Radiosynthesis and preliminary biological evaluation of a new 18F-labeled triethylene glycol derivative of triphenylphosphonium

Article abstract of DOI:10.1002/jlcr.3379

Delocalized lipophilic cations such as [¹⁸F]fluorobenzyltriphenylphosphonium ([¹⁸F]FBnTP) can accumulate in mitochondria and have been used in myocardial perfusion imaging (MPI). In this study, we established a simplified method for

Full text of DOI:10.1002/jlcr.3379

Research article
Received 22 August 2015,
Revised 02 January 2016,
Accepted 14 January 2016
Published online 10 February 2016 in Wiley Online Library
(wileyonlinelibrary.com) DOI: 10.1002/jlcr.3379
Radiosynthesis and preliminary biological
1
8
evaluation of a new F-labeled triethylene
†
glycol derivative of triphenylphosphonium
a,b
a,b
a
a
Takahiro Tominaga, Hiroaki Ito, Yoichi Ishikawa, Ren Iwata,
c,d
a,b
Kiichi Ishiwata, and Shozo Furumoto *
1
8
18
Delocalized lipophilic cations such as [ F]fluorobenzyltriphenylphosphonium ([ F]FBnTP) can accumulate in mitochondria
1
8
and have been used in myocardial perfusion imaging (MPI). In this study, we established a simplified method for [ F]FBnTP
synthesis using triphenylphosphine hydrobromide (PPh •HBr) without preparing an intermediate that contains benzyl
3
1
8
bromide structure. Applying this new method, we synthesized and evaluated a novel F-labeled PEGylated BnTP derivative
[ F]FPEGBnTP). In vitro cellular uptake study demonstrated that [ F]FPEGBnTP accumulated in cells in proportion to the
1
8
18
(
relative intensity of mitochondrial membrane potential. Biodistribution study revealed that the heart : liver uptake ratio of
1
8
18
18
[
F]FPEGBnTP (4.00 at 60min) was superior to that of [ F]FBnTP (1.50 at 60min). However, [ F]FPEGBnTP showed slow blood
18
clearance and high radioactivity uptake in bone at 120-min post-injection. These results imply the possibility of [ F]FPEGBnTP
being used as a MPI agent. However, there is a need of further structural optimization and flow-dependent uptake study.
Keywords: mitochondria; FBnTP; DLCs; positron emission tomography; PET
Introduction
9–11
MPI,
but its performance can be improved. This tracer
accumulates in the liver to the same extent as that seen in the heart
and exhibits slow clearance from the liver, which can adversely
affect MPI. To optimize pharmacokinetics, several research teams
According to a report from the World Health Organization, the
leading cause of death worldwide is ischemic heart disease,
and 7.4 million people died in 2012. Early and accurate
1
12,13
have developed new phosphonium derivatives.
detection of cardiac ischemia enables appropriate therapy to
be instigated and reduces the risk of disease progression.
Myocardial perfusion imaging (MPI) is a tool for non-invasive
evaluation of cardiac ischemia. Tracers based on technetium-99m
This study aims to develop a new BnTP derivative showing lower
liver uptake and higher heart-to-liver ratio in comparison with
1
8
18
[
F]FBnTP. We hypothesized that the lipophilicity of [ F]
99m
99m
99m
FBnTP would affect its accumulation in the liver and a lower
lipophilic derivative would indicate faster clearance from
the liver. To reduce the lipophilicity, we designed and synthesized
(
Tc), such as [ Tc]sestamibi and [ Tc]tetrofosmin, are used
as MPI agents for single-photon emission computed tomography
2
(
SPECT). With regard to the mechanism of accumulation of these
18
18
9
9m
a
F-labeled triethylene glycol derivative of BnTP ([ F]FPEGBnTP).
In the synthesis, we established a novel synthetic route using
triphenylphosphine hydrobromide (PPh • HBr) without preparing
Tc-labeled tracers, the association of mitochondrial membrane
potentials (MMPs) in myocardial cells has been considered.
[2,3]
It
has been suggested that these tracers remain in the heart for a
long time and that tracers based on thallium-201 are not
3
an intermediate that contains benzyl bromide structure. Then,
3
redistributed. However, these tracers also accumulate in organs
3,4
a
adjacent to the heart and interfere with diagnostic imaging.
Cyclotron and Radioisotope Center (CYRIC), Tohoku University, Sendai 980-8578,
Compared with SPECT, positron emission tomography (PET) has Japan
technical advantages: special resolution, attenuation correction,
b
Graduate School of Pharmaceutical Sciences, Tohoku University, Sendai 980-8578,
Japan
5
13
sensitivity, and quantitation. PET tracers such as [ N]NH and
3
15
6,7
[
O]H O are used for MPI, but application of these tracers is
2
c
Institute of Cyclotron and Drug Discovery Research, Southern Tohoku Research
limited to institutions with on-site cyclotrons because of their
1
/2
Institute for Neuroscience, Koriyama 963-8563, Japan
short half-lives (T = 9.97 min and 122 s, respectively). Conversely,
18
PET tracers labeled with fluorine-18 ( F) have a suitable half-life
d
Department of Biofunctional Imaging, Fukushima Medical University,
1/2
(T
= 109.8 min), which enables their delivery to many PET Fukushima 960-1295, Japan
centers and has contributed to their increased use. Accordingly,
18
*Correspondence to: Shozo Furumoto, Cyclotron and Radioisotope Center
(
drug design involving F for MPI has been undertaken.
CYRIC), Tohoku University, 6-3 Aoba, Aramaki, Aoba-ku, Sendai 980-8578,
Japan.
reported first in 2004. This tracer accumulates in the myocardium E-mail: furumoto@cyric.tohoku.ac.jp
18
18
Use of [ F]fluorobenzyltriphenylphosphonium ([ F]FBnTP) was
8
9
9m
99m
†
to the same extent as [ Tc]sestamibi and [ Tc]tetrofosmin.
[
Additional supporting information may be found in the online version of this
F]FBnTP has been used as a ‘gold standard’ PET tracer for article at the publisher’s web-site.
18
J. Label Compd. Radiopharm 2016, 59 117–123
Copyright © 2016 John Wiley & Sons, Ltd.

T. Tominaga et al.
we preliminarily evaluated in vitro cellular uptake and in vivo removed by azeotropic means with acetonitrile by heating at 110 °C
18
biodistribution of [ F]FPEGBnTP to elucidate the possibility for an with helium gas flow. After drying, precursor 1 (15 mg) in dimethyl
MPI agent.
sulfoxide (DMSO) (0.6 mL) was added and heated at 150 °C for
0 min, followed by the addition of water for quenching. The
1
Materials and methods
product was extracted by solid-phase extraction (SPE) with a
Sep-Pak tC18 Plus Long Cartridge (Waters, Milford, MA, USA). Then,
an aqueous solution of NaBH (20 mg/mL) was added to the
4
Synthesis of chemical compounds
cartridge to reduce the aldehyde to an alcohol. The cartridge was
washed with water and the water drained off. Then, the product
was eluted with o-xylene (4.1 mL). The eluent was dried by passing
through an EXtrelut® NT1 column (Merck, Whitehouse Station, NJ,
Methods for the synthesis and characterization data of chemical
compounds (shown in Schemes 1–4) are noted in detail in
Supporting Information.
18
USA). The anhydrous [ F]BnOH solution eluted from the EXtrelut
NT1 column was added to a brown vial containing PPh •HBr
Measurement of octanol/water partition coefficient
3
As an index for lipophilicity, octanol/water partition coefficient (126 mg). The reaction mixture was heated at 150 °C for 10 min,
LogP) values of FBnTP and FPEGBnTP were determined by followed by addition of CHCl (6 mL). The product was extracted
(
3
conventional shake flask method using n-octanol saturated with by SPE with a Sep-Pak tC18 Plus Short Cartridge (Waters). The
water and water saturated with n-octanol. Briefly, an octanol cartridge was washed with CHCl3 (5 mL) and Et O (5mL). The
2
18
solution of FBnTP or FPEGBnTP (1mL, 1 mg/mL) was mixed with product was eluted with ethanol. The F-labeled product was
octanol (2mL) and water (3mL) in a conical tube. After shaking purified from the eluent by semi-preparative RP-HPLC (column:
for 30 min, the mixture was centrifuged to separate the octanol– Inertsil® ODS-4 (10 × 250 mm); mobile phase: acetonitrile/1× PBS
18
water phases (500 rpm, 20 min). Relative amount of test compound (52/48); flow rate: 5.0 mL/min; UV: 254 nm). The F-labeled product
in each phase was determined by analytical reverse-phase high- was isolated from the collected fraction by SPE with a Sep-Pak tC18
performance liquid chromatography (RP-HPLC; column: Inertsil® Plus Short Cartridge (Waters) and finally dissolved in saline for
ODS-4 (4.6× 150 mm); mobile phase: acetonitrile/1× phosphate- biologic evaluation. At HPLC, the radiochemical yield was 12–14%
buffered saline (PBS; 50/50); flow rate: 1.5 mL/min; UV: 254 nm).
(n= 5). Radiochemical purity of the drug solution was >99%
Figure 3a).
(
Radiochemistry
1
8
Radiosynthesis of [ F]FPEGBnTP
1
8
18
18
No carrier-added F was produced by the O(p, n) F reaction
1
8
18
À
on enriched [ O]H
2
2 3
O (Taiyo Nippon Sanso, Tokyo, Japan) with The aqueous [ F]F contained in K CO solution (3.7–5.6 GBq)
an HM-12 cyclotron (Sumitomo Heavy Industries, Tokyo, Japan) and Kryptofix 2.2.2 (16 mg) were placed in a brown vial. Then,
installed in the Cyclotron and Radioisotope Center of Tohoku water was removed by azeotropic means with acetonitrile by
18
University (Sendai, Japan). The specific activity of F was heating at 110 °C and helium gas flow. After drying, precursor
4–740 GBq/mmol at the end of bombardment. Radiochemical 10 (3 mg) in DMSO (0.6 mL) was added and heated at 110 °C
purity was determined by radioanalytical HPLC under the same for 10 min. Then, 0.5 M HCl was added to the solution and heated
7
HPLC conditions used for LogP measurement.
for an additional 3 min. After neutralization with 1 M AcOK, the
product was extracted by SPE with a Sep-Pak tC18 Plus Short
Cartridge (Waters). Then, the product was eluted with o-xylene
18
Radiosynthesis of [ F]FBnTP
18
À
The aqueous [ F]F contained in K CO solution (3.7–5.6 GBq) and (4.1 mL). The eluent was dried by passing through an EXtrelut
2
3
Kryptofix 2.2.2 (16 mg) were placed in a brown vial. Then, water was NT1 column (Merck). The anhydrous solution eluted from the
1
8
18
Scheme 1. Synthesis of [ F]FBnTP. Reaction conditions: (a) [ F]KF, K.2.2.2, DMSO, 150 °C, 10 min; (b) NaBH
o-xylene, reflux, 10 min.
4 3
aq on a Sep-Pak tC18 Plus Short Cartridge; and (c) PPh • HBr,
2 3
Scheme 2. Synthesis of FPEGBnTP. Reaction conditions: (a) 2-[2-(2-Chloro-ethoxy)-ethoxy]-ethanol, K CO , DMF, 100 °C, 15 h; (b) p-TsCl, NMM, DCM, r.t., 2 h; (c) TBAF, THF,
reflux, 2 h; (d) NaBH , MeOH, r.t., 2 h; and (e) PPh ·HBr, MeCN, reflux, 11 h.
4 3
www.jlcr.org
Copyright © 2016 John Wiley & Sons, Ltd.
J. Label Compd. Radiopharm 2016, 59 117–123

T. Tominaga et al.
4
Scheme 3. Synthesis of precursor of FPEGBnTP. Reaction conditions: (a) TBS-Cl, DMAP, DMF, 100 °C, 15 h; (b) NaBH , MeOH, 0 °C, 30 min; (c) 2,3-DHP, PPTs, DCM, r.t., 1 h;
(
d) TBAF, THF, r.t., 15 min; and (e) p-TsCl, NMM, DCM, r.t., 11 h.
1
8
18
3
Scheme 4. Synthesis of [ F]FPEGBnTP. Reaction conditions: (a) [ F]KF, K.2.2.2, DMSO, 110 °C, 10 min; (b) HClaq, 110 °C, 3 min; and (c) PPh ·HBr, o-xylene, reflux, 10 min.
3
column was added to a brown vial containing PPh •HBr (126 mg). Assay kit (Microplate; Abcam, Cambridge, UK) and four human
The reaction mixture was heated at 150 °C for 10 min, followed cancer cell lines (TFK1, MCF7, MB231, and MB453). Cells in 90 μL
5
by addition of CHCl
a Sep-Pak tC18 Plus Short Cartridge (Waters). The cartridge was cells/well, n= 8). The medium not containing cells was added as a
washed with CHCl (5 mL) and Et O (5 mL). The product was eluted control. Then, 10 μL of PBS was added to each well and incubated
with ethanol. The F-labeled product was purified from the eluent at 37 °C for 30 min. Fifty microliters of JC-10 solution diluted with
by semi-preparative RP-HPLC (column: Inertsil® ODS-4 (10 × Assay Buffer A was added to each well and incubated at 37 °C for
50 mm); mobile phase: acetonitrile/1× PBS (52/48); flow rate: 30 min. Then, 50 μL of Assay Buffer B was added to each well and
3
(6mL). The product was extracted by SPE with of RPMI-1640 medium were added to a 96-well plate (5.0 × 10
3
2
1
8
2
5
18
.0 mL/min; UV: 254 nm). The F-labeled product was isolated fluorescence intensity read with a microplate reader. Excitation
from the collected fraction by SPE with a Sep-Pak tC18 Plus Short was at 495 nm; emission was read at 520 and 590 nm. The ratio
Cartridge and finally dissolved in saline for biologic evaluation. At of Em590 to Em520 was calculated. The percentage of the control
HPLC, the radiochemical yield was 13–23% (n=8). Radiochemical value was used as an index of the relative intensity of MMP.
purity of the drug solution was >95% (Figure 3b).
Biodistribution assay in normal mice
Cellular uptake
Study protocols using animals were approved by the
Four types of cells were used: human cholangiocarcinoma cells
TFK1) and human breast cancer cells (MCF7, MB231, MB453).
Institutional Animal Care and Use Committee of the Tohoku
University Environmental & Safety Committee. Male ICR mice
(
18
These cells were cultured with RPMI-1640 medium and used
(
25–35 g) were injected in a lateral tail vein with F-labeled
6
for uptake studies (3.3 × 10 cells/mL of medium).
tracer (370 ± 30 kBq) in saline (0.2 mL). Mice were killed by
cervical dislocation after heart puncture to obtain blood samples
at designated time points post-injection (n = 4 at 60 and
A cell line was suspended in 150μL of serum-free RPMI-1640
medium (TFK1, MCF7, MB231, MB453) and added to a 96-well
5
18
plate (5.0 × 10 cells/well, n= 8). [ F]FPEGBnTP in PBS (50 μL;
.85 MBq/mL at the start of experiments; radiochemical purity
99%) was added to each well and incubated at 37 °C for
0 min. Then, cells were harvested through suction filtration with
120 min). Tissues of interest were excised and weighed, and
1
>
3
the radioactivity was counted in an automatic gamma counter.
Data relating to radioactivity uptake are expressed as percent
injected dose per gram of tissue (%ID/g).
a vacuum manifold (Millipore, Bedford, MA, USA). Trapped cells
were washed three times with PBS (200 mL/well). Filters holding
the trapped cells were separated, and radioactivity was measured
with a γ-counter (AccuFLEX g7000; Hitachi Aloka Medical, Tokyo,
Japan). Then, percentage of the radioactivity of the cell to total
radioactivity of a solution in a blank well (containing media but
no cell) was calculated as an index of cellular uptake.
Results
1
8
Radiosynthesis of [ F]FBnTP
18
Total synthesis of [ F]FBnTP was planned as shown in Scheme 1.
First, the solvent condition of the phosphonium formation step
was examined in the non-radioactive condition (Figure 1). The
reaction was carried out so that FBnOH (9.16 μmol) and excess
Assessment of relative intensity of mitochondrial membrane
potential
PPh •HBr (400 μmol) in various solvents were heated at reflux
3
for 30 min. Results revealed that o-xylene (non-polar solvent with
Correlation between the relative intensity of mitochondrial a high boiling point) gave the best yield. We estimated that
18
membrane potential (MMP) and cellular uptake of [ F]FPEGBnTP using o-xylene would also give the best radiochemical yield in
was evaluated using a JC-10 Mitochondrial Membrane Potential the radioactive condition.
J. Label Compd. Radiopharm 2016, 59 117–123
Copyright © 2016 John Wiley & Sons, Ltd.
www.jlcr.org

T. Tominaga et al.
Then, we worked on the first nucleophilic aromatic
substitution step in Scheme 1. Compound 1 was prepared by a
14
slight modification of a procedure reported previously. After
18
the reaction, the crude product, [ F]FBnCHO, was extracted by
SPE with a Sep-Pak tC18 Plus Long Cartridge (Waters).
4
Subsequently, an aqueous solution of NaBH was added to the
cartridge to reduce the aldehyde to an alcohol. The crude
18
product containing [ F]FBnOH was eluted with o-xylene. HPLC
analyses revealed that the reduction reaction proceeded
quantitatively. With respect to the two steps so far, various
reaction conditions were tested. When dimethylformamide was
18
used as a solvent at 150 °C, it gave [ F]FBnOH with the best
radiochemical yield.
At the next phosphonium formation step using excess
1
8
Figure 2. Synthesis of
[
F]FBnTP. Reaction conditions of phosphonium
formation. R.Y., radiochemical yield (decay-corrected) measured by HPLC.
PPh • HBr, the reaction did not proceed at all. A possible cause
3
was the water contained in the reaction mixture. Hence, after Synthesis of FPEGBnTP and its precursor
drying the mixture with magnesium sulfate, the reaction was
1
8
The reference compound FPEGBnTP was prepared in the five steps
shown in Scheme 2. The precursor was also synthesized according
to Scheme 3 by five steps from 2. These compounds were
characterized by infrared, H-NMR, C-NMR, and high-resolution
mass spectrometry or fast atom bombardment mass spectrometry.
tried again and [ F]FBnTP was obtained. To simplify this
dehydration step, various commercially available column
cartridges were tested. Among them, EXtrelut NT1 (Merck)
could be used to dehydrate the reaction mixture to give
1
13
18
[
F]FBnTP.
Finally, we examined the amount of PPh •HBr and the
3
Measurement of octanol/water partition coefficient
reaction temperature under radioactive conditions. At first, a
small amount of PPh •HBr was applied to the reaction, but
3
To compare the lipophilicity between FBnTP and FPEGBnTP,
octanol/water partition coefficient (LogP) of them were measured.
LogP values of FBnTP and FPEGBnTP were À0.38 and À0.92,
respectively, indicating that FPEGBnTP is relatively less lipophilic
than FBnTP.
18
[
F]FBnTP was not detected (entries 1–3, Figure 2). Hence,
the amount of PPh •HBr and reaction temperature were
3
18
increased (entries 4–7, Figure 2). [ F]FBnTP could be obtained
at the best yield when 126 mg of PPh •HBr was used. Under this
3
condition, a large amount of unreacted PPh
3
remained in the
reaction mixture. Fortunately, the retention value of FBnTP is
1
8
18
Radiosynthesis of [ F]FPEGBnTP
much lower than that of PPh
after PPh was first removed. Hence, the reaction mixture was
placed on a Sep-Pak tC18 Plus Short Cartridge and washed with
3
, so [ F]FBnTP could be obtained
3
The radiosynthetic route was planned and undertaken as shown in
Scheme 4. Briefly, precursor 10 was reacted with the [ F]F
activated with Kryptofix 2.2.2. An aqueous solution of HCl was
added to deprotect the hydroxyl group. At the next phosphonium
formation step, our method using PPh •HBr was applied. After
purification with RP-HPLC, analytical HPLC was undertaken and
showed a single product peak. Radiochemical yield was 13–23%
18
À
1
8
CHCl
3 2
and Et O followed by extraction of [ F]FBnTP with
1
8
MeOH. [ F]FBnTP was purified further with semi-preparative
RP-HPLC using an acetonitrile/1× PBS phase. After the
appropriate fraction was collected, analytical HPLC was carried
out and showed a single product peak, which suggested that
F]FBnTP was obtained at high purity. Radiochemical yield
was 12–14% (n = 5, decay-corrected). Radiochemical purity was
99%.
3
18
[
(
n = 8, decay-corrected). Radiochemical purity was >95%.
>
Cellular uptake
The relative intensity of MMP was assayed with a JC-10 that is a
kind of fluorescent delocalized lipophilic cation (DLC) and
accumulates in mitochondria. JC-10 has usually maximal
fluorescence at 520 nm with an excitation wavelength at 495 nm.
However, when the concentration of JC-10 become high in
mitochondria, an aggregate of JC-10 is formed and the emission
peak shifts to 590 nm. The relative intensity of MMP can be
evaluated by measuring the ratio of fluorescence intensity between
at 520 and 590 nm. The relative intensity of MMP was expressed as
percent Em590/Em520 of the control and is summarized in Table 1.
18
Results of percent applied activity of [ F]FPEGBnTP uptakes were
summarized in Table 2. These data are plotted on a scatter diagram
and showed a linear correlation between the relative intensity of
18
MMP and cellular uptake of [ F]FPEGBnTP (Figure 3).
Biodistribution study in normal mice
Figure 1. Solvent condition of phosphonium formation. Superscript lowercase a
indicates that yield was measured by HPLC; superscript lowercase b indicates that
yield was not determined because of the product peak overlapping the peak for
by-products.
18
Properties of [ F]FPEGBnTP were evaluated by a biodistribution
study in normal mice (n = 4 at each time point). Radioactivity
www.jlcr.org
Copyright © 2016 John Wiley & Sons, Ltd.
J. Label Compd. Radiopharm 2016, 59 117–123

T. Tominaga et al.
structures allow the charge to be ‘delocalized’ over a large
molecular area by resonance stabilization. The lipophilicity of
DLCs enables them to cross the lipid bilayers of cells and to
Table 1. Cellular uptake. JC-10 assay for mitochondrial
membrane potential
Cellular
% Ratio (Em590/Em520) of control
15
access the internal environments of mitochondria. At the inner
membranes of mitochondria, a pH gradient is formed by
electron transport systems, so a strongly negative membrane
potential develops. After passing through the lipid bilayers of
cells, DLCs accumulate preferentially in the inner membranes
of mitochondria driven by the MMP.
TFK1
154 ± 2
150 ± 1
130 ± 5
124 ± 2
100 ± 2
MB231
MB453
MCF7
Control
18
[
F]FBnTP is phosphonium salt DLC that shows promising
18
11–13
characteristics as a F-labeled MPI tracer.
Additionally, this
Values of percent ratio (Em590/Em520) of the control are mean
SD (n = 8).
tracer has been used for studies on imaging of brown adipose
±
1
6–18
18
tissue and apoptosis.
In a synthetic route for [ F]FBnTP
10
reported in 2004, a corrosive reagent, PPh Br , was used in
the step for [ F]FBnBr synthesis. This reagent is not suitable
for automatic synthetic apparatus or conventional use.
3
2
18
18
Table 2. Cellular uptake of [ F]FPEGBnTP
We proposed a new synthetic route using PPh •HBr, which is
3
Cellular
% Applied activity
18
not corrosive (Scheme 1). Our route could lead to [ F]FBnTP
18
from [ F]FBnOH in one step. However, from the result of the
TFK1
1.96 ± 0.16
1.76 ± 0.10
0.78 ± 0.17
0.09 ± 0.07
phosphonium formation step, excess PPh •HBr turned out to
MB231
MB453
MCF7
3
be necessary (Figure 2). We suspected that if a small amount of
18
PPh •HBr was used, the hydroxyl group of [ F]FBnOH would
3
not be activated adequately, and the reaction would not
proceed completely. After this reaction, we were able to isolate
Values of percent applied activity are mean ± SD (n = 8).
1
8
[
F]FBnTP from excess PPh3 on a Sep-Pak tC18 Plus Short
18
Cartridge. At the HPLC separation step, [ F]FBnTP could be
obtained within a reasonable range and not be affected by
uptake in the blood, heart, and other organs are summarized in PPh . Radiochemical yield was 12–14% (n = 5, decay-corrected),
3
Table 3. With regard to heart uptake, each time point showed in accordance with that obtained by the conventional method.
adequate uptake (3.33%ID/g at 60 min). By contrast, the Our new method is thought to be practical and environmentally
neighboring organs such as liver and lung exhibited lower friendly. Nevertheless, a complicated dehydration procedure is
uptakes (0.83 and 0.81%ID/g at 60 min, respectively), resulting necessary after the reduction step. There is room for further
in higher ratios of heart : liver and heart : lung. However, improvement that does not require dehydration, such as using
radioactivity uptakes in blood were relatively higher than those a NaBH -containing alumina column.
4
of liver and lung. Kidney showed the highest radioactivity uptake
at 60 min post-injection. The bone uptake increased with [ F]FBnTP: This tracer accumulated in the liver. The liver is
There was a problem regarding the pharmacokinetics of
18
elapsed time from 60 to 120 min post-injection.
adjacent to the heart, which could adversely affect MPI. To
18
improve the pharmacokinetics of [ F]FBnTP, we designed
FPEGBnTP, which increased hydrophilicity by introduction of
Discussion
18
a polyethylene glycol (PEG) chain. In the synthesis of [ F]
•HBr
1
8
Accumulation of [ F]FBnTP in the heart is because of abundant FPEGBnTP, our phosphonium formation method with PPh
3
18
mitochondria in myocardial cells. [ F]FBnTP is a DLC. DLCs are could be applied and produce a good yield (13–23%, n = 8,
small molecules that carry a positive charge. Their chemical decay-corrected).
18
Table 3. Biodistribution of [ F]FPEGBnTP at 60 and 120 min after intravenous injection
%
ID/g
Heart : tissue ratio
60 min
120 min
60 min
120 min
Heart
Blood
Liver
Lung
Spleen
Kidney
Small intestine
Muscle
Bone
3.3 ± 0.22
1.08 ± 0.05
0.83 ± 0.02
0.81 ± 0.01
0.89 ± 0.04
3.64 ± 0.18
1.94 ± 0.10
1.28 ± 0.06
2.11 ± 0.08
0.69 ± 0.02
3.61 ± 0.26
1.00 ± 0.05
0.81 ± 0.04
0.95 ± 0.04
0.91 ± 0.06
1.08 ± 0.05
1.54 ± 0.18
1.38 ± 0.13
3.88 ± 0.29
0.73 ± 0.04
3.13 ± 0.35
4.00 ± 0.28
4.10 ± 0.37
3.74 ± 0.27
0.92 ± 0.09
1.74 ± 0.19
2.60 ± 0.20
1.58 ± 0.14
4.81 ± 0.31
3.61 ± 0.02
4.46 ± 0.02
3.80 ± 0.06
3.97 ± 0.06
3.34 ± 0.03
2.34 ± 0.08
2.62 ± 0.07
0.93 ± 0.02
4.95 ± 0.02
Brain
Values of %ID/g and heart : tissue ratio are mean ± SD for four mice.
J. Label Compd. Radiopharm 2016, 59 117–123
Copyright © 2016 John Wiley & Sons, Ltd.
www.jlcr.org

T. Tominaga et al.
1
8
18
Figure 3. Radioanalytical HPLC chromatograms. (a) [ F]FBnTP and (b) [ F]FPEGBnTP.
1
8
18
Biological features of [ F]FPEGBnTP were evaluated by in vitro pharmacokinetics of [ F]FPEGBnTP was improved. Conversely,
18
cellular uptake assay and ex vivo biodistribution assay in mice. the heart : blood ratio of [ F]FPEGBnTP was low compared
Using four cancer lines, the relationship between the relative with that of [ F]FBnTP, suggesting that the clearance of [ F]
intensity of MMP and cellular uptake of [ F]FPEGBnTP was FPEGBnTP from blood was slower as compared with [ F]FBnTP.
18
22
18
18
18
examined. Generally, MB231 cells show significant reduction in The slow blood clearance of the activity might be caused by non-
18
mitochondrial functions including electron transport chain specific binding of [ F]FPEGBnTP or its metabolites to the
19
activities, oxygen consumption, and ATP synthesis rates. These plasma constituents because the activity level showed little
features would raise an expectation that MMP of MB231 cells is change from 60 to 120 min after injection. In addition, increasing
lower than that of MCF7 cells. However, our study using a accumulation of radioactivity in bone suggests a low tolerance to
18
fluorescent dye JC-10 for MMP measurement resulted in reversed defluorination of [ F]FPEGBnTP in vivo. There is a possibility that
18
relationship between MB231 and MCF7 cells. We speculate that if these metabolism natures of [ F]FPEGBnTP have caused the
5
the total amount of mitochondria included in 5.0 × 10 cells of prolonged recirculation of activity of metabolites in the vascular
MB231 is greater than that of MCF7, it would be possible to be system and/or the relatively lower heart uptake. Hence, although
18
greater fluorescent signal of MB231 than MCF7. Similar to our the high heart : liver ratio of [ F]FPEGBnTP implies the possibility
results, Cheng et al. reported that JC-1, an analogue of JC-10, signal for using it as a MPI agent, there is a need of a detailed
2
0
ratio (% of control) of MB231 cells was greater than that of MCF7.
metabolism study, flow-dependent uptake in heart, and further
21
However, a result contrary to that was also reported. The chemical modification to show better in vivo stability leading to
importance of the cellular uptake assay is to reveal the correlation more favorable pharmacokinetics.
18
between cellular MMP and uptakes. As shown in Figure 4, [ F]
FPEGBnTP accumulated in cells in proportion to the relative
intensity of MMP, suggesting that this tracer shows an MMP-
dependent cellular uptake in vitro.
Conclusion
18
We developed a new PEGylated phosphonium derivative [ F]
FPEGBnTP, which accumulated in a particular cancer line in
proportion to the MMP. Biodistribution study demonstrated the
high uptake in the heart and good heart : liver and heart : lung
18
In vivo behavior of [ F]FPEGBnTP was investigated by a
18
biodistribution assay. The heart : liver ratio of [ F]FPEGBnTP
1
8
(
4.00 at 60 min) was superior to that of [ F]FBnTP (1.50
18
22
ratios, suggesting that [ F]FPEGBnTP would be a possible
at 60 min) reported by Madar et al., indicating that the liver
18
candidate for a new F-labeled MPI agent. However, there is still
room for optimization of the radiosynthesis procedures and the
chemical structure and is a need of studies on metabolism and
flow-dependent uptake in the heart to prove the usefulness for
using as a MPI agent.
References
[
1] The World Health Organization Website. http://www.who.int/
mediacentre/factsheets/fs310/en/ [28 July 2015].
[2] P. G. Camici, O. E. Rimoldi, J. Nucl. Med. 2009, 50, 1076.
[3] R. D. Okada, D. Glover, T. Gaffney, S. Williams, Circulation 1988, 77, 491.
[4] K. Nakajima, J. Taki, N. Shuke, H. Bunko, S. Takata, K. Hisada, J. Nucl.
Med. 1993, 34, 1478.
[
[
[
[
5] M. Dondi, G. Fagioli, M. Salgarello, S. Zoboli, C. Nanni, C. Cidda, Q. J.
Nucl. Med. Mol. Imaging 2004, 48, 181.
6] S. Nagamachi, J. Czernin, A. S. Kim, K. T. Sun, M. Böttcher, M. E.
Phelps, H. R. Schelbert, J. Nucl. Med. 1996, 37, 1626.
7] P. A. Kaufmann, T. Gnecchi-Ruscone, J. T. Yap, O. Rimoldi, P. G.
Camici, J. Nucl. Med. 1999, 40, 1848.
8] H. T. Ravert, I. Madar, R. F. Dannals, J. Label. Compd. Radiopharm.
2
004, 47, 469.
1
8
Figure 4. Correlation of cellular uptake of JC-10 and [ F]FPEGBnTP. A linear
correlation was observed (R = 0.9972).
[9] I. Madar, H. T. Ravert, Y. Du, J. Hilton, L. Volokh, R. F. Dannals, J. J.
2
Frost, J. M. Hare, J. Nucl. Med. 2006, 47, 1359.
www.jlcr.org
Copyright © 2016 John Wiley & Sons, Ltd.
J. Label Compd. Radiopharm 2016, 59 117–123

T. Tominaga et al.
[
10] I. Madar, H. Ravert, A. DiPaula, Y. Du, R. F. Dannals, L. Becker, J. Nucl.
Med. 2007, 48, 1021.
11] T. Higuchi, K. Fukushima, C. Rischpler, T. Isoda, M. S. Javadi, H. Ravert,
[19] Y. Ma, R. K. Bai, R. Trieu, L. J. Wong, Biochim. Biophys. Acta 2010, 1797, 29.
[20] L. Cheng, T. S. Xia, Y. F. Wang, W. Zhou, X. Q. Liang, J. Q. Xue, L. Shi, Y.
Wang, Q. Ding, M. Wang, Int. J. Oncol. 2014, 45, 757.
[
D. P. Holt, R. F. Dannals, I. Madar, F. M. Bengel, J. Nucl. Med. 2011, 52, [21] L. Zhao, Y. Li, M. He, Z. Song, S. Lin, Z. Yu, X. Bai, E. Wang, M. Wei,
965.
Int. J. Oncol. 2014, 45, 129.
[
12] T. M. Shoup, D. R. Elmaleh, A. Brownell, A. Zhu, J. L. Guerrero, A. J.
[22] I. Madar, H. Ravert, B. Nelkin, M. Abro, M. Pomper, R. Dannals, J. J.
Frost, Eur. J. Nucl. Med.Mol. Imaging 2007, 34, 2057.
Fischman, Mol. Imaging Biol. 2011, 13, 511.
[
[
13] D. Kim, H. Kim, K. Yu, J Min, Bioconjugate Chem 2012, 23, 431.
14] O. Åberg, F. Pisaneschi, G. Smith, Q. Nguyen, E. Stevens, E. O.
Aboagye, J. Fluorine Chem. 2012, 135, 200.
[
[
[
15] L. F. Yousif, K. M. Stewart, S. O. Kelley, ChemBioChem 2009, 10, 1939.
16] I. Madar, T. Isoda, P. Finley, J. Angle, R. Wahl, J. Nucl. Med. 2011, 52, 808.
17] I. Madar, Y. Huang, H. Ravert, S. L. Dalrymple, N. E. Davidson, J. T. Additional supporting information may be found in the online
Supporting information
Isaacs, R. F. Dannals, J. J. Frost, J. Nucl. Med. 2009, 50, 774.
18] I. Madar, E. Naor, D. Holt, H. Ravert, R. Dannals, R. Wahl, PLoS One
version of this article at the publisher’s web-site.
[
2
015, 10, e0129627.
J. Label Compd. Radiopharm 2016, 59 117–123
Copyright © 2016 John Wiley & Sons, Ltd.
www.jlcr.org