Synthesis and in vivo evaluation of [18F]2-(4-(4-(2-(2- fluoroethoxy)phenyl)piperazin-1-yl)butyl)-4-methyl-1,2,4-triazine-3,5(2H,4H) -dione ([18F]FECUMI-101) as an imaging probe for 5-HT1A receptor agonist in nonhuman prim

Article abstract of DOI:10.1016/j.bmc.2013.05.050

The 5-HT_1AR partial agonist PET radiotracer, [ ¹¹C]CUMI-101, has advantages over an antagonist radiotracer as it binds preferentially to the high affinity state of the receptor and thereby provides more functionally meaningful inform

Full text of DOI:10.1016/j.bmc.2013.05.050

Bioorganic & Medicinal Chemistry 21 (2013) 5598–5604
Contents lists available at SciVerse ScienceDirect
Bioorganic & Medicinal Chemistry
journal homepage: www.elsevier.com/locate/bmc
Synthesis and in vivo evaluation of [18F]2-(4-(4-(2-(2-fluoroeth-
oxy)phenyl)piperazin-1-yl)butyl)-4-methyl-1,2,4-triazine-3,5(2H,4H)-
dione ([18F]FECUMI-101) as an imaging probe for 5-HT1A receptor
agonist in nonhuman primates
Vattoly J. Majo a, Matthew S. Milak a,b, Jaya Prabhakaran a, Pratap Mali b, Lyudmila Savenkova a,
Norman R. Simpson b, J. John Mann a,b,c, Ramin V. Parsey d, J. S. Dileep Kumar a,b,
⇑
a Department of Psychiatry, Columbia University Medical Center, Columbia University, 1051 Riverside Drive, Box #42, New York, NY 10032, United States
b Division of Molecular Imaging and Neuropathology, New York State Psychiatric Institute, New York, United States
c Department of Radiology, Columbia University, New York, NY 10032, United States
d Department of Psychiatry and Behavioral Science, Stony Brook University, New York, United States
a r t i c l e i n f o
a b s t r a c t
The 5-HT1AR partial agonist PET radiotracer, [11C]CUMI-101, has advantages over an antagonist radio-
tracer as it binds preferentially to the high affinity state of the receptor and thereby provides more func-
tionally meaningful information. The major drawback of C-11 tracers is the lack of cyclotron facility in
many health care centers thereby limiting widespread clinical or research use. We identified the fluoro-
Article history:
Received 12 March 2013
Revised 9 May 2013
Accepted 17 May 2013
Available online 12 June 2013
ethyl
derivative,
2-(4-(4-(2-(2-fluoroethoxy)phenyl)piperazin-1-yl)butyl)-4-methyl-1,2,4-triazine-
3,5(2H,4H)dione (FECUMI-101) (Ki = 0.1 nM; Emax = 77%; EC50 = 0.65 nM) as a partial agonist 5-HT1A
R
Keywords:
PET
FECUMI-101
5-HT1AR
Agonist
ligand of the parent ligand CUMI-101. FECUMI-101 is radiolabeled with F-18 by O-fluoroethylation of
the corresponding desmethyl analogue (1) with [18F]fluoroethyltosylate in DMSO in the presence of
1.6 equiv of K2CO3 in 45 5% yield (EOS). PET shows [18F]FECUMI-101 binds specifically to 5-HT1A
R
enriched brain regions of baboon. The specificity of [18F]FECUMI-101 binding to 5-HT1AR was confirmed
by challenge studies with the known 5-HT1AR ligand WAY100635. These findings indicate that [18F]FEC-
UMI-101 can be a viable agonist ligand for the in vivo quantification of high affinity 5-HT1AR with PET.
Ó 2013 Elsevier Ltd. All rights reserved.
Brain
Radiotracer
1. Introduction
antipsychotics like aripiprazole are also partial agonists at the
5-HT1AR and used as antidepressants.9,10 5-HT1AR desensitization
The serotonin 1A receptor (5-HT1AR) is a G protein-coupled
receptor (GPCR) coupled to Gi/Go.1 5-HT1AR is both a somatoden-
dritic autoreceptor on serotonin neurons in the midbrain and
pons, and a widespread and heterogeneously distributed post-
synaptic receptor in the terminal fields.2 Somatodendritic 5-
HT1AR autoreceptors are present on serotonergic neurons in
brainstem raphe nuclei and inhibit serotonin neuron firing after
5-HT release from recurrent fibers. 5-HT1AR heteroreceptors
modulate or regulate release of other neurotransmitters, pro-
teins, hormones and growth factors in brain.3–5 5-HT1AR play a
role in cognition, learning, memory, aggression, sociability,
impulsivity, addictive behavior, sexual behavior, food intake or
appetite, prolongation of rapid eye movement (REM), sleep la-
and increased serotonin neuron firing is observed in animal
studies and hypothesized to be part of the mechanism of antide-
pressant action of selective serotonin reuptake inhibitors (SSRIs),
serotonin-norepinephrine reuptake inhibitors (SNRIs), tricyclic
antidepressants (TCAs), tetracyclic antidepressants (TeCAs), and
monoamine oxidase inhibitors (MAOIs).3–7
Quantitative imaging of the serotonergic system is of critical
importance for understanding the pathophysiology of psychiatric
and neurological illnesses.11–13,17–21 Positron Emission Tomogra-
phy (PET) has the sensitivity to measure 5-HT1ARs in living brain,
and antagonist PET ligands (e.g., [11C]WAY-100635, [18F]MPPF,
[
11C]FCWAY and [18F]MeFWAY) are the commonly used radiotra-
cers for imaging 5-HT1ARs14–17. However, 5-HT1AR antagonist
ligands bind with equal affinity to high agonist affinity (HA) and
low agonist affinity (LA) sites of GPCRs, and cannot detect changes
or differences in HA sites that are only a proportion of the total
number of receptors.22 In contrast, agonist ligands preferentially
tency, regulation of respiration, emesis and analgesia.6 5-HT1A
R
partial agonists (e.g., buspirone and tandospirone) are currently
approved drugs for anxiety and depression.7,8 Some atypical
bind to the HA receptor conformation thereby providing
more meaningful functional measure of 5-HT1AR transduction
a
⇑
Corresponding author. Tel.: +1 (212) 543 1163; fax: +1 (212) 543 1054.
E-mail address: kumardi@nyspi.columbia.edu (J. S. Dileep Kumar).
0968-0896/$ - see front matter Ó 2013 Elsevier Ltd. All rights reserved.
http://dx.doi.org/10.1016/j.bmc.2013.05.050

V. J. Majo et al. / Bioorg. Med. Chem. 21 (2013) 5598–5604
5599
capacity.22 The ratio of [3H]WAY100,635: [3H]8-OH-DPAT or
2. Results
antagonist: agonist receptor binding in human neocortex is 1.7–
2.7 across different brain regions, using quantitative autoradiogra-
phy).23 In rat hippocampal tissue, the LA: HA ratio average is
ꢀ1.5.22 Alterations in the HA affinity site of 5-HT1AR receptors
may be an important factor in the pathophysiology of diseases,
and is part of the hypothesized action of many antidepressants.
Therefore, obtaining PET scans with an agonist ligand in the same
subject would enable the quantification of binding by GPCRs, that,
in turn, is related to the level of signal transduction by endogenous
5-HT. Moreover, an agonist 5-HT1AR radiotracer is likely to be sen-
sitive to changes in intra-synaptic concentrations of the endoge-
nous serotonin, measures desensitization (downregulation) or
sensitization (upregulation) of GPCRs, provides a better estimate
of receptor occupancy for agonist therapeutic agents and for eval-
uation of the efficacy of SSRI treatment.24–30 The translocator pro-
2.1. Chemistry, pharmacology and radiochemistry
The reference standard FECUMI-101 was synthesized by heat-
ing desmethyl-CUMI-101 (1) with bromoethyl fluoride (1.5 equiv)
for 4 h at 80 °C in the presence of excess K2CO3 (5 equiv) in DMF in
65% yield. Synthesis of 1 was achieved using a procedure devel-
oped by us.36 The affinity of FECUMI-101 for 5-HT1AR and various
other biogenic amine receptors and transporters and functional as-
says were determined through the National Institute of Mental
Health Psychoactive Drug Screening Program (NIMH-PDSP). Bind-
ing assay shows that FECUMI-101 has 0.1 nM binding affinity (Ki)
for 5-HT1AR. FECUMI-101 has a Ki of 21.4 0.6 nM for alpha1A
R
and exhibits nanomolar affinities for 5-HT7R (Ki = 17.2 nM), D2R
(Ki = 37.1 nM), D3R (Ki = 22 nM) and D4R (Ki = 71 nM) receptors
(Table 1). The Kis for various other brain receptors and transporters
tein 18 kDa (TSPO) radioligand [18F]DPA-714,
radioligand [11C]GR103545, D2/D3 radiotracers [11C]PHNO and
11C]NPA, 5-HT2A radioligand [11C]CIMBI-36 are some of the ago-
nist PET tracers that appears to be superior to the corresponding
antagonist tracers in vivo.31–35 11C]CUMI-101, developed by us,
j-Opioid receptor
were low (0.1–10
HT1AR were evaluated using [35S]GTP
of CHO cells stably expressing the human 5-HT1AR. FECUMI-101
produced a dose-dependent increase in [35S]GTP
S binding. Maxi-
mal FECUMI-101 stimulated [35S]GTP
S binding for 5-HT1AR was
lM). Agonist properties of FECUMI-101 on 5-
[
cS formation in membranes
[
c
is the only successful partial agonist radiotracer that binds to the
HA state of 5-HT1AR in nonhuman primates and human
subjects.36–39 We have also demonstrated the ability of
c
77% that of serotonin (Emax) with an EC50 of 0.65 nM. The affinity,
cross selectivity and functional properties of CUMI-101 and FEC-
UMI-101 are presented in Table 1.
[
11C]CUMI-101 to measure robust increases in intrasynaptic
Radiosynthesis of
two-step fluroalkylation of the phenolate of 1 (Scheme 1). For this
purpose, initially we synthesized
18F]fluoroethyltosylate by
radioflourination of ethyleneglycol ditosylate with K[18F], kryptofix
[ a
18F]FECUMI-101 was performed by
endogenous serotonin in baboons by iv citalopram and fenflura-
mine administration.40,41 As predicted, lower BPF was observed
for [11C]CUMI-101 scans in comparison with [11C]WAY-100635
scans (ꢀ50%) for 5-HT1AR—rich regions such as hippocampus,
amygdala, cortex and cingulate regions.42 These findings are
consistent with autoradiography studies using tritiated ligands
8-OH-DPAT and WAY100635.22,23 Although [11C]CUMI-101 is a
promising radiotracer in human, its major limitation in serving
clinical populations is the requirement for an onsite cyclotron to
produce C-11. However, F-18 tracer can be shipped to multiple
medical centers that are within a radius of 4 h transportation time
from the production site and the 110-min half-life of F-18 labeled
agonist tracer also permits imaging for a longer duration facilitat-
ing tracer kinetic modeling studies. Moreover, the F-18 positron
energy is the lowest of the first row positron emitters and imaging
can be potentially done at the highest resolution45–47. Hence devel-
opment of an F-18 labeled 5-HT1AR agonist tracer may have many
advantages. We have developed several fluoro analogues of various
5-HT1AR agonists as PET ligands with limited success. However, 2-
(4-(4-(2-(2-fluoro-ethoxy)phenyl)piperazin-1-yl)butyl)-4-methyl-
1,2,4-triazine-3,5(2H,4H)-dione (FECUMI-101), the fluoroethyl
analogue of CUMI-101 shows promising characteristics as a 5-
HT1AR imaging agent. Herein we describe radiosynthesis and
in vivo evaluation of [18F]FECUMI-101 in nonhuman primates with
PET (Scheme 1).
[
K222 and K2CO3 in 90% crude yield.43
[
18F]fluoroethyltosylate
([18F]FETos) was then purified via semiprep RP HPLC (75%). When
LiChrolut EN column was used instead of C-18 Sep-PakÒ to trap
[
18F]FETos, a 15% better efficiency was observed. [18F]FETos reacted
with 1 in the presence of a base to afford [18F]FECUMI-101 ([18F]2).
The starting material was unreacted in acetonitrile at temperatures
below 100 °C. When aqueous NaOH was used as the base,
[
18F]FCH2CH2OTs decomposed at temperatures above 125 °C. The
anhydrous sodium salt of precursor obtained by azeotropic distilla-
tion of precursor with 5 M NaOH (1.8 equiv) in acetonitrile reacted
with [18F]FCH2CH2OTs in DMSO at 110 °C to yield [18F]2 in 25%
yield (EOS). Better yields were obtained by using 1.6 equiv of aque-
ous K2CO3 (25 lL, 5%) as a base at 110 °C. The crude product was
purified by RP-HPLC followed by C-18 Sep-PakÒ purification to ob-
tain [18F]FECUMI-101 in 45 5% yield at EOS (n = 10). Specific
activity obtained for [18F]FECUMI-101 was 92.5 18.5 GBq/
lmol
(n = 10) based on a standard mass curve with >95% chemical and
radiochemical purities. The total time required for the radiolabel-
ing process was 60 min at EOS. A one-pot modification of the pro-
cedure was later developed to provide [18F]FECUMI-101, albeit, in
slightly lower yields (20–25%). The logPo/w of [18F]2 from shake
Scheme 1. Radiosynthesis of [18F]FECUMI-101.

5600
V. J. Majo et al. / Bioorg. Med. Chem. 21 (2013) 5598–5604
2.3. PET imaging studies with [18F]FECUMI-101 in baboon
Table 1
Comparison of binding affinities and GTP
and CUMI-101
c
functional assay results of FECUMI-101
PET studies in anesthetized baboon (Papio anubis) showed that
18F]FECUMI-101 (111 18 MBq) penetrated the BBB and was re-
Binding assays
Ki (nM)
[
tained in 5-HT1AR rich areas such as hippocampus, insula, cingulate
cortex, prefrontal cortex, and amygdala, whereas the striatum
exhibited lower binding and cerebellum had the least amount of
binding (Fig. 2). No skull image due to in vivo radio[18F]defluorina-
tion was observed. The time activity curves (TACs) show radioac-
tivity in all regions reached a peak by 3–13 min post injection
and a rapid clearance was observed for cerebellum. The binding
ratios of hippocampus, anterior cingulate cortex, insular cortex,
cingulate cortex, prefrontal cortex, amygdala, and dorsal raphe
nucleus to cerebellum were 4.0, 2.5, 2.3, 1.8, 1.8, 2.4 and 1.75 at
175 min. Other cortical regions with known 5-HT1AR binding show
ꢀtwofold binding in comparison to cerebellum. Surprisingly,
thalamus showed moderate binding with [18F]FECUMI-101. The
binding ratio of thalamus to cerebellum was 2.0 at 175 min. TACs
show that radioligand reached equilibrium in 40 min, hence
120 min scan will be suitable for baboon PET studies with [18F]
FECUMI-101 (Fig. 3). The specificity of the radioligand uptake
was determined by chase studies by the administration of 5-HT1AR
antagonist WAY-100,635 (0.5 mg/kg/i.v) 50–60 min after the injec-
tion of [18F]FECUMI-101 (Figs. 2 and 4). No significant changes in
metabolism or protein binding were observed in chase challenge
administration studies with [18F]FECUMI-101. The blocking study
showed displacement of radioactivity in all regions close to
cerebellar level.
FECUMI-101
CUMI-101
5HT1A
5HT1B
5HT1D
5HT2A
5HT2B
5HT5A
5HT7
Adrenergic a1A
Adrenergic a1B
Adrenergic a1D
Adrenergic a2A
Adrenergic a2B
Adrenergic a2C
BZP rat brain site
Dopamine1
Dopamine2
Dopamine3
Dopamine4
Histamine1
Histamine2
Sigma 1
0.1
411
0.15
643
88.3
4,975
73.6
>10,000
12.9
6.75
>10,000
>10,000
96
234.6
>10,000
300
2128
17.2
21.4
26.8
62.6
2751
56.2
173.7
6238
>10,000
37.1
22
71
238
15
>10,000
>10,000
>10,000
>10,000
21
1030
2126
59
512
1767
>10,000
1726
Sigma 2
123
Agonist assay (GTP-cS)
Emax
EC50
77%
0.65 nM
80%
0.1 nM
flask method was 1.2.44 Stability of [18F]2 formulation (10% etha-
nol-90% saline) used for in vivo studies was analyzed using analyt-
ical HPLC and radio-TLC and indicated that the radioproduct was
stable for up to 4 h with no significant de[18F]fluorination in the
formulation.
3. Discussion
Alteration of serotonin 1A receptor (5-HT1AR) binding has been
implicated in a number of neuropsychiatric and neurodegenerative
disorders. It is the HA site that binds to g proteins and is responsible
for signal transduction. Hence imaging using an agonist PET tracer is
required to measure 5-HT1AR in vivo. [11C]WAY100,635, has been
widely used for in vivo quantification of 5-HT1AR with PET. One of
the major impediments to using this technology in clinics is the lim-
ited number of centers that can produce C-11 labeled radiotracers.
We identified [11C]CUMI-101 as an agonist PET tracer for imaging
HA 5-HT1ARs in human and nonhuman primates.36–42
2.2. Determination of unchanged radioligand in plasma
Figure 1 shows the time course of the unmetabolized fraction of
[
18F]FECUMI-101 in baboon plasma. [18F]FECUMI-101 showed
35 1.5% (n = 12) and 11.5 2.5% (n = 12) binding in standardized
baboon and human samples respectively. HPLC analyses of the
plasma samples indicated only polar metabolites and the percent-
age of unmetabolized [18F]FECUMI-101 was 98 1.0% at 2 min,
74 2.5% at 12 min, 51% at 30 min, 37 4.0% at 60 min,
32.5 0.5% at 90 min, 23.5 2% at 120 min, 21.5 2% at 150 min
and 18 2% at 180 min, respectively.
We identified FECUMI-101 as a candidate agonist PET tracer
through structure activity relationship studies of CUMI-101. FEC-
UMI-101 has comparable binding affinity and agonist properties
for 5-HT1AR (Ki = 0.1 nM; Emax = 77%, EC50 = 0.65 nM) to those of
the parent ligand CUMI-101 (Ki = 0.15 nM; Emax = 80%, EC50
=
0.1 nM) (Table 1). FECUMI-101 was also found to be highly selec-
tive for 5-HT1AR over the tested panel of brain biogenic amine
receptors, transporters and proteins that were tested through
NIMH PDSP. Both FECUMI-101 and CUMI-101 displayed compara-
ble affinities for most of the target receptors and transporters.
Some advantages of FECUMI-101 versus CUMI-101 is its lower
binding to alpha1AR (Ki = 6.75 nM) and sigma1R (Ki = >10,000 nM)
in comparison to a Ki of 21.4 and 59 nM, respectively for CUMI-
101. However, FECUMI-101 has Kis of 37.1 and 22 nM for D2R
and D3Rs, whereas CUMI-101 did not show any significant binding
to these receptors. However, despite its relatively higher affinity
for D2R and D3Rs, negligible binding was observed in the striatum
and therefore, the excellent affinity, selectivity and agonistic
properties of FECUMI-101 for 5-HT1AR and the availability of
F-18 labeling site makes it a promising PET ligand for 5-HT1AR.
The tethering of [18F]fluoroethyl group to phenolate of desm-
ethyl-CUMI-101 was achieved in 45 5% yield (EOS) with excellent
chemical, radiochemical purities and specific activity (Scheme 1).
The total time required for the radiosynthesis was 60 min at EOS.
100
90
80
70
60
50
40
30
20
10
0
0
20
40
60
80
Time (min.)
Figure 1. Unmetabolized parent fraction of [18F]FECUMI-101 in baboon plasma.
100
120
140
160
180

V. J. Majo et al. / Bioorg. Med. Chem. 21 (2013) 5598–5604
5601
Figure 2. Sum of 60–180 min of [18F]FECUMI-101 image in baboon. 1st row: Base line; 2nd row: Chase with WAY100635. First column: sagittal, middle column: coronal, last
column: axial views.
CER
AMY
0.3
0.25
0.2
CIN
HIP
PFC
TEM
INS
CAU
THA
ACN
0.15
0.1
0.05
0
0
20
40
60
80
100
120
140
160
180
Time (min.)
Figure 3. Baseline time activity curves of
[
18F]FECUMI-101 in baboon. AMY = Amygdala, CAU = Caudate, CER = cerebellum, CIN = cingulate, HIP = hippocampus,
PFC = prefrontal cortex, TEM: temporal cortex, INS: insular cortex, THA: thalamus, ACN: anterior cingulate
The lipophilicity value of 1.2 0.5, measured in terms of logPo/w, is
favorable for BBB penetration of [18F]FECUMI-101. Radiolabeled
metabolites found in baboon plasma during PET scan were ana-
Highest uptake of [18F]FECUMI-101 radioactivity in baboon was
found in hippocampus, anterior cingulate cortex, insular cortex
and amygdala. This is consistent with our previous reports of
lyzed using a reverse phase HPLC column coupled with
c-detector.
[
11C]CUMI-101 binding in baboon and human.37–40 Dorsal raphe
Only polar metabolites were detected in the analyses and these
metabolites are unlikely to cross the BBB because of their high
polarity and hence the image obtained could be attributed to the
parent [18F]FECUMI-101. The radioligand also has a measurable
free fraction and this will enable receptor quantification using free
fraction corrected arterial input functions. Distribution of [18F]FEC-
UMI-101 binding in baboon brain corresponds to the known distri-
bution of 5-HT1AR for human and nonhuman primates.36,42,48
nucleus, temporal cortex, parahippocampal gyrus, pareital cortex,
and prefrontal cortex show moderate binding of radioligand. Con-
sistent with previous reports for 5-HT1AR PET tracers, striatum and
occipital cortex regions show lower binding, whereas, cerebellum
shows the least binding of the radiolgand.36,42,48 To our surprise
thalamus shows higher binding of [18F]FECUMI-101 in comparison
to [11C]CUMI-101 and [11C]WAY100635. The specificity of the
[
18F]FECUMI-101 binding was determined via a chase study with

5602
V. J. Majo et al. / Bioorg. Med. Chem. 21 (2013) 5598–5604
0.25
0.2
CER
AMY
CIN
HIP
PFC
CAU
INS
0.15
0.1
THA
TEM
ACN
0.05
0
0
20
40
60
80
100
120
140
160
180
Time (min.)
Figure 4. Time activity curves of [18F]FECUMI-101 in baboon after WAY100635 chase.
WAY100635 (0.5 mg/kg iv) and we obtained excellent blockade of
radioligand binding throughout the brain including thalamus. The
administration of WAY100635 was performed at a time point clo-
ser to the time frame in which radioligand reached equilibrium as
evident from baseline TACs (Fig. 3).
spectrometer. Spectra were recorded in CDCl3 or CD3OD and chem-
ical shifts and reported in ppm relative to TMS as internal standard.
The mass spectra were recorded on JKS-HX 11UHF/HX110 HF Tan-
dem Mass Spectrometer in the FAB+ mode. Thin layer chromatog-
raphy (TLC) was performed using silica gel 60 F254 plates from E
Merck. HPLC analyses were performed using a Waters 1525 binary
HPLC system (analytical: Phenomenex, Prodigy ODS(3)
We believe radioligand binding to thalamus is not due to its 5-
HT1AR binding as it is well established that thalamus has low 5-
HT1AR concentration. Since the displaceable PET signal consists of
density multiplied by affinity, a high Bmax and high Kd will give a
large PET signal compared to the corresponding signal from a rel-
atively low abundance receptor. Receptor distribution (Bmax) of
colocalized targets 5-HT1A and 5-HT7 in thalamus are reported as
4.16 and 12 fmol/mg of tissue, respectively.49 Therefore, corre-
sponding ratios of 5-HT7 to 5-HT1A receptor in thalamus is 2.88.
In order to avoid more than 10% PET signal interference from the
5-HT7 receptors in thalamus, a 29-fold Kd difference over the 5-
HT7 receptor is sufficient. Since FECUMI-101 is 172-fold selective
to 5-HT1A over 5-HT7, no contribution to the PET signal is expected
from 5-HT7R in thalamus. Radioligand binding in striatum is very
low, ruling out the possibilities of D2R and D3R binding by [18F]FEC-
UMI-101.50,51The ratios of FECUMI-101 binding to 5-HT1AR in
comparison with adrenergic a1 receptor (a1 AR) sub types ranges
from 210 to 620 with Ki values 21–62 nM. In the case of in vivo
studies with CUMI-101, we did not find much binding of
4.6 Â 250 mm, 5
l
column; semipreparative: Phenomenex, Prod-
column). Flash column chroma-
igy ODS-prep 10 Â 250 mm, 10
l
tography was performed on silica gel (Fisher 200–400 mesh)
using the solvent system indicated. [18F]fluoride was produced
from an RDS112 cyclotron (Siemens, Knoxville, TN). For detection
of radiolabeled [18F]FECUMI-101, gamma ray detector (Bioscan
Flow-Count fitted with a NaI detector) was used in series with
the UV detector (Waters Model 996 set at 254 nm). Data acquisi-
tion for both the analytical and preparative systems was accom-
plished using a Waters Empower Chromatography System. PET
studies were performed in baboon using an ECAT EXACT HR+ scan-
ner (Siemens, Knoxville, TN). All animal experiments were carried
out with the approval of the Institutional Animal Care and Use
Committee of Columbia University Medical Center and New York
State Psychiatric Institute. Metabolite analyses were performed
using Phenomenex Prodigy column (ODS3, 4.6 Â 250 mm, 5
l).
The free fractions and metabolites were measured using Packard
[
11C]CUMI-101 in high a1 AR regions, despite its 6.45 nM Ki for
Instruments Gamma Counter (Model E5005, Downers Grove, IL).
the a1A AR, which is 45-fold less affinity in comparison to 5-HT1AR.
For example, there is low binding of [18F]FECUMI-101 in cerebel-
lum and striatum areas with moderate a1 ARs.52–54
4.2. Synthesis of 2-(4-(4-(2-(2-fluoroethoxy)phenyl)piperazin-1-
yl)butyl)-4-methyl-1,2,4-triazine-3,5(2H,4H)-dione (2)
The challenge doses of WAY100635 we used were 0.5 mg/kg
based on our previous experience36,48 that this dose would provide
substantial blockade of 5-HT1AR. Although WAY100635 is a gold
standard ligand for 5-HT1AR, it has considerable affinity for several
other receptors.55 Apart from the thalamic binding, [18F]FECUMI-
101 brain binding distribution corresponds to that of 5-HT1AR. Fur-
ther studies are required to establish the target responsible for the
thalamic binding of [18F]FECUMI-101.
1-Bromo-2-fluoroethane (26.4 mg, 0.208 mmol) and potassium
carbonate (96 mg, 0.694 mmol) were added to the solution of 2-(4-
(4-(6-hydroxypyridin-2-yl)piperazin-1-yl)butyl)-4-methyl-1,2,4-
triazine-3,5(2H,4H)dione (1) (50 mg, 0.139 mmol) in DMF (vol-
ume: 2 mL). The reaction mixture was heated at 80 °C for 4 h.
The progress of the reaction was followed by reverse phase HPLC
(Phenomenex, Prodigy ODS(3) 4.6 Â 250 mm, 5
l; mobile phase
40: 60 acetonitrile: 25 mM Na2HPO4, flow rate 2 mL/min; the pre-
cursor appeared at 4.9 min and the fluoride appeared at 7.6 min).
After the complete consumption of the phenolic precursor by
HPLC, the reaction mixture was cooled to rt, diluted with water, ex-
tracted with ethyl acetate (3 Â 30 mL). The combined organic lay-
ers were washed with brine, dried over MgSO4 and concentrated
under vacuum. The crude residue was column chromatographed
using 2% methanol in CH2Cl2 to give the fluoroethyl-CUMI-101
(2) as a viscous liquid (37 mg, 65%).
4. Materials and methods
4.1. General
The commercial chemicals and solvents used in the synthesis
were purchased from Sigma–Aldrich Chemical Co. (St. Louis, MO),
Fisher Scientific Inc. (Springfield, NJ), or Lancaster (Windham,
NH) and were used without further purification. Analytical grade
reagents were purchased from standard commercial sources. 1H
NMR spectra were recorded on a Bruker PPX 300 and 400 MHz
1H NMR (400 MHz, CDCl3) d 7.39 (s, 1H), 6.96 (dq, J = 5.8, 3.2 Hz,
3H), 6.84 (dd, J = 5.8, 3.2 Hz, 1H), 4.87–4.80 (m, 1H), 4.75–4.67 (m,

V. J. Majo et al. / Bioorg. Med. Chem. 21 (2013) 5598–5604
5603
1H), 4.34–4.26 (m, 1H), 4.25–4.16 (m, 1H), 4.02 (t, J = 7.1 Hz, 2H),
4.4. PET studies in baboons
3.34 (s, 3H), 3.13 (br s, 4H), 2.65 (br s, 4H), 2.45 (t, J = 7.6 Hz,
2H), 1.86–1.74 (m, 2H), 1.59 (tt, J = 9.8, 6.1 Hz, 2H); 13C NMR
PET studies were performed according to a protocol approved
by the Institutional Animal Care and Use Committee of Columbia
University Medical Center and New York State Psychiatric Insti-
tute. PET scans were performed in two male baboons with an ECAT
EXACT HR+ scanner (CPS/Knoxville, TN). The fasted animal was
immobilized with ketamine (10 mg/kg, im) and anesthetized with
1.5–2.0% isoflurane via an endotracheal tube. Core temperature
was kept constant at 37 °C with a heated water blanket. An intra-
venous infusion line with 0.9% NaCl was maintained during the
experiment and used for hydration and radiotracer injection. An
arterial line was placed for obtaining arterial samples for the input
function. The head was positioned at centre of the field of view,
and a 10 min transmission scan was performed before the tracer
injection. For each scan, 111 18 MBq of [18F]FECUMI-101 (specific
(100 MHz, CDCl3)
d 156.24, 150.98, 148.85, 141.97, 133.77,
122.73, 122.18, 118.47, 113.72, 82.88, 81.19, 67.70, 67.50, 58.08,
53.50, 51.76, 50.46, 26.98, 26.24, 23.72; HPLC (phenomenex, prod-
igy ODS(3) 4.6 Â 250 mm, 5
l; mobile phase; 40:60 acetonitrile/
25 mM Na2HPO4, flow rate 2 mL/min, tR ꢀ7.6 min); HRMS Calcd
for C20H29O3N5F (MH+): 406.2254; Found: 406.2268.
4.3. Radiosynthesis of [18F]2-(4-(4-(2-(2-
fluoroethoxy)phenyl)piperazin-1-yl)butyl)-4-methyl-1,2,4-
triazine-3,5(2H,4H)-dione([18F]FECUMI-101)
An aqueous solution of [18F]fluoride was treated with 200
lL
of 15:1 acetonitrile: water containing kryptofix K222 (36 mg) and
potassium carbonate (2 mg). The reaction mixture was azeotrop-
ically heated and dried at 98° C under a stream of argon by the
repeated addition of acetonitrile (3 Â 0.5 mL). A solution of 4–
5 mg of diethyl tosylate in 1 mL of anhydrous acetonitrile was
then added to the reaction vial, sealed, and heated for 10 min
at 95 °C. The reaction mixture was allowed to cool to room tem-
perature and then injected onto a semipreparative HPLC column
activity of 93 18 GBq/lmol) was injected as an iv bolus and emis-
sion data were collected for 180 min in 3-D mode. Plasma samples
were taken every 10 s for the first 2 min, using an automatic sys-
tem, and thereafter manually for a total of 30 samples over
180 min. Blocking studies were performed by pretreatment with
WAY100635 (maleate salt, 0.5 mg/kg/iv) in 12 mL saline, at 50–
60 min of the scan. Images were transferred into the image analy-
sis software MEDx (Sensor Systems, Inc., Sterling, Virginia) for
drawing and storing regions of interest. All PET images were co-
registered within a dynamic study to the previous frame using
the Functional Magnetic PET frames are then co-registered to the
MRI using FLIRT. Regions of interest drawn on the animal’s MRI
scan were transferred to co-registered automated image registra-
tion (AIR) frames of PET data. Radioactivity levels in the right
and left regions were averaged.
(Phenomenex, Prodigy ODS-Prep 10 Â 250 mm, 10
l; and eluted
with 45:55 acetontrile: 1 M ammonium formate. The [18F]fluor-
oroethyltosylate eluted at 7–8 min was collected based on the
c
-detector, diluted with 200 mL of deionized water and passed
through a Lichrolut ENÒ column. The column was dried under
a stream of argon and eluted with 3 mL of diethyl ether. The
ether layer was concentrated under argon without heating and
the residue was treated with 2 mg of precursor in 150
lL of
DMSO followed by 25 of 5% aqueous K2CO3 solution
lL
(1.6 equiv). The reaction mixture was then heated at 110 °C for
4.5. Protein binding and metabolite analyses
20 min, allowed to cool to rt and injected onto a semi-prepara-
tive HPLC column (Prodigy ODS-Prep 10 Â 250 mm, 10
l
; mobile
The protein binding of [18F]FECUMI-101 in baboon blood sam-
ples were determined as described elsewhere.31,43 Briefly, blood
samples were taken at 2, 12, 30, 60, 90, 120, 150 and 180 min after
radioactivity injection for metabolite analysis. The supernatant li-
quid obtained after centrifugation of the blood sample at
2000 rpm for 1 min was transferred (0.5 mL) into a tube and mixed
with acetonitrile (0.7 mL). The resulting mixture was vortexed for
10 s, and centrifuged at 14,000 rpm for 4 min. The supernatant li-
quid (1 mL) was removed and the radioactivity was measured in a
well-counter and the majority (0.8 mL) was subsequently injected
onto the HPLC column (Phenomenex, Prodigy ODS (3)
phase: 40:60 acetonitrile: 25 mM Na2HPO4, flow rate 10 mL/
min). The product fraction with a retention time (tR) of 10–
11 min based on
c-detector was collected, diluted with 100 mL
of deionized water and subsequently passed through a classic
C-18 Sep-Pak cartridge, washed with 10 mL of deionized water
and eluted with 1 mL of ethanol. Reconstitution of the product
in 1 mL of absolute ethanol afforded
45 5% yield (EOS) with specific activity in the range
92.5 18.5 GBq/ mol. A portion of the ethanol solution was ana-
lyzed by analytical HPLC (Phenomenex, Prodigy ODS(3)
mobile phase; 40:60 acetonitrile/25 mM
[
18F]FECUMI-101 in
l
4.6 Â 250 mm,
5
l;
4.6 Â 250 mm, 5
25 mM Na2HPO4, 40:60 (v/v), flow rate: 2 mL/min, retention time:
7 min) connected to a Waters guard column (Resolve™ 10 m,
l; mobile phase: mobile phase: acetonitrile/
Na2HPO4, flow rate 2 mL/min, tR = 7–8 min) to determine the
specific activity, chemical and radiochemical purities. The etha-
nol solution was then diluted to a volume of 10 mL with saline
and filtered through a sterile environment, and a portion of this
solution was formulated for injection.
l
90 A°) equipped with a radioactivity detector. The metabolite and
free fractions were collected using a Bioscan gamma detector. All
the acquired data were then subjected to correction for back-
ground radioactivity and physical decay to calculate the percent-
age of the parent compound in the plasma at different time
points. In order to reaffirm that the retention time of the parent
had not shifted during the course of the metabolite analysis, a
quality control sample of [18F]FECUMI-101 was injected at the
beginning and the end of the study. The percentage of radioactive
parent obtained was used for the measurement of metabolite-cor-
rected arterial input functions.
A one-pot modification of the procedure was later developed to
provide [18F]FECUMI-101, albeit, in slightly lower yields. Accord-
ingly, a solution of 4–5 mg of diethyl tosylate in 1 mL of anhydrous
acetonitrile was added to azeotropically dried [18F]KF/kryptofix
and heated for 10 min at 95 °C. The reaction mixture was then al-
lowed to cool to room temperature, diluted with 20 mL of deion-
ized water and passed through a Phenex nylon filter (22 l)
connected to a classic C-18 Sep-PakÒ cartridge. The Sep-PakÒ was
washed with 10 mL of water followed by 1 mL of hexane and dried
under a stream of argon for 2 min and eluted with 3 mL of diethyl
ether. The ether layer was concentrated under argon without heat-
ing and the residue was heated at 110 °C for 20 min with 2 mg of
5. Conclusion
The synthesis, radiosynthesis and in vitro pharmacological eval-
uation of [18F]FECUMI-101, a 5-HT1AR agonist has been achieved.
Total time required for the synthesis of [18F]FECUMI-101 is
60 min from EOB using [18F]fluoroethyl tosylate in 45% yield at
precursor in 150 lL of DMSO and 25 lL of 5% aqueous K2CO3 solu-
tion (1.6 equiv) and column chromatographed as in the two-step
procedure to yield [18F]FECUMI-101 in 20–25% yield (EOS).

5604
V. J. Majo et al. / Bioorg. Med. Chem. 21 (2013) 5598–5604
26. Rabiner, E. A.; Messa, C.; Sargent, P. A.; Husted-Kjaer, K.; Montgomery, A.;
Lawrence, A. D.; Bench, C. J.; Gunn, R. N.; Cowen, P.; Grasby, P. M. Neuroimage
2002, 15, 620.
27. Udo, D. H. J. I.; Bosker, F. J.; Van Waarde, A.; Pruim, J.; Willemsen, A. T. M.;
Vaalburg, W.; Den, B. J. A. Synapse 2002, 46, 108.
28. Rabiner, E. A.; Gunn, R. N.; Wilkins, M. R.; Sargent, P. A.; Mocaer, E.; Sedman, E.;
Cowen, P. J.; Grasby, P. M. Nucl. Med. Biol. 2000, 27, 509.
29. Passchier, J.; van Waarde, A.; Pieterman, R. M.; Willemsen, A. T.; Vaalburg, W.
Psychopharmacology 2001, 155, 193.
30. Riad, M.; Zimmer, L.; Rbah, L.; Watkins, K. C.; Hamon, M.; Descarries, L. J.
Neurosci. 2004, 24, 5420.
31. Arlicot, N.; Vercouillie, J.; Ribeiro, M.-J.; Tauber, C.; Venel, Y.; Baulieu, J.-L.;
Maia, S.; Corcia, P.; Stabin, M. G.; Reynolds, A.; Kassiou, M.; Guilloteau, D. Nucl.
Med. Biol. 2012, 39, 570.
32. Ettrup, A.; Hansen, M.; Santini, M.; Paine, J.; Gillings, N.; Palner, M.; Lehel, S.;
Herth, M.; Madsen, J.; Kristensen, J.; Begtrup, M.; Knudsen, G. Eur. J. Nucl. Med.
Mol. Imaging 2011, 38, 681.
33. Narendran, R.; Martinez, D.; Mason, N. S.; Lopresti, B. J.; Himes, M. L.; Chen, C.-
M.; May, M. A.; Price, J. C.; Mathis, C. A.; Frankle, W. G. Synapse 2011, 65, 1344.
34. Shotbolt, P.; Tziortzi, A. C.; Searle, G. E.; Colasanti, A.; van der Aart, J.; Abanades,
S.; Plisson, C.; Miller, S. R.; Huiban, M.; Beaver, J. D.; Gunn, R. N.; Laruelle, M.;
Rabiner, E. A. J. Cereb. Blood Flow Metab. 2012, 32, 127.
EOS with excellent chemical and radiochemical purities and high
specific activity. PET studies of [18F]FECUMI-101 in anesthetized
baboon showed that the tracer penetrates the BBB and is retained
in brain regions corresponding to 5-HT1AR binding, except for thal-
amus. The specificity of radioligand binding was established by
chase studies with WAY100635. Our studies suggests that
[
18F]FECUMI-101 can be a useful PET tracer for imaging 5-HT1A
R
in baboon. However, further studies are needed to establish the
source of thalamus binding of [18F]FECUMI-101.
Acknowledgments
This work was supported by research Grants from the National
Institute of Health (R21 MH077161 and K08 MH76258). The
authors thank Dr. Bryan Roth and the NIMH-PDSP program for
the competitive receptor-transporter binding and functional
assays.
35. Tomasi, G.; Nabulsi, N.; Zheng, M.-Q.; Weinzimmer, D.; Ropchan, J.; Blumberg,
L.; Brown-Proctor, C.; Ding, Y.-S.; Carson, R. E.; Huang, Y. J. Nucl. Med. 2013, 54,
600.
36. Kumar, J. S. D.; Prabhakaran, J.; Majo, V. J.; Milak, M. S.; Hsiung, S.-C.; Tamir, H.;
Simpson, N. R.; Van Heertum, R. L.; Mann, J. J.; Parsey, R. V. Eur. J. Nuc. Med. Mol.
Imaging 2007, 34, 1050.
A. Supplementary data
Supplementary data associated with this article can be found, in
the online version, at http://dx.doi.org/10.1016/j.bmc.2013.05.050.
37. Milak, M. S.; Mann, J. J.; Ogden, T. R.; Kumar, J. S. D.; Prabhakaran, P.; Majo, V. J.;
Pratap, M.; Parsey, R. V. J. Nucl. Med. 2008, 49, 587.
38. Milak, M. S.; Severance, A. J.; Ogden, T. R.; Prabhakaran, J.; Kumar, J. S. D.; Majo,
V. J.; Mann, J. J.; Parsey, R. V. J. Nucl. Med 2010, 51, 1892.
References and notes
39. Kumar, J. S. D.; Bai, B.; Ng, H. H.; Mirsalis, J. C.; Eralandsson, K.; Milak, M. S.;
Majo, V. J.; Prabhakaran, J.; Mann, J. J.; Parsey, R. V. Int. J. Toxicol. 2011, 30, 611.
40. Milak, M. S.; Severance, A. J.; Ogden, T. R.; Prabhakaran, J.; Kumar, J. S. D.; Majo,
V. J.; Mann, J. J.; Parsey, R. V. JCBFM 2011, 31, 241.
41. Palner, M.; Underwood, M. D.; Kumar, J. S. D.; Arango, V.; Knudsen, G. M.;
Mann, J. J.; Parsey, R. V. Synapse 2011, 65, 715.
42. Kumar, J. S. D.; Majo, V. J.; Milak, M. S.; Prabhakaran, J.; Mali, P.; Savenkova, L.;
Mann, J. J.; Parsey, R. V. J. Pharm. Sci. 2012, 120, 254.
43. Bauman, A.; Piel, M.; Schirrmacher, R.; Rosch, F. Tetrahedron Lett. 2003, 44,
9165.
1. Hoyer, D.; Hannon, J. P.; Martin, G. R. Pharmacol. Biochem. Behav. 2002, 71, 533.
2. Hoyer, D.; Clarke, D. E.; Fozard, J. R.; Hartig, P. R.; Martin, G. R.; Mylecharane, E.
J.; Saxena, P. R.; Humphrey, P. P. Pharmacol. Rev. 1994, 46, 157.
3. Cohn, J. B.; Rickels, K. Curr Med Res Opin. 1989, 11, 304.
4. Cryan, J. F.; Redmond, A. M.; Kelly, J. P.; Leonard, B. E. Eur.
Neuropsychopharmacol. 1997, 7, 109.
5. Blier, P.; Abbott, F. V. J. Psychol. Neurosci. 2001, 26, 37.
6. Tomiki, S.; Vera, B.-V.; Jiri, H.; Bettina, B. Adv. Ther. 2008, 25, 1037.
7. Frederico, F. G.; Helio, Z., Jr. CNS Agents Med. Chem. 2010, 10, 207.
8. Elena, A.; Rupert, L.; Siegfried, K. Biol. Psychiatry 2009, 66, 627.
9. Robinson, D. S. Prim. Psychiatry 2007, 14, 22.
44. Wilson, A. A.; Jin, L.; Garcia, A.; DaSilva, J. N.; Houle, S. Appl. Radiat. Isot. 2001,
54, 203.
10. Simon, J. S.; Nemeroff, C. B. J. Clin. Psychiatry 2005, 66, 1216.
11. Moresco, R. M.; Matarrese, M.; Fazio, F. Curr. Top. Med. Chem. 2006, 6, 2027.
12. Veltman, D. J.; Ruhe, H. G.; Booij, J. Curr. Pharm. Des. 2010, 16, 1979.
13. Fujita, M. F.; Dennis, S. C.; Innis, R. B. Biol. Psychiatry 2000, 48, 801.
14. Kumar, J. S. D.; Mann, J. J. Drug Dev. Today 2007, 12, 748.
15. Passchier, J.; van Waarde, A. Eur. J. Nucl. Med. 2001, 28, 113.
16. Savitz, J.; Lucki, I.; Drevets, W. C. Prog. Neurobiol. 2009, 88, 17.
17. Paterson, L. M.; Kornum, B. R.; Nutt, D. J.; Pike, V. W.; Knudsen, G. M. Med. Res.
Rev. 2013, 33, 54.
18. Drevets, W. C.; Thase, M. E.; Moses-Kolko, E. L.; Price, J.; Frank, E.; Kupfer, D. J.;
Mathis, C. Nucl. Med. Biol. 2007, 34, 865.
19. Didelot, A.; Ryvlin, P.; Lothe, A.; Merlet, I.; Hammers, A.; Mauguière, F. Brain
2008, 131, 2751.
20. Aznavour, N.; Zimmer, Luc Neuropharmacology 2007, 52, 695.
21. Sullivan, G. M.; Ogdan, T.; Ocqendo, M.; Kumar, J. S. D.; Simpson, N.; Huang, Y.;
Mann, J. J.; Parsey, R. V. Biol. Psychiatry 2009, 66, 223.
22. Gozlan, H.; Thibault, S.; Laporte, A. M.; Lima, L.; Hamon, M. Eur. J. Pharmacol.
1995, 288, 173.
23. Burnet, P. W.; Eastwood, S. L.; Harrison, P. J. Neurochem. Int. 1997, 30, 565.
24. Bantick, R. A.; Rabiner, E. A.; Hirani, E.; de Vries, M. H.; Hume, S. P.; Grasby, P.
M. Neuropsychopharmacology 2004, 29, 847.
45. Fluorine and Health; Tressaud, A., Haufe, G., Eds.; Elsevier, 2008.
46. Champion, C.; Le Loirec, C. Phys. Med. Biol. 2007, 52, 6605.
47. Sorensen, J. A.; Phelps, M. E. Physics in Nuclear Medicine;
W B Sanders:
Philadelphia, 1987.
48. Kumar, J. S. D.; Majo, V. J.; Tamir, H.; Millak, M. S.; Hsing, S.-C.; Prabhakaran, J.;
Simpson, N. R.; Van Heertum, R. L.; Mann, J. J.; Parsey, R. V. J. Med. Chem. 2006,
49, 125.
49. Herth, M. M.; Volk, B.; Pallagi, K.; Kofoed Bech, L.; Antoni, F. A.; Knudsen, G. M.;
Kristensen, J. L. ACS Chem. Neurosci. 2012, 3, 1002.
50. Missale, C.; Nash, R. S.; Robinson, S. W.; Jaber, M.; Caron, C. G. Physiol. Rev.
1998, 78, 189.
51. Patel, N. H.; Vyas, N. R.; Puri, B. K.; Nijran, K. S.; Al-Nahhas, A. J. Nucl. Med. 2010,
51, 511.
52. Jones, L. S.; Miller, G.; Gauger, L. L.; Davis, J. N. Life Sci. 1985, 36, 45.
53. Gross-Isseroff, R.; Dillon, K. A.; Fieldust, S. J.; Biegon, A. Arch. Gen. Psychiatry
1990, 47, 1049.
54. Szot, P.; White, S. S.; Greenup, J. L.; Leverenz, J. B.; Peskind, E. R.; Raskind, M. A.
Mol. Brain Res. 2005, 139, 367.
55. Chemel, B.; Roth, B.; Armbruster, B.; Watts, V.; Nichols, D. Psychopharmacology
2006, 188, 244.
25. Laruelle, M. J. Cereb. Blood Flow Metab. 2000, 20, 423.