Benzyl vinylogous amide substituted aryldihydropyridazinones and aryldimethylpyrazolones as potent and selective PDE3B inhibitors

Article abstract of DOI:10.1016/j.bmcl.2003.08.056

Aryldihydropyridazinones and aryldimethylpyrazolones with 2-benzyl vinylogous amide substituents have been identified as potent PDE3B subtype selective inhibitors. Dihydropyridazinone 8a (PDE3B IC₅₀=0.19 nM, 3A IC₅₀=1.3 nM) was selec

Full text of DOI:10.1016/j.bmcl.2003.08.056

Bioorganic & Medicinal Chemistry Letters 13 (2003) 3983–3987
Benzyl Vinylogous Amide Substituted Aryldihydropyridazinones
and Aryldimethylpyrazolones as Potent and Selective
PDE3B Inhibitors
Scott D. Edmondson,a,* Anthony Mastracchio,a Jiafang He,a Christine C. Chung,b
Michael J. Forrest,c Scott Hofsess,c Euan MacIntyre,c Joseph Metzger,c
Naphtali O’Connor,a Kajal Patel,c Xinchun Tong,a Michael R. Tota,b
Lex H. T. Van der Ploeg,b Jeff P. Varnerin,b Michael H. Fisher,a Matthew J. Wyvratt,a
Ann E. Webera and Emma R. Parmeea
aDepartment of Medicinal Chemistry, Merck & Co., Inc., PO Box 2000, Rahway, NJ 07065, USA
bDepartment of Metabolic Disorders, Merck & Co., Inc., PO Box 2000, Rahway, NJ 07065, USA
cDepartment of Pharmacology, Merck & Co., Inc., PO Box 2000, Rahway, NJ 07065, USA
Received 27 June 2003; revised 14 August 2003; accepted 27 August 2003
Abstract—Aryldihydropyridazinones and aryldimethylpyrazolones with 2-benzyl vinylogous amide substituents have been identi-
fied as potent PDE3B subtype selective inhibitors. Dihydropyridazinone 8a (PDE3B IC50=0.19 nM, 3A IC50=1.3 nM) was selec-
ted for in vivo evaluation of lipolysis induction, metabolic rate increase, and cardiovascular effects.
# 2003 Elsevier Ltd. All rights reserved.
Cyclic nucleotide phosphodiesterases (PDEs) are
responsible for catalyzing the hydrolysis and deactiva-
tion of secondary cellular messengers’ cAMP and
cGMP. The 11 isozymes in this gene family are respon-
sible for regulating a number of physiologically relevant
functions via downstream signaling cascades.1 PDE3
inhibitors have been investigated clinically for their car-
diotonic properties (vasodilation and inotropic activity)
and antithrombotic properties.2 Research in this field
has led to the discovery of some well described isozyme
selective PDE3 inhibitors like milrinone (1),3 cilostamide
(2),4 SK&F 95654 (3),5 and NSP-513 (4).6
inhibitors of PDE3 have been reported to stimulate
lipolysis in adipocytes.8 A potent subtype inhibitor of
PDE3B would be an excellent tool to gauge the effects
of PDE3B inhibition on lipolysis and whether such
changes subsequently evoke an increase in metabolic
rate. Such an inhibitor might find applications for the
treatment of obesity. To date, limited work has surfaced
describing potent PDE3B inhibitors.8a,9 Consequently,
we set out to identify a series of potent inhibitors that
are selective for PDE3B over 3A (Fig. 1).
Many known PDE3 inhibitors (e.g., 1, 3, and 4) possess
a pyridazinone core that is important for selectivity over
the other PDE isozymes (i.e., PDE2, PDE4, PDE5,
PDE7).10 Unfortunately, few known PDE3 inhibitors
have been tested against the PDE3B subtype. Two
inhibitors that have been tested against PDE3B are
milrinone (1) and cilostamide (2). Although 1 is less
potent against PDE3B than 2, it is about two-fold
selective for PDE3B over 3A (Table 1).8a Dihydropyri-
dazinones such as 3 and 4 are reported to be more
potent PDE3 inhibitors than 1, but no reports exist of
this structural motif with respect to inhibition against
the 3B subtype. We hoped to take advantage of the
To date, two highly homologous subtypes of PDE3
have been identified and each of these subtypes is local-
ized in different tissues.7 PDE3A was identified in vas-
cular smooth muscle tissue and is located primarily in
the heart and blood vessels. On the other hand, the
PDE3B subtype is found most prominently in brown
and white adipocytes. In addition to cardiotonic effects,
*Corresponding author. Tel.: +1-732-594-0287; fax: +1-732-594-
5350; e-mail: scott_edmondson@merck.com
0960-894X/$ - see front matter # 2003 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2003.08.056

3984
S. D. Edmondson et al. / Bioorg. Med. Chem. Lett. 13 (2003) 3983–3987
mediated Buchwald-type coupling with 12 to afford
vinylogous amides 13.13 These b-ketoesters were
exposed to hot methanolic hydrazine to produce dihy-
dropyridazinones 14 in good yields.
Encouragingly, the acylated dihydropyridazinones 6 are
uniformly more potent at PDE3B than 1 and they
maintain selectivity for 3B over 3A (Table 1).14 The
simple Cbz derivative 6d is especially potent at PDE3B
with an IC50=1.0 nM. Indeed this compound is 8.5
times more potent than the corresponding ethyl carba-
mate 6c. We reasoned that the distance between the aryl
groups in 6d might be important for improving PDE3B
potency. In order to apply this to the more potent
vinylogous amide structure class, we replaced the propyl
side chain of 4 with a benzyl groupto give 8a. We were
pleased to find that 8a is significantly more potent
against PDE3B than 6d, with a potency in the picomolar
range (IC50=0.19 nM). This compound is 46 times
more potent against PDE3B than 8b, supporting the
hypothesis that the pendant aryl group is important for
potency. We were also pleased to observe that 8a is 7-
fold selective for PDE3B over 3A. Importantly, the S
enantiomer of 8a was tested and found to be more than
20 times less potent at PDE3B.15
Figure 1. Known PDE3 inhibitors.
enhanced potency displayed by some of these inhibitors
in our search for potent PDE3B inhibitors. Conse-
quently, we decided to initiate a more comprehensive
investigation of substituents on the dihydropyri-
dazinone core.
Acylations of aniline 511 under standard conditions
afforded 6 in good yields (Scheme 1). Substituted benzyl
vinylogous amides 8 were formed by heating 5 with b-
substituted 1,3-cyclohexanediones 712 (Scheme 1).
Suzuki couplings with derivatives of 8 (Ar=3-BrPh)
afforded biaryl analogues such as 9.
In order to access analogues with substitution on the
central phenyl ring, a different synthetic route was uti-
lized (Scheme 2). A series of benzoic acids (10) were
converted to the corresponding Weinreb amides then
alkylated with ethyl Grignard. Enolates of the resulting
ethyl ketones were then alkylated with methyl bromo-
acetate to afford 11. Bromide 11 underwent palladium
Substitution on the pendant phenyl ring has little effect
on PDE3B potency with most compounds retaining
IC50’s in the subnanomolar range (Table 2). While 4-
substitution on this ring typically reduces PDE3B
potency (14h), sterically demanding groups at the 3-
position afford the most potent PDE3B inhibitors to
date (i.e., biphenyl derivatives 9a IC50=0.068 and 9b
IC50=0.049 nM). Regardless of substitution around the
pendant phenyl ring, 3A/3B selectivity remains com-
parable or inferior to the parent compound 8a.
Substitution on the central phenyl ring had a more dra-
matic effect on the potency and selectivity of PDE3B
inhibitors (Table 2). Larger groups at the 2-position
(e.g., 14i and 14n) result in a marked reduction of
potency at PDE3B, but they retain or even improve the
3A/3B subtype selectivity. On the other hand, fluorine
substituents on the central phenyl ring result in com-
pounds with subnanomolar potency against PDE3B
and a comparable selectivity profile to that of 8a.
Scheme 1. Synthesis of substituted dihydropyridazinones: (a) Et3N,
RCOCl, DCM; (b) p-TsOH (cat), tol, 7, DMSO, Á; (c) ArB(OH)2,
Pd(PPh3)4, Na2CO3 (aq), EtOH, dioxane, Á.
Table 1. Inhibition of PDE3A and PDE3B by dihydropyridazinones.
All compounds are >94% ee (R) configuration except 6d
Entry
R=
PDE3B IC50 (nM)a
PDE3A IC50 (nM)a
1b
2b
—
—
280
18
440
16
6a
6b
6c
6dc
8a
8b
Me
37
96
15
24
2.9
1.3
44
2-CF3Ph
OEt
OBn
Ph
5.0
8.5
1.0
0.19
8.8
Scheme 2. Synthesis of central-ring substituted dihydropyridazinones
14: (a) MeONHMe–HCl, HOBt, EDC, DIEA, DCM; (b) EtMgCl,
THF, 0 ꢀC; (c) LHMDS, BrCH2CO2Me, THF, À78 ꢀC; (d) 2-(2-
Me2NPh)PhP(c-hex)2, Pd2(dba)3, 12, THF, Cs2CO3, Á; (e) NH2NH2,
H2O, MeOH, Á.
H
aValues are means of two or more experiments.
bThese values are cited from ref 8a.
cRacemic.

S. D. Edmondson et al. / Bioorg. Med. Chem. Lett. 13 (2003) 3983–3987
3985
4,4-Dimethylpyrazolones have been reported as ring-
contracted mimics of dihydropyridazinone PDE3 inhi-
bitors.16 Using chemistry illustrated in Scheme 3, we
next examined a series of dimethylpyrazolone vinyl-
ogous amides with respect to PDE3B potency and sub-
type selectivity. The addition of methyl isobutyrate
enolate to substituted bromobenzaldehydes 15 followed
by oxidation of the resulting benzylic alcohol afforded
b-ketoesters 16. Buchwald coupling of 16 with benzyl
carbamate17 followed by cyclization gave the dimethyl-
pyrazolone 17. Deprotection to form the aniline was
followed by condensation with 7 to afford the desired
vinylogous amides 18. Alternatively, benzyl protection
of phenol 19 followed by zinc promoted addition of
ethyl 2-bromoisobutyrate to the nitrile afforded b-
ketoester 20. Deprotection followed by triflate forma-
tion generated 21, which was converted to 18 following
the previous sequence.
Although the pyrazolone vinylogous amides are gen-
erally less potent than the dihydropyridazinone analo-
gues (Table 3), they still maintain potencies against
PDE3B in the low nanomolar range. Furthermore, this
series is more selective for 3B over 3A. Entry 18f is the
most selective PDE3B subtype inhibitor yet reported,
with a 3A/3B selectivity of 33 and an IC50=4.5 nM
against PDE3B. Potency against PDE3B could be
enhanced by halide substitutions on the ortho position
of the pendant phenyl ring, such as the 2,6-dichloride
18h which exhibits a potency in the subnanomolar range
(IC50=0.13 nM). Although analogues with sterically
demanding substituents on the central phenyl ring are
typically less potent at PDE3B, they retain much of the
3A/3B selectivity. On the other hand, fluoride substitu-
tion on the central phenyl ring appears to increase
PDE3B potency, with 2,3-difluorophenyl analogue 18n
exhibiting an excellent overall in vitro profile (PDE3B
IC50=0.33 nM, 3A/3B=23).
Table 2. Inhibition of PDE3A and PDE3B by dihydropyridazinones
Entry
Ar=
R= PDE3B IC50 (nM)a PDE3A IC50 (nM)a
14ab
14bb
14cb
14db
14eb
14fb
9ab
2-MeOPh
2-CF3Ph
3-FPh
3-BrPh
3-IPh
3-NO2Ph
3-Thiophene
3-NO2Ph
4-MeOPh
4-NO2Ph
Ph
H
H
H
H
H
H
H
H
H
0.13
0.35
0.70
0.30
0.11
0.94
0.068
0.049
0.70
5.5
2.6
0.61
21
0.24
0.26
4.8
0.39
0.83
2.8
0.94
0.33
3.1
0.11
0.11
2.3
24
22
3.7
101
1.6
1.2
46
1.6
When dosed in the rat (0.5 mg/kg iv), the half-life of 8a
was 1.0 h with a clearance of 6.8 mL/min/kg. Oral
administration of 8a (1.0 mg/kg) led to good overall
blood levels (AUCnorm=3.6 mM h) and an oral bioa-
vailability of 61%. In order to test the effect of a potent
PDE3B inhibitor on lipolysis and metabolic rate, ana-
logue 8a was selected as a benchmark for further biolo-
gical evaluation.
9bb
14gb
14hb
14ic
H
2-Me
3-Me
2-Et
2-F
3-F
2-Cl
2,5-F2
14jc
Ph
Ph
Ph
Ph
Ph
Ph
14kc
14lc
14mc
14nc
14oc
To determine whether 8a would evoke lipolysis in vivo,
anesthetized rats were administered the compound in
increasing doses via an intravenous infusion. Elevation
of blood glycerol levels compared to vehicle was mea-
sured at 60 min to provide an index of lipolytic activity.
We were pleased to find that 8a caused a dose depen-
dent increase in glycerol levels, with an ED50=0.60 mg/
kg. Over a 90 min time period, the increase in plasma
glycerol after administration of 8a was time dependent
and observed to be maximum at 60 min post dose (1.0
mg/kg iv).
0.62
aValues are means of two or more experiments.
b>94% ee (R).
cRacemic.
Table 3. Inhibition of PDE3 subtypes by vinylogous amide pyr-
azolones
Entry
Ar=
R= PDE3B IC50 (nM)a PDE3A IC50 (nM)a
18a
18b
18c
18d
18e
18f
18g
18h
18i
18j
18k
18l
18m
18n
Ph
Ph
3-CNPh
Ph
3-CNPh
3-NO2Ph
2-ClPh
2,6-Cl2Ph
2-FPh
3-FPh
4-FPh
3-NO2Ph
2-ClPh
H
2-Me
2-Me
2-F
2-F
2-F
2-F
2-F
2-F
2-F
10
4.8
10
2.5
3.2
4.5
0.53
0.13
1.1
5.6
3.1
53
38
130
20
69
150
7.2
2.2
16
43
63
550
77
Scheme 3. Synthesis of dimethylpyrazolones 18: (a) Me2CHCO2Me,
LDA, THF, À78 ꢀC; (b) Dess–Martin periodinane, DCM; (c) Xant-
phos, Pd2(dba)3, THF, Cs2CO3, CbzNH2, Á; (d) NH2NH2, EtOH, Á;
(e) Pd(OH)2, H2, EtOH; (f) 7, toluene, TsOH, DMSO, Á; (g) BnBr,
THF, NaH, Bu4NI; (h) Me2CBrCO2Et, Zn, THF, Á then HCl, H2O;
(i) 2-PyrNTf2, THF, KHMDS, À78 ꢀC.
2-F
2-Cl
2-Cl
30
6.3
0.33
3-CNPh 2,3-F2
7.6
aValues are means of at least two experiments.

3986
S. D. Edmondson et al. / Bioorg. Med. Chem. Lett. 13 (2003) 3983–3987
The observed increase of plasma glycerol indicates an
increase of lipolysis leading to higher blood levels of
glycerol and free fatty acids. Since free fatty acids
undergo thermogenic oxidation in brown adipose tissue
(where PDE3B is localized), we next sought to measure
increases in metabolic rate after exposure to 8a. We thus
administered 8a orally to male rats in increasing doses (1,
3, and 10 mg/kg) and measured percentage increases in
thermogenesis compared to the predose metabolic rate.
The potent PDE3B inhibitor 8a showed a significant
increase in metabolic rate at 3.0 and 10.0 mg/kg (14 and
21%, respectively) over an 8-h period post dosing.
2. (a) Manganiello, V. C.; Taira, M.; Degerman, E.; Bel-
frage, P. Cell. Signal. 1995, 7, 445. (b) Alvarez, R.; Bane-
rjee, G. L.; Bruno, J. J.; Jones, G. H.; Littschwager, K.;
Strosberg, A. M.; Venuti, M. C. Mol. Pharmacol. 1986, 29,
554.
3. For a lead reference describing this compound, see: Jaski,
B. E.; Filer, M. A.; Wright, R. F.; Braunwald, E.; Colucci,
W. S. J. Clin. Invest. 1985, 75, 643.
4. Nishi, T.; Tabusa, F.; Tanaka, T.; Ueda, H.; Shimizu, T.;
Kanabe, T.; Kimura, Y.; Nakagawa, K. Chem. Pharm. Bull.
1983, 31, 852.
5. For a lead reference, see: Murray, K. J.; Eden, R. J.; Dolan,
J. S.; Grimsditch, D. C.; Stutchbury, C. A.; Patel, B.;
Knowles, A.; Worby, A.; Lynham, J. A.; Coates, W. J. Br. J.
Pharmacol. 1992, 107, 463.
In order to determine the effects of 8a on cardiovascular
activity, we administered this compound intravenously
to anesthetized rats in increasing doses (0.01 to 0.1 mg/
kg) and measured blood pressure over 60 min. A rapid
(earliest time point=5 min) and sustained (60 min=
duration of study) decrease in mean arterial pressure
was observed in response to 8a. The maximal decrease
in blood pressure was 40%, obtained at 0.1 mg/kg, 60
min post dosing and a 20% decrease was observed at
0.02 mg/kg. These cardiovascular effects are likely due
to the off target PDE3A inhibition, which is known to
induce vasodilation and is localized primarily in blood
vessels and heart. Nevertheless, it is also possible that
PDE3B inhibition could lead to the above effects.
6. Hirose, H.; Mashiko, S.; Kimura, T.; Ishida, F.; Mochi-
zuki, N.; Nishibe, T.; Nishikibe, M. J. Cardiovasc. Pharm.
2000, 35, 586.
7. (a) Maurice, D. H.; Liu, H. Br. J. Pharmacol. 1998, 125,
1501. (b) Degerman, E.; Belfrage, P.; Manganiello, V. C. J.
Biol. Chem. 1997, 272, 6823. (c) Reinhardt, R. R.; Chin, E.;
Zhou, J.; Taira, M.; Murata, T.; Manganiello, V. C.; Bondy,
C. A. J. Clin. Invest. 1995, 95, 1528. (d) Taira, M.; Hockman,
S. C.; Calvo, J. C.; Taira, M.; Belfrage, P.; Manganiello, V. C.
J. Biol. Chem. 1993, 268, 18573. (e) Meacci, E.; Taira, M.;
Moos, M.; Smith, C. J.; Movesian, M. A.; Degerman, E.;
Belfrage, P.; Manganiello, V. Proc. Natl. Acad. Sci. U.S.A.
1992, 89, 3721.
8. For a lead reference, see: (a) Snyder, P. B. Emerg. Ther.
Targets 1999, 3, 587. (b) Ekolm, D.; Hemmer, B.; Gao, G.;
Vergelli, M.; Martin, R.; Manganiello, V. C. J. Immunol. 1997,
20, 1529. (c) Murata, T.; Taira, M.; Manganiello, V. C. FEBS
Lett. 1996, 390, 29.
9. (a) Leroy, M.-J.; Degerman, E.; Taira, M.; Murata, T.;
Wang, L. H.; Movsesian, M. A.; Meacci, E.; Mangianello,
V. C. Biochemistry 1996, 35, 10194. (b) Snyder, P. B.; Beaton,
G.; Rueter, J. K.; Fanning, D. L.; Warren, S. D.; Hadidaruah,
S. S. W.O. Patent 070,469, 2002.
10. (a) Mertens, A.; Friebe, W.-G.; Muller-Beckmann, B.;
Kampe, W.; Kling, L.; von der Saal, W. J. Med. Chem. 1990,
33, 2870. (b) Coates, W. J.; Prain, H. D.; Reeves, M. L.;
Warrington, B. H. J. Med. Chem. 1990, 33, 1735. (c) Moos,
W. H.; Humblet, C. C.; Sircar, I.; Rithner, C.; Weishaar, R. E.;
Bristol, J. A.; McPhail, A. T. J. Med. Chem. 1987, 30, 1963.
(d) Sircar, I.; Weishaar, R. E.; Kobylarz, D.; Moos, W. H.;
Bristol, J. A. J. Med. Chem. 1987, 30, 1955.
In summary, a series of potent and subtype selective 2-
benzylvinylogous amide derived PDE3B inhibitors were
prepared and tested in vitro. Analysis of a series of
dihydropyridazinones led to the discovery of the most
potent PDE3B inhibitors yet reported (9a and 9b).
Investigations into a series of dimethylpyrazolones led
to the discovery of several potent PDE3B inhibitors that
are the most subtype selective compounds reported to
date (e.g., 18f, 3A/3B=33 and 18n, 3A/3B=23).
Benchmark compound 8a was then shown to stimulate
lipolysis in adipocytes with an ED50=0.6 mg/kg. Orally
administered 8a increased metabolic rate in the rat at 3
and 10 mg/kg. Finally, 8a showed a significant lowering
of blood pressure in the rat at doses as low as 0.01 mg/
kg. The results described above suggest that an orally
active, potent, and selective PDE3B inhibitor may be
therapeutically useful for the treatment of obesity.
Nevertheless, a compound substantially more selective
than 8a will be necessary to ultimately determine if
hypotension is an effect of PDE3B inhibition.
11. Owings, F. F.; Fox, M.; Kowalski, C. J.; Baine, N. H. J.
Org. Chem. 1991, 56, 1963.
12. Rajamannar, T.; Palani, N.; Balasubramanian, K. K. Syn.
Commun. 1993, 23, 3095.
13. Edmondson, S. D.; Mastracchio, A.; Parmee, E. R. Org.
Lett. 2000, 2, 1109.
14. The catalytic regions of PDE3B and PDE3A were expres-
sed as soluble proteins in Escherichia coli. The human PDE3B
gene fragment was inserted into a pET30a vector (Novagen)
such that an S-tag and a poly-histidine tag would be added to
the N-terminus of the protein starting at amino acid 387. The
human PDE3A gene fragment was inserted into a pET23a
vector such that a T7 tag would be added to the N-terminus of
PDE3A starting at amino acid 388. The proteins were expres-
sed in E. coli BL21 (DE3) pLysS following an IPTG induction.
The soluble protein extract from E. coli was precipitated by
ammonium sulfate (50% saturation) and resuspended in an
equivalent volume of buffer (20 mM Tris, pH 7.4, 75 mM
NaCl, 2 mM MgCl2, 1 mM DTT, 0.5 mM EDTA, 1:1000
poly-histidine tagged protein protease inhibitors (Sigma-
Aldrich), and 35% ethylene glycol). After an overnight dia-
lysis against the above buffer, the sample was purified over
DEAE Sepharose FF (40–70 mg protein/mL resin, Amersham
Acknowledgements
We thank Dr. Steven L. Colletti and Mr. Richard Ber-
ger for experimental procedures helpful for the conver-
sion of 10 to 11.
References and Notes
1. (a) Soderling, S. H.; Beavo, J. A. Curr. Opin. Cell Biol.
2000, 12, 174. (b) Beavo, J. A. Phsyiol. Rev. 1995, 75, 725. (c)
Weishaar, R. E.; Cain, M. H.; Bristol, J. A. J. Med. Chem.
1985, 28, 537.

S. D. Edmondson et al. / Bioorg. Med. Chem. Lett. 13 (2003) 3983–3987
3987
Biosciences) equilibrated in the same buffer. The column was
washed with buffer plus an additional 70 mM NaCl and eluted
with a 70–250 mM NaCl gradient. The partially purified
PDE3A and PDE3B was stored at À80 ꢀC until used.
mmol). The reactions were conducted in a total volume of 100
mL at room temperature for 2–15 min. Data was analyzed in
Graph Pad Prism software.
15. The S enantiomer of 8a was 5.1 nM at PDE3B and 26 nM
at 3A.
16. Sircar, I.; Morrison, G. C.; Burke, S. E.; Skeean, R.;
Weishaar, R. E. J. Med. Chem. 1987, 30, 1724.
17. Yin, J.; Buchwald, S. L. Org. Lett. 2000, 2, 1101.
PDE3 enzyme activity was determined using the phospho-
diesterase [3H]cAMP SPA enzyme assay (Amersham Life
Science). The assay buffer was 50 mM Tris pH 7.5, 8.3 mM
MgCl2, 1.7 mM EGTA and 10 nM [3H] cAMP (ꢁ40 Ci/