Synthesis and evaluation of 11C-labeled coumarin analog as an imaging probe for detecting monocarboxylate transporters expression

Article abstract of DOI:10.1016/j.bmcl.2017.09.033

Upregulated monocarboxylate transporters (MCTs) in tumors are considered diagnostic imaging targets. Herein, we synthesized the positron emission tomography probe candidates coumarin analogs 2 and 3, and showed 55 times higher affinity of 2 for MCTs than a representative MCT inhibitor. Whereas [¹¹C]2 showed low tumor accumulation, probably due to adduct formation with plasma proteins, [¹¹C]2 showed high initial brain uptake, suggesting that the scaffold of 2 has properties that are preferable in imaging probes for the astrocyte–neuron lactate shuttle. Although further optimization of 2 is required, our findings can be used to inform the development of MCT-targeted imaging agents.

Full text of DOI:10.1016/j.bmcl.2017.09.033

Bioorganic & Medicinal Chemistry Letters xxx (2017) xxx–xxx
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
Synthesis and evaluation of 11C-labeled coumarin analog as an imaging
probe for detecting monocarboxylate transporters expression
b,c,
a
a
c
Hiroyuki Tateishi a,e, Atsushi B. Tsuji a, , Koichi Kato
, Hitomi Sudo , Aya Sugyo , Takashi Hanakawa ,
⇑
⇑
Ming-Rong Zhang b, Tsuneo Saga a,d, Yasushi Arano e, Tatsuya Higashi a
a Experimental Nuclear Medicine Team, Department of Molecular Imaging and Theranostics, National Institute of Radiological Sciences, National Institutes for Quantum
and Radiological Science and Technology (QST-NIRS), 4-9-1 Anagawa, Inage-ku, Chiba 263-8555, Japan
b Department of Radiopharmaceuticals Development, National Institute of Radiological Sciences, National Institutes for Quantum and Radiological Science and Technology (QST-NIRS),
4-9-1 Anagawa, Inage-ku, Chiba 263-8555, Japan
c Department of Integrative Brain Imaging, National Center of Neurology and Psychiatry, 4-1-1 Ogawahigashi-cho, Kodaira, Tokyo 187-5551, Japan
d Department of Diagnostic Radiology, Kyoto University Hospital, 54 Shogoinkawahara-cho, Sakyo-ku, Kyoto 606-8507, Japan
e Department of Molecular Imaging and Radiotherapy, Graduate School of Pharmaceutical Sciences, Chiba University, 1-8-1 Inohana, Chuo-ku, Chiba 260-8675, Japan
a r t i c l e i n f o
a b s t r a c t
Article history:
Upregulated monocarboxylate transporters (MCTs) in tumors are considered diagnostic imaging targets.
Herein, we synthesized the positron emission tomography probe candidates coumarin analogs 2 and 3,
and showed 55 times higher affinity of 2 for MCTs than a representative MCT inhibitor. Whereas [11C]
2 showed low tumor accumulation, probably due to adduct formation with plasma proteins, [11C]2
showed high initial brain uptake, suggesting that the scaffold of 2 has properties that are preferable in
imaging probes for the astrocyte–neuron lactate shuttle. Although further optimization of 2 is required,
our findings can be used to inform the development of MCT-targeted imaging agents.
Received 30 June 2017
Revised 11 September 2017
Accepted 14 September 2017
Available online xxxx
Keywords:
Positron emission tomography
Fluorescence
Ó 2017 The Author(s). Published by Elsevier Ltd. This is an open access article under the CC BY license
(http://creativecommons.org/licenses/by/4.0/).
Cancer
Astrocyte–neuron lactate shuttle
Radiosynthesis
Monocarboxylate transporters (MCTs) belong to the SLC16 gene
family, which comprises 14 members. Among these, MCT1, MCT2,
MCT3, and MCT4 are proton symporters that mediate bidirectional
transmembrane transport of short chain monocarboxylates such as
pyruvate, lactate and ketone bodies. Most living cells consume
these monocarboxylates as fuel for the tricarboxylic-acid cycle,1,2
indicating that MCTs play important roles in cell homeostasis.
In many tumors, MCT1 and MCT4 highly express and mediate
lactate transport.3 Lactate is produced from aggressive tumor
metabolism such as glycolysis,4,5 glutaminolysis,6–8 and serinoly-
sis,6 and acidifies tumor microenvironments, promoting tumor
malignancy and invasiveness, and leading to poor clinical out-
comes.9–14 Recently, lactate has been recognized as a key energy
source for tumor metabolic symbiosis, in which oxidative tumor
cells consume lactate from glycolytic cells.15–17 Thus, lactate is
not merely a waste metabolite but also a key intermediate for
tumor metabolism. Consequently, aberrant lactate metabolism is
considered a hallmark of cancer and the lactate transporters
MCT1 and MCT4 are attractive targets for cancer diagnosis and
therapy. Hence, radiolabeled inhibitors of MCT1 and/or MCT4 with
positron emitting radionuclides such as 11C and 18F have potential
as imaging agents for cancer diagnoses using positron emission
tomography (PET), which is a non-invasive quantitative imaging
technique with high sensitivity that can be used to monitor patho-
physiological changes and disease progression.18
Although
a-cyano-4-hydroxycinamic acid (CHC) is a well-
known MCT inhibitor (Fig. 1), it has low affinity (11
l
M) and speci-
ficity for MCTs.19–21 Recently, Draoui et al. synthesized several
3-carboxy-coumarin derivatives and evaluated their inhibitory
activities against MCTs.21 These investigators reported that 7-alky-
lamino substituents on 3-carboxy-coumarin scaffolds were
required for significant inhibition of MCTs.21 Among presented
compounds, that with a 7-benzyl(methyl)amino substituent 1
⇑
Corresponding authors at: Department of Molecular Imaging and Theranostics,
National Institute of Radiological Sciences, National Institutes for Quantum and
Radiological Science and Technology (QST-NIRS), 4-9-1 Anagawa, Inage-ku, Chiba
263-8555, Japan (A.B. Tsuji); Department of Integrative Brain Imaging, National
Center of Neurology and Psychiatry, 4-1-1 Ogawahigashi-cho, Kodaira, Tokyo 187-
5551, Japan (K. Kato).
(Fig. 1) showed the strongest inhibition of lactate influx (IC50
,
11 nM) and the corresponding 11C-labeled compound [N-11C-
methyl]1 was considered a candidate PET probe. However, a
distinct synthesis scheme is required for a dimethyl precursor for
E-mail addresses: tsuji.atsushi@qst.go.jp (A.B. Tsuji), katok@ncnp.go.jp (K. Kato).
https://doi.org/10.1016/j.bmcl.2017.09.033
0960-894X/Ó 2017 The Author(s). Published by Elsevier Ltd.
This is an open access article under the CC BY license (http://creativecommons.org/licenses/by/4.0/).
Please cite this article in press as: Tateishi H., et al. Bioorg. Med. Chem. Lett. (2017), https://doi.org/10.1016/j.bmcl.2017.09.033

2
H. Tateishi et al. / Bioorganic & Medicinal Chemistry Letters xxx (2017) xxx–xxx
Fig. 1. Structure of
a
-cyano-4-hydroxycinamic acid (CHC) and monocarboxylate transporter (MCT)-targeting candidates 1, 2, and 3. *Represents the 11C-labeling site.
11C labeling of 1 that is directed our attention to more accessible
targets. Thus, we considered the use of another benzyl(methyl)
amino analog 2 (Fig. 1) for the development of an MCT-targeted
imaging probe, because both 2 and the appropriate non-radioac-
tive precursor to synthesize 11C-labeled 2 ([11C]2) can be prepared
via the common aldehyde intermediate 7. In addition, both the
diethylamino analog 3 (Fig. 1) and the appropriate non-radioactive
precursor 11 could be prepared using a similar synthetic scheme.22
Furthermore, 11C-labeled 2 and 3 can be synthesized by our label-
ing method to construct versatile 11C-labeled olefin frameworks
based on Horner-Wadsworth-Emmons (HWE) reactions.22 In the
present study, we synthesized the benzyl(methyl)amino analog 2
and the diethylamino analog 3 and performed in vitro studies using
human cancer cell lines. We also demonstrated 11C-labeling syn-
thesis of 2 and evaluated temporal radioactivity change of [11C]2
in a tumor mouse model by dynamic PET study.
Compounds 2 and 3 were synthesized as shown in Schemes 1
and 2, respectively, according to previous reports.21,22 During the
synthesis of 2, Knoevenagel condensation of 7-amino substituted
salicylaldehyde 5 with Merdrum’s acid was used to construct the
3-carboxy-coumarin skeleton 1. Salicylaldehyde 5 was obtained
from two steps, involving Buchwald–Hartwig cross coupling of 3-
bromophenol with N-methyl-1-phenylmethanamine, followed by
a Vilsmeier–Haack reaction. The carboxyl group of 1 was then con-
verted to the corresponding aldehyde 7. The subsequent HWE
reaction afforded 8 and 2 was obtained after hydrolysis of the ethyl
ester. Subsequently, 3 was synthesized according to a similar
scheme (Scheme 2). Overall, a yield of 23.3% was achieved from
the 7-step synthesis for 2 and a yield of 45.1% was achieved from
the 5-step synthesis for 3. The E/Z isomeric mixture of 2 and 3
was used for in vitro and in vivo studies due to their ease of
photo-isomerization (Supplementary Fig. 1). Compounds 2 and 3
showed strong blue–green fluorescence (Table 1).
CHC, which was similar to that of 3. We therefore chose to use 2
in further studies.
Expression analyses of MCT1 and MCT4 were performed using
quantitative RT-PCR in Becker, AsPC-1 and MDA-MB-231 human
cancer cell lines. These experiments showed that Becker cells
express MCT1 at the highest levels, followed by AsPC-1 and
MDA-MB-231 cells (Fig. 3 left panel). MCT4 expression was great-
est in MDA-MB-231 cells, followed by Becker and AsPC-1 cells
(Fig. 3 right panel).
As shown in Fig. 4, Becker cells showed the highest uptake of 2,
followed by AsPC-1 and MDA-MB-231 cells. Taken together, these
data suggest that 2 was transported into cells in an MCT1-expres-
sion-dependent manner, likely reflecting differences in the MCT1
and MCT4 binding affinity of monocarboxylates such as 2. Accord-
ingly, compared with MCT1, the affinity of MCT4 for monocarboxy-
lates was lower, and the affinity of MCT4 for L-lactate (Km, 28.0–
34.0 mM) was also lower than that of MCT1 (Km, 2.2–4.5 mM).3
As shown in Scheme 3, [11C]2 was synthesized using a 3-step
synthesis protocol. Both 11C-methylation of phosphonate and
subsequent HWE reaction of the aldehyde precursor 7 with the
11C-methylated phosphonate were performed in the presence of
tetrabutylammonium fluoride as a base, and subsequent alkaline
hydrolysis of the ethyl ester afforded [11C]2.22 Analytical HPLC
chromatograms of each step were shown in Supplementary
Fig. 2. The radiosynthesis time of [11C]2 was approximately
50 min and radiochemical yields were 6.6% and 5.2% (n = 2,
decay-uncorrected). The radiochemical purity was over 95% and
E/Z ratio was approximately 9:1 (Supplementary Fig. 2).
To evaluate the pharmacokinetics of [11C]2 in vivo, 11.8 MBq of
[
11C]2 was intravenously administered to a mouse bearing Becker
tumor xenograft and dynamic PET study was conducted. Temporal
PET images and time activity curves of interested organs are pre-
sented in Fig. 5. The radiotracer [11C]2 was rapidly cleared from
the blood and was excreted through the kidneys and hepatobiliary
tract (Fig. 5A and B). High liver uptake of [11C]2 was observed, with
a peak value of 16.0% ID/g at 270 s, and radioactivity decreased
thereafter to 6.5% ID/g at 3300 s (Fig. 5C). Contrary to expectations,
tumor accumulation of [11C]2 was low with a peak value of only
Lactate uptake inhibition assays of 2 and 3 were conducted
using [14C]lactate and Becker tumor cells. As shown in Fig. 2, com-
pounds 2 and 3 dose-dependently inhibited lactate uptake with
IC50 values of 0.2
lM for 2, 9.3 lM for 3, and 11 lM for CHC. The
IC50 value for 2 was approximately 55 times greater than that for
Scheme 1. Synthesis of compound 2. Reagents and conditions: (i) N-methyl-1-phenylmethanamine, Pd2(dba)3, (2-biphenyl)di-tert-butylphosphine, LHMDS in THF, reflux for
16 h; (ii) POCl3 in DMF, 0–60 °C for 7 h; (iii) Meldrum’s acid, piperidine, AcOH in EtOH, 90 °C for 8 h; (iv) 18% HCl aq., reflux for 6 h; (v) POCl3 in DMF, 0–60 °C for 7 h. (vi)
triethylphosphonopropionate, NaH in THF, 90 °C for 6 h; (vii) 2 M NaOH in MeCN, rt for 12 h.
Please cite this article in press as: Tateishi H., et al. Bioorg. Med. Chem. Lett. (2017), https://doi.org/10.1016/j.bmcl.2017.09.033

H. Tateishi et al. / Bioorganic & Medicinal Chemistry Letters xxx (2017) xxx–xxx
3
Scheme 2. Synthesis of compound 3. Reagents and conditions: (i) Meldrum’s acid, piperidine, AcOH in EtOH, 90 °C for 6 h; (ii) 18% HCl aq., reflux for 6 h; (iii) POCl3 in DMF, 0–
60 °C for 7 h; (iv) triethylphosphonopropionate, NaH in THF, 90 °C for 6 h; (v) 2 M NaOH in MeCN, rt for 12 h.
Table 1
Fluorescence properties of compound 2 and 3.
Compound
CHC
2
3
Excitation/Emission
Fluorescence quantum yield (/)
No fluorescence
–
395/504
0.16
410/508
0.15
Fig. 4. Cell uptake of 2 in three cell lines. Uptake of 2 into Becker, AsPC-1, and MDA-
MB-231 cells (n = 3); cells were incubated with 10À5 M of 2 in 1 mL of PBS for 1, 15,
30, 60, 180 and 360 min. Uptake of 2 into cells was then measured according to
fluorescence intensities with excitation at 395 nm and emission at 504 nm.
Fluorescence intensities were normalized to protein contents, which were deter-
mined using Bradford protein quantification methods. At the vertical axis, ‘‘F–F0”
represents ‘‘(fluorescence intensity at each time point)–(fluorescence intensity at
0 min).”
1.9% ID/g at 210 s (Fig. 5D). According to metabolite analysis (Sup-
plementary Fig. 3), approximately 27% of radioactivity were
observed as radiometabolites. Although PET images using [11C]2
mainly reflect the intact form, we need to consider radiometabo-
lites for more precise implication in further study.
To determine why [11C]2 accumulation was low in tumor, we
then performed plasma protein binding test and demonstrated a
plasma protein binding rate of 92.0% 4.1% (n = 3) after 30-min
incubation with mouse plasma, indicating almost complete bind-
ing of [11C]2 to plasma proteins. This high protein binding may
reflect reversible binding as well as Michael addition reaction of
thiols of plasma protein23,24 since [11C]2 contains electron deficient
olefin. Most of [11C]2 would bind to plasma proteins and the
remaining [11C]2 would accumulate highly in the liver, which is
considered to be the cause of low tumor uptake of [11C]2.
Fig. 2. Competitive inhibition assays of lactate influx with
a
-cyano-4-hydroxyci-
namic acid (CHC) (A), compounds 2 (B) and compound 3 (C). [14C]lactate uptake in
Becker cells was determined by measuring intracellular radioactivity after 60-min
incubation with each inhibitor (n = 3) at 37 °C. Radioactivity was normalized by
protein amount. IC50 is the compound concentration (mol/L) that reduces lactate
uptake by 50%.
In contrast with the low tumor accumulation, [11C]2 was read-
ily taken up into brain region and uptake of radioactivity after
injections peaked at 7.6% ID/g at 35 s post administration, although
the high radioactivity in the brain region, [11C]2 disappeared
quickly (Fig. 5E). Because MCT1 expresses at blood-brain barrier
and cerebral blood flow is greater than tumor blood flow, [11C]2
may be transported into the brain before conjugation with plasma
proteins. We therefore considered the molecular properties of the
scaffold of 2 as preferable for passage through the blood-brain bar-
rier. However, insufficient affinity of [11C]2 for MCT1 may cause
low retention, resulting in rapid clearance from the brain. Brain
MCT1 is predominantly expressed in astrocytes and is involved
in the astrocyte–neuron lactate shuttle (ANLS).25,26 The ANLS plays
Fig. 3. Expression of MCT1 and MCT4 mRNA in three cell lines. Relative mRNA
expression levels of MCT1 and MCT4 in Becker, AsPC-1, and MDA-MB-231 cells
(n = 3); mRNA expression levels in AsPC-1 cells were taken as 100. Fold changes in
expression were determined using the DDCt method and were normalized to those
of the 18S rRNA internal control.
Please cite this article in press as: Tateishi H., et al. Bioorg. Med. Chem. Lett. (2017), https://doi.org/10.1016/j.bmcl.2017.09.033

4
H. Tateishi et al. / Bioorganic & Medicinal Chemistry Letters xxx (2017) xxx–xxx
Scheme 3. Synthetic of [11C]2. Reagents and conditions: (a) TBAF, THF, 30 °C, 3 min; (b) compound 7, 70 °C, 7 min; (c) 2 M NaOH aq., 25 °C, 8 min.
Fig. 5. PET images and time activity curves (TAC) in organs of interest. A, PET images (maximum intensity projection) at 30, 210, and 3300 s after administration of [11C]2; B,
TAC of bladder and intestine; C, TACs of liver, heart, kidney and muscle; D, TAC of Becker tumors; E, Expanded widow of TAC of brain and heart.
important roles in energy metabolism and survival of neurons like
tumor cells, and decreased activity of ANLS is considered a hall-
mark of neurodegenerative disease.27 In addition, several studies
indicate that lactate acts as a key neuroprotective substance to
CNS disorders.28–30 Therefore, MCT1-targeted imaging may facili-
tate further studies of pathological mechanisms of CNS disorders.
Thus, further improvements of the inhibitory ability of 2 against
MCT1 will follow structure optimization of the scaffold of 2 by
changing the 7-amino substituent, which significantly impacts
affinity for MCTs.21
In conclusion, 7-amino carboxycoumarin derivatives 2 and 3
were synthesized, and 2 had approximately 55 times higher affin-
ity for MCTs than the representative MCT inhibitor CHC. Com-
pound 2 was also transported into cells predominantly by MCT1.
Although 2 was radiolabeled with 11C using our labeling method,
the high protein binding affinity prevented uptake of [11C]2 into
tumors. However, we observed high initial [11C]2 uptake into the
brain and subsequent rapid clearance. The present data warrant
further development of MCT-targeted PET imaging agents based
on 7-amino-3-carboxy-coumarin.
Please cite this article in press as: Tateishi H., et al. Bioorg. Med. Chem. Lett. (2017), https://doi.org/10.1016/j.bmcl.2017.09.033

H. Tateishi et al. / Bioorganic & Medicinal Chemistry Letters xxx (2017) xxx–xxx
5
6. Mazurek S, Drexler HCA, Troppmair J, Eigenbrodt E, Rapp UR. Anticancer Res.
2007;27:3963.
Acknowledgments
7. DeBerardinis RJ, Mancuso A, Daikhin E, et al. Proc Natl Acad Sci USA.
2007;104:19345.
8. Mazurek S, Zwerschke W, Jansen-Durr P, Eigenbrodt E. Biochem J.
2001;356:247.
9. Schwickert G, Walenta S, Sundfor K, Rofstad EK, Mueller-Klieser W. Cancer Res.
1995;55:4757.
10. Spector JA, Mehrara BJ, Greenwald JA, et al. Am J Physiol Cell Physiol. 2001;280:
C72.
11. Dhup S, Dadhich RK, Porporato PE, Sonveaux P. Curr Pharm Des. 2012;18:1319.
12. Walenta S, Salameh A, Lyng H, et al. Am J Pathol. 1997;150:409.
13. Walenta S, Wetterling M, Lehrke M, et al. Cancer Res. 2000;60:916.
14. Kumar VBS, Viji RI, Kiran MS, Sudhakaran PR. J Cell Physiol. 2007;211:477.
15. Hirschhaeuser F, Sattler UG, Mueller-Klieser W. Cancer Res. 2011;71:6921.
16. Sonveaux P, Vegran F, Schroeder T, et al. J Clin Invest. 2008;118:3930.
17. Vegran F, Boidot R, Michiels C, Sonveaux P, Feron O. Cancer Res. 2011;71:2550.
18. Vaquero JJ, Kinahan P. Annu Rev Biomed Eng. 2015;17:385.
19. Hildyard JCW, Halestrap AP. Biochem J. 2003;374:607.
20. Wang X, Levi AJ, Halestrap AP. Am J Physiol. 1996;270:476.
21. Draoui N, Schicke O, Fernandes A, et al. Bioorg Med Chem. 2013;21:7107.
22. Takashima M, Kato K, Ogawa M, Magata Y. RSC Adv. 2013;3:21275.
23. Kwon H, Lee K, Kim HJ. Chem Commun (Camb). 2011;47:1773.
24. Sun YQ, Chen M, Liu J, Lv X, Li JF, Guo W. Chem Commun (Camb).
2011;47:11029.
We thank Naoko Kuroda and Yuriko Ogawa for technical assis-
tance, Nobuki Nengaki and Hisashi Suzuki for producing the
radioisotope and providing technical support for remote-con-
trolled synthesis, the staff of the Cyclotron Operation section for
cyclotron operation, and Tomoya Uehara and Hiroyuki Suzuki for
MS operation. This work was supported in part by KAKENHI
15K09976, 15K19831, and 16K10304.
A. Supplementary data
Supplementary data associated with this article can be found, in
the online version, at http://dx.doi.org/10.1016/j.bmcl.2017.09.
033.
References
25. Pellerin L, Magistretti PJ. Proc Natl Acad Sci USA. 1994;91:10625.
26. Rouach N, Koulakoff A, Abudara V, Willecke K, Giaume C. Science.
2008;322:1551.
27. Newington JT, Harris RA, Cumming RC. J Neurodegener Dis. 2013;2013:234572.
28. Cureton EL, Kwan RO, Dozier KC, Sadjadi J, Pal JD, Victorino GP. J Surg Res.
2010;159:468.
1. Pérez-Escuredo J, Van Hée VF, Sboarina M, et al. Biochim Biophys Acta.
2016;1863:2481.
2. Halestrap AP, Price NT. Biochem J. 1999;343:281.
3. Pinheiro C, Longatto-Filho A, Azevedo-Silva J, Casal M, Schmitt FC, Baltazar F. J
Bioenergy Biomembr. 2012;44:127.
29. Coco M, Caggia S, Musumeci G, et al. J Neurosci Res. 2013;91:313.
30. Voss MW, Erickson KI, Prakash RS, et al. Brain Behav Immun. 2013;28:90.
4. Vander Heiden MG, Cantley LC, Thompson CB. Science. 2009;324:1029.
5. Lu H, Forbes RA, Verma A. J Biol Chem. 2002;277:23111.
Please cite this article in press as: Tateishi H., et al. Bioorg. Med. Chem. Lett. (2017), https://doi.org/10.1016/j.bmcl.2017.09.033