Novel selective phosphodiesterase (PDE4) inhibitors. 4. Resolution, absolute configuration, and PDE4 inhibitory activity of cis-tetra- and cis-hexahydrophthalazinones

Article abstract of DOI:10.1021/jm0110338

Recently, we reported that 4-catechol-substituted cis-(±)-4a,5,6,7,8,8a-hexa- and cis-(±)- 4a,5,8,8a-tetrahydro-2H-phthalazin-1-ones show potent inhibition of phosphodiesterase (PDE4) activity, while the corresponding trans racemic mixtures exhibit only w

Full text of DOI:10.1021/jm0110338

2526
J . Med. Chem. 2002, 45, 2526-2533
Novel Selective P h osp h od iester a se (P DE4) In h ibitor s. 4. Resolu tion , Absolu te
Con figu r a tion , a n d P DE4 In h ibitor y Activity of cis-Tetr a - a n d
cis-Hexa h yd r op h th a la zin on es
Margaretha Van der Mey,*,†,‡ Hildegard Boss,§ Dennis Couwenberg,‡ Armin Hatzelmann,§ Geert J . Sterk,‡
Kees Goubitz,| Henk Schenk,| and Hendrik Timmerman†
Leiden/ Amsterdam Center for Drug Research, Division of Medicinal Chemistry, Department of Pharmacochemistry, Vrije
Universiteit, De Boelelaan 1083, 1081 HV Amsterdam, The Netherlands, Byk Nederland, Zwanenburg, The Netherlands,
Laboratory for Crystallography, Institute for Molecular Chemistry (IMC), University of Amsterdam, Amsterdam, The
Netherlands, and Byk Gulden, Konstanz, Germany
Received September 3, 2001
Recently, we reported that 4-catechol-substituted cis-(()-4a,5,6,7,8,8a-hexa- and cis-(()-
4a,5,8,8a-tetrahydro-2H-phthalazin-1-ones show potent inhibition of phosphodiesterase (PDE4)
activity, while the corresponding trans racemic mixtures exhibit only weak to moderate activity.
To determine the absolute configuration and PDE4 inhibitory activity of the individual cis-
enantiomers, several optically active phthalazinones have been synthesized. The enantiomers
of the various γ-keto acids, used as starting materials, were resolved in a classical way by the
formation of diastereomeric salts, and each was converted to optically active phthalazinone in
an enantioselective manner. The absolute configuration of the (+)-enantiomer of cis-hexahy-
drophthalazinone (+)-12 was determined by X-ray crystallography. The carbon atoms at the
4a and 8a positions were found to have the S- and R-configuration, respectively. In the present
series of hexa- and tetrahydrophthalazinones, stereoselectivity for PDE4 inhibition is observed;
the cis-(+)-enantiomers of the phthalazinones display high inhibitory activity, whereas their
(-)-counterparts exhibit only weak to moderate activity. It is likely that all cis-(+)-phthalazi-
nones have a (4aS,8aR)-configuration and vice versa for the cis-(-)-analogues. In the current
series, the N-adamantan-2-yl analogue (+)-14 shows the most potent inhibition of PDE4 (pIC50
) 9.3); the corresponding (-)-enantiomer is 250-fold less active. In addition, the N-substituted
tetrahydrophthalazinones under study were investigated for their in vivo antiinflammatory
activities by examining the suppression of arachidonic acid (AA) induced mouse ear edema
formation. In this assay analogues (+)-14 and (+)-15 were found to be potent antiinflammatory
agents showing about 50% inhibition at 30 µmol/kg po.
In tr od u ction
Recently, we presented a novel class of chiral
4a,5,6,7,8,8a-hexa- and 4a,5,8,8a-tetrahydro-2H-phthal-
azin-1-ones as cAMP-specific phosphodiesterase (PDE4)
inhibitors, for which the synthesis and preliminary
structure-activity relationships (SARs) have been
F igu r e 1. Target cis-hexa- and cis-tetrahydrophthalazinone
reported.1-6 Studying the influence of several fused
enantiomers (general structures).
hydrocarbon rings on PDE4 inhibition, we found that
the (4a,8a)-cis-racemates of these hexa- and tetrahy-
drophthalazinones show potent inhibition of PDE4
activity.4 In contrast, the corresponding trans racemic
mixtures exhibit only weak to moderate inhibitory
activities. Besides potent PDE4 inhibition, several ra-
cemic cis-hexa- and tetrahydrophthalazinones were
found to possess moderate to high in vitro and in vivo
antiinflammatory potencies.6 Therefore, these analogues
are considered as promising agents for the treatment
of rheumatoid arthritis, asthma, and other inflamma-
tory diseases.
This paper describes the resolution of various cis-
hexa- and tetrahydrophthalazinones (Figure 1) in order
to examine the pharmacological profile of the individual
cis-hexa- and tetrahydrophthalazinone enantiomers and
to determine their absolute configurations.
Most target phthalazinones were selected from previ-
ous papers.4-6 Because of their easy accessibility, ini-
tially a number of optically active phthalazinones in
which R4 is H (Figure 1) were synthesized to establish
the influence of the configuration on PDE4 inhibitory
activity. Subsequently, some optically active N-substi-
tuted tetrahydrophthalazinones have been prepared and
tested for their PDE4 inhibitory and in vivo antiinflam-
matory activity.
* To whom correspondence should be addressed. Phone: +31 20
4449720. Fax: +31 20 4449121. E-mail: mmeij@rnc.vu.nl. Current
address: Vrije Universiteit, Radionuclidencentrum, De Boelelaan
1085c, 1081 HV Amsterdam, The Netherlands.
† Vrije Universiteit.
Ch em istr y
‡ Byk Nederland.
The target optically active cis-hexa- and tetrahydro-
phthalazinones with variations in the substituents at
§ Byk Gulden.
| University of Amsterdam.
10.1021/jm0110338 CCC: $22.00 © 2002 American Chemical Society
Published on Web 02/26/2002

Phosphodiesterase Inhibitors
J ournal of Medicinal Chemistry, 2002, Vol. 45, No. 12 2527
Sch em e 1a
a
Reagents: (chiral base A) (R)-(+)-R-methylbenzylamine; (B) (S)-(-)-R-methylbenzylamine; (C) quinine; (D) (1R,2S)-(-)-ephedrine.
Ta ble 1. Data for the Resolution of the cis-Cyclohex(a)ene
γ-Keto Acids
Ta ble 2. PDE4 Inhibition Data of the Racemic and
Corresponding Resolved 4-Aryl-Substituted Hexa- and
Tetrahydrophthalazinones
a
(A) (R)-(+)-R-methylbenzylamine; (B) (S)-(-)-R-methylbenzyl-
b
amine; (C) quinine; (D) (1R,2S)-(-)-ephedrine. (Yield/maximum
yield for each enantiomer) × 100. c The concentrations for the
solutions applied are c ) 1 except where noted / c ) 5.2.
both N2 and the 4-catechol moiety were prepared from
the corresponding racemic γ-keto acids as outlined in
Scheme 1. The structures of the target phthalazinones
are summarized in Tables 2 and 3. In the first step, the
enantiomers of the γ-keto acids cis-(()-1-75 (Table 1)
were obtained by resolution. Subsequent treatment of
the resolved γ-keto acids with the selected hydrazine
in ethanol or pyridine provided the optically active
a
pIC50 ) -log IC50. Inhibition of PDE4 was investigated in the
cytosol of human neutrophils. The data are the means of two
independent determinations in triplicate.
target compounds (+)- and (-)-8-15 in variable yields.
Phthalazinone (+)-16 was achieved after an additional
reaction step.

2528 J ournal of Medicinal Chemistry, 2002, Vol. 45, No. 12
Van der Mey et al.
Sch em e 2a
Ta ble 3. Summary of the PDE4 Inhibitory and in Vivo
Antiinflammatory Activities of the Racemic and Analogous
Resolved N-Substituted Tetrahydrophthalazinones
% inhibition
of AA-induced
a
PDE4
mouse ear
Reagents: (i) 1, SOCl2; 2, L-(-)-menthol; (ii) p-TosOH, L-(-)-
a
compd
R1 R2
R3
pIC50
edemab
menthol.
(()-14
Me Me adamantan-2-yl
Me Me adamantan-2-yl
Me Me adamantan-2-yl
Et Et 4-HO2CC6H4
Et Et 4-HO2CC6H4
Et Et 4-EtO2CC6H4
Et Et 4-EtO2CC6H4
9.2
9.3
6.9
7.7
8.3
8.0
8.2
7.3
7.7
7.0
51
49
c
48
47
c
27
55
48
33
(+)-14
contamination of one enantiomer by the other could be
detected in the resolved samples. Unfortunately, we
were not able to distinguish the two enantiomers of
(-)-14
(()-15
(+)-15
1
carboxylic acid 15 by H NMR spectroscopy. However,
(()-16
(+)-16
the antipodes of ethyl ester 16 could easily be identified
with 1H NMR, using the europium shift reagent Eu-
(hfc)3; since the benzoic acid (+)-15 was used as a
starting material for ester (+)-16 (>96% ee), comparable
enantiomeric purity is likely.
(-)-Rolipram
CDP840
Ariflo
a
b
See corresponding footnote from Table 2. Percent inhibition
of the formation of AA-induced mouse ear edema after pretreat-
The preparation of phthalazinones (()-8-14 (Tables
2 and 3) has been described previously.4-6 Condensation
of γ-keto acid (()-7 with 4-hydrazinobenzoic acid in
pyridine at reflux temperature afforded phthalazinone
(()-15. Ethyl ester (()-16 was prepared from (()-15 as
explained above for (()-15.
The optical rotation values for the resolved γ-keto
acids 1-7 and phthalazinones 8-6 are summarized in
Table 1 and the Experimental Section, respectively.
Attempts were made to synthesize diastereomeric
esters by acid-catalyzed condensation of γ-keto acids
with a chiral alcohol (e.g., L-(-)-menthol) or by conver-
sion of the γ-keto acids into the corresponding acid
chlorides, which were then coupled with a chiral alcohol
(Scheme 2). However, these efforts failed because the
γ-keto acids and the corresponding acid chlorides un-
dergo intramolecular lactonization as shown in Scheme
2. A similar lactonization was published for analogous
oxybutyric acids.8
ment with the target compound (1 h before AA) at
a drug
concentration of 30 µmol/kg po. Results are given as mean from
two independent experiments. c Not determined.
The racemates of targets 8-15 were synthesized, if
not already available,4-6 for pharmacological compari-
son.
The preparation of γ-keto acids cis-(()-1-6 has been
described in preceding papers.4,5 Friedel-Crafts acyla-
tion of 1,2-diethoxybenzene with cis-1,2,3,6-tetrahydro-
phthalic anhydride gave the corresponding cyclohex-3-
enecarboxylic acid cis-(()-7.
The conditions used for the resolution of γ-keto acids
(-)-1-6 and (+)-1-7 ((-)-7 was not isolated) are sum-
marized in Table 1. From a number of chiral bases
tested as potential resolving agents, (1R,2S)-(-)-ephe-
drine, quinine, and (R)- and (S)-(+)-R-methylbenzyl-
amine were found to be the most suitable. The most
diastereomerically pure salts were obtained using ethyl
acetate or ethanol as solvents for crystallization. The
diastereomeric salts were purified by recrystallization
until a diastereomeric excess (de) of more than 96% was
reached.7 The diastereomeric excess was determined by
means of 1H NMR spectroscopy; the diastereomers could
easily be distinguished because equimolar amounts of
quinine and racemic γ-keto acid show duplication of the
resonance of the H5 aromatic proton.
X-r a y Str u ctu r e a n d Absolu te Con figu r a tion of
(+)-12
The ideal method for determination of the absolute
configuration of the antipodes is X-ray crystallography.
From a solution of benzofuran (+)-12 in diethyl ether,
crystals suitable for an X-ray analysis could be grown.
It was found that the asymmetric unit cells contain two
identical independent molecules and one solvent mol-
ecule (Figures 2 and 3). Matching both molecules led to
a root-mean square (rms) value of 0.33 Å. The crystal
structure depicted in Figure 2 revealed that two hydro-
gen bonds are formed between the two target molecules
in a unit cell, one from N2A (N2, molecule A) to O9B
(O9, molecule B) and the other from N2B to O9A with
lengths of 2.943(5) and 2.890(5) Å, respectively. In the
crystal, the cis-fused cyclohexane ring of (+)-12 pos-
sesses an undistorted chair conformation. The carbonyl
moieties in both molecules are oriented axially, whereas
the NdC-carbon atom (C4) attached to C4a occupies the
equatorial position, thus confirming the cis configura-
tion. The benzofuran moiety, the 4-methoxy group, and
the heterocyclic ring are nearly coplanar. According to
Treatment of the enantiomerically pure γ-keto acids
1-6 with the suitable hydrazines provided the corre-
sponding optically active phthalazinones 8-15. Esteri-
fication of benzoic acid (+)-15 with acetyl chloride in
ethanol yielded derivative (+)-16. The enantiomeric
purity of the resolved phthalazinones was determined
1
by means of H NMR spectroscopy using a chiral shift
1
reagent. The H NMR spectra of the racemic phthalazi-
nones 8-14 and 16 in the presence of 0.7-2 equiv of
europium shift reagents Eu(hfc)3, Eu(tfc)3, or Eu(dcm)3
(full names in Experimental Section) showed duplication
of the resonances of the methoxy group(s) and/or the
H5 aromatic proton, whereas samples of the two sepa-
rate enantiomers each exhibited single resonances, thus
indicating a high enantiomeric purity (>96% ee).7 No

Phosphodiesterase Inhibitors
J ournal of Medicinal Chemistry, 2002, Vol. 45, No. 12 2529
rings with the binding site play an important role in
PDE4 inhibition.4 Now, this hypothesis is further
strengthened by the observation that the cis-(+)-enan-
tiomers of phthalazinones 8-14 display potent PDE4
inhibition, whereas their (-)-counterparts exhibit only
weak to moderate PDE4 inhibitory activities (Tables 2
and 3); up to a 250-fold difference in activity (compound
14) is observed. The cis-(+)-phthalazinones most likely
have the same absolute configuration.
In Vivo An tiin fla m m a tor y Activity. The in vivo
antiinflammatory properties of selected N-substituted
phthalazinones (Table 3) have been determined in a
standard model of inflammation. In this assay, the
suppression of the formation of arachidonic acid (AA)
induced mouse ear edema was measured after pretreat-
ment with the target compound (1 h before AA) at a
drug concentration of 30 µmol/kg po.6
Analogues ((/+)-14 and ((/+)-15 are potent inhibitors
of arachidonic acid (AA) induced mouse ear edema
formation. Esterification of benzoic acid (+)-15 leads to
a considerable decrease of in vivo activity, while the
PDE4 inhibitory activity remains equal (compare (+)-
15 with (+)-16). This reduction in inhibition of mouse
ear edema may be caused by an increase in lipophilicity;
however, the N-adamantyl derivatives ((/+)-14 are
more lipophilic than (+)-16 but have higher in vivo
antiinflammatory activities.
F igu r e 2. Crystal structure of benzofuran (+)-12. The unit
cells contain one solvent molecule (not shown) and two target
molecules (A and B) that form two hydrogen bonds with one
another.
Con clu sion
In this study, various γ-keto acids were successfully
resolved by resolution. Subsequently, the resolved γ-ke-
to acids were converted into optically active phthalazi-
nones in an enantioselective manner, as confirmed by
1H NMR spectroscopy using a suitable europium shift
reagent. The absolute configuration of cis-(+)-(4aS,8aR)-
hexahydrophthalazinone (+)-12 was determined by
X-ray crystallography.
In the present series of hexa- and tetrahydrophthalazi-
nones, stereoselectivity for PDE4 inhibition is observed
like for the well-known chiral PDE4 inhibitors including
rolipram,9 CDP840,10 and ariflo.11 The cis-(+)-enanti-
omers of the phthalazinones display high PDE4 inhibi-
tory activity, whereas their (-)-counterparts exhibit
only weak to moderate activity. Therefore, we assume
that all cis-(+)-phthalazinones have a (4aS,8aR)-con-
figuration, the arrangement found for (+)-12. The
N-substituted tetrahydrophthalazinones under study
also possess in vivo antiinflammatory activity. In par-
ticular, analogues ((/+)-14 and ((/+)-15 were found to
be potent antiinflammatory agents showing about 50%
inhibition of formation of mouse ear edema at a dose of
30 µmol/kg po. We plan to investigate an extended series
of N-substituted cis-(+)-phthalazinones for their in vitro
and in vivo antiinflammatory properties, applying vari-
ous assays.
F igu r e 3. Structure of molecule B as found in the unit cells
of a crystal of (+)-12. The positions of the hydrogen atoms are
calculated.
the Cahn-Ingold-Prelog convention, the annealed
carbon atoms at the 4a and 8a positions have the (S)-
and (R)-configuration, respectively.
The crystal structure of benzofuran (+)-12 is, not
surprisingly, very similar to the calculated (modeled)
3-D structures of the analogous phthalazinones we
described earlier.4
P h a r m a cology
The PDE4 inhibitory activities of the hexa- and
tetrahydrophthalazinones under investigation were de-
termined as described previously4 and are listed in
Tables 2 and 3. Structure-activity relationships for the
present series of optically active cis-phthalazinones are
discussed below.
Str u ctu r e-Activity Rela tion sh ip s. Recently, we
discussed the modeling and superimposition of a num-
ber of phthalazinones with various fused hydrocarbon
rings and proposed that steric interactions of these fused
Exp er im en ta l Section
I. Cr ysta llogr a p h ic Stu d ies. A crystal of (+)-12 with
approximate dimensions 0.30 mm × 0.45 mm × 0.50 mm was
used for data collection on an Enraf-Nonius CAD-4 diffrac-
tometer with graphite-monochromated Cu KR radiation and
ω-2θ scan. A total of 4656 unique reflections was measured
within the range -13e he 13, 0e ke 11, 0e le 26. Of these,
4323 were above the significance level of 2.5 σ(I). The range
of (sin θ)/λ was 0.046-0.626 Å-1 (4.1° < θ < 74.7°). Two

2530 J ournal of Medicinal Chemistry, 2002, Vol. 45, No. 12
Van der Mey et al.
reference reflections (0 1 5; 3h 1 4) were measured hourly and
showed no decrease during the 72 h collecting time. In
addition, around 969 “Friedel” reflections were measured,
which were used in the determination of the absolute config-
uration. Unit-cell parameters were refined by a least-squares
fitting procedure using 23 reflections with 80° e 2θ e 84°.
Corrections for Lorentz and polarization effects were applied.
The structure was solved by the program package CRUNCH.12
The hydrogen atoms were calculated. Full-matrix least-squares
refinement on F (anisotropic for the non-hydrogen atoms and
isotropic for the hydrogen atoms (the ADP’s of the solvent
hydrogen atoms were kept fixed at U ) 0.1 Å2) with restraint
of the latter in such a way that the distance to their carrier
remained constant at approximately 1.0 Å) converged to R )
0.040, Rw ) 0.040, (∆/σ)max ) 0.15, s ) 1.101. A weighting
scheme w ) [0.5 + 0.5(σ(Fobs)2 + 0.005/(σ(Fobs))]-1 was used.
The secondary isotropic extinction coefficient13,14 was refined
to EXT ) 0.67(2). The absolute structure parameter15 was
refined to Xabs ) 0.03, thus confirming the correct enantiomer.
A final difference Fourier map revealed a residual electron
density between -0.15 and 0.17 e Å-3. Scattering factors were
taken from Cromer and Mann16 and International Tables for
X-ray Crystallography.17 All calculations were performed with
XTAL18 unless stated otherwise. The asymmetric unit contains
two identical independent molecules and one solvent molecule.
Matching both molecules led to rms ) 0.33 Å. There are two
hydrogen bonds between the two molecules. One from N2a to
O9b and one from N2b to O9a with lengths of 2.943(5) and
salt was dissolved in dichloromethane (15 mL) and washed
twice with 1 N HCl (15 mL). The organic layer was dried over
MgSO4, filtered, and concentrated in vacuo to afford pure
γ-keto acid (-)-5: >96% ee; [R]25 -8.98° (c 1.1, CHCl3).
D
Crude γ-keto acid (+)-5 was obtained from the combined
and concentrated mother liquors containing oily diastereomeric
salt using the same procedure as mentioned above (CH2Cl2, 1
N HCl). The oily residue (6.85 g, 19.1 mmol) was dissolved in
absolute ethanol (150 mL), and (1R,2S)-(-)-ephedrine (3.16
g, 19.1 mmol) was added. The solution was cooled to 0 °C and
kept at this temperature for 72 h. The precipitate was filtered
off, washed twice with ethanol, and dried in vacuo to give 6.51
g (72%) of (1R,2S)-(-)-ephedrine salt of pure (+)-5: >96% de;
mp 143-146 °C; [R]25D -2.82° (c 1.0, CHCl3). The mother liquor
was concentrated to 75 mL, and after it was stirred overnight,
a second crop of pure diastereomeric salt (1.00 g, 11%) was
collected. The γ-keto acid (+)-5 was obtained as described
above: >96% ee; [R]25 10.7° (c 1.0, CHCl3).
D
The enantiomeric purity of (-)-5 and (+)-5 was determined
1
by H NMR (CDCl3). Quinine (1 equiv) was added to racemic
(()-5, resulting in a separation of the resonance of aromatic
H5 (doublet) into two resonances that appeared at 6.54 and
6.69 ppm for (+)-5 and (-)-5. Independent treatment of
samples of (-)-5 and (+)-5 with quinine (1 equiv) resulted in
retention of single resonances for the aromatic H5 resonance,
and no detectable contamination by the other enantiomer was
found (>96% de).
cis-(-)-4-(2,3-Dih yd r o-7-m et h oxyb en zofu r a n -2-sp ir o-
1′-cyclop en ta n -4-yl)-4a ,5,6,7,8,8a -h exa h yd r o-2H-p h th a l-
a zin -1-on e ((-)-12). The quinine salt of (-)-5 (1.91 g, 2.80
mmol) was dissolved in dichloromethane (25 mL) and washed
twice with 1 N HCl (20 mL). The organic layer was dried over
MgSO4 and concentrated in vacuo. The remainder was dis-
solved in ethanol (25 mL), hydrazine monohydrate (0.21 g, 4.2
mmol) was added, and the solution was acidified with acetic
acid and refluxed for 24 h. Next, the reaction mixture was
diluted with ethyl acetate and washed with 1 N HCl and water.
The organic layer was dried over MgSO4, and the solvents were
removed under reduced pressure. The remaining oil was
purified by flash column chromatography using 1:2 ethyl
acetate/petroleum ether (40-60 °C) to give enantiopure (-)-
12, which was crystallized from diethyl ether. Yield 0.40 g
(40%). A second crop of 0.28 g (28%) of enantiopure product
crystallized after concentration of the filtrate. Total yield 68%;
2.890(5) Å, respectively. Crystal data are the following: C21H26
N2O3‚0.5(CH3CH2)2O, Mr ) 354.5, monoclinic, P21, a
-
)
11.0427(7) Å, b ) 9.355(2) Å, c ) 21.186(2) Å, â ) 102.326(5)°,
V ) 2138.2(5) Å3, Z ) 4, Dx ) 1.22 g cm-3, λ(Cu KR) ) 1.5418
Å, µ(Cu KR) ) 6.19 cm-1, F(000) ) 844, room temperature.
Final R is 0.040 for 4323 observed reflections.
II. Ch em istr y. Gen er a l Meth od s a n d Ma ter ia ls.
Pyridine was stored over 4 Å molecular sieves. All the
other solvents were used as received. Starting materials
were commercially available. Tris[3-(trifluoromethylhydroxymeth-
ylene)-d-camphorato]europium(III) [Eu(tfc)3], tris(d,d-dicam-
pholylmethanato)europium(III) [Eu(dcm)3], and tris[3-(hep-
tafluoropropylhydroxymethylene)-d-camphorato]europium(III)
[Eu(hfc)3] were purchased from Fluka. Reactions were per-
formed under anhydrous conditions unless noted otherwise.
Friedel-Crafts acylations were performed under an N2 atmo-
sphere. Reactions were followed by TLC analysis on Merck
TLC aluminum sheets Silicagel 60 F254. Flash column chro-
matography was performed on silicagel, 30-60 µm (J . T.
Baker). Melting points were measured on an Electrothermal
IA9200 apparatus or a Mettler FP-5 + FP-052 apparatus
equipped with a microscope and are uncorrected. 1H NMR
spectra were recorded on a Bruker AC 200 (1H NMR: δ 200.1
mp 87-88 °C; >96% ee; [R]25 -339.5° (c 0.96, CHCl3). Anal.
D
(C21H26N2O3‚0.6C4H10O) C, H, N.
cis-(+)-4-(2,3-Dih yd r o-7-m et h oxyb en zofu r a n -2-sp ir o-
1′-cyclop en ta n -4-yl)-4a ,5,6,7,8,8a -h exa h yd r o-2H-p h th a l-
a zin -1-on e ((+)-12). (+)-12 was prepared from the (1R,2S)-
(-)-ephedrine salt of pure (+)-5 (1.10 g, 2.10 mmol) as
described for the preparation of (-)-12. The title compound
was crystallized from diethyl ether. Yield 0.10 g (13%). Another
0.40 g (53%) of enantiopure product crystallized upon concen-
tration of the filtrate. Total yield 66%; mp 87-88 °C; >96%
1
MHz). The H NMR chemical shifts (δ) are expressed in ppm
values relative to CDCl3 (δ ) 7.26 ppm). Abbreviations used
in description of NMR spectra are the following: s ) singlet,
d ) doublet, t ) triplet, q ) quartet, dd ) double doublet, dt
) double triplet, m ) multiplet, and bs ) broad singulet. The
limit of detection is <4% of one enantiomer in the other using
1H NMR spectroscopy. Specific rotations were measured on a
Perkin-Elmer 241 polarimeter. All phthalazinones had an
elemental analysis (C, H, and N) within (0.4% of the theoreti-
cal value. The synthesis of γ-keto acids (()-1-64,5 and phthal-
azinones (()-8-144-6 has been described in previous papers.
ee; [R]25 336.8° (c 1.0, CHCl3). Anal. (C21H26N2O3‚0.5C4H10O)
D
C, H, N.
The enantiomeric purity of (+)-12 and (-)-12 was deter-
1
mined by H NMR (CDCl3). Chiral shift reagent Eu(hfc)3 (12
mg, 0.7 equiv) was added to racemic (()-12 (5 mg), resulting
in a separation of the resonance of the methoxy group and
aromatic H5 into two resonances each. The resonances ap-
peared at 4.15 ppm (OCH3) and 7.34 ppm (H5-arom) for (-)-
12 and at 4.16 and 7.37 ppm for (+)-12. Independent treatment
of samples of (-)-12 and (+)-12 with Eu(hfc)3 (0.7 equiv)
resulted in retention of single resonances, and no detectable
contamination by the other enantiomer was found (>96% ee).
Resolu tion of cis-(()-2-(3,4-Dim eth oxyben zoyl)cyclo-
h exa n eca r boxylic Acid ((-)-1 a n d (+)-1). The resolution
was performed in a similar way as described for γ-keto acid
(()-5 using (R)-(+)-R-methylbenzylamine (2.01 g, 16.6 mmol),
γ-keto acid (()-1 (4.85 g, 16.6 mmol), and ethanol (200 mL).
Within 15 min a white precipitate was formed. The crystals
(2.51 g) were collected after 16 h. A second crop (0.64 g) was
obtained after concentration of the mother liquor to 100 mL
Resolu tion of cis-(()-2-(2,3-Dih yd r o-7-m eth oxyben zo-
fu r a n -2-sp ir o-1′-cyclop en ta n e-4-ca r bon yl)cycloh exa n e-
ca r boxylic Acid ((-)-5 a n d (+)-5). A stirred solution of
γ-keto acid (()-5 (12.5 g, 34.8 mmol) in ethyl acetate (250 mL)
was treated with quinine (11.3 g, 34.8 mmol). Crystallization
occurred within 30 min. After 4 h the crystalline precipitate
was filtered off and washed twice with ethyl acetate. The
product was dried in vacuo to give 10.4 g of quinine salt of
1
crude (-)-5, as determined by H NMR (see below). The solid
was recrystallized from ethyl acetate to afford 9.65 g (81%) of
pure (-)-5 as quinine salt: >96% de; mp 165-167 °C; [R]25
D
-101.8° (c 1.0, CHCl3). A small sample of the diastereomeric

Phosphodiesterase Inhibitors
J ournal of Medicinal Chemistry, 2002, Vol. 45, No. 12 2531
and stirring the mixture for 4 h. The combined diastereomeric
salts were recrystallized from ethanol. Yield 2.31 g (67%);
The other antipode was obtained using crude γ-keto acid
(-)-4, ethyl acetate (250 mL), and quinine (7.64 g, 23.6 mmol).
The crude diastereomeric salt was filtered off after the mixture
was stirred overnight and was recrystallized from ethyl
>96% de; mp 160-162 °C; [R]25 24.8° (c 0.97, CHCl3).
D
(+)-1: [R]25 9.87° (c 1.0, CHCl3).
D
acetate. Yield 11.5 g (73%); >96% de; mp 167-170 °C; [R]25
D
The other antipode was obtained using crude γ-keto acid
(-)-1, ethanol (100 mL), and (S)-(-)-R-methylbenzylamine
(1.00 g, 8.25 mmol). The pure diastereomeric salt was collected
-103.6° (c 1.1, CHCl3).
(-)-4: [R]25 -20.4° (c 0.92, CHCl3).
D
1H NMR (CDCl3) of (()-4 plus quinine (1 equiv) showed
resonances for aromatic H5 at 6.45 ppm for (+)-4 and 6.66
ppm for (-)-4. No detectable contamination by the other
enantiomer was found (>96% ee) upon independent treatment
of samples of (-)-4 and (+)-4 with quinine (1 equiv).
after 16 h. Yield 2.35 g (69%); >96% de; mp 153-156 °C; [R]25
D
-25.3° (c 1.0, CHCl3).
(-)-1: [R]25 -9.51° (c 0.86, CHCl3).
D
1H NMR (CDCl3) of (()-1 plus quinine (1 equiv) showed
resonances for aromatic H5 at 6.73 ppm for (+)-1 and 6.49
ppm for (-)-1. No detectable contamination by the other
enantiomer was found (>96% ee) upon independent treatment
of samples of (-)-1 and (+)-1 with quinine (1 equiv).
Resolu tion of cis-(()-2-(2,3-Dih yd r o-7-m eth oxyben zo-
fu r an -2-spir o-1′-cyclopen tan e-4-car bon yl)-1,2,3,6-tetr ah y-
d r oben zoic Acid ((-)-6 a n d (+)-6). The resolution was
performed in a similar way as described for γ-keto acid (()-5,
using quinine (9.02 g, 27.8 mmol), γ-keto acid (()-6 (9.90 g,
27.8 mmol), and ethyl acetate (350 mL). Within 1 h a white
precipitate was formed. The crude crystals (8.74 g) were
collected after the mixture was stirred overnight and were
recrystallized from ethyl acetate. Yield 7.25 g (77%); >96% de;
Resolu tion of cis-(()-6-(3,4-Dim eth oxyben zoyl)cyclo-
h ex-3-en eca r boxylic Acid ((-)-2 a n d (+)-2). The resolution
was performed in a similar way as described for γ-keto acid
(()-5, using (S)-(-)-R-methylbenzylamine (2.29 g, 18.9 mmol),
γ-keto acid (()-2 (5.49 g, 18.9 mmol), and ethyl acetate (100
mL). Within 30 min a white precipitate was formed. The
crystals (2.00 g) were collected after 16 h. A second crop (0.41
g) was obtained after concentration of the mother liquor and
stirring the mixture for 72 h. The combined diastereomeric
salts were recrystallized from ethyl acetate. Yield 2.26 g (58%);
mp 163-165 °C; [R]25 -107.2° (c 0.99, CHCl3).
D
(-)-6: [R]25 -40.1° (c 1.0, CHCl3).
D
The other antipode was obtained using crude γ-keto acid
(+)-6, ethanol (200 mL), and (1R,2S)-(-)-ephedrine (2.30 g,
13.9 mmol). The pure diastereomeric salt was collected after
the mixture was stirred for 72 h. Yield 2.03 g (28%); >96%
>96% de; mp 136-139 °C; [R]25 5.09° (c 0.98, CHCl3).
D
(+)-2: [R]25 29.1° (c 0.97, CHCl3).
D
de; mp 126-128 °C; [R]25 13.8° (c 1.0, CHCl3).
D
The other antipode was obtained using crude γ-keto acid
(-)-2, ethyl acetate (100 mL), and (R)-(+)-R-methylbenzyl-
amine (1.15 g, 9.49 mmol). The crude diastereomeric salt was
filtered off after the mixture was stirred overnight and was
recrystallized from ethyl acetate. Yield 2.95 g (76%); >96% de;
(+)-6: [R]25 37.5° (c 1.0, CHCl3).
D
1H NMR (CDCl3) of (()-6 plus quinine (1 equiv) showed
resonances for aromatic H5 at 6.62 ppm for (+)-6 and 6.65
ppm for (-)-6. No detectable contamination by the other
enantiomer was found (>96% ee) upon independent treatment
of samples of (-)-6 and (+)-6 with quinine (1 equiv).
mp 138-141 °C; [R]25 -6.57° (c 1.1, CHCl3).
D
(-)-2: [R]25 -27.4° (c 0.98, CHCl3).
D
cis-6-(3,4-Diet h oxyben zoyl)cycloh ex-3-en eca r b oxylic
Acid ((()-7). Aluminum chloride (73.3 g, 0.55 mol) was
suspended in dichloromethane (800 mL), the mixture was
cooled to 5 °C, and 1,2-diethoxybenzene (83.1 g, 0.50 mol) was
added. After 30 min, cis-1,2,3,6-tetrahydrophthalic anhydride
(76.1 g, 0.50 mol) was added, and the reaction mixture was
allowed to reach room temperature and was refluxed for 4 h.
The mixture was poured into ice-water, the organic layer was
dried over magnesium sulfate, and the solvent was removed
in vacuo. The remainder was dissolved in CH2Cl2 and filtered
over silicagel to remove the dicarboxylic acid formed during
workup. The organic layer was extracted with 1 N NaOH, and
the combined basic extracts were acidified and extracted with
diethyl ether. The associated organic extracts were dried over
MgSO4 and concentrated under reduced pressure. The title
compound crystallized from diethyl ether as a white solid.
Yield 93.7 g (59%). Mp 125-127 °C. 1H NMR (CDCl3): δ 1.38-
1.59 (m, 6H, CH3), 2.32-2.60 (m, 3H, CH2, CHH′), 2.72-3.10
(m, 2H, CHH′, H1), 3.89-4.04 (m, 1H, H6), 4.05-4.27 (m, 4H,
1H NMR (CDCl3) of (()-2 plus quinine (1 equiv) showed
resonances for aromatic H5 at 6.59 ppm for (+)-2 and 6.70
ppm for (-)-2. No detectable contamination by the other
enantiomer was found (>96% ee) upon independent treatment
of samples of (-)-2 and (+)-2 with quinine (1 equiv).
Resolu tion of cis-(()-2-(3-Cyclop en tyloxy-4-m eth oxy-
ben zoyl)cycloh exa n eca r boxylic Acid ((-)-3 a n d (+)-3).
The resolution was performed in a similar way as described
for γ-keto acid (()-5, using quinine (7.32 g, 22.6 mmol), γ-keto
acid (()-3 (7.81 g, 22.6 mmol), and ethyl acetate (200 mL).
Within 30 min a white precipitate was formed. The crystals
(5.59 g) were collected after 16 h. A second crop (2.13 g) was
obtained after concentration of the mother liquor and stirring
the mixture for 16 h. The combined diastereomeric salts were
recrystallized from ethyl acetate. Yield 5.30 g (70%); >96% de;
mp 165-168 °C; [R]25 -91.0° (c 0.97, CHCl3).
D
(+)-3: [R]25 15.0° (c 0.96, CHCl3).
D
The other antipode was obtained using crude γ-keto acid
(-)-3, ethanol (100 mL), and (1R,2S)-(-)-ephedrine (1.86 g,
11.3 mmol). The crude diastereomeric salt was filtered off after
the mixture was stirred for 72 h and was recrystallized from
3
OCH2), 5.58-5.88 (m, 2H, HCdCH), 6.87 (d, 1H, J ) 8.2 Hz,
H5-arom), 7.43-7.59 (m, 2H, H2-arom, H6-arom).
Resolu tion of cis-(()-6-(3,4-Dieth oxyben zoyl)cycloh ex-
3-en eca r boxylic Acid ((+)-7). The resolution was performed
in a similar way as described for γ-keto acid (()-5, using (S)-
(-)-R-methylbenzylamine (2.29 g, 18.9 mmol), γ-keto acid (()-7
(6.01 g, 18.9 mmol), and ethyl acetate (100 mL). The solution
was kept at 5 °C for 72 h. Subsequently, the mixture was
allowed to reach room temperature, upon which within 30 min
a white precipitate was formed. The pure diastereomeric salt
(1.89 g) was collected after 16 h. A second crop (0.96 g) was
collected after concentration of the mother liquor. Total yield
ethanol. Yield 2.01 g (35%); >96% de; mp 143-145 °C; [R]25
D
-32.4° (c 1.1, CHCl3).
(-)-3: [R]25 -13.9° (c 1.0, CHCl3).
D
1H NMR (CDCl3) of (()-3 plus quinine (1 equiv) showed
resonances for aromatic H5 at 6.76 ppm for (+)-3 and 6.47
ppm for (-)-3. No detectable contamination by the other
enantiomer was found (>96% ee) upon independent treatment
of samples of (-)-3 and (+)-3 with quinine (1 equiv).
Resolu tion of cis-(()-6-(3-Cyclop en tyloxy-4-m eth oxy-
ben zoyl)cycloh ex-3-en eca r boxylic Acid ((-)-4 a n d (+)-
4). The resolution was performed in a similar way as described
for γ-keto acid (()-5, using (1R,2S)-(-)-ephedrine (7.78 g, 47.1
mmol), γ-keto acid (()-4 (16.2 g, 47.1 mmol), and ethyl acetate
(350 mL). Within 1 h a white precipitate was formed. The
crude product (15.4 g) was collected after 72 h and was
recrystallized from ethyl acetate. Yield 3.40 g (28%); >96% de;
69%; >96% de; mp 116-118 °C; [R]25 1.87° (c 0.99, CHCl3).
D
(+)-7: [R]25 27.4° (c 5.2, CHCl3). The 1H NMR (CDCl3)
D
spectrum of (()-7 plus quinine (1 equiv) showed resonances
for aromatic H5 at 6.56 ppm for (+)-7 and 6.71 ppm for (-)-7.
No detectable contamination by the other enantiomer was
found (>96% ee) upon treatment of a sample of (+)-7 with
quinine (1 equiv).
mp 133-135 °C; [R]25 -1.46° (c 1.0, CHCl3).
D
cis-4-(3,4-Dim eth oxyp h en yl)-4a ,5,6,7,8,8a -h exa h yd r o-
2H-p h th a la zin -1-on e ((-)-8 a n d (+)-8). A mixture of γ-keto
(+)-4: [R]25 20.8° (c 1.0, CHCl3).
D

2532 J ournal of Medicinal Chemistry, 2002, Vol. 45, No. 12
Van der Mey et al.
acid (+)-1 (0.84 g, 2.9 mmol), hydrazine monohydrate (0.15 g,
3.0 mmol), and ethanol (30 mL) was acidified with acetic acid
and refluxed for 4 h. Workup was performed as described for
(-)-12. The title compound ((-)-8) was crystallized from diethyl
For 1H NMR (CDCl3), chiral shift reagent Eu(dcm)3 (0.7
equiv) was used to determine the enantiomeric purity of the
enantiomers. The resonances of the methoxy group appeared
at 3.99 ppm for (-)-11 and 3.97 ppm for (+)-11. No detectable
contamination by the other enantiomer was found (>96% ee)
upon independent treatment of samples of (-)-11 and (+)-11
with Eu(dcm)3 (0.7 equiv).
cis-4-(2,3-Dih yd r o-7-m eth oxyben zofu r a n -2-sp ir o-1′-cy-
clop en t a n -4-yl)-4a ,5,8,8a -t et r a h yd r o-2H -p h t h a la zin -1-
on e ((-)-13 a n d (+)-13). (-)-13 was synthesized analogously
to the preparation of (-)-8 using γ-keto acid (-)-6 (1.57 g, 4.41
mmol) and hydrazine monohydrate (0.33 g, 6.6 mmol). The title
compound ((-)-13) was crystallized from diethyl ether. Yield
0.90 g (58%); >96% ee; mp 185-186 °C; [R]25D -616.2° (c 0.99,
CHCl3). Anal. (C21H24N2O3‚0.2H2O) C, H, N.
ether. Yield 0.24 g (29%); >96% ee; mp 153-155 °C; [R]25
D
-452.0° (c 0.96, CHCl3). Anal. (C16H20N2O3) C, H, N.
Analogue (+)-8 was synthesized analogously to the prepara-
tion of (-)-8 using γ-keto acid (-)-1 (0.81 g, 2.8 mmol). The
product was purified by flash column chromatography using
3:1 petroleum ether (60-80 °C)/ethyl acetate. The title com-
pound ((+)-8) was crystallized from diethyl ether. Yield 0.31
g (39%); >92% ee; mp 153-155 °C; [R]25D 441.1° (c 1.1, CHCl3).
Anal. (C16H20N2O3) C, H, N.
1
For H NMR (CDCl3), chiral shift reagent Eu(tfc)3 (0.9 equiv)
was used to determine the enantiomeric purity of the enan-
tiomers. The resonances of the methoxy groups appeared at
4.05 and 4.16 ppm for (-)-8 and 4.03 and 4.15 ppm for (+)-8.
No detectable contamination by the other enantiomer was
found (>96% ee) upon independent treatment of samples of
(-)-8 and (+)-8 with Eu(tfc)3 (0.9 equiv).
Analogue (+)-13 was synthesized analogously to the prepa-
ration of (-)-8 using γ-keto acid (+)-6 (1.36 g, 3.82 mmol) and
hydrazine monohydrate (0.29 g, 5.8 mmol). The title compound
((+)-13) was crystallized from diethyl ether. Yield 0.89 g (66%);
>96% ee; mp 182-183 °C; [R]25 598.7° (c 1.1, CHCl3). Anal.
D
(C21H24N2O3‚0.2H2O) C, H, N.
cis-4-(3,4-Dim eth oxyp h en yl)-4a ,5,8,8a -tetr a h yd r o-2H-
p h th a la zin -1-on e ((-)-9 a n d (+)-9). (-)-9 was synthesized
analogously to the preparation of (-)-8 using γ-keto acid (-)-2
(1.20 g, 4.14 mmol). The title compound ((-)-9) was crystallized
from diethyl ether. Yield 0.84 g (71%); >80% ee; mp 151-154
For 1H NMR (CDCl3), chiral shift reagent Eu(hfc)3 (0.9
equiv) was used to determine the enantiomeric purity of the
enantiomers. The resonance of the methoxy group and aro-
matic H5 appeared at 4.25 and 7.55 ppm, respectively, for (-)-
13 and at 4.26 and 7.57 ppm, respectively, for (+)-13. No
detectable contamination by the other enantiomer was found
(>96% ee) upon independent treatment of samples of (-)-13
and (+)-13 with Eu(hfc)3 (0.9 equiv).
°C; [R]25 -678.4° (c 0.98, CHCl3). Anal. (C16H18N2O3) C, H,
D
N.
Analogue (+)-9 was synthesized analogously to the prepara-
tion of (-)-8 using γ-keto acid (+)-2 (1.64 g, 5.66 mmol) and
hydrazine monohydrate (0.42 g, 8.4 mmol). The title compound
((+)-9) was crystallized from diethyl ether. Yield 1.41 g (87%);
cis-2-Ad a m a n ta n -2-yl-4-(3,4-d im eth oxyp h en yl)-4a ,5,8,-
8a -tetr a h yd r o-2H-p h th a la zin -1-on e ((+)-14 a n d (-)-14).
A solution of γ-keto acid (+)-1 (0.64 g, 2.2 mmol) and adaman-
tylhydrazine hydrochloride (0.54 g, 2.7 mmol) in pyridine (50
mL) was refluxed for 5 h. The reaction mixture was concen-
trated in vacuo, and the remainder was dissolved in ethyl
acetate and washed with 1 N HCl and a solution of NaHCO3.
The organic layer was dried over MgSO4 and concentrated
under reduced pressure, and the title compound ((+)-14) was
crystallized from diethyl ether. Yield 60%; >96% ee; mp 157-
159 °C; [R]25D 574.0° (c 0.99, CHCl3). Anal. (C26H32N2O3) C, H,
N.
>96% ee; mp 152-156 °C; [R]25 726.1° (c 0.97, CHCl3). Anal.
D
(C16H18N2O3) C, H, N.
For 1H NMR (CDCl3), chiral shift reagent Eu(dcm)3 (0.9
equiv) was used to determine the enantiomeric purity of the
enantiomers. The resonances of the methoxy groups appeared
at 4.02 and 4.11 ppm for (-)-9 and at 4.07 and 4.13 ppm for
(+)-9. No detectable contamination by the other enantiomer
was found (>96% ee) upon independent treatment of samples
of (-)-9 and (+)-9 with Eu(dcm)3 (0.9 equiv).
cis-4-(3-Cyclop en tyloxy-4-m eth oxyp h en yl)-4a ,5,6,7,8,-
8a -h exa h yd r o-2H-p h th a la zin -1-on e ((-)-10 a n d (+)-10).
(-)-10 was synthesized analogously to the preparation of (-)-8
using γ-keto acid (+)-3 (1.46 g, 4.22 mmol) and hydrazine
monohydrate (0.32 g, 6.4 mmol). The title compound ((-)-10)
was crystallized from diethyl ether. Yield 1.19 g (82%); >96%
Analogue (-)-14 was synthesized analogously to the prepa-
ration of (+)-14 using γ-keto acid (-)-1 (2.0 g, 3.3 mmol) and
adamantylhydrazine hydrochloride (1.0 g, 5.0 mmol). The title
compound ((-)-14) was crystallized from petroleum ether (60-
80 °C)/diethyl ether. Yield 58%; >96% ee; mp 157-159 °C;
ee; mp 119-120 °C; [R]25 -383.6° (c 1.0, CHCl3). Anal.
[R]25 -582.9° (c 1.0, CHCl3). Anal. (C26H32N2O3) C, H, N.
D
D
1
(C20H26N2O3) C, H, N.
For H NMR (CDCl3), chiral shift reagent Eu(tfc)3 (0.9 equiv)
Analogue (+)-10 was synthesized analogously to the prepa-
ration of (-)-8 using γ-keto acid (-)-3 (1.50 g, 4.34 mmol) and
hydrazine monohydrate (0.43 g, 8.6 mmol). The title compound
((+)-10) was crystallized from diethyl ether. Yield 1.19 g (80%);
was used to determine the enantiomeric purity of the enan-
tiomers. The resonances of the aromatic H5 appeared at 7.08
ppm for (-)-14 and 7.11 ppm for (+)-14. No detectable
contamination by the other enantiomer was found (>96% ee)
upon treatment of a sample of (+)-14 with Eu(tfc)3 (0.9 equiv).
cis-4-[4-(3,4-Dieth oxyp h en yl)-1-oxo-4a ,5,8,8a -tetr a h y-
d r o-1H-p h th a la zin -2-yl]ben zoic Acid ((()-15). (()-15 was
prepared in a similar way as described for (+)-14 using γ-keto
acid (()-7 (10.0 g, 31.4 mmol), 4-hydrazinobenzoic acid (9.0 g,
59 mmol), and pyridine hydrochloride (8.0 g, 69 mmol). The
title compound was crystallized from ethyl acetate. Yield 71%.
Mp 204-207 °C. 1H NMR (CDCl3): δ 1.49 (t, 6H, 3J ) 6.9 Hz,
CH3), 2.17-2.44 (m, 3H, H5, H8), 2.98-3.19 (m, 2H, H8a, H8′),
3.39-3.57 (m, 1H, H4a), 4.15 (q, 4H, 3J ) 6.9 Hz, OCH2), 5.65-
5.91 (m, 2H, HCdCH), 6.91 (d, 1H, 3J ) 8.5 Hz, H5-arom),
>96% ee; mp 119-120 °C; [R]25 384.7° (c 1.0, CHCl3). Anal.
D
(C20H26N2O3) C, H, N.
1
For H NMR (CDCl3), chiral shift reagent Eu(tfc)3 (0.7 equiv)
was used to determine the enantiomeric purity of the enan-
tiomers. The resonance of the methoxy group and aromatic
H5 appeared at 4.04 and 7.20 ppm for (-)-10 and at 4.03 and
7.18 ppm for (+)-10. No detectable contamination by the other
enantiomer was found (>96% ee) upon independent treatment
of samples of (-)-10 and (+)-10 with Eu(tfc)3 (0.7 equiv).
cis-4-(3-Cyclop en t yloxy-4-m et h oxyp h en yl)-4a ,5,8,8a -
tetr a h yd r o-2H-p h th a la zin -1-on e ((-)-11 a n d (+)-11). (-)-
11 was synthesized analogously to the preparation of (-)-8
using γ-keto acid (-)-4 (0.98 g, 2.9 mmol). The title compound
((-)-11) was crystallized from ethyl acetate/petroleum ether
4
3
7.34 (dd, 1H, J ) 2.0 Hz, J ) 8.5 Hz, H6-arom), 7.53 (d, 1H,
3
4J ) 2.0 Hz, H2-arom), 7.83 (d, 2H, J ) 8.8 Hz, H-Ph), 8.15
3
(d, 2H, J ) 8.8 Hz, H-Ph). Anal. (C25H26N2O5) C, H, N.
(60-80 °C). Yield 0.65 g (67%); >92% ee; mp 98-102 °C; [R]25
-605.1° (c 1.0, CHCl3). Anal. (C20H24N2O3) C, H, N.
D
cis-4-[4-(3,4-Dieth oxyp h en yl)-1-oxo-4a ,5,8,8a -tetr a h y-
d r o-1H-p h th a la zin -2-yl]ben zoic Acid ((+)-15). (+)-15 was
prepared in a similar way as described for (+)-14 using γ-keto
acid (+)-7 (1.84 g, 5.79 mmol), 4-hydrazinobenzoic acid (1.50
g, 9.86 mmol), and pyridine hydrochloride (1.0 g, 8.7 mmol).
The title compound was crystallized from ethanol/water. Yield
Analogue (+)-11 was synthesized analogously to the prepa-
ration of (-)-8 using γ-keto acid (+)-4 (1.50 g, 4.36 mmol) and
hydrazine monohydrate (0.44 g, 8.8 mmol). The title compound
((+)-11) was crystallized from diethyl ether. Yield 1.11 g (75%);
>96% ee; mp 125-126 °C; [R]25 626.4° (c 0.98, CHCl3). Anal.
59%; >96% ee; mp 204-207 °C; [R]25 516.4° (c 0.96, CHCl3).
D
D
(C20H24N2O3) C, H, N.
Anal. (C25H26N2O5) C, H, N.

Phosphodiesterase Inhibitors
J ournal of Medicinal Chemistry, 2002, Vol. 45, No. 12 2533
For 1H NMR (CDCl3), no suitable shift reagent was found
(4) Van der Mey, M.; Hatzelmann, A.; Van der Laan, I. J .; Sterk,
G. J .; Thibaut, U.; Timmerman, H. Novel Selective PDE4
Inhibitors. 1. Synthesis, Structure-Activity Relationships, and
Molecular Modeling of 4-(3,4-Dimethoxyphenyl)-2H-phthalazin-
1-ones and Analogues. J . Med. Chem. 2001, 44, 2511-2522.
(5) Van der Mey, M.; Hatzelmann, A.; Van Klink, G. P. M.; Van
der Laan, I. J .; Sterk, G. J .; Thibaut, U.; Ulrich, W. R.;
Timmerman, H. Novel Selective PDE4 Inhibitors. 2. Synthesis
and Structure-Activity Relationships of 4-Aryl-Substituted cis-
Tetra- and cis-Hexahydrophthalazinones. J . Med. Chem. 2001,
44, 2523-2535.
(6) Van der Mey, M.; Boss, H.; Hatzelmann, A.; Van der Laan, I.
J .; Sterk, G. J .; Timmerman, H. Unpublished results. Novel
Selective PDE4 Inhibitors. 3. In Vivo Anti-inflammatory Activity
of a New Series of N-Substituted cis-Tetra- and cis-Hexahydro-
phthalazinones. J . Med. Chem. 2002, 45, 2520-2525.
(7) About 4% contamination of one enantiomer by the other in the
resolved samples could be detected by 1H NMR.
(8) J onas, R.; Klockow, M.; Lues, I.; Pru¨cher, H.; Schliep, H. J .;
Wurziger, H. Synthesis and biological activities of meribendan
and related heterocyclic benzimidazolo-pyridazinones. Eur. J .
Med. Chem. 1993, 28, 129-140.
(9) Marivet, M. C.; Bourguignon, J . J .; Lugnier, C.; Mann, A.;
Stoclet, J . C.; Wermuth, C. G. Inhibition of cyclic adenosine-
3′,5′-monophosphate phosphodiesterase from vascular smooth
muscle by rolipram analogues. J . Med. Chem. 1989, 32, 1450-
1457.
to ascertain the enantiomeric purity of (+)-15. However, it was
1
possible to determine the enantiomeric purity of (+)-16 by H
NMR (CDCl3) (>96% ee, see below), and because acid (+)-15
was used as a starting material for (+)-16, it is assumed that
it has the same or higher enantiomeric purity.
cis-(()-4-[4-(3,4-Diet h oxyp h en yl)-1-oxo-4a ,5,8,8a -t et -
r a h yd r o-1H -p h t h a la zin -2-yl]b en zoic Acid E t h yl E st er
((()-16). Benzoic acid (()-15 (1.00 g, 2.30 mmol) was dissolved
in ethanol (50 mL), and acetyl chloride (2.0 mL, 2.2 g, 28.1
mmol) was added. After the mixture was stirred overnight,
the solvent was removed in vacuo and the title compound was
crystallized from ethanol. Yield 85%. Mp 128-129 °C. 1H NMR
(CDCl3): δ 1.40 (t, 3H, 3J ) 7.1 Hz, CH3), 1.48 (t, 3H, 3J ) 7.0
Hz, CH3), 1.49 (t, 3H, 3J ) 7.0 Hz, CH3), 2.17-2.42 (m, 3H,
H5, H8), 2.95-3.18 (m, 2H, H8a, H8′), 3.38-3.56 (m, 1H, H4a),
4.14 (q, 2H, 3J ) 7.0 Hz, OCH2), 4.15 (q, 2H, 3J ) 7.0 Hz,
3
OCH2), 3.88 (q, 2H, J ) 7.1 Hz, CH2CO2), 5.65-5.91 (m, 2H,
HCdCH), 6.90 (d, 1H, 3J ) 8.5 Hz, H5-arom), 7.33 (dd, 1H, 4J
) 2.1 Hz, 3J ) 8.5 Hz, H6-arom), 7.52 (d, 1H, 4J ) 2.0 Hz,
3
3
H2-arom), 7.77 (d, 2H, J ) 8.7 Hz, H-Ph), 8.08 (d, 2H, J )
8.7 Hz, H-Ph). Anal. (C27H30N2O5) C, H, N.
cis-(+)-4-[4-(3,4-Diet h oxyp h en yl)-1-oxo-4a ,5,8,8a -t et -
r a h yd r o-1H -p h t h a la zin -2-yl]b en zoic Acid E t h yl E st er
((+)-16). (+)-16 was prepared as described for (()-16 from
benzoic acid (+)-15 (1.00 g, 2.30 mmol). The title compound
was crystallized from ethanol at -20 °C. Yield 65%; >96% ee;
mp 128-129 °C; [R]25D 543.4° (c 0.99, CHCl3). Anal. (C27H30N2O5)
C, H, N.
(10) Hughes, B.; Owens, R.; Perry, M.; Warrellow, G.; Allen, R. PDE
IV inhibitors: the use of molecular cloning in the design and
development of novel drugs. Drug Discovery Today 1997, 2, 89-
101.
(11) Christensen, S. B.; Guider, A.; Forster, C. J .; Gleason, J . G.;
Bender, P. E.; Karpinski, J . M.; De Wolf, W. E., J r.; Barnette,
M. S.; Underwood, D. C.; Griswold, D. E.; Cieslinski, L. B.;
Burman, M.; Bochnowicz, S.; Osborn, R. R.; Manning, C. D.;
Grous, M.; Hillegas, L. M.; Bartus, J . O.; Ryan, M. D.; Eggleston,
D. S.; Haltiwanger, R. C.; Torphy, T. J . 1,4-Cyclohexanecarboxy-
lates: potent and selective inhibitors of phosphodiesterase 4 for
the treatment of asthma. J . Med. Chem. 1998, 41, 821-835.
(12) De Gelder, R.; De Graaff, R. A. G.; Schenk, H. Automatic
Determination of Crystal Structures Using Karle-Hauptman
Matrices. Acta Crystallogr. 1993, A49, 287-293.
(13) Larson, A. C. The inclusion of secondary extinction in least-
squares refinement of crystal structures. In Crystallographic
Computing; Ahmed, F. R., Hall, S. R., Huber, C. P., Eds.;
Munksgaard: Copenhagen, 1969; pp 291-294.
(14) Zachariasen, W. H. A general theory of X-ray diffraction in
crystals. Acta Crystallogr. 1967, A23, 558.
(15) Flack, H. D. On Enantiomorph-Polarity Estimation. Acta Crys-
tallogr. 1983, A39, 876.
(16) Cromer, D. T.; Mann, J . B. X-ray scattering factors computed
from numerical Hartree-Fock wavefunctions. Acta Crystallogr.,
Sect. A 1968, 24, 321-324.
(17) International Tables for X-ray Crystallographyl D. Reidel: Dor-
drecht, The Netherlands, 1974; Vol. IV, p 55.
1
For H NMR (CDCl3), chiral shift reagent Eu(hfc)3 (2 equiv)
was used to determine the enantiomeric purity of (+)-16. The
resonances of the aromatic Hortho-Bz appeared at 8.68 ppm
for (-)-16 and 8.70 ppm for (+)-16. No detectable contamina-
tion by the other enantiomer was found (>96% ee) upon
independent treatment of a sample of (+)-16 with Eu(hfc)3 (2
equiv).
Refer en ces
(1) Van der Mey, M.; Van der Laan, I. J .; Timmerman, H.;
Hatzelmann, A.; Boss, H.; Ha¨fner, D.; Beume, R.; Kley, H. P.;
Sterk, G. J . Preparation of arylphthalazinones as inhibitors of
cyclic nucleotide phosphodiesterase. Patent WO 9831674,
1998.
(2) Hatzelmann, A.; Boss, H.; Ha¨fner, D.; Beume, R.; Kley, H. P.;
Van der Mey, M.; Van der Laan, I. J .; Timmerman, H.; Sterk,
G. J . Preparation of phthalazinones as phosphodiesterase IV
inhibitors. Patent WO 9931071, 1999.
(3) Thibaut, U.; Hatzelmann, A.; Boss, H.; Ha¨fner, D.; Beume, R.;
Kley, H. P.; Timmerman, H.; Van der Laan, I. J .; Ulrich, W. R.;
Sterk, G. J .; Van der Mey, M. Preparation of phthalazinyldihy-
drobenzofurans as cyclic nucleotide phosphodiesterase inhibitors.
Patent WO 9931090, 1999.
(18) XTAL3.4 User’s Manual; Hall, S. R., King, G. S. D., Stewart, J .
M., Eds.; University of Western Australia: Lamb, Perth, Aus-
tralia, 1995.
J M0110338