Der iva tives F r ee of P osition a l Isom er s. P oten t In h ibitor s of
cAMP -Sp ecific P h osp h od iester a se a n d of Ma lign a n t Tu m or Cell Gr ow th
Karl-Heinz Merz,† Doris Marko,† Thomas Regiert,† Guido Reiss,‡ Walter Frank,‡ and Gerhard Eisenbrand*,†
Departments of Chemistry, Division of Food Chemistry and Environmental Toxicology and Division of Inorganic Chemistry,
University of Kaiserslautern, Erwin-Schro¨dinger-Str. 52, D-67663 Kaiserslautern, Germany
Received April 4, 1998
7-Benzylamino-6-chloro-2-piperazino-4-pyrrolidinopteridine (7a ) is a potent inhibitor of the
cAMP-specific phosphodiesterase isoenzyme family PDE4 and induces growth inhibition in a
panel of tumor cell lines. In this study, we describe a synthesis that yields 7a and novel
derivatives free of positional isomers. The synthesis of alkylamino substituted pteridines is
based on the successive nucleophilic aromatic substitution of the chlorine atoms of 2,4,6,7-
tetrachloropteridine. For the reaction with secondary amines, the positional order of reactivity
was found to be C4 > C7 > C2 > C6. Final structural proof is given by X-ray crystallography.
To unravel structural elements of 7a crucial for the interaction with the target enzyme, the
compound was modified systematically. The impact of the modifications on activity was tested
by evaluating the ability of the compounds to inhibit cAMP hydrolysis by cAMP-specific
phosphodiesterase (PDE4) purified from the solid human large cell lung tumor xenograft
LXFL529. Growth inhibitory properties were determined by in vitro treatment of the respective
cell line LXFL529L using the sulforhodamine B assay (SRB). The results show that for high
activity, the heterocyclic substituent in position 2 of the pteridine ring system requires the
presence of a basic nitrogen in 4′-position, as represented by piperazine.
In tr od u ction
elements of the highly effective PDE4 inhibitor 7a were
modified systematically. The impact of the modifica-
tions was evaluated by testing the ability of the com-
pounds to inhibit cAMP hydrolysis of PDE4 purified
from solid tumor tissue of the human large cell lung
tumor xenograft LXFL529, a tumor which we have
found to express very high PDE activity in both solid
tumor tissue grown in nude mice as well as in cell
culture. About 80% of overall PDE activity of LXFL529
was represented by PDE4; very minor additional activi-
ties (e10%, respectively) could be ascribed to PDE1 and
PDE3 (data not shown). Treatment-induced growth
inhibitory properties were determined with the sulfo-
rhodamine B assay (SRB).
The second messenger cyclic adenosine monophos-
phate (cAMP) is known to play a pivotal role in the
regulation of cell growth and differentiation. Hydrolysis
of cAMP, resulting in the formation of 5′-AMP, is
catalyzed by phosphodiesterases (PDEs), a superfamily
of isoenzymes. According to substrate specificity and
sensitivity to endogenous modulators, at present there
are at least nine different PDE isoenzyme families
known.1 In the last years, enormous progress has been
achieved to elucidate the complex expression patterns
of PDEs in different cell types and tissues. Neverthe-
less, only limited information is available concerning
PDE patterns of malignant cells. In the murine mela-
noma B16, the murine spindle cell carcinoma CarB, and
the human mammary carcinoma MCF-7, the cAMP-
specific rolipram-sensitive isoenzyme family PDE4 has
been shown by us to represent the highest cAMP
hydrolyzing activity.2,3 Treatment of these cell lines
with the specific PDE4 inhibitor 7-benzylamino-6-chloro-
2-piperazino-4-pyrrolidinopteridine (7a ) induces dose-
dependent growth inhibition.2,3
Ch em istr y
The synthesis of 7a and its 4,7-bis(dimethylamino)
analogue 7f has been described already in patents4,5
without addressing, however, the formation of positional
isomers generated when the described procedure is
followed. The present paper describes an approach to
obtain alkylamino substituted chloropteridines free of
positional isomers. Since with some of these compounds
1
Human PDE4 is encoded by four different genes, the
catalytic domain being highly conserved. Since crystal
structures of PDE4 and PDE4 inhibitor complexes are
not yet available, knowledge about structural require-
ments for interaction of potential inhibitors with the
enzyme is limited. To get more insight into potential
sites of interaction with the target enzyme, structural
unequivocal structural proof was not achievable by H
and 13C NMR spectroscopy, X-ray crystallography was
used for structure characterization of two pivotal rep-
resentatives.
The synthesis of alkylamino substituted pteridines is
based on the successive nucleophilic aromatic substitu-
tion of the chlorine atoms of 2,4,6,7-tetrachloropteridine
at different temperatures and in various solvents.
Following the synthesis instructions of the patent
literature,4,5 the sequence of substitutions is believed
to occur according to Scheme 1.
* To whom correspondence should be addressed.
† Department of Chemistry, Division of Food Chemistry and Envi-
ronmental Toxicology.
‡ Department of Chemistry, Division of Inorganic Chemistry.
10.1021/jm981021v CCC: $15.00 © 1998 American Chemical Society
Published on Web 11/04/1998

4734 J ournal of Medicinal Chemistry, 1998, Vol. 41, No. 24
Merz et al.
Sch em e 1a
by fractionated crystallization. Hence the product
mixture of the first substitution also contained 7-pyr-
rolidino-2,4,6-trichloropteridine in substantial amounts.
As a consequence of hydrolysis during the chromato-
graphic process, this product could, however, not be
isolated. The mass spectrum revealed the signal of a
molecular ion showing the presence of a pyrrolidino, two
chloro, and one hydroxyl substituents (m/e 285 M+,
typical chlorine isotope distribution).
The preparation of 4,7-bis(dimethylamino)-2,6-dichloro-
pteridine (6f) was accomplished by three routes (Scheme
4): (i) The reaction of 4 with an aqueous solution of
dimethylamine at 0 °C in aqueous hydrogen carbonate/
chloroform gave a mixture, from which 4-(dimethyl-
amino)-2,6,7-trichloropteridine (5e) and 6f could be
separated by column chromatography on silica gel. A
compound analogous to 5z could not be detected. (ii)
The reaction of 4 with dimethylamine in ether at 0 °C
provided 6f as the main product, and only traces of 5e
were detected by TLC. In contrast to our findings, these
reaction conditions had been reported earlier to afford
6,7-bis(dimethylamino)-2,4-dichloropteridine.9 (iii) 6f
was also prepared by reaction of 5e with a solution of
dimethylamine in THF at room temperature.
a
Reagents and conditions: (i) pyrrolidine, 1 M KHCO3, CHCl3,
0 °C; (ii) benzylamine, dioxane, 25 °C; (iii) piperazine, dioxane,
triethylamine, 100 °C.
Sch em e 2a
Reaction of 5a and 5e with benzylamine and 5a with
dimethylamine at 20 °C exclusively afforded 4,7-di-
(subst)amino-2,6-dichloropteridines (6a ,d ,e) as products
(Scheme 1, step ii). Structures, formulas, and melting
points of the intermediates are listed in Table 1.
The substitution of the third chlorine atom was
accomplished by reaction at 100 °C under atmospheric
pressure to provide 2,4,7-tri(subst)amino-6-chloropteri-
dines (7 a -l) (Scheme 1, step iii). The generated HCl
was scavenged either by triethylamine or by adding the
reacting amine in excess.
Firm proof concerning the stepwise alkylaminolysis
of chloropteridines is not available from the literature,9-12
and therefore only a putative positional order of reactiv-
ity has been ascribed to the chloro substituents toward
aminolysis until now.13
a
Reagents and conditions: (i) 1. NaOEt, EtOH, 4 h reflux, 2.
HOAc; (ii) 1. NaNO2, 2. Na2S2O4, 3. HCl; (iii) oxalic acid, bath
temperature 240 °C; (iv) PCl5/POCl3, reflux 16 h.
2,4,6,7-Tetrachloropteridine (4) is the key intermedi-
ate for the preparation of these compounds. It was
prepared as shown in Scheme 2: Condensation of
ethylcyanoacetate with urea gives 6-aminouracil (1), and
subsequent nitrosation and reduction yields 5,6-diamino-
uracil hydrochloride (2).6 Melting condensation of 2
with oxalic acid dihydrate produces 2,4,6,7-tetrahydroxy-
pteridine (3).7 Chlorination of 3 with a mixture of
phosphorus pentachloride and phosphorus oxychloride
affords 2,4,6,7-tetrachloropteridine (4) in a moderate
yield.8
The reaction of 4 with pyrrolidine at 0 °C in aqueous
hydrogen carbonate/chloroform resulted in a mixture of
products as summarized in Scheme 3. Separation by
column chromatography on silica gel (ethyl acetate/
hexane ) 3/7) provided 4-pyrrolidino-2,6,7-trichloro-
pteridine (5a ) (28%), as has been described in the patent
literature. However, in substantial yields, 2-pyrroli-
dino-4,6,7-trichloropteridine (5z, 29%) and 4,7-bis(pyr-
rolidino)-2,6-dichloropteridine (6c, 16%) were also present
in the mixture. 5z and 6c were separated by column
chromatography on silica gel (100% dichloromethane).
The reaction of the unseparated raw reaction mixture
with benzylamine at room temperature provided equal
amounts of 6a and 6b, from which 7-pyrrolidino-4-
benzylamino-2,6-dichloropteridine (6b) could be isolated
In view of the absence of hydrogen atoms at the
heterocyclic ring system with exception of 7o and also
of splittings due to long-range C-H couplings in the 13
C
NMR spectra, confirmation of the positional isomers was
done by X-ray crystallography. The most convenient
way would have been to determine the structures of 7a
and 7b, because these structures would also provide
direct proof of the configuration of 6a , 6b, and 5a ,
respectively. We succeeded in crystallizing 7b as a bis-
(hydroperchlorate) hydrate and in determining its struc-
ture. Figure 1 shows the structure of the dication of
this salt. Unfortunately, 7a failed to crystallize in a
couple of solvents and by several methods. Alterna-
tively, 6a was crystallized from ethanol as a hemisolvate
(Figure 2). Due to positional disorder of the solvent
molecules in the crystal lattice and due to conforma-
tional disorder of the pyrrolidino group, the results of
the structure determination are of limited quality but
give an unambiguous proof of all relevant structural
features. The experimental evidence from the reaction
of 6b to 7b with its proven structure allows the
conclusion that under the same reaction conditions the
2-chloro substituent reacts, resulting in the conversion
of 6a to 7a .

Synthesis of Potent Inhibitors of PDE4
J ournal of Medicinal Chemistry, 1998, Vol. 41, No. 24 4735
Sch em e 3a
a
b
Reagents and conditions: (i) pyrrolidine, 1 M KHCO3, CHCl3, 0 °C; (ii) benzylamine, dioxane, 25 °C. Calculated as difference to
100%.
Sch em e 4a
found for the formation of 2-(dimethylamino)-4,6,7-
trichloropteridine. The 1H NMR spectrum of 5e (Figure
4, i), in analogy to 5a , shows two sharp singlets for the
methyl protons, indicating steric fixation of the sub-
stituent in 4-position. The introduction of the second
dimethylamino group in 7-position causes an attenua-
1
tion of the C4-Npyrrolidine bond strength. The H NMR
spectrum of 6f (Figure 4, ii) shows a very broad signal
for the methyl protons of the substituent in 4-position
and one sharp singlet for the methyl protons of the
substituent in 7-position, indicative for slow and fast
rotation around the pteridine-substituent axis. The
substitution of the chlorine atom in 7-position of 5a by
benzylamine is expected to reduce the π-electron defi-
ciency of the pteridine system, too. Remarkably, how-
ever, the pattern and shape of the signals of the
pyrrolidine protons in the 1H NMR spectrum of 6a were
unchanged, although a slight upfield shift became
apparent (Figure 5, i). This may be reconciled with a
smaller pKa value of benzylamine (pKa ) 9.37).15
Replacement of the chloro substituent in 2-positon by a
third amino substituent (7a ) has a stronger impact on
the electron deficiency in the pyrimidine part of the
molecule. This results in attenuation of the C4-
Npyrrolidine bond strength, enabling the pyrrolidine sub-
stituent to rotate, as indicated by the broad peaks at
1.90, 3.73, and 3.95 ppm in the 1H NMR spectrum
(Figure 5, ii).
a
Reagents and conditions: (i) dimethylamine, 40% in H2O, 1
M KHCO3, CHCl3, 0 °C; (ii) dimethylamine, 2 M in THF, Et2O, 0
°C; (iii) dimethylamine, 2 M in THF, dioxane, 25 °C.
Further insight into the reactivity toward nucleophilic
attack of 2,4,6,7-tetrachloropteridine is given by the 1H
NMR spectra, which are indicative for the strength of
the bonds at position 2 and 4 of the pteridine ring
formed by reacting with pyrrolidine (pKa ) 11.31)14 and
dimethylamine (pKa ) 10.72).15 The H NMR spectra
1
of 5a and 5z (Figure 3) show that the C4-Npyrrolidine bond
of 5a has almost double-bond character. At room
temperature, a rotation around this bond does not take
place, as indicated by distinct triplets and quintets for
the 2′- and 3′-protons (Figure 3, ii). In 5z, slow rotation
around the pteridine-substituent axis is allowed at
room temperature, giving rise to the broad peaks in the
1H NMR spectrum (Figure 3, i).
The isomeric dimethylamino compounds 7m and 7n
were prepared by the reaction of gaseous ammonia with
a suspension of 6a in dimethylformamide in a sealed
Substitution with dimethylamine, a weaker nucleo-
phile, exclusively gave 5e and 6f. No indication was

4736 J ournal of Medicinal Chemistry, 1998, Vol. 41, No. 24
Merz et al.
Ta ble 1. Structure, Formulas, and Melting Points of Di- and Trichloropteridines
compd
R1
R2
R3
formula
analysesa
mp (°C)
5a
5e
5z
6a
6b
6c
6d
6e
6f
pyrrolidino
dimethylamino
Cl
pyrrolidino
benzylamino
pyrrolidino
pyrrolidino
dimethylamino
dimethylamino
Cl
Cl
Cl
Cl
Cl
C
10H8Cl3N5
C, H, N
C, H, N
C, H, N
C, H, N
C, H, N
C, H, N
C, H, N
C, H, N
C, H, N
245
225
182
152
170
169-170
239
C8H6Cl3N5
C10H8Cl3N5
pyrrolidino
benzylamino
pyrrolidino
pyrrolidino
dimethylamino
benzylamino
dimethylamino
Cl
Cl
Cl
Cl
Cl
Cl
C
C
C
C
C
C
17H16Cl2N6
17H16Cl2N6
14H16Cl2N6
12H14Cl2N6
15H14Cl2N6
10H12Cl2N6
225-227
244
a
All values were within (0.4% of the calculated values.
F igu r e 1. Diagram of the dication in 7b‚2HClO4‚H2O. Dis-
placement ellipsoids are drawn at the 50% propability level,
radii of hydrogen atoms are chosen arbitrarily, and the
hydrogen atom labels are omitted for clarity. Note that a 1:1
conformational disorder of the pyrrolidino group in the crystal
was found. Only one of the conformations is drawn.
F igu r e 3. Section (1.7-4.6 ppm) of the 1H NMR spectra at
293 K of (i) 5z and (ii) 5a in CDCl3.
atoms, indicating the presence of the dimethylamino
substituent in 6-position.
The hydrogenation of 7a in the presence of 10%
palladium on carbon under atmospheric pressure af-
forded 7o within 2 h at room temperature.
F igu r e 2. Diagram of 6a in 6a ‚0.5EtOH (displacement
ellipsoids and further details as described for Figure 1). As
for 7b, a 1:1 conformational disorder of the pyrrolidino group
in the crystal was found. Only one of the conformations is
drawn.
In summary, these results demonstrate that the
reactivity of the carbon atoms in 2,4,6,7-tetrachloro-
pteridine for nucleophilic aromatic substitution with
secondary amines follows the order of C4 > C7 > C2 >
C6. Using the synthetic procedure as described in the
patent literature,5 the first substitution step produces
not only 5a but also 5z and 6c in substantial amounts.
We found that chromatographic isolation of these prod-
ucts is mandatory for the consecutive synthesis of
alkylamino substituted (chloro)pteridines free of posi-
tional isomers.
glass tube at 100 °C and 12 bar and subsequent isolation
by column chromatography. No indication for the
formation of 2- or 6-NH2 substituted compounds was
found. Obviously under these conditions DMF serves
as donor for dimethylamine, the formyl group of DMF
presumably ending up as formamide. The structure of
7m was established by the shifts of the C atoms in the
13C NMR spectrum, showing the same pattern as 7a .
In contrast, the 13C NMR spectrum of 7n exhibits a
different set of resonances for the pteridine carbon

Synthesis of Potent Inhibitors of PDE4
J ournal of Medicinal Chemistry, 1998, Vol. 41, No. 24 4737
F igu r e 5. Section (1.5-5.5 ppm) of the 1H NMR spectra at
293 K of (i) 6a in CDCl3 and (ii) 7a in DMSO-d6.
F igu r e 4. Section (3.0-4.2 ppm) of the 1H NMR spectra at
293 K of (i) 5e in CDCl3 and (ii) 6f in CDCl3.
SRB assay (Table 2). The IC50 values for growth
inhibition in cell culture exceed the IC50 values for PDE4
inhibition approximately by a factor of 150. This leaves
the question whether PDE4 is the only relevant target
for the growth inhibitory effects of 7a . In previous
studies we have shown in different tumor cell models
that treatment of tumor cells with 7a efficiently inhibits
the intracellular PDE4 activity, resulting in a dose-
dependent increase of the intracellular cAMP level and
in growth inhibition with IC50 values in the same order
of magnitude.2,3 Thus, the potency of the compound to
inhibit isolated PDE4 from tumor tissue is different
from its potential to inhibit intracellular PDE4 activity
as measured by incubation of tumor cells. However,
comparison of PDE4 isolated from solid tumors of
LXFL529 with PDE4 from the respective cell line
reveals no difference in sensitivity to 7a (data not
shown). The discrepancy between inhibition of isolated
PDE4 and growth inhibition of tumor cells might be
ascribed to subcellular localization of the drug which
we found to concentrate to a great extent in cellular
compartments other than the cytosol. It is, however,
the cytosol where more than 80% of total PDE4 of
LXFL529L cells is localized.18 Moreover, cellular targets
other than PDE4 might also be affected by 7a which
might be of influence to the observed biological effects.
Biology
Compounds listed in Table 2 were tested for their
ability to inhibit cAMP hydrolysis of purified PDE4
protein isolated from the cytosol of solid tumor tissue
of the human large cell lung tumor xenograft LXFL529,
grown in nude mice. The isolated protein was cAMP-
specific, showing no significant cGMP hydrolysis at a
nucleotide concentration of 1 µM, and exhibited high
sensitivity to the selective PDE4 inhibitor rolipram (IC50
) 0.1 ( 0.06 µM; n ) 3), characteristic for isoenzymes
of the PDE4 family. No significant changes of the cAMP
hydrolysis rate were observed in the presence of cGMP
or Ca2+/calmodulin, thus proving the absence of a
contamination with PDE1, PDE2, or PDE3 isoenzymes.
SDS-PAGE and subsequent silver staining of the
isolated protein revealed only one band at approxi-
mately 93 kDa. Western blotting with a monoclonal
antibody for the PDE4D subfamily showed an immuno-
reactive band at the same molecular weight (data not
shown). Therefore, the purified protein belongs to the
PDE4D subfamily. A member of the PDE4D subfamily
with the respective molecular weight has been reported
as PDE4D3.16,17
Resu lts a n d Discu ssion
Str u ctu r e-Activity of Su bstitu ted P ter id in es.
(a ) P osition 4 a n d 7 (F igu r e 6). Replacement of the
benzylamino substituent in 7-position by pyrrolidine
(7c) results in a reduction of PDE4 inhibitory properties
almost by a factor of 4. A dimethylamino substituent
in 7-position (7d ) provides approximately the same
attenuating effect. Concomitantly, the growth inhibi-
tory potential of 7c and 7d is slightly decreased com-
7a is a potent inhibitor of PDE4 isolated from solid
tumors of the human large cell lung tumor xenograft
LXFL529. Concomitantly, growth inhibition of the
respective cell line is induced in the low micromolar
range (Table 2). However, there is a discrepancy
between the potency of the compound to inhibit isolated
PDE4 and the growth inhibitory concentrations in the

4738 J ournal of Medicinal Chemistry, 1998, Vol. 41, No. 24
Merz et al.
Ta ble 2. Structure, Formulas, Melting Points, Tumor Cell Growth Inhibition Data, and PDE4 Inhibitory Properties of Tested
Compounds
growth inhibition
PDE4 inhibition
compd
R1
R2
R3
R4
formulaa
C21H25ClN8
mp (°C)
IC50 (µM)b
IC50 (µM)c
7a
7b
7c
7d
7e
7f
7g
7h
7i
7j
7k
7l
7m
7n
7o
i
ii
i
ii
i
i
iii
iii
iii
iii
iii
iii
-Cl
-Cl
-Cl
-Cl
-Cl
-Cl
-Cl
-Cl
-Cl
-Cl
-Cl
-Cl
-Cl
iv
192-194
170-172
194
146-148
138-140
131-132
204
2.3 ( 0.3
4.0 ( 0.2
3.1 ( 0.1
5.8 ( 0.4
5.9 ( 0.6
>20
0.016 ( 0.005 (4)
0.083 ( 0.037 (2)
0.059 ( 0.005 (2)
0.046 ( 0.006 (3)
0.30 ( 0.14 (2)
1.2 ( 0.5 (3)
C
C
C
C
C
C
C
C
C
C
C
C
C
C
21H25ClN8
18H25ClN8‚0.5H2O
16H23ClN8
i
iv
ii
iv
ii
ii
ii
ii
ii
ii
ii
ii
ii
iv
iv
i
i
i
i
i
i
i
19H23ClN8‚0.5H2O
14H21ClN8‚0.5H2O
22H27ClN8‚0.5H2Od
21H24ClN7O‚0.5H2O
23H27ClN8O
4-methylpiperazino
morpholino
6.5 ( 0.5
>20
3.1 ( 0.7 (3)
186
2.5 ( 0.4 (3)
4-acetylpiperazino
3-oxopiperazino
2-aminoethylamino
2-hydroxyethylamino
iv
146-148
13.0 ( 1.1
9.5 ( 2.2
3.2 ( 0.4
15.2 ( 1.4
11.8 ( 0.8
>20
5.9 ( 1.1 (2)
21H23ClN8Oe
19H23ClN8
19H22ClN7O
19H22ClN7
19H22ClN7
>250
4.2 ( 0.5 (3)
177-179
198
3.7 ( 0.5 (3)
2.8 ( 0.1 (3)
196
187
218
1.8 ( 1.2 (2)
i
i
-Cl
iii
0.65 ( 0.05 (2)
0.36 ( 0.09 (3)
-H
21H26N8
7.3 ( 0.7
a
b
Analyses for C, H, and N were within (0.4% of the expected values for the formula unless otherwise noted. Growth inhibition of
LXFL529L cells was determined using the sulforhodamine B assay, incubation time 3 d. IC50 values were calculated as survival of treated
cells over control cells × 100 [T/C %]; values are given as mean ( SEM of two separate assays, each done in quadruplicate. c PDE4 was
isolated from solid tumor tissue of the human large cell lung tumor xenograft LXFL529, grown in nude mice. Inhibition of cAMP hydrolysis
was determined in the presence of 1 µM cAMP according to the method of Po¨ch.20 IC50 values were determined from the logarithmic
concentration-inhibition curve (at least three points). Values are given as mean ( SEM, number of separate experiments in parentheses,
each done in triplicate. C: calcd, 59.0; found, 59.5. e N: calcd, 25.5; found, 24.9.
d
dimethylamine (7f), PDE4 inhibitory properties are
reduced by a factor of 75, compared to 7a . The latter
modification also totally abolishes growth inhibitory
effectiveness.
Since compounds 7b-e are still able to induce growth
inhibition, albeit less effective than 7a , it appears that
space-filling substitutents larger than dimethylamine
are required for optimum activity.
(b) P osition 6. Removal of the chloro substituent
in 6-position, which might potentially be involved in H
bond formation, causes a dramatic loss in PDE4 inhibi-
tion, the resulting analogue (7o) being about 20-fold less
F igu r e 6. Inhibition of PDE4 activity isolated from solid
tumor tissue of LXFL529 by 7a and derivatives with modifica-
tions in 4- and/or 7-position of the pteridine ring system.
Inhibition of cAMP hydrolysis was determined in the presence
of 1 µM cAMP according to the method of Po¨ch.20 Values are
given as mean ( SD of at least two separate experiments, each
done in triplicate.
effective than 7a (Figure 6). However, growth inhibi-
tory properties are reduced only by a factor of 3 (Table
2). This suggests that 7o might influence also other
cellular targets potentially relevant for growth.
(c) P osition 2. Modifications at position 2 of the
pteridine ring system (7g-m ) bring about the highest
impact on PDE4 inhibition (Figures 7 and 8). The
substituent in position 2 of 7a contains a strong basic
center at N4′, representing also a hydrogen bond donor.
N4′ methylation removes the hydrogen bond donor (7g).
This results in a 200-fold reduction of PDE4 inhibitory
properties, compared to those of 7a . Introduction of an
oxogroup in C3′-position (7j) removes the basicity of N4′
but preserves H bond donor qualities, showing slightly
reduced PDE4 inhibition compared to that of 7g.
Acetylation of N4′ (7i) eliminates both basicity of N4′
and H bond donor abilities, resulting in a further loss
of PDE4 inhibitory properties, 7i being 400-fold less
potent than 7a . Complete removal of the basic center
and of H bond donor activity at position 4′ is achieved
by exchange of piperazine versus dimethylamine (7m ).
pared to that of 7a . Interchange of substituents at
positions 4 and 7 of 7a induces a 5-fold decrease in
PDE4 inhibition (7b ). The comparison of 7b and 7c
shows that exchange of the pyrrolidine residue in
4-position versus exchange of an aromatic group has no
significant impact on PDE4 inhibitory activity. Deriva-
tives 7b, 7c, and 7d are similar in their PDE4 inhibitory
properties and are somewhat less potent compared to
7a .
Replacement of pyrrolidine in 4-position by dimethyl-
amine (7e) results in a stronger attenuation of PDE4
inhibition. 7e is almost 20-fold less inhibitory toward
PDE4 than 7a and about 6-fold less inhibitory than 7d .
Both compounds induce, however, cell growth inhibition
to the same extent. When both positions 4 and 7 carry

Synthesis of Potent Inhibitors of PDE4
J ournal of Medicinal Chemistry, 1998, Vol. 41, No. 24 4739
group (7l) has no impact on PDE4 inhibition but
considerable influence on tumor cell growth inhibition,
7l exhibiting only 20% of the activity of 7k .
Interchanging the dimethylamino substituent in 2-po-
sition and the chloro substituent in 6-position leads to
a molecule (7n ) with different spatial requirements,
compared to those of 7m . Interestingly, 7n exhibited
3-fold higher potency to inhibit PDE4 compared to 7m
but showed no growth inhibitory properties. Since the
2-chloro substituent in 7n shows a relatively high
reactivity toward nucleophilic substitution, a reaction
of 7n with nucleophiles from the cell culture medium
used in the SRB assay is believed to be the reason for
the observed loss of growth inhibitory activity.
In summary, reduction of PDE4 inhibitory potency
always induces a decrease of the growth inhibitory
properties. The most potent compounds, 7a -d , exhibit
some degree of correlation between these two param-
eters. Regarding the derivates 7e-o, which are less
potent with respect to PDE4 inhibition by one or 2
orders of magnitude when compared to 7a , the results
of the SRB assay do not reflect variations in PDE4
inhibitory properties. In addition to potential differ-
ences in cellular uptake and distribution, some of the
molecular modifications apparently inflict on cellular
targets other than PDE4, and this might result in
differential influence on growth.
F igu r e 7. Inhibition of PDE4 activity isolated from solid
tumor tissue of LXFL529 by 7a and derivatives with modifica-
tions at the piperazino residue. Inhibition of cAMP hydrolysis
was determined in the presence of 1 µM cAMP according to
the method of Po¨ch.20 Values are given as mean ( SD of at
least two separate experiments, each done in triplicate.
Con clu sion
In this study, we have elucidated the order of reactiv-
ity of pteridine C atoms toward secondary amines and
have established a synthetic route that allows the
preparation of alkylamino-substituted pteridines free of
positional isomers. The impact of structural elements
necessary for PDE4 inhibition has been investigated.
Changes in 7-position (7c,d ) had minor impact on the
inhibitory properties, compared to modifications in the
other pteridine ring positions. Reduction of the size of
the substituent in 4-position (7e) induces a significant
reduction of PDE4 inhibition. Exchange of the chloro
substituent versus hydrogen (7o) diminishes PDE4
inhibition by the factor of 20. The highest impact is
found for modifications at position 2 of the pteridine ring
system, where the parent compound carries a piperazino
residue. Potent PDE4 inhibitors are also found to be
potent inhibitors of tumor cell growth. However, with
compounds exhibiting significantly diminished PDE4
inhibitory properties, no correlation between PDE4 and
growth inhibition can be observed any longer. This
indicates that inhibition or interaction with cellular
targets different from PDE4 might become relevant for
tumor cell growth inhibition displayed by such deriva-
tives.
F igu r e 8. Inhibition of PDE4 activity isolated from solid
tumor tissue of LXFL529 by 7a and derivatives with modifica-
tions in 2-position of the pteridine ring system. Inhibition of
cAMP hydrolysis was determined in the presence of 1 µM
cAMP according to the method of Po¨ch.20 Values are given as
mean ( SD of at least two separate experiments, each done
in triplicate.
This modification induces a strong reduction of PDE4
inhibitory properties, similar to those of 7g. 7m is
slightly more potent than 7i, which might be due to the
different spatial requirements when interacting with the
target enzyme.
The growth inhibitory efficacy of 7g is about one-third
that of 7a on the basis of IC50 values. With respect to
PDE4 inhibition, 7g displays about 0.5% of 7a and 10%
of 7e, whereas the IC50 values for growth inhibition of
the latter two derivatives are very similar. Thus 7g
might affect other cellular targets.
Replacement of the H bond donor piperazine (7a ) by
the H bond acceptor morpholine (7h ) also induces a
strong loss of PDE4 inhibitory capacity, comparable to
7g. But in contrast to 7g, 7h exhibits no growth
inhibitory activity anymore. Apparently, the H bond
donor function has a greater impact on cell growth than
on PDE4 inhibition.
Substitution of the piperazine by 2-amino-ethylamine
(7k ) introduces a substituent with enhanced possibility
of free rotation with the basic center in approximately
the same distance to the pteridine ring system as the
piperazine residue. PDE4 inhibitory property of 7k is
comparable to that of 7g-j, the other derivatives with
modified 4′-position. In terms of growth inhibition
however, 7k is very close to 7a . Replacing the 2-amino-
ethylamino substituent in 7k by a 2-hydroxyethylamino
Exp er im en ta l Section
Biology. Ma ter ia ls. [3H]-cAMP (specific activity 32.9 Ci/
mmol) and [3H]-cGMP (specific activity 8.5 Ci/mmol) were
obtained from E. I. du Pont de Nemours & Co., (Inc). Column
HR 10/2, Q-Sepharose FF gel suspension and gel filtration
column Hiload 16/60 Superdex 200 were purchased from
Pharmacia (Freiburg, Germany). Rolipram (4-[3-cyclopentyl-
oxy-4-methoxyphenyl]-2-pyrrolidone) was kindly provided by
Schering AG (Berlin, Germany).
Hu m a n Xen ogr a ft Tu m or Tissu e. Human tumors es-
tablished by serial passage in nude mice (NMRI genetic
background) were used for obtaining tumor xenograft material.

4740 J ournal of Medicinal Chemistry, 1998, Vol. 41, No. 24
Merz et al.
The tumors have been described and characterized in detail.19
Tumors were implanted subcutaneously into both flanks of the
animals. When a median diameter of 10-15 mm was reached,
the mouse-grown tumors were resected under sterile condi-
tions and frozen in liquid nitrogen immediately.19 The non-
small cell lung tumor xenograft LXFL529 was established from
a primary tumor of a 34 year old female without receiving
chemotherapy prior to tumor resection. The survival time of
the patient after resection was 115 days.
Cytosol P r ep a r a tion . All procedures were performed at
4 °C. Tumor tissue was homogenized in homogenization buffer
(50 mM Tris/HCl, pH 7.4, 10 mM MgCl2, 0.1 mM EDTA, 0.1
mM EGTA, 4 mM benzamidine hydrochloride, 0.5 µM trypsin
inhibitor (soy beans), 0.1 mM phenylmethylsulfonyl fluoride,
1 mM â-mercaptoethanol, 0.1 mM N-R-p-tosyl-L-lysine chloro-
methyl ketone, 1 µM pepstatin, 1 µM leupeptin). After
centrifugation (100000g, 1 h) the supernatant was carefully
removed, pooled, and applied to anion exchange chromatog-
raphy.
points were determined on a Bu¨chi melting point apparatus
and are uncorrected. Elemental analyses were performed
using a Perkin-Elmer 2400 CHN elemental analyzer. Unless
1
otherwise indicated, H NMR spectra were recorded at room
temperature at 400 MHZ on a Bruker AMX 400 with tetra-
methylsilane, DMSO (δ 2.49), or CDCl3 (δ 7.26) used as
internal standard. 13C NMR spectra were obtained at 100
MHZ using tetramethylsilane, CDCl3 (δ 77.0), or DMSO (δ
39.5) as internal standard. J values are reported in hertz.
Apparent multiplicities are designated as s, singlet; d, doublet;
t, triplet; q, quartet; qu, quintet; m, multiplet; b, broad. Mass
spectra were taken in the positive ion mode under electron
impact (EI), 70 eV.
2,4,6,7-Tetr a ch lor op ter id in e (4). This compound was
prepared as described by Scho¨pf.8 For most preparations, it
is not necessary to sublimate the product in vacuo, as during
the next reaction step a purification by column chromatogra-
phy is required: 13C NMR (CHCl3) δ 129.8, 149.2, 153.1, 156.1,
160.2, 165.1. Anal. (C6Cl4N4) C, H, N.
4-P yr r olid in o-2,6,7-tr ich lor op ter id in e (5a ). A suspen-
sion of 4 (1.23 g, 4.56 mmol) in 40 mL of CHCl3 was cooled to
0 °C and mixed with a solution of 912 mg of KHCO3 (9.12
mmol) in 10 mL of water. A precooled solution of 325 mg of
pyrrolidine (4.56 mmol) in 10 mL of CHCl3 was added slowly,
and the reaction mixture was stirred at 0 °C for 1 h. Water
(40 mL) was added, and the organic layer was separated, dried
over Na2SO4, and concentrated in vacuo to afford a yellow-
brownish residue (921 mg). Column chromatography (ethyl
acetate/hexane ) 3/7 v/v) provided a uniform fraction of 416
mg (45.2%) with Rf ) 0.30 and 259 mg (28.1%) of 5a with Rf
) 0.54 as pale-yellow crystals; mp 245 °C: 1H NMR (CDCl3) δ
An ion Exch a n ge Ch r om a togr a p h y. The supernatant
was loaded on a Q Sepharose column (HR 10/2, Pharmacia),
equilibrated with buffer B (50 mM Tris/HCl, pH 7.4, 10 mM
MgCl2, 0.1 mM EDTA, 4 mM benzamidine hydrochloride, 0.5
µM trypsin inhibitor (soy beans), 0.1 mM phenylmethylsulfonyl
fluoride, 1 mM â-mercaptoethanol, 0.1 mM N-R-p-tosyl-L-lysine
chloromethyl ketone, 1 µM pepstatin, 1 µM leupeptin). After
loading, the column was washed with 20 mL of buffer B, flow
rate 2 mL/min. PDE activity was eluted at the same flow rate
with buffer A containing 1 M NaCl using a step gradient.
Fractions of 5 mL were collected and assayed for PDE activity.
Fractions containing rolipram-sensitive PDE activity were
concentrated by ultrafiltration (Amicon chamber 8010, mem-
brane YM3) under nitrogen to a volume of 1 mL. The filtration
step was performed at 4 °C. The resulting samples were
immediately applied to gel filtration.
3
3
3
2.02 (qu, J ) 6.8 Hz, 2), 2.16 (qu, J ) 6.8 Hz, 2), 3.88 (t, J
3
) 6.9 Hz, 2), 4.24 (t, J ) 6.9 Hz, 2); 13C NMR (CDCl3) δ 23.6,
26.9, 51.1, 51.4, 124.5, 141.3, 151.4, 154.5, 157.3, 161.8; MS
m/e 303 (M+). Anal. (C10H8Cl3N5) C, H, N.
4-Dim eth yla m in o-2,6,7-tr ich lor op ter id in e (5e) was pre-
pared in the same manner as 5a . Starting with 1.5 g of 4 the
reaction was performed using a 40% aqueous solution of
dimethylamine to give 967 mg of raw product. Column
chromatography yielded 294 mg (30.4%) of 5e, light-yellow fine
crystals; Rf ) 0.52 (ethyl acetate/hexane ) 1/3 v/v), mp 225
°C: 1H NMR (CDCl3) δ 3.45 (s, 3), 3.82 (s, 3); 13C NMR (CDCl3)
δ 41.7, 42.3, 124.5, 140.6, 151.3, 154.8, 159.1, 161.3; MS m/e
277 (M+). Anal. (C8H6Cl3N5) C, H, N.
2-P yr r olid in o-4,6,7-tr ich lor op ter id in e (5z). The frac-
tion with Rf ) 0.30 from the preparation of 5a was submitted
to a further separation by column chromatography on silica
gel (100% dichloromethane) to provide 264 mg (28.7%) of 5z
with Rf ) 0.38. The compound crystallized in bright-yellow
needles; mp 182 °C: 1H NMR (CDCl3) δ 2.08 (br, 4), 3.89 (br,
2), 4.22 (br, 2); 13C NMR (CDCl3) δ 24.0, 27.3, 51.8, 124.4,
138.8, 152.6, 155.6, 158.8, 161.1; MS m/e 303 (M+). Anal.
(C10H8Cl3N5) C, H, N.
Gel F iltr a tion . The column was equilibrated with two bed
volumes of buffer C (50 mM Tris/HCl, pH 7.4, 0.1 mM EDTA,
10 mM MgCl2, 0.3 mM glycerol, 0.5 mM NaCl). Gel filtration
was performed at a flow rate of 0.5 mL/min. Fractions of 2
mL were collected and immediately assayed for PDE activity
and sensitivity to rolipram.
P h osp h od iester a se Assa y. PDE activity was measured
as described previously.2,20 Briefly, PDE-containing prepara-
tions were incubated with [3H]-cAMP in buffer A, containing
50 mM Tris/HCl, pH 7.4, 10 mM MgCl2, and 1 mM AMP. The
final cAMP concentration in the assay was 1 µM. Incubations
were routinely carried out in triplicate at 37 °C. Reaction was
stopped by adding ZnSO4. [3H]-5′-AMP was precipitated by
addition of Ba(OH)2 and centrifuged at 10000g. Nonhydro-
lyzed [3H]-cAMP was determinated by liquid scintillation
counting of the supernatant. Calculation of PDE activity was
carried out with reference to controls.
Su lfor h od a m in B Assa y. Cells were cultivated in humi-
fied incubators (37 °C, 5% CO2). LXFL529 cells were grown
in RPMI 1640, with additional glutamine (300 mg/L), 1%
penicillin/streptomycin, and 10% newborn calf serum. The
cells were free from mycoplasm contamination, as tested
routinely. Cells were seeded in 24-well plates and allowed to
grow 24 h before treatment. The tests were performed by
continuous drug incubation. Cytotoxicity was determined as
described previously21 and calculated as survival of treated
cells over control cells × 100 [% T/C].
7-Ben zylam in o-2,6-dich lor o-4-pyr r olidin opter idin e (6a).
A suspension of 964 mg (3.17 mmol) of 5a in 25 mL of dioxane
was stirred for 1 h with 488 µL (3.48 mmol) of triethylamine
and 380 µL of (3.48 mmol) benzylamine at room temperature.
The reaction mixture was evaporated almost to dryness and
thoroughly washed with 50 mL of water, filtered, and dried
over phosphorus pentoxide to afford 1083 mg (91%) of 6a ; mp
152 °C. 1H NMR (CDCl3) δ 1.93 (qu, 3J ) 6.8 Hz, 2), 2.07 (qu,
3J ) 6.8 Hz, 2), 3.80 (t, 3J ) 6.8 Hz, 2), 4.12 (t, 3J ) 6.8 Hz, 2),
3
3
4.81 (d, J ) 5.3 Hz, 2), 6.05 (t, J ) 5.3 Hz, 1), 7.28-7.38 (m,
5); 13C NMR (CDCl3) δ 23.7, 26.9, 45.6, 50.2, 50.9, 117.3, 128.0,
128.1, 128.3, 128.9, 131.3, 137.2, 151.5, 156.3, 157.2, 160.3;
MS m/e 374 (M+). Anal. (C17H16Cl2N6) C, H, N.
Ch em istr y. Except for dimethylamine, amines were dis-
tilled from potassium hydroxide. THF, dioxane, and ether
were distilled over natrium under a dry argon atmosphere.
Other solvents and reagents obtained from commercial sup-
pliers were at least of reagent grade and were used without
further purification. All reactions involving oxygen or moisture-
Cr ysta l Gr ow in g of 6a . Boiling ethanol (0.5 mL) was
added to 5 mL of a solution of 6a in ethanol, saturated at 50
°C. The solution was cooled slowly to room temperature and
was kept at this temperature for 3 months under light
protection. Then the mixture was stored at 4 °C until crystal
growth was completed.
sensitive compounds were performed under
a dry argon
atmosphere. All reaction mixtures and column chromato-
graphic fractions were analyzed by thin layer chromatography
on TLC plates “Alugram Sil G/UV254” purchased from Mach-
erey & Nagel. Column chromatography was carried out using
silica gel 60, Macherey & Nagel, (0.063-0.2 mm). Melting
Cr ysta l Str u ctu r e Deter m in a tion of 6a ‚0.5EtOH. Mr
(C18.5H18.5Cl2N6O0.5) ) 403.79, monoclinic space group C2/c, a

Synthesis of Potent Inhibitors of PDE4
J ournal of Medicinal Chemistry, 1998, Vol. 41, No. 24 4741
) 21.937(4) Å, b ) 8.893(2) Å, c ) 21.586(4) Å, â ) 114.72(3)°,
V ) 3825.2(13) Å3, Z ) 8, Dx )1.402 g cm-3, µ ) 0.358 mm-1
(λ ) 0.71073 Å), T ) 293 K, pale-yellow crystal of dimensions
0.5 mm × 0.3 mm × 0.2 mm in a thin walled glass capillary,
3642 reflection data were collected on an Enraf-Nonius CAD4
diffractometer. Lp corrections were applied. The structure
was solved by direct methods22 and refined (216 parameters)
by full-matrix least-squares calculations on F2,23 using all but
three of the most negative of 1858 unique reflections to final
4,7-Bis(dim eth ylam in o)-2,6-dich lor opter idin e (6f). This
compound was prepared by three ways: (a) during the
preparation of 5e, 416 mg (43.0%) of a second component (6f)
could be isolated from the column with Rf ) 0.25, which
crystallized upon evaporating the solvent; (b) the preparation
analogous to 6d where the reaction of 5e (152 mg, 0.55 mmol)
with 2 M THF solution of dimethylamine (300 µL, 0.6 mmol)
in the presence of triethylamine (83 µL, 0.6 mmol) at room
temperature yielded 142 mg (91%) of 6f; and (c) the prepara-
tion according to Schenker9 where a sample of 4 (270 mg, 1
mmol) was dissolved in 30 mL of ether and cooled to 0 °C.
Precooled 2 M THF solution of dimethylamine (5 mL, 10 mmol)
was added by a syringe, and the mixture was stirred for 1 h.
The reaction mixture was evaporated to dryness and treated
with 25 mL of water to remove dimethylamine hydrochloride.
The yellow residue (225 mg, 78.4%) contained traces of 5e, as
seen by TLC, and was chromatographed on a column (silica
gel, ethyl acetate/hexane ) 1:2 v/v) to afford 190 mg (66.2%)
of 6f; greenish yellow needles, mp 244 °C: 1H NMR (CDCl3)
298 K δ 3.33 (s, 6), 3.53 (br, 6); 223 K δ 3.35 (s, 6), 3.38 (s, 3),
3.73 (s, 3); 13C NMR (CDCl3) δ 41.10, 41.14, 118.0, 130.5, 154.3,
155.0, 158.8, 159.9; MS m/e 286 (M+). Anal. (C10H12Cl2N6) C,
H, N.
2
R1[Fo > 2σ(Fo2)] ) 0.067, wR2 ) 0.198 (all data used for
2
refinement), w ) 1/[σ2(Fo2) + (0.10P)2 + 6.9P] where P ) (Fo
+ 2Fc2)/3, S ) 1.068.23 Largest peak and hole in the final
difference map are 0.257 e/Å3 and -0.395 e/Å2, respectively.
Anisotropic displacement parameters were refined for the
atoms of the central dichloropteridine unit and the benzyl-
amino group. An empirical extinction parameter was refined.23
With two exceptions all hydrogen atoms were included in the
refinement in calculated postions and were allowed to ride on
the atom to which they are attached. The positions of the H
atom at N71 and at the OH group of ethanol were taken from
the difference map and refined, applying distance restraints
only. Depending on the nature of the group, the isotropic
displacement parameters of the H atoms were kept equal to
120%, 130%, and 150% of the equivalent isotropic displacement
parameters of the parent atoms, respectively.
7-Ben zyla m in o-6-ch lor o-2-p ip er a zin o-4-p yr r olid in o-
p ter id in e (7a ). A sample of 1 g (2.66 mmol) of 6a and 920
mg (10.64 mmol) of finely ground piperazine in 30 mL of
dioxane were heated under reflux until complete disappear-
ance of educt, as monitored by TLC (ethyl acetate/hexane )
1/1 v/v). After the solvent was removed in vacuo, the residue
was washed with 50 mL of water and the slurry spun down
at 5000 rpm. The pellet was dissolved in 0.1 N HCl, filtered,
and precipitated by 2% NH3 solution. The solid was collected
by filtration and washed. Drying over KOH provided 983 mg
(87%) of a light-yellow solid; mp 192-194 °C: 1H NMR
4-Ben zylam in o-2,6-dich lor o-7-pyr r olidin opter idin e (6b).
A 1.35 g (5 mmol) sample of 4, purified by sublimation in
vacuo, was used as described for the preparation of 5a to yield
1.13 g (73.5%) of a product mixture. This mixture (3.71 mmol,
calculated as pyrrolidinotrichloropteridine) was suspended in
20 mL of dioxane, and benzylamine (890 µL, 874 mg, 8.16
mmol) was added at 20 °C. The reaction mixture was stirred
for 1 h, the solvent was removed in vacuo, and the yellow
residue was washed with 50 mL of water. Filtration yielded
1.2 g (3.2 mmol, 86.3%) of raw product. Fractionated crystal-
lization in ethanol yielded 330 mg (23.7%) of 6b, which
crystallized first; yellow platelets, mp 170 °C: 1H NMR
3
(DMSO-d6) δ 1.90 (br, 4), 2.69 (t, J ) 4.4 Hz, 4), 3.62 (br, 1),
3
3
3.66 (t, J ) 4.4 Hz, 4), 3.73 (br, 2), 3.95 (br, 2), 4.67 (d, J )
6.1 Hz, 2), 7.23-7.32 (m, 5), 7.98 (t, J ) 6.1 Hz, 1); 13C NMR
3
(DMSO-d6) δ 43.4, 44.6, 45.6, 113.8, 125.5, 126.6, 126.7, 128.2,
139.2, 151.5, 156.3, 156.7, 159.5; MS m/e 424 (M+). Anal.
(C21H25ClN8) C, H, N.
3
(DMSO-d6) δ 1.93 (br, 4), 3.79 (br, 4), 4.64 (d, J ) 6.3 Hz, 2),
7.23-7.35 (m, 5), 9.11 (t, 3J ) 6.3 Hz, 1); 13C NMR (DMSO-d6)
δ 25.0, 43.6, 50.4, 115.1, 126.8, 127.4, 128.2, 132.9, 138.6, 152.5,
153.2, 159.3, 159.6; MS m/e 374 (M+). Anal. (C17H16Cl2N6) C,
H, N.
4-Ben zyla m in o-6-ch lor o-2-p ip er a zin o-7-p yr r olid in o-
p ter id in e (7b). The product mixture from the preparation
of 6b (448 mg, 1.2 mmol) and finely ground piperazine (460
mg, 5.3 mmol) were suspended in 15 mL of dioxane and
refluxed for 1 h. The solvent was removed in vacuo, and the
residue was washed with water and dried to afford 460 mg
(90.4%) of a yellow solid. Repeated column chromatography
(column length 80 cm, silica gel, 2-propanol/triethylamine )
9/1 v/v) provided 24.5% of 7a (Rf ) 0.30) and 25% of 7b (Rf )
0.17), yellow solid, mp 170-172 °C: 1H NMR (DMSO-d6) δ
4,7-Bis-(p yr r r olid in o)-2,6-d ich lor op ter id in e (6c). The
second component of the separation of 5z by column chroma-
tography on silica gel (100% dichloromethane) with Rf ) 0.08
was 6c, which was eluted by adding 10% methanol to the
dichloromethane. On concentrating the solution, the com-
pound crystallized in yellow plates; mp 169-170 °C, yield 150
mg (16.3%): 1H NMR (CDCl3) δ 1.94 (qu, 3J ) 6.9 Hz, 2), 1.98-
3
3
2.01(m, 4H), 2.08 (qu, J ) 6.9 Hz, 2), 3.80 (t, J ) 6.9 Hz, 2),
3
1.88 (br, 4), 2.69 (br, 4), 3.70 (br, 9), 4.67 (d, J ) 5.6 Hz, 2),
3.92 (br, 4), 4.14 (t, J ) 6.9 Hz, 2); 13C NMR (CDCl3) δ 23.6,
3
7.21-7.37 (m, 5), 8.32 (br s, 1); 13C NMR (DMSO-d6) δ 25.1,
25.5, 26.8, 49.9, 50.5, 50.7, 116.9, 130.1, 151.5, 155.2, 157.0,
43.5, 44.6, 45.4, 50.1, 105.4, 112.9, 126.6, 127.5, 128.1, 139.9,
160.1; MS m/e 338 (M+). Anal. (C14H16Cl2N6) C, H, N.
152.7, 154.2, 158.2, 160.2; MS m/e 424 (M+). Anal. (C21H25
-
2,6-Dich lor o-7-(d im et h yla m in o)-4-p yr r olid in op t er i-
d in e (6d ). The product was analogously prepared as described
for 6a , starting with 132 mg (0.43 mmol) of 5a and using 66.4
µL (0.47 mmol) of triethylamine. The 1.1-fold molar ratio of
dimethylamine was added as 2 M solution in THF (238 µL);
yield 130 mg (95.9%), yellow platelets, mp 239 °C: 1H NMR
ClN8) C, H, N.
Cr ysta l Gr ow in g of 7b. A sample of 7b (20.8 mg, 0.049
mmol) was suspended in 4 mL of 0.1 N HClO4. Ethanol (5
mL) was added, and the mixture was warmed to 50 °C, until
complete solution had taken place. The solution was cooled
slowly to room temperature and was kept at this temperature
for 3 months under light protection. Then the mixture was
stored at 4 °C until crystal growth was completed.
3
3
(CDCl3) δ 1.94 (qu, J ) 6.8 Hz, 2), 2.07 (qu, J ) 6.8 Hz, 2),
3
3
3.31 (s, 6), 3.79 (t, J ) 6.8 Hz, 2), 4.14 (t, J ) 6.8 Hz, 2); 13
C
NMR (CDCl3) δ 23.7, 26.8, 41.2, 50.0, 50.8, 118.1, 131.3, 154.6,
154.7, 157.0, 160.3; MS m/e 312 (M+). Anal. (C12H14Cl2N6) C,
H, N.
Cr ysta l Str u ctu r e Deter m in a tion of 7b‚2HClO4‚H2O.
Mr (C21H23Cl3N8O9) ) 637.82, orthorhombic space group
P212121, a ) 8.645(2) Å, b ) 12.338(2) Å, c ) 26.471(5) Å, V )
2823.5(10) Å3, Z ) 4, Dx ) 1.500 g cm-3, µ ) 0.388 mm-1 (λ )
0.71073 Å), T ) 293 K, colorless crystal of dimensions 0.3 mm
× 0.25 mm × 0.2 mm in a thin walled glass capillary, 20517
reflection (Θmax ) 25.05°), data were collected on a STOE IPDS
diffractometer. The structure was solved by direct methods22
and refined (438 parameters) by full-matrix least-squares
calculations on F2,23 using all 4823 unique reflections to final
7-Be n zyla m in o-2,6-d ich lor o-4-d im e t h yla m in op t e r i-
d in e (6e). Preparation as described for 6a , starting with 160
mg (0.574 mmol) of 5e, adding 88 µL (0.632 mmol) of triethyl-
amine and 69 µL (0.632 mmol) of benzylamine; yield 190 mg
(94.7%), recrystallization from ethanol, mp 225-227 °C: 1H
3
NMR (CDCl3) δ 3.52 (br, 6), 4.80 (d, J ) 5.5 Hz, 2), 6.06 (s,
1), 7.28-7.35 (m, 5); 13C NMR (CDCl3) δ 41.4, 45.5, 117.3,
127.9, 128.1, 128.8, 130.7, 137.0, 151.3, 156.6, 158.9, 159.8;
MS m/e 348 (M+). Anal. (C15H14Cl2N6) C, H, N.
R1[Fo > 2σ(Fo2)] ) 0.0515, wR2 ) 0.1160, w ) 1/[σ2(Fo2) +
2
(0.033P)2 + 0.7P] where P ) (Fo + 2Fc2)/3, S ) 1.083.23 The
2

4742 J ournal of Medicinal Chemistry, 1998, Vol. 41, No. 24
Merz et al.
absolute structure has been determined by refinement of a
Flack parameter (-0.04(9)).24 Largest peak and hole in the
final difference map are 0.383 e/Å3 and -0.261 e/Å2, respec-
tively. Anisotropic displacement parameters were refined for
all non-hydrogen atoms. All CH2 hydrogen atoms were
included in the refinement in calculated postions and were
allowed to ride on the atom to which they are attached. The
hydrogen atom positions belonging to NH2 and NH groups
were refined free with a restrained N distance. Depending
on the nature of the group, the isotropic displacement param-
eters of the H atoms were kept equal to 120% and 130% of
the equivalent isotropic displacement parameters of the parent
atoms, respectively.
solid, mp 186 °C: 1H NMR (CDCl3) δ 1.94 (br, 2), 2.01 (br, 2),
3.70 (br, 2), 3.74 (t, 3J ) 4.7 Hz, 4), 3.91 (t, 3J ) 4.7 Hz, 4),
3
4.11 (br, 2), 4.81 (d, J ) 5.2 Hz, 2), 5.66 (s, 1), 7.30-7.37 (m,
5); 13C NMR (CDCl3) δ 23.6, 27.0, 44.5, 45.6, 49.4, 50.2, 67.1,
115.1, 126.1, 127.7, 128.0, 128.8, 137.8, 151.4, 157.0, 157.4,
160.4; MS m/e 425 (M+). Anal. (C21H24ClN7O‚0.5 H2O) C, H,
N.
2-(4-Acetylp ip er a zin o)-7-ben zyla m in o-6-ch lor o-4-p yr -
r olid in op ter id in e (7i). To a suspension of 7a (228 mg, 0.536
mmol) in 10 mL of dioxane was added triethylamine (80 µL,
0.574 mmol). A solution of 101 µL (1.07 mmol) of acetic
anhydride in 5 mL of dioxane was added slowly, and the
mixture was stirred until no educt could be monitored by TLC.
The reaction mixture was evaporated to dryness, washed with
water, and dried. The yellow residue was purified by column
chromatography (silica gel, ethanol/ethyl acetate ) 1/1 v/v, Rf
) 0.58) to yield 186 mg (74.3%) of fine light-yellow needles;
mp 146-148 °C: 1H NMR (CDCl3) δ 1.93 (br, 2), 2.02 (br, 2),
2.13 (s, 3), 3.50 (t, 3J ) 5.2 Hz, 2), 3.67 (t, 3J ) 5.2 Hz, 2), 3.73
(br, 2), 3.90 (t, 3J ) 5.2 Hz, 2), 3.96 (t, 3J ) 5.2 Hz, 2), 4.11
4,7-Bis(pyr r olidin o)-6-ch lor o-2-piper azin opter idin e (7c).
A sample of 394 mg (1.16 mmol) of 6c was suspended in 20
mL of dioxane. Ground piperazine (400 mg, 4.64 mmol) was
added, and the mixture was heated under reflux for 1 h. The
solvent was removed in vacuo. The residue was thoroughly
washed with 30 mL of water, filtered, and dried in vacuo over
KOH, yield 90.2%. Column chromatography (silica gel, etha-
nol + 2.5% triethylamine) provided 271 mg (60.1%) of 7c. For
analytical specimen, the product was dissolved in 0.1 N HCl
and neutralized with 2% NH3. The precipitate was filtered
and dried over KOH; yellow solid, mp 194 °C: 1H NMR (CDCl3)
3
3
(br, 2), 4.81 (d, J ) 5.3 Hz, 2), 5.71 (t, J ) 5.3 Hz, 1), 7.29-
7.36 (m, 5); 13C NMR (CDCl3) δ 21.4, 23.7, 26.9, 41.5, 43.7,
44.0, 45.6, 46.4, 49.4, 50.2, 115.0, 126.3, 127.7, 128.0, 128.8,
137.7, 151.4, 157.0, 157.2, 160.1, 169.1; MS m/e 466 (M+). Anal.
(C23H27ClN8O) C, H, N.
3
δ 1.90-1.96 (m, 9), 2.88 (t, J ) 5.1 Hz, 4), 3.71 (br, 2), 3.84-
3.86 (m, 8), 4.08 (br, 2); 13C NMR (CDCl3) δ 23.7, 25.6, 27.0,
45.1, 46.2, 49.1, 50.1, 115.1, 125.2, 152.1, 156.3, 156.8, 160.5;
MS m/e 388 (M+). Anal. (C18H25ClN8‚0.5 H2O) C, H, N.
2-Oxop ip er a zin e. The compound was synthesized accord-
ing to the method of Aspinall.25 Yield of the crude product
was 57.8%. Recrystallization afforded hygroscopical colorless
platelets; mp 132-133 °C: 1H NMR (CDCl3) δ 1.94 (br, 1), 3.02
6-Ch lor o-7-(dim eth ylam in o)-2-piper azin o-4-pyr r olidin o-
p ter id in e (7d ). See 7c. Preparation started with 125 mg (0.4
mmol) of 6d in 15 mL of dioxane, and the yield after column
chromatography was 77 mg (53.0%) of 7d ; yellow solid, mp
146-148 °C: 1H NMR (CDCl3) δ 1.70 (br, 1), 1.91 (br, 2), 2.01
3
3
4
(t, J ) 5.4 Hz, 2), 3.36 (dt, J ) 5.4 Hz, J ) 2.3 Hz, 2), 3.51
(s, 2), 7.49 (br, 1); 13C NMR (CDCl3) δ 42.2, 42.8, 49.7, 170.6.
Anal. (C4H8N2O) C, H, N.
7-Ben zyla m in o-6-ch lor o-2-(3-oxop ip er a zin o)-4-p yr r oli-
d in op ter id in e (7j). To a suspension of 6a (375 mg, 1 mmol)
in 15 mL of dioxane was added 2-oxopiperazine (400 mg, 4
mmol). The mixture was stirred at 80 °C until no educt could
be monitored by TLC (1:1 ethyl acetate/ hexane v/v). The
solvent was removed in vacuo, and the residue was washed
well with water and purified by column chromatography (silica
gel, 5:1 ethyl acetate/ethanol v/v); pale-yellow solid, mp >250
°C: 1H NMR (CDCl3) δ 1.92 (br, 2), 2.05 (br, 2), 3.47 (m, 2),
3.76 (br, 2), 4.13 (br, 2), 4.18 (t, 3J ) 5.3 Hz, 2), 4.54 (s, 2),
4.82 (d, 3J ) 5.2 Hz, 2), 5.71 (t, 3J ) 5.2 Hz, 1), 6.27 (s, 1),
7.31-7.39 (m, 5); 13C NMR (CDCl3) δ 23.7, 27.0, 40.6, 41.5,
45.7, 48.4, 49.6, 50.4, 115.0, 126.7, 127.8, 128.1, 128.9, 137.6,
151.4, 156.6, 156.8, 160.5, 168.8; MS m/e 438 (M+). Anal.
(C21H23ClN8O) C, H. N: calcd, 25.5%; found, 24.9%.
3
(br, 2), 2.89 (t, J ) 5.1 Hz, 4), 3.20 (s, 6), 3.72 (br, 2), 3.87 (t,
3J ) 5.1 Hz, 4), 4.11 (br, 2); 13C NMR (CDCl3) δ 23.8, 27.0,
41.0, 45.2, 46.3, 49.2, 50.2, 116.5, 127.2, 155.6, 156.0, 156.9,
160.5; MS m/e 362 (M+). Anal. (C16H23ClN8) C, H, N.
7-Ben zyla m in o-6-ch lor o-4-(d im et h yla m in o)-2-p ip er -
a zin op ter id in e (7e). See 7c. Preparation started with 176
mg (0.504 mol) of 6e in 15 mL of dioxane, and the yield after
column chromatography was 102 mg (50.7%) of 7e; yellow
solid, mp 138-140 °C: 1H NMR (CDCl3) δ 1.78 (br, 1), 2.91 (t,
3
3
3J ) 5.0 Hz, 4), 3.45 (s, 6), 3.88 (t, J ) 5.0 Hz, 4), 4.81 (d, J
) 5.2 Hz, 2), 5.66 (t, 3J ) 5.2 Hz, 1), 7.28-7.38 (m, 5); 13C
NMR (CDCl3) δ 41.0, 45.1, 45.5, 46.2, 114.9, 125.2, 127.7, 128.1,
128.8, 137.7, 151.2, 157.7, 158.9, 159.9; MS m/e 398 (M+). Anal.
(C19H23ClN8‚0.5 H2O) C, H, N.
4,7-Bis(d im et h yla m in o)-6-ch lor o-2-p ip er a zin op t er i-
d in e (7f). See 7c. Preparation started with 240 mg (0.836
mmol) of 6f in 20 mL of dioxane, and the yield after column
chromatography was 196 mg (69.6%) of 7f; yellow solid, mp
2-(2-Am in oeth yla m in o)-7-ben zyla m in o-6-ch lor o-4-p yr -
r olid in op ter id in e (7k ). See 7g. Preparation started with
139 mg (0.37 mmol) of 6a in 10 mL of dioxane. Ethylenedi-
amine (200 µL, 3 mmol) was added, and the mixture was
refluxed for 3 h; yield 120 mg (81.3%), bright-yellow solid, mp
177-179 °C: 1H NMR (CDCl3) δ 1.89 (br, 4), 2.00 (br, 2), 2.90
(t, 3J ) 5.6 Hz, 2), 3.53 (br, 2), 3.69 (br, 2), 4.09 (br, 2), 4.82 (d,
3
131-132 °C: 1H NMR (CDCl3) δ 2.12 (br, 1), 2.89 (t, J ) 5.0
Hz, 4), 3.20 (s, 6), 3.42 (s, 6), 3.85 (t, 3J ) 5.0 Hz, 4); 13C NMR
(CDCl3) δ 40.8, 40.9, 45.1, 46.1, 116.3, 126.3, 155.2, 156.2,
158.8, 160.0; MS m/e 336 (M+). Anal. (C14H21ClN8‚0.5 H2O)
C, H, N.
3J ) 4.6 Hz, 2), 5.44 (br, 1), 5.70 (br, 1), 7.28-7.34 (m, 5); 13
C
NMR (CDCl3) δ 23.7, 26.9, 41.9, 44.3, 45.3, 49.4, 50.3, 115.3,
125.7, 127.6, 127.9, 128.7, 137.8, 151.2, 157.2, 157.3, 161.6.
Anal. (C19H23ClN8) C, H, N.
7-Ben zyla m in o-6-ch lor o-2-(4-m eth ylp ip er a zin o)-4-p yr -
r olidin opter idin e (7 g). 6a (600 mg, 1.6 mmol) and 1-methyl-
piperazine (640 mg, 6.4 mmol) in 30 mL of dioxane were heated
under reflux until complete disappearance of educt, as moni-
tored by TLC (ethyl acetate/hexane ) 1/1 v/v). The solvent
was removed in vacuo, the residue was washed two times with
40 mL of water, filtered, and dried over KOH; yellow solid,
yield 626 mg (89.1%), mp 204 °C: 1H NMR (DMSO-d6) δ 1.90
7-Ben zyla m in o-6-ch lor o-2-(2-h yd r oxyet h yla m in o)-4-
p yr r olid in op ter id in e (7l). See 7c. Preparation started with
195 mg (0.52 mmol) of 6a in 10 mL of dioxane. Ethanolamine
(260 µL, 4.3 mmol) was added, and the mixture was refluxed
for 5 h. Yield was 181 mg (87.1%), and yield after column
chromatography (silica gel, ethyl acetate/ethanol ) 5/1 v/v) was
117 mg (56.4%); pale-yellow crystals, mp 198 °C: 1H NMR
(CDCl3) δ 1.92 (br, 2), 2.01 (br, 2), 3.61-3.65 (m, 2), 3.72 (br,
2), 3.81 (t, 3J ) 4.6 Hz, 2), 4.11 (br, 2), 4.77 (d, 3J ) 4.9 Hz, 2),
5.75 (t, 3J ) 4.9 Hz, 1), 5.93 (br, 1), 7.29-7.37 (m, 5); 13C NMR
(CDCl3) δ 23.8, 26.8, 44.7, 45.4, 49.6, 50.7, 64.5, 115.2, 126.1,
127.7, 128.0, 128.8, 137.7, 151.4, 157.0, 157.2, 162.0; MS m/e
399 (M+). Anal. (C19H22ClN7O) C, H, N.
3
(br, 4), 2.19 (s, 3), 2.30 (t, J ) 4.9 Hz, 4), 3.67 (br, 2), 3.72 (t,
3J ) 4.4 Hz, 4), 3.97 (br, 2), 4.67 (d, 3J ) 6.4 Hz, 2), 7.22-7.32
(m, 5), 8.01 (t, 3J ) 6.4 Hz, 1); 13C NMR (DMSO-d6) δ 23.1,
26.4, 43.4, 43.5, 45.8, 49.0, 49.9, 54.6, 113.8, 125.5, 126.7, 126.8,
128.3, 139.2, 151.5, 156.4, 156.8, 159.6; MS m/e 438 (M+). Anal.
(C22H27ClN8‚0.5 H2O) H, N. C: calcd, 59.0%; found, 59.5%.
7-Ben zyla m in o-6-ch lor o-2-m or p h olin o-4-p yr r olid in o-
p ter id in e (7h ). See 7g. Preparation started with 173 mg
(0.461 mmol) of 6a in 12 mL of dioxane, and morpholine (160
µL, 1.84 mmol) was added; yield 175 mg (89.1%) of 7h , yellow
7-Ben zyla m in o-6-ch lor o-2-(d im eth yla m in o)-4-p yr r oli-
d in op t er id in e (7m ) a n d 7-Ben zyla m in o-2-ch lor o-6-(d i-
m eth yla m in o)-4-p yr r olid in op ter id in e (7n ). In a 60 mL

Synthesis of Potent Inhibitors of PDE4
J ournal of Medicinal Chemistry, 1998, Vol. 41, No. 24 4743
(4) Roch, J .; Nickl, J .; Mu¨ller, E.; Narr, B.; Weisenberger, J .-M.;
Zimmermann, R.; Haarmann, W. Neue 2-Piperazino-pteridine,
Verfahren zu ihrer Herstellung und diese Verbindungen enthal-
tende Arzneimittel. Deutschland Patent, DE 3323932, 1985.
(5) Heckel, A.; Nickl, J .; Mu¨ller, E.; Narr, B.; Weisenberger, H.;
Zimmermann, R. Neue 2-Piperazino-pteridine, Verfahren zu
ihrer Herstellung und diese Verbindungen enthaltende Arznei-
mittel. Deutschland Patent, DE 3540952, 1987.
(6) Sherman, W. R.; Taylor, E. C. Diaminouracil Hydrochloride. Org.
Synth. 1957, 37, 15-17.
(7) Taylor, E. C.; Sherman, W. R. Pteridines. XVII. Reactions of
2,4,6,7-Tetrachloropteridine. The Synthesis of 5,6,7,8-Tetrahy-
dropteridine. J . Am. Chem. Soc. 1959, 81, 2464-2471.
(8) Scho¨pf, C.; Reichert, R.; Riefstahl, K. Zur Kenntnis des Leuko-
pterins. Liebigs Ann. Chem. 1941, 548, 82-94.
(9) Schenker, C. An Investigation of the Reactivity of Halogenated
Pteridines. Ph.D. Thesis, Cornell University, Ithaca, NY, 1949.
(10) Cain, C. K.; Schenker, C. Relative Reactivity of Chlorine Atoms
in 2,4,6,7-Tetrachloropteridine. In Abstracts of Papers, 117th
ACS Meeting, March-April 1950; American Chemical Society:
Washington, DC, 1950; pp 41L-42L.
(11) Roch, J . Verfahren zur Herstellung von Pteridinderivaten.
Deutschland Patent, 1088969, 1960.
(12) Roch, J . Tri- and tetra-substituted pteridine derivatives. United
States Patent, 2940972, 1960.
(13) Brown, D. J . Halogenopteridines. In Fused Pyrimidines. Part
Three. Pteridines, The Chemistry of Heterocyclic Compounds;
Taylor, E. C., Weissberger, A., Eds.; J ohn Wiley & Sons: New
York, 1988; pp 187-203.
(14) Hetzer, H. B.; Bates, R. G.; Robinson, R. A. Dissociation constant
of pyrrolidinium ion and related thermodynamic quantities from
0 to 50°. J . Phys. Chem. 1963, 67, 1124-1127.
(15) Kaltofen, R.; Opitz, R.; Schumann, K.; Ziemann, J . Tabellenbuch
Chemie, 5th ed.; VEB Deutscher Verlag fu¨r Grundstoffindus-
trie: Leipzig, 1968; p 305.
(16) Sette, C.; Conti, M. Phosphorylation and activation of a cAMP-
specific phosphodiesterase by the cAMP-dependent protein
kinase. J . Biol. Chem. 1996, 271 (28), 16526-16534.
(17) Bolger, G. B.; Erdogan, S.; J ones, R. E.; Loughney, K.; Scotland,
G.; Hoffmann, R.; Wilkinson, I.; Farrell, C.; Houslay, M. D.
Characterization of five different proteins produced by alterna-
tively spliced mRNAs from the human cAMP-specific phosphodi-
esterase PDE4D gene. Biochem. J . 1997, 328 (Pt2), 539-548.
(18) Marko, D.; Merz, K. H.; Tarasova, N.; Eisenbrand, G. Subcellular
localization of a pteridine-based inhibitor of cyclic nucleotide
monophosphate specific phosphodiesterase. J . Cancer Res. Clin.
Oncol. 1997, Suppl. 1, 123.
(19) Berger, D. P.; Fiebig, H. H.; Winterhalter, B. R. Establishment
and characterization of human tumor xenograft models in nude
mice. In Immundeficient Mice in Oncology; Fiebig, H. H., Berger,
D. P., Eds.; Karger: Basel, Switzerland, Contrib. Oncol. 1992;
Vol. 42, pp 23-46.
(20) Po¨ch, N. Assay of PDE with radioactively labeled cAMP as
substrate. Naunyn Schmiedebergs Arch. Pharmacol. 1971, 268,
272-299.
(21) Skehan, P.; Storeng, R.; Scudiero, D.; Monks, A.; McMahon, J .;
Vistica, D.; Warren, J . T.; Bokesch, H.; Kenney, S.; Boyd, M. R.
New colorometric assay for anticancer-drug screening. J NCI,
1990, 82, 1107-1112.
glass tube for a maximum pressure of 20 bar, 386 mg (1.03
mmol) of 6a were suspended in 25 mL of DMF. Gaseous
ammonia was passed for 1 min through the tube, then the
system was closed, and the tube was heated under thorough
stirring with a magnetic stirrer at 100 °C for 4 h. Thereafter
the pressure was removed, and the mixture was poured into
100 mL of water. The precipitate was collected on a filter,
washed with 2% NH3 and water, and dried. Column chroma-
tography on silica gel (L 2.5 cm; l ) 70 cm; ethyl acetate/
hexane ) 1/1 v/v) yielded 112 mg (28.4%) of 7n with Rf ) 0.55
and 151 mg (38.2%) of 7m with Rf ) 0.30.
7-Ben zyla m in o-6-ch lor o-2-(d im eth yla m in o)-4-p yr r oli-
d in op ter id in e (7m ): light-yellow needles, mp 196 °C; 1H
NMR (CDCl3) δ 1.93 (br, 2), 2.01 (br, 2), 3.24 (s, 6), 3.76 (br,
2), 4.11 (br, 2), 4.83 (d, 3J ) 4.9 Hz, 2), 5.62 (t, 3J ) 4.9 Hz, 1),
7.30-7.37 (m, 5); 13C NMR (CDCl3) δ 23.8, 27.2, 37.3, 45.6,
49.2, 50.2, 114.8, 125.3, 127.7, 128.1, 128.8, 138.0, 151.3, 156.9,
157.5, 161.2; MS m/e 383 (M+). Anal. (C19H22ClN7) C, H, N.
7-Ben zyla m in o-2-ch lor o-6-(d im eth yla m in o)-4-p yr r oli-
d in op ter id in e (7n ): fine colorless needles, mp 187 °C; 1H
NMR (CDCl3) δ 1.93 (br, 2), 2.02 (br, 2), 2.79 (s, 6), 3.77 (br,
2), 4.26 (br, 2), 4.79 (d, 3J ) 5.5 Hz, 2), 5.77 (t, 3J ) 5.5 Hz, 1),
7.29-7.38 (m, 5); 13C NMR (CDCl3) δ 23.7, 27.0, 41.1, 45.3,
49.6, 50.7, 115.8, 127.6, 128.1, 128.8, 138.0, 145.9, 149.8, 154.7,
157.4, 157.9; MS m/e 383 (M+). Anal. (C19H22ClN7) C, H, N.
7-Ben zyla m in o-2-p ip er a zin o-4-p yr r olid in op t er id in e
(7o). A suspension of 356 mg (0.84 mmol) of 7a in 7.5 mL of
ethanol was stirred vigorously with 2.5 mL of 0.5 N HCl
containing 350 mg of 10% Pd/C for 2 h with a slight excess
pressure of H2. Then water (15 mL) and ethanol (25 mL) were
added, the mixture was filtered, and the ethanol was removed
in vacuo. The aqueous solution was adjusted to pH 9 with
0.2 N NH4OH, and the precipitate was allowed to stand
overnight. The solid was filtered, washed with water, and
dried in vacuo to afford 259 mg (79.1%) of crude product.
Column chromatography (silica gel, ethanol + 2.5% triethyl-
amine) yielded 178 mg (54.4%) of light-yellow crystals of 7o;
mp 218 °C: 1H NMR (CDCl3) δ 1.77 (s, 1), 1.93 (br, 4), 2.79 (t,
3
3
3J ) 5.0 Hz, 4), 3.88 (t, J ) 5.0 Hz, 4), 4.70 (d, J ) 5.6 Hz,
2), 5.29 (br, 1), 7.24-7.33 (m, 5), 7.58 (s, 1); 13C NMR (CDCl3)
δ 24.0, 45.1, 45.3, 46.2, 49.8, 117.0, 126.6, 127.5, 127.8, 128.7,
138.3, 155.2, 157.7, 157.9, 160.3; MS m/e 390 (M+). Anal.
(C21H26N8) C, H, N.
Ack n ow led gm en t. Thanks are due to Prof. Dr. H.
H. Fiebig, Freiburg, Germany, for generously providing
us with tumor tissue of LXFL529. This study was
supported by the Deutsche Forschungsgemeinschaft,
Grant Ei 172/5-1 + 2.
Refer en ces
(1) Beavo, J . A.; Conti, M.; Heaslip, R. J . Multiple Cyclic Nucleotide
Phosphodiesterases. Mol. Pharmacol. 1994, 46, 399-405.
(2) Drees, M.; Zimmermann, R.; Eisenbrand, G. 3′,5′-Cyclic Nucleo-
tide Phosphodiesterase in Tumor Cells as Potential Target for
Tumor Growth Inhibition. Cancer Res. 1993, 53, 3058-3061.
(3) Marko, D.; Romanakis, K.; Zankl, H.; Fu¨rstenberger, G.; Stein-
bauer, B.; Eisenbrand, G. Induction of apoptosis by an inhibitor
of cAMP-specific phosphodiesterase (PDE) in malignant murine
carcinoma cells overexpressing PDE activity in comparison to
their nonmalignant counterparts. Cell Biochem. Biophys. 1998,
28, 75-101.
(22) Sheldrick, G. M. SHELXS86. Program for the Solution of Crystal
Structures. University of Go¨ttingen, Germany, 1985.
(23) Sheldrick, G. M. SHELXl97. Program for the Refinement of
Crystal Structures. University of Go¨ttingen, Germany, 1997.
(24) Flack, H. D. On Enantiomorph-Polarity Estimation. Acta Crys-
tallogr. Sect. A 1983, 39, 876-881.
(25) Aspinall, S. R. A Synthesis of Monoketopiperazines. J . Am.
Chem. Soc. 1940, 62, 1202-1204.
J M981021V