Synthesis and biological evaluation of an 18fluorine-labeled COX inhibitor - [18F]fluorooctyl fenbufen amide - For imaging of brain tumors

Article abstract of DOI:10.3390/molecules21030387

Molecular imaging of brain tumors remains a great challenge, despite the advances made in imaging technology. An anti-inflammatory compound may be a useful tool for this purpose because there is evidence of inflammatory processes in brain tumor micro-envi

Full text of DOI:10.3390/molecules21030387

molecules
Article
Synthesis and Biological Evaluation of an
18Fluorine-Labeled COX Inhibitor—[18F]Fluorooctyl
Fenbufen Amide—For Imaging of Brain Tumors
Ying-Cheng Huang 1, Yu-Chia Chang 2, Chun-Nan Yeh 3 and Chung-Shan Yu 2,4,
*
1
Department of Neurosurgery, Chang-Gung Memorial Hospital at Chiayi and Chang Gung University,
Taoyuan 33302, Taiwan; ns3068@gmail.com
Department of Biomedical Engineering and Environmental Sciences, National Tsinghua University,
Hsinchu 300, Taiwan; flyinken0221@hotmail.com
Department of Surgery, Chang-Gung Memorial Hospital at Linkou and Chang Gung University,
Guei-shan 33305,Taiwan; yehchunnan@gmail.com
Institute of Nuclear Engineering and Science, National Tsing-Hua University, Hsinchu 300, Taiwan
Correspondence: csyu@mx.nthu.edu.tw; Tel.: +886-3-571-5131 (ext. 35582)
2
3
4
*
Academic Editor: Derek J. McPhee
Received: 5 February 2016 ; Accepted: 14 March 2016 ; Published: 21 March 2016
Abstract: Molecular imaging of brain tumors remains a great challenge, despite the advances made in
imaging technology. An anti-inflammatory compound may be a useful tool for this purpose because
there is evidence of inflammatory processes in brain tumor micro-environments. Fluorooctylfenbufen
amide (FOFA) was prepared from 8-chlorooctanol via treatment with potassium phthalimide,
tosylation with Ts2O, fluorination with KF under phase transfer catalyzed conditions, deprotection
using aqueous hydrazine, and coupling with fenbufen. The corresponding radiofluoro product
[18F]FOFA, had a final radiochemical yield of 2.81 mCi and was prepared from activated [18F]F´
(212 mCi) via HPLC purification and concentration. The radiochemical purity was determined to be
99%, and the specific activity was shown to exceed 22 GBq/µmol (EOS) based on decay-corrected
calculations. Ex-vivo analysis of [18F]FOFA in plasma using HPLC showed that the agent had a
half-life of 15 min. PET scanning showed significant accumulation of [18F]FOFA over tumor loci
with reasonable contrast in C6-glioma bearing rats. These results suggest that this molecule is
a promising agent for the visualization of brain tumors. Further investigations should focus on
tumor micro-environments.
Keywords: Inflammation; molecular imaging; NSAIDs; cyclooxygenase
1. Introduction
Brain disorders such as Alzheimer’s and Parkinson’s diseases are highly associated with
inflammation. Several lines of evidence suggest a critical role for cyclooxygenase type 2 (COX-2) in
tumorigenesis [
1,2]. It is also known that permanent inactivation of platelet COX-1 restores antitumor
reactivity [ ]. Thus, COX-1 and COX-2 both constitute interesting targets for the development of
3,4
inhibitors. In contrast to the housekeeping gene COX-1, COX-2 is an inducible enzyme that is expressed
at elevated levels at sites of inflammation and malignant transformations [5,6].
COX is a class of enzymes that transforms arachidonic acid (AA) into a series of prostaglandin
(PG) analogues. Upon transformation by the aldo-keto reductase superfamily (AKR), the products
generated, such as PGs or thromboxane, can induce subsequent inflammation-related signaling
pathways. COX enzymes are sequence homodimers composed of tightly associated monomers with
identical primary structures. However, COX enzymes function as conformational heterodimers with
one allosteric and one catalytic monomer [
7]. In addition to catalyzing the transformation of AA, COX-2
Molecules 2016, 21, 387; doi:10.3390/molecules21030387
www.mdpi.com/journal/molecules

Molecules 2016, 21, 387
2 of 17
also metabolizes endocanabinoids such as donoylethanolamine (AEA) and 2-arachidonoylglycerol
(2-AG). The metabolites of these endocanabinoids are analogues of PGs that are involved in the
regulation of the arachidonic acid metabolic pathway. COX inhibitors such as traditional non-steroid
anti-inflammatory drugs (NSAIDs). COX-2 inhibitors called “coxibs” and both substrate and
non-substrate fatty acids have been shown to bind within COX channels [8].
COX inhibitors are characterized into two main groups: rapid reversible inhibitors, such
as ibuprofen and mefenamate, and slow irreversible inhibitors, such as celecoxib, flurbiprofen,
and aspirin [9]. A fenbufen analogue, derived from NSAIDs, exhibited cytotoxic effects against
cancer cells with an IC50 value in the submicromolar range [10,11]. A recent study indicated
that 8-[123I]iodooctylfenbufen {[123I]IOFA} in combination with single photon emission computed
tomography (SPECT) could be used to image cholangiocarcinoma in rats [12]. However, imaging
of tumors with high mortality rates in clinical settings remains challenging (Figure 1) [13–16] HPLC
analyses of the binding affinities of [123I]IOFA to COX enzymes showed a non-selective pattern with a
preference for COX-2 over COX-1 (3:2). In vivo application of [123I]IOFA have been restricted due to its
unfavorable aqueous solubility. This physical chemical feature has limited its use in animal studies,
as well as in human applications.
Figure 1. NSAIDs analogue derived radiotracers [123I]IOFA and [18F]FOFA {[18F]F-
1} studied for
SPECT (previous study) and PET imaging (this study) of tumors, respectively.
To address this issue, the structure was modified by incorporating a fluorine atom (Figure 1).
Fluoro groups may form H-bonds with the H2O from the aqueous medium facilitating the solvation.
Radioactive 18F (t1/2 = 110 min), a positron emitter, is used in positron emission tomography (PET)
for molecular imaging purposes and clinical applications [17–19]. Due to its better sensitivity and
resolution, the use of PET is preferred over the use of SPECT. Its higher sensitivity is particularly
valuable for detecting the fewest possible number of cells per unit volume while emitting the least
amount of radioactivity. Secondly, the spatial/temporal resolution of PET is significantly higher
than that of SPECT and allows for dynamic scanning and the detection of small lesions [20]. Hence,
PET combines various chemical probes and offers a tool to investigate biochemical mechanisms
in vivo [21–23].
A number of 18F-labeled tracers have been developed and some of these have been demonstrated
to be useful for imaging ongoing inflammation [24]. Marnett and co-workers performed a series of
sophisticated studies to decipher COX-2 expression and monitor inflammation using 18F-coxibs [
5].
Encouraged by this advancement, we introduced the hydrogen bond-accepting F atom to improve
the hydroaffinity of [18F]FOFA for PET applications. This study reports the preparation of [18F]FOFA
{[18F]F-1}and assessment of its biological activity in vitro and in vivo.
2. Results and Discussion
Starting with the reaction of 8-chlorooctanol (2) with potassium phthalimide via a Gabriel
synthesis (Scheme 1), product was formed; however, the fluorination step of compound
5
3
with
diethylamino sulfur trifluoride (DAST) gave an unexpectedly low yield. This lower yield may have
been caused by the purity of DAST or the reaction conditions. The yield of the fluorination reaction
could be improved by using a different synthetic route that employed a precursor, tosylate 6, under

Molecules 2016, 21, 387
3 of 17
phase-transfer catalyzed reaction conditions. Hence, with reference compound for FOFA (
1
) in hand,
subsequent radiochemical fluorination of tosylate 6 to generate [18F]FOFA {[18F]F-1} was performed.
Scheme 1. Schematic depiction of the preparation of target [18F]FOFA ([18F]F-
non-radioactive standard FOFA (1).
1) and the reference
The purity of precursor
reaction. Residual chloride ions can compete with fluoride ions and hamper the radiochemical
yields [25]. In the present work, tosylate , prepared from TsCl, did not meet the elemental analysis
criteria. This purity concern was resolved by using Ts2O as the sources for tosylate moieties.
Radiofluorination of was optimized (n = 11) using two sources of tosylate coupled with two bases.
6 was a decisive factor for the radiochemical yield for the radiofluorination
6
6
As shown in Table 1, the radiochemical yield of the fluorination reaction of tosylate
Ts2O was obviously higher than that obtained using TsCl.
6 obtained using
Table 1. Comparison of radiochemical yield (RCY) using different reagent combinations.
Entry
Reagent Combination
RCY
Experiment No.
1
2
3
TsCl/pyridine
TsCl/DMAP
Ts2O/DMAP
33% ˘ 4%
39% ˘ 4%
53% ˘ 1%
4
3
4
Radiofluorination of tosylate 6
afforded [18F]F-4 with a radiochemical yield of 53%, as determined
by an HPLC chromatogram (decay corrected, Figure 2). The probable mechanism for the appearance
of an unknown compound (tR = 8.7 min) was illustrated in Scheme 2. Subsequent restoration of the
amino groups was accomplished using aqueous hydrazine. However, after several purification trials
using either normal phase or reverse phase HPLC under various elution conditions, we failed to
generate a satisfactory HPLC chromatogram. Thus, purification was performed after the subsequent
amide-formation step. Starting from the radioactivity of activated [18F]F´ (212 mCi) obtained by
repeated distillation with Kryptofix[2,2,2], a three-hour manipulation that included radiofluorination,
deprotection, amide formation, HPLC purification, drying and finally dissolving the reaction product
in aqueous 20% EtOH for subcutaneous injection afforded the target compound [18F]FOFA {[18F]-
4% radiochemical yield (2.81 mCi). The final product had a radiochemical purity of 99% and specific
activity of 22 GBq/
mol (EOS), based on decay-corrected calculations (n = 3). The identity of [18F]FOFA
1} in
µ

Molecules 2016, 21, 387
4 of 17
{[18F]-1} was confirmed by HPLC analysis in the presence of the nonradioactive FOFA (
1
) reference
compound (Supplementary Materials, Figure S1). Cold synthesis using the same procedure through
column chromatography generated the desired product FOFA ( ) in sufficient purity. No specific
impurities with similar tRs was observed in UV chromatogram of [18F]FOFA {[18F]-
} because the
polarities between the precursor, fenbufen, and the product FOFA (1), were significantly different.
1
1
1.2
1
10000
5000
0
[
18F]4
tR: 13.4 min
UV
Radio
0.8
0.6
0.4
0.2
0
tR: 8.7 min
0
5
10
15
20
25
30
min
Figure 2. HPLC analysis of the intermediate mixture of [18F]F-4; unknown at tR = 8.7 min is suggested
to be an eliminated alkene product as shown below. RCY of eliminated product was ca. 15%. ZORBAX
SIL 5 µm 9.4 ˆ 250 mm, EtOAc:n-hexane = 1:2, 3 mL/min.
Scheme 2. Proposed eliminated product derived from [18F]4.
Our previous report of an analogue of [18F]FOFA {[18F]-1}, [123I]IOFA, showed comparable binding
affinities to COX-1 and COX-2 enzymes in micromolarities. Due to their structural similarities, the
present work of [18F]FOFA {[18F]-
1} focused on assessing its aqueous solubility and bioavailability.
Its radiotracer cell uptake increased with prolonged incubation times (Figure 3). In contrast, cellular
uptake of the tracer showed marginal selectivity, and image contrast could be optimized if the clearance
rate was sufficiently rapid.
4
C6 glioma
3
fibroblast
2
1
0
0
20
40
60
80
100
120
Incubation time (min)
Figure 3. Comparison of internalization rate of [18F]FOFA {[18F]-1} between C6-glioma brain tumour
and the control fibroblast cells. Increasing uptake rate is remarkable in the lateral stage. Data are from
one experiment with triplicate samples and are expressed as mean ˘ SD.

Molecules 2016, 21, 387
5 of 17
As shown in Figure 4, a cell viability test of FOFA
1
in both C6 glioma and fibroblast cells showed
comparable inhibitory activities, but C6 glioma cells exhibited increased cytotoxicity in (IC50 < 100 uM).
Together, these features may be indicative of its application in vivo.
1.2
1.1
1.0
0.9
0.8
0.7
0.6
0.5
0.4
0.3
0.2
0.1
0.0
Concentration of FOFA
(A)
1.2
1.1
1.0
0.9
0.8
0.7
0.6
0.5
0.4
0.3
0.2
0.1
0.0
Concentration of FOFA
(B)
Figure 4. Cell survival assay of FOFA {[18F]-
1} in (A) fibroblast (a control) and (B) C6 glioma tumour
cells after 48 h.
To assess the in vivo stability of [18F]FOFA, an alternative environment containing a flask of
plasma was used as an ex vivo model. The ex vivo test was performed by first mixing the tracer
with plasma samples for specific time intervals, i.e., 10 s, 2 min, 10 min and 30 min. After injecting
the soluble fractions of the plasma mixture of [18F]FOFA {[18F]F-
[18F]FOFA {[18F]F-
1
} into an HPLC system, the intact
1
} and the released [18F]F´ were observed concomitantly in the chromatogram
(Figure 5). The two species were further illustrated using their time activity profiles, as shown in
Figure 6. Its short half life of less than 1 min implied its instability.

Molecules 2016, 21, 387
6 of 17
8000
6000
4000
2000
0
0.6
0.4
0.2
0.0
[
18F]F-
[
18F]FOFA
UV
counts
0
4
8
12
16
20
min
(A)
8000
6000
4000
2000
0
0.3
[
18F]F-
[
18F]FOFA
0.2
0.1
0.0
UV
counts
0
4
8
12
16
20
min
(B)
15000
0.3
[
18F]F-
[
18F]FOFA
10000
5000
0
0.2
0.1
0.0
UV
conuts
0
4
8
12
16
20
min
(C)
15000
10000
5000
0
0.6
0.4
0.2
0.0
[
18F]F-
UV
[
18F]FOFA
counts
0
4
8
12
16
20
min
(D)
Figure 5. HPLC profiles of the soluble fraction of the [18F]FOFA {[18F]F-1} mixture in plasma at different
times post injection, λabs = 260 nm. For comparison, a small amount of authentic nonradioactive FOFA
1
[
was added to the mixture as a referential ultraviolet signal. tR = 3.41 min, [18F]F´; tR = 14.51 min,
18F]FOFA {[18F]F-
(D). The time delay of 3 min in the chromatogram (B) was denoted due to an unexpected status.
1}. Samples were taken from mixture after 10 s (A); 2 min (B); 10 min (C) and 30 min

Molecules 2016, 21, 387
7 of 17
100
80
60
40
20
0
[18F]F-
[18F]FOFA
0
5
10
15
20
25
30
time (min.)
Figure 6. Illustration of [18F]FOFA {[18F]F-
decline of [18F]FOFA {[18F]F- }and growth of the metabolites [18F]F´ from peak integration in HPLC
1}stability in terms of time activity curve by observing the
1
chromatogram. The ex vivo half-life is approximately t1/2 < 1 min.
As expected, 20% EtOH (aq) was capable of dissolving [18F]FOFA {[18F]F-
1} during injection
preparation for animal PET study. This impressive solubility reminded us the contradictory experience
of observing a lot of suspended particles with [123I]IOFA. Additionally, a biodistribution experiment
was performed in two normal rats (Figure 7). The results showed significant accumulation in bone
tissues at 60 min post-injection. Approximately 1%–3% ID/g of activity was distributed over the brain.
The relatively nonpolar feature of FOFA {[18F]F-
1} did not impact its penetration across the blood brain
barrier prior to defluorination because the significant activity observed in bone tissues was denoted
60 min after injection. These results were consistent with the findings from the HPLC analyses.
32
2min
60min
24
16
8
0
Figure 7. Biodistribution of [18F]FOFA {[18F]F-
1} in healthy male SD rats. The two rats were injected
intravenously with 0.1 mCi and were euthanized and dissected after injection at 2 min and at 60 min,
respectively. Data are presented as %ID/g. The measurement was performed in triplicate.
The in vitro stability test using plasma indicated that a significant amount of [18F]F-
1 was released
at 2 min post-injection. However, the radiotracer had not significantly accumulated in the bone tissue
at 2-min resection from the in vivo test. This may have been caused by short observation times which
may have been insufficient for reaching an equilibrium within 2 min.
To assess the molecular imaging applicability, a rat that was inoculated with a tumor on the
right hemisphere of the brain was injected with a dose of 2.81 mCi/1 mL via a tail-vein injection.
The tumor loci was easily identified by MRI before PET scanning, as reported previously [26
,27].
The animal was assessed using static PET imaging over a 60-min time frame as indicated in Figure 8.

Molecules 2016, 21, 387
8 of 17
The results showed that most of the radioactivity was accumulated in the bone region. An obvious
accumulation over the tumor region with reasonable contrast from the background was also noted at
60 min post-injection.
Coronal
Sagittal
axial
Figure 8. A static PET/CT image collected over a period of 0 to 60 min of [18F]FOFA {[[18F]F-
the C6 glioma-bearing rat with an injection dose of 2.81 mCi/1 mL of [18F]FOFA{[18F]F-
}. Organs
1} in
1
scanned are shown for the head part. The first row shows CT images; the second row shows PET
images; the third shows their overlay.
Imaging data indicated that the brain tumor absorbed higher amount of the radiotracer to maintain
COX metabolism. Regions where COX metabolism is active enough to be traced by PET at early stages
of tumor progression require a more stable radiotracer. Although C-F bonding is strong enough to
resist bond cleavage [28], catalytic protonation of poor leaving groups could readily transform F´ to
HF, which is a good leaving group. A hepatic enzyme with a housekeeping role such as cytochrome
P450, may be able to address this mechanism (Scheme 3) [29]. Carbonyl and acyl halides have been
suggested to form after initial hydroxylation at the
subsequent removal of hydrogen fluoride [28,30].
α-position of carbon–fluorine bonds promoting the
In brief, the present work demonstrates the preparation of an octylfenbufen analogue by
modifying the terminal end of the octyl group with a fluorine atom for use as a PET tumor-imaging
probe for tumors and for inflammation processing. Modifying the terminal end group was shown to
improve the solubility of the compound in a 20% EtOH aqueous solution according to the observation
of injection preparation before PET scanning.

Molecules 2016, 21, 387
9 of 17
Scheme 3. The mechanism for removal of fluoride from [18F]FOFA{[18F]F-
1} in vivo was proposed by
lipase-catalysed hydrolysis through proton donation from the crboxyl or thiol group from amino acid
side chains e.g., aspartate, glutamate or cysteine.
3. Materials and Methods
3.1. General Information
All reagents and solvents were purchased from Sigma-Aldrich (St. Louis, MO, USA), Fluka
(St. Louis, MO, USA), Acros (Geal, Belgium), Alfa (Binfield, Berkshire, UK), Malingkrodt (Phillipsburg,
NJ, USA), or Tedia (Farfield, OH, USA). Preparation of all nonradioactive compounds, except for
compound 5, was routinely conducted in dried glassware under a positive pressure of nitrogen at room
temperature unless otherwise noted. CH2Cl2, toluene, CH3CN, and pyridine were dried over CaH2
and CH3OH was dried over Mg and distilled prior to their use in reactions. Solvents (e.g., DMF and
N(i-Pr)2Et) were distilled under reduced pressure. All the reagents and solvents were of reagent grade.
Dimethylaminopyridine (DMAP) was purified by recrystallizing the reagent from a combination of
EtOAc and n-hexane before use. The eluents for flash chromatography (e.g., EtOAc, acetone, and
n-hexane) were of industrial grade and were distilled prior to use; CH3OH and CHCl3 were of reagent
grade and used without further purification. Nylone filters with 0.45 µM pore size were supplied by
Waters (Nilford, MA, USA). NMR spectroscopy including 1H-NMR (500 MHz),13C-NMR (125 MHz,
DEPT-135) and 19F-NMR (470 or 564 MHz) was performed on a Unity Inova 500 MHz instrument
(Varian, USA). Deuterated-solvents employed for NMR spectroscopy including CD3OD, CDCl3,
C6D6, and DMSO-d6 were purchased from Aldrich. Low-Resolution Mass Spectrometry (LRMS) was
performed on an ESI-MS spectrometer using a Varian 901-MS Liquid Chromatography Tandem Mass
Q-TOF Spectrometer at the Department of Chemistry of National Tsing-Hua University (NTHU) or the
Department of Applied Chemistry of National Chio-Tung University (NCTU). High-Resolution Mass
Spectrometry (HRMS) was performed using a Varian HPLC (prostar series ESI/APCI) system coupled
with a Varian 901-MS (FT-ICR Mass) mass detector and a triple quadrapole instrument. Elemental
analyses were performed using an Elemental Analysis, CHN-O-RAPID apparatus (Foss Heraeus,
Burladingen, Germany). Thin layer chromatography (TLC) was performed with TLC silica gel 60 F254
pre-coated plates (Machery-Nagel, Dueren, Germany) to monitor the starting materials and products
with visualization under UV light (254 nm). Further confirmation was carried out by staining the TLC
plates with 5% p-anisaldehyde, ninhydrin or ceric ammonium molybdate under heating. Celite 545
was purchased from Macherey-Nagel Inc. (Dueren, Germany). Flash chromatography was performed
using Silicycle 60 silica gel (70–230 mesh, Quebec City, QC, Canada). Melting points were measured
with a MEL-TEMP instrument (Barnstead International, Dubuque, IA, USA) and were uncorrected.
[18F]HF was produced on a GE PET tracer cyclotron (Milwaukee, WI, USA) by an 18O(p,n)
nuclear reaction (NERI, Longtan, 32546, Taiwan). The radiolabeling experiment was performed on
a GE TracerLAB FXFN synthesis module (GE Medical Systems, Milwaukee, WI, USA). PET imaging
was performed with a NanoPET/CT (MEDISO Inc., Knoxville, TN, USA) instrument at the Nuclear

Molecules 2016, 21, 387
10 of 17
Energy Research Institute. The intermediate product, [18F]F-4, was analyzed with an Waters HPLC
system (Milford, MA, USA) consisting of a Waters 510 pump, a linear UVIS detector (254 nm) in
series with a Berthhold (Bad Wildbad, Baden Wuetermberg, Germany) γ-flow detector and a ZORBAX
SILcolumn (9.4 mm
ˆ
250 mm, 5 µm) using isocratic EtOAc/n-hexane 1:2 as the mobile phase with
a flow rate of 3 mL/min. The mixture of [18F]FOFA {[18F]F-1} obtained from the FXFN synthesis
module (TracerLAB, Amersham, UK) was purified with the same HPLC settings as described above;
however, a reverse phase Develosil ODS-7 (5
µ
m 10
ˆ
250 mm) column and a mobile phase consisting
of H2O–CH3CN = 30:70 in isocratic mode was utilized for these HPLC analyses. The identity of the
radiolabeled compound was confirmed by co-injecting it with an authentic standard into an HPLC for
analysis. The area under the UV absorbance peak measured at 260 nm, which corresponded to the
carrier product, was calculated and compared to a standard curve to relate mass to UV absorbance.
Only specific activity below 40 GBq/
using a R15C dose calibrator (Capintec, Ramsey, NJ, USA). Ex-vivo stability studies were performed on
the system as described, but using a CHEMCOSORB 7-ODS-H column (10 250 mm, 5 m) and an
µmol could be measured accurately. Radioactivity was measured
ˆ
µ
eluent system that had the following gradient program: CH3CN/0.05% trifluoracetic acid = 20/80 at
0 min to CH3CN /0.05% trifluoracetic acid = 95/5 at 10 min and a further gradient to CH3CN (100%)
at 20 min.
3.2. Chemical Syntheses
2-(8-Hydroxyoctyl)isoindoline-1,3-dione (3)
A mixture of ClCH2(CH2)7OH (
28.4 mmol) in dimethyl formamide (DMF, 12 mL) was stirred under reflux (160 C) for 2 h. TLC analysis
(EtOAc–n-hexane = 4:6) indicated the consumption of starting material (Rf = 0.48) and the formation
of product (Rf = 0.3). The mixture was then transferred to a separation funnel for an extraction
with CH2Cl2 (130 mL) and 1N HCl (50 mL 2). The organic layer was washed with saturated
2, 4 mL, 23.7 mmol) and potassium phthalimide (5.26 g, 1.2 eq,
˝
2
3
ˆ
NaHCO3 (aq) followed by drying with Na2SO4. After filtration through a pad of celite, the filtrate was
˝
sequentially concentrated using a membrane pump and an oil pump, under reduced pressure at 50 C.
A brown viscous oil was obtained and was purified using flash chromatography under the following
conditions: a column that was 6 cm in diameter and using 250 g of silica gel and EtOAc–n-hexane
= 4:6 eluent system. The product
3 was obtained in 70% yield (4.5 g). After recrystallization from
˝
˝
CH3OH, a colorless, slightly white crystal was obtained in 67% yield (4.41 g). m.p. 63–64 C (60–62 C,
42% yield; [31] white wax, 72%) [32]. Calcd. C16H21NO3 MW = 275.3, ESI + Q-TOF MS, M = 275.2
(m/z), [M + Na]+ = 298.1; HRMS-ESI, Calcd. [M + H]+ = 276.1600 [M + Na]+ = 298.1419; found:
,
[M + H]+ = 276.1597, [M + Na]+ = 298.1419; Elemental analysis: Calcd. C, 69.79; H, 7.69; N, 5.09.
1
Found C, 69.46; H, 7.81; N, 5.36. H-NMR (500 MHz, CDCl3):
δ 1.32–1.33 (m, 8H, CH2), 1.52–1.56 (m,
2H, CH2), 1.65–1.69 (m, 2H, CH2), 2.16 (s, 1H, OH), 3.62 (t, J = 6.5 Hz, 2H, H-8), 3.67 (t, J = 7.5 Hz, 2H,
H-1 ), 7.68–7.72 (m, 2H, Ar), 7.81–7.85 (m, 2H, Ar); 13C-NMR (125 MHz, CDCl3):
25.58 ( H2), 26.70
H2), 28.50 ( H2), 29.04 ( H2), 29.18 ( H2), 32.67 ( H2), 37.98 ( H2), 62.96 ( H2), 123.13 (CH, arom),
δ
C
(C
C
C
C
C
C
C
132.12 (C, arom), 133.82 (CH, arom), 168.47 (C, COamide).
8-(1,3-Dioxoisoindolin-2-yl)octyl 4-methylbenzenesulfonate (6)
Bartholomä et al. [32] utilized TsCl to prepare compound
6
. Dimethylaminopyridine (DMAP,
was
181 mg, 1.48 mmol, 1.9 eq) was dried by co-distillation with toluene (1 mL) thrice. Compound
3
also co-distilled with toluene (1 mL) thrice and was dissolved in CH2Cl2 (3 mL) in a two-necked round
bottom flask. DMAP was then added to this solution and the mixture was stirred in an ice bath. Then,
toluene sulfonic anhydride (Ts2O, 473 mg, 1.45 mmol, 1.9 eq) was added and the stirring continued
for 30 min. After removal of the ice bath, the reaction was allowed to continue at RT for 2 h. A TLC
(EtOAc–n-hexane = 4:6) analysis indicated the consumption of starting material
3 (Rf = 0.25) and
the formation of product (Rf = 0.55). The mixture was then partitioned between 1N HCl (30 mL)
6

Molecules 2016, 21, 387
11 of 17
and CH2Cl2 (60 mL) in a separation funnel. The organic layer was dried with Na2SO4 followed by
filtration through a celite pad. The filtrate was concentrated under reduced pressure to give dark
brown colored viscous oil (310 mg). The residue was purified by flash chromatography in a column
that was 2 cm in diameter with 20 g silica gel using EtOAc–n-hexane = 2:8 as an eluent to afford the
product
obtained as a colorless oil in 45% yield. and elemental analysis data were not available. Preparation
of the compound using toluene sulfonyl chloride (TsCl) did not provide product with satisfactory
6 as a pale yellow oil in 75% yield (246 mg). The product synthesized by Bartholomä et al. was
6
purity according to the elemental analysis: H, 5.82 (6.34, calcd.). Calcd. C23H27NO5S, MW = 429.5,
ESI + Q-TOF MS, M = 429.2 (m/z), [M + H]+ = 430.1 (100%), [M + Na]+ = 452.1, [2M + Na]+ = 881.5;
HRMS-ESI, Calcd. [M + H]+ = 430.1688, [M + Na]+ = 452.1508; found: [M + H]+ = 430.1688, [M+Na]+
= 452.1506; Elemental analysis: Calcd. C, 64.31; H, 6.34; N, 3.26. Found C, 64.32; H, 6.33; N, 3.40.
1H-NMR (500 MHz, CDCl3):
(t, J = 7.0 Hz, 2H, H-1), 4.00 (t, J = 6.5 Hz, 2H, H-8), 7.34 (d, J = 8.5 Hz, 2H, Ar), 7.70–7.72 (m, 2H, Ar),
7.78 (d, J = 8.0 Hz, 2H, Ar), 7.82–7.84 (m, 2H, Ar); 13C-NMR (125 MHz, CDCl3):
21.59 ( H3), 25.20
H2), 26.63 ( H2), 28.46 ( H2), 28.71 ( H2), 28.73 ( H2), 28.86 ( H2), 37.91 ( H2), 70.57 ( H2), 123.13
δ 1.20–1.30 (m, 8H, CH2), 1.58–1.65 (m, 4H, CH2), 2.44 (s, 3H, CH3), 3.65
δ
C
(C
C
C
C
C
C
C
C
(CH, arom), 127.85 (CH, arom), 129.77 (CH, arom), 132.12 (C, arom), 133.21 (C, arom), 133.84 (CH,
arom), 144.59 (C, arom), 168.42 (C, COamide).
2-(8-Fluorooctyl)isoindoline-1,3-dione (4)
Route 1 via fluorination with diethyl aminosulfur trifluoride (DAST)
A similar product with a truncated heptyl chain was reported by Li et al. [33]. A mixture of
˝
compound
3
(2 g, 7.3 mmol) in CH2Cl2 (5 mL) was stirred at
´
40 C. DAST (1.6 mL, 13 mmol, 1.8 eq)
was added portionwise within 20 s, and the reaction was stirred for 2 h. TLC (EtOAc–n-hexane = 4:6)
analysis indicated consumption of the starting material (Rf = 0.28) and formation of the product
(Rf = 0.65). The mixture was then partitioned between CH2Cl2 (60 mL) and NaHCO3(aq)
in a separation funnel. The organic layer was dried using Na2SO4 followed by filtration through
a celite pad. The filtrate was concentrated under reduced pressure to give a yellow solid (650 mg).
This crude compound was purified by flash chromatography in a column that was 3 cm in diameter
3
4
(
30 mL ˆ 2
)
with 50 g of silica gel using a mixture of EtOAc–n-hexane = 1:9 as an eluent, product 4 was obtained
as a white solid in 27% yield (550 mg). After recrystallization from CH3OH, a colorless crystal was
obtained in 26% yield (523 mg). m.p. 51–52 ˝C. Calcd. C16H20FNO2, MW = 277.3, ESI + Q-TOF MS,
M = 277.1 (m/z), [M]+ = 277.1, [M + Na]+ = 300.1, [M + K]+ = 316.1; HRMS-ESI , Calcd. M = 277.1418
(m/z), [M + H]+ = 278.1556, [M + Na]+ = 300.1376, [M + K]+ = 316.1115; found: [M + H]+ = 278.1543
,
[M + Na]+ = 300.1363, [M + K]+ = 316.1284; Elemental analysis: Calcd. C, 69.29; H, 7.27; N, 5.05. Found
1
C, 69.19; H, 7.63; N, 5.35. H-NMR (500 MHz, CDCl3):
δ 1.30–1.41 (m, 8H, CH2), 1.62–1.71 (m, 4H,
CH2), 3.67 (t, J = 7.5 Hz, 2H, H-1), 4.42 (dt, JH´8,F = 47.5 Hz, JH´8,H´7 = 6.5 Hz, 2H, H-8), 7.70–7.71
(m, 2H, Ar), 7.83–7.84 (m, 2H, Ar); 13C-NMR (125 MHz, CDCl3): δ 25.04 (d, J = 5.5 Hz, CH2), 26.73 (d,
J = 4.5 Hz, CH2), 28.52 (
C
H2), 28.82 (d, J = 33.2 Hz,
CH2, C-6), 29.03 (d, J = 4.1 Hz,
CH2, C-5), 30.32 (d,
J = 19.4 Hz H2, C-7), 37.98 (
,
C
C
H2), 84.15 (d, J = 163.9 Hz, C
H2F) , 123.14 (CH, arom), 132.16 (C, arom),
133.83 (CH, arom), 168.46 (C, COamide). 19F-NMR (564 MHz, CDCl3): δ ´218.04 (tt, JF,H´8 = 47.5
JF,H´7 = 24.9 Hz).
,
Route 2 via fluorination with KF and a phase-transfer catalyst
A mixture of KF (20 mg, 0.35 mmol, 5 eq) and K2CO3 (3.5 mg, 0.025 mmol, 0.4 eq) in CH3CN
(1 mL) was stirred at RT. This was followed by addition of a solution of Kryptofix [2.2.2] (30 mg,
0.08 mmol, 1.1 eq) in CH3CN (0.5 mL), which was previously dried twice using co-distillation in
CH3CN (1 mL) under reduced pressure. After addition of the solution of compound
6 (30 mg,
0.07 mmol, 1 eq) in CH3CN (0.5 mL) which was previously dried twice using CH3CN (1 mL) through
˝
co-distillation under reduced pressure, the mixture was heated under reflux (100 C) for 30 min.

Molecules 2016, 21, 387
12 of 17
TLC (EtOAc–n-hexane = 2:8) analysis indicated formation of the product
4
(Rf = 0.50). A portion of
CH3CN (5 mL) was added, and the mixture was sequentially filtered through an Al cartridge and
RC 18 cartridge. The filtrate was concentrated at 40 ˝C under reduced pressure to afford a yellow
colored crude residue (65 mg). This residue was purified by flash chromatography using a 1.5 cm
diameter column with 10 g of silica gel under a gradient eluent condition of EtOAc–n-hexane =
1:9 EtOAc–n-hexane = 2:8. Product 4 was obtained as a white solid in 62% yield (12 mg). After
Ñ
recrystallization from CH3OH, a colorless crystal was obtained in 40% yield (8 mg). m.p. 51–52 ˝C.
The spectroscopic data including 1H- and 19F-NMR and HRMS were consistent with the data obtained
using route 1. Elemental analysis for C16H20FNO2: Calcd. C, 69.29; H, 7.27; N, 5.05. Found C, 69.27; H,
7.34; N, 5.37.
8-Fluorooctan-1-amine (5)
Pattison et al. previously reported a method for the preparation of compound
of compound (300 mg, 1.08 mmol) and EtOH (1 mL) in a two-necked round bottomed flask (25 mL)
was heated at 80 ˝C. To this mixture, NH2NH2
H2O (162 L, 64 wt %, 3.24 mmol, 3 eq) was added.
The colorless solution was then stirred for 4 h. TLC analysis was performed using two types of eluting
conditions and indicated the consumption of the starting material Rf = 0.62, EtOAc–n-hexane = 4:6)
and the formation of product (Rf = 0.22, CH3OH:CHCl3:NH3 = 50%:49%:1%). An anionic exchange
5 [34,35]. A mixture
4
¨
µ
4
(
5
resin (OH´) was added to the mixture whi˝ch was then filtered by gravity filtration. The filtrate was
concentrated under reduced pressure at 40 C to afford a dark brown oil (200 mg) with a strong odor.
The residue was purified using flash chromatography in a 2.5 cm-diameter column with 30 g of silica gel
using CH3OH:CHCl3:Et3N = 50%:49%:1% as an eluent to afford the product
5 as a yellowish oil in 60%
yield (95 mg) Pattison et al. reported an 81% yield from the hydrogenation of 8-fluorooctane nitrile using
LiAlH4; a 56% yield was reported by Windhorst et al. from the hydrogenation of 7-fluoroheptane nitrile
using NaBH4. Calcd. C8H18FN, MW = 147.2, ESI + Q-TOF MS, M = 147.1 (m/z), [M + H]+ = 148.1;
HRMS-ESI , Calcd. [M + H]+ = 148.1502; found: [M + H]+ = 148.1497; Elemental analysis: Calcd. C,
1
65.26; H, 12.32; N, 9.51. Found C, 65.16; H, 12.16; N, 9.68. H-NMR (500 MHz, CD3OD):
δ
1.34–1.40
(m, 8H, CH2), 1.47–1.49 (m, 2H, CH2), 1.62–1.69 (m, 2H, CH2), 2.65 (bs, 2H, CH2NH2), 4.39 (dt,
JH´8,F = 47.5 Hz, JH´8,H´7 = 6 Hz, 2H, H-8); 13C-NMR (125 MHz, CD3OD):
δ
26.27 (
C
H2), 26.31 (
CH2),
27.87 (
C
H2), 30.38 (d, J = 21.2 Hz,
C
H2), 31.54 (d, J = 20.0 Hz, CH2), 33.26 (
CH2), 42.34 (C
H2), 84.88 (d,
J = 163.4 Hz, CH2F). 19F-NMR (470 MHz, CD3OD): δ ´218.44 (tt, JF,H´8 = 47.5, JF,H´7 = 24.5 Hz).
8-Fluorooctyl Fenbufen Amide (FOFA, 1)
A mixture of HBTU (68 mg, 0.17 mmol, 1.2 eq), fenbufen (68 mg, 0.15 mmol, 1 eq) and
diisopropyl-ethylamine (49.5
Compound (24 mg, 0.16 mmol, 1.1 eq) was then added to this mixture and the reaction was allowed
to stir for another 5 min. TLC (acetone–n-hexane = 5:5) analysis indicated the consumption of fenbufen
(Rf = 0.33) and formation of the intermediate ester (Rf = 0.65). Finally, the product FOFA Rf = 0.38
was observed through TLC analysis using a developing solvent of acetone–n-hexane = 3: 7. The mixture
was then partitioned between saturated aqueous NaHCO3(aq) (10 mL 2) and CH2Cl2 (20 mL) in a
µL, 0.3 mmol, 2 eq) in DMF (0.5 mL) was stirred at RT for 5 min.
5
1
(
)
ˆ
separation funnel. The organic layer was dried using Na2SO4 followed by filtration through a celite
pad. The filtrate was concentrated under reduced pressure to give a dark brown viscous oil (100 mg).
The residue was purified by flash chromatography using a 2.5 cm-diameter column with 20 g of silica
gel under eluting conditions using acetone–CHCl3 = 1:49 to give the product FOFA (1) as a white
solid in 75% yield (45 mg). After recrystallization from CH3CH2OH, off-white needle-shaped crystals
were obtained in 67% yield (39 mg). m.p.126–128 ˝C. Calcd. C24H30FNO2, MW = 383.5, ESI + Q-TOF
MS, M = 383.2 (m/z), [M + H]+ = 384. 3, [M + Na]+ = 406.2; HRMS-ESI, Calcd. [M + H]+ = 384.2339
[M + Na]+ = 406.2158, [M + K]+ = 422.1898; found: [M + H]+ = 384.2327, [M + Na]+ = 406.2147,
[M + K]+ = 422.1883; Elemental analysis: Calcd. C, 75.16; H, 7.88; N, 3.65. Found C, 75.28; H, 7.74;
,
N, 3.77. 1H-NMR (500 MHz, CDCl3):
δ 1.35–1.39 (m, 8H, CH2), 1.49–1.52 (m, 2H, CH2), 1.63–1.71

Molecules 2016, 21, 387
13 of 17
(m, 2H, CH2), 2.63 (t, J = 6.5 Hz, 2H, CH2), 3.23–3.27 (m, 2H, CH2), 3.40–3.41 (m, 2H, CH2), 4.42 (dt,
JH´8,F = 47.5 Hz, JH´8,H´7 = 6.0 Hz, 2H, H-8), 5.78 (bs, 1H, CON ), 7.39–7.41 (m, 1H, Ar), 7.46–7.49
(m, 2H, Ar), 7.63 (d, J = 7.0 Hz, 2H, Ar), 7.69 (d, J = 9.0 Hz, 2H, Ar), 8.06 (d, J = 8.5 Hz, 2H, Ar);
13C-NMR (125 MHz, CDCl3):
25.04 ( H2), 25.08 ( H2), 26.75 ( H2), 29.09 ( H2), 29.12 ( H2), 29.57
H2), 30.30 (d, J = 20.7 Hz, H2), 34.23 ( H2), 39.61 ( H2), 84.16 (d, J = 163.9 Hz, H2F), 127.25 (CH,
H
δ
C
C
C
C
C
(C
C
C
C
C
arom), 128.24 (CH, arom), 128.66 (CH, arom), 128.94 (CH, arom), 135.26 (C, arom), 139.82 (C, arom),
145.93 (C, arom), 171.97 (C, COamide), 198.76 (C, C=O). 19F-NMR (470 MHz, CDCl3): δ ´218.02 (tt,
JF,H´8 = 47.5, JF,H´7 = 24.9 Hz).
3.3. Radiochemistry
Preparation of 8-[18F]fluorooctyl-1-amine ([18F]F-5)
A reaction vessel containing 315 mCi of K[18F]F was obtained from enriched [18O]H2O via proton
irradiation followed by elution through a column of AGI X 8 resin. A solution of Kryprtofix[2.2.2] (20
mg) in CH3CN (1 mL) was added to the vessel followed by heating at 100 ˝C under fumes of He gas
according to a previously described procedure [26]. The formed Kryptofix-+K-[18F]´F complex was
dried via repeated dissolution and distillation. A mixture of tosylate 6 (20 mg) and t-BuOH (0.4 mL)
in CH3CN (1 mL) was added to this reaction vessel. The reaction was allowed to stir at 100 ˝C and a
˝
pressure of 250 kpa for 10 min. After cooling to 90 C, the mixture was purged with He fumes for 2 min
and was concentrated under reduced pressure for 2 min. In a probe analysis, the obtained intermediate,
[18F]F-4, was purified using normal phase HPLC as stated in the general section with a retention
time (tR) of 13.4 min. Upon cooling the sample down to 30 ˝C, a mixture of aqueous H2NNH2
H2O
¨
(40
µ
L) and EtOH (2 mL) was˝injected. The reaction was then carried out at 70 ˝C for 15 min and
was subsequently cooled to 30 C. The mixture was eluted through a set up comprising two alumina
cartridges and an RC-18 cartridge. The eluents were filtered through a 0.45 µM nylon filter and were
then washed with EtOH (1.5 mL). The filtrates were combined to give crude 8-[18F]fluorooctyl-1-amine
[18F]F-5 in 60% yield (131.2 mCi). Further purification using either normal or reverse phase HPLC
was unsuccessful. Thus purification was peformed at the next stage of conjugation with fenbufen.
The mixture (3 mL) was transferred to a round bottomed flask (10 mL) and was concentrated under
˝
under reduced pressure (20 mbar) at 40 C for 15 min to afford [18F]F-5 with a radioactivity of 14 mCi;
this compound was then used for the next conjugation experiment.
Preparation of 8-[18F]fluorooctylfenbufen amide ([18F]F-1)
A mixture of fenbufen (20 mg, 0.078 mmol, 1 eq), HBTU (38 mg, 0.1 mmol, 1.3 eq), diisopropyl
ethylamine (26 µL, 0.157 mmol, 2 eq) and DMF (150 µL) in a microcentrifuge tube was sonicated for
5 min. A brown solution of activated fenbufen ester was formed and was added to the flask containing
[18F]F-5. The reaction mixture was stirred at RT for 15 min. The mixture was then purified through a set
up comprising an Al cartridge, an RC-18 cartridge, and a 0.45 µm Nylon Millipore filter. The filtrates,
along with washings of CH2Cl2 (5 mL), were combined and concentrated under reduced pressure
at 50 ˝C to give the crude product [18F]F-1 with a radioactivity of 4.82 mCi. Followed by addition of
EtOH (0.5 mL), the mixture was purified by RP-HPLC under eluting condition of H2O:CH3CN = 30:70
in isocratic mode, as described in the general section. The desired fraction (tR = 15.4 min) was isolated
and concentrated. After dissolving the sample in CH3CH2OH (200 µL) and saline (800 µL), the solution
was shown to have a radioactivity of 2.81 mCi and was used for the PET imaging study.
3.4. Cell Culture for C6 Glioma and Fibroblasts
The procedure was performed as described previously [26] A rat glioma cell line C6 was obtained
from the American Type Cell Collection (ATCC, Manassas, VA, USA). The C6 glioma cells were
cultured with Dulbecco’s modified Eagles medium (DMEM) containing 10% fetal calf serum (Gibco,
Thermo Fisher Scientific, MA, USA) under 5% of CO2 at 37 ˝C in a 96-well microtiterplate. The cells

Molecules 2016, 21, 387
14 of 17
were subcultured upon reaching 80%–90% confluency. The fibroblast cell line, 3T3, was provided by
our collaborator, Dr. Ya-Hwei Wu who purchased it from ATCC (American Type Cell Collection).
The cells were maintainedin the same culture conditions described for the C6 cells.
3.5. Animal Model of Cells Implanted in Rat Brains
Sprague-Dawley (SD) rats (8 weeks of age) were obtained from the BioLasco Animal Co. (Taipei,
Taiwan). All studies involving animals were conducted in compliance with federal and institutional
guidelines, including proper animal housing sanitation, clean water supply, sufficient feeding, 12-h
day/night light control, and humidity/temperature control. The Institutional Animal Care and Use
Committee of Chang Gung University approved the animal experiments (grant no. CGU 14-164).
Two normal rats were used for the biodistribution experiment. One tumor rat was used for PET
study. Three weeks before performing PET imaging study, adult Sprague-Dawley male rats (ages
6–8 weeks, body weight 250–300 g) were stereotactically xenografted with 5
µL (5
ˆ
105) of C6 cells in
the right hemisphere under isoflurane anesthesia. (American Type Tissue Collection). The animals
were placed on heating pad until they have entirely recovered. The animals were transferred to the
animal facility. The research staffs monitored the animal body weight, water intake, signs of pain or
distress every morning before the PET imaging. If the animals appear lethargic, do not appear to be
eating or drinking over 24 h, or weight loss greater than 20% body weight, euthanasia will be carried
out to avoid further suffering. The animals were monitored regularly with care in respect with the
feeding quality, interaction, and symptom of dystrophy. The tumor growth was monitored by IVIS at
the first and second week after implantation procedure. MRI was then used to monitor the tumor size
once before the PET imaging study. The tumor can usually grow with a size of 3 mm–5 mm in diameter
two to three weeks after implantation. Prior to PET imaging, all rats were affixed with venous and
arterial catheters.
3.6. MTT Assay (Cell Viability Assay)
The MTT assay was performed as described previously [27]. Cell toxicity was assessed by
observing the reducing potential of mitochondria organelle in the cells upon addition of the MTT
reagent (3-(4,5-dimethyl-thiazol-2-yl)-2,5-diphenyl tetrazolium bromide). In brief, C-6 glioma cells
and fibroblast cells were plated in a 96-well plate (5000/well) and incubated with MEM containing
10% fetal calf serum (Gibco) at 37 ˝C under a 5% CO2 atmosphere, followed by treatment of FOFA.
After a 48-h culture period, 50
µL of MTT (5 mg/mL) was added to each well and the cells were
incubated for an additional 4 h. The supernatant was removed and 100
µL of DMSO was added to
each well to dissolve the formazan crystals formed. The optical absorbance at 570 nm was read by
a spectrophotometer.
3.7. Analysis of the Stability of [18F]FOFA {[18F]F-1} in Plasma
Before carrying out the in vivo imaging experiment, the ex vivo stability of [18F]F-
using HPLC analysis of the mixture containing [18F]F-
in blood plasma according to a reported
procedure [36]. An aliquot (10 radiotracer
L, 50 uCi) drawn from the solution containing the [18F]F-
in EtOH (100 uCi/500 uL) was mixed with a volume of 0.2 mL of heparin-pretreated plasma taken
1 was assessed
1
µ
1
from the SD rats (5 IU/mL). After standing for 10 s, 2 min, 10 min and 30 min, an aliquot (20
of the plasma mixture was added to a tube containing a solution of CH3CN and H2O (7:1, 180
µ
L)
µL).
After centrifugation (6500 rpm/200 g, Qik Spin, Narellan NSW, Australia) for 1 min, a volume of
100 L was drawn from the supernatant and was filtered through a membrane filter (0.45 M, PTFE,
µ
µ
Millipore). The filtrate was chromatographed using a reverse phase HPLC system equipped with a
semi-preparative column. The elution condition was the same as that described for the radiochemical
purification process. The generated radioactivity chromatogram was analyzed to establish a temporal
plot for the concentration of the radioactive product and its metabolites.

Molecules 2016, 21, 387
15 of 17
3.8. Biodistribution of [18F]F-1 in Normal Rats
The experimental procedure was similar to that used previously [26]. In brief, [18F]FOFA {[18F]-
1}
dissolved in 20% EtOH (aq, 0.3 mL) with an activity of approximately 1 mCi was injected into two
rats via the tail vein. The two rats were sacrificed subsequently after 2 and 60 min, respectively, to
excise various specimens. These included twelve organ tissues such as the brain, liver, spleen, heart,
kidney, lung, colon, small bowel, stomach, testes, skull, and muscle. These specimens were submitted
for quantification of radioactivity using a solid scintillation gamma counter (Packard 5000, Packard
Instrument Co. Laboratory, Meriden, CT, USA). The counting value of each specimen was further
divided by the sample weight to give the final expression as a percentage of the injection dose per
sample weight (%ID/g).
3.9. PET/CT Imaging Study Using [18F]F-1 in an Animal Model
PET scanning experiments were performed within 72 h of the MRI experiments that were used
to confirm successful inoculation of the tumors by administering [18F]F-
1 via tail vein injections [27].
NanoPET/CT in static mode was employed over 60 min scanning sessions. Data were acquired in
list-mode. The raw data within each frame were then binned into three-dimensional sonograms with a
span of three and ring difference of 47. The data were corrected for scattering and attenuation using a
two-dimensional ordered-subsets expectation-maximization algorithm, which had 16 subsets and four
iterations. The sonograms were reconstructed into tomographic images (128
sizes of 0.095 ˆ 0.095 ˆ 0.08 cm3.
ˆ
128
ˆ
95) with voxel
Supplementary Materials: Supplementary materials can be accessed at: http://www.mdpi.com/1420-3049/
21/3/387/s1.
Acknowledgments: We are grateful to the National Science Council of Taiwan, CGMH_NTHU Joint
Research, and Chang-Gung Medical Research Project for providing financial support via the following grants:
NSC-100-2113-M-007-003, NSC-97-2314-B-182A-020-MY3, NSC-97-2314-B-182A-020-MY3, CGTH96N2342E1,
CMRPG390931, CMRPG3A0512 and CMRPG3B0361. We acknowledge Ho-Lien Huang for his technical
assistance. We are also thankful for the technical assistance provided by Jenn-Tzong Chen, Yean-Hung Tu,
and Gang-Wei Chang at the Nuclear Energy Research Institute, Taiwan.
Author Contributions: Y.-C. Huang performed the biological study and partial radiochemical labeling.
Y.-C. Chang performed the chemical synthesis. C.-N. Yeh analyzed the imaging data and C.-S. Yu conducted the
whole plan and drafted the work.
Conflicts of Interest: The authors declare no conflict of interest.
References
1.
2.
3.
4.
Phillis, W.; Horrocks, L.A.; Farooqui, A.A. Cyclooxygenases, lipoxygenases, and epoxygenases in CNS:
Their role and involvement in neurological disorders Brain Res. Rev. 2006, 52, 201–243. [CrossRef] [PubMed]
De Vries, E.F.J. Imaging of cyclooxygenase-2 (COX-2) expression: Potential use in diagnosis and drug
evaluation. Curr. Pharm. Des. 2006, 12, 3847–3856. [PubMed]
Patrignani, P.; Tacconelli, S.; Sciulli, M.G.; Capone, M.L. New insights into COX-2 biology and inhibition.
Brain Res. Rev. 2005, 48, 352–359. [CrossRef] [PubMed]
Uddin, Md.J.; Elleman, A.V.; Ghebreselasie, K.; Daniel, C.K.; Crews, B.C.; Nance, K.D.; Huda, T.; Marnett, L.J.
Design of fluorine-containing 3,4-diarylfuran-2(5H)-ones as selective COX-1 inhibitors. ACS Med. Chem. Lett.
2014, 5, 1254–1258. [CrossRef] [PubMed]
5.
Uddin, Md.J.; Crews, B.C.; Ghebreselasie, K.; Huda, I.; Kingsley, P.J.; Ansari, M.S.; Tantawy, M.N.;
Reese, J.; Marnett, L.J. Fluorinated COX-2 Inhibitors as Agents in PET Imaging of Inflammation and Cancer.
Cancer Prev. Res. 2011, 4, 1536–1545. [CrossRef] [PubMed]
6.
7.
Taketo, M.M. Cyclooxygenase-2 inhibitors in tumorigenesis (Part II). J. Natl. Can. Inst. 1998, 90, 1529–1536.
[CrossRef]
Alhouayek, M.; Muccioli, G.G. COX-2-derived endocannabinoid metabolites as novel inflammatory
mediators. Trends Pharmacol. Sci. 2014, 35, 284–292. [CrossRef] [PubMed]

Molecules 2016, 21, 387
16 of 17
8.
9.
Blobaum, A.L.; Marnett, L.J. Structural and functional basis of cyclooxygenase inhibition. J. Med. Chem. 2007
50, 1425–1441. [CrossRef] [PubMed]
Smith, W.L.; Urade, Y.; Jokobsson, P.-J. Enzymes of the cyclooxygenase pathways of prostanoid biosynthesis.
,
Chem. Rev. 2011, 111, 5821–5865. [CrossRef] [PubMed]
10. Lin, K.-I.; Yang, C.-H.; Huang, C.-W.; Jian, J.-Y.; Huang, Y.-C.; Yu, C.-S. Synthesis and structure-activity
relationships of fenbufen amide analogs. Molecules 2010, 15, 8796–8803. [CrossRef] [PubMed]
11. Su, Y.-H.; Chiang, L.-W.; Jeng, K.-C.; Huang, H.-L.; Chen, J.; Lin, W.; Huang, C.-W.; Yu, C.-S. Solution-phase
parallel synthesis and screening of anti-tumor activities from fenbufen and ethacrynic acid libraries.
Bioorg. Med. Chem. Lett. 2011, 21, 1320–1324. [CrossRef] [PubMed]
12. Huang, H.-L.; Yeh, C.-N.; Lee, W.-Y.; Huang, Y.-C.; Chang, K.-W.; Lin, K.-J.; Tien, S.-F.; Su, W.-C.; Yang, C.-H.;
Chen, J.-T.; et al. [123I]Iodooctyl fenbufen amide as a SPECT tracer for imaging tumors that overexpress COX
enzymes. Biomaterials 2013, 34, 3355–3365. [CrossRef] [PubMed]
13. Chang, K.Y.; Chang, J.Y.; Yen, Y. Increasing incidence of intrahepatic cholangiocarcinoma and its relationship
to chronic viral hepatitis. J. Natl. Compr. Cancer Netw. 2009, 7, 423–427.
14. Yeh, C.N.; Maitra, A.; Lee, K.F.; Jan, Y.Y.; Chen, M.F. Thioacetamide-induced intestinal-type
cholangiocarcinoma in rat: An animal model recapitulating the multi-stage progression of human
cholangiocarcinoma. Carcinogenesis 2004, 25, 631–636. [CrossRef] [PubMed]
15. Yeh, C.N.; Chiang, K.C.; Juang, H.H.; Pang, J.H.S.; Yu, C.S.; Lin, K.J.; Yeh, T.S.; Jan, Y.Y. Reappraisal of the
therapeutic role of celecoxib in cholangiocarcinoma. PLoS ONE 2013, 8. [CrossRef]
16. Yeh, C.N.; Weng, W.H.; Lenka, G.; Tsao, L.C.; Chiang, K.C.; Pang, S.T.; Chen, T.W.; Jan, Y.Y.; Chen, M.F. cDNA
microarray profiling of rat cholangiocarcinoma induced by thioacetamide. Mol. Med. Rep. 2013, 8, 350–360.
[PubMed]
17. Cai, L.; Lu, S.; Pike, V.W. Chemistry with [18F]fluoride ion. Eur. J. Org. Chem. 2008, 2853–2873. [CrossRef]
18. Lasne, M.C.; Perrio, C.; Rouden, J.; Barre, L.; Roeda, D.; Dolle, F.; Crouzel, C. Chemistry of beta(+)-emitting
compounds based on fluorine-18. Top. Curr. Chem. 2002, 222, 201–258.
19. Jacobson, O.; Kiesewetter, D.O.; Chen, X. Fluorine-18 radiochemistry, labeling strategies and synthetic routes.
Bioconjugate Chem. 2015, 26, 1–18. [CrossRef] [PubMed]
20. Yang, Y.-S.; Zhang, X.; Xiong, Z.; Chen, X. Comparative in vitro and in vivo evaluation of two 64Cu-labeled
bombesin analogs in a mouse model of human prostate adenocarcinoma. Nucl. Med. Biol. 2006, 33, 371–380.
[CrossRef] [PubMed]
21. Van de Bittner, G.C.; Ricq, E.L.; Hooker, J.M.A. Philosophy for CNS Radiotracer Design. Acc. Chem. Res.
2014, 47, 3127–3134. [CrossRef] [PubMed]
22. Fletcher, J.W.; Djulbegovic, B.; Soares, H.P.; Siegel, B.A.; Lowe, V.J.; Lyman, G.H.; Coleman, R.E.; Wahl, R.;
Paschold, J.C.; Avril, N.; et al. Recommendations on the Use of 18F-FDG PET in Oncology. J. Nucl. Med. 2008
,
49, 480–508. [CrossRef] [PubMed]
23. Dunet, V.; Rossier, C.; Buck, A.; Stupp, R.; Prior, J.O. Performance of 18F-Fluoro-Ethyl-Tyrosine (18F-FET) PET
for the Differential Diagnosis of Primary Brain Tumor: A Systematic Review and Metaanalysis. J. Nucl. Med.
2012, 53, 207–214. [CrossRef] [PubMed]
24. Laube, M.; Kniess, T.; Pietz, J. Radiolabeled COX-2 Inhibitors for Non-Invasive Visualization of COX-2
Expression and Activity—A Critical Update. Molecules 2013, 18, 6311–6355. [CrossRef] [PubMed]
25. Yu, C.-S.; Oberdorfer, F. Synthesis of 4-O-methyl-protected 5-(2-hydroxyethy)-2’-deoxyuridine derivatives
and their nucleophilic fluorination to 5-(2-fluoroethyl)-2’-deoxyuridine. Synthesis 1999, 2057–2064. [CrossRef]
26. Huang, H.-L.; Huang, Y.-C.; Lee, W.-L.; Yeh, C.-N.; Lin, K.-J.; Yu, C.-S. 18F-Glutathione conjugate as a
PET tracer for imaging tumors that overexpress L-PGDS enzyme. PLoS ONE 2014, 9, e104118. [CrossRef]
[PubMed]
27. Huang, Y.-C.; Huang, H.-L.; Yeh, C.-N.; Lin, K.-J.; Yu, C.-S. Investigation of brain tumors using 18F-fluorobutyl
ethacrynic amide and its metabolite with positron emission tomography. Oncotarg. Ther. 2015, 8, 1877–1885.
28. Park, B.K.; Kitteringham, N.R.; O’Neill, P.M. Metabolism of fluorine-containing drugs. Annu. Rev.
Pharmacol. Toxicol. 2001, 41, 443–470. [CrossRef] [PubMed]
29. Rendic, S.; Di Car, F.J. Human cytochrome p450 enzymes: A status report summarizing their reactions,
substrates, inducers, and inhibitors drug. Metab. Rev. 1997, 29, 413–580. [CrossRef]

Molecules 2016, 21, 387
17 of 17
30. Kharasch, E.D.; Armstrong, A.S.; Gunn, K. Clinical sevoflurane metabolism and dispodition. II. The role of
cytochrome P450 2E1 in fluoride and hexafluoroisopropanol formation. Anesthesiology 1995, 82, 1379–1388.
[CrossRef] [PubMed]
31. Fletcher, D.I.; Ganellin, C.R.; Piegentili, A.; Dunn, P.M.; Jenkinson, D.H. Synthesis and pharmacological
testing of polyamino quinolones as blockers of the apamin-sensitive Ca2+-activated K+ channel (SKCa).
Bioorg. Med. Chem. 2007, 15, 5457–5479. [CrossRef] [PubMed]
32. Bartholomä, M.D.; Vortherms, A.R.; Hillier, S.; Ploier, B.; Joyal, J.; Babich, J.; Doyle, R.P.; Zubieta, J. Synthesis,
cytotoxicity, and insight into the mode of action of Re(CO)3 thymidine complexes. Chem. Med. Chem. 2010, 5,
1513–1529. [CrossRef] [PubMed]
33. Li, Z.; Wang, Z.; Zhu, L.; Tan, X.; Li, C. Silver-catalyzed radical fluorination of alkylboronates in azueous
solution. J. Am. Chem. Soc. 2014, 136, 16439–16443. [CrossRef] [PubMed]
34. Pattison, F.L.M.; Howell, W.C.; White, R.W. Toxic fluorine compounds. VI w-Fluoro alkylamine. J. Am.
Chem. Soc. 1956, 78, 13487–13489.
35. Windhorst, A.D.; Bechger, L.; Visser, G.W.M.; Menge, W.P.M.B.; Leurs, R.; Timmerman, H.; Herscheid, J.D.M.
Approaches towards the synthesis of fluoro(cyclo)alkylamines. J. Fluo. Chem. 1996, 80, 35–40. [CrossRef]
36. Tang, D.; McKinley, E.T.; Hight, M.R.; Uddin, M.I.; Harp, J.M.; Fu, A.; Nickels, M.L.; Buck, J.R.; Manning, H.C.
Synthesis and Structure-Activity Relationships of 5,6,7-Substituted Pyrazolopyrimidines: Discovery of a
Novel TSPO PET Ligand for Cancer Imaging. J. Med. Chem. 2013, 56, 3429–3433. [CrossRef] [PubMed]
Sample Availability: Not available.
©
2016 by the authors; licensee MDPI, Basel, Switzerland. This article is an open access
article distributed under the terms and conditions of the Creative Commons by Attribution
(CC-BY) license (http://creativecommons.org/licenses/by/4.0/).