Imaging the trace amine-associated receptor 1 by positron emission tomography

Article abstract of DOI:10.1016/j.tetlet.2021.153007

Trace amine-associated receptor 1 (TAAR1) has been implicated in drug addiction, schizophrenia, depression and Parkinson's disease (PD). To date, there are no reports on TAAR1-targeted probes for non-invasive quantification of receptor density in vivo. Herein, we report the synthesis of a ¹¹C-labeled TAAR1 high-affinity antagonist N-(3-methoxyphenyl)-6-(pyrrolidin-1-yl)-5-(trifluoromethyl)nicotinamide [¹¹C]4 (also named [¹¹C]TAAR1-1911), as well as its physicochemical and preclinical evaluations for positron emission tomography (PET) imaging. This PET ligand was afforded using [¹¹C]CH₃I with the base NaOH in good radiochemical yield (non-decay corrected 14% relative to starting [¹¹C]CO₂), excellent radiochemical purities (>99%) and high molar activities (>37 GBq/μmol). Despite promising in vitro performance characteristics, [¹¹C]4 did not exhibit in vivo specificity, potentially owing to fast metabolic degradation. Further studies are warranted to identify a suitable TAAR1 PET tracer, which would ultimately aid the development of TAAR1-directed therapeutic agents.

Full text of DOI:10.1016/j.tetlet.2021.153007

Tetrahedron Letters 70 (2021) 153007
Contents lists available at ScienceDirect
Tetrahedron Letters
journal homepage: www.elsevier.com/locate/tetlet
Imaging the trace amine-associated receptor 1 by positron emission
tomography
Jiyun Sun a,b, Jiahui Chen a,b, Katsushi Kumata c, Zhiwei Xiao a,b, Jian Rong a, Ahmed Haider a, Tuo Shao a,
Lu Wang b, Hao Xu b, , Ming-Rong Zhang c, , Steven H. Liang a,
⇑
⇑
⇑
a Department of Radiology, Division of Nuclear Medicine and Molecular Imaging Massachusetts General Hospital and Harvard Medical School, 55 Fruit Street, Boston,
MA 02114, United States
b Center of Cyclotron and PET Radiopharmaceuticals, Department of Nuclear Medicine and PET/CT-MRI Center, the First Affiliated Hospital of Jinan University, Guangzhou 510630,
China
c Department of Advanced Nuclear Medicine Sciences, National Institute of Radiological Sciences, National Institutes for Quantum and Radiological Science and Technology, Chiba
263-8555, Japan
a r t i c l e i n f o
a b s t r a c t
Article history:
Trace amine-associated receptor 1 (TAAR1) has been implicated in drug addiction, schizophrenia, depres-
sion and Parkinson’s disease (PD). To date, there are no reports on TAAR1-targeted probes for non-inva-
sive quantification of receptor density in vivo. Herein, we report the synthesis of a 11C-labeled TAAR1
high-affinity antagonist N-(3-methoxyphenyl)-6-(pyrrolidin-1-yl)-5-(trifluoromethyl)nicotinamide [11C]
4 (also named [11C]TAAR1-1911), as well as its physicochemical and preclinical evaluations for positron
emission tomography (PET) imaging. This PET ligand was afforded using [11C]CH3I with the base NaOH in
good radiochemical yield (non-decay corrected 14% relative to starting [11C]CO2), excellent radiochemical
purities (>99%) and high molar activities (>37 GBq/mmol). Despite promising in vitro performance char-
acteristics, [11C]4 did not exhibit in vivo specificity, potentially owing to fast metabolic degradation.
Further studies are warranted to identify a suitable TAAR1 PET tracer, which would ultimately aid the
development of TAAR1-directed therapeutic agents.
Received 19 January 2021
Revised 9 March 2021
Accepted 12 March 2021
Available online 16 March 2021
Keywords:
Trace amine-associated receptor 1 (TAAR1)
Positron emission tomography (PET)
Neuroimaging
Radiochemistry
Drug development
Ó 2021 Elsevier Ltd. All rights reserved.
Trace amine-associated receptor 1 (TAAR1) is a G protein-cou-
pled receptor (GPCR) that is activated by endogenous monoamines
and amphetamine-related psychostimulants, including metham-
phetamine [1–5]. TAAR1 is widely expressed in the stomach, pan-
creas, intestine and the central nervous system (CNS). In the
mammalian brain, TAAR1 is highly expressed in the ventral
tegmental area and dorsal raphe nuclei, where it functions as a
negative modulator of monoamine neurotransmission [6]. Along
this line of reasoning, TAAR1 has been suggested as a potential
therapeutic target for CNS and metabolic disorders, particularly
owing to the implications in the pathophysiology of schizophrenia,
depression, sleep disorders, Parkinson’s disease, type 2 diabetes
and eating disorders [7]. Several TAAR1 agonists have been shown
to attenuate monoaminergic downstream signaling [3,8]. Of note,
some endogenous amines and their derivatives can activate TAAR1
for the development of TAAR1 ligands, chemical libraries of various
structural scaffolds have been employed to facilitate the screening
of potential drug candidates, ultimately resulting in the successful
identification of highly selective and potent TAAR1 agonists (Fig. 1)
[1,4,5,9,10]. In sharp contrast, the identification of promising
antagonist has proven to be challenging, primarily due to the lack
of selectivity of the vast majority of TAAR1 antagonists. To date,
EPPTB (RO5212773) is the only well-characterized TAAR1 antago-
nist with high in vitro binding affinity and selectivity, as well as
appropriate lipophilicity and functional activity [11,12]. EPPTB
was developed by high throughput screening (HTS) and subse-
quent structure–activity refinements. Of note, studies with EPPTB
in wild-type and TAAR1À/À mice unveiled a key regulatory mecha-
nism of TAAR1 in dopaminergic neurons of the mesolimbic system.
Positron emission tomography (PET) is a highly sensitive and
non-invasive imaging technology that allows the in vivo
quantification of biological processes [13–16]. To date, a suitable
TAAR1-selective probe is lacking. However, the development of a
TAAR1-targeted PET radioligand would enable the non-invasive
assessment of TAAR1 expression under physiological and patho-
logical conditions, thereby shedding light on the mechanistic
with nanomolar (nM) to micromolar (lM) potency [1,3,6]. While
the structural scaffold of endogenous amines has been exploited
⇑
Corresponding authors.
E-mail addresses: txh@jnu.edu.cn (H. Xu), zhang.ming-rong@qst.go.jp
(M.-R. Zhang), liang.steven@mgh.harvard.edu (S.H. Liang).
https://doi.org/10.1016/j.tetlet.2021.153007
0040-4039/Ó 2021 Elsevier Ltd. All rights reserved.

J. Sun, J. Chen, K. Kumata et al.
Tetrahedron Letters 70 (2021) 153007
Fig. 1. Selected TAAR1 agonists.
involvement of TAAR1 in various animal disease models, as well as
in patients. Further, it would accelerate the development of TAAR1
antagonists, which is a current unmet medical need. Indeed, TAAR1
antagonists harbor enormous potential for the treatment of
hypodopaminergic pathologies such as Parkinson’s disease [17].
Despite the high potency and selectivity of EPPTB, its chemical
structure is not amenable to conventional carbon-11 labeling. In
contrast, the previously reported analog, N-(3-methoxyphenyl)-6-
(pyrrolidin-1-yl)-5-(trifluoromethyl)nicotina-mide 4, which also
demonstrated high in vitro binding affinity (Ki = 2.0 nM) [11],
was suitable for 11C-labeling via O-alkylation. Accordingly, we
envisioned the development of the first TAAR1-targeted PET radi-
oligand based on the structure of amide 4 – with the aim to allow
non-invasive visualization of TAAR1 in vivo.
Target compound 4 and the respective demethylated precursor
7 were synthesized as previously reported [11], however, with
minor modification as outlined in Scheme 1. Briefly, amide 4 was
obtained in two steps via treatment of chloro-pyridine precursor
1 with pyrrolidine and subsequent condensation reaction with
m-anisidine in an overall yield of 55% (2 steps). Similarly, precursor
7 was synthesized from intermediate acid 2 via DMAP-mediated
amide formation, followed by cleavage of the protection group in
an overall yield of 45% (3 steps).
The ADME properties and SAR of EPPTB have been previously
reported, along with preliminary testing of target compound 4,
which included in vitro binding affinity and metabolism [11], In
this work, we evaluated the physicochemical properties of com-
pound 4 and predicted its blood–brain barrier (BBB) permeability
using in silico predictive algorithms (CNS multiparameter opti-
mization (CNS MPO) [18] and blood–brain barrier (BBB) Score
[19]). Pharmacology and ADME results are summarized in Fig. 2.
CNS MPO score and BBB score (4.62 and 4.54, respectively, using
a scale of 0–6, respectively) indicated that compound 4 is a CNS-
favored molecule [18,19]. The predicted ratio of Cbrain to Cblood
was expressed as logBB = À0.08 (ACD Percepta program), which
is reasonable for brain penetration [20].
As for the assessments of metabolism and in vitro safety pro-
files, compound 4 exhibited a reasonably low CYP inhibition, and
was found to be inactive in the hERG assay (IC50 = 10.5
lM, see
Figs. S1 and S2 in Supporting Information). These encouraging
results prompted us to carry on with the carbon-11 labeling of
compound 4 and subsequent in vivo evaluation.
As shown in Fig. 3, the hydroxyl group of precursor 7 is acces-
sible for conventional carbon-11 labeling with [11C]CH3I [21].
Accordingly, [11C]4 was synthesized with an automated module
sequence that includes radiolabeling, purification and formulation
in an overall synthetic time of 35 min. To optimize the reaction
conditions, different amounts of precursor, type of base and reac-
tion temperatures were employed, as depicted in Fig. 3A. After
testing different combinations, 0.5 N NaOH(aq) in DMF at 80 °C
was identified as the optimal reaction condition. The highest
non-decay corrected RCY (14.0%, relative to [11C]CO2 at the end
of bombardment, total synthetic time ca. 35 min) was obtained
with the lowest amount of precursor 7 (Fig. 3A, entry 5). The reac-
tion mixture was purified by semi-prep radio-HPLC to afford [11C]4
with excellent radiochemical purity (>99%) and high molar activi-
ties (37 > GBq/mmol). No radiolysis was observed up to 90 min after
formulation (Fig. 3B). The in vitro stability assay (5, 30, 60 min)
Scheme 1. Synthesis of target compound 4 and the respective precursor 7 for
carbon-11 labeling.
2

J. Sun, J. Chen, K. Kumata et al.
Tetrahedron Letters 70 (2021) 153007
Fig. 2. Summary of physicochemical and safety properties of TAAR1 antagonist EPPTB and target compound 4.
Fig. 3. (A) Optimization of reaction parameters for the radiosynthesis of [11C]4. (i)
a
[
11C]MeI, DMSO, 100 °C, 5 min. Non-decay corrected RCY relative to the starting
activity of [11C]CO2. 1.0 M aqueous solution of Cs2CO3 was used. Solid reagent
b
c
d
was used. 0.5 M aqueous solution of NaOH was used. (B) Stability of radioligand
[
11C]4 in saline containing 5% of ethanol at three different time point (30, 60 and
90 min). HPLC conditions: Xselect Hss T3, 4.6 mm i.d. Â 150 mm, UV at 254 nm;
CH3CN/H2O (v/v, 70/30) + 0.1% Et3N at a flow rate of 1.0 mL/min.
proved that [11C]4 has good in vitro stability (>99% parent compo-
nent) in mouse serum and blood, as well as in human serum. The
lipophilicity of [11C]4 was determined by the ‘shake-flask’ method.
As such, a LogD7.4 valve of 3.36 0.22 (n = 3) was obtained, indicat-
ing that compound 4 possesses the appropriate lipophilicity to
cross the BBB.
Fig. 4. (A) Representative PET images of radioligand [11C]4 (0–60 min averaged
image); (B) Time-activity curves of [11C]4 in mice brain (baseline, n = 3; EPPTB
blocking, 3 mg/kg, n = 3; self-blocking, 3 mg/kg, n = 2); (C) Quantification (0–
60 min) using total area under curve of TAC.
Dynamic PET imaging with [11C]4 was carried out in CD-1 mice.
Representative PET images (sagittal and coronal, averaged from 0
to 60 min), time-activity curves (TAC) and total areas under the
curves of whole brain at baseline conditions are shown in Fig. 4.
The limited radioactivity uptake in the whole brain (ca. 0.3 SUV
at peak) indicated that [11C]4 did not reach the brain in sufficient
amounts to allow appropriate TAAR1 imaging. Further, pretreat-
ments with non-radioactive reference compound 4 or EPPTB
(TAAR1 selective antagonist) resulted in higher brain uptake of
[
11C]4 compared to the baseline, which might be attributed to
3

J. Sun, J. Chen, K. Kumata et al.
Tetrahedron Letters 70 (2021) 153007
Fig. 5. (A) Ex vivo whole body biodistribution in CD-1 mice at four different time points (5, 10, 30 and 60 min) post injection of [11C]4. The results are expressed as the
percentage of the injected dose per gram of wet tissue (% ID/g); (B) Ex vivo blocking studies of [11C]4 in the mouse brain at two different time points (5 and 30 min). The
blocking group were pretreated with EPPTB (3 mg/kg) for 2 min followed by [11C]4. All data are mean SD, n = 3. (C) Radiometabolism in the mouse brain and plasma at
30 min post injection (n = 2). Asterisks indicate statistical significance: ****p < 0.0001, ***p < 0.001, **p < 0.01, and *p < 0.05.
several confounding factors, including nonspecific binding,
increased systemic tracer availability due to the blockade of TAAR1
in peripheral organs and rapid in vivo metabolism (vide infra).
Another explanation could be that the blocker might have resulted
in a pharmacological effect that would increase cerebral perfusion
and hence, delivery of the tracer (or radiometabolites) to the brain.
To verify our PET findings, ex vivo whole body biodistribution of
of [11C]4 were evaluated by PET imaging, ex vivo biodistribution
and radiometabolite analysis in mice. Despite the promising
in vitro performance characteristics including high in silico brain
penetration, low brain uptake and metabolic instability, as well
as the presence of brain radiometabolites, hampered the further
development of this ligand.
[
11C]4 was performed in CD-1 mice at four different time points (5,
Declaration of Competing Interest
15, 30 and 60 min post injection). The data are presented as the
percentage of injected dose per gram of wet tissue (Fig. 5A,
%ID/g). The limited brain uptake was consistent with the PET
results. In contrast, high radioactivity uptake (>5% ID/g) was found
in the small intestine, kidneys and liver at 5 min post injection.
Further, high tracer uptake was found in the heart, lungs and pan-
creas. Of note, radioactivity levels in the liver reached a plateau at
5 min post injection, followed by a slow washout, thus suggesting
that hepatobiliary elimination might be the primary clearance
pathway. Studies were also performed under blocking conditions
with EPPTB (3 mg/kg) at 5 min and 30 min time points (see
Table S1 and S2 in Supporting Information for more detail). In
accordance with the PET experiments, no specific binding was
observed in the brain (Fig. 5B).
The authors declare that they have no known competing finan-
cial interests or personal relationships that could have appeared
to influence the work reported in this paper.
Acknowledgments
We gratefully acknowledge the generous support provided by
the National Natural Science Foundation of China (NO. 81901802,
NO. 82071974, NO. 81871383). AH was supported by the Swiss
National Science Foundation.
Appendix A. Supplementary data
In vivo stability of [11C]4 was evaluated in mice at 30 min post
injection. In fact, it was found that only very small fractions of the
intact parent tracer were found in the brain (5.3 1.0% in the brain)
and in the plasma (17.4 0.5%), which may explain the overall low
brain uptake, as well as the high amount of nonspecific binding.
In summary, we have developed the first TAAR1-targeted PET
radioligand. The [11C]TAAR1-1911 ([11C]4) was obtained with
excellent radiochemical purity (>99%) and high molar activity
(>37 GBq/mmol). Brain uptake, clearance and binding specificity
Supplementary data to this article can be found online at
https://doi.org/10.1016/j.tetlet.2021.153007.
References
[1] R. Zucchi, G. Chiellini, T.S. Scanlan, D.K. Grandy, Br. J. Pharmacol. 149 (2006)
967–978.
[2] E. Cichero, M. Tonelli, Future Med. Chem. 9 (2017) 1507–1527.
[3] M.D. Schwartz, J.J. Canales, R. Zucchi, S. Espinoza, I. Sukhanov, R.R. Gainetdinov,
Expert Opin. Ther. Targets 22 (2018) 513–526.
4

J. Sun, J. Chen, K. Kumata et al.
Tetrahedron Letters 70 (2021) 153007
[13] S.S. Tee, K.R. Keshari, Cancer J. (United States) 21 (2015) 165–173.
[14] A. Gallamini, C. Zwarthoed, A. Borra, Cancers (Basel) 6 (2014) 1821–1889.
[15] P. Brust, J. van den Hoff, J. Steinbach, Neurosci. Bull. 30 (2014) 777–811.
[16] W. Mier, D. Mier, Front. Hum. Neurosci. 9 (2015) 1–6.
[17] A. Alvarsson, X. Zhang, T.L. Stan, N. Schintu, B. Kadkhodaei, M.J. Millan, T.
Perlmann, P. Svenningsson, J. Neurosci. 35 (2015) 14057–14069.
[18] T.T. Wager, X. Hou, P.R. Verhoest, A. Villalobos, ACS Chem. Neurosci. 1 (2010)
435–449.
[4] Z. Xu, Q. Li, Cell. Mol. Neurobiol. 40 (2020) 257–272.
[5] M. Tonelli, E. Cichero, Expert Opin. Ther. Pat. 30 (2020) 137–145.
[6] G. Rutigliano, A. Accorroni, R. Zucchi, Front. Pharmacol. 8 (2018) 1–18.
[7] M.D. Berry, R.R. Gainetdinov, M.C. Hoener, M. Shahid, Pharmacol. Ther. 180
(2017) 161–180.
[8] A.A. Aleksandrov, V.M. Knyazeva, A.B. Volnova, E.S. Dmitrieva, N.V. Polyakova,
R.R. Gainetdinov, Front. Pharmacol. 10 (2019) 1–8.
[9] R.R. Gainetdinov, M.C. Hoener, M.D. Berry, Pharmacol. Rev. 70 (2018) 549–620.
[10] K.S. Koblan, J. Kent, S.C. Hopkins, J.H. Krystal, H. Cheng, R. Goldman, A. Loebel,
New Engl. J. Med. 382 (2020) 1497–1506.
[19] M. Gupta, H.J. Lee, C.J. Barden, D.F. Weaver, J. Med. Chem. 62 (2019) 9824–
9836.
[11] H. Stalder, M.C. Hoener, R.D. Norcross, Bioorg. Med. Chem. Lett. 21 (2011)
1227–1231.
[12] A. Bradaia, G. Trube, H. Stalder, R.D. Norcross, L. Ozmen, J.G. Wettstein, A.
Pinard, D. Buchy, M. Gassmann, M.C. Hoener, B. Bettler, Proc. Natl. Acad. Sci. U.
S. A. 106 (2009) 20081–20086.
[20] D.E. Clark, J. Pharm. Sci. 88 (1999) 815–821.
[21] X. Deng, J. Rong, L. Wang, N. Vasdev, L. Zhang, L. Josephson, S.H. Liang, Angew.
Chem. Int. Ed. 58 (2019) 2580–2605.
5