Synthesis, structure-affinity relationships, and radiolabeling of selective high-affinity 5-HT4 receptor ligands as prospective imaging probes for positron emission tomography

Article abstract of DOI:10.1021/jm100668r

In a search for high-affinity receptor ligands that might serve for development as radioligands for the imaging of brain 5-HT₄ receptors in vivo with positron emission tomography (PET), structural modifications were made to the high-affinity 5-HT₄ antagonist (1-butylpiperidin-4-yl) methyl 8-amino-7-iodo-2,3-dihydrobenzo[b][1,4]dioxine-5-carboxylate (1, SB 207710). These modifications were made mainly on the aryl side of the ester bond to permit possible rapid labeling of the carboxylic acid component with a positron emitter, either carbon-11 (t_1/2 = 20.4 min) or fluorine-18 (t_1/2 = 109.7 min), and included (i) replacement of the iodine atom with a small substituent such as nitrile, methyl, or fluoro, (ii) methylation of the 8-amino group, (iii) opening of the dioxan ring, and (iv) alteration of the length of the N-alkyl goup. High-affinity ligands were discovered for recombinant human 5-HT₄ receptors with amenability to labeling with a positron emitter and potential for development as imaging probes. The ring-opened radioligand, (([methoxy-¹¹C]1-butylpiperidin-4-yl)methyl 4-amino-3-methoxybenzoate; [¹¹C]13), showed an especially favorable array of properties for future evaluation as a PET radioligand for brain 5-HT₄ receptors. This article not subject to U.S. Copyright. Published 2010 by the American Chemical Society.

Full text of DOI:10.1021/jm100668r

J. Med. Chem. 2010, 53, 7035–7047 7035
DOI: 10.1021/jm100668r
Synthesis, Structure-Affinity Relationships, and Radiolabeling of Selective High-Affinity
5-HT4 Receptor Ligands as Prospective Imaging Probes for Positron Emission Tomography
Rong Xu, Jinsoo Hong, Cheryl L. Morse, and Victor W. Pike*
Molecular Imaging Branch, National Institute of Mental Health, National Institutes of Health, Building 10, Room B3 C346A,
10 Center Drive, Bethesda, Maryland 20892
Received June 2, 2010
In a search for high-affinity receptor ligands that might serve for development as radioligands for the
imaging of brain 5-HT4 receptors in vivo with positron emission tomography (PET), structural modifica-
tions were made to the high-affinity 5-HT4 antagonist (1-butylpiperidin-4-yl)methyl 8-amino-7-iodo-2,3-
dihydrobenzo[b][1,4]dioxine-5-carboxylate (1, SB 207710). These modifications were made mainly on the
aryl side of the esterbond to permit possible rapid labeling of the carboxylic acid component with a positron
emitter, either carbon-11 (t1/2 = 20.4 min) or fluorine-18 (t1/2 = 109.7 min), and included (i) replacement
of the iodine atom with a small substituent such as nitrile, methyl, or fluoro, (ii) methylation of the 8-amino
group, (iii) opening of the dioxan ring, and (iv) alteration of the length of the N-alkyl goup. High-affinity
ligands were discovered for recombinant human 5-HT4 receptors with amenability to labeling with
a positron emitter and potential for development as imaging probes. The ring-opened radioligand,
(([methoxy-11C]1-butylpiperidin-4-yl)methyl 4-amino-3-methoxybenzoate; [11C]13), showed an especially
favorable array of properties for future evaluation as a PET radioligand for brain 5-HT4 receptors.
Introduction
in vivo imaging of some of the 5-HT receptors, especially the
5-HT1A, 5-HT2A, and5-HT1B subtypes, radioligandsfor some
other subtypes are not yet well explored (e.g., 5-HT4).
The 5-HT4 receptor is a well-characterized G-protein-
coupled receptor that exists abundantly in brain, especially
in limbic and striatonigral regions.3 This receptor population
is implicated in dopamine, serotonin, and acetylcholine re-
lease and possibly plays a significant role in normal cognition,
learning, and memory. 5-HT4 receptors have also been im-
plicated in neuropsychiatric disorders, such as Alzheimer’s
disease, anxiety, and depression.4-6
A reason for the relatively slow development of 5-HT4 PET
radioligands is the wide array of properties that must ideally
be found in any candidate. These include high target receptor
affinity, high selectivity, generally moderate lipophilicity,7
appropriate intrinsic activity, ability to cross the blood-brain
barrier,8 absence of troublesome radiometabolites,8 and
amenability to labeling with a suitable radioisotope.9,10
Suitable radioisotopes are generally short-lived carbon-11
(t1/2=20.4 min) or fluorine-18 (t1/2=109.7 min) for PET and
iodine-123 (t1/2=13.2 h) for SPECT.
Suitably effective radioligands, when applied with molecu-
larimagingtechniques, suchas positronemissiontomography
(PETa) or single photon emission computed tomography
(SPECT), provide a unique means for measuring brain neuro-
transmitter receptor concentrations in living subjects, and
therefore, they constitute important clinical research tools.1
Such radioligands may also be used with imaging techniques
to assess the receptor binding of unlabeled ligands in vivo,
whether exogenous (e.g., a developmental or therapeutic drug
or a substance of abuse) or endogenous (the neurotransmitter),
and hence, they are also useful for drug discovery and
development.2
Serotonin (5-hydroxytryptamine, 5-HT) is an important
neurotransmitter that is known to act on at least 14 receptors
in seven major subclasses. Many of these receptors in brain
have been implicated in neuropsychiatric disorders, and
hence, they have become targets for deep biomedical investi-
gation and also for drug discovery and development pro-
grams. Although many effective radioligands exist for the
(1-Butylpiperidin-4-yl)methyl 8-amino-7-iodo-2,3-dihy-
drobenzo[b][1,4]dioxine-5-carboxylate (1, SB 207710; Chart 1)
is an exceptionally high-affinity 5-HT4 receptor antagonist.11,12
The use of [123I]1 with SPECT provided the first demonstration
of 5-HT4 receptor imaging in primate brain in vivo.13 Some
analogues of 1, labeled with either carbon-11 or fluorine-18 in
the terminal N-alkyl group, have been prepared, and one of
these, [11C]2 ([11C]SB 207145, Chart 1) has shown promise for
PET imaging in animal14 and human subjects.15,16 This radi-
oligand is an ester and rapidly metabolized in vivo by hydro-
lysis. Here, we aimed to develop alternative radioligands for
5-HT4 receptor imaging with PET, again based on 1, in which
*To whom correspondence should be addressed. Phone: 301-594-
5986. Fax: 301-480-5112. E-mail: pikev@mail.nih.gov.
a Abbreviations: BZP, peripheral benzodiazepine receptor; Cbz, benzyl-
oxycarbonyl; CDI, N,N0-carbonyldiimidazole; D, dopamine; DAT, dopa-
mine transporter; DOR, δ opiate receptor; DMAP, 4-(dimethylamino)-
pyridine; DMF, dimethylformamide; EOS, end of synthesis; H, histamine;
K 2.2.2, 4,7,13,18-tetraoxo-1,10-diazabicyclo[8,8,8]hexacosane; M, mus-
carinic; MOR, μ opiate receptor; NBS, N-bromosuccinimide; NCA, no-
carrier-added; NET, noradrenalin transporter; NMP, N-methyl-2-pyrro-
lidinone; NXS, N-halosuccinimide (X = halo atom); PET, positron
emission tomography; rt, room temperature; RCY, decay-corrected radio-
chemical yield; SERT, serotonin transporter; SPECT, single photon emis-
sion computed tomography; THF, tetrahydrofuran; TMS, trimethylsilyl;
TPP, triphenylphosphine; 5-HT, serotonin.
This article not subject to U.S. Copyright. Published 2010 by the American Chemical Society
Published on Web 09/02/2010
pubs.acs.org/jmc

7036 Journal of Medicinal Chemistry, 2010, Vol. 53, No. 19
Xu et al.
Chart 1. Current Analogues of 1 for 5-HT4 Receptor Imaging
in Vivo
Scheme 1. Synthesis of Ring Methyl, Halo, and Nitrile Ligands
and N-Methyl Ligands 5-11a
the radiolabel is located on the carbonyl side of the ester group
to avoid potential issues from possibly brain-penetrant radio-
metabolites and to permit control of lipophilicity through
adjustment of N-alkyl chain length. The ester hydrolysis of
such radioligands outside brain would produce carboxylic acids
as radiometabolites; these radiometabolites would be expected
to have poor entry into brain and therefore would not be
expected to interfere with the measurement of brain receptor
density with the parent radioligand.8 The 5-HT4 radioligand
[11C]2 has been shown to be amenable to quantification in
human brain,16 implying that ester hydrolysis to generate
benzoates as radiometabolites within brain should not be a
major concern for ligands related to 1. Control of lipophilicity
can be important for adjusting properties that may have
an impact on the potential success of a radioligand, including
its plasma free fraction, ability to penetrate the blood-brain
barrier (permeability parameter),17 and susceptibility to meta-
bolism.8-10 Our synthetic strategies included replacement of the
aryl halo group and/or alkylation of the aryl amino group,
opening of the dioxan ring, and manipulation of the N-alkyl
chain length. As a result, several new high-affinity ligands were
discovered and radiolabeled as prospective PET radioligands
for 5-HT4 receptors. The 3-methoxy compound, [11C]13, was
found to have an especially favorable array of properties for
further evaluation as a new 5-HT4 receptor PET radioligand.
a Reagents and conditions: (i) NXS (X = Cl, Br, or I), AcOH;
(ii) KCN, CuI, 1,10-phenanthroline, DMF, 42 h, 110 °C; (iii) (1) para-
formaldehyde, EtOH, 20 h, 60 °C; (2) NaBH4; (iv) Me4Sn, CuI, Pd2-
(dba)3, TPP, NMP, 70 °C, 48 h.
Scheme 2. Synthesis of Ring Methoxy Ligands 12 and 13a
Results
Chemistry. We considered several approaches for modify-
ing the structure of 1 to allow carbon-11 to be introduced
as a radiolabel into the structure on the carbonyl side of the
ester bond, using readily accessible labeling agents such as
[11C]methyl iodide,18 [11C]methyl triflate,19 or [11C]cyanide
ion.20 These strategies included (i) replacing the iodo group
in 1 with a group of similar or smaller size, namely, nitrile, as
in 5 and 11, or methyl as in 7 (Scheme 1), (ii) N-methylation,
as in 6 and 8-11 (Scheme 1), and (iii) replacement of the
dioxan ring by a single O-methyl group, as in 12 and 13
(Scheme 2). Ligand 5 was obtained in low but useful yield
(19%) by treating 1 with potassium cyanide. The N-methyl
compound 6 was obtained in good yield (60%) by treat-
ing the primary arylamine 4 with paraformaldehyde and then
sodium borohydride. Treatment of 6 with N-halosuccini-
mides (NXS, X = Cl, Br, or I) gave the respective halo
derivatives 8-10 in moderate yields (30-56%). Ligand 7
was obtained from 1 in moderate yield (44%) through
Pd-catalyzed methylation with tetramethyltin. Ligand 11
was similarly obtained from ligand 10 but in much lower
yield (7%). The ring O-methyl derivatives, 12 and 13, were
obtained in moderate (32%) and low (8%) overall yields,
respectively, from the corresponding methyl esters 12a and
13a in two steps, namely, N-protection and trans-esterifica-
tion (Scheme 2). Analogues of 13, in which either the ester
a Reagents and conditions: (i) CbzCl, NaHCO3, overnight; (ii) (1)
lithium (l-butylpiperidin-4-yl)methanolate, THF, 0 °C, then rt over-
night; (2) H2,Pd/C.
group was replaced with an amido group, as in 14, or in
which the length of the N-alkyl group was altered to vary
lipophilicity, as in 15 and 16, were also prepared in two steps
and moderate overall yields (30-69%) from 4-amino-3-
methoxybenzoic acid (Scheme 3).
For the purpose of creating a radioligand that might be
labeled with fluorine-18, we succeeded in preparing 17, a
7-fluoro analogue of 1, in six steps from 5-fluoro-2-hydroxy-
benzoic acid (Scheme 4). All steps proceeded in >75% yield
except the final two-stage esterification, involving Cbz pro-
tection (34%) and trans-esterification (59%). Also, two
fluoroalkoxy analogues of 13, the fluoromethoxy compound
18 and the 2-fluoroethoxy compound 19, were prepared in
two steps from methyl 4-amino-3-hydroxybenzoate in mod-
erate overall yields of 37% and 20%, respectively (Scheme 5).
Potential precursors for radiolabeling were also synthe-
sized, including the phenol precursor 20 by demethylation of

Article
Journal of Medicinal Chemistry, 2010, Vol. 53, No. 19 7037
13 (42% yield) (Scheme 6), and the nitro compound 21
in three steps from 8-amino-7-nitro-2,3-dihydrobenzo[b]-
[1,4]dioxine carboxylic acid methyl ester (21a) in 42% overall
yield (Scheme 7).
the N-alkyl chain length in 13 gave ligands of 5- and 7-fold
lower affinity, respectively.
Replacement of the ester group in 13 with an amido group
drastically reduced binding affinity.
Pharmacological Assays and Screen. Assay of compounds
1, 5, 8, 13, 15, and 17-19 for binding to 5-HT4 receptors in
guinea pig striatal membranes showed all these compounds
to have subnanomolar Ki values (Table 1). However, assay of
the same ligand set against human recombinant 5-HT4
receptors (h5-HT4), expressed in HEK293T cells, revealed
a greater variation in Ki, with only some of the compounds
(1, 8, 13, and 17) showing low nanomolar values. This assay
was applied to the full range of new ligands and revealed
interesting structure-activity information (Table 1).
Replacement of the iodine atom in 1 with a nitrile, methyl,
or nitro group resulted in approximately 15-, 3-, and 3.5-fold
decrease in binding affinity, respectively, whereas replace-
ment with fluorine retained binding affinity and replacement
with hydrogen slightly increased affinity. N-Methylation of 1
resulted in about a 4-fold reduction in binding affinity, while
the N-methylation of the corresponding nitrile, 5, resulted in
only a marginal reduction in binding affinity. The N-methyl
8-chloro analogue 8 showed a binding affinity comparable to
that of 1.
Replacement of the methoxy group in 13 with fluoro-
methoxy or 2-fluoroethoxy led to 7- and 5-fold reduction in
binding affinity, respectively.
A selection of the compounds was also assessed for
intrinsic activity (Table 1). Most of these ligands were found
to be quite potent partial or full inverse agonists with pEC50
values in the range 7.70-8.23. The 2-methoxy ligand 12 was
found to be a potent full agonist with a pEC50 value of 9.75.
In stark contrast, the high-affinity regional isomer 13 was
found to be an antagonist. The arylmethyl analogue of 1 was
also found to be an antagonist, as was the previously known
ligand, 4.
Ligand 13 showed greater than 2000-fold selectivity
for h5-HT4 receptors versus 12 other h5-HT receptors
and binding sites (Table 2). Ligands 8 and 17 were not quite
as selective among h5-HT receptors and binding sites.
Scheme 5. Synthesis of Fluoroalkoxy Ligands 18 and 19a
Replacement of the dioxan ring in 1 with a 2-methoxy
group gave ligand 12 with much reduced affinity, whereas
replacement with a 3-methoxy group gave ligand 13 with
affinity comparable to that of 1. Shortening or lengthening of
Scheme 3. Synthesis of Amide (14) and N-Alkyl Analogues
(15 and 16) of 13a
a Reagents and conditions: (i) F(CH2)nC1, Cs2CO3, DMF, rt;
(ii) lithium (l-butylpiperidin-4-yl)methanolate, THF, 0 °C, then rt over-
night.
Scheme 6. Synthesis of Phenol Precursor 20a
a Reagents and conditions: (i) CDI; (ii) (l-butylpiperidin-4-
yl)methanamine or lithium (1-propylpiperidin-4-yl)methanolate or
lithium (l-pentylpiperidin-4-yl)methanolate.
a Reagents and conditions: (i) AlCl3, Nal, MeCN, reflux, overnight.
Scheme 4. Synthesis of Ring Fluoro Ligand 17a
a Reagents and conditions: (i) NBS, AcOH; (ii) NaOH, CuSO4; (iii) (1) TMSCHN2; (2) Cs2CO3, 1,2-dibromoethane, DMF, 80 °C, 16 h; (iv) HNO3,
-50 °C, 16 min; (v) 10% Pd/C, HCOOK, MeOH, 80 °C, 2 h; (vi) (1) CbzCl, aqueous NaHCO3; (2) CDI, MeCN, rt, 2 h; (3) lithium (l-butylpiperidin-4-yl)-
methanolate, THF, 0 °C, then rt overnight; (4) H2, Pd/C.

7038 Journal of Medicinal Chemistry, 2010, Vol. 53, No. 19
Scheme 7. Synthesis of Nitro Analogue 21a
Xu et al.
a Reagents and conditions: (i) (Boc)2O, DMAP, 50 °C, 1 h; (ii) lithium (l-butylpiperidin-4-yl)methanolate, THF, 0 °C, then rt overnight; (iii) 4 M HC1
in dioxane, overnight.
Table 1. Binding Affinities (1/Ki),a Efficacies, and cLogD of New 5-HT4 Ligands
compd
R1
R2
R3
R4
X
n
5-HT4
Ki (nM)a
h5-HT4
Ki (nM)b
efficacyc,d
pEC50
7.9
cLogDe
d
1
I
H
OCH2
OCH2
OCH2
OCH2
OCH2
OCH2
OCH2
OCH2
OCH2
H
OCH2
OCH2
OCH2
OCH2
OCH2
OCH2
OCH2
OCH2
OCH2
OMe
H
O
O
O
O
O
O
O
O
O
O
O
N
O
O
O
O
O
O
2
2
2
2
2
2
2
2
2
2
2
2
1
3
2
2
2
2
0.20
2.2 ( 0.3
1.4 ( 0.2
33 ( 5
9.1 ( 0.7
6.3 ( 1
2.0 ( 0.2
4.5 ( 0.4
7.4 ( 0.5
37 ( 5
inv ag
antagonist
n.m.
3.20
1.74
2.59
2.37
2.20
3.27
3.36
3.41
2.52
1.77
2.07
0.16
1.54
2.60
2.42
1.97
2.30
3.04
4
H
H
5
CN
H
H
0.45
n.m.
8.05
6
Me
H
inv.ag
antagonist
inv ag
inv ag
inv ag
inv ag
ag.
7
Me
Cl
Br
I
8
Me
Me
Me
Me
H
0.577
7.85
8.0
9
10
11
12
13
14
15
16
17
18
19
21
7.7
7.8
CN
H
50 ( 7
9.75
H
H
OMe
OMe
0.738
0.901
2.4 ( 0.3
8664 ( 857
17 ( 1
12 ( 1
2.1 ( 0.3
17 ( 1
antagonist
n.m.
n.m.
H
H
H
n.m.
n.m.
n.m.
8.23
n.m.
n.m.
n.m.
H
H
OMe
OMe
H
H
H
H
n.m.
F
H
OCH2
OCH2F
O(CH2)2F
OCH2
OCH2
H
0.334
0.334
0.215
inv ag
n.m.
n.m.
H
H
H
NO2
H
H
OCH2
11 ( 1
7.8 ( 0.9
H
n.m.
a For guinea pig striatal membrane 5-HT4 receptors. Values are averages of triplicate measurements. b For h5-HT4 receptors. Binding assay results are
averages of triplicate measurements. c In agonist/inverse agonist assay: ag = agonist; inv ag = inverse agonist. d n.m. = not measured. e cLogD was
calculated with ACD software. Estimated errors are approximately (1.0.
Table 2. Binding Affinities of 5-HT4 Ligands 8, 13, and 17 for Other
Ligand 13 also exhibited high selectivity for h5-HT4 receptors
5-HT Receptors/Binding Sites
versus a wide range of nonserotonergic receptors and binding
sites (Table 3). For 13, the lowest selectivities for h5-HT4
binding affinity (Ki) (nM)
receptors were versus σ1 and σ2 receptors, namely, 88- and
62-fold, respectively. Ligands 8 and 17 were also generally
selective for h5-HT4 receptors versus other receptors and
binding sites except that they displayed much lower selectivities
versus σ1 and especially σ2 receptors (Table 3). Ligand 8 also
showed a quite low 22-fold selectivity versus D4 receptors
(Table 3).
binding site or receptor
ligand 8
2.0
ligand 13
ligand 17
h5-HT4
2.4
>10000
>10000
5686
2.1
>10000
>10000
2856
h5-HT1A
h5-HT1B
h5-HT1D
h5-HT1E
h5-HT2A
h5-HT2B
h5-HT2C
h5-HT3
300
820
1304
>10000
1776
66
>10000
>10000
2861
>10000
>10000
168
Computation of cLogD. The cLogD values of ligands
ranged from 1.74 to 3.27 (Table 1).
685
>10000
3511
8363
1072
8589
489
>10000
>10000
>10000
>10000
>10000
Syntheses of Radioligands. Some ligands, namely, ligands
5, 8, 13, 17, and 18, were selected to test their amenability for
rapid labeling with a positron emitter, either carbon-11 or
fluorine-18. Ligand [11C]5, a [11C]nitrile, was produced in an
average of 26% decay-corrected radiochemical yield (RCY)
by Pd-mediated exchange of the iodine atom with no-carrier-
added (NCA) [11C]cyanide ion in the presence of potassium
h5-HT5A
h5-HT6
h5-HT7
>10000
>10000
>10000
>10000
5702
9006
hSERT
The lowest selectivity for 8 was 33-fold versus h5-HT2B recep-
tors and for 17, 80-fold versus the same receptors (Table 2).

Article
Journal of Medicinal Chemistry, 2010, Vol. 53, No. 19 7039
Scheme 9. Radiosynthesis of [11C]8a
Table 3. Binding Affinities of 5-HT4 Ligands 8, 13, and 17 for Non-
Serotonergic Receptors and Binding Sites
binding affinity (Ki) (nM)
binding site or receptor
ligand 8
ligand 13
ligand 17
hR1A
hR1B
hR1D
hR2A
hR2B
hR2C
522
>10000
2276
>10000
4228
87.6
9076
>10000
>10000
>10000
989
>10000
>10000
773
2239
671.2
556
593
a Reagents, conditions and yield: (i) [11C]Mel, Li3N, DMF, rt, ultra-
sound, 10 min. RCY = 11%.
β1 (rat)
β2 (rat)
β3 (rat)
σ1 (rat)
σ2 (rat)
BZP (rat brain site)
hDAT
>10000
>10000
>10000
60
4407
>10000
>10000
>10000
55
>10000
>10000
211
Scheme 10. Radiosyntheses of [11C]13 and [18F]18a
8
148
12.9
>10000
2020
>10000
>10000
>10000
>10000
>10000
6272
>10000
>10000
>10000
>10000
615
hDOR
>10000
2291
220
D1 (rat)
D2 (rat)
D3 (rat)
D4 (rat)
hD5
336
43
1011
316
1272
972
>10000
2727
>10000
>10000
>10000
>10000
1649
>10000
>10000
>10000
>10000
5038
a Reagents, conditions and yields: (i) [11C]MeI, DMF, 1.0 M (n-Bu)4-
NOH in MeOH, 80 °C, 5 min, RCY = 36%; (ii) [11C]MeOTf, MeCN,
0.5 M NaOH, heat, 5 min, RCY = 27%; (iii) [18F]FCD2Br, MeCN, 0.5
M NaOH, 100 °C, 15 min. RCY = 13% from [18F]fluoride ion.
hGABAA
H1 (guinea pig)
H2 (guinea pig)
H3 (guinea pig)
H4 (guinea pig)
hKOR
>10000
>10000
>10000
>10000
>10000
>10000
433
>10000
>10000
>10000
>10000
>10000
1095
>10000
>10000
>10000
>10000
>10000
>10000
>10000
>10000
>10000
Table 5. RCYs of [18F]18 under Various Conditions
hMOR
hM1
hM2
solvent
base
T (°C)
time (min)
RCY a(%)
MeCN
DMF
MeCN
0.5 M NaOH
K2CO3, 18-crown-6
0.5 M NaOH
80
110
100
5
10
15
1
7
hM3
hM4
823
760
2579
4058
13
hM5
hNET
972
3226
a Overall starting from [11C]fluoride ion.
>10000
iodide or [11C]methyl triflate in either 36% or 27% RCY,
respectively (Scheme 9). An experiment based on 13C/11C
colabeling, followed by 13C NMR,22 confirmed the position
of the radiolabel. This radioligand was obtained in moder-
ately high specific radioactivity, namely, 2848 mCi/μmol
from [11C]methyl iodide and 2517 mCi/μmol from [11C]-
methyl triflate at the end of radiosynthesis.
Scheme 8. Radiosynthesis of [11C]5a
All attempts to prepare [18F]17 by substitution of the nitro
group in precursor 21 or the N-Boc-protected analogue 21c
with cyclotron-produced [18F]fluoride ion were unsuccessful.
However, the [18F]fluorodideuteromethoxy ligand, NCA
[18F]18 was obtained in 13% RCY from [18F]fluoride ion
by treating the phenol 21 with derived [18F]fluorobromo-
methane-d2 in acetonitrile with NaOH as base at 100 °C
(Scheme 10). Other tested conditions gave inferior RCYs
(Table 5).
a Reagents, conditions and yield: (i) [11C]HCN, Pd(PPh3)4, K2CO3,
THF, K 2.2.2, 80 °C, 5 min, RCY = 26%.
Table 4. RCYs for [11C]5 under Various Conditions
solvent
base
KH2PO4
KH2PO4
RCY a(%)
n
THF
DMSO
THF
2
10
26
1
2
3
Discussion
K2CO3, K 2.2.2
a From [11C]HCN.
In this study, variation of the structure of 1 led to several
new high-affinity ligands for guinea pig and h5-HT4, recep-
tors, some of which proved amenable to labeling with a
positron emitter to provide candidate radioligands for ima-
ging brain 5-HT4 receptors with PET. Interestingly, all seven
new ligands (5, 8, 13, 15, 17-19) that were tested for binding
to guinea pig 5-HT4 receptors showed subnanomolar affinity
within a narrow range (Ki = 0.22-0.90 nM), while the same
set of ligands assayed against human recombinant receptors
(h5-HT4) showed somewhat lower binding affinity across
a wider range (Ki = 2.2-33 nM) (Table 1). This species
carbonate-K 2.2.2 in THF (Scheme 8). Use of potassium
dihydrogen phosphate as base in THF or DMSO21 gave
much lower RCY (Table 4). The N-[11C]methyl ligand,
[11C]8, was obtained in 11% RCY by 11C-methylation of 1
with NCA [11C]methyl iodide in the presence of the solid
base, Li3N, under the influence of ultrasound (Scheme 9).
Use of Li2O as solid base resulted in a lower RCY (7%) of
[11C]8. The O-[11C]methyl ligand [11C]13 was obtained from
the phenol 20 by methylation with either NCA [11C]methyl

7040 Journal of Medicinal Chemistry, 2010, Vol. 53, No. 19
Xu et al.
difference was unexpected, since 5-HT4 receptor ligands from
several other structural classes do not show such major
differences in binding affinity between guinea pig 5-HT4 and
h5-HT4 receptors.23 Also, the affinities of 1, the prototypic
ligand for our series, were previously reported to be quite
similar for human colon and guinea pig 5-HT4 receptors.12
Because any successfully developed PET radioligand would
ultimately be used for imaging 5-HT4 receptors in human
brain, all new ligands from this study were assayed against h5-
HT4 receptors (Table 1).
Replacement of the iodine atom in 1 with a methyl or nitro
substituent reduced h5-HT4 receptor binding affinity by less
than 1 order of magnitude, whereas replacement with a fluoro
substituent caused negligible change, and replacement with
hydrogen slightly improved affinity (Table 1). Replacement
with a nitrile or nitro group caused a 15- or 4-fold reduction in
affinity, respectively. Therefore, the h5-HT4 receptor was quite
tolerant of a small substituent ortho to the 8-amino group in 1,
irrespective of the electronic influence of the substituent.
Although a primary arylamino group is a frequent consti-
tuent of ligands reported for 5-HT4 receptors,24 the effect of
N-alkylation of this group on binding affinity had not been
examined previously. We found that the N-methylations of 1,
4, and 5 caused only small reductions of binding affinity of
about 3-, 6.5-, and 1.1-fold, respectively. The N-methylated 4
recovered its affinity to 2.0 nM after chlorination in the ortho
position as seen in 8. Another prepared N-methyl ligand, 9,
also exhibited low nanomolar affinity. Thus, the h5-HT4
receptor readily tolerates the secondary 8-N-methylamino
substituent in this structural class of ligand.
SB204070 (3) is the 7-chloro analogue of 1. Thenon-dioxan
2-methoxy analogue of 3 has nanomolar affinity for guinea
pig 5-HT4 receptors.25 Here, opening of the dioxan ring in the
proto analogue of 1, namely, ligand 4, by removal of either
OCH2 group gave the two O-methyl compounds 12 and 13.
Binding affinity was reduced 36-fold in the 2-methoxy com-
pound 12 but only 1.7-fold in the 3-methoxy compound 13.
Replacement of the 3-methoxy group in 13 with fluoro-
methoxy or 2-fluoroethoxy reduced affinity by 7- and 4.6-
fold, respectively. Shortening or lengthening of the N-alkyl
chain in 13 similarly reduced binding affinity by 7- and 5-fold,
respectively.
We noted that the conversion of the ester function in the
benzodioxan 3 into an amido function has negligible effect on
binding affinity (1/IC50)24 and in the ring-opened 2-methoxy
analogue causes only a 10-fold reduction in binding affinity.25
We therefore considered replacing the ester function in the
new radioligands with an amide function in order to confer
greater resistance to hydrolysis in vivo. However, replacement
of the ester group in the 3-methyl ether 13 with an amido
group dramatically reduced affinity by several-thousand-fold.
The maximal receptor-specific signal to be expected from
the use of a radioligand with PET is related to the binding
potential (BP), expressed as Bmax/KD, where Bmax is the local
concentration of receptors and KD is the equilibrium dissocia-
tion constant of radioligand from the receptor.9,10 Therefore,
high affinity (1/KD or as a surrogate measure, 1/Ki) is a key
parameter in determining whether a particular radio-
ligand can be successful. Previously, it has been found that
moderately sizable receptor-specific signals can be obtained
in minipig,14 monkey,13 and human16 subjects in vivo with
5-HT4 receptor radioligands having KD values in the low or
subnanomolar range, such as [123I]1 and [11C]2. Our new
ligands 8, 13, and 17 showed Ki values less than or comparable
with that of 1 for h5-HT4 receptors, and they therefore met the
high-affinity criterion for development as PET radioligands.
Since target selectivity, intrinsic activity,26 and lipophilicity
also bear on the likely success of candidate PET radio-
ligands,9,10 these three ligands were also assessed for these
parameters. For comparison, lipophilicity was also computed
for the other synthesized ligands, and intrinsic activity was
also assessed for 1, 4, 6-13, and 17.
Ligands 1, 8, and 17 appeared to be similarly potent inverse
agonists in the GloSensor assay, whereas ligands 4 and 13
wereantagonists(inactive) in the same assay (Table1). Ligand
1 is widely considered to be an antagonist at 5-HT4 recep-
tors.11,12 However, it is noted that ligands characterized as
competitive antagonists often express inverse agonism in
assays where constituitive receptor activity is present.27 This
fact likely explains why 1 appears to be an inverse agonist
in the Glo-sensor assay. Ligand 7 was also inactive in the
Glo-sensor assay, whereas ligands 6 and 9-11 were inverse
agonists. Remarkably, ligand 12, the close methoxy positional
isomer of the antagonist 13, was a potent agonist with quite
high binding affinity.
Ligands 8, 13, and 17 showed generally high selectivity for
binding to h5-HT4 receptors versus binding to other h5-HT
receptors and binding sites (Table 2). The lowest selectivity
was 33-fold for ligand 8 versus h5-HT2B receptors. Although
there is strong evidence that 5-HT2B receptors exist in brain,
their distribution and density in human brain remain un-
known and are suspected to be low.28,29 Thus, the low-affinity
binding of8 to5-HT2B receptorsin humanbrainin vivowould
likely be insignificant compared to its high-affinity binding to
relatively abundant 5-HT4 receptors. Therefore, this off-
target binding would probably not be a serious impediment
to the development of 8 as a PET radioligand. Ligands 8, 13,
and 17 showed only low affinity for a wide range of other
receptors and binding sites except for σ1 and σ2 receptors in
the cases of 8 and 17 and for D4 receptors in the case of 8
(Table 3). Both σ receptors are present in brain,30 and indeed
σ1 receptors may be imaged with PET in vivo with moderately
high-affinity radioligands, such as [11C]SA4503 (Ki = 17.4
nM).30 Thus, the affinities of ligands 8 (Ki = 60 nM) and 17
(Ki = 55 nM) for σ1 receptors may be insufficiently low for
successful 5-HT4 receptor PET imaging. Imaging data suggest
that the density of σ2 receptors in human brain is much lower
than that of σ1 receptors.30 Hence, it is uncertain whether the
high-affinities of 8 (Ki = 8 nM) and 17 (Ki = 13 nM) for σ2
receptors may be problematic for their development as PET
5-HT4 receptor radioligands. The moderate affinity of 8 for
D4 receptors is unlikely to be problematic, since D4 receptors
only exist in very low density in human brain.31,32 As a
candidate for development as a PET radioligand, 13 expressed
excellent overall 5-HT4 receptor selectivity, with greater than
60-fold selectivity against the full range of tested receptors and
binding sites (Table 3).
Ligands 8, 13, and 17 had computed lipophilicities (cLogD
values) between 2.0 and 3.3, and these are within the range
considered desirable for achieving adequate brain entry from
blood without incurring excessive nonspecific binding.7-10
Given the array of favorable properties expressed by
ligands 8, 13, and 17, it appeared especially attractive to
attempt to label these ligands with a positron emitter, either
carbon-11 or fluorine-18, to give radioligands that might be
tested and evaluated in vivo with PET.
Variousconditionswereattemptedforthe labelingof8 with
carbon-11 in itsN-methylgroup, all based on 11C-methylation

Article
Journal of Medicinal Chemistry, 2010, Vol. 53, No. 19 7041
of the primary arylamine 3. However, 3, as is quite usual for
primary arylamines, proved stubbornly reactive toward
[11C]methyl iodide or [11C]methyl triflate under conventional
conditions. Recently, we have developed methodology for
labeling such precursors in the presence of a strong inorganic
baseundertheinfluence of ultrasoundagitation. Thismethod,
using DMF as solvent and Li3N as solid base, gave [11C]8 in
low but still useful RCY (11%) from [11C]methyl iodide after
HPLC separation (Scheme 9).
By contrast to 8, the methoxy compound 13 was readily
labeled by reaction of the phenol 20 with [11C]methyl iodide or
[11C]methyl triflate under basic conditions. No protection of
the anilino nitrogen was necessary because of its low reactivity
toward these labeling agents. The selectivity of the 11C-
methylation reaction for the phenol oxygen versus the anilino
nitrogen was confirmed through a 11C/13C colabeling experi-
ment and subsequent 13C NMR spectroscopy.22 [11C]13 was
readily separated by HPLC and was shown to be radio-
chemically pure by analytical HPLC.
The labeling of the fluoro compound 17 was first attempted
through treatment of the N-Boc-protected 21 with NCA
[18F]fluoride ion because the presence of unprotected amino
group and the electron-rich nature of the aryl ring were
expected to oppose facile aromatic nucleophilic substi-
tution.33 The N-Boc group was found to be unstable when
the temperature exceeded 120 °C. Nitro substitution occurred
in deprotected 21 at high temperature (200 °C) but not with-
out concomitant ester hydrolysis. The direct labeling of 21
also only gave [18F]fluoride ion substituted benzoic acid at
similarly high temperature. Another attempt based on halo-
gen exchange in methyl 7-iodo-8-nitro-2,3-dihydrobenzo[b]-
[1,4]dioxine-5-carboxylate precursor resulted in the replace-
ment of the nitro group instead of the iodo group by the
[18F]fluoride ion. We considered other strategies for 18F-
labeling, including the production of a diaryliodonium salt
precursor for radiofluorination.34 However, we were unable
to synthesize a suitable iodonium salt via a metalated-ring
intermediate, since the aryl ring resisted clean stannylation,
boronation, or mercuration by conventional reagents and
methods. Curiously, the n-tributylstannyl analogue of 1 is
known13 but its synthesis, as far as we can ascertain, has never
been published. Therefore, the radiosynthesis of [18F]17 re-
mains a major challenge.
radioligand for the study of 5-HT4 receptors in vivo with PET.
This radioligand is currently under evaluation in monkey, and
findings will be published elsewhere.
Experimental Section
Materials. Methyl 3-methoxy-4-nitrobenzoate, 4-(amino-
methyl)-1-(n-butyl)piperidine, and tetrakis(triphenylphosphine)-
palladium(0) were purchased from Alfa Aesar (Ward Hill, MA).
Chlorofluoromethane and 1-chloro-2-fluoroethane were pur-
chased from SynQuest (Alachua, FL). Other chemicals were
purchased from Aldrich Chemical Co. (Milwaukee, WI) and
used as received. Compound 3 (SB 204070, 1-butylpiperidin-4-yl)-
methyl 8-amino-7-chloro-2,3-dihydrobenzo[b][1,4]dioxane-5-
carboxylate) and the des-chloro analogue 4 were synthesized from
2,3-dihydrobenzodioxin by modified literature procedures.25,38,39
1
General Methods. H (400.13 MHz), 13C (100.62 MHz), and
19F (376.46 MHz) NMR spectra were recorded at room tem-
perature on an Avance-400spectrometer (Bruker, Billerica, MA).
Chemicalshiftsare reportedinδ units(ppm) downfield relative to
the chemical shift for tetramethylsilane. Abbreviations s, d, t, dd,
dt, and bs denote singlet, doublet, triplet, doublet of doublet,
doublet of triplet, and broad singlet. Thin layer chromatography
was performed with POLYGRAM SIL G/UV254 layers (0.2 mm
silica gel with fluorescent indicator; Grace Davison Discovery
Sciences; Deerfield, IL); compounds were visualized under UV
light (λ = 254 nm).
High resolution mass spectra were acquired from the Mass
Spectrometry Laboratory, University of Illinois at Urbana;
Champaign (Urbana, IL) under electron ionization conditions
using a double-focusing high-resolution mass spectrometer
(Autospec, Micromass Inc.) with samples introduced through
a direct insertion probe.
LC-MS analyses of synthesized compounds were performed
on an LCQ Deca model instrument (Thermo Fisher Scientific
Corp.; Waltham, MA). A gradient or isocratic LC analysis of
sample was carried out with binary solvents (A/B, 150 μL/min)
composed of water-methanol-acetic acid (90:10:0.5 by volume)
(A) and methanol-acetic acid (100:0.5, v/v) (B) on a Luna C18
column (3 μm, 50 mm ꢀ 2 mm; Phenomenex; Torrance, CA).
Following electrospray ionization of the column effluent, ions
m/z 150-750 were acquired.
Melting points were measured with a Mel-Temp manual appa-
ratus (Electrothermal, Fisher Scientific) and were uncorrected.
γ-Radioactivity from 11C and 18F was measured using a cali-
brated dose calibrator (Atomlab 300, Biodex Medical Systems).
Radioactivity measurements were corrected for physical decay.
All radiochemistry was performed in lead-shielded hot cells for
personnel protection from radiation.
Radioactive products were separated by HPLC on a Gemini or
Gemini-NX C18 column (5 μm, 10 mm ꢀ 250 mm; Phenomenex)
and eluted with 10 mM NH4OH-MeCN or 100 mM
HCOONH4-MeCN at the stated composition and flow rate.
Eluates were monitored for radioactivity (pin diode detector,
Bioscan) and absorbance at 294 nm (System Gold 166 detector,
Beckman).
The radiolabeling of the relatively lower affinity ligands 5
and 18 with carbon-11 and fluorine-18, respectively, was
shown to be feasible. Thus, [11C]5 was obtained by palla-
dium-mediated 11C-cyanation of 1 and [18F]18 by 18F-fluoro-
methylation of the phenol 20 with [18F]fluoromethyl-d2 bro-
mide. Deuterium was incorporated into [18F]18 to provide for
greater resistance to defluorination in vivo.35-37
Conclusions
The purity of each new nonradioactive compound was assessed
by reverse phase HPLC under the conditions tabulated in Sup-
porting Information. Each compound was shown to have a
chemical purity of >98%. Radioactive compounds were analyzed
with HPLC on a Gemini C18 column (5 μm, 4.6 mm ꢀ 150 mm) or
Gemini-NX C18 column (5 μm, 4.6 mm ꢀ 250 mm; Phenomenex)
and eluted with 10 mM NH4OH-MeCN or 100 mM
HCOONH4-MeCN at the later stated composition and flow
rate. Eluates were monitored for radioactivity (pin diode detector,
Bioscan). Samples were injected alone and then co-injected with
the reference nonradioactive compound to check for coelution.
RCYs were calculated for labeled products isolated with HPLC.
Computation of cLogP and cLogD. cLogP and cLogD (at
pH 7.4) values for ligands were computed with ACD software.
In this study, manipulations of the structure of 1 led to
several new ligands with high affinity toward guinea pig
5-HT4 receptors and a few (8, 13, and 17) with comparably
high affinity toward h5-HT4 receptors. Both 8 and 13 were
amenable to labeling with carbon-11, whereas the labeling of
17 with NCA fluorine-18 remains a challenge. [11C]8 has
affinity, selectivity, intrinsic activity, and computed lipophili-
city comparable to those of [123I]1 (Table 1) and should
prove to be similarly effective for imaging 5-HT4 receptors
in monkey in vivo. [11C]13 is an easily labeled, highly selec-
tive, high-affinity, and moderate lipophilicity antagonist for
5-HT4 receptors and therefore merits evaluation as a PET

7042 Journal of Medicinal Chemistry, 2010, Vol. 53, No. 19
Xu et al.
(1-Butylpiperidin-4-yl)methyl 8-amino-7-iodo-2,3-dihydrobenzo-
[b][1,4]dioxine-5-carboxylate (1). N-Iodosuccinimide (498 mg,
2.21 mmol) was added portionwise to a solution of 4 (770 mg,
2.21 mmol) in acetic acid (5 mL) at 0 °C and stirred at rt for 2 h.
The acetic acid was evaporated, and the residue was basified with
NaHCO3 solution and extracted twice with CH2Cl2. The com-
bined organic layers were evaporated to dryness. Silica gel chro-
matography (MeOH-CH2Cl2, 1:20 v/v) of the residue gave 1 as a
light yellow oil (410 mg, 40%). 1H NMR (CDCl3): δ 0.92 (3H, t,
J = 7.2 Hz), 1.25-1.53 (6H, m), 1.74-1.79 (3H, m), 1.94 (2H, t,
J = 11.2 Hz), 2.31-2.34 (2H, m), 2.96 (2H, d, J = 11.2 Hz), 4.10
(2H, d, J = 6.4Hz), 4.30-4.38 (4H, m), 4.54 (2H, bs), 7.82 (1H, s).
13C NMR (CDCl3): δ 14.09, 20.92, 29.02, 29.17, 35.53, 53.44,
58.89, 63.97, 64.44, 68.86, 71.21, 110.80, 129.77, 133.29, 141.06,
144.75, 164.14.
(1-Butylpiperidin-4-yl)methyl 8-amino-7-cyano-2,3-dihydrobenzo-
[b][1,4]dioxine-5-carboxylate (5).40 A mixture of 1 (200 mg, 0.42
mmol), KCN (42 mg, 0.64 mmol), CuI (17 mg, 0.084 mmol), and
1,10-phenanthroline (30 mg, 0.17 mmol) in DMF (500 μL) was
stirred at 110 °C in an oven-dried sealed tube under Ar for 42 h.
Then the mixture was cooled to rt and filtered through Celite. The
Celite pad was rinsed twice with CH2Cl2. The combined filtrates
were evaporated to dryness and dissolved in MeOH. Separation by
HPLC on an XTerra RP18 column (10 μm, 19 mm ꢀ 250 mm;
Waters) eluted with MeOH-aqueous NH4OH (0.025%) (9:1 v/v)
at 10 mL/min) gave 5 as a pale yellow solid (tR = 14 min; 30 mg,
19%). Mp: 108-110 °C. 1H NMR (CDCl3): δ 0.92 (3H, t, J = 7.2
Hz), 1.28-1.50 (6H, m), 1.81-1.84 (3H, m), 2.15 (2H, t, J = 11.4
Hz), 2.51 (2H, m), 3.16 (2H, d, J = 11.6 Hz), 4.12 (2H, d, J = 6.0
Hz), 4.34-4.44 (4H, m), 4.87 (2H, s), 7.67 (1H, s). 13C NMR
(CDCl3): δ 14.09, 20.92, 29.06, 29.23, 35.53, 53.42, 58.89, 63.73,
64.85, 69.13, 88.00, 110.11, 116.81, 128.80, 129.90, 143.32, 147.47,
163.64. LC-MS m/z: [M þ H]þ, 374.2. HRMS: calcd for
C20H27N3O4 (Mþ þ H), 374.2080; found, 374.2070.
14.02, 16.48, 20.84, 28.69, 35.39, 53.34, 58.72, 63.87, 64.45,
68.27, 107.79, 113.41, 125.34, 130.17, 138.86, 143.34, 165.46.
LC-MS: m/z [M þ H]þ, 363.1. HRMS: calcd for C20H30N2O4
(Mþ þ H), 363.2284; found, 363.2279.
(1-Butylpiperidin-4-yl)methyl 7-chloro-8-(methylamino)-2,3-
dihydrobenzo[b][1,4]dioxine-5-carboxylate (8). N-Chlorosuccini-
mide (70 mg, 0.52 mmol) was added in portions to a stirred
solution of 6 (190 mg, 0.52 mmol) in acetic acid (5 mL) at rt and
left for 1 h. The acetic acid was evaporated, and the residue
was basified with NaHCO3 solution and extracted thrice
with CH2Cl2. The combined organic layers were evaporated to
dryness. Silica gel chromatography (MeOH-CH2Cl2, 1:20
v/v) of the residue gave 8 as a pale yellow oil (80 mg, 38%).
1H NMR (CDCl3): δ 0.92 (3H, t, J = 7.2 Hz), 1.29-1.35 (2H,
m), 1.44-1.53 (4H, m), 1.75-1.80 (3H, m), 1.98 (2H, t, J = 11.2
Hz), 2.34-2.38 (2H, m), 3.00 (2H, d, J = 11.2 Hz), 3.12 (3H, d,
J = 4.8 Hz), 4.10 (2H, d, J = 6.4 Hz), 4.28-4.36 (5H, m), 7.45
(1H, s). 13C NMR (CDCl3): δ 14.03, 20.84, 28.83, 28.97, 33.82,
35.42, 53.35, 58.76, 63.63, 64.26, 68.72, 109.54, 112.30, 124.79,
133.42, 140.14, 144.06, 164.27. LC-MS: m/z [M þ H]þ, 397.6.
HRMS: calcd for C20H29ClN2O4 (Mþ þ H), 397.1894; found,
397.1886.
(1-Butylpiperidin-4-yl)methyl 7-bromo-8-(methylamino)-2,3-
dihydrobenzo[b][1,4]dioxine-5-carboxylate (9). The procedure
for the synthesis of 8 was used with a solution of 6 (140 mg,
0.39 mmol) in acetic acid (4 mL) and with N-bromosuccinimide
(70 mg, 0.39 mmol) replacing N-chlorosuccinimide and gave 9 as
a pale yellow oil (50 mg, 30%). 1H NMR (CDCl3): δ 0.92 (3H, t,
J = 7.2 Hz), 1.29-1.52 (6H, m), 1.73-1.78 (3H, m), 1.92 (2H,
dt, J = 2.4, 11.6 Hz), 2.29-2.33 (2H, m), 2.96 (2H, d, J = 11.2
Hz), 3.10 (3H, d, J = 5.2 Hz), 4.10 (2H, d, J = 6.4 Hz),
4.27-4.37 (5H, m), 7.62 (1H, s). 13C NMR (CDCl3): δ 14.09,
20.93, 29.11, 29.27, 34.11, 35.58, 53.48, 58.93, 63.60, 64.31, 68.96,
102.04, 110.64, 127.61, 133.67, 141.18, 144.76, 164.21. LC-MS:
m/z [M þ H]þ 441.1. HRMS: calcd for C20H2980BrN2O4 (Mþ þ
H), 441.1389; found, 441.1368.
(1-Butylpiperidin-4-yl)methyl 7-iodo-8-(methylamino)-2,3-
dihydrobenzo[b][1,4]dioxine-5-carboxylate (10). The procedure
for the synthesis of 8 was used with a solution of 6 (195 mg,
0.54 mmol) in acetic acid (4 mL) and with N-iodosuccinimide
(122 mg, 0.54 mmol) replacing N-chlorosuccinimide to give 10 as
a colorless oil (147 mg, 56%). 1H NMR (CDCl3): δ 0.92 (3H, t,
J = 7.2 Hz), 1.19-1.54 (6H, m), 1.72-1.80 (3H, m), 1.96 (2H, t,
J = 11.2 Hz), 2.33-2.37 (2H, m), 3.00 (2H, d, J = 11.2 Hz), 3.07
(3H, d, J = 3.2 Hz), 4.05 (1H, bs), 4.09-4.11 (3H, m), 4.28-4.38
(4H, m), 7.84 (1H, s). 13C NMR (CDCl3): δ 14.06, 20.89, 28.91,
29.06, 34.50, 35.47, 53.39, 58.83, 63.57, 64.35, 68.88, 112.25,
133.12, 133.80, 143.57, 145.67, 164.02. LC-MS: m/z [M þ H]þ,
489.2. HRMS: calcd for C20H29IN2O4 (Mþ þ H), 489.1250;
found, 489.1258.
(1-Butylpiperidin-4-yl)methyl 7-cyano-8-(methylamino)-2,3-
dihydrobenzo[b][1,4]dioxine-5-carboxylate (11).40 A mixture of
10 (140 mg, 0.29 mmol), KCN (21 mg, 0.32 mmol), CuI (6.0 mg,
0.029 mmol), 1,10-phenanthroline (11 mg, 0.057 mmol) in DMF
(300 μL) was stirred at 110 °C in an oven-dried sealed tube under
Ar for 48 h. The mixture was cooled to rt and then filtered
through Celite. The Celite pad was rinsed twice with CH2Cl2.
The combined filtrates were evaporated to dryness and then
dissolved in MeOH. Separation by HPLC on an XTerra RP18
column (10 μm, 19 mm ꢀ 250 mm; Waters) and elution with
MeOH-aqueous NH4OH (0.025%) (9:1 v/v) at 15 mL/min
gave 11 (tR = 10.31 min; 8 mg, 7%). 1H NMR (CDCl3): δ 0.92
(3H, t, J = 7.2 Hz), 1.25-1.52 (7H, m), 1.73-1.79 (3H, m),
1.90-1.96 (2H, t, J = 11.2 Hz), 2.30-2.34 (2H, m), 2.96 (2H, d,
J = 11.2 Hz), 3.33 (3H, d, J = 5.2 Hz), 4.09 (2H, d, J = 6.0 Hz),
4.30-4.40 (4H, m), 5.04-5.06 (1H, m), 7.71 (1H, s). 13C NMR
(CDCl3): δ 14.09, 20.92, 29.10, 29.26, 31.63, 35.53, 53.44, 58.91,
63.79, 64.61, 69.07, 85.21, 109.42, 119.52, 130.19, 131.81, 144.86,
145.98, 163.61. LC-MS: m/z [M þ H]þ, 388.2. HRMS: calcd for
C21H29N3O4 (Mþ þ H), 388.2236; found, 388.2219.
(1-Butylpiperidin-4-yl)methyl 8-(methylamino)-2,3-dihydrobenzo-
[b][1,4]dioxine-5-carboxylate (6).41 A mixture of 4 (2.24 g, 6.44
mmol) and paraformaldehyde (580 mg) in ethanol (58 mL) was
heated at 60 °C overnight, and then NaBH4 (245 mg, 6.44 mmol)
was added. The mixture was heated at 70 °C for 2 h and then
evaporated to dryness. The residue was diluted with water and
extracted thrice with CH2Cl2. The combined organic layers were
dried on MgSO4 and evaporated to dryness. Silica gel chroma-
tography of the residue (MeOH-CH2Cl2, 1:20 v/v) gave 6 as a
colorless oil (1.4 g, 60%). 1H NMR (CDCl3): δ 0.92 (3H, t, J =
7.2 Hz), 1.29-1.35 (2H, m), 1.52-1.53 (4H, m), 1.78-1.82 (3H,
m), 1.99 (2H, t, J = 11.2 Hz), 2.39-2.35 (2H, m), 2.90(2H, d, J =
5.2 Hz), 3.02 (2H, d, J = 11.2 Hz), 4.10 (2H, d, J = 6.0 Hz),
4.28-4.30 (2H, m), 4.34-4.36 (2H, m), 4.46-4.52 (1H, m), 6.18
(1H, d, J = 8.8 Hz), 7.51 (1H, d, J = 8.8 Hz). 13C NMR (CDCl3):
δ 14.02, 20.85, 28.78, 28.88, 29.85, 35.47, 53.38, 58.75, 63.83,
64.50, 68.18, 100.99, 107.36, 125.60, 129.78, 143.15, 144.11,
165.40. LC-MS m/z: [M þ H]þ, 363.3. HRMS: calcd for
C20H30N2O4 (Mþ þ H), 363.2284; found, 363.2281.
(1-Butylpiperidin-4-yl)methyl 8-amino-7-methyl-2,3-dihydro-
benzo[b][1,4]dioxine-5-carboxylate (7).42 1 (210 mg, 0.44 mmol),
N-methylpyrrolidone (560 μL), Pd2(dba)3 (14 mg, 0.015 mmol),
and PPh3 (28 mg, 0.11 mmol) were added to an oven-dried sealed
tube. The mixture was heated at 50 °C for 10 min, and then CuI
(6.0 mg, 0.031 mmol) was added. The mixture was stirred for
another 10 min, and then Me4Sn (91 μL, 0.65 mmol) was added.
The mixture was heated at 70 °C for 48 h. The solvent was
evaporated, and the residue was diluted with water and ex-
tracted thrice with CH2Cl2. The combined organic layers were
dried over MgSO4 and evaporated to dryness. Silica gel chro-
matography (MeOH-CH2Cl2, 1:20 v/v) of the residue gave 7 as
a pale yellow oil (70 mg, 44%). 1H NMR (CDCl3): δ 0.92 (3H, t,
J = 7.2 Hz), 1.25-1.58 (7H, m), 1.72-1.81 (3H, m), 1.78-1.82
(3H, m), 1.95 (2H, t, J = 11.2 Hz), 2.11 (3H, s), 2.31-2.35 (2H,
m), 2.98 (2H, d, J = 11.6 Hz), 4.08 (2H, s), 4.10 (2H, d, J = 6.4
Hz), 4.30-4.36 (4H, m), 7.28 (1H, s). 13C NMR (CDCl3): δ

Article
Methyl 4-(Benzyloxycarbonylamino)-2-methoxybenzoate (12b).43
Journal of Medicinal Chemistry, 2010, Vol. 53, No. 19 7043
stirred for 2 h at rt. (1-Butylpiperidin-4-yl)methanamine
(510 mg, 2.99 mmol) in MeCN (10 mL) was added dropwise.
The above solution was stirred overnight and evaporated to
dryness. Silica gel chromatography (from 9% to 33% MeOH in
CH2Cl2) of the residue gave an oil which was then dissolved in
MeCN and filtered through an Iso-Disc filter (PTFE 25-4,
25 mm ꢀ 0.45 μm). The filtrate was dried to give 14 as a light
brown foamlike oil (660 mg, 69%). 1H NMR (CDCl3): δ 0.91
(3H, t, J = 7.2 Hz), 1.26-1.38 (4H, m), 1.42-1.48 (2H, m),
1.57-1.64 (1H, m), 1.88 (2H, dt, J = 2.0, 11.6 Hz), 2.27-2.31
(2H, m), 2.92 (2H, d, J = 11.6 Hz), 3.30 (2H, t, J = 6.4 Hz),
3.86 (3H, s), 4.14 (2H, bs), 6.34 (1H, t, J = 5.6 Hz), 6.63 (1H, d,
J = 8.0 Hz), 7.14 (1H, dd, J = 2.0, 8.4 Hz), 7.37 (1H, d, J = 1.6
Hz). 13C NMR (CDCl3): δ 12.58, 19.41, 27.72, 28.58, 34.79,
43.98, 52.06, 54.08, 57.36, 108.34, 111.67, 118.05, 122.76,
138.15, 145.21, 166.04. LC-MS: m/z [M þ H]þ, 320.2. HRMS:
calcd for C18H30N3O2 (Mþ þ H), 320.2338; found, 320.2328.
(1-Propylpiperidin-4-yl)methyl 4-Amino-3-methoxybenzoate
(15). CDI (485 mg, 2.99 mmol) was added in portions to a
suspension of 4-amino-3-methoxybenzoic acid (500 mg, 2.99
mmol) in MeCN (15 mL) in an oven-dried flask and stirred for
30 min at rt. Solvent was then evaporated and the residue
dissolved in anhydrous THF (7 mL). In another oven-dried
flask, (1-propylpiperidin-4-yl)methanol (319 mg, 2.03 mmol)
was dissolved in anhydrous THF (7 mL) and cooled (ice bath). A
THF solution of n-BuLi (1.6M, 1.27 mL, 2.03 mmol) was added
dropwise to this solution and stirred for 10 min. The solution of
CDI-activated acid was then added, stirred overnight, and
evaporated to dryness. The residue was diluted with water and
extracted thrice with CH2Cl2. The combined organic layers were
dried on MgSO4 and evaporated to dryess. Silica gel chroma-
tography (MeOH-CH2Cl2, 1:30 v/v) of the residue gave 15 as a
yellow oil (312 mg, 50%). 1H NMR (CDCl3): δ 0.89 (3H, t, J =
7.2 Hz), 1.37-1.56 (4H, m), 1.72-1.78 (2H, m), 1.94 (2H, dt,
J = 2.0, 12.0 Hz), 2.25-2.29 (2H, m), 2.94 (2H, d, J = 11.6 Hz),
3.86 (3H, s), 4.13 (2H, d, J = 6.0 Hz), 4.36 (2H, bs), 6.63 (1H, d,
J = 8.4 Hz), 7.44 (1H, d, J = 2.0 Hz), 7.54 (1H, dd, J = 1.6, 8.0
Hz). 13C NMR (CDCl3): δ 12.05, 20.15, 29.06, 35.67, 53.40,
55.50, 61.09, 68.76, 111.08, 112.90, 119.35, 124.03, 141.31,
146.03, 166.83. LC-MS: m/z [M þ H]þ, 307.2. HRMS: calcd
for C17H26N2O3 (Mþ þ H), 307.2022; found, 307.2022.
Saturated NaHCO3 solution (50 mL) and then CbzCl (1.70 mL,
12.1 mmol) were added to a solution of methyl 4-amino-2-methoxy-
benzoate (12a, 2.0 g, 11 mmol) in THF (50 mL). The mixture was
stirred at rt for 4 h and then filtered through Celite. The filtrate was
acidified to pH < 1 and extracted with EtOAc. After concentration
of this solution, the residue was purified by recrystallization from
EtOAc-hexane to give 12b as a white solid (3.09 g, 89%). Mp:
132-134 °C. 1H NMR (CDCl3): δ 3.84 (3H, s), 3.87 (3H, s), 5.19
(2H, s), 6.77 (1H, dd, J = 1.6, 8.4 Hz), 6.66, 7.11 (1H, s), 7.35-7.39
(5H, m), 7.79 (1H, d, J = 8.4 Hz). 13C NMR (CDCl3): δ 51.84,
55.97, 67.30, 101.72, 109.43, 114.05, 128.32, 128.52, 128.68, 133.06,
135.67, 143.19, 152.96, 160.76, 166.07.
(1-Butylpiperidin-4-yl)methyl 4-Amino-2-methoxybenzoate
(12).44 A THF solution of n-BuLi (1.6 M, 612 μL, 0.98 mmol)
was added dropwise to a solution of (4-butylpiperidin-1-yl)-
methanol (190 mg, 1.11 mmol) in THF (3 mL) in an oven-dried
flask under Ar at 0 °C. After the mixture was stirred for 10 min, a
solution of 12b (350 mg, 1.11 mmol) in THF (3 mL) was added
dropwise. The mixturewas stirrredfor 2 h, pouredintowater, and
extracted with CH2Cl2. The combined organic layers were dried
on MgSO4 and evaporated to dryness. Silica gel chromatog-
raphy of the residue (MeOH-CH2Cl2, 1:15 v/v) gave a white
solid (200 mg). MeOH (15 mL) was added to this solid (170 mg)
plus Pd/C (10%, 25 mg). The mixture was degassed with H2
for 30 min, stirred at rt overnight, filtered, and evaporated
to dryness. Silica gel chromatography (MeOH-CH2Cl2, 1:20
v/v) of the residue gave 12 as a colorless oil (108 mg, 36%). 1H
NMR (CDCl3): δ 0.91 (3H, t, J = 7.6 Hz), 1.28-1.53 (6H, m),
1.71-1.80 (3H, m), 1.95 (2H, t, J = 11.2 Hz), 2.31-2.35 (2H,
m), 2.97 (2H, d, J = 11.6 Hz), 3.81 (3H, s), 4.09 (2H, d,
J = 6.0 Hz), 4.22 (2H, br), 6.18-6.22 (2H, m), 7.72 (1H, d,
J = 8.4 Hz). 13C NMR (CDCl3): δ 14.06, 20.88, 28.94, 29.07,
35.53, 53.45, 55.66, 58.84, 68.33, 97.58, 106.23, 108.78, 134.06,
152.29, 161.87, 165.63. LC-MS: m/z [M þ H]þ, 321.1. HRMS:
calcd for C18H28N2O3 (Mþ þ H), 321.2178; found, 321.2181.
Methyl 4-Amino-3-methoxybenzoate (13a). Pd/C (10%, 1.8 g)
and HCOOK (7.4 g, 88 mmol) were added to a solution of
methyl 3-methoxy-4-nitrobenzoate (2.0 g, 9.5 mmol) in MeOH
(50 mL). The mixture was refluxed at 80 °C for 1 h. The sus-
pension was cooled and filtered through Celite. The filtrate was
evaporated to dryness. Silica gel chromatography (30% EtOAc
in hexane) of the residue gave 13a as a white solid (1.72 g, 100%).
Mp: 127-129 °C. Lit. mp: 127-128 °C.45 1H NMR (CDCl3): δ
3.86 (3H, s), 3.90 (3H, s), 4.22 (2H, br), 4.45 (1H, s), 6.66 (1H, d,
J = 8.4 Hz), 7.55 (1H, d, J = 8.4 Hz). 13C NMR (CDCl3): δ
51.69, 55.58, 111.15, 113.09, 119.41, 124.09, 141.21, 146.10,
167.35.
(1-Pentylpiperidin-4-yl)methyl 4-Amino-3-methoxybenzoate
(16). As described for 15, compound 16 was obtained from
4-amino-3-methoxybenzoic acid and 1-pentylpiperidin-4-yl)-
methanol as a pale yellow oil (200 mg, 30%). 1H NMR
(CDCl3): δ 0.89 (3H, t, J = 7.2 Hz), 1.24-1.52 (8H, m), 1.77-
1.79 (3H, m), 1.89-1.95 (2H, dt, J = 2.4, 12 Hz), 1.95-2.32
(2H, m), 2.96 (2H, d, J = 11.6 Hz), 3.89 (3H, s), 4.13 (2H, d, J =
6.4 Hz), 4.26 (2H, bs), 6.65 (1H, d, J = 8 Hz), 7.45 (1H, d, J = 1.6
Hz), 7.55 (1H, dd, J = 2, 8.4 Hz). 13C NMR (CDCl3): δ 14.06,
22.64, 26.78, 29.10, 29.94, 35.71, 53.49, 55.59, 59.23, 68.82, 111.16,
113.00, 119.61, 124.03, 141.17, 146.12, 166.86. LC-MS: m/z [M þ
H]þ, 335.2. HRMS: calcd for C19H30N2O3 (Mþ þ H), 335.2335;
found, 335.2325.
3-Bromo-5-fluoro-2-hydroxybenzoic Acid (17a). 5-Fluoro-2-
hydroxybenzoic acid (20 g, 0.13 mol) and N-bromosuccinimide
(23 g, 0.13 mol) were added to acetic acid (200 mL). The mixture
was heated at 80 °C for 24 h. After evaporation of all acetic acid,
the residue was recrystallized from EtOAc and hexane to give
17a (27 g, 90%). Mp 233-235 °C. Lit. mp 233 °C.46 1H NMR
(MeOD): δ 7.48-7.50 (2H, br). 13C NMR (MeOD): δ 111.91 (d,
J = 10.0 Hz), 114.92 (d, J = 7.0 Hz), 116.22 (d, J = 24.0 Hz),
127.13 (d, J = 26.0 Hz), 155.70 (d, J = 239 Hz), 156.56, 172.24.
5-Fluoro-2,3-dihydroxybenzoic Acid (17b).47 NaOH solution
(2.5 M, 300 mL) was stirred under an aspirator for 2 h. Then
CuSO4 (250 mg, 1.57 mmol) was added and the solution further
stirred for 1 h. To the filtrate of this solution was added 17a
(22 g, 0.094 mol), and the mixture was refluxed overnight. The
reaction mixture was cooled, acidified with 37% HCl to pH < 2,
and then evaporated to dryness. The residue was dissoved in
Methyl 4-(Benzyloxycarbonylamino)-3-methoxybenzoate (13b).
As described for 12b from 12a, 13b was obtained from 13a as
a white solid (58%). Mp: 95-96 °C. 1H NMR (CDCl3): δ 3.89
(6H, 2s), 5.22 (2H, s), 7.34-7.52 (6H, m), 7.69 (1H, dd, J = 1.6,
8.4 Hz), 8.20 (1H, d, J = 8.4 Hz). 13C NMR (CDCl3): δ 52.05,
55.90, 67.27, 110.73, 116.92, 123.48, 124.25, 128.43, 128.49,
128.68, 132.06, 135.84, 146.96, 152.93, 166.81.
(1-Butylpiperidin-4-yl)methyl 4-Amino-3-methoxybenzoate (13).
As described for 12 from 12b, compound 13 was obtained from 13b
as a white solid in 13% yield. Mp: 162-164 °C. 1HNMR (CDCl3):
δ 0.94 (3H, t, J = 7.2 Hz), 1.31-1.39 (2H, m), 1.72-1.80 (2H, m),
1.95-1.99 (5H, br), 2.55 (2H, br), 2.75-2.79 (2H, m), 3.37 (2H, d,
J = 11.6 Hz), 3.90 (3H, s), 4.18 (2H, d, J = 4.0 Hz), 4.38 (2H, br),
6.69 (1H, d, J = 8.0 Hz), 7.43 (1H, d, J = 2 Hz), 7.52-7.53 (1H,
dd, J =1.6, 8Hz). 13CNMR:δ(CDCl3) 14.09, 20.94, 29.10, 29.25,
35.71, 53.50, 55.63, 58.94, 68.84, 111.21, 113.05, 119.75, 124.02,
141.09, 146.16, 166.86. LC-MS: m/z [M þ H]þ, 321.3. HRMS:
calcd for C18H28N2O3 (Mþ þ H), 321.2178; found, 321.2190.
4-Amino-N-((1-butylpiperidin-4-yl)methyl)-3-methoxybenzamide
(14).38 CDI (485 mg, 2.99 mmol) was added in portions to
a suspension of 4-amino-3-methoxybenzoic acid (500 mg, 2.99
mmol) in MeCN (30 mL) in an oven-dried flask and then

7044 Journal of Medicinal Chemistry, 2010, Vol. 53, No. 19
Xu et al.
MeOH and filtered through Celite. The combined filtrates were
evaporated to dryness. Silica gel chromatography (30% EtOAc
in hexane with 1% HOAc) of the residue gave 17b as a white
solid (12 g, 75%). Mp 184-186 °C. 1H NMR (MeOD): δ 6.58
(1H, dd, J = 3.2, 9.6 Hz), 7.01 (1H, dd, J = 3.0, 9.0 Hz). 13C
NMR (MeOD): δ 105.80 (d, J = 24.0 Hz), 109.44 (d, J = 26.0
Hz), 113.48 (d, J = 9.0 Hz), 148.36, 148.50 (d, J = 9.0 Hz),
156.24 (d, J = 234 Hz), 173.03. 19F NMR (CDCl3): δ -125.39.
HRMS: calcd for C7H5FO4 (Mþ þ H), 172.017 19; found,
172.017 17.
Methyl 7-Fluoro-2,3-dihydrobenzo[b][1,4]dioxine-5-carboxy-
late (17c).48 TMSCHN2 (2.0 M, 45 mL, 90 mmol) was added
in portions to a solution of 17b (6.0 g, 35 mmol) in MeOH (150
mL) and Et2O (150 mL). After 20 min the solvent was evapo-
rated off. Silica gel chromatography (15% EtOAc in hexane) of
the residue gave the methyl ester (5.6 g, 93%). A mixture of this
ester (3.9 g, 20 mmol) and Cs2CO3 (16.2 g, 49.8 mmol) in DMF
(40 mL) was stirred at rt for 0.5 h. Then 1,2-dibromoethane
(5.07 g, 27.0 mmol) was added and the mixture stirred at 80 °C
for 16 h. The mixture was cooled to rt and filtered through
Celite, which was then rinsed twice with DMF. The combined
DMF rinses were evaporated to dryness under high vacuum to
give a dark red residue, which after silica gel chromatography
(20% EtOAc in hexane with 1% HOAc) gave 17c as a white solid
(3.37 g, 77%). Mp 108-110 °C. 1H NMR (MeOD): δ 3.89 (3H,
s), 4.28-4.34 (4H, m), 6.77 (1H, dd, J = 3.2, 8.8 Hz), 7.11 (1H,
dd, J = 3.2, 8.8 Hz). 13C NMR (MeOD): δ 52.27, 64.10, 64.30,
108.71 (d, J = 26.0 Hz), 109.73 (d, J = 24.0 Hz), 120.07 (d, J =
9.0 Hz), 140.65, 144.69 (d, J = 12.0 Hz), 155.69 (d, J = 238 Hz),
164.99. 19F NMR (CDCl3): δ -121.14. HRMS: calcd for
C10H9FO4 (Mþ þ H), 213.0563; found, 213.0561.
Cbz-protected 17e (420 mg, 1.21 mmol), CDI (196 mg, 1.21
mmol), and MeCN (30 mL) were added to an oven-dried flask
under Ar. The mixture was stirred for 2 h and evaporated to
dryness. The residue was dissolved in THF (10 mL). In another
oven-dried flask, a solution of (4-butylpiperidin-1-yl)methanol
(207 mg, 1.21 mmol) in THF (10 mL) was added dropwise to a
THF solution of n-BuLi (1.6 M, 760 μL, 1.21 mmol) under Ar at
0 °C and stirred at this temperature for 10 min. Then the
activated acid in THF was added dropwise to the prepared
lithium alkoxide solution. The mixture was warmed to rt and
stirred overnight. The reaction mixture was evaporated to
remove THF, diluted with water, and extracted with CH2Cl2.
The residue was purified by silica gel chromatography (MeOH-
CH2Cl2, 1:15 v/v) to give an oil. Pd/C (10%, 25 mg) was then
added to a solution of the oil in MeOH (15 mL). The suspen-
sion was degassed for 30 min with H2 and then stirred at rt
overnight under H2. The mixture was filtered through Celite
and evaporated to dryness. Silica gel chromatography (MeOH-
CH2Cl2, 1:15 v/v) of the residue gave 17 as a yellow solid
1
(260 mg, 59%). Mp 160-162 °C. H NMR (CDCl3): δ 0.95
(3H, t, J = 7.4 Hz), 1.33-1.39 (2H, m), 1.70-1.95 (7H, m),
2.39-2.42 (2H, m), 2.67-2.71 (2H, m), 3.31 (2H, d, J = 11.6
Hz), 4.19-4.27 (6H, m), 4.88 (2H, s), 6.71 (1H, d, J = 11.6 Hz).
13C NMR (CDCl3): δ 13.74, 20.46, 26.97, 34.35, 52.56, 57.84,
63.84, 64.56, 68.14, 105.60 (d, J = 4.0 Hz), 108.19 (d, J = 23.0
Hz), 132.20 (d, J = 15.0 Hz), 133.43 (d, J = 11.0 Hz), 139.81 (d,
J = 3.0 Hz), 145.24 (d, J = 232 Hz), 166.72 (d, J = 4 Hz). 19F
NMR (CDCl3): δ -141.99. LC-MS: m/z [M þ H]þ, 367.1.
HRMS: calcd for C19H27FN2O4 (Mþ þ H), 367.2033; found,
367.2033.
Methyl 4-Amino-3-(fluoromethoxy)benzoate (18a).49 Chloro-
fluoromethane was bubbled into a tube (30 mL) containing
anhydrous DMF (10 mL), methyl 4-amino-3-hydroxybenzoate
(500 mg, 2.99 mmol), and Cs2CO3 (1.95 g, 5.98 mmol) for 14 min
at -70 °C. The tube was then sealed and slowly warmed to rt.
The mixture was stirred for 5 days and then filtered through
Celite which was then rinsed thrice with EtOAc. The combined
rinses were evaporated to dryness. Silica gel chromatography
(20% EtOAc in hexane) of the residue gave 18a as a pale yellow
solid (368 mg, 62%). Mp: 69-70 °C. 1H NMR (CDCl3): δ 3.84
(3H, s), 4.46 (2H, br), 5.71 (2H, d, J = 54.4 Hz), 6.69 (1H, d, J =
8 Hz), 7.63 (1H, dd, J = 8.4, 1.6 Hz), 7.68 (1H, s). 13C NMR
(CDCl3): δ 51.73, 100.22, 102.40, 114.25, 116.72, 119.23, 126.79,
142.44, 142.92, 142.95, 166.90. 19F NMR (CDCl3): δ -147.56.
LC-MS: m/z [M þ H]þ 200.1. HRMS: calcd for C9H11FNO3
(Mþ þ H), 200.0723; found, 200.0723.
Methyl 7-Fluoro-8-nitro-2,3-dihydrobenzo[b][1,4]dioxine-5-
carboxylate (17d). 17c (3.4 g, 15.9 mol) was added in portions to a
flask cooled between -50 and -60 °C and containing HNO3
(90%, 30 mL). After 16 min, the reaction mixture was warmed
gradually and water was added. The precipitate was filtered off
and washed with water to give 17d as a yellow solid (3.58 g, 88%).
17d was used in the next step without further purification. Mp
133-135 °C. 1H NMR (CDCl3): δ 3.92 (3H, s), 4.33-4.40 (4H,
m), 6.83 (1H, d, J = 11.2 Hz). 13C NMR (CDCl3): δ 53.59, 64.22,
64.80, 107.12 (d, J = 25.0 Hz), 119.75 (d, J = 1.0 Hz), 137.57 (d,
J = 3.0 Hz), 148.50 (d, J = 12.0 Hz), 150.10 (d, J = 257 Hz),
162.83 (d, J = 3.0 Hz). 19F NMR (CDCl3): δ -124.76. HRMS:
calcd for C10H8FNO6 (Mþ þ Na), 280.0233; found, 280.0225.
8-Amino-7-fluoro-2,3-dihydrobenzo[b][1,4]dioxine-5-carboxylic
Acid (17e). Pd/C (10%, 340 mg) and potassium formate (1.4 g,
17 mmol) were added to a solution of 17d (500 mg, 1.94 mmol) in
MeOH (10 mL). The mixture was refluxed at 80 °C for 2 h and
then cooled and filtered through Celite. Then 37% HCl was
added to the filtrate until no CO2 was released. The white solid
was filtered off and the filtrate evaporated to dryness. Silica gel
chromatography (MeOH-CH2Cl2, 1:30 v/v) of the residue gave
(1-Butylpiperidin-4-yl)methyl 4-Amino-3-(fluoromethoxy)-
benzoate (18). n-BuLi (1.6 M, 1.6 mmol, 1 mL) was added
dropwise to a solution of (4-butylpiperidin-1-yl)methanol (280
mg, 1.64 mmol) in THF (2 mL) contained in an oven-dried flask
under Ar at 0 °C. The solution was stirred for 10 min, and 18a
(160 mg, 0.80 mmol) in THF (2 mL) was added dropwise. The
mixture was stirred overnight, poured into water, and extracted
with CH2Cl2. The combined organic layers were dried on MgSO4
and evaporated to dryness. Silica gel chromatography (MeOH-
CH2Cl2, 1:20 v/v) of the residue gave 18 as a pale yellow oil (160
mg, 59%). 1H NMR (CDCl3): δ 0.92 (3H, t, J = 7.6 Hz),
1.43-1.53 (6H, m), 1.75-1.80 (3H, m), 1.91-1.98 (2H, t, J =
12.0 Hz), 2.33 (2H, t, J = 8 Hz), 2.98 (2H, d, J = 11.6 Hz), 4.14
(2H, d, J = 6.0 Hz), 4.28 (2H, br), 5.76 (2H, d, J = 54.4 Hz), 6.71
(1H, d, J = 8.0 Hz), 7.67 (1H, dd, J = 2.0, 8.4 Hz), 7.70 (1H, s).
13C NMR (CDCl3): δ 12.81, 19.65, 27.73, 27.89, 34.37, 52.18,
57.63, 67.66, 100.17 (d, J = 219 Hz), 113.04, 115.81, 118.78,
125.58, 140.86, 141.85 (d, J = 2.0 Hz), 165.00. 19F NMR
(CDCl3): δ -147.50. LC-MS m/z [M þ H]þ, 214.1. HRMS:
calcd for C18H28FN2O3 (Mþ þ H), 339.2084; found, 339.2088.
Methyl 4-Amino-3-(2-fluoroethoxy)benzoate (19a). 1-Chloro-
2-fluoroethane (760 mg, 8.97 mmol) was added to a tube (15 mL)
containing anhydrous DMF (5 mL), methyl 4-amino-3-hydroxy-
benzoate (500 mg, 2.99 mmol), and Cs2CO3 (1.95 g, 5.98 mmol).
1
17e as light brown solid (350 mg, 85%). Mp 174-176 °C. H
NMR (MeOD): δ 4.08-4.10 (2H, m), 4.18-4.21 (2H, m), 6.67
(1H, d, J = 11.6 Hz). 13C NMR (MeOD): δ 65.06, 66.29, 106.06
(d, J = 4.0 Hz), 109.21 (d, J = 23.0 Hz), 134.30 (d, J = 11.0 Hz),
134.44, 140.54 (d, J = 3.0 Hz), 147.01 (d, J = 232 Hz), 169.25.
19F NMR (MeOD): δ -143.14. HRMS: calcd for C9H8FNO4
(Mþ þ H), 214.0516; found, 214.0510.
(1-Butylpiperidin-4-yl)methyl 8-Amino-7-fluoro-2,3-dihydro-
benzo[b][1,4]dioxine-5-carboxylate (17). Saturated NaHCO3 so-
lution (5 mL) and THF (20 mL) followed by CbzCl (662 mg,
3.87 mmol) were added to a flask containing 17e (750 mg, 3.52
mmol). After the mixture was stirred overnight, the THF was
evaporated off. The solution was diluted with water, acidified
with 37% HCl to pH < 1, and extracted thrice with EtOAc.
The combined organic layers were dried on MgSO4 and evapo-
rated to dryness. Silica gel chromatography (30% EtOAc in
hexane) of the residue gave Cbz-protected 17e as a white solid
(420 mg, 34%).

Article
Journal of Medicinal Chemistry, 2010, Vol. 53, No. 19 7045
The tube was sealed, and the mixture was stirred for 88 h. The
mixture was then filtered through Celite which was rinsed thrice
with EtOAc. The combined filtrate and rinses were evaporated to
dryness. Silica gel chromatography (20% EtOAc in hexane) of the
residue gave 19a as a whitesolid (270 mg, 42%). Mp 86-87 °C. 1H
NMR (CDCl3): δ 3.84 (3H, s), 4.18-4.27 (2H, dt, J = 28.4, 4 Hz),
4.39 (2H, br), 4.65-4.79 (2H, dt, J = 47.2, 3.7 Hz), 6.66 (1H,
d, J = 8 Hz), 7.43 (1H, d, J = 1.6 Hz), 7.55 (1H, dd, J = 8.4, 1.6
Hz). 19F NMR (CDCl3): δ -147.56. LC-MS: m/z [M þ H]þ,
214.1. HRMS: calcd for C10H13FNO3 (Mþ þ H), 214.0879;
found, 214.0876.
(1-Butylpiperidin-4-yl)methyl 4-Amino-3-(2-fluoroethoxy)-
benzoate (19). As described for 18 from 18a, compound 19 was
obtained from 19a as a white solid in 48% yield. Mp 72-74 °C.
1H NMR (CDCl3): δ 0.92 (3H, t, J = 7.6 Hz), 1.27-1.53 (6H, m),
1.74-1.79 (3H, m), 1.94 (2H, t, J = 12.0 Hz), 2.33 (2H, t, J = 7.6
Hz), 2.97 (2H, d, J = 11.2 Hz), 4.13 (2H, d, J = 6 Hz), 4.24-4.33
(2H, dt, J = 28.4, 4.0 Hz), 4.36 (2H, br), 4.69-4.83 (2H, dt, J =
47.2, 4.0 Hz), 6.67 (1H, d, J = 8.0 Hz), 7.45 (1H, d, J = 2.0 Hz),
7.57 (1H, dd, J = 1.6, 8.4 Hz). 13C NMR (CDCl3): δ 14.07, 20.89,
29.03, 29.17, 35.65, 53.45, 58.88, 67.92 (d, J = 20.0 Hz), 68.82,
81.77 (d, J = 170 Hz), 112.94, 113.39, 119.45, 124.79, 141.70,
144.70, 166.67. 19F NMR (CDCl3): δ -147.49. LC-MS m/z
[M þ H]þ, 353.2. HRMS: calcd for C19H30FN2O3 (Mþ þ H),
353.2240; found, 353.2238.
21c (100 mg, 0.20 mmol) was stirred overnight in a dioxane
solution of HCl (4 M). The mixture was evaporated to dryness,
diluted with water, and then neutralized to pH 7 with aqueous
NH4OH (1 M). The solution was extracted thrice with CH2Cl2.
The organic layers were combined and dried on MgSO4. Eva-
poration of solvent gave 21 as a yellow solid (71 mg, 89%). Mp
108-110 °C (n = 2). 1H NMR (CDCl3): δ 0.92 (3H, t, J = 7.2
Hz), 1.29-1.51 (7H, m), 1.79 (3H, d, J = 8.4 Hz), 1.92 (2H, t,
J = 11.6 Hz), 2.33 (2H, t, J = 7.6 Hz), 2.98 (2H, d, J = 11.6 Hz),
4.14 (2H, d, J = 6.4 Hz), 4.38-4.46 (4H, m), 8.45 (1H, s). 13C
NMR (CDCl3): δ 14.06, 20.89, 28.99, 29.16, 35.47, 53.38, 58.85,
63.82, 64.84, 69.28, 108.62, 123.12, 125.47, 130.17, 139.42,
147.16, 163.70. LC-MS: m/z [M þ H]þ, 394.2. HRMS: calcd
for C19H28N3O6 (Mþ þ H), 394.1978; found, 394.1972.
Pharmacological Assay and Screen. Some ligands (1, 5, 8, 13,
15, 17-19) were evaluated at Caliper Life Sciences (Hanover,
MD) for binding to 5-HT4 receptors in guinea pig striatal mem-
branes at 37 °C with [3H]GR 113808 as reference radioligand.53
All new ligands (2, 3, and 5-21) plus the already known
ligands 1 and 4 were submitted to the National Institute of
Mental Health Psychoactive Drug Screening Program (NIMH-
PDSP) for assessment of binding affinity to human recombi-
nant 5-HT4 receptors (reference radioligand, [3H]GR 113808)
and a wide range of other receptors and binding sites (5-
HT1A-E,2A-C,3,5A,6,7, R1A,B,D,2A-C, BZP (rat brain site), β1-3
,
(1-Butylpiperidin-4-yl)methyl 4-Amino-3-hydroxybenzoate
(20).50,51 AlCl3 (311 mg, 2.34 mmol) and NaI (351 mg, 2.34
mmol) were added to a solution of 13 (500 mg, 1.56 mmol) in
MeCN (10 mL). The mixture was refluxed overnight and ex-
tracted thrice with EtOAc. The combined organic layers were
dried on MgSO4 and evaporated to dryness. Silica gel chroma-
tography (MeOH/CH2Cl2, 1:20 v/v) of the residue gave 20 as an
orange-brown solid (200 mg, 42%). Mp: 210-212 °C. 1H NMR
(CDCl3):δ0.98(3H, t, J= 7.6 Hz), 1.31-1.43(2H, m), 1.51-1.63
(4H, m), 1.87-1.93 (3H, m), 2.28 (2H, t, J = 11.2 Hz), 2.54-2.58
(2H, m), 3.17 (2H, d, J = 12.0 Hz), 4.14 (2H, d, J = 5.6 Hz), 6.68
(1H, d, J = 8.0 Hz), 7.34 (1H, d, J = 2.0 Hz), 7.38-7.41 (1H, dd,
J = 1.6, 8.0 Hz); 13C NMR (CDCl3): δ 13.96, 21.05, 27.46, 34.81,
53.45, 58.09, 68.05, 114.47, 116.05, 119.06, 124.34, 143.61, 144.84,
168.55. LC-MS: m/z [M þ H]þ, 307.2. HRMS: calcd for
C17H26N2O3 (Mþ þ H), 307.2022; found, 307.2025.
σ
1,2, D1-5, DAT, DOR, GABAA, H1-4, KOR, M1-5, MOR,
NET, and SERT). Selected ligands (1, 4, 6-13, and 17) were also
assessed for agonist/partial agonist activity in a GloSensor L9
assay for cAMP response and for antagonist activity. Detailed
assay protocols are available at the NIMH-PDSP Web site
(http://pdsp.cwru.edu).
Radiochemistry. Production of NCA [11C]Carbon Dioxide.
No-carrier-added (NCA) [11C]carbon dioxide was produced with
a PEtrace cyclotron (GE; Milwaukee, WI) according to the
14N(p,R)11C reaction20 by irradiation of nitrogen gas (300 psi)
containing 1% oxygen with a proton beam (16.5 MeV, 45 μA) for
either 20 or 40 min. A 40 min irradiation produced about 2.0 Ci of
[11C]carbon dioxide.
Production of NCA [11C]Hydrogen Cyanide.20 A PETrace
radiotracer production system was used to produce this labeling
agent as follows. Cyclotron-produced NCA [11C]carbon dioxide
˚
(1-Butylpiperidin-4-yl)methyl 8-Amino-7-nitro-2,3-dihydrobenzo-
[b][1,4]dioxine-5-carboxylate (21). A mixture of 21a52 (200 mg,
0.79 mmol), (Boc)2O (515 mg, 2.36 mmol), and DMAP (44 mg,
0.039 mmol) in CH2Cl2 (15 mL) was refluxed at 50 °C for 1 h,
then quenched with water and extracted thrice with CH2Cl2.
The combined organic layers were dried on MgSO4 and then
condensed to a crude product, which after silica gel chroma-
tography (30% EtOAc in hexane) gave di-Boc-protected 21a as
was trapped on molecular sieves (13 A, 80-100 mesh, 0.55 g) at
40 °C, while residual [13N]nitrogen was directed to waste. The
[11C]carbon dioxide was then released with a stream of nitrogen
(250 mL/min), mixed with a stream of hydrogen (30 mL/min),
and passed through a heated (400 °C) glass tube (10 mm ꢀ
200 mm) containing nickel catalyst (Ni-3266 Engelhardt). The
effluent containing the generated [11C]methane was passed
through an OXY-TRAP (part no. 4001R, Alltech), mixed
with anhydrous ammonia (research grade, 20 mL/min), and
then passed over a wad of platinum wire (d 0.127 mm, 2.6 g) at
920 °C. The generated [11C]hydrogen cyanide was delivered
to a hot cell for subsequent radiochemistry in a mixture of
hydrogen, ammonia, and nitrogen carrier gas at 300 mL/min.
Production of NCA [11C]Methyl Iodide. NCA [11C]methyl
iodide was produced from NCA [11C]carbon dioxide (∼2.0 Ci)
via reduction to [11C]methane and then vapor phase iodination,18
either in a TRACERlab FX C Pro module (GE; Milwaukee, WI)
for the synthesis of [11C]8 or a MeI MicroLab apparatus (GE;
Milwaukee, WI) for the synthesis of [11C]13.
1
a yellow solid (21b, 343 mg, 95%). H NMR (CDCl3): δ 1.41
(18H, s), 3.94 (3H, s), 4.37-4.49 (4H, m), 8.30 (1H, s). 13C NMR
(CDCl3): δ 27.78, 52.63, 63.77, 64.80, 83.66, 118.16, 120.06,
126.55, 138.19, 140.71, 148.53, 149.49, 163.61.
A THF solution of n-BuLi (1.6 M, 0.95 mL, 1.52 mmol) was
added dropwise to a solution of (4-butylpiperidin-1-yl)methanol
(260 mg, 1.52 mmol) in THF (2 mL) in an oven-dried flask,
under Ar at 0 °C. After this solution was stirred for 10 min, a
solution of 21b (343 mg, 0.76 mmol) in THF (4 mL) was added
dropwise. The mixture was stirred overnight, poured into water,
and extracted with CH2Cl2. The combined organic layers were
dried on MgSO4 and evaporated to dryness. Silica gel chroma-
tography (MeOH/CH2Cl2, 1:15 v/v) of the residue gave mono-
Boc-protected 21 as a white solid (21c, 186 mg, 50%). 1H NMR
(CDCl3): δ 0.92 (3H, t, J = 7.2 Hz), 1.29-1.49 (15H, m), 1.78
(3H, t, J = 12.4 Hz), 1.93 (2H, t, J = 10.0 Hz), 2.31 (2H, t, J =
7.6 Hz), 2.96 (2H, d, J = 11.6 Hz), 4.16 (2H, d, J = 6.4 Hz),
4.39-4.48 (4H, m), 7.20 (1H, br), 8.15 (1H, s). 13C NMR
(CDCl3) δ 13.06, 19.88, 27.05, 28.01, 28.24, 34.43, 52.34,
57.85, 62.79, 63.70, 68.82, 81.43, 119.75, 124.57, 134.69,
135.44, 146.92, 150.51, 162.21.
Production of NCA [11C]Methyl Triflate. A quartz column
(2.6 mm i.d., 26 cm length) was packed around its center with a
6 cm length of AgOTf/Graphac (50:50 w/w) held in place at each
end with glass wool. NCA [11C]methyl triflate (∼350 mCi) was
produced by passing [11C]methyl iodide in helium gas (17 mL/
min) into the heated (180 °C) column.19
Production of NCA [18F]Fluoromethyl-d2 Bromide.54 Cyclo-
tron-produced [18F]fluoride ion (∼150 mCi) in [18O]water was
delivered into a glass vial containing K 2.2.2 (5.0 mg, 13.3 μmol)
and potassium carbonate (0.50 mg, 3.6 μmol) in MeCN-H2O

7046 Journal of Medicinal Chemistry, 2010, Vol. 53, No. 19
Xu et al.
(0.1 mL, 9:1 v/v). This solution was transferred to a modified
version of a TRACERlab FXF-N module and diluted with
MeCN (1 mL). The mixture was evaporated to dryness at
90 °C under reduced pressure with a nitrogen flow. MeCN
(2 mL) was again added and then evaporated to dryness. The
vessel was sealed, and then CD2Br2 (100 μL) in MeCN (1.0 mL)
was added to the dry [18F]fluoride ion-K 2.2.2-Kþ complex
which was then heated at 95 °C for 15 min. The reaction vessel
was then cooled to 35 °C. Nitrogen gas was used to transfer the
volatile [18F]fluoromethyl-d2 bromide through a series of four
silica gel cartridges (SepPak Plus) and then into a V-vial (1-mL)
having a crimp-sealed silicon-Teflon septum cap. The RCY of
[18F]fluoromethyl-d2 bromide was typically about 28%.54
Radiosynthesis of [11C]5. NCA [11C]hydrogen cyanide (∼200
mCi) was trapped in a V-vial (5 mL) containing THF (500 μL),
precursor (1, ∼1.0 mg), Pd(PPh3)4 (1.5-2.0 mg), and base
[K2CO3 (∼2.0 mg) plus K 2.2.2 (∼5.0 mg), or usually KH2PO4
(∼2.0 mg) only]. The reaction mixture was heated at 80 °C for
5 min. HPLC mobile phase (3 mL) was added to the V-vial and
[11C]5 isolated with radio-HPLC on a Gemini C18 column
(5 μm, 10 mm ꢀ 250 mm; MeCN-aqueous NH4OH (10 mM,
3:2 v/v, 6 mL/min, tR = 7.2 min). The identity of [11C]5 was
confirmed by analytical radio-HPLC on a Gemini C18 column
(5 μm, 4.6 mm ꢀ 150 mm) eluted at 1 mL/min with MeCN-
10 mM NH4OH (16:9 v/v, tR = 5.8 min) and also by LC-MS of
the associated carrier. The radiochemical purity of [11C]5 was
>99%. The decay-corrected radiochemical yield (RCY) from
[11C]hydrogen cyanide was 26% (n = 3).
The following experiment22 was performed to confirm the
position of radiolabel in [11C]13. [11C]Methyl iodide (∼47 mCi)
was trapped in a solution of 20 (∼1.0 mg) in DMF (300 μL).
Then [13C]methyl iodide (10 μL from 47.7 mM stock solution in
DMF) was added. [11C]/[13C]13 was isolated by HPLC, as
described above. A sample of the collected radioactive fraction
was then analyzed by HPLC and LC-MS. The radioactive
fraction was then evaporated to dryness, dissolved in CDCl3,
and analyzed by 13C NMR {DEPT 135}.
Radiosynthesis of [18F]18. Phenol 20 (∼0.15 mg, 0.49 umol),
MeCN (300 μL), and 0.5 M NaOH (1.5 μmol, 3 μL) were added
to a V-vial (1 mL). [18F]FCD2Br (∼20 mCi) was transferred to the
solution under computer control from a TRACERlab FXF-N
module. Radioactivity transfer was monitored by two external
radioactivity detectors (Bioscan) and was stopped when radio-
activity was maximized. The mixture was heated at 100 °C for
15 min and then diluted with water (700 μL). [18F]18 was isolated
with HPLC (MeCN-HCOONH4, 2:3 v/v, 3 mL/min, tR = 10.1
min). The identity of [18F]18 was confirmed by analytical HPLC on
a Gemini-NX C18 column (5 μm, 4.6 mm ꢀ 250 mm) eluted with
MeCN-100 mM HCOONH4 (1:1 v/v) at 1 mL/min (tR = 4.2 min)
and also by LC-MS. The RCY of [18F]18 was 13% (n = 1) from
[18F]fluoride ion and the radiochemical purity >99%.
Acknowledgment. This study was supported by the Intra-
mural Research Program of the National Institutes of Health
(NIH), specifically the National Institute of Mental Health
(NIMH). We thank the NIH PET Department for radio-
isotope production and the NIMH Psychoactive Screening
Program (PDSP) for performing assays. The PDSP is directed
by Bryan L. Roth, Ph.D., with project officer Jamie Driscol
(NIMH), at the University of North Carolina at Chapel Hill
(Contract No. NO1MH32004). We thank our colleagues,
Drs. Lisheng Cai and Joong-Hyun Chun, for useful sugges-
tions and technical assistance in some areas of this work.
Radiosynthesis of [11C]8. [11C]Methyl iodide (∼15 mCi)
was trapped in a capped fluoropolymer custom-made reaction
vial (1.5-mL) containing anhydrous DMF (300 μL), precursor
(3, ∼1.0 mg) and base (Li3N or Li2O, ∼5.0 mg). The mixture was
sonicated in an ultrasound apparatus (UIS250L, Hielscher
Ultrasonics, Germany) for 10 min and then filtered through
an Iso-Disc filter (PFTE13-4, 13 mm ꢀ 0.45 μm) which was then
rinsed twice with DMF. The combined filtrates were diluted
with water, and then the [11C]8 was isolated with radio-HPLC
(MeCN-HCOONH4, 2:3 v/v, 3 mL/min, tR = 13.1 min). The
identity of [11C]8 was confirmed by analytical radio-HPLC on a
Gemini-NX C18 column (5 μm, 4.6 mm ꢀ 250 mm) eluted with
MeCN-100 mM HCOONH4 (16:9 v/v) at 1 mL/min (tR = 7.0
min) and also by LC-MS of the associated carrier. The RCYs
from [11C]methyl iodide were 7% (Li2O) and 11% (Li3N).
Radiosynthesis of [11C]13 from [11C]Methyl Triflate. [11C]-
Methyl triflate (30-20 mCi) was trapped in a tapered bottom
vial (0.9 mL) containing 20 (0.15-0.8 mg), 0.5 M NaOH
(3 equiv), and MeCN (300 μL). The mixture was heated at
80 °C for 5 min and diluted with HPLC mobile phase (500 μL),
and then the [11C]13 (tR = 8.7 min) was isolated with radio-
HPLC (MeCN-HCOONH4, 2:3 v/v, 4 mL/min). The identity of
[11C]13 was confirmed by analytical radio-HPLC on a Gemini-
NX C18 column (5 μm, 4.6 mm ꢀ 250 mm) eluted with
MeCN-100 mM HCOONH4 (2:3 v/v) at 1 mL/min (tR = 4.4
min) and also by LC-MS of the associated carrier. The RCY was
27% (n = 12) from [11C]methyl triflate and the radiochemical
purity>99%. The absorbance detector responseofthe analytical
HPLC system was calibrated for the mass of carrier ligand 13.
This allowed the mass of 13 in measured samples of [11C]13 to
be determined and the specific radioactivity to be calculated.
The specific radioactivity was 2517 mCi/μmol at the end of syn-
thesis (EOS).
Supporting Information Available: Chemical purities of com-
pounds 1-21 and their HPLC methods of determination. This
material is available free of charge via the Internet at http://
pubs.acs.org.
References
€
ꢀ ꢁ
(1) Wong, D. F.; Grunder, G.; Brasic, J. R. Brain imaging research:
does the science serve clinical practice? Int. Rev. Psychiatry 2007,
19, 541–558.
(2) Gibson, R. E.; Burns, H. D.; Hamill, T. G.; Eng, W. S.; Francis,
B. E.; Ryan, C. Non-invasive radiotracer imaging as a tool for drug
development. Curr. Radiopharm. Des. 2000, 6, 973–989.
(3) Eglen, R. M.; Wong, E. H. F.; Dumuis, A.; Bockaert, J. Central
5-HT4 receptors. TIPS 1995, 16, 391–398.
(4) Langlois, M.; Fischmeister, R. 5-HT4 receptor ligands: applica-
tions and new perspectives. J. Med. Chem. 2003, 46, 319–344.
(5) Eglen, R. M.; Hegde, S. S. 5-Hydroxytryptamine (5-HT4) recep-
tors: physiology, pharmacology and therapeutic potential. Exp.
Opin. Invest. Drugs 1996, 5, 373–388.
(6) Warner-Schmidt, J. L.; Flajolet, M.; Maller, A.; Chen, E. Y.; Qi,
H. S.; Svenningsson, P.; Greengard, P. Role of p11 in cellular and
behavioral effects of 5-HT4 receptor stimulation. J. Neurosci. 2009,
29, 1937–1946.
(7) Waterhouse, R. N. Determination of lipophilicity and its use as a
predictor of blood brain barrier penetration of molecular imaging
agents. Mol. Imaging Biol. 2005, 5, 376–389.
(8) Pike, V. W. PET Radiotracers: crossing the blood-brain barrier
Radiosynthesis of [11C]13 from [11C]Methyl Iodide. [11C]Methyl
iodide (30-550 mCi) was trapped in a tapered bottom vial
(0.9 mL) containing 20 (∼1.0 mg, 3.2 μmol), 1.0 M (n-Bu)4NOH
(3 equiv, 10 μL, 3.2 μmol), and DMF (300 μL). The mixture was
heated at 80 °C for 5 min and then diluted with water (500 μL).
[11C]13 was separated andanalyzed asdescribed above. The RCY
of[11C]13 from [11C]iodomethane was 36% (n = 6) andthe radio-
chemical purity >99%. The specific radioactivity was 2848
mCi/μmol at EOS.
and surviving metabolism. TIPS 2009, 30, 431–440.
(9) Pike, V. W. Positron-emitting radioligands for studies in vivo;
probes for human psychopharmacology. J. Psychopharmacol.
1993, 7, 139–158.
(10) Laruelle, M.; Slifstein, M.; Huang, Y. Relationships between
radiotracer properties and image quality in molecular imaging of
the brain with positron emission tomography. Mol. Imaging Biol.
2003, 5, 363–375.
(11) Kaumann, A. J.; Gaster, L. M.; King, F. D.; Brown, A. M.
Blockade of human atrial 5-HT4 receptors by SB 207710, a selective

Article
and high affinity 5-HT4 receptor antagonist. Nauyn-Schmiede-
Journal of Medicinal Chemistry, 2010, Vol. 53, No. 19 7047
(32) Marazziti, D.; Lucacchini, A.; Baroni, S.; Betti, L.; Catena, M.;
Ginnaccini, G.; Dell’Osso, B.; Masala, I.; Mungai, F.; Dell’Osso,
L. Presence of D4 dopamine receptors in human prefrontal cortex:
a post mortem study. Rev. Bras. Psiquiatria 2007, 29, 148–152.
(33) Cai, L.; Lu, S.; Pike, V. W. Chemistry with [18F]fluoride ion. Eur. J.
Org. Chem. 2008, 17, 2853–2873.
berg’s Arch. Pharmacol. 1994, 349, 546–548.
(12) McLean, P. G.; Coupar, I. M. 5-HT4 receptor antagonist affinities
of SB207710, SB205008, and SB203186 in the human colon, rat
oesophagus, and guinea pig ileum peristaltic reflex. Nauyn-Schmie-
deberg’s Arch. Pharmacol. 1995, 352, 132–140.
(13) Pike, V. W.; Halldin, C.; Nobuhara, K.; Hiltunen, J.; Mulligan,
R. S.; Swahn, C.-G.; Karlsson, P.; Olsson, H.; Hume, S. P.; Hirani,
E.; Whalley, J.; Pilowsky, L. S.; Larson, S.; Schnell, P.-O.; Ell, P. J.;
Farde, L. Radioiodinated SB 207710 as a radioligand in vivo:
imaging of brain 5-HT4 receptors with SPET. Eur. J. Nucl. Med.
Mol. Imaging 2003, 30, 1520–1528.
(14) Kornum, B. R.; Lind, N. M.; Gillings, N.; Marner, L.; Andersen,
F.; Knudsen, G. M. Evaluation of the novel 5-HT4 receptor PET
ligand [11C]SB207145 in the Gottingen mini-pig. J. Cereb. Blood
Flow Metab. 2009, 29, 186–196.
(34) Pike, V. W.; Aigbirhio, F. I. Reactions of cyclotron-produced
[
18F]fluoride with diaryliodonium salts;a novel single-step route
to no-carrier-added [18F]fluoroarenes. J. Chem. Soc., Chem. Com-
mun. 1995, 2215–2216.
(35) Hamill, T.; Burns, H.; Eng, W.; Ryan, C.; Krause, S.; Gibson, R.;
Hargreaves, R. An improved fluorine-18 labeled neurokinin-1
receptor ligand. Mol. Imaging Biol. 2002, 4 (Suppl. 1), S34.
(36) Schou, M.; Halldin, C.; Sovago, J.; Pike, V. W.; Gulyas, B.;
Mozley, D.; Dobson, D.; Johnson, P. D.; Innis, R. B.; Farde, L.
Evaluation of two fluorinated reboxetine analogs as potential
norepinephrine transporter probes in the monkey brain with
PET. Synapse 2004, 53, 57–67.
(37) Terry, G. E.; Hirvonen, J.; Liow, J.-S.; Zoghbi, S. S.; Gladding, R.;
Tauscher, J. T.; Schaus, J. M.; Phebus, L.; Felder, C. C.; Morse,
C. L.; Donohue, S. R.; Pike, V. W.; Halldin, C.; Innis, R. B.
Imaging and quantitation of cannabinoid CB1 receptors in human
and monkey brain using 18F-labeled inverse agonist radioligands.
J. Nucl. Med. 2010, 51, 112–120.
(38) King, F. D.; Gaster, L. M.; Mulholland, K. R.; Rahman, S. K.;
Wyman, P. A.; Sanger, G. J.; Wardle, K. A.; Baxter, G. S.; Kennett,
G. A.; Kaumann, A. J. 5-HT4 Receptor Antagonists. U.S. Patent
5,580,885, 1996.
(39) Kowalczyk, B. A.; Robinson, J., III; Gardner, J. O. Process
development of the synthetic route to sulamserod hydrochloride.
Org. Process Res. Dev. 2001, 5, 116–121.
(40) Cristau, H.-J.; Ouali, A.; Spindler, J.-F.; Taillefer, M. Mild and
efficient copper-catalyzed cyanation of aryl iodides and bromides.
Chem.;Eur. J. 2005, 11, 2483–2492.
(41) Sellarajah, S.; Lekishvili, T.; Bowring, C.; Thompsett, A. R.;
Rudyk, H.; Birkett, C. R.; Brown, D. R.; Gilbert, I. H. Synthesis
of analogues of Congo Red and evaluation of their anti-prion
activity. J. Med. Chem. 2004, 47, 5515–5534.
(42) Hudgens, T. L.; Turnbull, K. D. C-Methylation of phenols,
tyrosine derivatives, and a tyrosine containing peptide. Tetrahe-
dron Lett. 1999, 40, 2719–2722.
(43) Qiao, C.; Gupte, A.; Boshoff, H. I.; Wilson, D. J.; Bennett,
E. M.; Somu, R. V.; Barry, C. E.; Aldrich, C. C. 50-O-[(N-Acyl)-
sulfamoyl]adenosines as antitubercular agents that inhibit MbtA:
an adenylation enzyme required for siderophore biosynthesis of the
mycobactins. J. Med. Chem. 2007, 50, 6080–6094.
(44) Meth-Cohn, O. Transesterification of methyl esters of aromatic
and R,β-ansaturated acids with bulky alcohols: (D)-menthyl cinna-
mate and (D)-menthyl nicotinate. Org. Synth. 1990, 68, 155–157.
(45) Eaborn, C.; Hornfeld, H. L.; Walton, D. R. M. Preparation of
some aryltrialkyl- and tetraaryl stannanes. J. Organomet. Chem.
1967, 10, 529–530.
(46) Duda, H.; Ostaszynski, A.; Urbanski, T. Halosalicylohydroxamic
acids. I. Dihalosalicylohydroxamic acids. Bull. Acad. Pol. Sci., Ser.
Sci. Chim. 1965, 13, 341–347.
(47) Weller, D. D.; Stirchak, E. P. Quassinoid synthesis via o-quinone
Diels-Alder reactions. J. Org. Chem. 1983, 48, 4873–4879.
(48) Fukuda, Y.; Seto, S.; Furuta, H.; Ebisu, H.; Oomori, Y.;
Terashima, S. Novel seco cyclopropa[c]pyrrolo[3,2-e]indole bisal-
kylators bearing a 3,30-arylenebisacryloyl group as a linker. J. Med.
Chem. 2001, 44, 1396–1406.
(49) Hamill, T. G.; McCauley, J. A.; Burns, H. D. The synthesis of a
benzamidine-containing NR2B-selective NMDA receptor ligand
labelled with tritium or fluorine-18. J. Labelled Compd. Radio-
pharm. 2005, 48, 1–10.
(15) Gee, A. D.; Martarello, L.; Passchier, J.; Wishart, M.; Parker, C;
Matthews, J.; Comley, R.; Hopper, R.; Gunn, R. Synthesis and
evaluation of [11C]SB207145 as the first in vivo serotonin 5-HT4
receptor radioligand for PET imaging in man. Curr. Radiopharm.
2008, 1, 110–114.
ꢁ
(16) Marner, L.; Gillings, N.; Comley, R. A.; Baarre, W. F. C.; Rabiner,
E. A.; Wilson, A. A.; Houle, S.; Hasselbalch, S. G.; Svarer, C.;
Gunn, R. N.; Laruelle, M.; Knudsen, G. M. Kinetic modeling of
11C-SB207145 binding to 5-HT4 receptor in the human brain
in vivo. J. Nucl. Med. 2009, 50, 900–908.
(17) Rappaport, S. I.; Ohno, K.; Pettigrew, K. D. Drug entry into the
brain. Brain Res. 1979, 172, 254.
€
(18) Larsen, P.; Ulin, J.; Dahlstrom, K.; Jensen, M. Synthesis of
[
11C]iodomethane by iodination of [11C]methane. Appl. Radiat.
Isot. 1997, 48, 153–157.
(19) Jewett, D. M. A simple synthesis of [11C]methyl triflate. Appl.
Radiat. Isot. 1992, 43, 1383–1385.
€
(20) Christman, D. R.; Finn, R. D.; Karlstrom, K.; Wolf, A. P. The
production of ultra high specific activity 11C-labeled hydrogen
cyanide, carbon dioxide, carbon monoxide and methane via the
14N(p,R)11C reaction. Int. J. Appl. Radiat. Isot. 1975, 26, 435–442.
(21) Donohue, S. R.; Pike, V. W.; Finnema, S. J.; Truong, P.; Anders-
ꢁ
son, J.; Gulyas, B.; Halldin, C. Discovery and labeling of high
affinity 3,4-diarylpyrazolines as candidate radioligands for in vivo
imaging of cannabinoid subtype-1 (CB1) receptors. J. Med. Chem.
2008, 51, 5608–5616.
(22) Luthra, S. K.; Pike, V. W.; Brady, F.; Turton, D. R.; Wood, B.;
Matthews, R. W.; Hawkes, G. E. The utility of 13C/11C-co-labelling
and subsequent 13C-NMR in the characterisation of 11C-labelled
products. J. Labelled Compd. Radiopharm. 1987, 25, 1070–1072.
(23) Vickery, R. G.; Mai, N.; Kaufman, E.; Beattie, D. T.; Pulido-Rios,
T.; O’Keefe, M.; Humphrey, P. P. A.; Smith, J. A. M. A compar-
ison of the pharmacological properties of guinea pig and human re-
combinant 5-HT4 receptors. Br. J. Pharmacol. 2007, 150, 782–791.
(24) Clark, R. D. Medicinal Chemistry of 5-HT4 Receptor Ligands. In
5-HT4 Receptors in the Brain and Periphery; Eglen, R. M., Ed.;
Landes Bioscience: Austin, TX, 1997; Chapter 1, pp 1-48.
(25) Gaster, L. M.; Jennings, A. J.; Joiner, G. F.; King, F. D.; Mulhol-
land, K. R.; Rahman, S. K.; Starr, S.; Wyman, P. A.; Wardle,
K. A.; Wardle, E. S. E.; Gareth, J. S. (1-Butyl-4-piperidinyl)methyl
8-amino-7-chloro-1,4-benzodioxane-5-carboxylate hydrochloride:
a highly potent and selective 5-HT4 receptor antagonist derived
from metoclopramide. J. Med. Chem. 1993, 36, 4121–4123.
(26) McCarron, J. A.; Zoghbi, S. S.; Shetty, S. S.; Ichise, M.; Vermeulen,
€
E. S.; Wikstrom, H. V.; Halldin, C.; Innis, R. B.; Pike, V. W.
Synthesis and preliminary evaluation of [11C](-)-RWAY in
monkey;a new simply labeled PET radioligand for imaging
brain 5-HT1A receptors. Eur. J. Nucl. Med. Mol. Imaging 2007,
34, 1670–1682.
(27) Kenakin, T. Efficacy as a vector: the relative prevalence and
paucity of inverse agonism. Mol. Pharmacol. 2004, 66, 2–11.
(28) Kursar, J. D.; Nelson, D. L.; Wainscott, D. B.; Baez, M. Molecular
cloning, functional expression, and mRNA tissue distribution of
the human 5-hydroxytryptamine2B receptor. Mol. Pharmacol.
1994, 46, 227–234.
(50) Yadav, J. S.; Reddy, B. V. S.; Madan, Ch.; Hashim, S. R. A mild
and chemoselective dealkylation of alkyl aryl ethers by cerium(III)
chloride-NaI. Chem. Lett. 2000, 29, 738–739.
(51) Ghiaci, M.; Asghari, J. Dealkylation of alkyl and aryl ethers with
AlCl3-NaI in the absence of solvent. Synth. Commun. 1999, 29,
973–979.
(29) Duxon, M. S.; Flanigan, T. P.; Reavley, A. C.; Baxter, G. S.;
Blackburn, T. P.; Fone, K. C. F. Evidence for expression of the
5-hydroxytryptamine-2B receptor in the rat central nervous sys-
tem. Neurosci. 1997, 76, 323–329.
(30) Hashimoto, K.; Ishiwata, K. Sigma receptor ligands: possible
application as therapeutic drugs and as radiopharmaceuticals.
Curr. Pharm. Des. 2006, 12, 3857–3876.
(31) Marazziti, D.; Baroni, S.; Masala, I.; Ginnaccini, G.; Betti,
L.; Palego, L.; Dell’Osso, M. C.; Consoli, G.; Castagna, M.;
Lucacchini [3H]-YM-09151-2 binding sites in human brain post
mortem. Neurochem. Int. 2009, 55, 643–647.
(52) Gaster, L. M.; Wyman, P. A. Preparation of Aminoalkyl Tricyclic
Heterocyclecarboxylates as 5-HT4 Receptor Antagonists. PCT Int.
Appl. WO 9417071 A1 19940804, 1994.
(53) Grossman, C. J.; Gale, J. D.; Bunce, K. T.; Kilpatrick, G. J.;
Whitehead, J. W. F.; Oxford, A. W.; Humphrey, P. P. A. Devel-
opment of a radioligand binding assay for 5HT4 receptors in guinea
pig and rat brain. Br. J. Pharmacol. 1993, 109, 618–624.
(54) Chin, F. T.; Morse, C. L.; Shetty, H. U.; Pike, V. W. Automated
radiosynthesis of [18F]SPA-RQ for imaging human brain NK1
receptors with PET. J. Labelled Compd. Radiopharm. 2006, 49,
17–31.