Design, synthesis, and biological evaluation of 3-[4-(2-hydroxyethyl) piperazin-l-yl]7-(6-methoxypyridin-3-yl)-l-(2-propoxyethyl)pyrido[3,4-b] pyrazin-2(lH)-one, a potent, orally active, brain penetrant inhibitor of phosphodiesterase 5 (PDE5)

Article abstract of DOI:10.1021/jm901781q

We recently described a novel series of aminopyridopyrazinones as PDE5 inhibitors. Efforts toward optimization of this series culminated in the identification of 3-[4-(2-hydroxyethyl)piperazin-l-yl]7-(6-methoxypyridin-3-yl)- l-(2-propoxyethyl)pyrido[3,4-è]pyrazin-2(l//)-one, which possessed an excellent potency and selectivity profile and demonstrated robust in vivo blood pressure lowering in a spontaneously hypertensive rat (SHR) model. Furthermore, this compound is brain penetrant and will be a useful agent for evaluating the therapeutic potential of central inhibition of PDE5. This compound has recently entered clinical trials.

Full text of DOI:10.1021/jm901781q

2656 J. Med. Chem. 2010, 53, 2656–2660
DOI: 10.1021/jm901781q
Design, Synthesis, and Biological Evaluation of 3-[4-(2-Hydroxyethyl)piperazin-1-yl]-
7-(6-methoxypyridin-3-yl)-1-(2-propoxyethyl)pyrido[3,4-b]pyrazin-2(1H)-one, a Potent,
Orally Active, Brain Penetrant Inhibitor of Phosphodiesterase 5 (PDE5)
Robert O. Hughes,*,† D. Joseph Rogier,† E. Jon Jacobsen,† John K. Walker,† Alan MacInnes,† Brian R. Bond,†
Lena L. Zhang,† Ying Yu,† Yi Zheng,† Jeanne M. Rumsey,† Jennie L. Walgren,† Sandra W. Curtiss,† Yvette M. Fobian,†
Steven E. Heasley,† Jerry W. Cubbage,† Joseph B. Moon,† David L. Brown,† Brad A. Acker,† Todd M. Maddux,†
Mike B. Tollefson,† Brent V. Mischke,† Dafydd R. Owen,‡ John N. Freskos,† John M. Molyneaux,† Alan G. Benson,† and
Rhadika M. Blevis-Bal†
†Pfizer Global Research and Development, Chesterfield Parkway West, St. Louis, Missouri 63017, and ‡Pfizer Global Research and Development,
Ramsgate Road, Sandwich CT139NJ, U.K.
Received December 1, 2009
We recently described a novel series of aminopyridopyrazinones as PDE5 inhibitors. Efforts toward
optimization of this series culminated in the identification of 3-[4-(2-hydroxyethyl)piperazin-1-yl]-
7-(6-methoxypyridin-3-yl)-1-(2-propoxyethyl)pyrido[3,4-b]pyrazin-2(1H)-one, which possessed an ex-
cellent potency and selectivity profile and demonstrated robust in vivo blood pressure lowering in a
spontaneously hypertensive rat (SHR) model. Furthermore, this compound is brain penetrant and will
be a useful agent for evaluating the therapeutic potential of central inhibition of PDE5. This compound
has recently entered clinical trials.
Introduction
that of 1. To implement this strategy, we focused on modifica-
tions to the southwest (SW) region of the molecule. As deter-
mined by crystallographic analysis, the SW substituent points
toward the solvent; therefore, modifications to this region of the
molecule would likely have minimal effect on the PDE potency
and selectivity. Herein, we described our efforts to identify a
basic compound from the aminopyridopyrazinone class. These
efforts culminated in the discovery of clinical candidate 13.
We considered the previously described basic compounds 2
and 35 as good starting points worthy of further optimization.
As summarized in Table 1 both compounds demonstrate
excellent potency against PDE5, derived from human plate-
lets, and good selectivity over PDE6 and PDE11. Both
compounds also possessed high (>100 μM) aqueous solubi-
lity. Encouragingly, in the case of the 3, the affinity for the
hERG channel (4.5 μM) was only modest. Unfortunately,
both compounds suffered from high clearance in the rat.
Accordingly, our goal was to improve the pharmacokinetic
profile of this subset of compounds while maintaining the
favorable aspects: potency, selectivity, solubility, and hERG
margin. As illustrated in Figure 2, we speculated that reducing
the flexibility and rotational degrees of freedom in the SW
substituent by constraining it into a ring system (as piperazine
or a homopiperazine) would result in an improvement in
metabolic stability. That is, we hypothesized that positive
impact of the removal of three rotatable bonds from 2 or 3 on
the pharmacokinetic profile of these compounds would out-
weigh any deleterious effects from the slight increase in
lipophilicity of the new molecules.
Gathered over the past decade, significant clinical and
commercial experience has demonstrated the safety, tolera-
tion, and efficacy of PDE5 inhibitors, such as sildenafil,
for the treatment of male erectile dysfunctionon (MEDa).1
Furthermore, experimental studies and clinical trials indicate
that PDE5 inhibitors are effective in the treatment of various
other diseases such as Raynaud’s disease, gastrointestinal
disorders, and stroke and exert cardioprotective effects.2
With this potential in mind and initially targeting the
discovery of compounds suitable for use as antihypertensive
agents, we initiated a program directed toward the discovery
of long-acting, selective inhibitors of PDE5. Previously, we
reported on the identification of the aminopyridopyrazinone
class of PDE5 inhibitors.3 Among the desirable features of
this class of PDE5 inhibitors was the excellent selectivity
over other PDE isoforms.4 Initial optimization of the profile
of this class of compounds led to neutral inhibitor 3-[(trans-4-
hydroxycyclohexyl)amino]-7-(6-methoxypyridin-3-yl)-1-(2-
propoxyethyl)pyrido[3,4-b]pyrazin-2(1H)-one,5 1 (Figure 1),
which possessed an excellent preclinical profile including
robust, sustained reduction in blood pressure (BP) in the
SHR model of hypertension. Upon further study, however,
this compound caused unacceptable toxicity6 whichprecluded
its further development. To overcome this shortcoming, one
strategy that we pursued was to develop compounds from
different physicochemical classes7 in the anticipation that
toxicological profile of these compounds would diverge from
*To whom correspondence should be addressed. Phone: (636) 247-
9303. Fax: (636) 247-5620. E-mail: Robert.O.Hughes@pfizer.com.
a Abbreviations: BP, blood pressure; SHR, spontaneously hypertensive
rat; PDE, phosphodiesterase; MED, male erectile dysfunction; CNS,
central nervous system; PK, pharmacokinetics; SD, Srague-Dawley.
Results and Discussion
The synthetic route into this class of molecules is summar-
ized in Scheme 1. Treatment of 4 with n-BuLi followed by
r
pubs.acs.org/jmc
Published on Web 03/02/2010
2010 American Chemical Society

Brief Article
Journal of Medicinal Chemistry, 2010, Vol. 53, No. 6 2657
Figure 1. Structure of 1.
Figure 2. Proposed cyclization of 2 and 3 to improve pharmacoki-
netic parameters. ΔclogP, ΔclogD, and Δrotbonds are from Proper-
tycyclized - Propertyacyclic
.
Table 1. Selected Data for 2 and 3g
and human microsomal systems. Additionally, all four com-
pounds displayed similar protein binding in rat and humans
with free fractions ranging from 18% to 30%. We were
gratified to find a dramatic improvement in the in vivo
clearance measured in the rat. Supporting our hypothesis that
reduced conformational flexibility would lead to improved
PK, 10 and 13 were significantly improved over the respective
acyclic counterparts. Interestingly, the improvement in the in
vivo PK profile was not predicted by the in vitro microsomal
assays.
During our optimization of the neutral alcohol subclass of
aminopyridopyrazinones, which led to 1, we noted that
methyl branching of the SW side chain resulted in improve-
ments in the PK profile.6 Accordingly, we sought to examine
the effect of this modification on 13. Table 4 summarizes the
key results of this effort. Overall, methylation of the hydroxyl
ethyl side chain of 13 was not beneficial. In the case of 17, 19,
and 21 selectivity over PDE6 was reduced (49-fold, 64-fold,
and 68-fold, respectively). Compound 20, which displayed a
similar PDE5, PDE6, and PDE11 profile to 13, was signifi-
cantly more potent at blocking the hERG channel (54 nM,
160-fold PDE5 IC50). Similar data were collected for 18: good
PDE5, PDE6, and PDE11 data but a significantly reduced
hERG margin (1000-fold) relative to 13. Likely the increase in
lipophilicity associated with the introduction of addition
methyl groups is driving increased affinity to the hERG
channel. As a result of these combined properties, 17-21
were not further studied.
In parallel with our efforts to expand the SAR around 13,
we elected to more extensively profile this analogue. Com-
pound 13 showed excellent selectivity versus all other PDE
isoforms and against a wide panel of binding assays.8 Addi-
tional pharmacokinetic studies were also encouraging. In the
dog, 13 exhibited low clearance (5.8 (mL/min)/kg) and high
bioavailability (92%). In a bile duct canulated rat study less
than 2% of parent was recovered over 6 h of monitoring.
Furthermore, in rats and dogs 13 showed negligible renal
clearance (<2%). Taken together, these data suggest that the
primary route of clearance for 13 is through hepatic metabo-
lism. The metabolite profile in rat, dog, and human micro-
somes and hepatocytes was similar, and the major metabolite
was loss of the n-propoxy group. Finally, consistent with this
compound’s excellent in vitro permeability, 13 demonstrated
4-fold higher brain concentrations than total plasma concen-
tration 2 h after a single subcutaneous dose of 10 mpk in rat,
suggesting the potential for significant CNS penetration of
this compound. As 13 demonstrated an attractive potency,
selectivity, and PK profile, we undertook a more detailed
evaluation of the PK/pharmacodynamic (PD) relationships
for this compound. For these studies we utilized reduction in
BP in the conscious SHR as the pharmacodynamic readout.
A single oral dose (2 mpk) of 13 reduced systolic BP by
a PDE5 IC50 (nM). b PDE6 selectivity defined as (PDE6 IC50)/
(PDE5 IC50). c PDE11 selectivity defined as (PDE11 IC50)/(PDE5
IC50). d Aqueous solubility measured at pH 7.4 (μM). e hERG IC50
(nM) determined in a patch-clamp experiment. f Clearance ((mL/min)/
kg) measured in the Srague-Dawley (SD) rat following a 1 mpk iv dose.
g PDE IC50 values are the averages of at least three determinations.
hERG IC50 is from one determination.
quenching with N-fluorobenzenesulfonimide (NFSI) gave the
desired fluoropyridine, 5, in 60% yield. Addition of propyloxy-
ethylamine to 5 was brought about in refluxing EtOH to give
the diamino ether 6 in 82% yield. The action of HCl to remove
the Boc group of 6 followed by treatment with methyl 2-chloro-
2-oxoacetate gave, upon heating, the desired dione 7. Reaction
of dione 7 with oxalyl chloride in the presence of a catalytic
amount of DMF gave chloroimidate 8 (66% for three steps)
which was reacted with cyclic amines to give 9 (50-90%). A
Suzuki reaction between the chloroaminopyridopyrazinone (9)
and 6-methoxypyridin-3-ylboronic acid completed the synth-
esis in good overall yield.
As shown in Table 2, constraint of the SW substituent was
well tolerated in terms of PDE5 potency. Indeed, comparison
of 2 and 10 (0.25 nM f 3.28 nM) and 3 and 13 (1.12 nM f
0.20 nM) reveals little impact of cyclization on potency. In
all cases selectivity over PDE11 was high. However, we
observed that in some cases selectivity over PDE6 eroded.
Specifically, homopiperazines 15 and 16 were less than
100-fold selective over PDE6 as was bicyclo derivative 11.
N-Ethyl derivative 12 and N-hydroxylethyl derivative 13
possessed excellent selectivity over PDE6. However, 10,
lacking a piperazine substituent, was less selective against
PDE6. Interestingly, homologation of 13 to 14 resulted
in similar potency against PDE5 but a 2.5-fold loss in
PDE6 selectivity. Piperazine analogues 10, 12, and 13 were
evaluated in the hERG patch-clamp experiment. Com-
pounds 12 (1450-fold) and 13 (8950-fold) showed excellent
selectivity for PDE5 over hERG activity.
Table 3 summarizes pharmacokinetic parameters for 2, 3,
10, and 13. All four compounds were found to be stable in rat

2658 Journal of Medicinal Chemistry, 2010, Vol. 53, No. 6
Hughes et al.
Scheme 1. General Synthetic Route for the Synthesis of 10-21a
a (a) TMEDA, n-BuLi, Et2O, -60 f -10 °C, 2 h; then NFSI, THF -60 f 0 °C, 1 h, 60%; (b) 2-propoxyethanamine, EtOH, reflux, 22 h, 82%; (c) 4 N
HCl, 1,4-dioxane, room temp, 1 h; (d) MeOC(O)C(O)Cl, i-Pr2NEt, CH2Cl2, 0 °C f room temp, 4 h; then PhMe, 105 °C; (e) (COCl) , DMF(cat),
2
CH2Cl2, room temp, 30 min, 66% for three steps; (f) amines, Et3N, THF, room temp, 4 h, 50-90%; (g) 6-methoxypyridin-3-ylboronic acid, (Ph3P)4P
(10 mol %), Na2CO3, 1,4-dioxane-EtOH, 100 °C, 2 h, 50-85%.
Table 2. PDE5 Potency and PDE6, PDE11, and hERG Selectivity Data
for Compounds 10-16
Table 3. Selected Metabolic and Pharmacokinetic Parameters for 2, 3,
10, and 13
microsomal stability
protein binding
rat PK
compd rat (%)a human (%)a rat (%) human (%) Clb F (%)c
2
98
100
86
97
95
86
85
83.4
77.4
84.2
82.9
69.4
78.8
81.2
67.8
170
582
42
ND
ND
ND
69
3
10
13
89
37
a Percent remaining after 30 min of incubation. b Clearance ((mL/min)/
kg) measured in the SD rat following a 1 mpk iv dose. c Bioavailability
measured the SD rat following a 1 mpk po dose.
25 mmHg, which was sustained for approximately 24 h (data
not shown). This study demonstrated that maximal BP low-
ering efficacy is achieved when free plasma exposures of 13 are
above ∼10-fold the PDE5 enzyme IC50. To further refine this
exposure-response relationship, changes in BP were moni-
tored following steady state iv infusion of escalating doses of
13. Data from this study enabled a detailed analysis of the
exposure-response relationship on an individual animal
basis. Figure 3 shows steady state exposure plotted against
the corresponding BP lowering efficacy for each individual
infusion period. These data enhanced our confidence that
maximal BP lowering efficacy would be achieved for 13 with a
free plasma concentration of 2.3 nM (11-fold PDE5 IC50).
Importantly, 13 was well tolerated in toxicological studies. On
the basis of the excellent, overall profile, 13 was selected for
clinical development. In healthy human volunteers 13 plasma
concentrations increased proportionally with dose. Absorp-
tion of 13 was rapid with a median Tmax estimate of 1-2 h
across all doses studied. Following attainment of Cmax
,
plasma 13 concentrations declined over time with a mean
apparent elimination half-life of approximately 17-25 h
across dose levels.
a PDE5 IC50 (nM). b PDE6 selectivity defined as (PDE6 IC50)/(PDE5
IC50). c PDE11 selectivity defined as (PDE11 IC50)/(PDE5 IC50).
d hERG IC50 (nM) determined in a patch-clamp experiment.

Brief Article
Journal of Medicinal Chemistry, 2010, Vol. 53, No. 6 2659
Figure 3. PK/PD relationship for 13 in the conscious SHR.
Table 4. PDE5 Potency and PDE6, PDE11, and hERG Selectivity Data
for Compounds 13-21
the role of PDE5 in the CNS. Additional data on this
compound will be reported in due course.9
Supporting Information Available: Procedures for the pre-
paration of 13; analytical data for 10-21; details of the PDE5,
PDE6, and PDE11 assays; additional selectivity data for 13;
procedures for the in vivo experiment summarized in Figure 3;
method for determining brain concentrations. This material is
available free of charge via the Internet at http://pubs.acs.org.
References
(1) Rashid, A. The efficacy and safety of PDE5 inhibitors. Clin.
Cornerstone 2005, 7, 47–56.
(2) Sandner, P.; Huetter, J.; Tinel, H.; Ziegelbauer, K.; Bischoff, E.
PDE5 inhibitors beyond erectile dysfunction. Int. J. Impotence Res.
2007, 19, 533–543.
(3) (a) Hughes, R. O.; Walker, J. K.; Cubbage, J. W.; Fobian, Y. M.;
Rogier, D. J.; Heasley, S. E.; Blevis-Bal, R. M.; Benson, A. G.; Owen,
D. R.; Jacobsen, E. J.; Freskos, J. N.; Molyneaux, J. M.; Brown,
D. L.; Stallings, W. C.; Acker, B. A.; Maddux, T. M.; Tollefson,
M. B.; Williams, J. M.; Moon, J. B.; Mischke, B. V.; Rumsey, J. M.;
Zheng, Y.; MacInnes, A.; Bond, B. R.; Yu, Y. Investigation of
aminopyridiopyrazinones as PDE5 inhibitors: evaluation of modifi-
cations to the central ring system. Bioorg. Med. Chem. Lett. 2009, 19,
4092–4096. (b) Owen, D. R.; Walker, J. K.; Jacobsen, E. J.; Freskos, J. N.;
Hughes, R. O.; Brown, D. L.; Bell, A. S.; Brown, D. G.; Phillips, C.;
Mischke, B. V.; Molyneaux, J. M.; Fobian, Y. M.; Heasley, S. E.; Moon,
J. B.; Stallings, W. C.; Rogier, D. J.; Fox, D. N. A.; Palmer, M. J.; Ringer,
T.; Rodriquez-Lens, M.; Cubbage, J. W.; Blevis-Bal, R. M.; Benson,
A. G.; Acker, B. A.; Maddux, T. M.; Tollefson, M. B.; Bond, B. R.;
MacInnes, A.; Yu, Y. Identification, synthesis and SAR of amino
substituted pyrido[3,2b]pyrazinones as potent and selective PDE5 inhi-
bitors. Bioorg. Med. Chem. Lett. 2009, 19, 4088–4091.
(4) Accordingly, we routinely only monitored the selectivity of com-
pounds against the closely related PDE6 and PDE11.
(5) Hughes, R. O.; Walker, J. K.; Rogier, J. D.; Heasley, S. E.;
Blevis-Bal, R. M.; Benson, A. G.; Jacobsen, E. J.; Cubbage, J. W.;
Fobian, Y. M.; Owen, D. R.; Freskos, J. N.; Molyneaux, J. M.;
Brown, D. L.; Acker, B. A.; Maddux, T. M.; Tollefson, M. B.;
Moon, J. B.; Mischke, B. V.; Rumsey, J. M.; Zheng, Y.; MacInnes,
A.; Bond, B. R.; Yu, Y. Optimization of the aminopyridopyrazi-
nones class of PDE5 inhibitors: discovery of 3-[(trans-4-hydro-
xycyclohexyl)amino]-7-(6-methoxypyridin-3-yl)-1-(2-propoxyethyl)-
pyrido[3,4-b]pyrazin-2(1H)-one. Bioorg. Med. Chem. Lett. 2009, 19,
5209–5213.
(6) In rats, 1 caused vacuolization of the pancreatic acinar cells.
(7) Hughes, J. D.; Blagg, J.; Price, D. A.; Bailey, S.; DeCrescenzo, G. A.;
Devraj, R. V.; Ellsworth, E.; Fobian, Y. M.; Gibbs, M. E.; Gilles, R.
W.; Greene, N.; Huang, E.; Krieger-Burke, T.; Loesel, J.; Wager, T.;
Whiteley, L.; Zhang, Y. Physiochemical drug properties associated
with in vivo toxicological outcomes. Bioorg. Med. Chem. Lett. 2008,
18, 4872–4875.
a PDE5 IC50 (nM). b PDE6 selectivity defined as (PDE6 IC50)/(PDE5
IC50). c PDE11 selectivity defined as (PDE11 IC50)/(PDE5 IC50).
d hERG IC50 (nM) determined in a patch-clamp experiment. e Racemic.
Relative chemistry known.
Conclusion
We have described the optimization of a basic series of
aminopyridopyrazinones as potent, selective, and safe inhibi-
tors of PDE5. This effort culminated with the identification of
the brain penetrant 13, 3-[4-(2-hydroxyethyl)piperazin-1-yl]-
7-(6-methoxypyridin-3-yl)-1-(2-propoxyethyl)pyrido[3,4-b]-
pyrazin-2(1H)-one, which willenable theclinicalevaluation of

2660 Journal of Medicinal Chemistry, 2010, Vol. 53, No. 6
Hughes et al.
(8) 13 was screened against other members of the PDE family (1A, 1B,
1C, 2, 3A, 3B, 4A, 4B, 4C, 4D, 7, 7B, 8A, 8B, and 9) and showed no
significant inhibition (IC50 > 2000 nM in all cases). Additional
selectivity data were obtained at CEREP. See Supporting Informa-
tion for additional details.
(9) Menniti, F. S.; Ren, J. M.; Coskran, T. M.; Liu, J.; Morton, D.;
Sietsma, D. K.; Som, A.; Stephenson, D. T.; Tate, B. A.; Finklestein, S.
P. Phosphodiesterase 5A inhibitors improve functional recovery after
stroke in rats: optimized dosing regimen with implications for mechan-
ism. J. Pharmacol. Exp. Ther. 2009, 331, 842–850.