Identification, synthesis and SAR of amino substituted pyrido[3,2b]pyrazinones as potent and selective PDE5 inhibitors

Article abstract of DOI:10.1016/j.bmcl.2009.06.012

A new class of potent and selective PDE5 inhibitors is disclosed. Guided by X-ray crystallographic data, optimization of an HTS lead led to the discovery of a series of 2-aryl, (N8)-alkyl substituted-6-aminosubstituted pyrido[3,2b]pyrazinones which show p

Full text of DOI:10.1016/j.bmcl.2009.06.012

Bioorganic & Medicinal Chemistry Letters 19 (2009) 4088–4091
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
Identification, synthesis and SAR of amino substituted pyrido[3,2b]pyrazinones
as potent and selective PDE5 inhibitors
b
b
b
Dafydd R. Owen a, John K. Walker b, , E. Jon Jacobsen , John N. Freskos , Robert O. Hughes ,
David L. Brown b, Andrew S. Bell a, David G. Brown a, Christopher Phillips a, Brent V. Mischke b,
John M. Molyneaux b, Yvette M. Fobian b, Steve E. Heasley b, Joseph B. Moon b, William C. Stallings b,
D. Joseph Rogier b, David N. A. Fox a, Michael J. Palmer a, Tracy Ringer a, Margarita Rodriquez-Lens a,
Jerry W. Cubbage b, Radhika M. Blevis-Bal b, Alan G. Benson b, Brad A. Acker b, Todd M. Maddux b,
Michael B. Tollefson b, Brian R. Bond c, Alan MacInnes c, Yung Yu c
*
a Department of Chemistry, Pfizer Global Research and Development, Ramsgate Road, Sandwich, Kent CT13 9NJ, UK
b Department of Chemistry, Pfizer Global Research and Development, Chesterfield Parkway West, St. Louis, MO 63017, USA
c Department of Biology, Pfizer Global Research and Development, Chesterfield Parkway West, St. Louis, MO 63017, USA
a r t i c l e i n f o
a b s t r a c t
Article history:
A new class of potent and selective PDE5 inhibitors is disclosed. Guided by X-ray crystallographic data,
optimization of an HTS lead led to the discovery of a series of 2-aryl, (N8)-alkyl substituted-6-aminosub-
stituted pyrido[3,2b]pyrazinones which show potent inhibition of the PDE5 enzyme. Synthetic details
and some structure–activity relationships are also presented.
Received 8 May 2009
Revised 29 May 2009
Accepted 2 June 2009
Available online 6 June 2009
Ó 2009 Elsevier Ltd. All rights reserved.
Keywords:
PDE5
PDE6
PDE11
Inhibitors
Cyclic guanidine monophosphate (cGMP) is a second messenger
for nitric oxide and atrial natriuretic peptide. Increasing levels of
cGMP activates protein kinase G which through phosphorylation
of other downstream proteins promotes vasorelaxation. Phospho-
diesterase type 5 (PDE5) is widely expressed in smooth muscle,
lung, kidney and platelets and plays a major role in regulating
intracellular levels of cGMP through degradation of cGMP to the
inactive 50-GMP.1 For this reason, PDE5 inhibitors were originally
studied as potential anti-hypertensive or antianginal agents. The
discovery of the potent PDE5 inhibitor sildenafil (ViagraÒ), Figure
1, led to breakthrough therapies for male erectile dysfunction
and has spurred development of additional PDE5 agents such as
vardenafil (LevitraÒ) and tadalafil (Cialis Ò).2,3 More recently, silde-
nafil was approved for the treatment of pulmonary hypertension
(RevatioÒ) further illustrating the potential of PDE5 inhibitors as
therapeutic agents.4
O
N
N
HN
O
O
N
O
N
S
N
N
N
OEt
Sildenafil
NH2
1
Figure 1.
class of inhibitors is a less than ideal selectivity profile against
other members of the PDE family.5–7
As part of our on-going effort8 to identify potent, selective, and
long-acting PDE5 inhibitors, a high-throughput screen of the Pfizer
compound file was initiated. The screen yielded the phenyl substi-
tuted quinolin-4-one derivative 1 (Fig. 1) as a potent PDE-5 inhib-
itor (IC50 = 51 nM) but with poor selectivity for PDE6 (IC50 = 6 nM).
To facilitate further optimization of this lead, we obtained an X-
ray crystal structure of 1 bound in the active site of PDE59 (see
Fig. 2). A comparison between 1 and sildenafil in the active site
revealed the n-propyl group of 1 fits into the lipophilic pocket
As the potential therapeutic uses of PDE5 inhibitors grow there
is an increasing need to develop new classes of inhibitors more
suitable for chronic usage. One liability of the current marketed
* Corresponding author. Tel.: +1 636 247 8895.
E-mail address: john.k.walker@pfizer.com (J.K. Walker).
0960-894X/$ - see front matter Ó 2009 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2009.06.012

D. R. Owen et al. / Bioorg. Med. Chem. Lett. 19 (2009) 4088–4091
4089
Various lipophilic groups
Optimize for
lipophilic pocket
R1
O
N
N
N
Ar
R2
N
H
4
Build out towards solvent front
Figure 4. Representative pyridopyrazinone scaffold to serve as the template for our
analog program.
solvent, the amino-pyrido[3,2b]pyrazinone (Fig. 4) was particularly
attractive as a starting point for further optimization.
Figure 2. Overlay of co-crystal structure for compound 1 (orange) and sildenafil
In addition to allowing for the incorporation of a third substitu-
ent, the pyrazinones could be obtained through a flexible synthesis
that would allow for rapid generation of SAR for each of the three
regions. Furthermore, compound 4 would strike a good balance be-
tween decreasing lipophilicity and maintaining the same binding
mode observed with the previous leads as depicted in Figure 5.
Toward these ends, we initiated efforts to prepare analogs in the
pyridazinone template. Synthesis of the pyridopyrazinones was
carried out according to Scheme 1. Reaction of a primary amine
(purple) in the active site.
occupied by the ethoxy group of sildenafil. Additionally, the phenyl
ring of 1 pointed into a deep lipophilic pocket not occupied by sil-
denafil. Finally, inspection of the crystal structure revealed that the
solvent exposed space, occupied by the sulfonylpiperazine group in
sildenafil, was unoccupied by 1 suggesting a potential handle for
physical property and PK modulation via the incorporation of polar
functionality (Fig. 3).
Our enthusiasm for 1 was tempered, however, by concerns over
the drug ability of the quinolin-4-one scaffold. In addition, 1 was
plagued by an inefficient and lengthy synthesis which presented
limited handles for systematic modification. The crystal structure
of 1 suggested that replacement of the quinolin-4-one system, to
address these concerns, with similarly substituted 6, 6 fused ring
systems should be tolerated. We first explored this idea by exam-
ining a small set of related 2-quinolones represented by 2 (Fig. 2).
We were pleased to find compound 2 maintained activity on PDE5
(IC50 = 25 nM) and, more importantly, demonstrated some selec-
tivity against PDE6 (sixfold). While encouraged by this result we
also viewed the 2-quinolone scaffold to be a non-optimal template
suffering from high lipophilicity and the presence of a potential
Michael acceptor. Building off compound 2, we next turned our
attention to the dihydropyrazinone core exemplified by compound
3. Gratifyingly, 3, which has a lower log D and lacks the Michael
acceptor, also maintained similar potency and selectivity to 2
(PDE5 IC50 = 2 nM, PDE6 IC50 = 20 nM); however the aqueous solu-
bility of this series was still poor. Similar to what we observed in
the analysis of 1 the crystal structure of 3 (not shown) bound to
PDE5 suggested that the lower left region of the molecule pointed
to solvent and would allow for the incorporation of an ionizable
center to potentially improve solubility. Of the 6, 6 ring systems
which we envisioned capable of projecting substituents toward
Figure 5. Overlay of co-crystal structure of compound
minimized conformation of 3 (orange) and prototype pyridazinone inhibitor 22
(blue).
1 (green) with the
R1
Cl
N
Cl
c
HN
N
Cl
a, b
O2N
H2N
5
6
R1
R1
O
N
N
N
Cl
d, e
O
O
N
N
Cl
R2
N
H
N
H
F
O
7
8
F
N
R1
O
N
N
O
N
O
N
N
Ar
f
N
R2
N
H
N
2
3
9-38
PDE5 IC50 = 25 nM
PDE6 IC50 = 150 nM
PDE11 IC50 = 41 nM
PDE5 IC50 = 43 nM
PDE6 IC50 = 1160 nM
PDE11 IC50 = > 2000 nM
Scheme 1. General synthetic scheme to the pyridopyrazinone scaffold. Reagents
and conditions: (a) R1CH2NH2, Na2CO3, iPrOH, 0–20 °C; (b) 55 PSI H2, Raney Ni; (c)
oxalic acid, 4 N HCl, reflux 16 h; (d) oxalyl chloride, DMF, CH2Cl2; (e) R2NH2, Et3N,
CH2Cl2; (f) aryl boronic acid, Pd(Ph3)4,Na2CO3 glyme:H20 (5:1), reflux 24 h.
Figure 3. Structure and PDE data for a set of rationally designed inhibitors.

4090
D. R. Owen et al. / Bioorg. Med. Chem. Lett. 19 (2009) 4088–4091
Table 1
with 2,6-dichloro-3-nitropyridine (5) occurred selectively at the 2-
position and gave amino-substituted pyridine 6. Reduction of the
nitro group followed by reaction with oxalic acid/HCl gave rise to
the intermediate dione 7. In turn, 7 was exposed to oxalyl chloride
in the presence of DMF to give a chloroimidate intermediate which
upon exposure to either a primary or secondary amine under
mildly basic conditions yielded amino-substituted pyridopyrazi-
none compound 8. With both the R1 and R2 variables introduced
the synthesis was completed by standard Suzuki coupling of 8 with
various aryl boronic acids to give analogs 9–38.
Structure–activity relationships for the R1 and R2 positions
R1
F
O
N
N
N
X
R2
N
H
PDE isoformsa
#
R1
R2
X
5
6
11
Table 1 presents structure–activity relationships (SAR) of a rep-
resentative set of analogs aimed at probing the breadth and scope
of both the R1 and R2 substituents. Initially, we focused on analogs
with the aryl substituent fixed as either the 4-fluorophenyl or 3,4-
difluorophenyl group. In general, the SAR highlighted that PDE5
activity was more sensitive to the nature of the R1 substituent than
to the R2 substituent. We identified several analogs with PDE5
potency in the low nanomolar range with varying degrees of selec-
tivity against PDE 6 and 11. For example, when R1 was p-methoxy-
phenyl (e.g., 9–10) we obtained potent analogs in the single digit
nanomolar range with good selectivity (ꢀ100ꢁ) against PDE11
but only modest selectivity against PDE6. The 3,5-difluorophenyl
analog, 13 was found to be potent against PDE5 but also shows
good selectivity against both PDE6 and PDE11. Pyridyl groups were
also explored with the 3-pyridyl analog, 11 being preferred as it
9
H
1.3
10.1
345
482
474
970
390
145
OMe
O
N
10
11
12
13
14
F
2.0
11.6
375
920
910
114
OMe
O
O
N
N
H
H
H
H
2.5
N
O
O
H
20.3
2.7
F
Table 2
F
Structure–activity relationships for the R1 and Ar positions when R1 is
cyclohexylmethyl
28.5
O
O
15
16
H
H
3.0
4.6
459
240
324
405
O
N
N
N
Ar
R2
N
H
N
N
PDE isoformsa
6
#
R2
O
O
Ar
5
11
17
18
F
1.6
195
871
213
558
O
N
N
O
25
1.0
41.6
33.5
72.0
32.0
25.2
36.3
112
286
N
H
13.4
N
26
27
28
29
30
31
0.9
0.3
0.3
0.4
0.3
2.9
36.5
19
20
H
F
3.5
>2000
979
90.9
N
O
N
N
N
O
N
7.7
8.1
52.5
1,660
O
O
O
O
O
N
O
O
O
21
22
23
24
H
H
H
H
149
22.4
33.3
224
NDb
NDb
O
5.9
>2000
1850
NDb
>2000
>2000
NDb
O
O
N
O
O
N
N
O
8.2
O
O
N
N
OH
O
79.2
a
Assay protocols can be found in Ref. 10 IC50 values are reported in nM.
ND; Not Determined.
b
a
Assay protocols can be found in Ref. 10 IC50 values are reported in nM.

D. R. Owen et al. / Bioorg. Med. Chem. Lett. 19 (2009) 4088–4091
4091
Table 3
4-methoxypyridyl group was found to be optimal in terms of
PDE5 potency (e.g., 27–29) yielding a number of analogs in the
sub-nanomolar range. Unfortunately, selectivity falls off especially
for PDE11 which was found to be significantly diminished in this
series. Additionally, these compounds suffered from poor aqueous
solubility. For example, the aqueous solubility of 29 was found to
Structure–activity relationships for the R2 and Ar positions when R1 is n-propylethoxy
O
O
N
N
N
Ar
be <3 lM, at pH 7.4.
R2
The results for when R1 was ethoxypropyl are summarized in
Table 3. In general, these analogs tended to be less potent (ꢀ5–
10-fold) than their corresponding cyclohexyl variants found in
Table 2 (see for e.g., 29 vs 35 for a comparison). We also noted
a similar trend with respect to the 2-methoxypyridine substitu-
ent which was again found to be optimal with regard to PDE5
potency as highlighted by examples 34–36. More importantly,
however, these compounds demonstrated good selectivity
against PDE6 (>100ꢁ) and excellent selectivity against PDE11
(>600ꢁ). Poor solubility was still as issue with this series as
N
H
PDE isoformsa
6
#
R2
Ar
5
11
N
O
32
36.5
1980
1170
1190
350
>2000
O
O
N
N
36 was measured to have a solubility of 4.6 lM, at pH 7.4. Final-
33
34
35
36
37
38
33.9
14.5
2.3
>2000
1210
1230
1540
>2000
NDb
ly, as a representative example for the series, we also assessed
the rat11,12 pharmacokinetic (PK) profile of 36. This compound
exhibited high clearance (67.6 ml/min/kg) resulting in a modest
half-life (ꢀ3 h) indicating further refinement of the template
would be required. Efforts to further optimize this scaffold to
improve the PK properties, as well as the physical–chemical
properties, will be discussed in the following Letter of this
journal.
In conclusion, we report the discovery of a novel series of 2-
amino substituted pyrido[3,2b]pyrazinone analogs that show po-
tent and selective inhibition of PDE5. Starting from a novel lead,
1, identified in a high-throughput screen and relying heavily on
X-ray crystal structures we developed the pyridopyrazinone tem-
plate 4 through a series of lead-hopping exercises. We further opti-
mized the template to identify a handful of potent and selective
analogs.
N
O
N
O
N
N
O
O
O
2.4
317
O
O
O
N
N
N
N
N
O
26.7
534
>2000
NDb
N
OH
References and notes
1. (a) Beavo, J. A. Physiol. Rev. 1995, 75, 725; (b) Corbin, J. D.; Francis, S. H. J. Biol.
Chem. 1999, 274, 13729.
2. Haning, H.; Niewöhner, U.; Bischoff, E. Prog. Med. Chem. 2003, 41, 249.
3. Stamford, A. W. Annu. Rep. Med. Chem. 2002, 37, 53.
a
Assay protocols can be found in Ref. 10. IC50 values are reported in nM.
ND; not determined.
b
4. Ghofrani, H. A.; Osterloh, I. H.; Grimminger, F. Nat. Rev. Drug Disc. 2006, 5, 689.
5. (a) Pomara, G.; Morelli, G. Int. J. Impot. Res. 2005, 17, 385; (b) Pomara, G.;
Morelli, G.; Canale, D.; Turchi, P.; Caglieresi, C.; Liguori, G.; Selli, C.; Macchia, E.;
Martino, E.; Frencesca, F. Fertil. Steril. 2007, 88, 860.
was found to be more active and show greater sparing of PDE6 and
PDE11 compared to 2-pyridyl analog 12.
6. (a) Goldstein, I.; Lue, T. F.; Padma-Nathan, H.; Rosen, R. C.; Steers, W. D.;
Wicker, P. A. N. Eng. J. Med. 1998, 338, 1397; (b) Cote, R. H. Int. J. Impot. Res.
2004(16), S28.
We also explored the saturated cyclohexyl ring which generally
gave potent analogs (e.g., 15–17) with good (>100ꢁ) selectivity for
PDE6 and PDE11. We also observed some impact on activity and/or
selectivity due to the nature of R2 substituent (e.g., 18–19). When
R1 was a simple alkyl group or alkyl ether a drop in PDE5 potency
was observed which was also accompanied by an equal or greater
loss in PDE6 and PDE11 activity. The chain length proved critical
for activity. The ethoxyisopropyl analog, 20, and ethoxyethyl ana-
log, 21, both exhibit only modest inhibition of PDE5. However,
when the chain was lengthened by one atom to ethoxypropyl, a
nearly sixfold improvement in activity, relative to 21, was observed
with only minimal or no increase in either PDE6 or PDE11 activity
(e.g., 22 and 23).
We next probed the SAR around the aryl substituent. We
elected to focus on the scaffolds where R1 was fixed as either the
cyclohexyl or ethoxypropyl group. Table 2 summarizes the SAR ob-
served in the template where R1 was cyclohexyl. Within this series
PDE5 potency was found to be relatively unaffected by the aryl
group but selectivity was found to be negatively impacted. The
7. Manallack, D. T.; Hughes, R. A.; Thompson, P. E. J. Med. Chem. 2005, 48, 3449.
8. (a) Palmer, M. J.; Bell, A. S.; Fox, D. N.; Brown, D. G. Curr. Top. Med. Chem. 2007,
7, 405; (b) Heasley, S. E.; Jacobsen, E. J.; Walker, J. K.; Hughes, R. O.; Freskos, J.;
Brown, D. L.; Fobian, Y. M.; Owen, D.; Mischke, B. V.; Tollefson, M. B.;
Molyneaux, J.; Blevis-Bal, R.; Maddux, T.; Brown, D. G.; Moon, J. B.; Phillips, C.;
Rogier, D. J.; Benson, A. Abstracts of Papers, 236th National Meeting of the
American Chemical Society, Philadelphia, PA, Aug 17–21, 2008; American
Chemical Society: Washington, DC, 2008; MEDI-397.
9. The coordinates have been deposited in the PDB under accession code 3HDZ.
10. Brown, D. L.; Owen, D. R.; Phillips, C.; Palmer, M. J.; Bell, A. S.; Freskos, J. N.;
Fobian, Y. M.; Walker. J. K.; Hughes, R.; Jacobsen, E. J.; Tollefson, M. B.; Brown,
D. G.; Mischke, B. V.; Molyneaux, J. M.PCTWO 07/020521.
11. The Pfizer Institutional Animal Care and Use Committee reviewed and
approved the animal use in these studies. The animal care and use program
is fully accredited by the Association for Assessment and Accreditation of
Laboratory Animal Care, International.
12. PK experiments were conducted in Sprague Dawley rats under the following
conditions; IV: 2 mpk (n = 3) vehicle: 70% PEG400/20% 0.05 M citrate buffer/
10% ethanol, pH 5; PO: 2 mpk (n = 3) vehicle: 0.5% methylcellulose/0.1% Tween
80 in 50 nM citric acid, pH 5.