Synthesis and evaluation of 18F-labeled mitiglinide derivatives as positron emission tomography tracers for β-cell imaging

Article abstract of DOI:10.1016/j.bmc.2014.04.059

Measuring changes in β-cell mass in vivo during progression of diabetes mellitus is important for understanding the pathogenesis, facilitating early diagnosis, and developing novel therapeutics for this disease. However, a non-invasive method has not been developed. A novel series of mitiglinide derivatives (o-FMIT, m-FMIT and p-FMIT; FMITs) were synthesized and their binding affinity for the sulfonylurea receptor 1 (SUR1) of pancreatic islets were evaluated by inhibition studies. (+)-(S)-o-FMIT had the highest affinity of our synthesized FMITs (IC₅₀ = 1.8 μM). (+)-(S)-o-[ ¹⁸F]FMIT was obtained with radiochemical yield of 18% by radiofluorination of racemic precursor 7, hydrolysis, and optical resolution with chiral HPLC; its radiochemical purity was >99%. In biodistribution experiments using normal mice, (+)-(S)-o-[¹⁸F]FMIT showed 1.94 ± 0.42% ID/g of pancreatic uptake at 5 min p.i., and decreases in radioactivity in the liver (located close to the pancreas) was relatively rapid. Ex vivo autoradiography experiments using pancreatic sections confirmed accumulation of (+)-(S)-o-[¹⁸F]FMIT in pancreatic β-cells. These results suggest that (+)-(S)-o-[¹⁸F]FMIT meets the basic requirements for an radiotracer, and could be a candidate positron emission tomography tracer for in vivo imaging of pancreatic β-cells.

Full text of DOI:10.1016/j.bmc.2014.04.059

Bioorganic & Medicinal Chemistry xxx (2014) xxx–xxx
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry
journal homepage: www.elsevier.com/locate/bmc
Synthesis and evaluation of 18F-labeled mitiglinide derivatives
as positron emission tomography tracers for b-cell imaging
Hiroyuki Kimura a, Hirokazu Matsuda a,c, Hiroyuki Fujimoto b, Kenji Arimitsu a, Kentaro Toyoda d,
Eri Mukai e, Hiroshi Nakamura c, Yu Ogawa a, Mikako Takagi c, Masahiro Ono a, Nobuya Inagaki b,
Hideo Saji a,
⇑
a Department of Patho-Functional Bioanalysis, Graduate School of Pharmaceutical Sciences, Kyoto University, 46-29 Yoshida Shimoadachi-cho, Sakyo-ku, Kyoto 606-8501, Japan
b Department of Diabetes, Endocrinology and Nutrition, Graduate School of Medicine, Kyoto University, 54 Shogoin Kawahara-cho, Sakyo-ku, Kyoto 606-8507, Japan
c Research & Development Division, Arkray, Inc., Yousuien-nai, 59 Gansuin-cho, Kamigyo-ku, Kyoto 602-0008, Japan
d Institute for Advancement of Clinical and Translational Science, Kyoto University Hospital, 54 Shogoin Kawahara-cho, Sakyo-ku, Kyoto 606-8507, Japan
e Department of Medical Physiology, Graduate School of Medicine, Chiba University, 1-8-1 lnohana, Chuo-ku, Chiba, Chiba 260-8677, Japan
a r t i c l e i n f o
a b s t r a c t
Article history:
Measuring changes in b-cell mass in vivo during progression of diabetes mellitus is important for under-
standing the pathogenesis, facilitating early diagnosis, and developing novel therapeutics for this disease.
However, a non-invasive method has not been developed. A novel series of mitiglinide derivatives
(o-FMIT, m-FMIT and p-FMIT; FMITs) were synthesized and their binding affinity for the sulfonylurea
receptor 1 (SUR1) of pancreatic islets were evaluated by inhibition studies. (+)-(S)-o-FMIT had the highest
Received 26 March 2014
Revised 26 April 2014
Accepted 28 April 2014
Available online xxxx
affinity of our synthesized FMITs (IC50 = 1.8 l
M). (+)-(S)-o-[18F]FMIT was obtained with radiochemical
Keywords:
yield of 18% by radiofluorination of racemic precursor 7, hydrolysis, and optical resolution with chiral
HPLC; its radiochemical purity was >99%. In biodistribution experiments using normal mice, (+)-
(S)-o-[18F]FMIT showed 1.94 0.42% ID/g of pancreatic uptake at 5 min p.i., and decreases in radioactivity
in the liver (located close to the pancreas) was relatively rapid. Ex vivo autoradiography experiments
using pancreatic sections confirmed accumulation of (+)-(S)-o-[18F]FMIT in pancreatic b-cells. These
results suggest that (+)-(S)-o-[18F]FMIT meets the basic requirements for an radiotracer, and could be a
candidate positron emission tomography tracer for in vivo imaging of pancreatic b-cells.
Ó 2014 Elsevier Ltd. All rights reserved.
Mitiglinide
Positron emission tomography tracers
18F
Sulfonylurea receptor 1
1. Introduction
preserve BCM in humans is not known. Therefore, a non-invasive
method for BCM measurement is required urgently for under-
The International Diabetes Federation has reported that the
total number of people with diabetes mellitus will rise from 285
million in 2009 to 435 million in 2030. There are two major types
of diabetes mellitus: type-1 diabetes (T1D) and type-2 diabetes
(T2D).1 T1D is characterized by selective destruction of b-cells by
the autoimmune reaction, and pancreatic b-cell mass (BCM) in
T1D is already lost at disease onset. T2D is characterized by the
decreased ability to secrete insulin and increased insulin
resistance, it also has been reported that BCM in T2D is decreased
significantly.2–5 Recently, incretin-related drugs were reported to
have proliferative and anti-apoptotic effects on pancreatic b-cells
in in vitro or rodent experiments.6–8 However, to date, non-inva-
sive quantification of BCM has not been possible, so how and when
a decrease in BCM begins and whether incretin-related drugs
standing the pathogenesis, facilitating early diagnosis, and devel-
oping novel therapeutics for diabetes.
Sulfonylurea (SU) agents are insulin secretagogues, and are
used widely in T2D treatment. They close the adenosine triphos-
phate-sensitive potassium ion channel (KATP channel) in the
pancreatic b-cell membrane, thereby depolarizing the cell and trig-
gering insulin secretion.9 The KATP channel is an octameric complex
of two protein subunits, an inwardly rectifying K+ (Kir6.x) ion
channel and a sulfonylurea receptor (SUR), which is the target for
SU agents.9,10 The pancreatic b-cell KATP channel is composed of
Kir6.2 and SUR1 subunits.11 Other KATP channels comprise Kir6.2
and SUR2A (heart muscle and skeletal muscle), Kir6.2 and SUR2B
(smooth muscle), and Kir6.1 and SUR2B (vascular smooth
muscle).9 In recent years, SUR2B expression in human liver tissue
has been described.12
Several radiolabeled tracers based on the core structure of SU
agents have been developed as imaging probes for SUR1 (Fig. 1)
⇑
Corresponding author. Tel.: +81 75 753 4566; fax: +81 75 753 4568.
E-mail address: hsaji@pharm.kyoto-u.ac.jp (H. Saji).
http://dx.doi.org/10.1016/j.bmc.2014.04.059
0968-0896/Ó 2014 Elsevier Ltd. All rights reserved.
Please cite this article in press as: Kimura, H.; et al. Bioorg. Med. Chem. (2014), http://dx.doi.org/10.1016/j.bmc.2014.04.059

2
H. Kimura et al. / Bioorg. Med. Chem. xxx (2014) xxx–xxx
O
O
O
O
O
S
O
18F
S
O
O
N
N
H
N
N
H
H
H
18F
N
H
O
18F]Tolubutamide
[
18F]Glibenclamide
[
Br
COOH
O
COOH
O
O
18F
11CH3
O
N
H
N
H
N
N
18F]Repaglinide
[
11C]Repaglinide
[
Figure 1. Chemical structures of radioligands for imaging of SUR1.
in recent years.13–18 However, these imaging probes are not avail-
able because of low uptake in pancreatic tissue and high uptake in
liver tissue located close to the pancreas. Promising candidates for
pancreatic b-cell imaging tracers based on SU agents need to have
relatively low lipophilicity and have high affinity for SUR1. In this
context, mitiglinide is a promising candidate because of its low
lipophilicity (calculated logD (clogD) value (pH 7.4) for mitigli-
nide, ꢀ0.29; glibenclamide, 1.14; repaglinide, 1.99) and high
affinity and selectivity for SUR1 (SUR1/Kir6.2:3.8 nM, SUR2A/
Kir6.2:3200 nM, SUR2B/Kir6.2:4600 nM).19
We synthesized a novel series of mitiglinide derivatives, 2-(flu-
oroethyloxybenzyl)-4-(cis-3a,4,5,6,7,7a-hexahydroisoindolin-2-yl)-
4-oxobutanoic acid (FMITs) and 2S-(2-[18F]fluoroethyloxybenzyl)-
4-(cis-3a,4,5,6,7,7a-hexahydroiso-indolin-2-yl)-4-oxobutanoic acid
[(+)-(S)-o-[18F]FMIT] and evaluated their potential as in vivo pan-
creatic b-cell imaging tracers (Fig. 2).
2. Results and discussion
2.1. Chemistry
The synthesis of mitiglinide derivatives containing fluorine
(o-, m- and p-FMIT) is shown in Scheme 1. The succinic acid deriv-
atives 2a–c were prepared, using a method based on the synthetic
strategy reported by Kamijo and co-workers20 by cross-aldol con-
densation of the benzaldehydes 1a, b, and c (which have a benzyl-
oxy group at the o-, m-, and p-position, respectively) with dimethyl
succinate in the presence of potassium hydroxide, followed by
hydrolysis of the ester. Acids 2a–c were treated with acetic
anhydride in dichloromethane to afford the cyclic acid anhydride
derivatives, followed by amidation with 3a,4,7,7a-tetrahydroiso-
indoline (3) and successive esterification to afford the amides
4a–c. Catalytic hydrogenation of 4a–c gave the phenols 5a–c.
F
18F
O
O
O
O
O
H
N
COOH
H
H
N
COOH
N
COOH
H
H
H
Mitiglinide
[
18F]FMIT
FMIT
Figure 2. Structure of mitiglinide and 18F-labeled mitiglinide derivatives.
1) Ac2O
2)
H
BnO
NH
, CH2Cl2
dimethyl succinate
BnO
HO2C
KOH, DMSO
3
H
O
BnO
H
then 4 N NaOH aq.
CO2Me
3) SOCl2, MeOH
CHO
N
CO2H
o- : 1a
1b
o- : 4a
4b
2a
o- :
m- : 2b
2c
H
m- :
p- : 1c
m- :
p- : 4c
p- :
F
(+)-FMIT
(o-, m- and p-)
HO
1) 2-fluoroethyl tosylate
Cs2CO3, DMF
O
separatin
with a chiral column
Pd-C
MeOH
O
+
O
H
H
2) 2 N NaOH aq.
1,4-dioxane
CO2Me
CO2H
N
N
(−)-FMIT
5a
(o-, m- and p-)
o- :
m- : 5b
5c
H
(
(
(
H
)-o-FMIT
)-m-FMIT
)-p-FMIT
p- :
Scheme 1. Syntheses of mitiglinide derivatives (+)- and (ꢀ)-FMIT.
Please cite this article in press as: Kimura, H.; et al. Bioorg. Med. Chem. (2014), http://dx.doi.org/10.1016/j.bmc.2014.04.059

H. Kimura et al. / Bioorg. Med. Chem. xxx (2014) xxx–xxx
3
Finally, 5a–c were alkylated with 2-fluoroethyl tosylate in the
presence of cesium carbonate in DMF, followed by saponification
with aqueous sodium hydroxide in 1,4-dioxane to give ( )-FMIT.
Optical resolution using HPLC with a chiral column gave enantio-
merically pure (+)-FMIT and (ꢀ)-FMIT (o-, m- and p-, respectively).
enantiomer suggested that the S-configuration of o-FMIT (which
has the same absolute configuration as mitiglinide) was important
for binding affinity.
2.4. Radiochemistry
2.2. In vitro binding assays
The preparation of precursor 7 and its radiofluorination, shown
in Scheme 3, were undertaken to obtain 18F-labeled (+)-(S)-o-FMIT.
The phenol 5a, with a phenolic hydroxyl group at the o-position,
was alkylated with 2-bromoethyl tosylate to give the bromo deriv-
ative 7. Although we attempted to label the chiral 7 (given by opti-
cal resolution) with 18F, we found that the resulting o-[18F]FMIT
was racemized. Therefore, racemic 7 was used as a precursor,
treated with [18F]KF and Kryptofix222 in acetonitrile at 140 °C in
a sealed tube, to afford the 18F-labeled methyl ester. This was
saponified successively with an aqueous solution of sodium
hydroxide in 1,4-dioxane. Then, optical resolution using HPLC with
a chiral column of the racemic o-[18F]FMIT gave the enantiomeri-
cally pure o-[18F]FMIT (Fig. 4). The radiochemical identity of
o-[18F]FMIT was verified by co-injection HPLC analyses with nonra-
dioactive o-FMIT. Each chiral (+)-(S)-o-[18F]FMIT showed a single
radioactive peak at the same retention time as corresponding non-
radioactive o-FMIT. The radiochemical yields of (+)-(S)-o-[18F]FMIT
and (–)-(R)-o-[18F]FMIT were 18% and 13%, respectively. Both of
their radiochemical purities after purification by HPLC with a chiral
column were >99%.
The FMITs were evaluated using dispersed islet cells from ddY
mice pancreas. The binding of [3H]glibenclamide was displaced
by (+)-o-FMIT, (+)-m-FMIT, and (ꢀ)-p-FMIT in a concentration-
dependent manner, similar to mitiglinide Ca2+ (Fig. 3). The results
of these assays are summarized in Table 1. IC50 values of (+)-
o-FMIT, (+)-m-FMIT, and (ꢀ)-p-FMIT were 1.8, 14.0, and 10.7
lM,
respectively. Namely, (+)-o-FMIT had the highest binding affinity
for pancreatic islet SUR1 in our synthesized FMITs, but its affinity
was lower than that of mitiglinide. The other enantiomers, (ꢀ)-
o-FMIT, (ꢀ)-m-FMIT and (+)-p-FMIT, did not show inhibitory
potency.
2.3. Stereochemistry
The absolute configuration of (+)-o-FMIT and (ꢀ)-o-FMIT was
determined by 1H NMR spectra using corresponding
a-methoxy-a-
(trifluoromethyl)phenylacetic acid (MTPA) esters with Eu(fod)3.21–24
(+)-o-FMIT and (ꢀ)-o-FMIT were transformed into corresponding
diastereomeric esters 6 by reduction of the carboxyl group and
esterification according to the method of Dale et al.25,26 The mag-
nitude of the lanthanide-induced shift (LIS) by Eu(fod)3 for the
methoxyl group of 6 in 1H NMR indicated that the configuration
of (+)-o-FMIT was S and that of (ꢀ)-o-FMIT was R (Scheme 2).
The result of the binding affinity for pancreatic b-cell SUR1 of each
2.5. Biodistribution studies
The tissue time-course distributions of (+)-(S)-o-[18F]FMIT in
male ddY mice are summarized in Table 2. (+)-(S)-o-[18F]FMIT
showed 1.94 0.42% ID/g of pancreatic uptake at 5 min p.i. and
radioactivity washout from the pancreas was relatively slow
(1.45 0.19% ID/g at 60 min p.i.). Radioactivity in the liver showed
rapid clearance (20.83 1.35% ID/g at 5 min p.i. and 2.18 0.34%
ID/g at 120 min p.i.). Radioactivity in the kidney showed rapid
clearance (7.77 3.30% ID/g at 5 min p.i. and 1.33 0.30% ID/g at
120 min p.i.). In addition, radioactivity in the bone was low, sug-
gesting that (+)-(S)-o-[18F18F]FMIT was relatively stable to in vivo
defluorination.
In the case of (–)-(R)-o-[18F]FMIT, maximum uptake in the pan-
creas was observed at 5 min p.i. and was 1.18 0.30% ID/g. Com-
parison of the two o-[18F]FMITs at 5, 15 and 60 min p.i. revealed
the amount of uptake of (+)-(S)-o-[18F]FMIT in the pancreas to be
approximately double that of (–)-(R)-o-[18F]FMIT (Fig. 5). The
difference derived from the presence or absence of the binding
affinity for SUR1 would suggest the specific binding of (+)-
(S)-o-[18F]FMIT. Pancreas-to-liver (P/L), pancreas-to-kidney (P/K),
and pancreas-to-blood (P/B) ratios at different time points are
shown in Figure 6. The P/L ratio was 60.57 at 60 min, the P/K ratio
was 60.81 at 120 min, and the P/B ratio was 60.64 at 60 min.
It is considered that the hepatic clearance of (+)-(S)-o-[18F]FMIT
is relatively rapid and that the P/L is high when compared with
other radiolabeled compounds based on SU agents.14–16
Figure 3. Inhibition of [3H]glibenclamide binding to islet cells at various concen-
trations of (+)-o-FMIT, (+)-m-FMIT, and (ꢀ)-p-FMIT. Specific binding of [3H]gliben-
clamide and displacement curves by FMITs are shown. Data are the mean SD
(n = 4).
Table 1
Binding inhibition study of mitiglinide and FMITs with [3H]glibenclamide
a
b
Compound
tR (min)
IC50 (lM)
2.6. Ex vivo autoradiography
Mitiglinide
o-FMIT
—
0.14 (0.12–0.17)
1.8 (1.2–2.6)
N.D.
14.0 (10.8–18.1)
N.D.
(+)-
(ꢀ)-
(ꢀ)-
(+)-
(ꢀ)-
(+)-
11.6
24.3
12.5
44.6
17.4
28.8
To further characterize the potential of (+)-(S)-o-[18F]FMIT as an
agent for imaging pancreatic b-cells, we undertook an ex vivo
autoradiography study in normal ddY mice (6-week-old, male).
Radioactive signals were observed in pancreatic sections, whose
localizations were relatively consistent with islets colored by
insulin immunostaining on serial sections, which suggested that
(+)-(S)-o-[18F]FMIT accumulated in pancreatic b-cells. However,
radioactive signals in other regions were also observed, which
suggested non-specific binding of (+)-(S)-o-[18F]FMIT (Fig. 7).
m-FMIT
p-FMIT
10.7 (7.0–16.1)
N.D.
a
Retention time using chiral HPLC (CHIRALPAK AY-H, 10 ꢁ 250 mm, stationary
phase acetonitrile/methanol/TFA = 95/5/0.1, flow rate 2.0 mL/min).
b
Best-fit values and 95% confidence intervals of 50% inhibitory concentrations
obtained from displacement of the [3H]glibenclamide assay. R2 values were >0.85 in
experiments used to determine the binding specificity of each compound.
Please cite this article in press as: Kimura, H.; et al. Bioorg. Med. Chem. (2014), http://dx.doi.org/10.1016/j.bmc.2014.04.059

4
H. Kimura et al. / Bioorg. Med. Chem. xxx (2014) xxx–xxx
O
O
i
F
F
1) ClCO2 Bu, Et3N, THF
(+)-o-FMIT
or
then NaBH4, H2O
Ph
Ph
CF3
OMe
CF3
+
O
O
R
S
H
H
O
O
2) (S)-Ph(CF3)(MeO)CCOCl
Et3N, DMAP, CH2Cl2
R
R
OMe
(−)-o-FMIT
N
N
O
O
(S,R)-6
from (+)-o-FMIT
(R,R)-6
from (−)-o-FMIT
H
H
ΔLISOMe (S,R)-6
<
(R,R)-6
Scheme 2. Syntheses of MTPA ester (S,R)-6 and (R,R)-6.
O
O
O
18F
18F
Br
TsO
18
K
Br
[
F]KF,
MeCN
1)
2)
222
Cs2CO3
O
O
H
O
H
H
+
5a
H
H
CO2
CO2
CO2Me
DMF
2 N NaOH aq.
1,4-dioxane
N
N
N
7
H
H
H
then
separation
a
−
with chiral column
-[18F]FMIT
-[18F]FMIT
o
o
R
S
(+)-( )-
( )-( )-
Scheme 3. Radiosynthesis of (+)-(S)- and (ꢀ)-(R)-o-[18F]FMIT.
tion at the 2-position of FMITs must be especially important to the
affinity for SUR1.
We synthesized (+)-(S)-o-[18F]FMIT and evaluated its potential
as an in vivo pancreatic b-cell imaging tracer. In in vivo experi-
ments, (+)-(S)-o-[18F]FMIT showed pancreatic uptake and rela-
tively rapid clearance in the liver and kidney. The detected
localization of the radioactive signals by ex vivo autoradiography
corresponded with that of pancreatic b-cells shown by insulin
immunostaining.
Further enhancements of pancreatic uptake and specific bind-
ing to pancreatic b-cells are required for development of a poten-
tial tracer for in vivo imaging of pancreatic b-cells. Nevertheless,
these results provide useful information for targeting SUR1 of pan-
creatic b-cells.
Figure 4. HPLC chromatograms of (+)-(S)-o-[18F]FMIT and (ꢀ)-(R)-o-[18F]FMIT.
Peaks at a retention time of 22.9 min and 46.4 min were (+)-(S)-o-[18F]FMIT and
(ꢀ)-(R)-o-[18F]FMIT, respectively.
4. Experimental sections
4.1. General
3. Conclusions
We synthesized a novel series of mitiglinide derivatives (o-
FMIT, m-FMIT, and p-FMIT) and evaluated their binding affinity
for pancreatic islet SUR1. (+)-o-FMIT showed the highest affinity
of our synthesized FMITs, which suggested that introduction of a
fluoroethoxy group at the o-position affects the affinity less than
at other positions. In addition, we determined the absolute config-
uration of each enantiomer of o-FMIT by 1H NMR spectra using
MTPA esters with Eu(fod)3, which showed that the exact configura-
[
18F]Fluoride was produced in a cyclotron (CYPRIS HM-18,
Sumitomo Heavy Industries Ltd., Tokyo, Japan). Radioactivity was
measured with an IGC-7 curiemeter (Aloka, Tokyo, Japan), an
Atomlab100+ (Biodex Medical Systems Inc., Shirley, NY, USA) or
CRC-15BETA (Capintec Inc., Ramsey, NJ, USA) dose calibrator or a
Wallac 1480 WIZARD 300 (PerkinElmer Inc., Waltham, MA, USA)
c
-counter.
Table 2
Tissue distribution of (+)-(S)-o-[18F]FMIT in ddY mice
Organ or tissue
5 min
15 min
30 min
60 min
120 min
Pancreas
Blood
Heart
1.94 0.42
3.93 0.76
2.63 0.42
2.81 0.65
2.22 1.06
12.25 4.46
20.83 1.35
1.91 0.28
7.77 3.30
1.40 0.31
1.65 0.15
2.76 0.26
2.16 0.19
2.25 0.20
5.36 3.32
26.72 3.65
5.40 0.46
1.71 0.17
5.98 1.20
1.22 0.11
1.30 0.12
1.45 0.19
2.27 0.28
2.10 0.26
1.84 0.33
2.07 1.51
31.79 6.13
2.55 0.20
1.61 0.30
2.16 0.48
1.80 0.19
1.08 0.17
2.16 0.51
1.72 0.11
1.37 0.14
2.66 1.47
13.84 5.08
2.18 0.34
1.22 0.12
1.33 0.30
3.01 0.62
2.07 0.18
1.82 0.21
1.71 0.22
2.54 0.71
35.27 4.73
3.44 1.73
1.36 0.17
3.16 1.07
1.22 0.08
Lung
Stomach
Intestine
Liver
Spleen
Kidney
Bone
Values are % ID/g (mean SD) for 5 animals at each interval.
Please cite this article in press as: Kimura, H.; et al. Bioorg. Med. Chem. (2014), http://dx.doi.org/10.1016/j.bmc.2014.04.059

H. Kimura et al. / Bioorg. Med. Chem. xxx (2014) xxx–xxx
5
Tokyo, Japan) chiral HPLC column. Specific rotations were recorded
using a SEPA-500 automatic digital polarimeter (Horiba, Kyoto,
Japan). All chemicals used were of reagent grade. Calculated logD
values were obtained by computational methods using ACD/Labs
v12.0 (Advanced Chemistry Development, Toronto, ON, Canada).
4.2. Animal experiments
Six-week-old male ddY mice were obtained from Japan SLC
(Kyoto, Japan). Animal studies were conducted in accordance with
institutional guidelines. Experimental procedures were approved
by the Animal Care Committee of Kyoto University (Kyoto, Japan).
4.3. Synthesis
4.3.1. Procedure for the preparation of dicarboxylic acids 2a–c
Potassium hydroxide (2.4 equiv) was added to a solution of o-,
m-, or p-benzyloxybenzaldehyde (1a–c) and dimethyl succinate
(3.0 equiv) in dimethyl sulfoxide, and the mixture stirred for 15 h
at 80 °C. An aqueous solution of sodium hydroxide (5 M, 30 mL)
was added and the mixture stirred for 5 h at 120 °C. After the reac-
tion, distilled water was added to the reaction mixture, which was
then washed with ethyl acetate. The aqueous layer was acidified
with 6 N hydrochloric acid, and extracted with ethyl acetate. The
organic layer was washed with brine, dried over magnesium sul-
fate, and evaporated under vacuum to give a pale-yellow solid.
The solid was washed with hot water and dried to afford dicarbox-
ylic acids 2a–c.
Figure 5. Comparative uptake of (+)-(S)- and (ꢀ)-(R)-o-[18F]FMIT at 5, 15 and
60 min p.i. in the pancreas.
1H NMR spectra were obtained on
a JEOL JNM-ECS400
(400 MHz) spectrometer (JEOL, Tokyo, Japan). Chemical shifts were
reported in ppm using tetramethylsilane or residual solvent as the
internal reference. Infrared (IR) spectra were recorded on an FTIR-
8300 diffraction grating infrared spectrophotometer (Shimadzu,
Kyoto, Japan). Mass spectra (MS) were determined on a JEOL
JMS-SX 102A QQ or a JEOL JMS-GC-mate mass spectrometer (JEOL).
Silica gel column chromatography was undertaken using a W-Prep
2XY system (Yamazen, Osaka, Japan) using a Hi Flash™ 40 mm
60 Å silica gel column (Yamazen). Kieselgel 60 F-254 plates
(0.25 mm; Merck, Darmstadt, Germany) were used for thin-layer
chromatography (TLC). Preparative thin-layer chromatography
(PTLC) was conducted with a silica gel 60 F-254 plate (0.5 mm;
Merck). An LC-20AD system (Shimadzu) was used for high-
performance liquid chromatography (HPLC), with SPD-20A (Shi-
madzu) ultraviolet (UV; k = 254 nm) and NDW-351D (Aloka,
4.3.1.1. 2-(2-Benzyloxybenzylidene)succinic acid (2a).
The
reaction was carried out with o-benzyloxybenzaldehyde (1a)
(5.00 g, 23.6 mmol) in dimethyl sulfoxide (24 mL) to afford 2a
(2.98 g, 40%). mp 216–217 °C (recrystallized from ethyl acetate).
1H NMR (DMSO-d6) d 3.30 (2H, s), 5.17 (2H, s), 7.00 (1H, t,
J = 7.7 Hz), 7.14 (1H, d, J = 8.5 Hz), 7.26–7.45 (7H, m), 7.86 (1H, s),
12.47 (2H, br s). IR (KBr)
m 3042, 2934, 2872, 2540, 2361, 2341,
1697, 1630, 1452, 1420, 1290, 1232 cmꢀ1. EI-MS m/z: 312 (M+).
HRMS m/z: 312.1002 (Calcd for C18H16O5: 312.0997). Anal. Calcd
for C18H16O5: C, 69.22; H, 5.16. Found: C, 69.28; H, 5.20.
Tokyo, Japan) radioisotope (RI) detectors, and
a Cosmosil
5C18-AR-II (10 ꢁ 250 mm; Nacalai Tesque, Kyoto, Japan) reverse-
phase HPLC column and a CHIRALPAK AY-H (10 ꢁ 250 mm; Daicel,
Figure 6. Uptake ratios of (+)-(S)-o-[18F]FMIT. (a) pancreas/liver, (b) pancreas/kidney, and (c) pancreas/blood. Ratios were calculated based on the % ID/g of each organ.
Please cite this article in press as: Kimura, H.; et al. Bioorg. Med. Chem. (2014), http://dx.doi.org/10.1016/j.bmc.2014.04.059

6
H. Kimura et al. / Bioorg. Med. Chem. xxx (2014) xxx–xxx
Figure 7. (a) Autoradiography of (+)-(S)-o-[18F]FMIT. (b) Localization with insulin by immunohistochemistry.
4.3.1.2. 2-(3-Benzyloxybenzylidene)succinic acid (2b).
The
4.3.2.1.
Methyl
2-(2-benzyloxybenzylidene)-4-oxo-4-(cis-
The
reaction was carried out with m-benzyloxybenzaldehyde (1b)
(7.50 g, 35.3 mmol) in dimethyl sulfoxide (30 mL) to afford 2b
(3.84 g, 35%). mp 197–198 °C (recrystallized from ethyl acetate).
1H NMR (DMSO-d6) d 3.40 (2H, s), 5.16 (2H, s), 7.01–7.09 (3H,
3a,4,7,7a-tetrahydro-isoindolin-2-yl) butanoate (4a).
reaction was carried out with 2a (2.83 g, 9.06 mmol) to afford 4a
(2.87 g, 73%). 1H NMR (CDCl3) d 1.68 (2H, m), 2.23–2.32 (3H, m),
2.43 (1H, m), 3.29 (1H, dd, J = 6.9, 9.6 Hz), 3.40 (1H, m), 3.42 (2H,
s), 3.53–3.61 (2H, m), 3.80 (3H, s), 5.14 (2H, s), 5.65 (2H, m),
7.27–7.42 (6H, m), 7.55 (1H, br d, J = 7.1 Hz), 8.10 (1H, s). IR (CHCl3)
m), 7.35–7.50 (6H, m), 7.75 (1H, s), 12.60 (2H, br s). IR (KBr)
m
2955, 2866, 2544, 1703, 1678, 1578, 1431, 1267, 1225 cmꢀ1. EI-
MS m/z: 312 (M+). HRMS m/z: 312.0990 (Calcd for C18H16O5:
312.0997). Anal. Calcd for C18H16O5: C, 69.22; H, 5.16. Found: C,
69.14; H, 5.20.
m
3007, 2949, 1705, 1634, 1485, 1449, 1437, 1242 cmꢀ1. FAB-MS
m/z: 432 (M+H+). HRMS m/z: 432.2172 (Calcd for C27H29NO4:
432.2175).
4.3.1.3. 2-(4-Benzyloxybenzylidene)succinic acid (2c).
The
4.3.2.2.
Methyl
2-(3-benzyloxybenzylidene)-4-oxo-4-(cis-
The
reaction was undertaken with p-benzyloxybenzaldehyde (1c)
(5.00 g, 23.5 mmol) in dimethyl sulfoxide (30 mL) to afford 2c
(2.80 g, 38%). mp 207–208 °C (recrystallized from ethyl acetate).
1H NMR (DMSO-d6) d 3.44 (2H, s), 5.18 (2H, s), 7.13 (2H, d,
J = 8.8 Hz), 7.14–7.51 (7H, m), 7.73 (1H, s), 12.51 (2H, br s). IR
3a,4,7,7a-tetrahydro-isoindolin-2-yl) butanoate (4b).
reaction was undertaken with 2b (3.00 g, 9.61 mmol) to afford 4b
(1.85 g, 45%). 1H NMR (CDCl3) d 1.87–1.95 (2H, m), 2.20–2.33
(3H, m), 2.42 (1H, m), 3.28 (1H, dd, J = 6.8, 9.6 Hz), 3.37 (1H, dd,
J = 5.5, 11.7 Hz), 3.41 (2H, s), 3.54 (1H, dd, J = 6.4, 11.7 Hz), 3.57
(1H, dd, J = 6.8, 9.6 Hz), 3.80 (3H, s), 5.05 (2H, s), 5.63 (2H, m),
6.94 (1H, dd, J = 2.1, 8.2 Hz), 7.02 (1H, d, J = 7.6 Hz), 7.09 (1H, br),
7.24–7.41 (6H, m), 7.85 (1H, s). IR (CHCl3)
1636, 1437, 1242 cmꢀ1 EI-MS m/z: 431 (M+). HRMS m/z:
431.2105 (Calcd for C27H29NO4: 431.2096).
(KBr)
m 2918, 2849, 2540, 1703, 1674, 1605, 1512, 1425, 1258,
1231, 1175 cmꢀ1. EI-MS m/z: 312 (M+). HRMS m/z: 312.1004 (Calcd
for C18H16O5: 312.0997). Anal. Calcd for C18H16O5: C, 69.22; H, 5.16.
Found: C, 69.05; H, 5.22.
m 3007, 2951, 1709,
.
4.3.2. Procedure for the preparation of methyl esters 4a–c
Acetic anhydride (2.0 equiv) was added to a solution of acids
2a–c in dichloromethane (40 mL), which was stirred for 1.5 h at
60 °C. The reaction mixture was evaporated under vacuum, and
dichloromethane (6 mL) added to give a suspension. A solution of
3 (1.0 equiv) in dichloromethane (4 mL) at 0 °C was added to this
suspension, and the mixture stirred for 5 h at room temperature.
After the reaction, the mixture was evaporated and dissolved in
methanol (20 mL). Thionyl chloride (1.5 mL) at 0 °C was added to
this solution, and the mixture stirred for 15.5 h at room tempera-
ture. The reaction mixture was evaporated, poured into a saturated
aqueous solution of sodium bicarbonate and extracted with ethyl
acetate. The organic layer was washed with brine, dried over mag-
nesium sulfate, and evaporated under vacuum. The residue was
purified using silica gel column chromatography (n-hexane/ethyl
acetate = 1/1) to afford 4a–c.
4.3.2.3.
Methyl
2-(4-benzyloxybenzylidene)-4-oxo-4-(cis-
The
3a,4,7,7a-tetrahydro-isoindolin-2-yl) butanoate (4c).
reaction was carried out with 2c (3.00 g, 9.61 mmol) to afford 4c
(2.89 g, 70%). 1H NMR (CDCl3) d 1.89–1.98 (2H, m), 2.23–2.36
(3H, m), 2.46 (1H, m), 3.34 (dd, 1H, J = 7.0, 9.5 Hz), 3.40 (1H, dd,
J = 5.3, 11.7 Hz), 3.46 (2H, br s), 3.56 (1H, dd, J = 6.2, 11.7 Hz),
3.63 (1H, dd, J = 6.8, 9.5 Hz), 3.79 (3H, s), 5.08 (2H, s), 5.67 (2H,
m), 6.97 (2H, d, J = 8.8 Hz), 7.31–7.45 (7H, m), 7.87 (1H, s). IR
(CHCl3)
m
3007, 2949, 1703, 1636, 1437, 1244 cmꢀ1. EI-MS m/z:
431 (M+). HRMS m/z: 431.2091 (Calcd for C27H29NO4: 431.2096).
4.3.3. Procedure for the preparation of phenol 5a–c
A catalytic amount of 10% Pd-C was added to a solution of 4a–c
in methanol and the mixture stirred for 10–40 h under a hydrogen
atmosphere (1 atom). The mixture was filtered through Hyflo
Please cite this article in press as: Kimura, H.; et al. Bioorg. Med. Chem. (2014), http://dx.doi.org/10.1016/j.bmc.2014.04.059

H. Kimura et al. / Bioorg. Med. Chem. xxx (2014) xxx–xxx
7
Super-CelÒ and the filtrate evaporated under vacuum. The residue
was purified using silica gel column chromatography (n-hexane/
ethyl acetate = 1/1) to afford 5a–c.
tion of sodium hydroxide (2 M, 5 mL) in 1,4-dioxane (20 mL) for
1 h to afford o-FMIT (315.4 mg, 85%). Two rotamers were observed
(ca. 1/1). 1H NMR (CDCl3) d 1.34–1.61 (8H, m), 2.18 (2H, m), 2.53
(2H, m), 2.85 (1H, m), 2.93 (0.5H, dd, J = 5.7, 10.5 Hz), 3.06–3.28
(3.5H, m), 3.32 (0.5H, dd, J = 6.0, 12.1 Hz), 3.38 (1H, m), 3.43 (0.5H,
dd, J = 7.1, 12.1 Hz), 4.22 (2H, br d, J = 27.5 Hz), 4.76 (2H, dt, J = 3.4,
47.4 Hz), 6.85 (1H, d, J = 8.2 Hz), 6.92 (1H, t, J = 7.3 Hz), 7.16 (1H, d,
4.3.3.1. Methyl 2-(2-hydroxybenzyl)-4-(cis-3a,4,5,6,7,7a-hexa-
hydro-isoindolin-2-yl)-4-oxobutanoate (5a).
The reaction
was undertaken with 4a (3.05 g, 7.07 mmol) and 10% Pd-C
(306.7 mg), in methanol (40 mL) for 10 h to afford 5a (2.18 g,
89%). Two rotamers were observed (ca. 1/1). 1H NMR (CDCl3) d
1.40–1.63 (8H, m), 2.21 (1H, m), 2.30 (1H, m), 2.51 (1H, m), 2.65
(1H, m), 2.95 (2H, m), 3.12 (1H, m), 3.26 (1H, m), 3.39–3.48 (3H,
m), 3.646 and 3.648 (each 1.5H, s), 6.76 (1H, t, J = 7.6 Hz), 6.90
(1H, br d, J = 8.0 Hz), 7.05 (1H, br d, J = 7.6 Hz), 7.11 (1H, m), 8.94
J = 7.6 Hz), 7.21 (1H, t, J = 7.6 Hz). IR (CHCl3)
m 3007, 2934, 2859,
1732, 1587, 1495, 1450, 1242 cmꢀ1. FAB-MSm/z: 378 (M+H+). HRMS
m/z: 378.2076 (Calcd for C21H29FNO4: 378.2081).
Racemic o-FMIT was optically resolved using preparative HPLC
(acetonitrile/methanol/TFA = 95/5/0.1, 2.0 mL/min) using a chiral
column. The enantiomer with a retention time of 11.6 min was
(+)-o-FMIT, and the enantiomer with a retention time of 24.3 min
and 8.95 (each 0.5H, s). IR (CHCl3)
m 3215, 2934, 2859, 1732,
1620, 1582, 1489, 1458, 1242 cmꢀ1. EI-MS m/z: 345 (M+). HRMS
m/z: 345.1936 (Calcd for C20H27NO4: 345.1940).
was (ꢀ)-o-FMIT. (+)-o-FMIT: [
a
] + 6.9° (c 1.01, MeOH). (ꢀ)-o-FMIT:
[a
] ꢀ5.7° (c 0.30, MeOH).
4.3.3.2. Methyl 2-(3-hydroxybenzyl)-4-(cis-3a,4,5,6,7,7a-hexa-
4.3.4.2. 2-(3-Fluoroethyloxybenzyl)-4-(cis-3a,4,5,6,7,7a-hexahy-
dro-isoindolin-2-yl)-4-oxobutanoic acid (m-FMIT). The
hydro-isoindolin-2-yl)-4-oxobutanoate (5b).
The reaction
was carried out with 4b (151.9 mg, 0.35 mmol) and 10% Pd-C
(15.2 mg) in methanol (4 mL) for 10 h to afford 5b (75.5 mg,
62%). Two rotamers were observed (ca. 1/1). 1H NMR (CDCl3) d
1.35–1.55 (8H, m), 2.05–2.29 (3H, m), 2.65 (2H, m), 3.11 (1.5H,
m), 3.27–3.45 (4.5H, m), 3.68 and 3.69 (each 1.5H, s), 6.64 (1H,
br d, J = 7.3 Hz), 6.72 (1H, m), 6.94 (1H, s), 7.09–7.14 (1H, m),
step-1 reaction was carried out with 5b (65.6 mg, 0.19 mmol) in
dimethyl formamide (2 mL) for 13 h to afford the corresponding
methyl ester (56.1 mg, 76%). The step-2 reaction was undertaken
with the methyl ester (56.1 mg, 0.14 mmol) and an aqueous solu-
tion of sodium hydroxide (2 M, 1 mL) in 1,4-dioxane (4 mL) for
1 h to afford m-FMIT (37.9 mg, 72%). Two rotamers were observed
(ca. 1/1). 1H NMR (CDCl3) d 1.35–1.56 (8H, m), 2.20 (2H, m), 2.42–
2.59 (2H, m), 2.77 (1H, m), 3.03 (0.5H, dd, J = 6.0, 10.0 Hz), 3.11–
3.23 (3.5H, m), 3.33 (0.5 H, dd, J = 6.0, 12.0 Hz), 3.38 (1H, m),
3.45 (0.5H, dd, J = 6.8, 12.0 Hz), 4.20 (2H, dt, J = 4.4, 28.0 Hz), 4.74
(2H, dt, J = 4.4, 47.2 Hz), 6.78 (3H, m), 7.22 (1H, br t, J = 7.6 Hz),
7.94 (1H, m). IR (CHCl3)
m 3302, 2932, 2859, 1728, 1624, 1599,
1456, 1240 cmꢀ1. EI-MS m/z: 345 (M+). HRMS m/z: 345.1932 (Calcd
for C20H27NO4: 345.1940).
4.3.3.3. Methyl 2-(4-hydroxybenzyl)-4-(cis-3a,4,5,6,7,7a-hexa-
hydro-isoindolin-2-yl)-4-oxobutanoate (5c).
The reaction
12.05 (1H, br). IR (CHCl3) m 3007, 2934, 2859, 1732, 1585, 1489,
was undertaken with 4c (161.5 mg, 0.37 mmol) and 10% Pd-C
(16.2 mg), in methanol (4 mL) for 40 h to afford 5c (127.7 mg,
99%). Two rotamers were observed (ca. 1.5/1). 1H NMR (CDCl3) d
1.33–1.61 (8H, m), 2.15 (1H, m), 2.21–2.32 (2H, m), 2.66 (2H, m),
2.96 (0.6H, dd, J = 4.9, 6.4 Hz), 2.99 (0.4H, dd, J = 4.9, 6.4 Hz), 3.17
(0.6H, dd, J = 6.2, 10.1 Hz), 3.24–3.47 (4.4H, m), 3.65 (1.8H, s),
1449, 1260, 1240, 1196 cmꢀ1. FAB-MS m/z: 378 (M+H+). HRMS
m/z: 378.2084 (Calcd for C21H29FNO4: 378.2081).
Racemic m-FMIT was optically resolved using preparative HPLC
(acetonitrile/methanol/TFA = 95/5/0.1, 2.0 mL/min) using a chiral
column. The enantiomer with a retention time of 12.5 min was
(ꢀ)-m-FMIT, and the enantiomer with
44.6 min was (+)-m-FMIT. (+)-m-FMIT: [ ] + 1.8° (c 1.88, MeOH).
(ꢀ)-m-FMIT: [ ] ꢀ1.7° (c 2.52, MeOH).
a retention time of
3.66 (1.2H, s), 6.75 (2H, m), 6.96–7.04 (3H, m). IR (CHCl3)
m
3312,
a
2932, 1728, 1626, 1516, 1450, 1258, 1173 cmꢀ1. EI-MS m/z: 345
(M+). HRMS m/z: 345.1946 (Calcd for C20H27NO4: 345.1940).
a
4.3.4.3. 2-(4-Fluoroethyloxybenzyl)-4-(cis-3a,4,5,6,7,7a-hexahy-
dro-isoindolin-2-yl)-4-oxobutanoic acid (p-FMIT). The
4.3.4. Procedure for the preparation of FMIT
Step 1: Cesium carbonate (1.4 equiv) was added to a solution of
5a–c and 2-fluoroethyl tosylate (1.5 equiv) in dimethyl formamide,
and the mixture stirred at 60 °C. After chilling the reaction mixture
to room temperature, the mixture was poured into water and
extracted with diethyl ether. The organic layer was washed with
a small amount of water, dried over magnesium sulfate, and evap-
orated under vacuum. The residue was purified using silica gel col-
umn chromatography (n-hexane/ethyl acetate = 1/1) to afford the
methyl ester.
Step 2: An aqueous solution of sodium hydroxide (2 M) was
added to a solution of the methyl ester in 1,4-dioxane, and the mix-
ture stirred at 50 °C. After chilling the reaction mixture to room
temperature, it was acidified with 1 N hydrochloric acid and
extracted with ethyl acetate. The organic layer was washed with
brine, dried over magnesium sulfate, and evaporated under vac-
uum. The residue was purified using silica gel column chromatog-
raphy (chloroform/methanol = 9/1) to afford FMIT.
step-1 reaction was undertaken with 5c (57.6 mg, 0.17 mmol) in
dimethyl formamide (2 mL) for 21 h to afford the corresponding
methyl ester compound (48.5 mg, 73%). The step-2 reaction was car-
ried out with the methyl ester (56.1 mg, 0.14 mmol) and an aqueous
solution of sodium hydroxide (2 M, 1 mL) in 1,4-dioxane (4 mL) for
1 h to afford p-FMIT (15.8 mg, 35%). Two rotamers were observed
(ca. 1/1). 1H NMR (CDCl3) d 1.35–1.55 (8H, m), 2.18 (2H, m), 2.38
(1H, dd, J = 4.4, 16.4 Hz), 2.55 (1H, dd, J = 6.0, 16.4 Hz), 2.75 (1H,
dd, J = 8.8, 13.6 Hz), 3.04–3.40 (6H, m), 4.16 (2H, dt, J = 4.0,
28.4 Hz), 4.73 (2H, dt, J = 4.0, 47.6 Hz), 6.84 (2H, d, J = 8.0 Hz), 7.12
(2H, d, J = 8.0 Hz), 11.62 (1H, br). IR (CHCl3)
m 3007, 2934, 2859,
1732, 1585, 1512, 1452, 1240, 1180 cmꢀ1. EI-MS m/z: 377 (M+).
HRMS m/z: 377.2008 (Calcd for C21H28FNO4: 377.2002).
Racemic p-FMIT was optically resolved using preparative HPLC
(acetonitrile/methanol/TFA = 95/5/0.1, 2.0 mL/min) using a chiral
column. The enantiomer with a retention time of 17.4 min was
(ꢀ)-p-FMIT, and the enantiomer with
28.8 min was (+)-p-FMIT. (+)-p-FMIT: [ ] + 12.2° (c 3.72, MeOH).
(ꢀ)-p-FMIT: [ ] + 12.2° (c 3.74, MeOH).
a retention time of
a
4.3.4.1. 2-(2-Fluoroethyloxybenzyl)-4-(cis-3a,4,5,6,7,7a-hexahy-
a
dro-isoindolin-2-yl)-4-oxobutanoicacid (o-FMIT).
The step-
1 reaction was carried out with 5a (386.9 mg, 1.12 mmol) in
dimethyl formamide (10 mL) for 17.5 h to afford the corresponding
methyl ester (386.0 mg, 88%). The step-2 reaction was undertaken
with the methyl ester (386.0 mg, 0.99 mmol) and an aqueous solu-
4.3.5. Procedure for the preparation of MTPA esters (S,R)-6 and
(R,R)-6
Isobutyl chloroformate (2.0 equiv) was added to a solution of
triethylamine (2.0 equiv) and (+)-o-FMIT (40.8 mg, 0.11 mmol) or
Please cite this article in press as: Kimura, H.; et al. Bioorg. Med. Chem. (2014), http://dx.doi.org/10.1016/j.bmc.2014.04.059

8
H. Kimura et al. / Bioorg. Med. Chem. xxx (2014) xxx–xxx
(ꢀ)-o-FMIT (40.1 mg, 0.11 mmol) in tetrahydrofuran (1 mL), and
the mixture stirred for 3 h at ꢀ10 °C. The reaction mixture was fil-
tered, and an aqueous solution of NaBH4 (5 equiv) added to the fil-
trate at 0 °C. The mixture was stirred for 1 h at room temperature.
After the reaction the mixture was poured into a saturated aqueous
solution of sodium bicarbonate and extracted with ethyl acetate.
The organic layer was washed with brine, dried over sodium sul-
fate, and evaporated under vacuum. The residue was purified using
preparative silica gel TLC (chloroform/methanol = 20/1) to afford
the alcohol derived from (+)-o-FMIT (11.7 mg, 30%) or the alcohol
derived from (ꢀ)-o-FMIT (9.8 mg, 25%). Triethylamine (1.2 equiv),
4.3.6. Methyl 2-(2-bromoethyloxybenzyl)-4-(cis-3a,4,5,6,7,7a-
hexahydro-isoindolin- 2-yl)-4-oxobutanoate (7)
Cesium carbonate (1.16 g, 3.6 mmol) was added to a solution of
5a (560 mg, 1.6 mmol) and 2-bromoethyl p-toluenesulfonate
(1.09 g, 3.9 mmol) in dimethyl formamide (8 mL), and the mixture
stirred at 40 °C over one weekend. After chilling the reaction mix-
ture to room temperature it was poured into water and extracted
with ethyl acetate. The organic layer was washed with water and
brine, dried over magnesium sulfate, and evaporated under vac-
uum. The residue was purified using silica gel column chromatog-
raphy (n-hexane/ethyl acetate = 4/1 to 1/1) to afford 7 (440 mg,
60%). Two rotamers were observed (ca. 1/1). 1H NMR (CDCl3) d
1.34–1.64 (8H, m), 2.14 (1H, m), 2.22 (1H, m), 2.33 (1H, m), 2.65
(0.5H, dd, J = 5.7, 9.8 Hz), 2.68 (0.5H, dd, J = 5.3, 9.6 Hz), 2.85 (1H,
m), 3.14 (1H, m), 3.17 (0.5H, dd, J = 6.2, 9.8 Hz), 3.26–3.42 (4.5H,
m), 3.63 and 3.64 (each 1.5H, s), 3.69 (2H, br t, J = 4.6 Hz), 4.30
(2H, br t, J = 4.6 Hz), 6.81 (1H, d, J = 8.2 Hz), 6.90 (1H, br t,
7.3 Hz), 7.11 (1H, d, J = 7.6 Hz), 7.19 (1H, br t, J = 8.2 Hz). IR (CHCl3)
a catalytic amount of DMAP, and (S)-
methyl)phenylacetyl chloride (1.2 equiv) were added to a solution
of the alcohol derived from (+)-o-FMIT (11.7 mg, 32.2
mol) or (ꢀ)-
o-FMIT (9.8 mg, 27.0 mol) at 0 °C, and stirred for 5 h at room tem-
a-methoxy-a-(trifluoro-
l
l
perature. The reaction mixture was poured into a saturated aque-
ous solution of sodium bicarbonate and extracted with ethyl
acetate. The organic layer was washed with brine, dried over
sodium sulfate, and evaporated under vacuum. The residue was
purified using preparative silica gel TLC (n-hexane/ethyl ace-
tate = 1/1) to afford (S,R)-6 (11.3 mg, 60%) or (R,R)-6 (11.7 mg, 75%).
m
3007, 2932, 2859, 1728, 1628, 1493, 1454, 1248, 1194 cmꢀ1. FAB-
MS m/z: 452 (M+H+). HRMS m/z: 452.1434 (Calcd for C22H31BrNO4:
452.1436).
4.3.5.1.
hexahydro-isoindolin-2-yl)-4-oxobutyl (R)-
fluoromethyl)phenylacetate (S,R-6). Two rotamers were
observed (ca. 1/1). [
] + 23.0° (c 0.55, CHCl3). 1H NMR (CDCl3) d
1.34–1.67 (8H, m), 2.20 (4H, m), 2.75 (3H, br), 3.04 (0.5H, dd,
J = 5.7, 9.8 Hz), 3.08 (0.5H, dd, J = 6.0, 9.8 Hz), 3.15 (0.5H, dd,
J = 6.9, 9.8 Hz), 3.23 (0.5 H, dd, J = 7.1, 9.8 Hz), 3.29–3.41 (2H, m),
3.55 and 3.56 (each 1.5H, s), 4.19 (2H, br d, J = 28.2 Hz), 4.27 (1H,
m), 4.42–4.48 (1H, m), 4.73 (2H, dt, J = 3.7, 47.4 Hz), 6.82 (1H, d,
J = 8.2 Hz), 6.88 (1H, br t, J = 7.3 Hz), 7.06 (1H, d, J = 7.3 Hz), 7.18
(2S)-2-(2-Fluoroethyloxybenzyl)-4-(cis-3a,4,5,6,7,7a-
4.4. In vitro binding assays
a
-methoxy- -(tri-
a
The displacement effect of the mitiglinide derivatives on SUR1
binding was assessed using dispersed islet cells as described previ-
ously.27 Pancreatic islets were isolated from male ddY mice by a
collagenase digestion method.14 Isolated islets were dispersed
using 0.05% trypsin/0.53 mM EDTA (Invitrogen, Carlsbad, CA,
USA) and PBS. Islet cells were incubated with [3H] glibenclamide
(3.7 kBq) in 1 mL of buffer containing 20 mM HEPES (pH 7.4),
1 mM MgCl2, 1 mg/mL bacitracin, and 1 mg/mL BSA for 1 h at room
temperature in the presence of varying concentrations of the non-
radioactive mitiglinide derivatives. Binding was terminated by
rapid filtration through Whatman GF/C filters (24 mm) followed
by washing thrice with 5 mL of ice-cold phosphate-buffered saline
(PBS). The radioactivity of the filters was measured with a liquid
scintillation analyzer. Results were expressed as the percentage
of the radioactivity of the bound [3H]glibenclamide remaining after
the addition of the nonradioactive compound. GraphPad Prism
software (v5.03 for Windows) was used to calculate 50% inhibitory
concentration (IC50) values.
a
(1H, t, J = 7.8 Hz), 7.41 (3H, m), 7.53 (2H, m). IR (CHCl3)
m 3007,
2932, 2857, 1748, 1624, 1493, 1450, 1244, 1194, 1171 cmꢀ1
.
FAB-MS m/z: 580 (M+H+). HRMS m/z: 580.2691 (Calcd for C31H38F4-
NO5: 580.2686).
The magnitude of LISOMe versus the molar ratio of Eu(fod)3 for
(S,R)-6 was 0.47 ppm.
4.3.5.2.
hexahydro-isoindolin-2-yl)-4-oxobutyl (R)-
fluoromethyl)phenylacetate (R,R-6). Two rotamers were
observed (ca. 1/1). [
] +23.9° (c 0.59, CHCl3). 1H NMR (CDCl3) d
(2R)-2-(2-Fluoroethyloxybenzyl)-4-(cis-3a,4,5,6,7,7a-
a
-methoxy- -(tri-
a
a
1.35–1.63 (8H, m), 2.16 (4H, m), 2.73 (3H, br), 3.07 (1H, m), 3.20
(1H, m), 3.31–3.43 (2H, m), 3.55 and 3.56 (each 1.5H, s), 4.20
(2H, br d, J = 27.9 Hz), 4.35 (2H, m), 4.74 (2H, dt, J = 3.9, 47.4 Hz),
6.81 (1H, d, J = 8.0 Hz), 6.88 (1H, t, J = 7.3 Hz), 7.04 (1H, d,
J = 7.3 Hz), 7.18 (1H, t, J = 7.8 Hz), 7.41 (3H, m), 7.53 (2H, m). IR
4.5. Radiochemistry
4.5.1. (2S)-2-(2-18Fluoroethyloxybenzyl)-4-(cis-3a,4,5,6,7,7a-
hexahydro-isoindolin- 2-yl)-4-oxobutanoic acid ((+)-(S)-o-
[
18F]FMIT) and (2R)-2-(2-18Fluoroethyloxybenzyl)-4-(cis-
(CHCl3)
m
3007, 2932, 2857, 1748, 1624, 1493, 1450, 1246, 1194,
3a,4,5,6,7,7a-hexahydro-isoindolin-2-yl)-4-oxobutanoic acid ((–
1171 cmꢀ1. FAB-MS m/z: 580 (M+H+). HRMS m/z: 580.2681 (Calcd
for C31H38F4NO5: 580.2686).
)-(R)-o-[18F]FMIT)
[
18F]Fluoride trapped on an anion-exchange cartridge was
The magnitude of LISOMe versus the molar ratio of Eu(fod)3 for
(R,R)-6 was 0.90 ppm.
eluted with an aqueous solution of potassium carbonate
(33 mM). Kryptofix222 (22.1 mg) was dissolved in the solution of
[
18F]fluoride (9.99 GBq) in water. The solvent was removed at
4.3.5.3. NMR studies for determination of absolute configura-
120 °C under a stream of nitrogen gas. The residue was dried with
anhydrous acetonitrile (0.5 mL) thrice at 120 °C under nitrogen. For
(+)-(S)-o-[18F]FMIT, a solution of the precursor 7 (2.83 mg) in ace-
tonitrile (0.15 mL) was added to the reaction vessel containing the
18F (4.78 MBq). The mixture was heated at 140 °C for 10 min. After
the solvent was removed at room temperature under a stream of
nitrogen gas, the residue was dissolved in 1,4-dioxane (0.05 mL),
then an aqueous solution of sodium hydroxide (2 M, 0.05 mL)
was added and the mixture stored for 1 h at room temperature.
After this, acetonitrile (0.2 mL) was added and the mixture purified
using HPLC with a reverse-phase column (Cosmosil 5C18-AR-II,
tion.
A solution of MTPA esters 6 in CDCl3 were taken with a
molar ratio of Eu(fod)3 to MTPA ester of 0.1–0.3 and their NMR
spectra recorded. The magnitudes of the induced chemical shift
of OMe signals were plotted against molar ratio (Eu(fod)3/MTPA
ester). In this range, the induced shifts were essentially linear with
respect to the molar ratio of the reagent. The lanthanide-induced
shift of
6 derived from (+)-o-FMIT versus molar ratio was
0.47 ppm, and that of 6 derived from (ꢀ)-o-FMIT was 0.90 ppm.
Therefore, (+)-o-FMIT was determined to have an S-configuration,
and (ꢀ)-o-FMIT was determined to have a R-configuration.
Please cite this article in press as: Kimura, H.; et al. Bioorg. Med. Chem. (2014), http://dx.doi.org/10.1016/j.bmc.2014.04.059

H. Kimura et al. / Bioorg. Med. Chem. xxx (2014) xxx–xxx
9
10 ꢁ 250 mm, acetonitrile/water/TFA = 70/30/0.1) and a chiral col-
Acknowledgments
umn (CHIRALPAK AY-H, 10 ꢁ 250 mm, acetonitrile/methanol/
TFA = 95/5/0.1) at a flow rate of 1.0 mL/min. Fractions containing
the product were evaporated. (+)-(S)- and (–)-(R)-o-[18F]FMIT were
obtained with a radiochemical yield of 18% and 13% (overall decay
corrected). The radiochemical purity of these radiotracers was
>98% after purification by high-performance liquid chromatogra-
phy (HPLC) and their specific activity estimated to be ꢂ100 GBq/
This work was supported in part by the Advanced Research for
Medical Products Mining Programme of the National Institute of
Biomedical Innovation (NIBIO),
a Grant-in-Aid for Scientific
Research (A) from the Japan Society for the Promotion of Science,
the Takeda Science Foundation, and a Research Grant on Nanotech-
nical Medicine from the Ministry of Health, Labour, and Welfare of
Japan.
lmol.
4.6. Biodistribution studies
References and notes
Biodistribution studies of (+)-(S)- and (–)-(R)-o-[18F]FMIT were
conducted using six-week-old male ddY mice. o-[18F]FMITs
(0.15–0.22 MBq/100 lL) were administered to mice through tail-
vein injection. At 5, 15, 30, 60, and 120 min after administration,
mice were sacrificed by exsanguination. Radioactivity in blood
and organs was measured using an auto well c-counter. Radioac-
tivity levels in tissue were expressed as a percentage of the
injected dose per gram of tissue (% ID/g).
1. Leonor, G.; David, W.; Clara, W.; Niegel, U. Diabetes Res. Clin. Pract. 2011, 94,
322.
2. American Diabetes Association Diabetes Care 2012, 35, S64.
3. Butler, A. E.; Janson, J.; Bonner-Weir, S.; Ritzel, R.; Rizza, R. A.; Butler, P. C.
Diabetes 2003, 52, 102.
4. Sakuraba, H.; Mizukami, H.; Yagihashi, N.; Hanyu, C.; Yagihashi, S. Diabetologia
2002, 45, 85.
5. Yoon, K. H.; Ko, S. H.; Cho, J. H.; Lee, J. M.; Ahn, Y. B.; Song, K. H.; Yoo, S. J.; Kang,
M. L.; Cha, B. Y.; Lee, K. W.; Son, H. Y.; Kang, S. K.; Kim, H. S.; Lee, I. K.; Bonner-
Weir, S. J. Clin. Endocrinol. Metab. 2003, 88, 2300.
6. Farilla, L.; Bulotta, A.; Hirshberg, B.; Calzi, S. L.; Khoury, N.; Noushmehr, H.;
Bertolotto, C.; Mario, U. D.; Harlan, D. M.; Perfetti, R. Endocrinology 2003, 144,
5149.
7. Xu, G.; Stoffers, D. A.; Habener, J. F.; Bonner-Weir, S. Diabetes 1999, 48, 2270.
8. Toyoda, K.; Okitsu, T.; Yamane, S.; Uonaga, T.; Liu, X.; Harada, N.; Uemoto, S.;
Seino, Y.; Inagaki, N. Biochem. Biophys. Res. Commun. 2008, 367, 793.
9. Seino, S. Annu. Rev. Physiol. 1999, 61, 337.
10. Mikhailov, M. V.; Campbell, J. D.; de Wet, H.; Shimomura, K.; Zadek, B.; Collins,
R. F.; Sansom, M. S.; Ford, R. C.; Ashcroft, F. M. EMBO J. 2005, 24, 4166.
11. Inagaki, N.; Gonoi, T.; Clement, J. P., IV; Namba, N.; Inazawa, J.; Gonzalez, G.;
Aguilar-Bryan, L.; Seino, S.; Bryan, J. Science 1995, 270, 1166.
12. Gabrielsson, B. G.; Karlsson, A. C.; Lönn, M.; Olofsson, L. E.; Johansson, J. M.;
Torgerson, J. S.; Sjöström, L.; Carlsson, B.; Edén, S.; Carlsson, L. M. S. Mol. Cell.
Biochem. 2004, 258, 65.
13. Shiue, G. G.; Schirrmacher, R.; Shiue, C.-Y.; Alavi, A. A. J. Labelled Compd.
Radiopharm. 2001, 44, 127.
14. Schmitz, A.; Shiue, C.-Y.; Feng, Q.; Shiue, G. G.; Deng, S.; Pourdehnad, M. T.;
Schirrmacher, R.; Vatamaniuk, M.; Doliba, N.; Matschinsky, F.; Wolf, B.; Rösch,
F.; Naji, A.; Alavi, A. A. Nucl. Med. Biol. 2004, 31, 483.
15. Schneider, S.; Feilen, P. J.; Schreckenberger, M.; Schwanstecher, M.;
Schwanstecher, C.; Buchholz, H. G.; Thews, O.; Oberholzer, K.; Korobeynikov,
A.; Bauman, A.; Comagic, S.; Piel, M.; Schirrmacher, E.; Shiue, C. Y.; Alavi, A. A.;
Bartenstein, P.; Rösch, F.; Weber, M. M.; Klein, H. H.; Schirrmacher, R. Exp. Clin.
Endocrinol. Diabetes 2005, 113, 388.
16. Wängler, B.; Schneider, S.; Thews, O.; Schirrmacher, E.; Comagic, S.; Feilen, P.;
Schwanstecher, C.; Schwanstecher, M.; Shiue, C.-Y.; Alavi, A.; Höhnemann, S.;
Piela, M.; Rösch, F.; Schirrmacher, R. Nucl. Med. Biol. 2004, 31, 639.
17. Wängler, B.; Beck, C.; Shiue, C. Y.; Schneider, S.; Schwanstecher, C.;
Schwanstecher, M.; Feilen, P. J.; Alavi, A.; Rösch, F.; Schirrmacher, R. Bioorg.
Med. Chem. Lett. 2004, 14, 5205.
18. Schirrmacher, R.; Weber, M.; Schmitz, A.; Shiue, C. Y.; Alavi, A. A.; Feilen, P.;
Schneider, S.; Kann, P.; Rösch, F. J. Labelled Compd. Radiopharm. 2002, 45, 763.
19. Reimann, F.; Proks, P.; Ashcroft, F. M. Br. J. Pharmacol. 2001, 132, 1542.
20. Yamaguchi, T.; Yanagi, T.; Hokari, H.; Mukaiyama, Y.; Kamijo, T.; Yamamoto, I.
Chem. Pharm. Bull. 1997, 45, 1518.
21. Yamaguchi, S.; Yasuhara, F.; Kabuto, K. Tetrahedron 1976, 32, 1363.
22. Yasuhara, F.; Yamaguchi, S. Tetrahedron Lett. 1977, 47, 4085.
23. Sugimoto, Y.; Tsuyuki, T.; Moriyama, Y.; Takahashi, T. Bull. Chem. Soc. Jpn. 1980,
53, 3723.
4.7. Ex vivo autoradiography experiments
(+)-(S)-o-[18F]FMIT was administered to non-anesthetized ddY
mice (male, weight: 20 g, n = 3) by intravenous injection
(185 MBq/100 lL). Mice were sacrificed by exsanguination
30 min after administration. Pancreases were removed, rinsed with
saline, and frozen immediately. Frozen pancreases were cut into
several sections (thickness, 10 lm), placed on a glass slide, and
covered with a glass cover. The radioactivity of a pancreatic section
on a glass slide was determined using an imaging analyzer
(BAS5000, Fujifilm Photo Film, Tokyo, Japan) (18-h exposure).
4.8. Immunohistochemistry
Serial sections of pancreas prepared for autoradiography were
fixed with 4% paraformaldehyde in 0.1 M phosphate buffer for
1 h, then permeabilized and blocked with 0.5% Triton X-100 in
PBS containing 1.5% goat serum for 1 h at room temperature. Spec-
imens were incubated with rabbit anti-insulin (1:50 dilution;
Santa Cruz Biotechnology, Santa Cruz, CA, USA) as a primary anti-
body overnight at 4 °C. After rinsing with PBS, specimens were
incubated with Alexa Fluor 488-labeled goat anti-rabbit IgG
(1:200; Molecular Probes, Eugene, OR, USA) as a secondary anti-
body for 1 h at room temperature. Fluorescence signals were
acquired through a fluorescence microscope (BZ-9000, KEYENCE,
Osaka, Japan).
4.9. Statistical analyses
24. Harada, H.; Yamaguchi, T.; Iyobe, A.; Tsubaki, A.; Kamijo, T.; Iizuka, K.; Ogura,
K.; Kiso, Y. J. Org. Chem. 1990, 55, 1679.
25. Dale, J. A.; Dull, D. L.; Mosher, H. S. J. Org. Chem. 1969, 34, 2543.
26. Dale, J. A.; Mosher, H. S. J. Am. Chem. Soc. 1973, 85, 512.
27. Mukai, E.; Ishida, H.; Kato, S.; Tsuura, Y.; Fujimoto, S.; Ishida-Takahashi, A.;
Horie, M.; Tsuda, K.; Seino, Y. Am. J. Physiol. Endocrinol. Metabol. 1998, 274, E38.
Data are the mean SD. The statistical significance of differ-
ences was evaluated using the Tukey–Kramer test. P <0.05 was
considered significant.
Please cite this article in press as: Kimura, H.; et al. Bioorg. Med. Chem. (2014), http://dx.doi.org/10.1016/j.bmc.2014.04.059