Identification of multiple 5-HT4 partial agonist clinical candidates for the treatment of alzheimers disease

Article abstract of DOI:10.1021/jm300953p

The cognitive impairments observed in Alzheimers disease (AD) are in part a consequence of reduced acetylcholine (ACh) levels resulting from a loss of cholinergic neurons. Preclinically, serotonin 4 receptor (5-HT₄) agonists are reported to modulate cholinergic function and therefore may provide a new mechanistic approach for treating cognitive deficits associated with AD. Herein we communicate the design and synthesis of potent, selective, and brain penetrant 5-HT₄ agonists. The overall goal of the medicinal chemistry strategy was identification of structurally diverse clinical candidates with varying intrinsic activities. The exposure-response relationships between binding affinity, intrinsic activity, receptor occupancy, drug exposure, and pharmacodynamic activity in relevant preclinical models of AD were utilized as key selection criteria for advancing compounds. On the basis of their excellent balance of pharmacokinetic attributes and safety, two lead 5-HT₄ partial agonist candidates 2d and 3 were chosen for clinical development.

Full text of DOI:10.1021/jm300953p

Article
pubs.acs.org/jmc
Identification of Multiple 5‑HT4 Partial Agonist Clinical Candidates
for the Treatment of Alzheimer’s Disease
Michael A. Brodney,* David E. Johnson, Aarti Sawant-Basak, Karen J. Coffman, Elena M. Drummond,
Emily L. Hudson, Katherine E. Fisher, Hirohide Noguchi, Nobuaki Waizumi, Laura L. McDowell,
Alexandros Papanikolaou, Betty A. Pettersen, Anne W. Schmidt, Elaine Tseng, Kim Stutzman-Engwall,
David M. Rubitski, Michelle A. Vanase-Frawley, and Sarah Grimwood
Groton Laboratories, Pfizer Global Research and Development, Eastern Point Road, Groton, Connecticut 06340, United States
S
* Supporting Information
ABSTRACT: The cognitive impairments observed in Alzheimer’s disease (AD) are in part a consequence of reduced
acetylcholine (ACh) levels resulting from a loss of cholinergic neurons. Preclinically, serotonin 4 receptor (5-HT4) agonists are
reported to modulate cholinergic function and therefore may provide a new mechanistic approach for treating cognitive deficits
associated with AD. Herein we communicate the design and synthesis of potent, selective, and brain penetrant 5-HT4 agonists.
The overall goal of the medicinal chemistry strategy was identification of structurally diverse clinical candidates with varying
intrinsic activities. The exposure−response relationships between binding affinity, intrinsic activity, receptor occupancy, drug
exposure, and pharmacodynamic activity in relevant preclinical models of AD were utilized as key selection criteria for advancing
compounds. On the basis of their excellent balance of pharmacokinetic attributes and safety, two lead 5-HT4 partial agonist
candidates 2d and 3 were chosen for clinical development.
memory deficits with anticholinergic drugs (e.g., atropine and
scopolamine) has been used to simulate the cognitive
impairments associated with the loss of cholinergic function
in AD, and 5-HT4 agonists have been shown to reverse these
deficits.9,10 On the basis of these reports, 5-HT4 receptor
agonists might provide a new mechanistic approach for
providing symptomatic relief in AD.
A 5-HT4 agonist that is suitable for clinical development
must be potent and highly selective. Early generation 5-HT4
agonists, such as tegaserod (I) and cisapride (II), although
clinically effective in the treatment of gastrointestinal motility
disorders, caused serious adverse events that were likely related
to off-target pharmacology, a property that has severely
restricted their use (Figure 1).11 Second generation agonists
such as prucalopride (III) were recently approved in Europe
and Canada for the treatment of constipation. From the
perspective of treating central nervous systems (CNS)
indications, a major limitation of these early generation 5-
INTRODUCTION
■
Alzheimer’s disease (AD) is the most common cause of
dementia and is characterized by cognitive decline, primarily
affecting memory.1 The societal impact of caring for people
with AD and dementia is enormous. In 2011, family members
and unpaid caregivers provided 17 billion hours of care in the
U.S. and worldwide costs for 2009 were estimated at $422
billion, highlighting the need for more effective therapies.2,3
The exact cause of AD is not known; however, some of the
earliest pathology that occurs involves loss of acetylcholine
(ACh) releasing cholinergic neurons in brain regions critical to
cognitive function, such as the frontal cortex and hippo-
campus.4,5 Currently, AD treatment centers on the use of
acetylcholinesterase inhibitors that nonselectively increase ACh
throughout the brain. An approach that targets the cholinergic
circuitry within specific brain regions involved in cognition
might provide improved therapeutic relief.
5-HT4 receptors are G-protein-coupled receptors (GPCRs)
positively coupled to adenylate cyclase and are found in brain
areas that are critical for cognitive function, including the cortex
and hippocampus.6 Preclinical evidence indicates that 5-HT4
receptor agonists increase Ach-release in these areas.7,8 In
various behavioral models, pharmacological induction of
Special Issue: Alzheimer's Disease
Received: July 5, 2012
© XXXX American Chemical Society
A
dx.doi.org/10.1021/jm300953p | J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
Article
Figure 1. Literature 5-HT4 agonists (I−III) and lead chemical series (IV−VI).
Table 1. 5-HT4d Binding, Functional Activity, Passive Permeability, and in Vitro Human Microsomal Clearances of Analogues
e
1a−f
a
Binding Ki values (geometric mean; at least n = 3 measurements) were determined in HEK293 cells stably expressing the human 5-HT4d isoform
b
using [3H]GR113808 as the radioligand . Functional activity was assessed as the EC50 and Emax values for stimulation of cAMP production in
HEK293 cells stably expressing the human 5-HT4d receptor isoform. EC50 values (geometric mean; at least n = 3 measurements) represent the
concentration producing 50% of the maximal cAMP response. Emax values (arithmetic mean) represent the maximal effect on cAMP production
c
expressed as a percentage of a maximal response of 5-HT (a full agonist). Predicted hepatic apparent clearance (h-CLint,scaled) from human liver
microsomal stability assay using well-stirred approach. CYP2D6 inhibition was obtained by measuring inhibition of dextromethorphan at 3 μM
d
substrate concentration. RRCK cells with low transporter activity were isolated from Madin−Darby canine kidney cells and were used to estimate
e
intrinsic absorptive permeability.17 ND: not determined.
HT4 agonists was poor brain penetration likely due to P-
glycoprotein (Pgp) mediated efflux.12
An additional concern is that prolonged exposure of GPCRs
to an agonist can result in receptor desensitization, thereby
limiting the agent’s therapeutic usefulness. Agonist-induced
desensitization is often dependent on the degree of intrinsic
activity; while weak partial agonists minimize the potential for
desensitization,12 they can in fact act as functional antagonists
when low receptor numbers are present. Therefore, it is critical
to identify compounds with appropriate intrinsic activities.
B
dx.doi.org/10.1021/jm300953p | J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
Article
Table 2. 5-HT4 Binding, Functional Activity, in Vitro Human Microsomal Clearance, and P-gp Liability of Analogues 2a−d
a
Binding Ki values (geometric mean; at least n = 3 measurements) were determined in HEK293 cells stably expressing the human 5-HT4d isoform
b
using [3H]GR113808 as the radioligand. Functional activity was assessed as the EC50 and Emax values for stimulation of cAMP production in
HEK293 cells stably expressing the human 5-HT4d receptor isoform. EC50 values (geometric mean; at least n = 3 measurements) represent the
concentration producing 50% of the maximal cAMP response. Emax values (arithmetic mean) represent the maximal effect on cAMP production
c
expressed as a percentage of a maximal response of 5-HT (a full agonist). Predicted hepatic apparent clearance (h-CLint,scaled) in HLM, CYP2D6
d
inhibition was obtained by measuring inhibition of dextromethorphan at 3 μM substrate concentration. RRCK cells with low transporter activity
e
were isolated from Madin−Darby canine kidney cells and were used to estimate intrinsic absorptive permeability.17 MDR efflux ratio (MDR Er)
from an MDR1-transfected MDCK cell line represents the ratio of permeability, Papp(BA)/Papp(AB).
On the basis of this information, our criteria for the selection
of a desirable CNS candidate were high binding affinity (<10
nM), excellent off-target selectivity (>100×), low intrinsic
activity (<50%), no significant barriers to CNS penetration, and
an acceptable safety (tolerability and cardiovascular) profile. As
part of a longstanding effort to discover 5HT4 agonists for the
treatment of gastrointestinal (GI) disorders, a diverse set of
chemical series possessing excellent oral drug properties had
been identified. Since 5-HT4 receptor activation in the GI is
peripherally mediated, optimization of physicochemical proper-
ties for CNS indications was not a key design element of the
previous program. To improve our chances of identifying
compounds with excellent CNS penetration, the known
chemical series were triaged to select cores with a reduced
number of hydrogen bond donors (≤1), low molecular weight
(<400), and weakly basic amine centers (pKa < 8.5).13 From
this analysis, multiple chemotypes emerged possessing the
above CNS criteria, as represented by benzisoxazole IV,
oxindole V, and benzimidazolidinone VI (Figure 1).14−16 We
postulated that the R1 and R2 positions in each chemical series
IV−VI would provide enabled and flexible handles to rapidly
modulate intrinsic activity and ADME properties. Herein we
disclose the SAR, pharmacological, pharmacodynamic, and
safety profiles of the lead clinical candidates derived from these
starting points.
were conducted in a HEK293 cell line expressing the 5-HT4d
isoform of the human 5-HT4 receptor. Analogues 1a−e with
cycloalkyl and branched alkyl R2 substituents showed high
binding and functional activity, with a range of Emax values from
50% to 95%. SAR at the R2 position of the benzisoxazole
indicated that the molecular shape and size of this group might
provide a means to modulate intrinsic activity. For example,
replacement of the cyclopentyl group (1b) with cyclobutyl (1a)
resulted in a reduction in Emax from 81% to 53%. This series of
compounds demonstrated high clearance from human liver
microsomes (HLM) and low passive permeability in the RRCK
cell line, likely due to the high lipophilicity (ClogP > 5) of
analogues 1a−e.17 To test whether replacing the piperidine
substituent (R1) was a viable strategy to introduce polarity, the
n-butyl side chain of compound 1d was modified to afford
hydroxyl-THP analogue 1f. Despite the overall reduction in
ClogP (5.5−3.5) and modified pKa, analogue 1f demonstrated
high intrinsic clearance in HLM and significant inhibition of
CYP2D6 at a single substrate concentration of 3 μM.
Despite the significant reduction in lipophilicity for
compound 1f relative to 1d, the similarly high microsomal
intrinsic clearance suggested that further reductions in ClogP
were required as a key design criteria. To this end, the isobutyl
side chain (R2) was replaced with more polar groups such as
analogues 2a−d (Table 2; see Chemistry section for synthesis).
The tertiary carbinol 2a exhibited excellent potency and
functional activity (Emax = 50%) with reduced clearance and
low CYP2D6 liability. In addition, the compound showed high
passive permeability and no predicted P-gp mediated efflux as
measured by MDR efflux ratio in MDCK-MDR1 cell lines.18
Further modification of the R2 group led to cyclopentyl
carbinol 2b and tetrahydropyran 2c which exhibited poor
RESULTS AND DISCUSSION
■
Our efforts to identify viable CNS development candidates
began by exploring alkyl and cycloalkyl ether substituents
(OR2) off the benzisoxazole ring while maintaining an n-butyl
substituent on the piperidine (examples 1a−e, Table 1; see
Chemistry section for synthesis). Initial in vitro binding studies
C
dx.doi.org/10.1021/jm300953p | J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
Article
metabolic stability. Interestingly, the degree of intrinsic activity
appeared to correlate with the size of the R2 group as
demonstrated by tetrahydropyran 2c which is now a full agonist
relative to 5-HT (Emax = 85%), in the 5-HT4d functional assay.
Reduction of the ring size from tetrahydropyran to
tetrahydrofuran provided (4-{4-[(R)-(tetrahydrofuran-3-yl)-
oxy]benzo[d]isoxazol-3-yloxymethyl}piperidin-1-ylmethyl)-
tetrahydropyran-4-ol (2d, PF-04995274), a potent (5-HT4d Ki
= 0.15 nM), partial agonist (Emax = 24%) with low clearance
and no predicted efflux from Pgp.16
To understand the brain penetration of analogue 2d, a single
dose was subcutaneously (sc) administered to rats and time
points over a range of 0−4 h were studied. Brain tissue, plasma,
and CSF were collected and analyzed for total drug exposure
levels using LC−MS/MS. Protein binding values were also
determined using equilibrium dialysis to calculate free drug
exposures in plasma and brain as previously described.19 As
shown in Table 3, compound 2d exhibited excellent brain
penetration, as measured by unbound brain to plasma ratio
(Cu,b/Cu,p) of 0.7 and CSF to unbound plasma ratio (CSF/Cu,p)
The key compounds from the oxindole and imidazolidione
series were analogue 6-fluoro-3,3-dimethyl-2-oxo-2,3-dihydroin-
dole-1-carboxylic acid [1-(4-hydroxytetrahydropyran-4-
ylmethyl)piperidin-4-ylmethyl]amide (3, PF-03382792) and
6-fluoro-N-({1-[(4-hydroxytetrahydro-2H-pyran-4-yl)methyl]-
piperidin-4-yl}methyl)-3-isopropyl-2-oxo-2,3-dihydro-1H-
benzo[d]imidazole-1-carboxamide (4, PF-01352968).14,15 Ana-
logues 3 and 4 were highly potent in binding and functional
assays with low predicted Pgp efflux (Figure 2). Assessment of
brain penetration for 3 (Cu,b/Cu,p = 1.8) and 4 (Cu,b/Cu,p = 1.6)
in rats, as shown in Table 3, confirmed excellent brain
penetration properties for both analogues.
To increase our understanding around the 5-HT4 isoform
and tissue selectivity, in vitro binding studies were conducted
using cells lines expressing the 4a, 4b, 4d, and 4e isoforms of
the human 5-HT4 receptor as well as brain membranes
prepared from rat, dog, and monkey striatal tissue (Table 4).
Table 4. In Vitro Binding Studies Performed Using
a
Transfected Cells and Native Tissue
of 0.5 for AUC0−4h
.
binding Ki (nM)
tissue
HEK293
receptor
human
2d
3
4
Table 3. Neuropharmacokinetic Parameters of Lead 5HT4
Partial Agonists in Rats
a
4a
4b
4d
4e
0.36
0.46
0.15
0.32
0.30
0.30
0.27
4.6
6.8
2.7
6.3
3.9
4.8
8.9
4.4
5.5
1.7
4.9
1.5
3.8
1.9
HEK293
human
human
human
rat
parameter
2d
3
4
HEK293
Cu,b (AUC0−4h, nM·h)
Cu,p (AUC0−4h, nM·h)
CSF (AUC0−4h, nM·h)
706
1054
519
0.7
2922
1669
1381
1.8
44.3
27.6
ND
1.6
CHO
striatal membranes
striatal membranes
striatal membranes
dog
C
u,b/Cu,p
monkey
a
CSF/Cu,b
CSF/Cu,p
1.4
2.1
ND
ND
Binding studies were conducted using human 5-HT4 receptor
0.5
0.8
isoforms stably expressed in cells lines or native receptors in brain
membranes prepared from striatal tissue. Ki values are expressed as the
geometric mean of three to six determinations. HEK is human
embryonic kidney. CHO is Chinese hamster ovary.
a
Ratios are derived from respective matrices by measuring brain,
plasma, and CSF concentrations over 0−4 h followed by calculating
AUC0−4h in Winnonlin (Pharsight, California) for 2−4 animals; Cb,u
=
total brain concentration *fu,b for each compound; Cp,u = total plasma
concentration*fu,p for each compound; ND: not determined. CSF:
cerebrospinal fluid.
Compound 2d displayed subnanomolar affinity for the various
human receptor isoforms (Ki of 0.15−0.46 nM) and showed
similar potencies in native tissue preparations (Ki of 0.27−0.30
nM). Analogues 3 and 4 were approximately 10-fold less potent
than 2d, with Ki values in the nanomolar range. The binding
potencies for all three compounds were generally consistent
across species.
The ability of compounds 2d, 3, and 4 to increase cAMP
production across multiple 5-HT4 isoforms and consistently
between human and rats was next examined (Table 5).
The excellent brain penetration, low partial agonism, and
high binding affinity for compound 2d led us to consider that
additional chemical matter containing the hydroxyl tetrahy-
dropyran R1 group might provide similar alignment of CNS
properties.13 To this end, piperidine−THP containing
analogues in the oxindole series (V) and imidazolidinone
series (VI) were selected for additional profiling (Figure 1).
Figure 2. Identification of higher intrinsic activity 5-HT4 agonists.
D
dx.doi.org/10.1021/jm300953p | J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
Article
Information Tables 1−3). Free brain drug concentrations
(Cb,u) were calculated by incorporation of free fractions in brain
to total brain exposures measured using LC−MS/MS. Plots of
Cb,u versus % RO were fitted to a simple Emax model (eq 1)
using NONMEM, via the naive-pooled approach.20
Table 5. Cyclic AMP Production in Cell Lines Expressing
Human and Rat 5-HT4 Receptor Isoforms
a
EC50, nM (Emax, %)
species
human
isoform
2d
3
4
4a
4b
4d
4e
4l
0.47 (33)
0.36 (6)
0.8 (53)
1.3 (40)
0.9 (35)
1.2 (21)
1.52 (46)
1.15 (63)
2.15 (37)
0.38 (89)
1.20 (61)
1.13 (88)
0.98 (58)
0.96 (105)
0.90 (106)
1.59 (104)
Emax × concentrationγ
% receptor occupancy = E0 +
γ
EC50 + concentrationγ
0.37 (15)
0.26 (7)
(1)
rat
0.65 (22)
0.59 (29)
0.62 (15)
The relationship between Cb,u vs RO is shown for
compounds 2d (Figure 3B), 3 (Figure 3C), and 4 (Figure
3D). The simple Emax model generally described the observed
data (Figure 3). By use of the model, free brain EC50 and Emax
values were determined for 2d (0.48 nM, 15%), 3 (19.7 nM,
53%), and 4 (10.9 nM, 86%). Relative to the 5-HT4d binding Ki
values (0.33 nM, 2.7 nM and 1.7 nM), these in vivo binding
values were respectively 1.45-, 7.3-, and 6.4-fold higher. We
have previously observed differences between in vitro and in
vivo derived Ki values for the high intrinsic activity of the 5-HT4
full agonist prucalopride (11-fold).12 Since the degree of
separation observed between in vitro and in vivo derived Ki
values appears to generally correlate with the 5-HT4 agonism,
this difference might be due to the degree of interaction
occurring with high- and low-affinity binding states, which have
been described for 5-HT4 receptors.12,21 The free brain % Emax
values determined using the model showed the same rank order
of effect that was observed for Emax values determined in vitro
for 5-HT4d receptors (2d, 24%; 3, 35%; 4, 86%).
The isolated rat esophagus has been used previously to test
the functional response of 5-HT4 receptor ligands at native
receptors in intact tissue.22 5-HT4 receptors are also located on
enteric neurons in the gastrointestinal tract, and agonist
activation of these receptors produces an increase in gastro-
intestinal transit by modulating contraction and relaxation.23,24
Functional activities of compounds 2d, 3, and 4, were
compared to the high intrinsic activity agonist prucalopride
(III) and to the endogenous full agonist 5-HT for their ability
4s
4e
a
In vitro studies comparing the compounds’ abilities to stimulate
cAMP production were conducted in HEK293 cells transiently
expressing human and rat receptor isoforms. EC50 and Emax values
are expressed as geometric and arithmetic means, respectively. Mean
values are based on three to four determinations. EC50, 50% of the
maximal compound response, values are reported in nM. Emax
maximal effect on cAMP production, is expressed as a percentage of
the response observed with 1 μM 5-HT, a full agonist.
,
Compound 2d displayed high functional potencies with EC50
values ranging from 0.26 to 0.65 nM across the human and rat
receptor isoforms, closely matching its binding affinity. For
compounds 3 and 4, EC50 values ranged from 0.38 to 2.15 nM
and were 2- to 10-fold lower than their respective binding Ki
values. For each compound, Emax values showed some variation
across the receptor isoforms tested. With respect to intrinsic
activity levels across compounds, the rank order for Emax values
was 2d < 3 < 4 for each receptor isoform.
In vivo RO was determined as previously described using
[3H]-(1-methylpiperidin-4-yl)methyl-8-amino-7-chloro-2,3-di-
hydro-1,4-benzodioxine-5-carboxylate-([3H]SB207145).12 One
hour following subcutaneous administration, 2d, 3, and 4 were
all shown to have penetrated the brain and bind to striatal 5-
HT4 receptors in a dose-dependent manner (Figure 3 A).
Whole brain tissue was collected from the animals tested for in
vivo RO and drug exposures measured (see Supporting
Figure 3. Compound 2d, compound 3, and compound 4 (A) displaced radioligand binding to rat striatal membranes. Data shown are the mean
SEM for RO and the mean SD for exposure, combined from two to three experiments; n = 4−8 animals. Free brain exposure measured from
individual animals (Cb,u) vs RO data (blue circles) were fitted to an Emax model, with resulting predicted values. Open circles represent observed data,
whereas the blue solid line represents the simulated trials for compound 2d (B), compound 3 (C), and compound 4 (D).
E
dx.doi.org/10.1021/jm300953p | J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
Article
to relax carbamylcholine contracted rat esophagus (Figure
Previous studies have shown that activation of 5-HT4
receptors increases brain ACh levels in rats.7,8 Microdialysis
studies were conducted in conscious rats to examine the effects
of 2d, 3, or 4 on brain ACh release in the prefrontal cortex.
Figure 4B shows the overall ACh response to compound
treatment as a fraction of baseline, expressed as the increase in
the area under the effect curve (AUC) for the time period 3 h
post-treatment. Compound 2d (1 mg/kg), 3 (5 mg/kg), and 4
(5 mg/kg) all produced moderate increases in cortical ACh that
were similar in magnitude to that observed with a single dose of
prucalopride (10 mg/kg, Figure 4B). On the basis of our
previous results, this dose of prucalopride would be expected to
produce a maximal ACh response and occupy approximately
50% of the 5-HT4 receptors at 1 h postdose.12 The % RO
required by GPCR agonist ligands for efficacy is generally
dependent on the degree of intrinsic activity, with full agonists
demonstrating in vivo activity at lower levels of RO than that
required by partial agonists and antagonists.26 Previously we
have shown this to be the case for the 5-HT4 partial agonists
prucalopride (Emax = 77%) and 4-hydroxy-7-isopropyl-6-oxo-N-
(3-piperidin-1-ylpropyl)-6,7-dihydrothieno[2,3-b]pyridine-5-
carboxamide PRX-03140 (Emax = 33%) in microdialysis studies,
where maximal ACh increases were observed at <40% RO and
>85% RO, respectively.12 This learning was extrapolated to
ACh release studies for 2d, 3, and 4 tested at 1 or 5 mg/kg. By
application of the Emax model described above and neuro-PK
studies described in the previous section, free brain
concentrations and thereby in vivo RO in rats were simulated
for all three compounds at the doses tested for the ACh release.
It was confirmed that at 1 and/or 5 mg/kg the % RO achieved
for all three compounds was greater than 90% at 0−4 h,
postdose (see Supporting Information Table 1−3 for simulated
output of these doses in rats). At the same or similar receptor
occupancies, the magnitude of ACh release was dependent on
intrinsic activity of the compounds tested, as demonstrated by
microdialysis data (Figure 4B).
5-HT4 agonists have been reported to reverse behavioral
deficits caused by treatment with anticholinergic agents.10,27
The Morris water maze (MWM) is a classical behavioral testing
paradigm that examines effects on spatial learning.28 The
cholinergic system appears to play an integral role in spatial
learning as treatment with anticholinergic agents (e.g.,
scopolamine) can cause significant learning impairments in
this task. Compound 2d was tested for its ability to reverse
scopolamine-induced learning deficits in the rat MWM (Figure
4C). Prior to compound testing, the ability to reverse
scopolamine-induced learning deficits in this assay was verified
using donepezil, a cholinesterase inhibitor, as a positive control.
Results indicate that scopolamine caused a significant learning
impairment during acquisition testing on days 1−4. Pretreat-
ment with 2d (0.032 mg/kg) significantly reversed the
scopolamine deficits on acquisition days 3 and 4. This dose
of 2d achieved >90% RO in rats as shown above and in the
Supporting Information.
4A).25 In comparison to 5-HT (10 μM), 2d (10 μM) and 3 (10
Figure 4. (A) 5-HT, compound 2d, compound 3, compound 4, and
prucalopride (PRU, III) relaxed carbamylcholine-contracted rat
esophagus in a manner that was consistent with the compounds’
intrinsic activity level. Data are the mean
represent the percent relaxation compared to vehicle. (B) Compounds
SEM (n = 4−17) and
2d, 3, 4, and prucalopride increased ACh release in the rat prefrontal
cortex. Data are expressed as the mean
SEM of the AUC as a
fraction of baseline (five to seven animals per group): unpaired t tests,
(∗) P < 0.05 vs PF-compound vehicle, (†) P < 0.05 vs PRU vehicle.
(C) Compound 2d reversed scopolamine-induced learning deficits in
the rat Morris water maze. The distance traveled to find the platform
during acquisition testing was used as the test parameter. Data were
analyzed by two-way ANOVA (days × treatment) followed by
Dunnett’s multiple comparison test: (∗) P < 0.05 or (∗∗) P < 0.01
versus scopolamine, n = 12 animals per group. PRU is prucalopride.
Scop is scopolamine.
μM) produced incomplete relaxations that were 21% and 22%
of the response observed with 5-HT, respectively, and were
indicative of partial agonism (Figure 4A). The relaxation
observed in response to 4 (10 μM) was reflective of its higher
intrinsic acticvity in that it produced a relaxation that was
approximately 70% of that observed with 5-HT (Figure 4A).
Intrinsic activity levels for prucalopride reportedly range from
52% to 77% in cell lines expressing the 4l, 4s, and 4e isoforms
of the rat 5-HT4 receptor.12 Under the test conditions used in
this study, prucalopride (10 μM) produced a relaxation that
was equivalent to the full agonist 5-HT (Figure 4A).
To predict human pharmacokinetic properties and clinical
doses, studies were conducted to project the metabolic
clearance, oral bioavailability, and plasma elimination half-life
in humans. Initial in vivo clearance studies in rat and dog
suggested that hepatic clearance (Clb) mediated by P450 was
the primary route of metabolism (Table 6). Therefore, blood
clearance in humans was projected by application of the well-
stirred model and scaling NADPH-dependent intrinsic
clearances observed in HLMs, via the substrate depletion
F
dx.doi.org/10.1021/jm300953p | J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
Article
regimen to maintain steady state plasma concentration at the 5-
HT4d in vitro binding Ki.
Table 6. Summary of Preclinical Pharmacokinetics and
Predicted Human PK for Lead 5HT4 Partial Agonists
a
With the primary design objectives for a CNS drug candidate
satisfied, the nonclinical safety profiles of 2d, 3, and 4 were
assessed in repeat dose toxicology studies up to 1 month in
duration in rats and dogs, genetic toxicology assays, and safety
pharmacology studies (see Supporting Information Table 4).
The safety profile of these compounds supported continued
development and identified the following toxicology as
neurological (convulsions), cardiovascular (increase in systolic
blood pressure), and gastrointestinal (emesis, decreased food
intake and body weight, and/or mucosal injury in small and
large intestine). The neurological and gastrointestinal findings
occurred at high multiples of the exposures predicted to be
efficacious (>900×). The non-dose-related increase in systolic
blood pressure in dogs was limited to the first dose and was not
observed after subsequent doses of 3 and 4 in the same animal.
These findings are also consistent with broad selectivity panel
data for 2d, 3, and 4 including excellent selectivity against other
5-HT receptor families (see Supporting Information Table 5).
parameter
2d
3
4
predicted Clb from HHEP
(HLM) (mL min−1 kg−1
11 (8.8)
ND (10.0)
10.4 (7.9)
)
predicted F (%)
55.8
6.0
50.0
4.3
12
60.5
6.5
10
Vdss (L/kg)
predicted plasma elimination ∼10
t1/2 (h)
rat (dog) iv Clb
(mL min−1 kg−1
81.8 (21.2)
104.5 (12.4)
12.6 (40)
)
rat (dog) F (%)
40 (47)
100 (51)
40 (100)
RLM (DLM) in vitro Clb
(mL min−1 kg−1
48.9 (27.9)
42.7 (35.4)
45.2 (<5.6)
)
a
Hepatic blood clearance from HLM was scaled using a well-stirred
method by incorporation of free fraction in plasma (fup), free fraction
in microsomes (fu,mics), and blood to plasma partitioning (Kb/p) to
intrinsic clearance (Clint, not shown above). Blood clearance from
hepatocytes was scaled using a well-stirred model but without
incorporation of binding factors, whereas in vivo clearance (Clb) was
scaled by incorporation of Kb/p into total plasma clearance during iv
PK single dose studies.29
CHEMISTRY
■
method.19 The predicted Clb values from HLM for 2d, 3, and 4
were 8.83, 10.0, and 7.9 mL min−1 kg−1, respectively. The oral
bioavailability was calculated to be 55.8%, 50.0%, and 60.5%,
respectively, with the assumption that hepatic elimination was
the primary clearance mechanism. The projected elimination
half-lives for 2d, 3, and 4 were ∼10−12 h using a moderate
volume of distribution (VDSS) of ∼4.3−6 L/kg from single
species allometry and in silico methods. Using the Css equation,
dose = (Ceff*τ*Clp)/F, where total plasma Ceff was predicted to
correspond to 50−70% RO in humans, the compounds were
predicted to have low daily doses of <30 mg using a q.d. dosing
The syntheses of analogues 1a−f and 2a−d (Tables 1 and 2)
are illustrated in Scheme 1. Mitsunobu couplings between
compound 5 and the appropriate alcohol afforded intermedi-
ates 6a−g.30 Methanolysis provided esters 7a−g, which were
converted to their hydroxamic acid derivatives 8a−f by
treatment with hydroxylamine sulfate and potassium carbonate.
Substituted hydroxylbenzisoxazoles 9 were prepared by
cyclization of hydroxamic acids using carbonyldiimidazole
(CDI). Mitsunobu coupling of 9a−e with (1-butylpiperidin-4-
yl)methanol afforded the n-butyl-substituted piperidine ana-
logues 1a−e. Alternatively, treatment of 9d (R = isobutyl) or
a
Scheme 1
a
Reagents and conditions: (a) R2OH, diethyl azodicarboxylate, PPh3, THF; (b) K2CO3, MeOH; (c) (NH2OH)2−H2SO4 (5 equiv), K2CO3, water,
MeOH; (d) CDI, THF, reflux; (e) 4-hydroxymethyl-N-BOC-piperidine or (1-butylpiperidin-4-yl)methanol, diethyl azodicarboxylate, PPh3, THF;
(f) HCl, ether; (g) 1,6-dioxa-spiro[2.5]octane, diisopropylethylamine, EtOH, reflux; (h) thioanisole, o-cresol, TFA; (i) NaOH, 2,2-dimethyloxirane,
EtOH, 50 °C; (j) BOC2O, TEA, THF; (k) HCl (conc), MeOH.
G
dx.doi.org/10.1021/jm300953p | J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
Article
a
Scheme 2
a
Reagents and conditions: (a) 4-nitrophenyl chloroformate, triethylamine, toluene, 0 °C, 10 min; (b) compound 21, triethylamine, 2-Me-
tetrahydrofuran, 5 h, rt; (c) MeOH, 55 °C, 5 h; (d) tetrahydrofuran, sponge Ni, H2, 30 psi, 12 h.
9g (R = (R)-tetrahydrofuran-3-yl) with 4-hydroxymethyl-N-
BOC-piperidine yielded piperidines 10d and 10g. Removal of
the BOC protecting group and epoxide opening with 1,6-dioxa-
spiro[2.5]octane in refluxing ethanol afforded analogues 1f and
2d. An alternative synthetic sequence was developed allowing
for the late stage incorporation of various R2 groups using
compound 10f, where R2 was now an appropriate protecting
group such as p-methoxybenzyl (PMB). Deprotection of 10f
with thioanisole provided compound 11. Compound 2a was
prepared from 11 by installation of the R2 ether following BOC
protection and epoxide opening on the piperidine with
dimethyloxirane in the presence of NaOH. By use of methods
described above, compound 11 was converted into piperidine
12. Mitsunobu coupling then afforded the cyclopentyl analogue
2b. By use of analogous procedures, tetrahydropyranyl
analogue 2c was prepared.
Compounds 3 and 4 were readily prepared via the route
outlined in Scheme 2. Activation of oxindole 13 or
imidazolidinone 15 with 4-nitrophenyl chloroformate afforded
intermediates 14 and 16.31,32 Treatment of the carbamate
intermediates with amine 21 provided final analogues 3 and 4.
The synthesis of amine 21 begins with spiro-epoxide 18 which
undergoes ring opening with piperidine 19 followed by
reduction of the nitrile group using hydrogenolysis.14
been reported to reverse learning deficits induced by
cholinergic agents. In this report we have demonstrated that
the partial agonist 2d reverses scopolamine-induced learning
deficits in the rat MWM, thus providing further evidence that
the cognitive effects of these compounds are at least partly
mediated via the cholinergic system. A safety assessment of
these compounds revealed large therapeutic margins between
efficacious concentrations and safety findings. On the basis of
the excellent balance of pharmacokinetic attributes and safety,
the two lead 5-HT4 partial agonist candidates 2d and 3 were
selected for further clinical development.
EXPERIMENTAL SECTION
■
Biology. Animal Care. All animal studies were conducted in
accordance with animal use protocols approved by Pfizer’s local
institutional animal care and use committee and in accordance with
the guidelines for the care and use of laboratory animals.33 Male
Sprague−Dawley rats, used for all studies unless otherwise noted, were
obtained from Charles River Laboratories (Raleigh, NC). Animals
were housed on a 12 h light/dark cycle with free access to food and
water and were allowed to acclimate for at least 5 days after arrival. For
radioligand administration during in vivo RO studies, rats were
purchased with preimplanted jugular vein catheters.
In Vitro Binding to Native 5-HT4 Receptors in Brain
Membranes. Binding assays on brain membranes prepared from
rat, dog, and monkey striatal tissue were performed as previously
described.12 Briefly, striatal tissue was homogenized in 50 mM Tris
buffer (pH 7.4 at 4 °C) and then centrifuged at 40000g for 10 min.
The resulting pellet was resuspended in Tris buffer, incubated at 37 °C
for 10 min to eliminate endogenous serotonin, then centrifuged again
at 40000g for 10 min. The final membrane pellet was resuspended in
50 mM HEPES assay buffer (pH 7.4) containing 2 mM MgCl2.
Binding assays were conducted in 96-well plates in a total assay volume
of 250 μL containing 225−250 μg of membrane protein and 0.5 nM
tritiated 1-methyl-1H-indole-3-carboxylic acid, (1-{2-
[(methylsulfonyl)amino]ethyl}piperidin-4-yl)methyl ester ([3H]-
GR113808), a 5-HT4 receptor antagonist. Nonspecific binding was
determined in the presence of 10 μM (1-{2-[(methylsulfonyl)amino]-
ethyl}piperidin-4-yl)methyl 5-fluoro-2-methoxy-1H-indole-3-carboxy-
late (GR125487), a 5-HT4 receptor antagonist. Test compounds
CONCLUSION
■
We have identified three highly potent, selective, and brain
penetrant 5HT4 partial agonists that possess a wide range of
intrinsic activities with ADME properties suitable for further
development. In vivo RO studies demonstrated a clear
relationship between the binding affinity, free brain exposure,
and RO levels. All three compounds increased ACh release in
the rat prefrontal cortex, a finding that is consistent with
previous reports. Since these studies were conducted under
conditions of high RO (>90%), additional testing is required to
further examine the relationships between intrinsic activity,
receptor occupancy, and ACh release. 5-HT4 agonists have
H
dx.doi.org/10.1021/jm300953p | J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
Article
were prepared as 10× stocks in 50 mM Tris buffer (pH 7.6)
containing 10% dimethyl sulfoxide and assayed in duplicate at each test
concentration. After a 30 min incubation period at 37 °C, the samples
were filtered over Whatman GF/B filters and rinsed with ice-cold 50
mM Tris buffer (pH 7.4). Membrane-bound [3H]GR113808 content
was determined by liquid scintillation counting. IC50 values
(concentration at which 50% inhibition of specific binding occurs)
were calculated by linear regression of the concentration response
data. Ki values were calculated according to the Cheng−Prusoff
equation, Ki = IC50/(1 + (L/Kd)), where L is the radioligand
concentration and the Kd is the dissociation constant for the
radioligand (determined previously by saturation binding analysis).
In Vitro Binding in Cell Lines Stably Expressing 5HT4
Receptors. Membranes from CHO or HEK293 cells transfected
and stably expressing the human 4a, 4b, 4d, or 4e receptor isoforms
were prepared from frozen cell paste by homogenization in 50 mM
Tris buffer (pH 7.4, 4 °C) containing 2 mM MgCl2, followed by
centrifugation at 40000g for 10 min. The resulting pellet was
resuspended in 50 mM HEPES assay buffer (pH 7.4) containing 2
mM MgCl2. Appropriate tissue levels for binding studies were based
on the Bmax and Kd values for each isoform which were established by
saturation binding analysis. Binding studies were conducted in 96-well
plates in a total volume of 250 μL containing 2−25 μg of membrane
protein. [3H]GR113808 was added to a final concentration of 0.1 nM
for membranes expressing the 4a, 4b, and 4e isoforms or 0.2 nM for
the 4d membranes. Nonspecific binding was determined in the
presence of 10 μM GR125487. Compound preparation, assay
procedures, and data analysis were performed as described for binding
to native receptors in brain membranes.
Cyclic AMP Production in Cell Lines Expressing 5-HT4
Receptors. The ability to increase cAMP production in cell lines
expressing human and rat 5-HT4 isoforms was used to establish in
vitro functional potency and intrinsic activity levels. Increases in cAMP
were determined according to methods previously described.12 For
general SAR, screening was conducted using HEK293 cells that were
stably transfected with the human 5-HT4d receptor isoform. For
functional comparisons at the human 4a, 4b, 4d, or 4e as well as the rat
4s, 4l, or 4e isoforms (DNA 2.0, Menlo Park, CA), HEK293 cells were
transduced to transiently express receptors using a BacMam expression
system (Invitrogen, Carlsbad, CA). For transient expression, cells were
incubated at 37 °C for 24 h post-transduction, then used immediately
or stored frozen for later use. Test compounds were prepared in 100%
DMSO and diluted to 2× their final concentration in phosphate-
buffered saline containing 1 mM isobutylmethylxanthine and 10 μM
HEPES buffer. Assays were performed in 384-well plates, and each
compound was tested in triplicate or quadruplicate. Test compound (5
μL) was added to each well, and the assay was initiated by the addition
of 5 μL of cell suspension containing either 8000 or 4000 cells for the
human or rat isoforms, respectively. Following a 30 min incubation at
37 °C, the assay was terminated and cAMP levels were determined
using a Dynamic 2 homogeneous time-resolved fluoresence cAMP
assay kit from Cisbio (Medford, MA). The maximal effects on cAMP
production (Emax) for each compound were expressed as the
percentage of the response produced by 1 μM 5-HT, a full agonist.
using a LS6500 scintillation counter (Beckman Coulter). The rest of
the brain was frozen on dry ice and reserved for compound exposure
analysis. Curve fit analyses were performed using the equation “one-
site binding (hyperbola)” from GraphPad Prism 5 software (GraphPad
Software Inc., Santa Clara, CA).
Rat Esophagus Assay. Male Sprague−Dawley rats were
euthanized by exsanguination under isoflurane anesthesia, and the
esophagus in the area of the thoracic aorta was quickly removed and
cleared of any extra tissue. The esophagus was cut through the middle,
laid flat, and then cut into two strips. Each strip was then suspended in
4 mL of tissue baths containing Krebs−Henseleit physiological salt
solution of the following composition (mM): NaCl, 118.4; KCl, 4.7;
CaCl2, 2.5; MgSO4·H2O, 1.2; KH2PO4, 1.2; NaHCO3, 25; dextrose,
10. The solution was maintained at 37 °C and continuously aerated
with 95% O2 and 5% CO2. The force of isometric contraction was
measured using Grass model FT03C force transducers connected to
AD Instruments bridge amplifiers. Data were collected using the AD
Instruments Chart data acquisition and analysis system (Unit 13, 22
Lexington Drive, Bella Vista, New South Wales 2153, Australia). After
an equilibration period of 90 min at a resting tension of 1 g, tissues
were exposed to 1 μM carbamylcholine to cause contraction. After 7
mins or when the tissue was no longer increasing the force of
contraction, solvent, 5-HT, and 2d, 3, or 4 were added to a final
concentration of 10 μM and the tissues were allowed to relax for 30
min. The greatest point of relaxation within the first 5 min after adding
the compound was used in calculations.
Acetylcholine (ACh) Determination in Rats. Microdialysis was
performed in conscious rats, and ACh levels were determined by
online HPLC as previously described.12 Briefly, microdialysis guide
cannulas were implanted into the prefrontal cortex of male Sprague−
Dawley rats (300−350 g), and on the day after implantation
microdialysis probes were inserted and perfused with artificial
cerebrospinal fluid overnight. On the day of testing, 100 nM
neostigmine was added to the perfusion fluid. Microdialysis samples
were collected at 15 min intervals and analyzed for ACh content by
online HPLC using a modification of the Bioanalytical Systems (BAS;
West Lafayette, IN) acetylcholine−choline assay kit (BAS MF-8910).
After baseline stabilization, animals were given compounds subcuta-
neously at the indicated dose and ACh levels were monitored for at
least 3 h postdose. The overall ACh response was quantified as the
increase in the area under the curve (AUC) for the 0−3 h period
postdose. Statistical analyses were performed using GraphPad Prism 5
software (GraphPad Software Inc., Santa Clara, CA). Chromatography
data were collected and analyzed using EZChrom Elite software
(Agilent Technologies, Inc., Santa Clara, CA).
Morris Water Maze Testing. Compound 2d was dissolved in 20%
hydroxylpropyl-β-cyclodextrin (HPCD), and scopolamine was dis-
solved in saline. At 60 min prior to place acquisition testing on days
1−4 in the water maze, rats received HPCD vehicle or 2d (0.032 mg/
kg, sc). At 30 min prior to testing, rats received a second injection with
scopolamine (0.32 mg/kg, sc). The total pretreatment time for 2d was
60 min. All doses were corrected for salt content.
Prescreening Procedure. Male Wistar rats, 160−180 g at arrival,
were acclimated for 5−7 days prior to testing and given free access to
food and water. Sixty rats (12 per group) were selected prior to
compound testing based on the following prescreening procedure: A
circular tank was filled with water (24 1 °C) to a height of 30−35
cm. A 12 cm2 visible escape platform, which was exposed 1 cm above
the surface of the water, was placed in the center of the pool. Each rat
was placed in the water facing the pool wall at one of four release
points (north, south, east and west quadrants) and allowed to swim a
maximum of 60 s in order to find the visible platform. When
successful, the rat was allowed a 30 s rest period on the platform. If
unsuccessful within the allotted time period, the rat was given a score
of 60 s and then hand-guided to the platform and also allowed the 30 s
rest period. A total of four trials were conducted for each rat, one from
each release point. Rats were selected for compound testing in the
MWM based on the criteria that animals found the visible platform
twice out of four trials within the 60 s cutoff time.
EC50 values represent the concentration that yields 50% of the Emax
.
Emax and EC50 values were calculated using nonlinear least-squares
curve fitting of the concentration response data using a four-parameter
sigmoidal fit.
In Vivo Receptor Binding. In vivo binding experiments were
performed as described.12 Briefly, rats (∼200 g at use, n = 4/group)
were subcutaneously administered 2d (0.0032−1 mg/kg), 3 (0.032−
3.2 mg/kg), or 4 (0.1−3 mg/kg). Forty five minutes later, rats were
administered a 1 mL/kg intravenous (iv) injection of [3H]-(1-
methylpiperidin-4-yl)methyl-8-amino-7-chloro-2,3-dihydro-1,4-benzo-
dioxine-5-carboxylate (100 μCi/mL, 20−25 μCi/rat) via a jugular vein
cannula. After a further 15 min, rats were euthanized by live
decapitation and brains removed. Trunk blood was reserved to obtain
plasma for compound exposure analysis. Striatum (from all subjects)
and cerebellum (from vehicle control group only, to define nonspecific
binding) were rapidly dissected, homogenized, filtered, and counted
I
dx.doi.org/10.1021/jm300953p | J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
Article
Place Acquisition Testing. A circular tank was filled with water to a
level of 1 cm above a platform, which was located in one of four
quadrants, midway between the center of the pool and the outer rim.
White nontoxic paint was added to the water to obscure the platform.
Animals were gently placed in the tank facing the pool wall and
released from starting points within one of four quadrants. The
animals were allowed to swim in the pool for a maximum of 90 s per
trial. When the rat reached the platform, it was allowed to remain on it
for 30 s. If an animal did not find the hidden platform within the 90 s,
then the experimenter gently guided it there by hand. The animals
were allowed to remain on the platform for 30 s before being returned
to the home cage, or a holding cage, for a brief period (1−2 min)
before the next trial. Animals received a total of four trials per day for 4
consecutive days. The location of the platform remained the same for
all tests. The distance traversed to reach the hidden platform was
recorded using a video monitoring system. For the hidden platform
test, distances traveled across the four trials for each rat were averaged
for each day. These results were then analyzed across the 4 days of
testing. Two-way (days × treatment) repeated measures ANOVAs
followed by Dunnett’s multiple comparisons were used to compare
performance between groups for the acquisition trials
Measurement of Fraction Unbound in Brain. The unbound
fraction of each compound was determined in brain tissue homogenate
using a 96-well equilibrium dialysis method as previously described
with the following exceptions.19 Brain homogenates were prepared
from freshly harvested rat brains following dilution with a 4-fold
volume of phosphate buffer and spiked with 1 μM compound. The
homogenates were dialyzed against an equal volume (150 μL) of
phosphate buffer at 37 °C for 6 h. Following the incubation, equal
volumes (50 μL) of brain homogenate and buffer samples were
collected and mixed with 50 μL of buffer or control homogenate,
respectively, for preparation of mixed matrix samples. All samples were
then precipitated with internal standard in acetonitrile (200 μL),
vortexed, and centrifuged. Supernatants were analyzed using an LC−
MS/MS assay. A dilution factor of 5 was applied to the calculation of
brain fraction unbound.
Chemistry. General Methods. Solvents and reagents were of
reagent grade and were used as supplied by the manufacturer. All
reactions were run under a N2 atmosphere. Organic extracts were
routinely dried over anhydrous Na2SO4. Concentration refers to rotary
evaporation under reduced pressure. Chromatography refers to flash
chromatography using Merck silica gel 60 (230−400 mesh) columns
or Redisep Rf disposable columns on a CombiFlash Companion
system. All target compounds were analyzed using ultrahigh
performance liquid chromatography/UV/evaporative light scattering
detection coupled to time-of-flight mass spectrometry (UHPLC/UV/
ELSD/TOFMS). UHPLC analysis was performed on a Waters
ACQUITY UHPLC system (Waters, Milford, MA), which was
equipped with a binary solvent delivery manager, column manager,
and sample manager coupled to ELSD and UV detectors (Waters,
Milford, MA). Detection was performed on a Waters LCT premier XE
mass spectrometer (Waters, Milford, MA). The instrument was fitted
with an Acquity BEH (bridged ethane hybrid) C18 column (30 mm ×
2.1 mm, 1.7 μm particle size, Waters, Milford, MA) operated at 60 °C.
Unless otherwise noted, all tested compounds were found to be ≥95%
pure by this method. NMR spectra were recorded as designated on
either a 270 MHz JEOL JNM-LA 270 spectrometer or a 400 MHz
Varian instrument. Chemical shifts (δ) are quoted in ppm relative to
tetramethylsilane (TMS); all coupling constants (J) are given in Hz.
High resolution electrospray ionization mass spectrometry was carried
out using a Bruker (USA) Daltronics BioApex II with a 7 T
superconducting magnet and an analytical ESI source.
g, 0.500 mol) in 600 mL of anhydrous THF. The resulting mixture was
stirred at room temperature for 18 h. The solvent was removed under
reduced pressure and the crude material was purified on a silica gel
flash column, eluting with petroleum ether/ethyl acetate (15:1 → 3:1).
An amount of 86.0 g (65% yield) of product was isolated as a colorless
oil. H NMR (400 MHz, CDCl3) δ 1.67 (s, 6H), 2.30 (m, 2H), 4.2
(m,4H), 4.97 (m, 1H), 6.49 (d, J = 8.4, 1H), 6.51 (d, J = 8.4, 1H), 7.39
(t, J = 8.4, 1H).
Methanolysis, Representative Procedure for Step b, Scheme
1. Methyl 2-Hydroxy-6-[(R)-(tetrahydrofuran-3-yl)oxy]-
benzoate (7g). Potassium carbonate (135 g, 0.980 mol) was added
to a solution of 2,2-dimethyl-5-[(R)-(tetrahydrofuran-3-yl)oxy]benzo-
[1,3]dioxin-4-one (86.0 g, 0.330 mol) in 1000 mL of methanol. The
mixture was stirred at room temperature for 2 h, then concentrated in
vacuo. The residue was dissolved in ethyl acetate and washed with
aqueous ammonium chloride solution. The organic layer was dried
(Na2SO4) and concentrated to afford 72.0 g of the product as a yellow
solid (92% yield). 1H NMR (400 MHz, CDCl3) δ 2.20 (m, 2H), 3.99
(s, 3H), 4.80 (m, 4H), 4.94 (m, 1H), 6.31 (dd, J = 8.4, 0.8, 1H), 6.59
(dd, J = 8.4, 0.8, 1H), 7.30 (t, J = 8.4, 1H).
1
Methyl 2-(Cyclobutyloxy)-6-hydroxybenzoate (7a). 7a was
prepared in a similar manner, telescoping the two steps described
above utilizing cyclobutanol as a replacement for (S)-tetrahydrofuran-
3-ol. Compound 6a was carried on without purification or character-
1
ization. Quantitative yield. H NMR (270 MHz, CDCl3) δ 1.70−1.90
(m, 2H), 2.17−2.25 (m, 2H), 2.41−2.51 (m, 2H), 4.63−4.68 (m, 1H),
6.24 (d, J = 8.4, 1H), 6.70 (d, J = 8.4, 1H), 7.26 (dd, J = 8.4, 8.4, 1H),
11.45 (s, 1H).
Methyl 2-(cyclopentyloxy)-6-hydroxybenzoate (7b). 7b was
prepared in a similar manner, telescoping the two steps described
above utilizing cyclopentanol as a replacement for (S)-tetrahydrofuran-
3-ol. Compound 6b was carried on without purification or
1
characterization. Quantitative yield. H NMR (270 MHz, CDCl3) δ
1.66−2.10 (m, 8H), 3.91 (s, 3H), 4.76−4.84 (m, 1H), 6.41 (d, J = 8.6,
1H), 6.54 (d, J = 8.2, 1H), 7.28 (dd, J = 8.6, 8.2, 1H).
Methyl 2-(Cyclobutylmethoxy)-6-hydroxybenzoate (7c). 7c
was prepared in a similar manner, telescoping the two steps described
above utilizing cyclobutylmethanol as a replacement for (S)-
tetrahydrofuran-3-ol. Compound 6c was carried on without
purification or characterization. Quantitative yield. 1H NMR (270
MHz, CDCl3) δ 1.86−2.11 (m, 6H), 2.72−2.82 (m, 1H), 3.93 (d, J =
5.6, 2H), 3.95 (s, 3H), 6.39 (d, J = 8.4, 1H), 6.55 (d, J = 8.4, 1H), 7.30
(dd, J = 8.4, 8.4, 1H), 11.46 (s, 1H).
Methyl 2-Hydroxy-6-(2-methylpropoxy)benzoate (7d). 7d
was prepared in a similar manner, telescoping the two steps described
above utilizing 2-methylpropan-1-ol as a replacement for (S)-
tetrahydrofuran-3-ol. Compound 6d was carried on without
purification or characterization. 82% yield. 1H NMR (270 MHz,
CDCl3) δ 1.06 (d, J = 6.8, 6H), 2.07−2.17 (m, 1H), 3.75 (d, J = 6.3,
1H), 3.94 (s, 3H), 6.38 (d, J = 8.2, 1H), 6.56 (d, J = 8.4, 1H), 7.30 (dd,
J = 8.4, 8.2, 1H).
Methyl 2-(2,2-Dimethylpropoxy)-6-hydroxybenzoate (7e).
7e was prepared in a similar manner, telescoping the two steps
described above utilizing 2,2-dimethylpropan-1-ol as a replacement for
(S)-tetrahydrofuran-3-ol. Compound 6e was carried on without
purification or characterization. 99% yield. 1H NMR (270 MHz,
CDCl3) δ 1.06 (s, 9H), 3.62 (s, 2H), 3.95 (s, 3H), 6.36 (d, J = 8.4,
1H), 6.56 (d, J = 8.2, 1H), 7.30 (dd, J = 8.4, 8.2, 1H).
Methyl 2-Hydroxy-6-[(4-methoxybenzyl)oxy]benzoate (7f).
7f was prepared in a similar manner, telescoping the two steps
described above utilizing (4-methoxyphenyl)methanol as a replace-
ment for (S)-tetrahydrofuran-3-ol. Compound 6f was carried on
without purification or characterization. 79% yield. 1H NMR (270
MHz, CDCl3) δ 3.80 (s, 3H), 3.87 (s, 3H), 5.05 (s, 2H), 6.49 (d, J =
8.2, 1H), 6.60 (d, J = 8.2, 1H), 6.91 (d, J = 8.5, 2H), 7.32 (dd, J = 8.2,
8.2, 1H), 8.51 (d, J = 7.4, 2H), 11.48 (s, 1H).
(4-{4-[(R)-(Tetrahydrofuran-3-yl)oxy]benzo[d]isoxazol-3-
yloxymethyl}piperidin-1-ylmethyl)tetrahydropyran-4-ol (2d).
Mitsunobu Coupling, Representative Procedure for Step a,
Scheme 1. 2,2-Dimethyl-5-[(R)-(tetrahydrofuran-3-yl)oxy]-
benzo[1,3]dioxin-4-one (6g). Diethyl azodicarboxylate (130 g,
0.750 mol) was added in a dropwise fashion to a mixture of 5-hydroxy-
2,2-dimethylbenzo[1,3]dioxin-4-one30 5 (100 g, 0.510 mol), triphe-
nylphosphine (196.5 g, 0.750 mol), and (S)-tetrahydrofuran-3-ol (44.0
Hydroxamic Acid Formation, Representative Procedure for
Step c, Scheme 1. 2,N-Dihydroxy-6-[(R)-(tetrahydrofuran-3-
yl)oxy]benzamide (8g). Potassium carbonate (121 g, 0.867 mol)
was added portionwise to a solution of hydroxylamine sulfate (120 g,
J
dx.doi.org/10.1021/jm300953p | J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
Article
0.732 mol) in 360 mL of water at 0 °C. After the mixture was stirred
for 30 min, sodium sulfite (3.74 g, 0.029 mol) and a solution of methyl
2-hydroxy-6-[(R)-(tetrahydrofuran-3-yl)oxy]benzoate (7g) (35.0 g,
0.146 mol) in 360 mL of methanol were added and the mixture was
stirred at 50 °C for 30 h. Methanol was removed from the cooled
reaction mixture under reduced pressure, and the resulting aqueous
layer was acidified with 2 N HCl. The aqueous layer was extracted with
EtOAc and the organic layer was dried (Na2SO4) and concentrated to
4-(Cyclobutylmethoxy)-1,2-benzoxazol-3-ol (9c). 9c was
prepared from compound 8c in the manner described above. 24%
yield. H NMR (270 MHz, CDCl3) δ 1.95−1.98 (m, 4H), 2.16−2.20
(m, 2H), 2.86−2.91 (m, 1H), 4.13 (d, J = 6.4, 2H), 6.61 (d, J = 7.9,
1H), 6.96 (d, J = 8.4, 1H), 7.45 (dd, J = 6.4, 7.9, 1H).
4-(2-Methylpropoxy)-1,2-benzoxazol-3-ol (9d). 9d was pre-
pared from compound 8d in the manner described above. 42% yield.
1H NMR (270 MHz, CDCl3) δ 1.04 (d, J = 6.8, 6H), 2.05−2.20 (m,
1H), 3.84 (d, J = 6.6, 2H), 6.71 (d, J = 8.4, 1H), 6.84 (d, J = 8.1, 1H),
7.06 (dd, J = 8.1, 8.4, 1H).
1
1
afford 25.0 g (76% yield) of the product as a yellow solid. H NMR
(400 MHz, CDCl3) δ 2.00 (m, 1H), 2.15 (m, 1H), 3.80 (m, 4H), 5.05
(m, 1H), 6.48 (d, J = 8, 1H), 6.49 (d, J = 8, 1H), 7.19 (t, J = 8, 1H),
10.41 (br s, 1H), 11.49 (br s, 1H). MS (APCI) m/z = 239 (M + 1).
2-(Cyclobutyloxy)-N,6-dihydroxybenzamide (8a). 8a was
prepared from compound 7a following the general procedure
4-(2,2-Dimethylpropoxy)-1,2-benzoxazol-3-ol (9e). 9e was
prepared from compound 8e in the manner described above. 53%
yield. 1H NMR (270 MHz, CDCl3) δ 1.12 (s, 9H), 3.76 (s, 2H), 6.59
(d, J = 8.1, 1H), 6.95 (d, J = 8.4, 1H), 7.41 (dd, J = 8.4, 8.1, 1H).
4-[(4-Methoxybenzyl)oxy]-1,2-benzoxazol-3-ol (9f). 9f was
prepared from compound 8f in the manner described above. 63%
1
described above. 95% yield. H NMR (270 MHz, CDCl3) δ 1.70−
2.04 (m, 2H), 2.20−2.32 (m, 2H), 2.48−2.58 (m, 2H), 4.74−4.79 (m,
1H), 6.25 (d, J = 8.2, 1H), 6.61 (d, J = 8.4, 1H), 7.26 (dd, J = 8.4, 8.2,
1H), 10.57 (br s 1H), 12.70 (s, 1H).
1
yield. H NMR (270 MHz, DMSO-d6) δ 4.69 (s, 3H), 5.18 (s, 2H),
6.85 (d, J = 8.2, 1H), 6.96 (d, J = 8.5, 2H), 7.06 (d, J = 8.4, 1H), 7.43
(d, J = 8.9, 2H), 7.48 (dd, J = 8.4, 8.2, 1H).
2-(Cyclopentyloxy)-N,6-dihydroxybenzamide (8b). 8b was
prepared from compound 7b following the general procedure
Mitsunobu Coupling, Representative Procedure for Step e,
Scheme 1. tert-Butyl 4-{4-[(R)-(Tetrahydrofuran-3-yl)oxy]-
benzo[d]isoxazol-3-yloxymethyl}piperidine-1-carboxylate
(10g). Diethyl azodicarboxylate (15.6 g, 0.09 mol) was added to a
mixture of 4-[(R)-(tetrahydrofuran-3-yl)oxy]benzo[d]isoxazol-3-ol, 9g
(10 g, 0.045 mol), tert-butyl 4-hydroxymethylpiperidine-1-carboxylate
(11.6 g, 0.054 mol), and triphenylphosphine (23.5 g, 0.090 mol) in
300 mL of tetrahydrofuran. After the addition was complete, the
mixture was heated at reflux for 18 h. After concentration in vacuo, the
crude product was purified on a silica gel flash column, eluting with
petroleum ether/ethyl acetate (15:1 → 5:1) to afford 22.0 g of the
product as an oil. 51% yield. 1H NMR (400 MHz, CDCl3) δ 1.25 (m,
2H), 1.39 (s, 9H), 1.76 (m, 2H), 1.99 (m, 1H), 2.15 (m, 2H), 2.70 (br
t, J = 11.6, 2H), 3.95 (m, 4H), 4.13 (m, 2H), 4.34 (d, J = 6.4, 2H),
4.98 (m, 1H), 6.43 (d, J = 8, 1H), 6.93 (d, J = 8, 1H), 7.31 (t, J = 8,
1H). MS (APCI) m/z = 319 [(M + 1) − 99 (loss of BOC)].
Designated compounds were isolated as the tosylate salts, which
were prepared by dissolving the free bases in ethyl acetate and adding
0.95 equiv of p-toluenesulfonic acid. The resulting white solids were
filtered and dried.
1
described above. 78% yield. H NMR (270 MHz, CDCl3) δ 1.76−
2.11 (m, 8H), 4.90−5.00 (m, 1H), 6.40 (d, J = 8.6, 1H), 6.60 (d, J =
8.4, 1H), 7.26 (dd, J = 8.6, 8.4, 1H), 10.58 (br s, 1H), 12.7 (s, 1H).
2-(Cyclobutylmethoxy)-N,6-dihydroxybenzamide (8c). 8c
was prepared from compound 7c following the general procedure
described above. An inseparable mixture was isolated and utilized in
the next step.
N,2-Dihydroxy-6-(2-methylpropoxy)benzamide (8d). 8d was
prepared from compound 7d. 1H NMR (270 MHz, CDCl3) δ 1.09 (d,
J = 6.4, 6H), 2.10−2.26 (m, 1H), 3.93 (d, J = 7.85, 2H), 6.40 (d, J =
7.6, 1H), 6.63 (d, J = 8.4, 1H), 7.26 (dd, J = 8.4, 7.6, 1H), 10.57 (br s,
1H), 12.71 (s, 1H).
2-(2,2-Dimethylpropoxy)-N,6-dihydroxybenzamide (8e). 8e
was prepared from compound 7e following the general procedure
1
described above. Quantitative yield. H NMR (270 MHz, CDCl3) δ
1.15 (s, 9H), 3.85 (s, 2H), 6.41 (d, J = 8.2 Hz, 1H), 6.64 (d, J = 7.6
Hz, 1H), 7.28 (dd, J = 8.2, 7.3 Hz, 1H), 12.69 (s, 1H).
N,2-Dihydroxy-6-[(4-methoxybenzyl)oxy]benzamide (8f). 8f
was prepared from compound 7f following the general procedure
1
described above. 95% yield. H NMR (270 MHz, CDCl3) δ 3.81 (s,
3-[(1-Butylpiperidin-4-yl)methoxy]-4-(cyclobutyloxy)-1,2-
benzoxazole (1a). 1a was prepared in the manner described above,
utilizing compound 9a rather than 9g and (1-butylpiperidin-4-
yl)methanol in place of tert-butyl 4-hydroxymethylpiperidine-1-
3H), 5.07 (s, 2H), 6.50 (d, J = 8.4, 1H), 6.65 (d, J = 8.4, 1H), 6.96 (d,
J = 8.7, 2H), 7.26−7.37 (m, 3H), 10.42 (br s, 1H), 12.77 (br s, 1H).
Cyclization, Representative Procedure for Step d, Scheme 1.
4-[(R)-(Tetrahydrofuran-3-yl)oxy]benzo[d]isoxazol-3-ol (9g). A
solution of 2,N-dihydroxy-6-[(R)-(tetrahydrofuran-3-yl)oxy]-
benzamide (25.0 g, 0.105 mol) in 250 mL of THF was heated to 50
°C. Carbonyldiimidazole was added portionwise, and the resulting
mixture was stirred at 50 °C for 14 h. After the mixture was cooled to
room temperature, 100 mL of 2 N HCl was added and the aqueous
layer was extracted with ethyl acetate. The combined organic layers
were then extracted three times with 10% aqueous potassium
carbonate. The aqueous extracts were washed with ethyl acetate and
then acidified to pH 2−3 with 2 N HCl. The acidified aqueous layer
was extracted with ethyl acetate. The ethyl acetate extracts were
washed with brine, dried (Na2SO4), and concentrated to afford 20.0 g
of product as a yellow solid (43% yield). 1H NMR (400 MHz, CDCl3)
δ 2.20 (m, 2H), 3.89 (m, 1H), 4.01 (m, 3H), 5.05 (m, 1H), 6.48 (d, J
= 7.6, 1H), 6.92 (d, J = 7.6, 1H), 7.37 (t, J = 7.6, 1H). MS (APCI) m/z
= 222 (M + 1).
1
carboxylate. 65% yield (tosylate salt). H NMR (400 MHz, MeOH-
d4) δ 0.97 (t, J = 7.2, 3H), 1.39 (dq, J = 14.8, 7.4, 2H), 1.51−1.92 (m,
6H), 2.06−2.20 (m, 4H), 2.23−2.38 (m, 4H), 2.43−2.58 (m, 2H),
2.90−3.17 (m, 4H), 3.63 (d, J = 10.9, 2H), 4.27 (d, J = 5.5, 2H), 4.67−
4.83 (m, 1H), 6.55 (d, J = 8.2, 1H), 6.96 (d, J = 8.6, 1H), 7.19 (d, J =
7.8, 2H), 7.42 (dd, J = 8.6, 8.2, 1H), 7.67 (d, J = 8.2, 2H). HRMS calcd
for C21H31N2O3 (M + 1): 359.2329. Found: 359.2326.
3-[(1-Butylpiperidin-4-yl)methoxy]-4-(cyclopentyloxy)-1,2-
benzoxazole (1b). 1b was prepared in the manner described above,
utilizing compound 9b rather than 9g and (1-butylpiperidin-4-
yl)methanol in place of tert-butyl 4-hydroxymethylpiperidine-1-
1
carboxylate. 41% yield. H NMR (400 MHz, MeOH-d4) δ 0.98 (t, J
= 7.4, 3H), 1.40 (dq, J = 14.8, 7.4, 2H), 1.53−2.03 (m, 12H), 2.16 (d, J
= 14.1, 2H), 2.26 (br s, 1H), 2.95−3.16 (m, 4H), 3.63 (d, J = 10.9,
2H), 4.26 (d, J = 6.2, 2H), 4.94 (tt, J = 5.5, 2.5, 1H), 6.69 (d, J = 8.2,
1H), 6.94 (d, J = 8.2, 1H), 7.43 (dd, J = 8.2, 1H). HRMS calcd for
C22H33N2O3 (M + 1): 373.2486. Found: 373.2481.
4-(Cyclobutyloxy)-1,2-benzoxazol-3-ol (9a). 9a was prepared
1
from compound 8a in the manner described above. 25% yield. H
3-[(1-Butylpiperidin-4-yl)methoxy]-4-(cyclobutylmethoxy)-
1,2-benzoxazole (1c). 1c was prepared in the manner described
above, utilizing compound 9c rather than 9g and (1-butylpiperidin-4-
yl)methanol in place of tert-butyl 4-hydroxymethylpiperidine-1-
NMR (270 MHz, CDCl3) δ 1.69−1.87 (m, 1H) 1.90−2.04 (m, 2H),
2.25−2.40 (m, 2H), 2.48−2.59 (m, 2H), 4.77−4.87 (m, 1H), 6.47 (d, J
= 8.1, 1H), 7.12 (d, J = 8.4, 1H), 7.41 (dd, J = 8.4, 8.1, 1H).
4-(Cyclopentyloxy)-1,2-benzoxazol-3-ol (9b). 9b was prepared
1
carboxylate. 55% yield (tosylate salt). H NMR (400 MHz, MeOH-
1
d4) δ 0.97 (t, J = 7.4, 3H), 1.39 (dq, J = 15.0, 7.37, 2H), 1.50−1.79 (m,
3H), 1.87−2.40 (m, 8H), 2.72−2.87 (m, 2H), 2.90−3.18 (m, 4H),
3.62 (d, J = 10.2, 2H), 4.04 (d, J = 5.9, 2H), 4.25 (d, J = 5.1, 2H), 6.70
(d, J = 7.8, 1H), 6.97 (d, J = 8.2, 1H), 7.19 (d, J = 7.8, 2H), 7.45 (dd, J
from compound 8b in the manner described above. 46% yield. H
NMR (270 MHz, CDCl3) δ 1.67−1.88 (m, 1H), 1.98−2.05 (m, 7H),
4.94−4.97 (m, 1H), 6.60 (d, J = 8.1, 1H), 6.94 (d, J = 8.4, 1H), 7.43
(dd, J = 8.4, 8.1, 1H).
K
dx.doi.org/10.1021/jm300953p | J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
Article
= 8.2, 7.8, 1H), 7.67 (d, J = 8.2, 2H). HRMS calcd for C22H33N2O3 (M
+ 1): 373.2486. Found: 373.2485.
washed with brine, dried (Na2SO4), and concentrated to provide 17.0
g of crude product as a yellow oil. The crude material was purified by
preparative HPLC to afford 10.0 g of the desired product as a white
solid. (50% yield). 1H NMR (400 MHz, CDCl3) δ 1.41−1.63 (m, 6H),
1.71−1.81 (m, 2H), 1.81−1.94 (m, 1H), 2.17−2.26 (m, 2H), 2.33 (s,
2H), 2.4 (td, J = 11.7, 2.3, 2H), 2.92 (d, J = 11.8, 2H), 3.46 (s, 1H),
3.71−3.84 (m, 4H), 3.91−4.10 (m, 4H), 4.24 (d, J = 5.9, 2H), 5.03−
5.08 (m, 1H), 6.50 (d, J = 8.2, 1H), 7.00 (d, J = 8.2, 1H), 7.38 (t, J =
8.2, 1H). 13C NMR (100 MHz, CDCl3) δ 29.11, 33.10, 35.20, 36.92,
36.96, 56.15, 63.93, 67.14, 67.46, 68.27, 72.94, 74.06, 78.37, 103.17,
105.15, 131.71, 152.71, 166.02, 166.28. Elemental anal. calcd: C 63.87,
H 7.46, N 6.48. Found: C 63.91, H 7.78, N 6.48. Optical rotation:
+5.6° (c 0.0104 g/mL, MeOH). Melting point 92.6 °C.
3-[(1-Butylpiperidin-4-yl)methoxy]-4-(2-methylpropoxy)-
1,2-benzoxazole (1d). 1d was prepared in the manner described
above, utilizing compound 9d rather than 9g and (1-butylpiperidin-4-
yl)methanol in place of tert-butyl 4-hydroxymethyl-piperidine-1-
1
carboxylate. 54% yield (tosylate salt). H NMR (400 MHz, MeOH-
d4) δ 0.97 (t, J = 7.2, 3H), 1.05 (d, J = 6.6, 6H), 1.39 (dq, J = 14.9, 7.5,
2H), 1.51−1.77 (m, 4H), 2.00−2.37 (m, 7H). 2.91−3.15 (m, 4H),
3.62 (d, J = 9.8, 2H), 3.86 (d, J = 6.2, 2H), 4.25 (d, J = 5.5, 2H), 6.61−
6.75 (m, 1H), 6.96 (d, J = 8.2, 1H), 7.19 (d, J = 7.8, 2H), 7.44 (t, J =
8.2, 1H), 7.67 (d, J = 8.2 Hz, 2H). HRMS calcd for C21H33N2O3 (M +
1): 361.2486. Found: 361.2483.
3-[(1-Butylpiperidin-4-yl)methoxy]-4-(2,2-dimethylpro-
poxy)-1,2-benzoxazole (1e). 1e was prepared in the manner
described above, utilizing compound 9e rather than 9g and (1-
butylpiperidin-4-yl)methanol in place of tert-butyl 4-hydroxymethylpi-
4-{[4-({[4-(2-Methylpropoxy)-1,2-benzoxazol-3-yl]oxy}-
methyl)piperidin-1-yl]methyl}tetrahydro-2H-pyran-4-ol (1f). 1f
was synthesized in the manner described above from 4-(2-
methylpropoxy)-3-(piperidin-4-ylmethoxy)-1,2-benzoxazole. 68%
1
1
peridine-1-carboxylate. 58% yield (tosylate salt). H NMR (400 MHz,
yield. H NMR (270 MHz, CDCl3) δ 1.17 (d, J = 6.6, 6H), 1.44−
MeOH-d4) δ 0.97 (t, J = 7.2, 3H), 1.03−1.11 (m, 9H), 1.31−1.46 (m,
2H), 1.70 (dt, J = 7.9, 4.05, 5H), 2.32 (s, 5H), 2.89−3.16 (m, 4H),
3.54−3.69 (m, 2H), 3.74 (s, 2H), 4.25 (d, J = 6.2, 2H), 6.62−6.72 (m,
1H), 6.96 (d, J = 8.2, 1H), 7.19 (d, J = 8.2, 2H), 7.45 (t, J = 8.2, 1H),
7.67 (d, J = 8.2, 2H). HRMS calcd for C22H35N2O3 (M + 1):
375.2642. Found: 375.2638.
1.65 (m, 7H), 1.80−1.90 (m, 3H), 2.11−2.20 (m, 1H), 2.34 (s, 2H),
2.42 (t, J = 10.2, 2H), 2.92 (d, J = 11.4, 2H), 3.77−3.87 (m, 6H), 4.27
(d, J = 5.9, 2H), 6.56 (d, J = 8.1, 1H), 6.97 (d, J = 8.2, 1H), 7.38 (dd, J
= 8.2, 8.1, 1H). HRMS calcd for C23H35N2O5 (M + 1): 419.2450.
Found: 419.2544.
4-{[4-({[4-(2-Hydroxy-2-methylpropoxy)-1,2-benzoxazol-3-
yl]oxy}methyl)piperidin-1-yl]methyl}tetrahydro-2H-pyran-4-ol
(2a). 2a was prepared from compound 10f in three steps as depicted
below.
(1). 3-(Piperidin-4-ylmethoxy)-1,2-benzoxazol-4-ol (11). tert-
Butyl 4-[({4-[(4-methoxybenzyl)oxy]-1,2-benzoxazol-3-yl}oxy)-
methyl]piperidine-1-carboxylate (10f) (2.29 g, 4.89 mmol) was
combined with thioanisole (6.0 mL, 46.0 mmol) and o-cresol (5.30
g, 49.0 mmol). Trifluoroacetic acid (3 mL) was added. After the
mixture was stirred at room temperature for 2 h, 50 mL of hexane was
added and the solvent was decanted. After two additional hexane
washes, 50 mL of diethyl ether was added. After the mixture was
stirred for 30 min, a white solid was filtered and air-dried to afford 1.82
g of the trifluoroacetic acid salt of the title compound as a white solid.
1H NMR (270 MHz, DMSO-d6) δ 1.45−1.53 (m, 2H), 1.91−1.96 (m,
2H), 2.10−2.30 m, 1H), 2.91−2.96 (m, 2H), 3.45−3.47 (m, 2H), 4.25
(d, J = 6.3, 2H), 6.65 (d, J = 7.7, 1H), 6.95 (d, J = 8.4, 1H), 7.38 (dd, J
= 8.4, 7.7, 1H), 8.61 (br s, 1H), 10.73 (s, 1H).
(2). 3-({1-[(4-Hydroxytetrahydro-2H-pyran-4-yl)methyl]-
piperidin-4-yl}methoxy)-1,2-benz-oxazol-4-ol (12). 12 was pre-
pared from compound 11 in the manner described in step g above. 1H
NMR (270 MHz, CDCl3) δ 1.47−1.65 (m, 7H), 1.79−1.84 (m, 2H),
1.97−2.04 (m, 1H), 2.34−2.45 (m, 2H), 2.90−2.95 (m, 2H), 3.77−
3.96 (m. 5H), 4.35 (d, J = 6.8, 2H), 6.66 (d, J = 7.9, 1H), 6.97 (d, J =
8.5, 1H), 7.41 (dd, J = 8.5, 7.9, 1H).
(3). Step i, Scheme 1. 4-{[4-({[4-(2-Hydroxy-2-methylpro-
poxy)-1,2-benzoxazol-3-yl]oxy}methyl)piperidin-1-yl]methyl}-
tetrahydro-2H-pyran-4-ol (2a). Compound 12 (109 mg, 0.3 mmol)
was stirred with 0.5 mL of 2 N NaOH (1.00 mmol) and 2 mL of
ethanol at 50 °C for 20 h. 2,2-Dimethyloxirane (139 mg, 1.93 mmol)
was then added, and the mixture was stirred at 50 °C for an additional
20 h. After cooling, the reaction mixture was diluted with ethyl acetate
and washed with brine. The organic layer was dried (Na2SO4) and
concentrated to an oil. The crude product was purified on an alumina
column to afford 89.8 mg of the title compound as a white solid in
80% yield. 1H NMR (400 MHz, MeOH-d4) δ 1.35 (s, 6H), 1.41−1.59
(m, 4H), 1.68 (ddd, J = 13.6, 11.4, 5.1, 2H), 1.79 (d, J = 12.1, 2H),
2.21−2.37 (m, 4H), 2.96 (d, J = 11.3, 2H), 3.61−3.80 (m, 5H), 3.87
(s, 2H), 4.19 (d, J = 5.9, 2H), 6.69 (d, J = 8.2, 1H), 6.97 (d, J = 8.6,
1H), 7.44 (dd, J = 8.6, 8.2, 1H). HRMS calcd for C23H35N2O (M + 1):
435.2490. Found: 435.2486.
tert-Butyl 4-({[4-(2-Methylpropoxy)-1,2-benzoxazol-3-yl]-
oxy}methyl)piperidine-1-carboxylate (10d). 10d was prepared
in the manner described above utilizing compound 9d rather than
compound 9g. 71% yield. 1H NMR (270 MHz, CDCl3) δ 1.07 (d, J =
6.8, 6H), 1.23−1.47 (m, 12H), 1.82−1.87 (m, 2H), 2.04−2.19 (m,
2H), 2.71− 2.76 (m, 2H), 3.84 (d, J = 6.4, 2H), 4.27 (d, J = 6.3, 2H),
6.55 (d, J = 7.9, 1H), 6.97 (d, J = 8.6, 1H), 7.39 (dd J = 8.6, 7.9, 1H).
tert-Butyl 4-[({4-[(4-Methoxybenzyl)oxy]-1,2-benzoxazol-3-
yl}oxy)methyl]piperidine-1-carboxylate (10f). 10f was prepared
in the manner described above utilizing compound 9f rather than
1
compound 9g. 68% yield H NMR (270 MHz, CDCl3) δ 1.23−1.32
(m, 2H), 1.47 (s, 9H), 1.78−1.82 (m, 2H), 1.98−2.14 (m, 1H), 2.66−
2.76 (m, 2H), 3.84 (s, 2H), 4.05−4.20 (m, 2H), 4.24 (d, J = 6.9, 2H),
6.67 (d, J = 7.9, 1H), 6.92 (d, J = 8.7, 2H), 7.02 (d, J = 8.4, 1H), 7.38−
7.44 (m, 3H).
Representative Procedure for Step f, Scheme 1. BOC
Deprotection, 3-(Piperidin-4-ylmethoxy)-4-[(R)-(tetrahydrofur-
an-3-yl)oxy]benzo-[d]isoxazole. A 0 °C solution of tert-butyl 4-{4-
[(R)-(tetrahydrofuran-3-yl)oxy]benzo-[d]isoxazol-3-yloxymethyl}-
piperidine-1-carboxylate (10g) in 500 mL of ether was treated with a
saturated solution of HCl (g) in 200 mL of ether. After addition was
complete, the mixture was warmed to room temperature and stirred
for 16 h. The reaction mixture was filtered. The white solid was
washed with ethyl acetate followed by ether and dried to yield 15.0 g
1
(81% yield) of the desired product as a white solid. H NMR (400
MHz, MeOH-d4) δ 1.51−1.69 (m, 2H), 2.04−2.19 (m, 3H), 2.22−
2.37 (m, 2H), 2.99−3.14 (m, 2H), 3.40−3.51 (m, 2H), 3.85−4.02 (m,
4H), 4.25−4.31 (m, 2H), 5.17 (td, J = 3.7, 1.6, 1H), 6.72 (d, J = 8.0,
1H), 7.01 (d, J = 8.6, 1H), 7.47 (t, J = 8.2, 1H). MS (APCI) m/z = 319
(M + 1).
4-(2-Methylpropoxy)-3-(piperidin-4-ylmethoxy)-1,2-benzox-
azole. The compound was prepared according to the general
procedure described above utilizing compound 10d as starting
material. 75% yield. 1H NMR (270 MHz, CDCl3) δ 1.05 (d, J =
6.8, 6H), 1.75−1.88 (m, 2H), 2.04−2.16 (m, 3H), 2.90−2.98 (m, 2H),
3.55−3.60 (m, 2H), 3.85 (d, J = 6.3, 2H), 4.32 (d, J = 7.1, 2H), 6.56
(d, J = 7.9, 1H), 6.98 (d, J = 8.4, 1H), 7.40 (dd, J = 8.4, 7.9, 1H).
Epoxide Opening, Representative Procedure for Step g,
Scheme 1. (4-{4-[(R)-(Tetrahydrofuran-3-yl)oxy]benzo[d]-
i s o x a z o l - 3 - y l o x y m e t h y l } p i p e r i d i n - 1 - y l m e t h y l ) -
tetrahydropyran-4-ol (2d). 1,6-Dioxaspiro[2.5]octane (9.70 g,
0.084 mol) and triethylamine (8.60 g, 0.084 mol) were added to a
solution of 3-(piperidin-4-ylmethoxy)-4-[(R)-(tetrahydrofuran-3-yl)-
oxy]benzo[d]isoxazole (15.0 g, 0.042 mol) in 200 mL of methanol.
The resulting solution was heated at reflux for 18 h. The cooled
mixture was concentrated, and ethyl acetate and water were added to
the residue. The layers were separated, and the organic extracts were
4-[(4-{[(4-{[(1R,3R)-3-Hydroxycyclopentyl]oxy}-1,2-benzoxa-
zol-3-yl)oxy]methyl}piperidin-1-yl)methyl]tetrahydro-2H-
pyran-4-ol (2b). 2b was prepared from compound 11 in four steps as
shown below.
(1). Step j, Scheme 1. tert-Butyl 4-{[(4-Hydroxy-1,2-benzox-
azol-3-yl)oxy]methyl}piperidine-1-carboxylate. Triethylamine
(2.10 mL, 15.0 mmol) was added to a solution of compound 11
L
dx.doi.org/10.1021/jm300953p | J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
Article
(1.82 g, 5 mmol) in 15 mL of THF. Di-tert-butyl dicarbonate (1.17 g,
5.36 mmol) was added, and the reaction mixture was stirred at room
temperature. After 18 h, the mixture was diluted with ethyl acetate and
washed three times with brine. The organic layer was dried (MgSO4)
and concentrated to afford 1.77 g of the desired product in quantitative
yield. 1H NMR (270 MHz,CDCl3) δ 1.26−1.56 (m, 13H), 1.75−1.86
(m, 2H), 2.05−2.15 (m, 1H), 2.65−2.83 (m, 2H), 4.34 (d, J = 6.6,
2H), 6.67 (d, J = 7.9, 1H), 6.95 (d, J = 8.4, 1H), 7.35 (dd, J = 8.4, 7.4,
1H).
pyran-4-yloxy)-1,2-benzoxazole for 3-(piperidin-4-ylmethoxy)-4-[(R)-
(tetrahydrofuran-3-yl)oxy]benzo[d]isoxazole. 1H NMR (400 MHz,
MeOH-d4) δ 1.40−1.61 (m, 5H), 1.63−1.73 (m, 3H), 1.73−1.89 (m,
6H), 1.98−2.07 (m, 3H), 2.25−2.37 (m, 5H), 2.97 (d, J = 11.3, 2H),
3.57−3.79 (m, 7H), 3.97 (ddd, J = 11.3, 7.8, 3.1, 3H), 4.20 (d, J = 6.2,
2H), 4.76−4.80 (m, 1H), 6.76 (d, J = 8.2, 1H), 6.96 (d, J = 8.6, 1H),
7.43 (dd, J = 8.6, 8.2, 1H). HRMS calcd for C24H35N2O6 (M + 1):
447.2490. Found: 447.2487.
Representative Procedure for Step a. 4-Nitrophenyl 6-
Fluoro-3,3-dimethyl-2-oxo-2,3-dihydroindole-1-carboxylate
(14). Triethylamine (34.0 g, 330 mmol) was added to a solution of 6-
fluoro-3,3-dimethyl-1,3-dihydroindol-2-one31 (10.0 g, 56.0 mmol) and
4-nitrophenyl chloroformate (22.0 g, 220 mmol) in toluene (150 mL).
The resulting reaction mixture was heated at reflux for 18 h, then
cooled to room temperature and concentrated under reduced pressure.
Cold methanol was added and the mixture was filtered, washed with
additional cold methanol, and dried to afford 15.0 g of the title
compound as a white solid (79% yield). 1H NMR (400 MHz, CDCl3)
δ 1.51 (s, 6H), 7.00−6.90 (m, 1H), 7.26−7.16 (m, 1H), 7.56−7.49 (m,
2H), 7.76−7.71 (m, 1H), 8.36−8.33 (m, 2H).
(2). tert-Butyl 4-{[(4-{[(1R,3R)-3-Hydroxycyclopentyl]oxy}-
1,2-benzoxazol-3-yl)oxy]methyl}piperidine-1-carboxylate. The
compound was prepared according to the representative procedure for
step e, substituting tert-butyl 4-{[(4-hydroxy-1,2-benzoxazol-3-yl)oxy]-
methyl}piperidine-1-carboxylate for 9g and (1S,3S)-cyclopentane-1,3-
diol for tert-butyl 4-hydroxymethylpiperidine-1-carboxylate. Quantita-
1
tive yield. H NMR (270 MHz, CDCl3) δ 1.43 (s, 9H), 1.67−2.22 (s,
10H), 2.73−3.01 (m, 3H), 4.08−4.37 (m, 5H), 5.05−5.09 (m, 1H),
6.61 (d, J = 7.9, 1H), 7.00 (d, J = 8.4, 1H), 7.41 (dd, J = 8.4, 7.9, 1H).
(3). Representative Procedure for Step k, Scheme 1. (1R,3R)-
3-{[3-(Piperidin-4-ylmethoxy)-1,2-benzoxazol-4-yl]oxy}-
cyclopentanol. tert-Butyl 4-{[(4-{[(1R,3R)-3-hydroxycyclopentyl]-
oxy}-1,2-benzoxazol-3-yl)oxy]methyl}piperidine-1-carboxylate (175
mg, 0.405 mmol) was treated with a solution of 200 μL of
concentrated HCl in 1.8 mL of methanol. After the mixture was
stirred for 3 h at room temperature, the solvent was evaporated to
Representative Procedure for Step b. 6-Fluoro-3,3-dimeth-
yl-2-oxo-2,3-dihydroindole-1-carboxylic Acid [1-(4-Hydroxyte-
trahydropyran-4-ylmethyl)piperidin-4-ylmethyl]amide (3).
Triethylamine (26 g, 256 mmol) was added to 4-(4-amino-
methylpiperidin-1-ylmethyl)tetrahydropyran-4-ol hydrochloride
(20)15 in 300 mL of dichloromethane. After the mixture was stirred
for 5 min, 4-nitrophenyl 6-fluoro-3,3-dimethyl-2-oxo-2,3-dihydroin-
dole-1-carboxylate 14 (15 g, 44 mmol) was added and the reaction
mixture was stirred at room temperature for 16 h. The mixture was
concentrated under reduced pressure and the crude material was
purified on a silica gel flash column, eluting with petroleum ether/ethyl
acetate (50:1 → 3:1) to yield the desired product as a white solid (13
1
afford the title compound as the HCl salt. 60% yield. H NMR (270
MHz, DMSO-d6) δ 1.54−1.99 (m, 10H), 2.18−2.39 (m, 1H), 2.51−
2.93 (m, 2H), 3.28−3.39 (m, 2H), 4.02−4.24 (m, 3H), 4.85−4.90 (m
1H), 6.75 (d, J = 8.1, 1H), 7.04 (d, J = 8.4, 1H), 7.51 (dd, J = 8.4, 8.1,
1H), 9.05 (br s, 1H).
(4). 4-[(4-{[(4-{[(1R,3R)-3-Hydroxycyclopentyl]oxy}-1,2-ben-
zoxazol-3-yl)oxy]methyl}piperidin-1-yl)methyl]tetrahydro-2H-
pyran-4-ol (2b). 2b was prepared according to the general procedure
for step g substituting (1R,3R)-3-{[3-(piperidin-4-ylmethoxy)-1,2-
benzoxazol-4-yl]oxy}cyclopentanol for 3-(piperidin-4-ylmethoxy)-4-
[(R)-(tetrahydrofuran-3-yl)oxy]benzo[d]isoxazole. 63% yield. 1H
NMR (400 MHz, MeOH-d4) δ 1.38−1.60 (m, 4H), 1.62−2.11 (m,
10H), 2.24−2.46 (m, 5H), 2.97 (d, J = 10.9, 2H), 3.62−3.79 (m, 4H),
4.18 (d, J = 6.2, 2H), 4.27 (dd, J = 11.2, 5.5, 1H), 4.90 (d, J = 3.1, 1H),
6.65 (d, J = 7.8, 1H), 6.94 (d, J = 8.2, 1H), 7.43 (dd, J = 8.4, 7.8, 1H).
HRMS calcd for C24H35N2O6 (M + 1): 447.2490. Found: 447.2488.
4-{[4-({[4-(Tetrahydro-2H-pyran-4-yloxy)-1,2-benzoxazol-3-
yl]oxy}methyl)piperidin-1-yl]methyl}tetrahydro-2H-pyran-4-ol
(2c). 2c was prepared in three steps from tert-butyl 4[(4-
hydroxybenzo[d]isoxazol-3-yloxy)methyl]piperidine-1-carboxylate as
described below.
1
g, 69% yield). H NMR (400 MHz, CDCl3) δ 1.25−1.37 (m, 2H),
1.40 (s, 6H), 1.45−1.47 (m, 1H), 1.50−1.63 (m, 4H), 1.72 (d, J =
12.7, 2H), 2.30 (s, 2H), 2.34 (td, J = 11.8, 2.34, 1H), 8.63 (t, J = 5.6,
1H). 13C NMR (100 MHz, CDCl3) δ 25.3, 30.7, 35.7, 37.3, 45.2, 45.6,
56.3, 64.2, 67.6, 68.4, 105.1, 105.4, 111.3, 111.6, 122.8, 122.9, 163.8,
184.2. Elemental anal. calcd: C 63.72, H 7.44, N 9.69. Found: C 63.70,
H 7.68, N 9.73. Melting point 155.1 °C.
6-Fluoro-N-({1-[(4-hydroxytetrahydro-2H-pyran-4-yl)-
methyl]piperidin-4-yl}methyl)-3-isopropyl-2-oxo-2,3-dihydro-
1H-benzo[d]imidazole-1-carboxamide (4). Analogue 4 was
prepared over two steps in an analogous fashion to compound 3
above utilizing 5-fluoro-1-isopropyl-1H-benzo[d]imidazol-2(3H)-
one14 (15) as starting material.
(1). 4-Nitrophenyl 6-Fluoro-3-isopropyl-2-oxo-2,3-dihydro-
1H-benzo[d]imidazole-1-carboxy-late (16). 16 was prepared
using the procedure from step l, substituting 5-fluoro-1-isopropyl-
1H-benzo[d]imidazol-2(3H)-one (15) for 6-fluoro-3,3-dimethyl-1,3-
(1). tert-Butyl 4-({[4-(Tetrahydro-2H-pyran-4-yloxy)-1,2-ben-
zoxazol-3-yl]oxy}methyl)piper-idine-1-carboxylate. The com-
pound was synthesized according to the general procedure in step a,
substituting tert-butyl 4-[(4-hydroxybenzo[d]isoxazol-3-yloxy)-
methyl]piperidine-1-carboxylate for 5-hydroxy-2,2-dimethylbenzo-
[1,3]dioxin-4-one and tetrahydro-2H-pyran-4-ol for (S)-tetrahydrofur-
1
dihydroindol-2-one. 72% yield. H NMR (400 MHz, CDCl3) δ 1.57
(d, J = 7.0, 6H), 4.68 (dt, J = 14.0, 7.0, 1H), 6.94−7.01 (m, 1H), 7.03−
7.10 (m, 1H), 7.47−7.53 (m, 2H), 7.73 (dd, J = 9.0, 2.5, 1H), 8.30−
8.38 (m, 2H).
1
an-3-ol. H NMR (270 MHz, CDCl3) δ 1.25−1.45 (m, 2H), 1.47 (s,
9H), 1.75−1.93 (m, 4H), 1.94−2.20 (m, 2H), 2.60−2.90 (m, 2H),
3.54−3.71 (m, 2H), 3.85−4.25 (m, 3H), 4.28 (d, J = 6.2, 2H), 4.62−
4.76 (m, 1H), 6.61 (d, J = 8.4, 1H), 7.00 (d, J = 8.4, 1H), 7.39 (dd, J =
8.4, 1H).
(2). 6-Fluoro-N-({1-[(4-hydroxytetrahydro-2H-pyran-4-yl)-
methyl]piperidin-4-yl}methyl)-3-isopropyl-2-oxo-2,3-dihydro-
1H-benzo[d]imidazole-1-carboxamide (4). 4 was prepared
according to the procedure for step m, substituting 4-nitrophenyl 6-
fluoro-3-isopropyl-2-oxo-2,3-dihydro-1H-benzo[d]imidazole-1-carbox-
ylate (16) for 4-nitrophenyl 6-fluoro-3,3-dimethyl-2-oxo-2,3-dihy-
droindole-1-carboxylate (14). 89% yield. 1H NMR (400 MHz,
CDCl3) δ 1.23−1.65 (m, 15H), 1.73 (d, J = 12.5, 2H), 2.26−2.39
(m, 4H), 2.87 (d, J = 11.7, 2H), 3.29 (t, J = 6.2, 2H), 3.42 (br s, 1H),
3.68−3.83 (m, 4H), 4.66 (spt, J = 7.0, 1H), 6.89 (td, J = 8.9, 2.6, 1H),
7.04 (dd, J = 8.8, 4.5, 1H), 8.03 (dd, J = 9.6, 2.7, 1H), 8.87 (t, J = 5.8,
1H). Elemental anal. calcd: C 61.59, H 7.42, N 12.49. Found: C 61.38,
H 7.52, N 12.31. Melting point 112.4 °C.
(2). 3-(Piperidin-4-ylmethoxy)-4-(tetrahydro-2H-pyran-4-
yloxy)-1,2-benzoxazole. The compound was prepared according
to the representative procedure of step k, substituting tert-butyl 4-({[4-
(tetrahydro-2H-pyran-4-yloxy)-1,2-benzoxazol-3-yl]oxy}methyl)-
piperidine-1-carboxylate for tert-butyl 4-{[(4-{[(1R,3R)-3-
hydroxycyclopentyl]oxy}-1,2-benzoxazol-3-yl)oxy]methyl}piperidine-
1
1-carboxylate. H NMR (270 MHz, CDCl3) δ 1.75−1.95 (m, 4H),
1.99−2.37 (m, 5H), 2.85−3.05 (m, 2H), 3.50−3.71 (m, 4H), 3.91−
4.05 (m, 2H), 4.34 (d, J = 6.9, 2H), 4.65−4.76 (m, 1H), 6.62 (d, J =
8.4, 1H), 7.00 (d, J = 8.4, 1H), 7.40 (dd, J = 8.4, 8.4, 1H).
1-[(4-Hydroxytetrahydro-2H-pyran-4-yl)methyl]piperidine-
4-carbonitrile (20). To a mixture of cyanopiperidine (148.7 g, 1.35
mol) in methanol (800 mL) was charged epoxide (167.3 g, 1.47 mol)
as a solution in methanol (200 mL). The resulting mixture was heated
to 55 °C for 5 h and then concentrated to an oil, which was crystallized
(3). 4-{[4-({[4-(Tetrahydro-2H-pyran-4-yloxy)-1,2-benzoxa-
zol-3-yl]oxy}methyl)piperidin-1-yl]methyl}tetrahydro-2H-
pyran-4-ol (2c). 2c was prepared according to the general procedure
for step g, substituting 3-(piperidin-4-ylmethoxy)-4-(tetrahydro-2H-
M
dx.doi.org/10.1021/jm300953p | J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
Article
from heptane (1 L), filtered, and dried on the filter to afford 283 g of
REFERENCES
■
4-(4-hydroxytetrahydropyran-4-ylmethyl)cyclohexanecarbonitrile as a
(1) Lopez, O. L. The growing burden of Alzheimer’s disease. Am. J.
Managed Care 2011, 17 (Suppl.13), S339−S345.
(2) 2012 Alzheimer’s disease facts and figures. Alzheimer's Dementia
2012, 8 (2), 131−168.
(3) Wimo, A.; Winblad, B.; Jonsson, L. The worldwide societal costs
of dementia: estimates for 2009. Alzheimer's Dementia 2010, 6 (2),
98−103.
(4) Bartus, R. T.; Dean, R. L., 3rd; Beer, B.; Lippa, A. S. The
cholinergic hypothesis of geriatric memory dysfunction. Science 1982,
217 (4558), 408−414.
(5) Francis, P. T.; Palmer, A. M.; Snape, M.; Wilcock, G. K. The
cholinergic hypothesis of Alzheimer’s disease: a review of progress. J.
Neurol. Neurosurg. Psychiatry 1999, 66 (2), 137−147.
(6) Eglen, R. M.; Wong, E. H.; Dumuis, A.; Bockaert, J. Central 5-
HT4 receptors. Trends Pharmacol. Sci. 1995, 16 (11), 391−398.
(7) Consolo, S.; Arnaboldi, S.; Giorgi, S.; Russi, G.; Ladinsky, H. 5-
HT4 receptor stimulation facilitates acetylcholine release in rat frontal
cortex. NeuroReport 1994, 5 (10), 1230−1232.
(8) Mohler, E. G.; Shacham, S.; Noiman, S.; Lezoualc’h, F.; Robert,
S.; Gastineau, M.; Rutkowski, J.; Marantz, Y.; Dumuis, A.; Bockaert, J.;
Gold, P. E.; Ragozzino, M. E. VRX-03011, a novel 5-HT4 agonist,
enhances memory and hippocampal acetylcholine efflux. Neuro-
pharmacology 2007, 53 (4), 563−573.
(9) Fontana, D. J.; Daniels, S. E.; Eglen, R. M.; Wong, E. H.
Stereoselective effects of (R)- and (S)-zacopride on cognitive
performance in a spatial navigation task in rats. Neuropharmacology
1996, 35 (3), 321−327.
(10) Galeotti, N.; Ghelardini, C.; Bartolini, A. Role of 5-HT4
receptors in the mouse passive avoidance test. J. Pharmacol. Exp. Ther.
1998, 286 (3), 1115−1121.
(11) De Maeyer, J. H.; Lefebvre, R. A.; Schuurkes, J. A. 5-HT4
receptor agonists: similar but not the same. Neurogastroenterol. Motil.
2008, 20 (2), 99−112.
(12) Johnson, D. E.; Drummond, E.; Grimwood, S.; Sawant-Basak,
A.; Miller, E.; Tseng, E.; McDowell, L. L.; Vanase-Frawley, M. A.;
Fisher, K. E.; Rubitski, D. M.; Stutzman-Engwall, K. J.; Nelson, R. T.;
Horner, W. E.; Gorczyca, R. R.; Hajos, M.; Siok, C. J. The 5-
hydroxytryptamine 4 receptor agonists prucalopride and PRX-03140
increase acetylcholine and histamine levels in the rat prefrontal cortex
and the power of stimulated hippocampal theta oscillations. J.
Pharmacol. Exp. Ther. 2012, 341 (3), 681−691.
(13) Wager, T. T.; Chandrasekaran, R. Y.; Hou, X.; Troutman, M.
D.; Verhoest, P. R.; Villalobos, A.; Will, Y. Defining desirable central
nervous system drug space through the alignment of molecular
properties, in vitro ADME, and safety attributes. ACS Chem. Neurosci.
2010, 1 (6), 420−434.
(14) Iguchi, S.; Katsu, Y.; Sone, H.; Uchida, C.; Kojima, T.
Preparation of Benzimidazolones Having 5-HT4 Receptor Agonistic
Activity. WO2005021539, 2005.
1
white solid. H NMR (CD3OD, 400 MHz) δ 3.77−3.64 (m, 4H),
2.77−2.73 (m, 3H), 2.50−2.46 (m, 2H), 2.31 (s, 2H), 1.94−1.87 (m,
2H), 1.83−1.75 (m, 2H), 1.68−1.61 (m, 2H), 1.49−1.48 (m, 2H). 13C
NMR (CD3OD, 100 MHz) δ 122.0, 69.1, 68.2, 63.6, 53.8, 35.6, 29.0,
25.6.
4-(4-Aminomethylpiperidin-1-ylmethyl)tetrahydropyran-4-
ol Maleate Acid (21). A solution of nitrile 20 in THF (3.36 L) was
charged with sponge nickel (JM type A, 1.67 mol, 98.00 g) under
nitrogen. The mixture was hydrogenated at 30 psi for 12 h and was
then filtered and treated with a solution of maleic acid (145 g, 1.25
mol) in 400 mL of THF over 1 h. The mixture was stirred for 2 h and
the resulting solids were collected by filtration, rinsed with THF (500
mL), and dried on the filter for 12 h to afford 415 g of 4-(4-
aminomethylpiperidin-1-ylmethyl)tetrahydropyran-4-ol, maleate salt
1
as a white solid, mp 135−137 °C. H NMR (CD3OD, 400 MHz) δ
6.22 (s, 2H), 3.78−3.67 (m, 4H), 3.08 (d, J = 11.6, 2H), 2.83 (d, J =
7.1, 2H), 2.45 (s, 2H), 2.39 (t, J = 11.6, 2H), 1.74−1.60 (m, 5H),
1.53−1.38 (m, 4H). 13C NMR (CD3OD, 100 MHz) δ 170.0, 135.1,
68.7, 68.0, 63.5, 55.4, 44.5, 35.8, 33.8, 29.0.
ASSOCIATED CONTENT
■
S
* Supporting Information
Measured and predicted brain concentrations, summary of
toxicology results, and Cerep profile for 2d, 3, and 4. This
material is available free of charge via the Internet at http://
pubs.acs.org.
AUTHOR INFORMATION
■
Corresponding Author
*Phone: 617-395-0706. E-mail: michael.a.brodney@pfizer.com.
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
The authors sincerely thank the legacy 5HT4 project team from
the Gastrointestinal Reseach Unit, Pfizer Global Research and
Development, from Nagoya, Japan, for the initial discovery,
SAR, and characterization of multiple chemical series. In
particular, we acknowledge the contributions in medicinal
chemistry from Kiyoshi Kawamura, Hiroki Sone, Satoru Iguchi,
Yasuhiro Katsu, Isao Sakurada, and Chikara Uchida as well as
Atsushi Nagahisa for helpful advice during the preparation of
this manuscript. In addition we thank Roxanne R. Gorczyca for
providing assistance in microdialysis studies; R. Scott Obach for
helpful discussions about the ADME data; James M. Duerr,
Robin T. Nelson, and Kathlene McGrath for 5HT4 cell lines
preparations; and Katherine Brighty and Bruce Rogers for
comments on this manuscript.
(15) Kawamura, K.; Sone, H.; Uchida, C. Synthesis of Oxyindolyl
Piperidine Derivatives as 5-HT4 Receptor Agonists. WO2006090279,
2006.
(16) Noguchi, H.; Sakurada, I.; Uchida, C.; Waizumi, N. Preparation
of Benzoisoxazole Derivatives as 5-HT4 Receptor Agonists.
WO2006090224, 2006.
(17) Callegari, E.; Malhotra, B.; Bungay, P. J.; Webster, R.; Fenner, K.
S.; Kempshall, S.; LaPerle, J. L.; Michel, M. C.; Kay, G. G. A
comprehensive non-clinical evaluation of the CNS penetration
potential of antimuscarinic agents for the treatment of overactive
bladder. Br. J. Clin. Pharmacol. 2011, 72 (2), 235−246.
(18) Feng, B.; Mills, J. B.; Davidson, R. E.; Mireles, R. J.; Janiszewski,
J. S.; Troutman, M. D.; de Morais, S. M. In vitro P-glycoprotein assays
to predict the in vivo interactions of P-glycoprotein with drugs in the
central nervous system. Drug Metab. Dispos. 2008, 36 (2), 268−275.
(19) Kalvass, J. C.; Maurer, T. S. Influence of nonspecific brain and
plasma binding on CNS exposure: implications for rational drug
discovery. Biopharm. Drug Dispos. 2002, 23 (8), 327−338.
ABBREVIATIONS USED
■
5-HT, serotonin; 5-HT4, serotonin receptor subtype 4; ACh,
acetylcholine; AD, Alzheimer’s disease; AUC, area under the
curve; cAMP, cyclic adenosine monophosphate; CHO, Chinese
hamster ovary; CNS, central nervous system; CSF, cerebrospi-
nal fluid; EC50, concentration giving a 50% response; Emax
,
maximal effect; GPCR, G-protein-coupled receptor; HEK,
human embryonic kidney; RRCK, HLM, human liver micro-
some; Ki, constant for inhibition of binding; MDCK, Madin−
Darby canine kidney; MWM, Morris water maze; Pgp, P-
gycloprotein; RO, receptor occupancy; SEM, standard error of
the mean; sc, subcutaneous; THP, tetrahydropyran
N
dx.doi.org/10.1021/jm300953p | J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
Article
(20) The following software was utilized: NONMEM, version 6;
ICON: Columbia, MD.
(21) Mikami, T.; Sugimoto, H.; Naganeo, R.; Ohmi, T.; Saito, T.;
Eda, H. Contribution of active and inactive states of the human 5-
HT4d receptor to the functional activities of 5-HT4-receptor agonists.
J. Pharmacol. Sci. 2008, 107 (3), 251−259.
(22) Cohen, M. L.; Bloomquist, W.; Calligaro, D.; Swanson, S.
Comparative 5-HT4 receptor antagonist activity of LY353433 and its
active hydroxylated metabolites. Drug Dev. Res. 1998, 43 (4), 193−
199.
(23) Langlois, M.; Fischmeister, R. 5-HT4 receptor ligands:
applications and new prospects. J. Med. Chem. 2003, 46 (3), 319−344.
(24) Talley, N. J. Serotoninergic neuroenteric modulators. Lancet
2001, 358 (9298), 2061−2068.
(25) Krobert, K. A.; Brattelid, T.; Levy, F. O.; Kaumann, A. J.
Prucalopride is a partial agonist through human and porcine atrial 5-
HT4 receptors: comparison with recombinant human 5-HT4 splice
variants. Naunyn Schmiedeberg's Arch. Pharmacol. 2005, 371 (6), 473−
479.
(26) Grimwood, S.; Hartig, P. R. Target site occupancy: emerging
generalizations from clinical and preclinical studies. Pharmacol. Ther.
2009, 122 (3), 281−301.
(27) Fontana, D. J.; Daniels, S. E.; Wong, E. H.; Clark, R. D.; Eglen,
R. M. The effects of novel, selective 5-hydroxytryptamine (5-HT)4
receptor ligands in rat spatial navigation. Neuropharmacology 1997, 36
(4−5), 689−696.
(28) Morris, R. Developments of a water-maze procedure for
studying spatial learning in the rat. J. Neurosci. Methods 1984, 11 (1),
47−60.
(29) Obach, R. S. Prediction of human clearance of twenty-nine
drugs from hepatic microsomal intrinsic clearance data: an
examination of in vitro half-life approach and nonspecific binding to
microsomes. Drug Metab. Dispos. 1999, 27 (11), 1350−1359.
(30) Furstner, A.; Thiel, O. R.; Blanda, G. Asymmetric synthesis of
the fully functional macrolide core of salicylihalamide: remote control
of olefin geometry during RCM. Org. Lett. 2000, 2 (23), 3731−3734.
(31) Fensome, A.; Adams, W. R.; Adams, A. L.; Berrodin, T. J.;
Cohen, J.; Huselton, C.; Illenberger, A.; Kern, J. C.; Hudak, V. A.;
Marella, M. A.; Melenski, E. G.; McComas, C. C.; Mugford, C. A.;
Slayden, O. D.; Yudt, M.; Zhang, Z.; Zhang, P.; Zhu, Y.; Winneker, R.
C.; Wrobel, J. E. Design, synthesis, and SAR of new pyrrole-oxindole
progesterone receptor modulators leading to 5-(7-fluoro-3,3-dimethyl-
2-oxo-2,3-dihydro-1H-indol-5-yl)-1-methyl-1H-pyrrole-2-carbonitrile
(WAY-255348). J. Med. Chem. 2008, 51 (6), 1861−1873.
(32) Tapia, I.; Alonso-Cires, L.; Lopez-Tudanca, P. L.; Mosquera, R.;
Labeaga, L.; Innerarity, A.; Orjales, A. 2,3-Dihydro-2-oxo-1H-
benzimidazole-1-carboxamides with selective affinity for the 5-HT(4)
receptor: synthesis and structure−affinity and structure−activity
relationships of a new series of partial agonist and antagonist
derivatives. J. Med. Chem. 1999, 42 (15), 2870−2880.
(33) Animal use reference: Institute of Laboratory Animal Resources.
Guide for the Care and Use of Laboratory Animals; National Institutes of
Health Publication 85-23; National Academy Press: Washington, DC,
1996.
O
dx.doi.org/10.1021/jm300953p | J. Med. Chem. XXXX, XXX, XXX−XXX