Novel cyclic compounds as potent phosphodiesterase 4 inhibitors

Article abstract of DOI:10.1021/jm970575f

The synthesis and biological activity of a novel series of 2,2- disubstituted indan-1,3-dione-based PDE4 inhibitors are described. This structurally unique class of PDE4 inhibitors is markedly different from the known PDE4 inhibitors such as RP 73401 (2) and CDP 840 (3). Structure- activity relationship (SAR) studies led to the identification of inhibitors with nanomolar potency and oral activity in a murine endotoxemia model for TNF-α inhibition. Unlike other classical PDE4 inhibitors, several analogues were found to be nonemetic in a canine emesis model at intravenous doses of up to 3 mg/kg.

Full text of DOI:10.1021/jm970575f

4216
J . Med. Chem. 1998, 41, 4216-4223
Novel Cyclic Com p ou n d s a s P oten t P h osp h od iester a se 4 In h ibitor s
Wei He,* Fu-Chih Huang, Barbara Hanney, J ohn Souness, Bruce Miller, Guyan Liang, J on Mason,† and
Stevan Djuric‡
Department of Medicinal Chemistry, SW 8 Rhoˆne-Poulenc Rorer Central Research, 500 Arcola Road,
Collegeville, Pennsylvania 19426
Received August 28, 1997
The synthesis and biological activity of a novel series of 2,2-disubstituted indan-1,3-dione-
based PDE4 inhibitors are described. This structurally unique class of PDE4 inhibitors is
markedly different from the known PDE4 inhibitors such as RP 73401 (2) and CDP 840 (3).
Structure-activity relationship (SAR) studies led to the identification of inhibitors with
nanomolar potency and oral activity in a murine endotoxemia model for TNF-R inhibition.
Unlike other classical PDE4 inhibitors, several analogues were found to be nonemetic in a
canine emesis model at intravenous doses of up to 3 mg/kg.
Ch a r t 1
In tr od u ction
Tumor necrosis factor-R (TNF-R) has been implicated
in the pathogenesis of a number of autoimmune and
inflammatory diseases including rheumatoid arthritis
and sepsis.1 Consequently, manipulation of the cytokine
signaling or biosynthetic pathways associated with
these proteins may provide therapeutic benefit in the
above-mentioned disease states. It has been demon-
strated that TNF-R production in proinflammatory cells
becomes attenuated by an elevation of intracellular
cyclic adenosine 3′,5′-monophosphate (cAMP).2 This
second messenger is regulated by the phosphodiesterase
(PDE) family of enzymes. Inhibition of one such en-
zyme, the calcium-independent and cAMP-specific PDE4,
represents an attractive approach toward immunoin-
flammatory disease therapy.3
The prototype PDE4 inhibitor rolipram (1)4 (Chart 1)
exhibits several untoward side effects, which have
precluded its clinical development as an antidepressant
drug.5 Most of the other PDE4 inhibitors3e,6b also
possess side effects such as emesis; therefore, their
therapeutic applications have been limited. PDE4 has
been shown to have two binding sites for rolipram,
consisting of a high-affinity site (Sr, Ki ) 10 nM) and a
catalytic site (Sc, Ki ) 300 nM).7 It has been suggested
that some of the side effects are associated with the
high-affinity binding site.8,6b-e,3e Recently, Warrellow
et al. have disclosed a new class of inhibitors that
display selectivity for Sc over Sr and claimed that no
adverse effects in phase I human clinical trials were
observed with the prototype inhibitor CDP 840 (3).9
Overall, the mechanism responsible for the side effects
of PDE4 inhibitors is still not well-understood.
RP 73401 (2) is a potent benzamide PDE4 inhibitor,6a
which shows no selectivity for inhibiting PDE4 (catalytic
site) over the displacement of rolipram from its binding
site. The orientation of the aromatic groups in 2 is
dependent on the amide linker as well as the dichloro
substitution.6a In addition, it is conceivable that the
amide bond can undergo hydrolysis, and thus undesir-
able metabolites might be generated. It was decided
to replace the amide linker of 2 with a conformationally
constrained indan ring linker which cannot be cleaved.
The indan ring in the initial template (4) dictated the
orientation of the pyridyl and 4-methoxy-3-cyclopentoxy-
phenyl rings. Since PDE4 has been shown to have two
different binding sites for rolipram, substitutions on the
aromatic ring of indan with functional groups could
provide different interactions (through hydrogen bond-
ing and van der Waals forces). In contrast, the benz-
amide-type PDE4 inhibitors such as 26a,9d cannot engage
in such interactions. Our ultimate goals were to
discover PDE4 inhibitors that maintain antiinflamma-
tory properties but with reduced side effects and to
assess the role of selectivity of the catalytic over high-
affinity binding sites in contributing to the side effects.
Several conformational studies of 1 and its analogues
have been published.6f,g However, it is not feasible to
use molecular modeling to study the interaction of PDE4
and its inhibitors since neither the three-dimensional
structure of PDE4 nor its homology model is available.
Molecular modeling was applied for the analysis of
conformational overlaps of target compounds and 2. It
was used as a tool to select target compounds for
synthesis and was not used to interpret structure-
activity relationships (SAR) in detail. A superposition
† Current address: Bristol-Myers Squibb, Princeton, NJ .
‡ Current address: Abbott Laboratories, Abbott Park, IL.
10.1021/jm970575f CCC: $15.00 © 1998 American Chemical Society
Published on Web 10/02/1998

Novel Cyclic Compounds as PDE4 Inhibitors
J ournal of Medicinal Chemistry, 1998, Vol. 41, No. 22 4217
Sch em e 1. Synthesis of the Cyclic Compoundsa
F igu r e 1. Superposition of low-energy conformations of indan
4 (purple) and 2 (green).
of low-energy conformations (from MOPAC/AM1 opti-
mizations) of 2 and indan 4 is shown in Figure 1. The
pyridine nitrogens are in very similar positions for both
compounds, and the aromatic rings can also be in
similar planes. Molecular modeling studies on 2 (MO-
PAC/AM1 calculations) found a series of structures with
lowest energy where the bulky nature of the ortho
substituents forces the aromatic ring and the amide
group out of planarity. The X-ray structure10 of the
norbornyl (replacing cyclopentyl) analogue of 2 had a
conformation similar to one of these MOPAC/AM1 low-
energy conformers. This shows that both aromatic rings
are out of conjugation with the amide bond and that
the final conformations are practically equienergy. The
carbonyl group faces in the same or opposite direction
to the cyclopentyloxy substituent (up or down), and the
o-dichloropyridyl ring ends up being at about 90° to the
other aromatic ring (as shown in Figures 1-4). It can
be noted that unlike for the 2 analogue, the indan 4
derivative does not require a dichloropyridyl ring to
have such a low-energy conformation. It is in this
context that we report upon the identification and
biological profiles of a structurally unique class of cyclic
PDE4 inhibitors.
a
(a) MeOH/H2SO4 (98% yield); (b) DEAD/Ph3P/cyclopentanol
(89% yield); (c) NaOH/MeOH, HCl (95% yield); (d) phthalic
anhydride/NaOAc/heat; (e) NaOMe/MeOH, HCl (78% yield over
two steps for 7 and 45% for 8); (f) K2CO3/NaI/4-picolyl chloride
(85% yield); (g) NaBH4/MeOH; (h) Et3SiH/TFA (90% yield over two
steps for 4); (i) 10% Pd/C/AcOEt (100% yield); (j) DEAD/Ph3P/4-
pyridinylcarbinol (38% yield).
Ta ble 1. Effects of Carbonyl Groups of the Cyclic Compounds
Ch em istr y
IC50
Ki
The synthesis of the cyclic compounds is exemplified
in Scheme 1. Acid 6 was obtained from commercially
available 3-hydroxy-4-methoxyphenylacetic acid via es-
terification, Mitsunobu reaction11 with cyclopentanol,
and saponification. Gabriel-modified Perkin reaction12
of acid 6 and various phthalic anhydrides afforded the
monosubstituted indan-1,3-diones 7 and 8 which were
then alkylated with 4-picolyl chloride. Selective hydro-
genation of the 2,2-disubstituted indan-1,3-dione 10
delivered phenol 20 which was a versatile intermediate
for further transformations. Reduction of indan-1,3-
dione 9 into the corresponding indan 4 was achieved
with sodium borohydride followed by triethylsilane/TFA.
(PDE4,
(rolipram
compd
X
Y
R
nM)a,b
bind, nM)b,c
2
4
9
16
(+)-10
(-)-17
(+)-18
19
1
750
30
1
1600
10
H,H
O
H,OH
O
H,H
O
H,OH
O
H
H
H
>10000
>10000
benzyloxy
benzyloxy
benzyloxy
benzyloxy
4
6
30
20
20
20
20
60
O
O
O
O
O
H,H
a
IC50 and Ki values were determined from concentration-
response curves (n ) 3) in which concentrations ranged from 0.1
b
nM to 10 µM. Errors were within (20%. PDE activity was
determined in macrophage homogenates by the two-step radio-
isotopic method of Thompson et al.18 c (()-[3H]Rolipram binding
assays were performed using guinea pig brain membranes as
described by Schneider et al.8b
Biologica l Resu lts a n d Discu ssion
The PDE4 inhibition and rolipram displacement data
are summarized in Tables 1-3. The initial target
compound 4, as shown in Table 1, proved to be a rather
weak PDE4 inhibitor with an IC50 of 750 nM despite
its structural resemblance to 3 (IC50 of 7 nM).9 The
monosubstituted indan-1,3-dione intermediate 7 did not
display any significant PDE4 inhibitory activity at
concentrations up to 10 µM (Table 2). However, several
alkylation products derived from 7 demonstrated in-
creased potency (e.g., 9, 13, and 15). The precursor
indan-1,3-dione 9 had much better inhibitory activity
(IC50 of 30 nM, still less potent than 2) than indan 4.
The initial approach based on indan 4 was abandoned,

4218 J ournal of Medicinal Chemistry, 1998, Vol. 41, No. 22
Ta ble 2. Effects of Alkylation of the Cyclic Compounds
He et al.
IC50
Ki
(PDE4,
(rolipram
compd
Z
nM)a,b
bind, nM)b,c
2
7
9
1
>10000
30
1
>10000
10
H
pyridin-4-ylmethyl
F igu r e 2. Superposition of low-energy conformations of 2
(green) and indan-1,3-dione 9 (light blue).
12
13
14
15
phenylmethyl
500
70
3000
50
2300
260
1700
20
4-hydroxyphenylmethyl
3,5-dichloropyridin-4-ylmethyl
1-oxypyridin-4-ylmethyl
a
IC50 and Ki values were determined from concentration-
response curves (n ) 3) in which concentrations ranged from 0.1
b
nM to 10 µM. Errors were within (20%. PDE activity was
determined in macrophage homogenates by the two-step radio-
isotopic method of Thompson et al.18 c (()-[3H]Rolipram binding
assays were performed using guinea pig brain membranes as
described by Schneider et al.8b
Ta ble 3. Effects of Substitution on the Indan Ring of the
Cyclic Compounds
F igu r e 3. Superposition of low-energy conformation of indan-
1,3-dione 9 (light blue), indan 4 (purple), and 2 (green).
the indan-1,3-dione 9 with 2 is shown in Figure 2,
together with indan 4 in Figure 3. The overlap of the
position and directionality of the nitrogens of pyridines
can be clearly seen, and the common region for potential
interaction of a chlorine atom on 2 and carbonyl oxygen
on indan-1,3-dione 9 can also be noted. In addition, the
indan-1,3-dione ring of compound 9 is slightly shifted
away from the indan ring of compound 4. This shift, in
part, might account for the potency difference bewteen
compounds 4 and 9 (Table 1). The poor activity of the
dichloropyridine analogue 14 in Table 2 was not sur-
prising because steric interactions of the bulky chlorine
atoms with the indan-1,3-dione ring destabilize confor-
mations where the pyridine ring nitrogen can overlap
with that of 2. A superposition of a low-energy structure
of 14 with 2 is shown in Figure 4. This result clearly
distinguishes the cyclic compounds from the benzamide
class of PDE4 inhibitors where the dichloropyridine
analogue 2 is the most potent compound in that series.6a
The dichloro analogue of indan 4 was not pursued due
to the poor activities of indan 1 and indan-1,3-dione 14.
Indan 4 was about 750-fold less potent than 2 as a
PDE4 inhibitor despite good conformational overlaps
according to molecular modeling. This observation is
understandable since the conformational modeling does
not take other interactions into account. The indan ring
of compound 4 might have unfavorable steric interaction
with PDE4 and thus destabilize the overall interaction
of compound 4 with PDE4. Furthermore, 2 is a much
more rigid molecule than compound 4, and its amide
bond might also provide additional interactions with
IC50
Ki
(PDE4, (rolipram
compd
R
nM)a,b bind, nM)b,c
2
9
1
30
4
100
12
14
50
30
30
1
10
20
45
20
30
20
60
20
H
10
20
21
22
23
24
25
benzyloxy
OH
ethoxy
4-pyridinylmethoxy
carboxymethoxy
1-methylpiperidin-4-yloxy
[1,4′]-bipiperidinyl-1′-carbonyloxy
a
IC50 and Ki values were determined from concentration-
response curves (n ) 3) in which concentrations ranged from 0.1
b
nM to 10 (M. Errors were within (20%. PDE activity was
determined in macrophage homogenates by the two-step radio-
isotopic method of Thompson et al.18 c (()-[3H]Rolipram binding
assays were performed using guinea pig brain membranes as
described by Schneider et al.8b
and indan-1,3-dione 9 was used as the initial lead
compound from which the novel series of PDE4 inhibi-
tors was developed. Although it was more selective than
compound 9 (Table 2), subsequently synthesized phenol
13 was not pursued further because the results below
seemed to indicate that the selectivity was not as critical
as anticipated.
The data presented in Table 2 suggested that inhibi-
tory potency was associated with the presence of func-
tional groups capable of forming hydrogen bond inter-
actions with the enzyme (e.g., pyridyl and phenol
groups). A superposition of a low-energy structure of

Novel Cyclic Compounds as PDE4 Inhibitors
J ournal of Medicinal Chemistry, 1998, Vol. 41, No. 22 4219
Ta ble 4. Effects of 1, 3, 9, 10, and 25 in the Murine
Endotoxemia Model
IC50 of TNF-R
ED50 or % of TNF-R
compd
inhibition (µM)a
inhibition (mg/kg/po)b
1
3
0.5
8
25
9
10
25
0.2
0.2
0% @ 30
40% @ 30
15
a
IC50 values were determined from concentration-response
curves (n ) 3) in which concentrations ranged from 0.1 nM to 10
b
µM. Errors were within (20%. ED50 values were determined
from dose-response curves of TNF-R inhibition. Compounds were
administered orally to mice 4 h prior to LPS challenge.
F igu r e 4. Superposition of low-energy conformation of indan-
1,3-dione 14 (purple) and 2 (green). The chlorines force a
different low-energy conformation; the conformation with the
pyridine in the same position as 2 is sterically hindered.
PDE4. The potency improvement from indan 4 to
indan-1,3-dione 9 could result from the previously
mentioned ring shift and/or additional bonding from the
carbonyl groups to PDE4.
Further potency enhancements were achieved through
functionalization of the aromatic ring of the indan-1,3-
dione. The benzyloxy derivative 10 with an IC50 of 4
nM proved to be the most potent inhibitor of PDE4
within this series synthesized so far (Tables 1 and 3).
The fact that compound 10, capable of interacting with
an additional binding site, is less potent than 2 in the
PDE4 assay seems to indicate there is still room for
potency improvement of the cyclic series. The confor-
mational freedom of compound 10 and perhaps the
different binding mode of the cyclic inhibitors with
PDE4 might account for the potency difference between
2 and indan-1,3-dione 10. Interestingly, the corre-
sponding indanone 19 (stereochemistry not confirmed)
also retained good potency. The carbonyl groups in 10
and 19 clearly had a significant effect on the inhibitory
potency. Whether the effect of the carbonyl groups is
steric or electronic in nature remains to be ascertained.
Racemic 10 was resolved by chiral HPLC to afford the
(-)-enantiomer 17 and the (+)-enantiomer 18 (absolute
stereochemistry not determined) with the former being
5-fold more potent against PDE4. Interestingly, the two
enantiomers were equipotent in displacing rolipram
from the high-affinity binding site.
As exemplified in Table 3, substitutions on the aro-
matic ring of the indan-1,3-dione proved to be reason-
ably well-tolerated. The selectivity profiles of the cyclic
PDE4 inhibitors were not affected by the nature of
substituents, ranging from acidic to basic and from
small to bulky. These observations allowed us to use
this site as a handle for modification to identify com-
pounds with improved pharmacodynamic and pharma-
cokinetic profiles relative to the initial lead compound
9 (discussion below). This observation also seemed to
indicate that additional pockets in the active site might
exist and that these need to be explored further.
The initial lead compound 9 was shown to have good
inhibitory potency against TNF-R release from human
mononuclear cells stimulated with LPS13 with IC50 of
0.2 µM, comparable with 1 (IC50 of 0.5 µM).13c Several
compounds from the cyclic series were selected for
further evaluation as potential antiinflammatory agents.
F igu r e 5. Dose-response curves of TNF-R inhibition for
compound 25 and rolipram. Compounds were administered
orally to mice 4 h prior to LPS challenge. Results are means
( SEM (n ) 6 animals/group).
They were tested in a murine endotoxemia model via
inhibition of TNF-R release (Table 4). To identify long-
acting antiinflammatory agents and to differentiate
PDE4 inhibitors within the series, the murine endo-
toxemia model was developed with the oral administra-
tion of test compounds 4 h prior to LPS challenge, very
different from those reported in the literature where
compounds were given 30 min before LPS injection.13b
For comparison, 1 was shown to have an ED50 of 8 mg/
kg/po in the murine endotoxemia model, while its ED50
was reported to be less than 3 mg/kg/po in a similar
model.13b The initial lead compound 9 was found to be
inactive in the murine endotoxemia model, while the
more potent analogue 10 demonstrated 40% inhibition
of TNF-R release at 30 mg/kg/po. This result was not
surprising since both compounds 9 and 10 were very
lipophilic (with log P values of 5.5 and 5.96, respec-
tively), and their bioavailability might be poor. 3 was
found to have an ED50 of 25 mg/kg/po in the same in
vivo assay. Compound 25, with the incorporation of ni-
trogen atoms to increase its polarity and water solubil-
ity, exhibited an ED50 of 15 mg/kg/po in a murine endo-
toxemia model (Figure 5). On the molar basis com-
pound 25 was equipotent to rolipram in the model since
its molecular weight is more than twice that of rolipram.

4220 J ournal of Medicinal Chemistry, 1998, Vol. 41, No. 22
Ta ble 5. Results of Dog Emesis Studies
He et al.
compounds was identified as a handle to improve the
pharmacodynamic and pharmacokinetic profiles of the
cyclic PDE4 inhibitors. The tolerance of substitutions
in such a region of the cyclic compounds seems to
suggest that there might be a pocket in the active site
of PDE4 which has not been explored by the known
PDE4 inhibitors such as 2 and 3, and this could be
exploited further in the future. In addition, some of
these cyclic compounds were found to be orally active
in a murine endotoxemia model via TNF-R inhibition.
Significantly, two of these compounds (10 and 25) were
shown to have no emetic activity in a dog model at
intravenous doses up to 3 mg/kg. Further studies on
this interesting class of PDE4 inhibitors are in progress.
dose tested
(mg/kg/iv)
ratio of
emetic dogs/total dogs
compd
vehicle (DMSO)
1
0/3
1/2
2/2
2/2
4/4
0/4
0/4
0/4
0/4
0.01
0.03
0.2
0.6
1
0.7
3
3
2
3
10
25
To evaluate the most common side effects, a dog
emesis model was chosen because dogs are the most
sensitive species to emesis. The intravenous adminis-
tration was applied to achieve initial high plasma con-
centration of compounds and to ensure that compounds
tested were in the circulation. Thus, cross-species dif-
ference such as oral bioavailability could be avoided
since efficacy and emesis studies were conducted in two
different species (mouse and dog). Interestingly, no
emesis was observed with any of four dogs when indan-
1,3-diones 10 and 25 were given at doses up to 3 mg/
kg/iv, while administration with rolipram caused emetic
symptoms even at a dose of 0.01 mg/kg/iv with one out
of two dogs (Table 5). Two out of two dogs were ob-
served to have emetic symptoms when 1 was adminis-
tered at doses of 0.03 and 0.2 mg/kg/iv. Intravenous
administration of 2 at a dose of 0.6 mg/kg also resulted
in emesis with four out of four dogs.14
Exp er im en ta l Section
Ch em istr y. Representative syntheses of the target com-
pounds are shown below. All final compounds were character-
ized by melting point, H NMR, mass spectra, and elemental
1
analyses. Some of compounds were hygroscopic and amor-
phous, as noted below for each compound. Melting points of
such amorphous solids were taken on a Thomas-Hoover (Uni-
Melt) capillary melting point apparatus and are uncorrected.
Elemental analyses were performed by QTI, Whitehouse, NJ .
The concentration of hygroscopic compounds for biological
evaluation was calculated based on the hydrated formula
weight derived from elemental analyses. 1H NMR spectra
were recorded on a Bruker AC 300 spectrometer. Mass spectra
were measured on Finnigan 4500 or VG 70 SE instruments.
Flash chromatography was conducted with 230-400 mesh
silica gel (EM Science). Optical rotations were recorded on a
Perkin-Elmer 241 polarimeter. Reagents were used as re-
ceived from commercial suppliers.
From the data presented above, we have identified a
novel series of inhibitors that display reasonable PDE4
inhibitory activity with certain members also showing
reduced rolipram binding affinity compared with 2. The
data does not demonstrate any correlation between
PDE4 inhibition and rolipram binding displacement.
More importantly, both 10 and 25 were still nonemetic
at the doses tested despite having different ratios for
the catalytic/rolipram binding affinity, which seems to
raise questions about the hypothesis concerning the role
of selectivity or the catalytic/rolipram binding sites in
their contribution to the side effects. It is also possible
that 10 and 25 cannot penetrate into the brain,15 al-
though some compounds do not have to cross the blood-
brain barrier to cause emesis. It has not been deter-
mined what effects compounds 10 and 25 will have at
doses higher than 3 mg/kg/iv in the emesis studies.
Additional studies are needed to address these issues.
Nevertheless, we have identified PDE4 inhibitors with
improved therapeutic windows as indicated by the
inhibitory dose of TNF-R release and the no-effect dose
of emesis.
4-Meth oxy-3-(cyclop en tyloxy)p h en yla cetic Acid (6).
To a solution of 3-hydroxy-4-methoxyphenylacetic acid (10 g,
55 mmol) in 50 mL of anhydrous methanol was added 1 mL
of concentrated sulfuric acid, and the resulting mixture was
refluxed for 2 h before being concentrated to dryness. The
residue was redissolved in 100 mL of ether; the organic layer
was washed with saturated sodium bicarbonate and then dried
over magnesium sulfate. Filtration and concentration pro-
vided 10.6 g (98%) of the corresponding methyl ester as an
oil.
The above ester was dissolved in 20 mL of THF and at room
temperature treated with cyclopentyl alcohol (5 g, 58 mmol),
triphenylphosphine (17 g, 65 mmol), and DEAD (11.3 g, 65
mmol), and the resulting brown solution was stirred overnight.
After being concentrated to dryness, the residue was diluted
with 100 mL of ether, and solids were removed by filtration.
The residue obtained by concentration was purified by chro-
matography on silica gel (30% ether/hexane) to give 12.6 g
(89%) of a liquid.
The above liquid was treated with 60 mL of methanol and
60 mL of 1 N sodium hydroxide, and the mixture was
vigorously stirred at room temperature overnight. The aque-
ous solution after removal of methanol was acidified with 1 N
hydrochloric acid. The solid was collected by filtration and
then dried under high vacuum to afford 11.4 g (95%) of the
desired acid.
2-(3-(Cyclop en tyloxy)-4-m eth oxyp h en yl)in d a n -1,3-d i-
on e (7). Gen er a l P r oced u r e 1: A mixture of 4-methoxy-3-
(cyclopentyloxy)phenylacetic acid (2.5 g, 10 mmol), phthalic
anhydride (4.5 g, 30 mmol), and sodium acetate (1 g, 12 mmol)
was heated at 200 °C for 4 h and then cooled to room
temperature before being dissolved in 100 mL of dichlo-
romethane. The organic phase was washed with saturated
sodium bicarbonate (100 mL) and dried over magnesium
sulfate. The residue obtained by filtration and concentration
was redissolved in 20 mL of methanol before being treated with
a 25% sodium methoxide/methanol solution (6.9 mL). The
reddish suspension was refluxed for 0.5 h and then concen-
trated to dryness. The residue was acidified with concentrated
Con clu sion s
In summary, we have elaborated a structurally novel
class of indan-1,3-dione-based PDE4 inhibitors with
reasonable (nanomolar) potencies in a PDE4 assay. This
series of PDE4 inhibitors is clearly different from the
known PDE4 inhibitors such as 2 and 3. The carbonyl
groups of the indandione or indanone compounds were
critical for the inhibitory activities. The indan ring
could be substituted to improve the PDE4 inhibitory
activity, and substitutions on this part of the molecules
did not affect selectivity. This region of the cyclic

Novel Cyclic Compounds as PDE4 Inhibitors
J ournal of Medicinal Chemistry, 1998, Vol. 41, No. 22 4221
hydrochloric acid at 0 °C and then extracted with dichlo-
romethane (3 × 50 mL). The organic layer was dried over
magnesium sulfate and then filtered. The solution was
concentrated to give the indan-1,3-dione (2.62 g, 78%): yellow
solid; mp 128-130 °C; MS m/z 336 (M+). Anal. (C21H20O4) C,
H.
mL) and then concentrated by rotary evaporation. The residue
was purified by chromatography on silica gel (50% ethyl
acetate/hexane) to give 192 mg (75%) of the desired product:
yellow solid; hygroscopic; mp 172-173 °C; MS m/z 443 (M+
1). Anal. (C28H26O5‚0.5H2O) C, H.
+
2-(3-(Cyclop en tyloxy)-4-m eth oxyp h en yl)-2-(3,5-d ich lo-
r op yr id -4-ylm et h yl)in d a n -1,3-d ion e (14). Compound 14
was prepared according to general procedure 2: yellow solid;
mp 124-126 °C; 98% yield; MS m/z 496 (M+ + 1). Anal.
(C27H23Cl2NO4) C, H, N.
6-(Ben zyloxy)-2-(3-(cyclopen tyloxy)-4-m eth oxyph en yl)-
in d a n -1,3-d ion e (8). A mixture of 4-(benzyloxy)phthalic
anhydride16 (4.50 g, 18.5 mmol), triethylamine (13 mL, 93.3
mmol), 3-(cyclopentyloxy)-4-methoxyphenylacetic acid (4.63 g,
18.5 mmol), and acetic anhydride (10 mL) was heated to reflux
for a total of 4 h. The mixture was concentrated to yield a
thick black tar which was dissolved in dichloromethane (100
mL) and then poured into ice water (200 mL). The organic
phase, after washing with water, was dried with magnesium
sulfate. The crude material obtained by filtration and con-
centration was dissolved in methanol before the addition of
25% sodium methoxide in methanol (30 mL). The resulting
reaction mixture was refluxed for 1 h and then worked up as
above to give 3.7 g (45%) of the corresponding indan-1,3-
dione: light-yellow solid; hygroscopic; MS m/z 442 (M+). Anal.
(C28H26O5‚0.2H2O) C, H.
2-(3-(Cyclopen tyloxy)-4-m eth oxyph en yl)-2-(1-oxopyr id-
4-ylm eth yl)in d a n -1,3-d ion e (15). To a solution of 9 (400 mg,
0.94 mmol) in 4 mL of dichloromethane at room temperature
was added MCPBA (74%, 300 mg, 1.3 mmol). The resulting
suspension was stirred at room temperature for 2 h before
being quenched with a saturated solution of sodium bisulfite
(1 mL). After being washed with 1 N sodium hydroxide (5 mL),
the organic phase was concentrated, and the residue was
purified by chromatography on silica gel (ethyl acetate) to
afford 400 mg (96%) of the desired product: yellow solid; mp
215-217 °C; MS m/z 443 (M+). Anal. (C27H25NO5) C, H, N.
2-(3-(Cyclop en tyloxy)-4-m eth oxyp h en yl)-2-(p yr id -3-yl-
m eth yl)in d a n -1,3-d iol (16). To a solution of 9 (370 mg, 0.87
mmol) in 10 mL of THF at 0 °C was added a 1 M LAH/THF
solution (1.7 mL, 1.7 mmol), and the mixture was stirred for
3 h before being quenched with 1 N hydrochloric acid and then
neutralized to pH 7. The mixture was extracted with dichlo-
romethane (2 × 50 mL). The residue obtained by concentra-
tion was purified by chromatography on silica gel (60% ethyl
acetate/hexane) to give 277 mg (74%) of a mixture of diols:
white solid; hygroscopic; mp 99-101 oC; MS m/z 432 (M+).
Anal. (C27H29NO4‚0.5H2O) C, H, N.
(-)-6-(Be n zyloxy)-2-(3-(cyclop e n t yloxy)-4-m e t h oxy-
p h en yl)-2-(p yr id -4-ylm eth yl)in d a n -1,3-d ion e (17). Com-
pound 17 was resolved by chiral HPLC from racemic 10
(CHIRACEL OD) eluted with heptane/ethanol containing 0.1%
diethyamine (1:3, v/v) detected at UV 230 nm at flow rate of
1 mL/min: retention time 6.5 min; yellow solid; mp 58-61 °C;
MS m/z 533 (M+); [R]D -8.8° (c ) 0.17, CHCl3).
(+)-6-(Be n zyloxy)-2-(3-(cyclop e n t yloxy)-4-m e t h oxy-
p h en yl)-2-(p yr id -4-ylm eth yl)in d a n -1,3-d ion e (18). Com-
pound 18 was obtained with a retention time of 9.3 min:
yellow solid; mp 55-60 °C; MS m/z 533 (M+); [R]D +10° (c )
0.17, CHCl3).
2-(3-(Cyclop en tyloxy)-4-m eth oxyp h en yl)-2-(p yr id -4-yl-
m eth yl)in d a n -1,3-d ion e (9). Gen er a l P r oced u r e 2:
A
suspension of 7 (1.34 g, 4.0 mmol), potassium carbonate (1.5
g), sodium iodide (300 mg), and 4-picolyl chloride hydrochloride
(800 mg, 4.8 mmol) in 30 mL of acetone was refluxed for 1 h
and then diluted with ether (50 mL) at room temperature. The
residue obtained by filtration and concentration was purified
by chromatography on silica gel (50% ether/hexane) to give
1.25 g (73%) of the desired product: yellow solid; mp 154-
155 °C; MS m/z 427 (M+). Anal. (C27H25NO4) C, H, N.
4-[2-(3-(Cyclop en tyloxy)-4-m eth oxyp h en yl)in d a n -2-yl-
m eth yl]p yr id in e (4). To a solution of 9 (270 mg, 0.63 mmol)
in methanol at room temperature was added sodium borohy-
dride (100 mg, 2.6 mmol). The resulting suspension was
stirred at room temperature for 1 h before being concentrated
to dryness. The residue was treated with trifluoroacetic acid
(4 mL) and triethylsilane (4 mL). The resulting mixture was
stirred vigorously overnight before being diluted with dichlo-
romethane (100 mL) and then washed with saturated sodium
bicarbonate (50 mL). The organic phase was dried over
magnesium sulfate and then filtered. The residue obtained
by concentration was purified by chromatography on silica gel
(30% ether/hexane) to give 225 mg (90%) of compound 4:
colorless oil; MS m/z 399 (M+). Anal. (C27H29NO2) C, H, N.
6-(Ben zyloxy)-2-(3-(cyclopen tyloxy)-4-m eth oxyph en yl)-
2-(p yr id -4-ylm eth yl)in d a n -1,3-d ion e (10). Compound 10
was prepared according to general procedure 2 starting with
8 and purified by chromatography on silica gel (50% ether/
hexane): 202 mg (85% yield); yellow solid; mp 65-68 °C; MS
5-(Ben zyloxy)-2-(3-(cyclopen tyloxy)-4-m eth oxyph en yl)-
2-(p yr id -4-ylm eth yl)in d a n -1-on e (19). A suspension of 4-[6-
(benzyloxy)-2-(3-(cyclopentyloxy)-4-methoxyphenyl)indan-2-
ylmethyl]pyridine (100 mg, 0.2 mmol, obtained from reduction
of 10) and 2,2′-bipyridinium chlorochromate (584 mg, 2 mmol)
in 20 mL of acetone was refluxed overnight before being
diluted with ether (100 mL) at room temperature. The
mixture was filtered, washed with a saturated solution of
copper(II) sulfate, and then concentrated. The residue was
purified by chromatography on silica gel (50% ether/hexane)
to give 25 mg (25%) of the title compound: white solid; mp
52-55 °C; MS m/z 520 (M+).
1
m/z 533 (M+); H NMR (CDCl3) δ 8.32 (bd, 2H), 7.81 (d, 1H),
7.48-7.31 (bm, 7H), 7.11 (bd, 3H), 7.05 (bd, 1H), 6.85 (dd, 1H),
6.77 (d, 1H), 5.13 (s, 2H), 4.77 (m, 1H), 3.80 (s, 3H), 3.48 (s,
2H), 1.96-1.63 (m, 7H). Anal. (C34H31NO5) C, H, N.
2-Ben zyl-(3-(cyclop en tyloxy)-4-m eth oxyp h en yl)in d a n -
1,3-d ion e (12). Compound 12 was prepared according to
general procedure 2: yellow solid; mp 116-119 °C; 53% yield;
MS m/z 426 (M+). Anal. (C28H26O4) C, H.
2-(3-(Cyclop en tyloxy)-4-m eth oxyp h en yl)-5-h yd r oxy-2-
(p yr id -4-ylm eth yl)in d a n -1,3-d ion e (20). A suspension of
10 (200 mg, 0.38 mmol) in 10 mL of ethyl acetate containing
200 mg of Pd/C (10%) was stirred at room temperature with a
hydrogen balloon for 2 h before the Pd/C was removed by
filtration. Concentration afforded 165 mg (100%) of compound
20: yellow solid; hygroscopic; mp 106-109 °C; MS m/z 443
4-[2-(3-(Cyclop e n t yloxy)-4-m e t h oxyp h e n yl)-1,3-d i-
oxoin d a n -2-ylm eth yl]p h en yl Aceta te. The title intermedi-
ate of compound 13 was prepared according to general
procedure 2: yellow solid; mp 130-132 °C; 74% yield; MS m/z
484 (M+). Anal. Calcd for C30H28O6: C, 74.36; H, 5.82.
Found: C, 74.01; H, 5.95.
2-(3-(Cyclop en tyloxy)-4-m eth oxyp h en yl)-2-(4-h yd r oxy-
b en zyl)in d a n -1,3-d ion e (13). To a solution of 4-[2-(3-(cy-
clopentyloxy)-4-methoxyphenyl)-1,3-dioxoindan-2-ylmethyl]phenyl
acetate (280 mg, 0.58 mmol) in 10 mL of methanol at 0 °C
was added a 25% sodium methoxide/methanol solution (0.15
mL, 0.67 mmol). The resulting reddish solution was stirred
at room temperature for 2 h before being concentrated to
dryness and then acidified with 1 N hydrochloric acid. The
acidified mixture was extracted with dichloromethane (2 × 50
(M+
). Anal. (C27H25NO5‚0.5H2O) C, H, N.
2-(3-(Cyclop en tyloxy)-4-m eth oxyp h en yl)-5-(p yr id in -4-
ylm eth oxy)-2-(p yr id in -4-ylm eth yl)in d a n -1,3-d ion e (22). A
mixture of 20 (140 mg, 0.32 mmol), 4-pyridinylcarbinol (140
mg, 1.28 mmol), triphenylphosphine (335 mg, 1.28 mmol), and
DEAD (0.2 mL, 1.28 mmol) in 10 mL of THF was stirred at
room temperature overnight. After being concentrated to
dryness, the residue was purified by chromatography on silica
gel (ethyl acetate) to give 65 mg (38%) of the desired product:
yellow solid; hygroscopic; mp 67-70 °C; MS m/z 534 (M+); 1H
NMR (CDCl3) δ 8.65 (d, 1H), 8.6 (d, 1H), 8.6 (d, 2H), 7.86 (d,

4222 J ournal of Medicinal Chemistry, 1998, Vol. 41, No. 22
He et al.
1H), 7.37-7.21 (m, 5H), 7.1 (s 1H), 7.05 (d, 1H), 6.95-6.89
(bd, 1H), 6.78 (d, 1H), 5.16 (s, 2H), 4.75 (m, 1H), 3.78 (s, 3H),
3.49 (s, 2H), 1.96-1.63 (m, 3H). Anal. (C27H25NO5‚0.5H2O)
C, H, N.
50% inhibition of substrate hydrolysis) for the compounds
examined were determined from concentration-response curves
in which concentrations ranged from 0.1 nM to 10 µM. At least
three concentration-response curves were generated for each
agent, and 2 was used as control.
Mea su r em en t of (+)-[3H]Rolip r a m Bin d in g to Br a in
Mem br a n es. (()-Rolipram was brominated in-house in CCl4
(Mr. Kenneth Clow, Discovery Chemistry Department, Rhoˆne-
Poulenc Rorer) and subsequently tritiated by catalytic reduc-
tion with palladium and charcoal by Amersham International.
The specific radioactivity of the (()-[3H]rolipram was 24.7 Ci/
mmol. Guinea pig brain membranes were prepared, and the
binding assay was performed as described by Schneider et al.8b
with (()-[3H]rolipram (2 nM) and membrane samples corre-
sponding to 500 µg of brain tissue.
Ca n in e Em esis Mod el. Male and female beagle dogs
(Marshall Farms, Rose, NY; 1-2 years of age, 8-10 kg) were
used. Animals were fed once daily (in the morning) and given
free access to water. Food was withheld the morning of a test.
Compounds to be tested were dissolved in 100% DMSO.
Compounds were administered intravenously into a cephalic
vein (injection volume of 0.03 mL/kg) by slow bolus injection
over a period of 30 s. Animals were monitored for emesis for
up to 2 h or until emesis occurred. For a typical test, a total
of 4 dogs (2 male, 2 female) were used.
Mu r in e En d otoxem ia Mod el. The ability of the test
compounds to inhibit TNF-R production in vivo was evaluated
in a mouse endotoxemia (i.e., lipopolysaccharide (LPS) chal-
lenge) model. Male, Balb/c mice (20-25 g; Harlan Sprague-
Dawley Inc., Indianapolis, IN) were used. Test compounds
were administered orally in a vehicle of 0.25% methylcellulose/
0.2% Tween-20 in water 4 h prior to LPS challenge. Mice were
then challenged with LPS (Escherichia coli 055:B5; Sigma
Chemical Co., St. Louis, MO) at a dose of 30 µg/mouse by
intraperitoneal injection. Ninety minutes following LPS injec-
tion, mice were anesthetized with isoflurane, and blood was
collected by cardiac puncture. The whole blood was allowed
to clot on ice, and serum was prepared by centrifugation at
500g for 10 min. Serum was assayed for TNF-R by ELISA
(Genzyme Corp., Cambridge, MA).
Molecu la r Mod elin g. Computational approach was used
to obtain low-energy conformations for 2, 4, 9, and 14. The
initial molecular structures were constructed using Sybyl 6.3
(from Tripos Inc., 1699 S. Hanley Rd, St. Louis, MO 63144)
by systematically rotating all three relevant torsion angles
between the two aromatic rings. The amide bond was included
to ensure the coverage of the conformational space. To avoid
transition states and to allow more conformational space to
be searched, the starting torsion angles of 45°, 135°, -45°, and
-135° were assigned to those rotatable bonds. After an initial
molecular mechanics optimization, a full geometric optimiza-
tion was carried out for each of those initial conformations
using MOPAC/AM1 6.00 (distributed by Tripos Inc., also
available from Quantum Chemistry Program Exchange, De-
partment of Chemistry, Indiana University, Bloomington, IN
47405). For low-energy conformations identified by geometric
optimization, frequency calculations were performed to ensure
that no negative second-derivative exists.
Several low-energy conformations were identified for 2. The
lowest-energy conformation was in a group of four conforma-
tions within an energy range of 1.0 kcal/mol. Two other
conformations were located about 1.5 and 4.5 kcal/mol above
the lowest-energy conformation. For indan 4, a group of five
conformations carried an energy close to one another. The
lowest-energy conformation was 0.77, 0.95, 1.05, and 1.18 kcal/
mol more stable than the rest of four conformations in the
group, respectively. Another conformation was identified with
an energy of 3.18 kcal/mol higher than that of the lowest-
energy conformation.
2-(3-(Cyclop en t yloxy)-4-m et h oxyp h en yl)-5-et h oxy-2-
(p yr id -4-ylm eth yl)in d a n -1,3-d ion e (21). Compound 21 was
prepared similarly using conditions above: 33% yield; yellow
solid; mp 60-65 °C; MS m/z 472 (M+ + 1). Anal. Calcd for
C
29H29NO5: C, 73.87; H, 6.2; N, 2.97. Found: C, 73.25; H,
6.19; N, 2.83.
[2-(3-(Cyclop en tyloxy)-4-m eth oxyp h en yl)-1,3-d ioxo-2-
(p yr id -4-ylm eth yl)in d a n -5-yl]a cetic Acid Tr iflu or oa cetic
Acid Sa lt (23). A suspension of 20 (300 mg, 0.68 mmol), tert-
butyl bromoacetate (0.1 mL, 0.68 mmol), and sodium hydride
(28 mg, 60%, 0.68 mmol) in 10 mL of THF was refluxed for 2
h. The residue, after being filtered and concentrated, was
purified by chromatography on silica gel (ether) to give 190
mg (50%) of the alkylated product: yellow solid; mp 51-54
°C; MS m/z 516 (M+).
A mixture of the alkylated product (180 mg) and TFA (1.5
mL) in 10 mL of methylene chloride was stirred at room
temperature for 3 h. The residue obtained by concentration
was purified by chromatography on silica gel (5% methanol/
ethyl acetate) to give 100 mg (47%) as the trifluoroacetic acid
salt: yellow solid; hygroscopic; mp 126-131 °C; MS m/z 501
(M+); 1H NMR (CDCl3) δ 8.4 (bs, 1H), 7.81 (bd, 2H), 7.45-
7.18 (bm, 5H), 7.05 (s, 1H), 6.90 (bd, 1H), 6.78 (d, 1H), 4.73
(bm, 3H), 3.79 (s, 3H), 3.57 (bd, 2H), 2.05-1.55 (bm, 8H). Anal.
(C29H27NO7‚C2HF3O2‚2H2O) C, H, N.
2-(3-(Cyclopen tyloxy)-4-m eth oxyph en yl)-5-(1-m eth ylpi-
per idin -4-yloxy)-2-(pyr id-4-ylm eth yl)in dan -1,3-dion e (24).
Compound 24 was prepared similarly to 22: 34.3% yield;
yellow solid; hygroscopic; mp 67-72 °C; MS m/z 541 (M+
1). Anal. (C33H36N2O5‚H2O) C, H, N.
+
2-(3-(Cyclop en t yloxy)-4-m et h oxyp h en yl)-1,3-d ioxo-2-
(p yr id -4-ylm eth yl)in d a n -5-yl [1,4′]-Bip ip er id in yl-1′-ca r -
boxyla te (25). A mixture of 20 (215 mg, 0.49 mmol), [1,4′]-
bipiperidinyl-1′-carbonyl chloride [generated in situ from
4-piperidinopiperidine (370 mg, 2.2 mmol) and phosgene/
toluene (1.93 M, 3.0 mL, 5.8 mmol)], and pyridine (0.1 mL) in
5 mL of THF was stirred overnight at room temperature before
being diluted with ether (20 mL) and then filtered. The
residue after concentration was purified by chromatography
on silica gel (ethyl acetate) to give 100 mg (32%) of the desired
product: yellow solid; hygroscopic; mp 75-78 °C; MS m/z 638
(M+ + 1). 1H NMR (CDCl3) δ 8.31 (bd, 2H), 7.89 (d, J ) 8.4
Hz, 1H), 7.62 (bd, 1H), 7.5 (bd, 1H), 7.09 (d, J ) 2.2 Hz, 1H),
7.02 (d, J ) 5.5 Hz, 2H), 6.92 (dd, J ) 2.2, 8.3 Hz, 1H), 6.76
(d, J ) 8.7 Hz, 1H), 4.77 (m, 1H), 4.36 (m, 2H), 3.80 (s, 3H),
3.48 (s, 2H), 2.90 (m, 3H), 2.35-2.15 (m, 2H), 2.0-1.63 (bm,
10H). Anal. (C38H43N3O6 ‚2.5H2O) C, H, N.
Biologica l Meth od s. P r ep a r a tion of P DE fr om Gu in ea
P ig Ma cr op h a ges. The method of Turner et al.17 was used.
Briefly, cells were harvested from the peritoneal cavity of horse
serum-treated (0.5 mL ip) Dunkin Hartley guinea pigs (250-
400 g), and the macrophages were purified by discontinuous
(55%, 65%, 70%, v/v) gradient (Percoll) centrifugation. Washed
macrophages were plated out in cell culture flasks and allowed
to adhere. The cells were washed with Hank’s balanced salt
solution, scraped from the flasks, and centrifuged (1000g). The
supernatant was removed, and the pellets were stored at -80
°C until used. The pellet was homogenized in 20 mM tris-
(hydroxymethyl)aminomethane HCl, pH 7.5, 2 mM MgSO4, 1
mM dithiothreitol, 5 mM ethylenediaminetetraacetic acid, 0.25
mM sucrose, 20 µM p-tosyl-L-lycine chloromethyl ketone, 10
µg/mL leupeptin, and 2000 U/mL aprotinin.
Mea su r em en t of P DE Activity. PDE activity was deter-
mined in macrophage homogenates by the two-step radioiso-
topic method of Thompson et al.18 The reaction mixture
contained 20 mM tris(hydroxymethyl)aminomethane HCl (pH
8.0), 10 mM MgCl2, 4 mM 2-mercaptoethanol, 0.2 mM ethyl-
enebis(oxyethylenenitrilo)tetraacetic acid, and 0.05 mg of
bovine serum albumin/mL. The concentration of the substrate
was 1 µM. The IC50 values (i.e., concentrations which produce
Refer en ces
(1) Beutler, B.; Cerami, A. Cachectin and tumor necrosis factor as
two sides of the same biological coin. Nature 1986, 320, 584-
589.

Novel Cyclic Compounds as PDE4 Inhibitors
J ournal of Medicinal Chemistry, 1998, Vol. 41, No. 22 4223
(2) (a) Strieter, R. M.; Remick, D. G.; Ward, P. A.; Spengler, R. N.;
Lynch, J . P.; Larrick, J .; Kunkel, S. L. Cellular and molecular
regulation of tumor necrosis factor-alpha production by pentoxi-
fylline. Biochem. Biophy. Res. Commun. 1988, 155, 1230-1236.
(b) Taffet, S. M.; Singhel, K. J .; Overholtzer, J . F.; Shurtleff, S.
A. Regulation of tumor necrosis factor expression in a macroph-
age-like cell line by lipopolysaccharide and cyclic AMP. Cell.
Immunol. 1989, 120, 291-300.
(3) (a) Dulpantier, A. J .; Cheng, J . B. Emerging opportunities in
the treatment of asthma and allergy. Annu. Rep. Med. Chem.
1994, 29, 73-81. (b) Nicholson, C. D.; Shahid, M. Inhibitors of
cyclic nucleotide phosphodiesterase isozymes -their potential
utility in the therapy of asthma. Pulmonary Pharmacol. 1994,
7, 1-17. (c) Torphy, T. J .; Livi, G. P.; Christensen, S. B. Novel
phosphodiesterase inhibitors for the therapy of asthma. Drug
News Perspect. 1993, 6, 203-214. (d) Lowe, J . A., III; Cheng, J .
B. The PDE IV family of calcium-independent phosphodiesterase
enzymes. Drugs Future 1994, 29, 73-81. (e) Palfreyman, M. N.;
Souness, J . E. Phosphodiesterase type IV inhibitors. Prog. in
Med. Chem. 1996, 33, 1-52.
(4) Schwabe, U.; Miyake, M.; Ohga, Y.; Daly, J . 4-(3-Cyclopentyloxy-
4-methoxyphenyl)-2-pyrrolidone (ZK62711): A potent inhibitor
of adenosine cyclic 3′,5′-monophosphophate phosphodiesterase
in homogenates and tissue slices from rat brain. Mol. Pharmacol.
1976, 28, 900-910.
(5) Zeller, E.; Stief, H. J .; Pflug, B.; Satre, Y.; Hernandez, M. Results
of a phase II study of the antidepressant effect of rolipram.
Pharmacopsychiatry 1984, 17, 188-190.
Schmiechen, R.; Brezinski, M.; Seidler, J . Stereospecific binding
of the antidepressant rolipram to brain protein structures. Eur.
J . Pharmacol. 1986, 127, 105-115. (c) Barnett, M. S.; O’Leary,
B. J .; Burman, M.; Christensen, S. B.; Cieslinski, L. B.; Esser,
K. M.; Prabhakar, U. S.; Rush, J . A.; Torphy, T. J . Association
of the antiinflammatory activity of phosphodiesterase 4 (PDE4)
inhibitors with either inhibition of PDE4 catalytic activity or
competition for [3H]rolipram binding. Biochem. Pharmacol. 1996,
51, 949-956. (d) Barnett, M. S.; Grous, M.; Cieslinski, L. B.;
Burman, M.; Christensen, S. B.; Torphy, T. J . Inhibitors of
phosphodiesterase IV (PDE IV) increase acid secretion in rabbit
isolated gastric glands: correlation between function and inter-
action with a high-affinity rolipram binding site. J . Pharmacol.
Exp. Ther. 1995, 273, 1396-1402.
(9) (a) Warrellow, G. J .; Alexander, R. P.; Boyd, E. C.; Eaton, M.
A.; Head, J . C.; Higgs, G. A. The design of an orally active PDE
IV inhibitor (CDP 840) for asthma. 8th RSC-SCI Medicinal
Chemistry Symposium, Cambridge, U.K., 1995. (b) Hughes, B.
CDP 840. 2nd World Congress on Inflammation, Brighton, U.K.,
1995. (c) Hughes, B. Fourth Annual Midwest Meeting of the
Inflammation Research Association, Chicago, IL, 1996. (d)
Hughes, B.; Howat, D.; Lisle, H. M.; J ames, T.; Gozzard, N.;
Blease, K.; Hughes, P.; Kingaby, R.; Warrellow, G.; Alexander,
R.; Head, J .; Boyd, E.; Eaton, M.; Perry, M.; Wales, M.; Smith,
B.; Owens, R.; Catterall, C.; Lumb, S.; Russell, A.; Allen, R.;
Merriman, M.; Bloxham, D.; Higgs, G. The inhibition of antigen-
induced eosinophilia and bronchoconstriction by CDP840, a novel
stereoselective inhibitor of phosphodiesterase type 4. Br. J .
Pharmacol. 1996, 118, 1183-1191.
(6) (a) Ashton, M. J .; Cook, D. C.; Fenton, G.; Karlsson, J . A.;
Palfreyman, M. N.; Raeburn, D.; Ratcliffe, A. J .; Souness, J . E.;
Thurairatnam, S.; Vicker, N. Selective type IV phosphodiesterase
inhibitors as antiasthmatic agents. The syntheses and biological
activities of 3-(cyclopentyloxy)-4-methoxybenamides and ana-
logues. J . Med. Chem. 1994, 37, 1696-1703. (b) Dulpantier, A.
J .; Biggers, M. S.; Chambers, R. J .; Cheng, J . B.; Cooper, K.;
Damon, D. B.; Eggler, J . F.; Kraus, K. G.; Marfat, A.; Masmune,
H.; Pillar, J . S.; Shirley, J . T.; Umland, J . P.; Watson J . W.
Biarylcarboxylic acids and amides: inhibition of phosphodi-
esterase type IV versus [3H]rolipram binding activity and their
relationship to emetic behavior in the ferret. J . Med. Chem.
1996, 39, 120-125. (c) Masamune, H.; Cheng, J . B.; Cooper, K.;
Eggler, J . F.; Marfat, A.; Marshall, S. C.; Shirley, J . T.; Tickner,
J . E.; Umland, J . P.; Vazquez, E. Discovery of micromolar PDE
IV inhibitors that exhibit much reduced affinity for the [3H]-
rolipram binding site: 3-norbornyloxy-4-methoxyphenylmeth-
ylene oxindoles. Bioorg. Med. Chem. Lett. 1995, 5, 1965-1968.
(d) Cheng, J . B.; Cooper, K.; Duplantier, A. J .; Eggler, J . F.;
Kraus, K. G.; Marshall, S. C.; Marfat, A.; Masamune, H.; Shirley,
J . T.; Tickner, J . E.; Umland, J . P. Synthesis and in vitro profile
of a novel series of catechol benimidazoles. The discovery of
potent, selective phosphodiesterase type IV inhibitors with
greatly attenuated affinity for the [3H]rolipram binding site.
Bioorg. Med. Chem. Lett. 1995, 5, 1969-1972. (e) Dal Piaz, V.;
Giovannoni, M. P.; Castellana, C. Novel heterocyclic-fused
pyridazinones as potent and selective phosphodiesterase IV
inhibitors. J . Med. Chem. 1997, 40, 1417-1421. (f) Lampe, J .
W.; Chou, Y.-L.; Hanna, R. G.; Di Meo, S. V.; Erhardt, P. W.;
Hagedorn, A. A., III; Ingebretesen, W. R.; Cantor, E. (Imida-
zolylphenyl)pyrrol-2-one inhibitor of cardiac cAMP phosphodi-
esterase. J . Med. Chem. 1993, 36, 1041-1047. (g) Brackeen, M.
F.; Cowan, D. J .; Stfford, J . A.; Schoenen, J . M.; Domanico, P.
L.; Rose, D.; Strickland, A. B.; Verghese, M.; Feldman, P. L.
Design and synthesis of conformationally constrained analogues
of 4-(3-butoxy-4-methoxy)imidazolidin-2-one (Ro 20-1724) as
potent inhibitors of cAMP-specific phosphodiesterase. J . Med.
Chem. 1995, 38, 4848-4854.
(10) Unpublished results of Dagenham Research Centre, RPR, U.K.
(11) Mitsunobu, O.; Yamada, M. Preparation of esters of carboxylic
and phosphoric acid via quaternary phosphonium salts. Bull.
Chem. Soc. J pn. 1967, 40, 2380-238.
(12) J ohnson, J . R. The Perkin reaction and related reactions. Org.
React. 1942, 1, 211-256.
(13) (a) Molnar-Kimber, K. L.; Yonno, L.; Heaslip, R. J .; Weichman,
B. M. Differential regulation of TNF-R and IL-1â production from
endotoxin stimulated human monocytes by phosphodiesterase
inhibitors. Mediat. Inflam. 1992, 1, 411-417. (b) Feldman, P.
L.; Brackeen, M. F.; Cowan, D. J .; Marron, B. E.; Schoenen, F.
J .; Stafford, J . A.; Suh, E. M.; Domanico, P. L.; Rose, D.;
Leesnitzer, M. A.; Brawley, E. S.; Strickland, A. B.; Verghese,
M. W.; Connolly, K. M.; Bateman-Fite, R.; Noel, L. S.; Sekut,
L.; Stimpson, S. A. Phosphodiesterase type IV inhibition.
Structure-activity relationships of 1,3-disubstituted pyrro-
lidines. J . Med. Chem. 1995, 38, 1505-1510. (c) Souness, J . E.;
Griffin, M.; Maslen, C.; Ebsworth, K.; Scott, L. C.; Pollock, K.;
Palfreyman, M. N.; Karlsson, J .-A. Evidence that cyclic AMP
phosphodiesterase inhibitors suppress TNFR generation from
human monocytes by interacting with a ‘low-affinity’ phosphodi-
esterase 4 conformer. Br. J . Pharmacol. 1996, 118, 649-658.
(14) For the therapeutic ratio of RP 73401 administered orally:
Escott, K. J .; McMillan, S. J .; Birrell, M.; Webber, S. E.; Souness,
J . E.; Maslen, C.; Temple, R.; Iqbal, S.; Mariucci, M.; Rose, J .;
Geiger, L. E.; Thurairatnam, S.; Bower, S.; Aldous, D.; Sargent,
C. A. Pharmacological profiling of phosphodiesterase 4 (PDE4)
inhibitors and analysis of the therapeutic ratio in rats and dogs.
Br. J . Pharmacol. 1998, 123, 40.
(15) (a) J ezequel, S. G. Central nervous system penetration of
drugs: importance of physicochemical properties. In Progress
in Drug Metabolism; Gibson, G. G., Ed.; Taylor & Francis:
London, U.K., 1992; Vol. 13, pp 141-178. (b) Lin, J . H.; Lu, A.
Y. H. Role of pharmacokinetics and metabolism in drug discovery
and development. Pharmacol. Rev. 1997, 49, 403-449.
(16) Menard, M.; Erichomovitch, L.; La Brooy, M.; Chubb, F. L.
Quelques metabolites possibles de la thalidomide. Can. J . Chem.
1963, 41, 1722-1724.
(17) Turner, N. C.; Wood, L. J .; Burns, F. M.; Gueremy, T.; Souness,
J . E. The effect of cyclic AMP and cyclic GMP phosphodiesterase
inhibitors on the superoxide burst of guinea-pig peritoneal
macrophages. Br. J . Pharmacol. 1993, 108, 876-883.
(7) (a) Muller, T.; Engels, P.; Fozard, J . R. Subtypes of the type 4
cAMP phospho- diesterase: structure, regulation and selective
inhibition. Trends Pharmacol. Sci. 1996, 17, 294-298. (b) Kelly,
J . J .; Barnes, P. J .; Giembycz, M. A. Phosphodiesterase 4 in
macrophages: relation between cAMP accumulation, suppres-
sion of cAMP hydrolysis and inhibition of [3H]-(-)-rolipram
binding by selective inhibitors. Biochem. J . 1996, 318, 425-436.
(8) (a) Schmiechen, R.; Schneider, H. H.; Watchtel, H. Close
correlation between behavioural response and binding in vivo
for inhibitors of the rolipram-sensitive phosphodiesterase. Psy-
chopharmacology 1990, 102, 17-20. (b) Schneider, H. H.;
(18) Thompson, W. J .; Terasaki, W.; Epstein, P. M.; Strada, S. J .
Assay of cyclic nucleotide phosphodiesterase and resolution of
multiple molecular forms of the enzyme. Adv. Cycl. Nucleotide
Res. 1979, 10, 69-92.
J M970575F